diff options
Diffstat (limited to 'arch/um/kernel')
99 files changed, 13673 insertions, 0 deletions
diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile new file mode 100644 index 00000000000..dc796c1bf39 --- /dev/null +++ b/arch/um/kernel/Makefile @@ -0,0 +1,58 @@ +# +# Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +extra-y := vmlinux.lds +clean-files := vmlinux.lds.S config.tmp + +obj-y = checksum.o config.o exec_kern.o exitcode.o \ + helper.o init_task.o irq.o irq_user.o ksyms.o main.o mem.o mem_user.o \ + physmem.o process.o process_kern.o ptrace.o reboot.o resource.o \ + sigio_user.o sigio_kern.o signal_kern.o signal_user.o smp.o \ + syscall_kern.o sysrq.o sys_call_table.o tempfile.o time.o time_kern.o \ + tlb.o trap_kern.o trap_user.o uaccess_user.o um_arch.o umid.o \ + user_util.o + +obj-$(CONFIG_BLK_DEV_INITRD) += initrd_kern.o initrd_user.o +obj-$(CONFIG_GPROF) += gprof_syms.o +obj-$(CONFIG_GCOV) += gmon_syms.o +obj-$(CONFIG_TTY_LOG) += tty_log.o +obj-$(CONFIG_SYSCALL_DEBUG) += syscall_user.o + +obj-$(CONFIG_MODE_TT) += tt/ +obj-$(CONFIG_MODE_SKAS) += skas/ + +# This needs be compiled with frame pointers regardless of how the rest of the +# kernel is built. +CFLAGS_frame.o := -fno-omit-frame-pointer + +user-objs-$(CONFIG_TTY_LOG) += tty_log.o + +USER_OBJS := $(user-objs-y) config.o helper.o main.o process.o tempfile.o \ + time.o tty_log.o umid.o user_util.o frame.o + +include arch/um/scripts/Makefile.rules + +targets += config.c + +# Be careful with the below Sed code - sed is pitfall-rich! +# We use sed to lower build requirements, for "embedded" builders for instance. + +$(obj)/config.tmp: $(objtree)/.config FORCE + $(call if_changed,quote1) + +quiet_cmd_quote1 = QUOTE $@ + cmd_quote1 = sed -e 's/"/\\"/g' -e 's/^/"/' -e 's/$$/\\n"/' \ + $< > $@ + +$(obj)/config.c: $(src)/config.c.in $(obj)/config.tmp FORCE + $(call if_changed,quote2) + +quiet_cmd_quote2 = QUOTE $@ + cmd_quote2 = sed -e '/CONFIG/{' \ + -e 's/"CONFIG"\;/""/' \ + -e 'r $(obj)/config.tmp' \ + -e 'a""\;' \ + -e '}' \ + $< > $@ diff --git a/arch/um/kernel/checksum.c b/arch/um/kernel/checksum.c new file mode 100644 index 00000000000..e69b2be951d --- /dev/null +++ b/arch/um/kernel/checksum.c @@ -0,0 +1,36 @@ +#include "asm/uaccess.h" +#include "linux/errno.h" +#include "linux/module.h" + +unsigned int arch_csum_partial(const unsigned char *buff, int len, int sum); + +unsigned int csum_partial(unsigned char *buff, int len, int sum) +{ + return arch_csum_partial(buff, len, sum); +} + +EXPORT_SYMBOL(csum_partial); + +unsigned int csum_partial_copy_to(const unsigned char *src, + unsigned char __user *dst, int len, int sum, + int *err_ptr) +{ + if(copy_to_user(dst, src, len)){ + *err_ptr = -EFAULT; + return(-1); + } + + return(arch_csum_partial(src, len, sum)); +} + +unsigned int csum_partial_copy_from(const unsigned char __user *src, + unsigned char *dst, int len, int sum, + int *err_ptr) +{ + if(copy_from_user(dst, src, len)){ + *err_ptr = -EFAULT; + return(-1); + } + + return arch_csum_partial(dst, len, sum); +} diff --git a/arch/um/kernel/config.c.in b/arch/um/kernel/config.c.in new file mode 100644 index 00000000000..c062cbfe386 --- /dev/null +++ b/arch/um/kernel/config.c.in @@ -0,0 +1,32 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include "init.h" + +static __initdata char *config = "CONFIG"; + +static int __init print_config(char *line, int *add) +{ + printf("%s", config); + exit(0); +} + +__uml_setup("--showconfig", print_config, +"--showconfig\n" +" Prints the config file that this UML binary was generated from.\n\n" +); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/dyn.lds.S b/arch/um/kernel/dyn.lds.S new file mode 100644 index 00000000000..715b0838a68 --- /dev/null +++ b/arch/um/kernel/dyn.lds.S @@ -0,0 +1,176 @@ +#include <asm-generic/vmlinux.lds.h> + +OUTPUT_FORMAT(ELF_FORMAT) +OUTPUT_ARCH(ELF_ARCH) +ENTRY(_start) +jiffies = jiffies_64; + +SECTIONS +{ + PROVIDE (__executable_start = START); + . = START + SIZEOF_HEADERS; + .interp : { *(.interp) } + /* Used in arch/um/kernel/mem.c. Any memory between START and __binary_start + * is remapped.*/ + __binary_start = .; + . = ALIGN(4096); /* Init code and data */ + _stext = .; + __init_begin = .; + .init.text : { + _sinittext = .; + *(.init.text) + _einittext = .; + } + + . = ALIGN(4096); + + /* Read-only sections, merged into text segment: */ + .hash : { *(.hash) } + .dynsym : { *(.dynsym) } + .dynstr : { *(.dynstr) } + .gnu.version : { *(.gnu.version) } + .gnu.version_d : { *(.gnu.version_d) } + .gnu.version_r : { *(.gnu.version_r) } + .rel.init : { *(.rel.init) } + .rela.init : { *(.rela.init) } + .rel.text : { *(.rel.text .rel.text.* .rel.gnu.linkonce.t.*) } + .rela.text : { *(.rela.text .rela.text.* .rela.gnu.linkonce.t.*) } + .rel.fini : { *(.rel.fini) } + .rela.fini : { *(.rela.fini) } + .rel.rodata : { *(.rel.rodata .rel.rodata.* .rel.gnu.linkonce.r.*) } + .rela.rodata : { *(.rela.rodata .rela.rodata.* .rela.gnu.linkonce.r.*) } + .rel.data : { *(.rel.data .rel.data.* .rel.gnu.linkonce.d.*) } + .rela.data : { *(.rela.data .rela.data.* .rela.gnu.linkonce.d.*) } + .rel.tdata : { *(.rel.tdata .rel.tdata.* .rel.gnu.linkonce.td.*) } + .rela.tdata : { *(.rela.tdata .rela.tdata.* .rela.gnu.linkonce.td.*) } + .rel.tbss : { *(.rel.tbss .rel.tbss.* .rel.gnu.linkonce.tb.*) } + .rela.tbss : { *(.rela.tbss .rela.tbss.* .rela.gnu.linkonce.tb.*) } + .rel.ctors : { *(.rel.ctors) } + .rela.ctors : { *(.rela.ctors) } + .rel.dtors : { *(.rel.dtors) } + .rela.dtors : { *(.rela.dtors) } + .rel.got : { *(.rel.got) } + .rela.got : { *(.rela.got) } + .rel.bss : { *(.rel.bss .rel.bss.* .rel.gnu.linkonce.b.*) } + .rela.bss : { *(.rela.bss .rela.bss.* .rela.gnu.linkonce.b.*) } + .rel.plt : { *(.rel.plt) } + .rela.plt : { *(.rela.plt) } + .init : { + KEEP (*(.init)) + } =0x90909090 + .plt : { *(.plt) } + .text : { + *(.text) + SCHED_TEXT + LOCK_TEXT + *(.fixup) + *(.stub .text.* .gnu.linkonce.t.*) + /* .gnu.warning sections are handled specially by elf32.em. */ + *(.gnu.warning) + } =0x90909090 + .fini : { + KEEP (*(.fini)) + } =0x90909090 + + .kstrtab : { *(.kstrtab) } + + #include "asm/common.lds.S" + + init.data : { *(.init.data) } + + /* Ensure the __preinit_array_start label is properly aligned. We + could instead move the label definition inside the section, but + the linker would then create the section even if it turns out to + be empty, which isn't pretty. */ + . = ALIGN(32 / 8); + .preinit_array : { *(.preinit_array) } + .init_array : { *(.init_array) } + .fini_array : { *(.fini_array) } + .data : { + . = ALIGN(KERNEL_STACK_SIZE); /* init_task */ + *(.data.init_task) + *(.data .data.* .gnu.linkonce.d.*) + SORT(CONSTRUCTORS) + } + .data1 : { *(.data1) } + .tdata : { *(.tdata .tdata.* .gnu.linkonce.td.*) } + .tbss : { *(.tbss .tbss.* .gnu.linkonce.tb.*) *(.tcommon) } + .eh_frame : { KEEP (*(.eh_frame)) } + .gcc_except_table : { *(.gcc_except_table) } + .dynamic : { *(.dynamic) } + .ctors : { + /* gcc uses crtbegin.o to find the start of + the constructors, so we make sure it is + first. Because this is a wildcard, it + doesn't matter if the user does not + actually link against crtbegin.o; the + linker won't look for a file to match a + wildcard. The wildcard also means that it + doesn't matter which directory crtbegin.o + is in. */ + KEEP (*crtbegin.o(.ctors)) + /* We don't want to include the .ctor section from + from the crtend.o file until after the sorted ctors. + The .ctor section from the crtend file contains the + end of ctors marker and it must be last */ + KEEP (*(EXCLUDE_FILE (*crtend.o ) .ctors)) + KEEP (*(SORT(.ctors.*))) + KEEP (*(.ctors)) + } + .dtors : { + KEEP (*crtbegin.o(.dtors)) + KEEP (*(EXCLUDE_FILE (*crtend.o ) .dtors)) + KEEP (*(SORT(.dtors.*))) + KEEP (*(.dtors)) + } + .jcr : { KEEP (*(.jcr)) } + .got : { *(.got.plt) *(.got) } + _edata = .; + PROVIDE (edata = .); + __bss_start = .; + .bss : { + *(.dynbss) + *(.bss .bss.* .gnu.linkonce.b.*) + *(COMMON) + /* Align here to ensure that the .bss section occupies space up to + _end. Align after .bss to ensure correct alignment even if the + .bss section disappears because there are no input sections. */ + . = ALIGN(32 / 8); + . = ALIGN(32 / 8); + } + _end = .; + PROVIDE (end = .); + /* Stabs debugging sections. */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } + /* DWARF debug sections. + Symbols in the DWARF debugging sections are relative to the beginning + of the section so we begin them at 0. */ + /* DWARF 1 */ + .debug 0 : { *(.debug) } + .line 0 : { *(.line) } + /* GNU DWARF 1 extensions */ + .debug_srcinfo 0 : { *(.debug_srcinfo) } + .debug_sfnames 0 : { *(.debug_sfnames) } + /* DWARF 1.1 and DWARF 2 */ + .debug_aranges 0 : { *(.debug_aranges) } + .debug_pubnames 0 : { *(.debug_pubnames) } + /* DWARF 2 */ + .debug_info 0 : { *(.debug_info .gnu.linkonce.wi.*) } + .debug_abbrev 0 : { *(.debug_abbrev) } + .debug_line 0 : { *(.debug_line) } + .debug_frame 0 : { *(.debug_frame) } + .debug_str 0 : { *(.debug_str) } + .debug_loc 0 : { *(.debug_loc) } + .debug_macinfo 0 : { *(.debug_macinfo) } + /* SGI/MIPS DWARF 2 extensions */ + .debug_weaknames 0 : { *(.debug_weaknames) } + .debug_funcnames 0 : { *(.debug_funcnames) } + .debug_typenames 0 : { *(.debug_typenames) } + .debug_varnames 0 : { *(.debug_varnames) } +} diff --git a/arch/um/kernel/exec_kern.c b/arch/um/kernel/exec_kern.c new file mode 100644 index 00000000000..49ddabe69be --- /dev/null +++ b/arch/um/kernel/exec_kern.c @@ -0,0 +1,91 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/slab.h" +#include "linux/smp_lock.h" +#include "linux/ptrace.h" +#include "asm/ptrace.h" +#include "asm/pgtable.h" +#include "asm/tlbflush.h" +#include "asm/uaccess.h" +#include "user_util.h" +#include "kern_util.h" +#include "mem_user.h" +#include "kern.h" +#include "irq_user.h" +#include "tlb.h" +#include "2_5compat.h" +#include "os.h" +#include "time_user.h" +#include "choose-mode.h" +#include "mode_kern.h" + +void flush_thread(void) +{ + CHOOSE_MODE(flush_thread_tt(), flush_thread_skas()); +} + +void start_thread(struct pt_regs *regs, unsigned long eip, unsigned long esp) +{ + CHOOSE_MODE_PROC(start_thread_tt, start_thread_skas, regs, eip, esp); +} + +extern void log_exec(char **argv, void *tty); + +static long execve1(char *file, char __user * __user *argv, + char *__user __user *env) +{ + long error; + +#ifdef CONFIG_TTY_LOG + log_exec(argv, current->tty); +#endif + error = do_execve(file, argv, env, ¤t->thread.regs); + if (error == 0){ + task_lock(current); + current->ptrace &= ~PT_DTRACE; + task_unlock(current); + set_cmdline(current_cmd()); + } + return(error); +} + +long um_execve(char *file, char __user *__user *argv, char __user *__user *env) +{ + long err; + + err = execve1(file, argv, env); + if(!err) + do_longjmp(current->thread.exec_buf, 1); + return(err); +} + +long sys_execve(char *file, char __user *__user *argv, + char __user *__user *env) +{ + long error; + char *filename; + + lock_kernel(); + filename = getname((char __user *) file); + error = PTR_ERR(filename); + if (IS_ERR(filename)) goto out; + error = execve1(filename, argv, env); + putname(filename); + out: + unlock_kernel(); + return(error); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/exitcode.c b/arch/um/kernel/exitcode.c new file mode 100644 index 00000000000..0ea87f24b36 --- /dev/null +++ b/arch/um/kernel/exitcode.c @@ -0,0 +1,73 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/init.h" +#include "linux/ctype.h" +#include "linux/proc_fs.h" +#include "asm/uaccess.h" + +/* If read and write race, the read will still atomically read a valid + * value. + */ +int uml_exitcode = 0; + +static int read_proc_exitcode(char *page, char **start, off_t off, + int count, int *eof, void *data) +{ + int len; + + len = sprintf(page, "%d\n", uml_exitcode); + len -= off; + if(len <= off+count) *eof = 1; + *start = page + off; + if(len > count) len = count; + if(len < 0) len = 0; + return(len); +} + +static int write_proc_exitcode(struct file *file, const char __user *buffer, + unsigned long count, void *data) +{ + char *end, buf[sizeof("nnnnn\0")]; + int tmp; + + if(copy_from_user(buf, buffer, count)) + return(-EFAULT); + tmp = simple_strtol(buf, &end, 0); + if((*end != '\0') && !isspace(*end)) + return(-EINVAL); + uml_exitcode = tmp; + return(count); +} + +static int make_proc_exitcode(void) +{ + struct proc_dir_entry *ent; + + ent = create_proc_entry("exitcode", 0600, &proc_root); + if(ent == NULL){ + printk("make_proc_exitcode : Failed to register " + "/proc/exitcode\n"); + return(0); + } + + ent->read_proc = read_proc_exitcode; + ent->write_proc = write_proc_exitcode; + + return(0); +} + +__initcall(make_proc_exitcode); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/gmon_syms.c b/arch/um/kernel/gmon_syms.c new file mode 100644 index 00000000000..2c86e7fdb01 --- /dev/null +++ b/arch/um/kernel/gmon_syms.c @@ -0,0 +1,34 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/module.h" + +extern void __bb_init_func(void *); +EXPORT_SYMBOL(__bb_init_func); + +/* This is defined (and referred to in profiling stub code) only by some GCC + * versions in libgcov. + * + * Since SuSE backported the fix, we cannot handle it depending on GCC version. + * So, unconditinally export it. But also give it a weak declaration, which will + * be overriden by any other one. + */ + +extern void __gcov_init(void *) __attribute__((weak)); +EXPORT_SYMBOL(__gcov_init); + +extern void __gcov_merge_add(void *) __attribute__((weak)); +EXPORT_SYMBOL(__gcov_merge_add); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/gprof_syms.c b/arch/um/kernel/gprof_syms.c new file mode 100644 index 00000000000..9244f018d44 --- /dev/null +++ b/arch/um/kernel/gprof_syms.c @@ -0,0 +1,20 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/module.h" + +extern void mcount(void); +EXPORT_SYMBOL(mcount); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/helper.c b/arch/um/kernel/helper.c new file mode 100644 index 00000000000..13b1f5c2f7e --- /dev/null +++ b/arch/um/kernel/helper.c @@ -0,0 +1,173 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <errno.h> +#include <sched.h> +#include <sys/signal.h> +#include <sys/wait.h> +#include "user.h" +#include "kern_util.h" +#include "user_util.h" +#include "os.h" + +struct helper_data { + void (*pre_exec)(void*); + void *pre_data; + char **argv; + int fd; +}; + +/* Debugging aid, changed only from gdb */ +int helper_pause = 0; + +static void helper_hup(int sig) +{ +} + +static int helper_child(void *arg) +{ + struct helper_data *data = arg; + char **argv = data->argv; + int errval; + + if(helper_pause){ + signal(SIGHUP, helper_hup); + pause(); + } + if(data->pre_exec != NULL) + (*data->pre_exec)(data->pre_data); + execvp(argv[0], argv); + errval = errno; + printk("execvp of '%s' failed - errno = %d\n", argv[0], errno); + os_write_file(data->fd, &errval, sizeof(errval)); + os_kill_process(os_getpid(), 0); + return(0); +} + +/* Returns either the pid of the child process we run or -E* on failure. + * XXX The alloc_stack here breaks if this is called in the tracing thread */ +int run_helper(void (*pre_exec)(void *), void *pre_data, char **argv, + unsigned long *stack_out) +{ + struct helper_data data; + unsigned long stack, sp; + int pid, fds[2], ret, n; + + if((stack_out != NULL) && (*stack_out != 0)) + stack = *stack_out; + else stack = alloc_stack(0, um_in_interrupt()); + if(stack == 0) + return(-ENOMEM); + + ret = os_pipe(fds, 1, 0); + if(ret < 0){ + printk("run_helper : pipe failed, ret = %d\n", -ret); + goto out_free; + } + + ret = os_set_exec_close(fds[1], 1); + if(ret < 0){ + printk("run_helper : setting FD_CLOEXEC failed, ret = %d\n", + -ret); + goto out_close; + } + + sp = stack + page_size() - sizeof(void *); + data.pre_exec = pre_exec; + data.pre_data = pre_data; + data.argv = argv; + data.fd = fds[1]; + pid = clone(helper_child, (void *) sp, CLONE_VM | SIGCHLD, &data); + if(pid < 0){ + printk("run_helper : clone failed, errno = %d\n", errno); + ret = -errno; + goto out_close; + } + + os_close_file(fds[1]); + fds[1] = -1; + + /*Read the errno value from the child.*/ + n = os_read_file(fds[0], &ret, sizeof(ret)); + if(n < 0){ + printk("run_helper : read on pipe failed, ret = %d\n", -n); + ret = n; + os_kill_process(pid, 1); + } + else if(n != 0){ + CATCH_EINTR(n = waitpid(pid, NULL, 0)); + ret = -errno; + } else { + ret = pid; + } + +out_close: + if (fds[1] != -1) + os_close_file(fds[1]); + os_close_file(fds[0]); +out_free: + if(stack_out == NULL) + free_stack(stack, 0); + else *stack_out = stack; + return(ret); +} + +int run_helper_thread(int (*proc)(void *), void *arg, unsigned int flags, + unsigned long *stack_out, int stack_order) +{ + unsigned long stack, sp; + int pid, status; + + stack = alloc_stack(stack_order, um_in_interrupt()); + if(stack == 0) return(-ENOMEM); + + sp = stack + (page_size() << stack_order) - sizeof(void *); + pid = clone(proc, (void *) sp, flags | SIGCHLD, arg); + if(pid < 0){ + printk("run_helper_thread : clone failed, errno = %d\n", + errno); + return(-errno); + } + if(stack_out == NULL){ + CATCH_EINTR(pid = waitpid(pid, &status, 0)); + if(pid < 0){ + printk("run_helper_thread - wait failed, errno = %d\n", + errno); + pid = -errno; + } + if(!WIFEXITED(status) || (WEXITSTATUS(status) != 0)) + printk("run_helper_thread - thread returned status " + "0x%x\n", status); + free_stack(stack, stack_order); + } + else *stack_out = stack; + return(pid); +} + +int helper_wait(int pid, int block) +{ + int ret; + + CATCH_EINTR(ret = waitpid(pid, NULL, WNOHANG)); + if(ret < 0){ + printk("helper_wait : waitpid failed, errno = %d\n", errno); + return(-errno); + } + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/init_task.c b/arch/um/kernel/init_task.c new file mode 100644 index 00000000000..cd7c85be0a1 --- /dev/null +++ b/arch/um/kernel/init_task.c @@ -0,0 +1,61 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/mm.h" +#include "linux/module.h" +#include "linux/sched.h" +#include "linux/init_task.h" +#include "linux/mqueue.h" +#include "asm/uaccess.h" +#include "asm/pgtable.h" +#include "user_util.h" +#include "mem_user.h" + +static struct fs_struct init_fs = INIT_FS; +struct mm_struct init_mm = INIT_MM(init_mm); +static struct files_struct init_files = INIT_FILES; +static struct signal_struct init_signals = INIT_SIGNALS(init_signals); +static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand); +EXPORT_SYMBOL(init_mm); + +/* + * Initial task structure. + * + * All other task structs will be allocated on slabs in fork.c + */ + +struct task_struct init_task = INIT_TASK(init_task); + +EXPORT_SYMBOL(init_task); + +/* + * Initial thread structure. + * + * We need to make sure that this is 16384-byte aligned due to the + * way process stacks are handled. This is done by having a special + * "init_task" linker map entry.. + */ + +union thread_union init_thread_union +__attribute__((__section__(".data.init_task"))) = +{ INIT_THREAD_INFO(init_task) }; + +void unprotect_stack(unsigned long stack) +{ + protect_memory(stack, (1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE, + 1, 1, 0, 1); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/initrd_kern.c b/arch/um/kernel/initrd_kern.c new file mode 100644 index 00000000000..fc568af468b --- /dev/null +++ b/arch/um/kernel/initrd_kern.c @@ -0,0 +1,59 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/init.h" +#include "linux/bootmem.h" +#include "linux/initrd.h" +#include "asm/types.h" +#include "user_util.h" +#include "kern_util.h" +#include "initrd.h" +#include "init.h" +#include "os.h" + +/* Changed by uml_initrd_setup, which is a setup */ +static char *initrd __initdata = NULL; + +static int __init read_initrd(void) +{ + void *area; + long long size; + int err; + + if(initrd == NULL) return 0; + err = os_file_size(initrd, &size); + if(err) return 0; + area = alloc_bootmem(size); + if(area == NULL) return 0; + if(load_initrd(initrd, area, size) == -1) return 0; + initrd_start = (unsigned long) area; + initrd_end = initrd_start + size; + return 0; +} + +__uml_postsetup(read_initrd); + +static int __init uml_initrd_setup(char *line, int *add) +{ + initrd = line; + return 0; +} + +__uml_setup("initrd=", uml_initrd_setup, +"initrd=<initrd image>\n" +" This is used to boot UML from an initrd image. The argument is the\n" +" name of the file containing the image.\n\n" +); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/initrd_user.c b/arch/um/kernel/initrd_user.c new file mode 100644 index 00000000000..cb90681e151 --- /dev/null +++ b/arch/um/kernel/initrd_user.c @@ -0,0 +1,46 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <sys/types.h> +#include <sys/stat.h> +#include <errno.h> + +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "initrd.h" +#include "os.h" + +int load_initrd(char *filename, void *buf, int size) +{ + int fd, n; + + fd = os_open_file(filename, of_read(OPENFLAGS()), 0); + if(fd < 0){ + printk("Opening '%s' failed - err = %d\n", filename, -fd); + return(-1); + } + n = os_read_file(fd, buf, size); + if(n != size){ + printk("Read of %d bytes from '%s' failed, err = %d\n", size, + filename, -n); + return(-1); + } + + os_close_file(fd); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/irq.c b/arch/um/kernel/irq.c new file mode 100644 index 00000000000..d71e8f00810 --- /dev/null +++ b/arch/um/kernel/irq.c @@ -0,0 +1,178 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + * Derived (i.e. mostly copied) from arch/i386/kernel/irq.c: + * Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar + */ + +#include "linux/config.h" +#include "linux/kernel.h" +#include "linux/module.h" +#include "linux/smp.h" +#include "linux/irq.h" +#include "linux/kernel_stat.h" +#include "linux/interrupt.h" +#include "linux/random.h" +#include "linux/slab.h" +#include "linux/file.h" +#include "linux/proc_fs.h" +#include "linux/init.h" +#include "linux/seq_file.h" +#include "linux/profile.h" +#include "linux/hardirq.h" +#include "asm/irq.h" +#include "asm/hw_irq.h" +#include "asm/atomic.h" +#include "asm/signal.h" +#include "asm/system.h" +#include "asm/errno.h" +#include "asm/uaccess.h" +#include "user_util.h" +#include "kern_util.h" +#include "irq_user.h" +#include "irq_kern.h" + + +/* + * Generic, controller-independent functions: + */ + +int show_interrupts(struct seq_file *p, void *v) +{ + int i = *(loff_t *) v, j; + struct irqaction * action; + unsigned long flags; + + if (i == 0) { + seq_printf(p, " "); + for_each_online_cpu(j) + seq_printf(p, "CPU%d ",j); + seq_putc(p, '\n'); + } + + if (i < NR_IRQS) { + spin_lock_irqsave(&irq_desc[i].lock, flags); + action = irq_desc[i].action; + if (!action) + goto skip; + seq_printf(p, "%3d: ",i); +#ifndef CONFIG_SMP + seq_printf(p, "%10u ", kstat_irqs(i)); +#else + for_each_online_cpu(j) + seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]); +#endif + seq_printf(p, " %14s", irq_desc[i].handler->typename); + seq_printf(p, " %s", action->name); + + for (action=action->next; action; action = action->next) + seq_printf(p, ", %s", action->name); + + seq_putc(p, '\n'); +skip: + spin_unlock_irqrestore(&irq_desc[i].lock, flags); + } else if (i == NR_IRQS) { + seq_putc(p, '\n'); + } + + return 0; +} + +/* + * do_IRQ handles all normal device IRQ's (the special + * SMP cross-CPU interrupts have their own specific + * handlers). + */ +unsigned int do_IRQ(int irq, union uml_pt_regs *regs) +{ + irq_enter(); + __do_IRQ(irq, (struct pt_regs *) regs); + irq_exit(); + return 1; +} + +int um_request_irq(unsigned int irq, int fd, int type, + irqreturn_t (*handler)(int, void *, struct pt_regs *), + unsigned long irqflags, const char * devname, + void *dev_id) +{ + int err; + + err = request_irq(irq, handler, irqflags, devname, dev_id); + if(err) + return(err); + + if(fd != -1) + err = activate_fd(irq, fd, type, dev_id); + return(err); +} +EXPORT_SYMBOL(um_request_irq); +EXPORT_SYMBOL(reactivate_fd); + +static DEFINE_SPINLOCK(irq_spinlock); + +unsigned long irq_lock(void) +{ + unsigned long flags; + + spin_lock_irqsave(&irq_spinlock, flags); + return(flags); +} + +void irq_unlock(unsigned long flags) +{ + spin_unlock_irqrestore(&irq_spinlock, flags); +} + +/* presently hw_interrupt_type must define (startup || enable) && + * disable && end */ +static void dummy(unsigned int irq) +{ +} + +static struct hw_interrupt_type SIGIO_irq_type = { + .typename = "SIGIO", + .disable = dummy, + .enable = dummy, + .ack = dummy, + .end = dummy +}; + +static struct hw_interrupt_type SIGVTALRM_irq_type = { + .typename = "SIGVTALRM", + .shutdown = dummy, /* never called */ + .disable = dummy, + .enable = dummy, + .ack = dummy, + .end = dummy +}; + +void __init init_IRQ(void) +{ + int i; + + irq_desc[TIMER_IRQ].status = IRQ_DISABLED; + irq_desc[TIMER_IRQ].action = NULL; + irq_desc[TIMER_IRQ].depth = 1; + irq_desc[TIMER_IRQ].handler = &SIGVTALRM_irq_type; + enable_irq(TIMER_IRQ); + for(i=1;i<NR_IRQS;i++){ + irq_desc[i].status = IRQ_DISABLED; + irq_desc[i].action = NULL; + irq_desc[i].depth = 1; + irq_desc[i].handler = &SIGIO_irq_type; + enable_irq(i); + } + init_irq_signals(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/irq_user.c b/arch/um/kernel/irq_user.c new file mode 100644 index 00000000000..6d6f9484b88 --- /dev/null +++ b/arch/um/kernel/irq_user.c @@ -0,0 +1,443 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <unistd.h> +#include <errno.h> +#include <signal.h> +#include <string.h> +#include <sys/poll.h> +#include <sys/types.h> +#include <sys/time.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "process.h" +#include "signal_user.h" +#include "sigio.h" +#include "irq_user.h" +#include "os.h" + +struct irq_fd { + struct irq_fd *next; + void *id; + int fd; + int type; + int irq; + int pid; + int events; + int current_events; + int freed; +}; + +static struct irq_fd *active_fds = NULL; +static struct irq_fd **last_irq_ptr = &active_fds; + +static struct pollfd *pollfds = NULL; +static int pollfds_num = 0; +static int pollfds_size = 0; + +extern int io_count, intr_count; + +void sigio_handler(int sig, union uml_pt_regs *regs) +{ + struct irq_fd *irq_fd, *next; + int i, n; + + if(smp_sigio_handler()) return; + while(1){ + n = poll(pollfds, pollfds_num, 0); + if(n < 0){ + if(errno == EINTR) continue; + printk("sigio_handler : poll returned %d, " + "errno = %d\n", n, errno); + break; + } + if(n == 0) break; + + irq_fd = active_fds; + for(i = 0; i < pollfds_num; i++){ + if(pollfds[i].revents != 0){ + irq_fd->current_events = pollfds[i].revents; + pollfds[i].fd = -1; + } + irq_fd = irq_fd->next; + } + + for(irq_fd = active_fds; irq_fd != NULL; irq_fd = next){ + next = irq_fd->next; + if(irq_fd->current_events != 0){ + irq_fd->current_events = 0; + do_IRQ(irq_fd->irq, regs); + + /* This is here because the next irq may be + * freed in the handler. If a console goes + * away, both the read and write irqs will be + * freed. After do_IRQ, ->next will point to + * a good IRQ. + * Irqs can't be freed inside their handlers, + * so the next best thing is to have them + * marked as needing freeing, so that they + * can be freed here. + */ + next = irq_fd->next; + if(irq_fd->freed){ + free_irq(irq_fd->irq, irq_fd->id); + free_irq_by_irq_and_dev(irq_fd->irq, + irq_fd->id); + } + } + } + } +} + +int activate_ipi(int fd, int pid) +{ + return(os_set_fd_async(fd, pid)); +} + +static void maybe_sigio_broken(int fd, int type) +{ + if(isatty(fd)){ + if((type == IRQ_WRITE) && !pty_output_sigio){ + write_sigio_workaround(); + add_sigio_fd(fd, 0); + } + else if((type == IRQ_READ) && !pty_close_sigio){ + write_sigio_workaround(); + add_sigio_fd(fd, 1); + } + } +} + +int activate_fd(int irq, int fd, int type, void *dev_id) +{ + struct pollfd *tmp_pfd; + struct irq_fd *new_fd, *irq_fd; + unsigned long flags; + int pid, events, err, n, size; + + pid = os_getpid(); + err = os_set_fd_async(fd, pid); + if(err < 0) + goto out; + + new_fd = um_kmalloc(sizeof(*new_fd)); + err = -ENOMEM; + if(new_fd == NULL) + goto out; + + if(type == IRQ_READ) events = POLLIN | POLLPRI; + else events = POLLOUT; + *new_fd = ((struct irq_fd) { .next = NULL, + .id = dev_id, + .fd = fd, + .type = type, + .irq = irq, + .pid = pid, + .events = events, + .current_events = 0, + .freed = 0 } ); + + /* Critical section - locked by a spinlock because this stuff can + * be changed from interrupt handlers. The stuff above is done + * outside the lock because it allocates memory. + */ + + /* Actually, it only looks like it can be called from interrupt + * context. The culprit is reactivate_fd, which calls + * maybe_sigio_broken, which calls write_sigio_workaround, + * which calls activate_fd. However, write_sigio_workaround should + * only be called once, at boot time. That would make it clear that + * this is called only from process context, and can be locked with + * a semaphore. + */ + flags = irq_lock(); + for(irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next){ + if((irq_fd->fd == fd) && (irq_fd->type == type)){ + printk("Registering fd %d twice\n", fd); + printk("Irqs : %d, %d\n", irq_fd->irq, irq); + printk("Ids : 0x%x, 0x%x\n", irq_fd->id, dev_id); + goto out_unlock; + } + } + + n = pollfds_num; + if(n == pollfds_size){ + while(1){ + /* Here we have to drop the lock in order to call + * kmalloc, which might sleep. If something else + * came in and changed the pollfds array, we free + * the buffer and try again. + */ + irq_unlock(flags); + size = (pollfds_num + 1) * sizeof(pollfds[0]); + tmp_pfd = um_kmalloc(size); + flags = irq_lock(); + if(tmp_pfd == NULL) + goto out_unlock; + if(n == pollfds_size) + break; + kfree(tmp_pfd); + } + if(pollfds != NULL){ + memcpy(tmp_pfd, pollfds, + sizeof(pollfds[0]) * pollfds_size); + kfree(pollfds); + } + pollfds = tmp_pfd; + pollfds_size++; + } + + if(type == IRQ_WRITE) + fd = -1; + + pollfds[pollfds_num] = ((struct pollfd) { .fd = fd, + .events = events, + .revents = 0 }); + pollfds_num++; + + *last_irq_ptr = new_fd; + last_irq_ptr = &new_fd->next; + + irq_unlock(flags); + + /* This calls activate_fd, so it has to be outside the critical + * section. + */ + maybe_sigio_broken(fd, type); + + return(0); + + out_unlock: + irq_unlock(flags); + kfree(new_fd); + out: + return(err); +} + +static void free_irq_by_cb(int (*test)(struct irq_fd *, void *), void *arg) +{ + struct irq_fd **prev; + unsigned long flags; + int i = 0; + + flags = irq_lock(); + prev = &active_fds; + while(*prev != NULL){ + if((*test)(*prev, arg)){ + struct irq_fd *old_fd = *prev; + if((pollfds[i].fd != -1) && + (pollfds[i].fd != (*prev)->fd)){ + printk("free_irq_by_cb - mismatch between " + "active_fds and pollfds, fd %d vs %d\n", + (*prev)->fd, pollfds[i].fd); + goto out; + } + memcpy(&pollfds[i], &pollfds[i + 1], + (pollfds_num - i - 1) * sizeof(pollfds[0])); + pollfds_num--; + if(last_irq_ptr == &old_fd->next) + last_irq_ptr = prev; + *prev = (*prev)->next; + if(old_fd->type == IRQ_WRITE) + ignore_sigio_fd(old_fd->fd); + kfree(old_fd); + continue; + } + prev = &(*prev)->next; + i++; + } + out: + irq_unlock(flags); +} + +struct irq_and_dev { + int irq; + void *dev; +}; + +static int same_irq_and_dev(struct irq_fd *irq, void *d) +{ + struct irq_and_dev *data = d; + + return((irq->irq == data->irq) && (irq->id == data->dev)); +} + +void free_irq_by_irq_and_dev(unsigned int irq, void *dev) +{ + struct irq_and_dev data = ((struct irq_and_dev) { .irq = irq, + .dev = dev }); + + free_irq_by_cb(same_irq_and_dev, &data); +} + +static int same_fd(struct irq_fd *irq, void *fd) +{ + return(irq->fd == *((int *) fd)); +} + +void free_irq_by_fd(int fd) +{ + free_irq_by_cb(same_fd, &fd); +} + +static struct irq_fd *find_irq_by_fd(int fd, int irqnum, int *index_out) +{ + struct irq_fd *irq; + int i = 0; + + for(irq=active_fds; irq != NULL; irq = irq->next){ + if((irq->fd == fd) && (irq->irq == irqnum)) break; + i++; + } + if(irq == NULL){ + printk("find_irq_by_fd doesn't have descriptor %d\n", fd); + goto out; + } + if((pollfds[i].fd != -1) && (pollfds[i].fd != fd)){ + printk("find_irq_by_fd - mismatch between active_fds and " + "pollfds, fd %d vs %d, need %d\n", irq->fd, + pollfds[i].fd, fd); + irq = NULL; + goto out; + } + *index_out = i; + out: + return(irq); +} + +void free_irq_later(int irq, void *dev_id) +{ + struct irq_fd *irq_fd; + unsigned long flags; + + flags = irq_lock(); + for(irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next){ + if((irq_fd->irq == irq) && (irq_fd->id == dev_id)) + break; + } + if(irq_fd == NULL){ + printk("free_irq_later found no irq, irq = %d, " + "dev_id = 0x%p\n", irq, dev_id); + goto out; + } + irq_fd->freed = 1; + out: + irq_unlock(flags); +} + +void reactivate_fd(int fd, int irqnum) +{ + struct irq_fd *irq; + unsigned long flags; + int i; + + flags = irq_lock(); + irq = find_irq_by_fd(fd, irqnum, &i); + if(irq == NULL){ + irq_unlock(flags); + return; + } + + pollfds[i].fd = irq->fd; + + irq_unlock(flags); + + /* This calls activate_fd, so it has to be outside the critical + * section. + */ + maybe_sigio_broken(fd, irq->type); +} + +void deactivate_fd(int fd, int irqnum) +{ + struct irq_fd *irq; + unsigned long flags; + int i; + + flags = irq_lock(); + irq = find_irq_by_fd(fd, irqnum, &i); + if(irq == NULL) + goto out; + pollfds[i].fd = -1; + out: + irq_unlock(flags); +} + +int deactivate_all_fds(void) +{ + struct irq_fd *irq; + int err; + + for(irq=active_fds;irq != NULL;irq = irq->next){ + err = os_clear_fd_async(irq->fd); + if(err) + return(err); + } + /* If there is a signal already queued, after unblocking ignore it */ + set_handler(SIGIO, SIG_IGN, 0, -1); + + return(0); +} + +void forward_ipi(int fd, int pid) +{ + int err; + + err = os_set_owner(fd, pid); + if(err < 0) + printk("forward_ipi: set_owner failed, fd = %d, me = %d, " + "target = %d, err = %d\n", fd, os_getpid(), pid, -err); +} + +void forward_interrupts(int pid) +{ + struct irq_fd *irq; + unsigned long flags; + int err; + + flags = irq_lock(); + for(irq=active_fds;irq != NULL;irq = irq->next){ + err = os_set_owner(irq->fd, pid); + if(err < 0){ + /* XXX Just remove the irq rather than + * print out an infinite stream of these + */ + printk("Failed to forward %d to pid %d, err = %d\n", + irq->fd, pid, -err); + } + + irq->pid = pid; + } + irq_unlock(flags); +} + +void init_irq_signals(int on_sigstack) +{ + __sighandler_t h; + int flags; + + flags = on_sigstack ? SA_ONSTACK : 0; + if(timer_irq_inited) h = (__sighandler_t) alarm_handler; + else h = boot_timer_handler; + + set_handler(SIGVTALRM, h, flags | SA_RESTART, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, -1); + set_handler(SIGIO, (__sighandler_t) sig_handler, flags | SA_RESTART, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + signal(SIGWINCH, SIG_IGN); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c new file mode 100644 index 00000000000..b41d3397d07 --- /dev/null +++ b/arch/um/kernel/ksyms.c @@ -0,0 +1,137 @@ +/* + * Copyright (C) 2001 - 2004 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/module.h" +#include "linux/string.h" +#include "linux/smp_lock.h" +#include "linux/spinlock.h" +#include "linux/highmem.h" +#include "asm/current.h" +#include "asm/delay.h" +#include "asm/processor.h" +#include "asm/unistd.h" +#include "asm/pgalloc.h" +#include "asm/pgtable.h" +#include "asm/page.h" +#include "asm/tlbflush.h" +#include "kern_util.h" +#include "user_util.h" +#include "mem_user.h" +#include "os.h" +#include "helper.h" + +EXPORT_SYMBOL(stop); +EXPORT_SYMBOL(uml_physmem); +EXPORT_SYMBOL(set_signals); +EXPORT_SYMBOL(get_signals); +EXPORT_SYMBOL(kernel_thread); +EXPORT_SYMBOL(__const_udelay); +EXPORT_SYMBOL(__udelay); +EXPORT_SYMBOL(sys_waitpid); +EXPORT_SYMBOL(task_size); +EXPORT_SYMBOL(flush_tlb_range); +EXPORT_SYMBOL(host_task_size); +EXPORT_SYMBOL(arch_validate); +EXPORT_SYMBOL(get_kmem_end); + +EXPORT_SYMBOL(page_to_phys); +EXPORT_SYMBOL(phys_to_page); +EXPORT_SYMBOL(high_physmem); +EXPORT_SYMBOL(empty_zero_page); +EXPORT_SYMBOL(um_virt_to_phys); +EXPORT_SYMBOL(__virt_to_page); +EXPORT_SYMBOL(to_phys); +EXPORT_SYMBOL(to_virt); +EXPORT_SYMBOL(mode_tt); +EXPORT_SYMBOL(handle_page_fault); +EXPORT_SYMBOL(find_iomem); +EXPORT_SYMBOL(end_iomem); + +#ifdef CONFIG_MODE_TT +EXPORT_SYMBOL(strncpy_from_user_tt); +EXPORT_SYMBOL(copy_from_user_tt); +EXPORT_SYMBOL(copy_to_user_tt); +#endif + +#ifdef CONFIG_MODE_SKAS +EXPORT_SYMBOL(strncpy_from_user_skas); +EXPORT_SYMBOL(copy_to_user_skas); +EXPORT_SYMBOL(copy_from_user_skas); +#endif +EXPORT_SYMBOL(uml_strdup); + +EXPORT_SYMBOL(os_stat_fd); +EXPORT_SYMBOL(os_stat_file); +EXPORT_SYMBOL(os_access); +EXPORT_SYMBOL(os_print_error); +EXPORT_SYMBOL(os_get_exec_close); +EXPORT_SYMBOL(os_set_exec_close); +EXPORT_SYMBOL(os_getpid); +EXPORT_SYMBOL(os_open_file); +EXPORT_SYMBOL(os_read_file); +EXPORT_SYMBOL(os_write_file); +EXPORT_SYMBOL(os_seek_file); +EXPORT_SYMBOL(os_lock_file); +EXPORT_SYMBOL(os_ioctl_generic); +EXPORT_SYMBOL(os_pipe); +EXPORT_SYMBOL(os_file_type); +EXPORT_SYMBOL(os_file_mode); +EXPORT_SYMBOL(os_file_size); +EXPORT_SYMBOL(os_flush_stdout); +EXPORT_SYMBOL(os_close_file); +EXPORT_SYMBOL(os_set_fd_async); +EXPORT_SYMBOL(os_set_fd_block); +EXPORT_SYMBOL(helper_wait); +EXPORT_SYMBOL(os_shutdown_socket); +EXPORT_SYMBOL(os_create_unix_socket); +EXPORT_SYMBOL(os_connect_socket); +EXPORT_SYMBOL(os_accept_connection); +EXPORT_SYMBOL(os_rcv_fd); +EXPORT_SYMBOL(run_helper); +EXPORT_SYMBOL(start_thread); +EXPORT_SYMBOL(dump_thread); + +EXPORT_SYMBOL(do_gettimeofday); +EXPORT_SYMBOL(do_settimeofday); + +/* This is here because UML expands open to sys_open, not to a system + * call instruction. + */ +EXPORT_SYMBOL(sys_open); +EXPORT_SYMBOL(sys_lseek); +EXPORT_SYMBOL(sys_read); +EXPORT_SYMBOL(sys_wait4); + +#ifdef CONFIG_SMP + +/* required for SMP */ + +extern void FASTCALL( __write_lock_failed(rwlock_t *rw)); +EXPORT_SYMBOL(__write_lock_failed); + +extern void FASTCALL( __read_lock_failed(rwlock_t *rw)); +EXPORT_SYMBOL(__read_lock_failed); + +#endif + +#ifdef CONFIG_HIGHMEM +EXPORT_SYMBOL(kmap); +EXPORT_SYMBOL(kunmap); +EXPORT_SYMBOL(kmap_atomic); +EXPORT_SYMBOL(kunmap_atomic); +EXPORT_SYMBOL(kmap_atomic_to_page); +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/main.c b/arch/um/kernel/main.c new file mode 100644 index 00000000000..a17c49703f9 --- /dev/null +++ b/arch/um/kernel/main.c @@ -0,0 +1,271 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <signal.h> +#include <errno.h> +#include <sys/resource.h> +#include <sys/mman.h> +#include <sys/user.h> +#include <asm/page.h> +#include "user_util.h" +#include "kern_util.h" +#include "mem_user.h" +#include "signal_user.h" +#include "time_user.h" +#include "irq_user.h" +#include "user.h" +#include "init.h" +#include "mode.h" +#include "choose-mode.h" +#include "uml-config.h" +#include "irq_user.h" +#include "time_user.h" +#include "os.h" + +/* Set in set_stklim, which is called from main and __wrap_malloc. + * __wrap_malloc only calls it if main hasn't started. + */ +unsigned long stacksizelim; + +/* Set in main */ +char *linux_prog; + +#define PGD_BOUND (4 * 1024 * 1024) +#define STACKSIZE (8 * 1024 * 1024) +#define THREAD_NAME_LEN (256) + +static void set_stklim(void) +{ + struct rlimit lim; + + if(getrlimit(RLIMIT_STACK, &lim) < 0){ + perror("getrlimit"); + exit(1); + } + if((lim.rlim_cur == RLIM_INFINITY) || (lim.rlim_cur > STACKSIZE)){ + lim.rlim_cur = STACKSIZE; + if(setrlimit(RLIMIT_STACK, &lim) < 0){ + perror("setrlimit"); + exit(1); + } + } + stacksizelim = (lim.rlim_cur + PGD_BOUND - 1) & ~(PGD_BOUND - 1); +} + +static __init void do_uml_initcalls(void) +{ + initcall_t *call; + + call = &__uml_initcall_start; + while (call < &__uml_initcall_end){; + (*call)(); + call++; + } +} + +static void last_ditch_exit(int sig) +{ + CHOOSE_MODE(kmalloc_ok = 0, (void) 0); + signal(SIGINT, SIG_DFL); + signal(SIGTERM, SIG_DFL); + signal(SIGHUP, SIG_DFL); + uml_cleanup(); + exit(1); +} + +extern int uml_exitcode; + +extern void scan_elf_aux( char **envp); + +int main(int argc, char **argv, char **envp) +{ + char **new_argv; + sigset_t mask; + int ret, i; + + /* Enable all signals except SIGIO - in some environments, we can + * enter with some signals blocked + */ + + sigemptyset(&mask); + sigaddset(&mask, SIGIO); + if(sigprocmask(SIG_SETMASK, &mask, NULL) < 0){ + perror("sigprocmask"); + exit(1); + } + +#ifdef UML_CONFIG_MODE_TT + /* Allocate memory for thread command lines */ + if(argc < 2 || strlen(argv[1]) < THREAD_NAME_LEN - 1){ + + char padding[THREAD_NAME_LEN] = { + [ 0 ... THREAD_NAME_LEN - 2] = ' ', '\0' + }; + + new_argv = malloc((argc + 2) * sizeof(char*)); + if(!new_argv) { + perror("Allocating extended argv"); + exit(1); + } + + new_argv[0] = argv[0]; + new_argv[1] = padding; + + for(i = 2; i <= argc; i++) + new_argv[i] = argv[i - 1]; + new_argv[argc + 1] = NULL; + + execvp(new_argv[0], new_argv); + perror("execing with extended args"); + exit(1); + } +#endif + + linux_prog = argv[0]; + + set_stklim(); + + new_argv = malloc((argc + 1) * sizeof(char *)); + if(new_argv == NULL){ + perror("Mallocing argv"); + exit(1); + } + for(i=0;i<argc;i++){ + new_argv[i] = strdup(argv[i]); + if(new_argv[i] == NULL){ + perror("Mallocing an arg"); + exit(1); + } + } + new_argv[argc] = NULL; + + set_handler(SIGINT, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1); + set_handler(SIGTERM, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1); + set_handler(SIGHUP, last_ditch_exit, SA_ONESHOT | SA_NODEFER, -1); + + scan_elf_aux( envp); + + do_uml_initcalls(); + ret = linux_main(argc, argv); + + /* Disable SIGPROF - I have no idea why libc doesn't do this or turn + * off the profiling time, but UML dies with a SIGPROF just before + * exiting when profiling is active. + */ + change_sig(SIGPROF, 0); + + /* Reboot */ + if(ret){ + int err; + + printf("\n"); + + /* stop timers and set SIG*ALRM to be ignored */ + disable_timer(); + + /* disable SIGIO for the fds and set SIGIO to be ignored */ + err = deactivate_all_fds(); + if(err) + printf("deactivate_all_fds failed, errno = %d\n", + -err); + + /* Let any pending signals fire now. This ensures + * that they won't be delivered after the exec, when + * they are definitely not expected. + */ + unblock_signals(); + + execvp(new_argv[0], new_argv); + perror("Failed to exec kernel"); + ret = 1; + } + printf("\n"); + return(uml_exitcode); +} + +#define CAN_KMALLOC() \ + (kmalloc_ok && CHOOSE_MODE((os_getpid() != tracing_pid), 1)) + +extern void *__real_malloc(int); + +void *__wrap_malloc(int size) +{ + void *ret; + + if(!CAN_KMALLOC()) + return(__real_malloc(size)); + else if(size <= PAGE_SIZE) /* finding contiguos pages can be hard*/ + ret = um_kmalloc(size); + else ret = um_vmalloc(size); + + /* glibc people insist that if malloc fails, errno should be + * set by malloc as well. So we do. + */ + if(ret == NULL) + errno = ENOMEM; + + return(ret); +} + +void *__wrap_calloc(int n, int size) +{ + void *ptr = __wrap_malloc(n * size); + + if(ptr == NULL) return(NULL); + memset(ptr, 0, n * size); + return(ptr); +} + +extern void __real_free(void *); + +extern unsigned long high_physmem; + +void __wrap_free(void *ptr) +{ + unsigned long addr = (unsigned long) ptr; + + /* We need to know how the allocation happened, so it can be correctly + * freed. This is done by seeing what region of memory the pointer is + * in - + * physical memory - kmalloc/kfree + * kernel virtual memory - vmalloc/vfree + * anywhere else - malloc/free + * If kmalloc is not yet possible, then either high_physmem and/or + * end_vm are still 0 (as at startup), in which case we call free, or + * we have set them, but anyway addr has not been allocated from those + * areas. So, in both cases __real_free is called. + * + * CAN_KMALLOC is checked because it would be bad to free a buffer + * with kmalloc/vmalloc after they have been turned off during + * shutdown. + * XXX: However, we sometimes shutdown CAN_KMALLOC temporarily, so + * there is a possibility for memory leaks. + */ + + if((addr >= uml_physmem) && (addr < high_physmem)){ + if(CAN_KMALLOC()) + kfree(ptr); + } + else if((addr >= start_vm) && (addr < end_vm)){ + if(CAN_KMALLOC()) + vfree(ptr); + } + else __real_free(ptr); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c new file mode 100644 index 00000000000..f156661781c --- /dev/null +++ b/arch/um/kernel/mem.c @@ -0,0 +1,359 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/kernel.h" +#include "linux/mm.h" +#include "linux/bootmem.h" +#include "linux/swap.h" +#include "linux/highmem.h" +#include "linux/gfp.h" +#include "asm/page.h" +#include "asm/fixmap.h" +#include "asm/pgalloc.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "mem_user.h" +#include "uml_uaccess.h" +#include "os.h" + +extern char __binary_start; + +/* Changed during early boot */ +unsigned long *empty_zero_page = NULL; +unsigned long *empty_bad_page = NULL; +pgd_t swapper_pg_dir[PTRS_PER_PGD]; +unsigned long highmem; +int kmalloc_ok = 0; + +static unsigned long brk_end; + +void unmap_physmem(void) +{ + os_unmap_memory((void *) brk_end, uml_reserved - brk_end); +} + +static void map_cb(void *unused) +{ + map_memory(brk_end, __pa(brk_end), uml_reserved - brk_end, 1, 1, 0); +} + +#ifdef CONFIG_HIGHMEM +static void setup_highmem(unsigned long highmem_start, + unsigned long highmem_len) +{ + struct page *page; + unsigned long highmem_pfn; + int i; + + highmem_pfn = __pa(highmem_start) >> PAGE_SHIFT; + for(i = 0; i < highmem_len >> PAGE_SHIFT; i++){ + page = &mem_map[highmem_pfn + i]; + ClearPageReserved(page); + set_bit(PG_highmem, &page->flags); + set_page_count(page, 1); + __free_page(page); + } +} +#endif + +void mem_init(void) +{ + unsigned long start; + + max_low_pfn = (high_physmem - uml_physmem) >> PAGE_SHIFT; + + /* clear the zero-page */ + memset((void *) empty_zero_page, 0, PAGE_SIZE); + + /* Map in the area just after the brk now that kmalloc is about + * to be turned on. + */ + brk_end = (unsigned long) UML_ROUND_UP(sbrk(0)); + map_cb(NULL); + initial_thread_cb(map_cb, NULL); + free_bootmem(__pa(brk_end), uml_reserved - brk_end); + uml_reserved = brk_end; + + /* Fill in any hole at the start of the binary */ + start = (unsigned long) &__binary_start & PAGE_MASK; + if(uml_physmem != start){ + map_memory(uml_physmem, __pa(uml_physmem), start - uml_physmem, + 1, 1, 0); + } + + /* this will put all low memory onto the freelists */ + totalram_pages = free_all_bootmem(); + totalhigh_pages = highmem >> PAGE_SHIFT; + totalram_pages += totalhigh_pages; + num_physpages = totalram_pages; + max_pfn = totalram_pages; + printk(KERN_INFO "Memory: %luk available\n", + (unsigned long) nr_free_pages() << (PAGE_SHIFT-10)); + kmalloc_ok = 1; + +#ifdef CONFIG_HIGHMEM + setup_highmem(end_iomem, highmem); +#endif +} + +static void __init fixrange_init(unsigned long start, unsigned long end, + pgd_t *pgd_base) +{ + pgd_t *pgd; + pmd_t *pmd; + pte_t *pte; + int i, j; + unsigned long vaddr; + + vaddr = start; + i = pgd_index(vaddr); + j = pmd_index(vaddr); + pgd = pgd_base + i; + + for ( ; (i < PTRS_PER_PGD) && (vaddr < end); pgd++, i++) { + pmd = (pmd_t *)pgd; + for (; (j < PTRS_PER_PMD) && (vaddr != end); pmd++, j++) { + if (pmd_none(*pmd)) { + pte = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE); + set_pmd(pmd, __pmd(_KERNPG_TABLE + + (unsigned long) __pa(pte))); + if (pte != pte_offset_kernel(pmd, 0)) + BUG(); + } + vaddr += PMD_SIZE; + } + j = 0; + } +} + +#ifdef CONFIG_HIGHMEM +pte_t *kmap_pte; +pgprot_t kmap_prot; + +#define kmap_get_fixmap_pte(vaddr) \ + pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), (vaddr)),\ + (vaddr)), (vaddr)) + +static void __init kmap_init(void) +{ + unsigned long kmap_vstart; + + /* cache the first kmap pte */ + kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN); + kmap_pte = kmap_get_fixmap_pte(kmap_vstart); + + kmap_prot = PAGE_KERNEL; +} + +static void init_highmem(void) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long vaddr; + + /* + * Permanent kmaps: + */ + vaddr = PKMAP_BASE; + fixrange_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, swapper_pg_dir); + + pgd = swapper_pg_dir + pgd_index(vaddr); + pud = pud_offset(pgd, vaddr); + pmd = pmd_offset(pud, vaddr); + pte = pte_offset_kernel(pmd, vaddr); + pkmap_page_table = pte; + + kmap_init(); +} +#endif /* CONFIG_HIGHMEM */ + +static void __init fixaddr_user_init( void) +{ +#if CONFIG_ARCH_REUSE_HOST_VSYSCALL_AREA + long size = FIXADDR_USER_END - FIXADDR_USER_START; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long paddr, vaddr = FIXADDR_USER_START; + + if ( ! size ) + return; + + fixrange_init( FIXADDR_USER_START, FIXADDR_USER_END, swapper_pg_dir); + paddr = (unsigned long)alloc_bootmem_low_pages( size); + memcpy( (void *)paddr, (void *)FIXADDR_USER_START, size); + paddr = __pa(paddr); + for ( ; size > 0; size-=PAGE_SIZE, vaddr+=PAGE_SIZE, paddr+=PAGE_SIZE){ + pgd = swapper_pg_dir + pgd_index(vaddr); + pud = pud_offset(pgd, vaddr); + pmd = pmd_offset(pud, vaddr); + pte = pte_offset_kernel(pmd, vaddr); + pte_set_val( (*pte), paddr, PAGE_READONLY); + } +#endif +} + +void paging_init(void) +{ + unsigned long zones_size[MAX_NR_ZONES], vaddr; + int i; + + empty_zero_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE); + empty_bad_page = (unsigned long *) alloc_bootmem_low_pages(PAGE_SIZE); + for(i=0;i<sizeof(zones_size)/sizeof(zones_size[0]);i++) + zones_size[i] = 0; + zones_size[0] = (end_iomem >> PAGE_SHIFT) - (uml_physmem >> PAGE_SHIFT); + zones_size[2] = highmem >> PAGE_SHIFT; + free_area_init(zones_size); + + /* + * Fixed mappings, only the page table structure has to be + * created - mappings will be set by set_fixmap(): + */ + vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK; + fixrange_init(vaddr, FIXADDR_TOP, swapper_pg_dir); + + fixaddr_user_init(); + +#ifdef CONFIG_HIGHMEM + init_highmem(); +#endif +} + +struct page *arch_validate(struct page *page, int mask, int order) +{ + unsigned long addr, zero = 0; + int i; + + again: + if(page == NULL) return(page); + if(PageHighMem(page)) return(page); + + addr = (unsigned long) page_address(page); + for(i = 0; i < (1 << order); i++){ + current->thread.fault_addr = (void *) addr; + if(__do_copy_to_user((void __user *) addr, &zero, + sizeof(zero), + ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)){ + if(!(mask & __GFP_WAIT)) return(NULL); + else break; + } + addr += PAGE_SIZE; + } + + if(i == (1 << order)) return(page); + page = alloc_pages(mask, order); + goto again; +} + +/* This can't do anything because nothing in the kernel image can be freed + * since it's not in kernel physical memory. + */ + +void free_initmem(void) +{ +} + +#ifdef CONFIG_BLK_DEV_INITRD + +void free_initrd_mem(unsigned long start, unsigned long end) +{ + if (start < end) + printk ("Freeing initrd memory: %ldk freed\n", + (end - start) >> 10); + for (; start < end; start += PAGE_SIZE) { + ClearPageReserved(virt_to_page(start)); + set_page_count(virt_to_page(start), 1); + free_page(start); + totalram_pages++; + } +} + +#endif + +void show_mem(void) +{ + int pfn, total = 0, reserved = 0; + int shared = 0, cached = 0; + int highmem = 0; + struct page *page; + + printk("Mem-info:\n"); + show_free_areas(); + printk("Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10)); + pfn = max_mapnr; + while(pfn-- > 0) { + page = pfn_to_page(pfn); + total++; + if(PageHighMem(page)) + highmem++; + if(PageReserved(page)) + reserved++; + else if(PageSwapCache(page)) + cached++; + else if(page_count(page)) + shared += page_count(page) - 1; + } + printk("%d pages of RAM\n", total); + printk("%d pages of HIGHMEM\n", highmem); + printk("%d reserved pages\n", reserved); + printk("%d pages shared\n", shared); + printk("%d pages swap cached\n", cached); +} + +/* + * Allocate and free page tables. + */ + +pgd_t *pgd_alloc(struct mm_struct *mm) +{ + pgd_t *pgd = (pgd_t *)__get_free_page(GFP_KERNEL); + + if (pgd) { + memset(pgd, 0, USER_PTRS_PER_PGD * sizeof(pgd_t)); + memcpy(pgd + USER_PTRS_PER_PGD, + swapper_pg_dir + USER_PTRS_PER_PGD, + (PTRS_PER_PGD - USER_PTRS_PER_PGD) * sizeof(pgd_t)); + } + return pgd; +} + +void pgd_free(pgd_t *pgd) +{ + free_page((unsigned long) pgd); +} + +pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) +{ + pte_t *pte; + + pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO); + return pte; +} + +struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address) +{ + struct page *pte; + + pte = alloc_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO); + return pte; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/mem_user.c b/arch/um/kernel/mem_user.c new file mode 100644 index 00000000000..4a663fd434b --- /dev/null +++ b/arch/um/kernel/mem_user.c @@ -0,0 +1,273 @@ +/* + * arch/um/kernel/mem_user.c + * + * BRIEF MODULE DESCRIPTION + * user side memory routines for supporting IO memory inside user mode linux + * + * Copyright (C) 2001 RidgeRun, Inc. + * Author: RidgeRun, Inc. + * Greg Lonnon glonnon@ridgerun.com or info@ridgerun.com + * + * This program is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License as published by the + * Free Software Foundation; either version 2 of the License, or (at your + * option) any later version. + * + * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESS OR IMPLIED + * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF + * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN + * NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, + * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT + * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF + * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON + * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF + * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + * + * You should have received a copy of the GNU General Public License along + * with this program; if not, write to the Free Software Foundation, Inc., + * 675 Mass Ave, Cambridge, MA 02139, USA. + */ + +#include <stdio.h> +#include <stdlib.h> +#include <stddef.h> +#include <stdarg.h> +#include <unistd.h> +#include <errno.h> +#include <string.h> +#include <fcntl.h> +#include <sys/types.h> +#include <sys/mman.h> +#include "kern_util.h" +#include "user.h" +#include "user_util.h" +#include "mem_user.h" +#include "init.h" +#include "os.h" +#include "tempfile.h" +#include "kern_constants.h" + +#define TEMPNAME_TEMPLATE "vm_file-XXXXXX" + +static int create_tmp_file(unsigned long len) +{ + int fd, err; + char zero; + + fd = make_tempfile(TEMPNAME_TEMPLATE, NULL, 1); + if(fd < 0) { + os_print_error(fd, "make_tempfile"); + exit(1); + } + + err = os_mode_fd(fd, 0777); + if(err < 0){ + os_print_error(err, "os_mode_fd"); + exit(1); + } + err = os_seek_file(fd, len); + if(err < 0){ + os_print_error(err, "os_seek_file"); + exit(1); + } + zero = 0; + err = os_write_file(fd, &zero, 1); + if(err != 1){ + os_print_error(err, "os_write_file"); + exit(1); + } + + return(fd); +} + +void check_tmpexec(void) +{ + void *addr; + int err, fd = create_tmp_file(UM_KERN_PAGE_SIZE); + + addr = mmap(NULL, UM_KERN_PAGE_SIZE, + PROT_READ | PROT_WRITE | PROT_EXEC, MAP_PRIVATE, fd, 0); + printf("Checking PROT_EXEC mmap in /tmp..."); + fflush(stdout); + if(addr == MAP_FAILED){ + err = errno; + perror("failed"); + if(err == EPERM) + printf("/tmp must be not mounted noexec\n"); + exit(1); + } + printf("OK\n"); + munmap(addr, UM_KERN_PAGE_SIZE); + + os_close_file(fd); +} + +static int have_devanon = 0; + +void check_devanon(void) +{ + int fd; + + printk("Checking for /dev/anon on the host..."); + fd = open("/dev/anon", O_RDWR); + if(fd < 0){ + printk("Not available (open failed with errno %d)\n", errno); + return; + } + + printk("OK\n"); + have_devanon = 1; +} + +static int create_anon_file(unsigned long len) +{ + void *addr; + int fd; + + fd = open("/dev/anon", O_RDWR); + if(fd < 0) { + os_print_error(fd, "opening /dev/anon"); + exit(1); + } + + addr = mmap(NULL, len, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0); + if(addr == MAP_FAILED){ + perror("mapping physmem file"); + exit(1); + } + munmap(addr, len); + + return(fd); +} + +int create_mem_file(unsigned long len) +{ + int err, fd; + + if(have_devanon) + fd = create_anon_file(len); + else fd = create_tmp_file(len); + + err = os_set_exec_close(fd, 1); + if(err < 0) + os_print_error(err, "exec_close"); + return(fd); +} + +struct iomem_region *iomem_regions = NULL; +int iomem_size = 0; + +static int __init parse_iomem(char *str, int *add) +{ + struct iomem_region *new; + struct uml_stat buf; + char *file, *driver; + int fd, err, size; + + driver = str; + file = strchr(str,','); + if(file == NULL){ + printf("parse_iomem : failed to parse iomem\n"); + goto out; + } + *file = '\0'; + file++; + fd = os_open_file(file, of_rdwr(OPENFLAGS()), 0); + if(fd < 0){ + os_print_error(fd, "parse_iomem - Couldn't open io file"); + goto out; + } + + err = os_stat_fd(fd, &buf); + if(err < 0){ + os_print_error(err, "parse_iomem - cannot stat_fd file"); + goto out_close; + } + + new = malloc(sizeof(*new)); + if(new == NULL){ + perror("Couldn't allocate iomem_region struct"); + goto out_close; + } + + size = (buf.ust_size + UM_KERN_PAGE_SIZE) & ~(UM_KERN_PAGE_SIZE - 1); + + *new = ((struct iomem_region) { .next = iomem_regions, + .driver = driver, + .fd = fd, + .size = size, + .phys = 0, + .virt = 0 }); + iomem_regions = new; + iomem_size += new->size + UM_KERN_PAGE_SIZE; + + return(0); + out_close: + os_close_file(fd); + out: + return(1); +} + +__uml_setup("iomem=", parse_iomem, +"iomem=<name>,<file>\n" +" Configure <file> as an IO memory region named <name>.\n\n" +); + +int protect_memory(unsigned long addr, unsigned long len, int r, int w, int x, + int must_succeed) +{ + int err; + + err = os_protect_memory((void *) addr, len, r, w, x); + if(err < 0){ + if(must_succeed) + panic("protect failed, err = %d", -err); + else return(err); + } + return(0); +} + +#if 0 +/* Debugging facility for dumping stuff out to the host, avoiding the timing + * problems that come with printf and breakpoints. + * Enable in case of emergency. + */ + +int logging = 1; +int logging_fd = -1; + +int logging_line = 0; +char logging_buf[512]; + +void log(char *fmt, ...) +{ + va_list ap; + struct timeval tv; + struct openflags flags; + + if(logging == 0) return; + if(logging_fd < 0){ + flags = of_create(of_trunc(of_rdwr(OPENFLAGS()))); + logging_fd = os_open_file("log", flags, 0644); + } + gettimeofday(&tv, NULL); + sprintf(logging_buf, "%d\t %u.%u ", logging_line++, tv.tv_sec, + tv.tv_usec); + va_start(ap, fmt); + vsprintf(&logging_buf[strlen(logging_buf)], fmt, ap); + va_end(ap); + write(logging_fd, logging_buf, strlen(logging_buf)); +} +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c new file mode 100644 index 00000000000..420e6d51fa0 --- /dev/null +++ b/arch/um/kernel/physmem.c @@ -0,0 +1,478 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/mm.h" +#include "linux/rbtree.h" +#include "linux/slab.h" +#include "linux/vmalloc.h" +#include "linux/bootmem.h" +#include "linux/module.h" +#include "asm/types.h" +#include "asm/pgtable.h" +#include "kern_util.h" +#include "user_util.h" +#include "mode_kern.h" +#include "mem.h" +#include "mem_user.h" +#include "os.h" +#include "kern.h" +#include "init.h" + +struct phys_desc { + struct rb_node rb; + int fd; + __u64 offset; + void *virt; + unsigned long phys; + struct list_head list; +}; + +static struct rb_root phys_mappings = RB_ROOT; + +static struct rb_node **find_rb(void *virt) +{ + struct rb_node **n = &phys_mappings.rb_node; + struct phys_desc *d; + + while(*n != NULL){ + d = rb_entry(*n, struct phys_desc, rb); + if(d->virt == virt) + return(n); + + if(d->virt > virt) + n = &(*n)->rb_left; + else + n = &(*n)->rb_right; + } + + return(n); +} + +static struct phys_desc *find_phys_mapping(void *virt) +{ + struct rb_node **n = find_rb(virt); + + if(*n == NULL) + return(NULL); + + return(rb_entry(*n, struct phys_desc, rb)); +} + +static void insert_phys_mapping(struct phys_desc *desc) +{ + struct rb_node **n = find_rb(desc->virt); + + if(*n != NULL) + panic("Physical remapping for %p already present", + desc->virt); + + rb_link_node(&desc->rb, (*n)->rb_parent, n); + rb_insert_color(&desc->rb, &phys_mappings); +} + +LIST_HEAD(descriptor_mappings); + +struct desc_mapping { + int fd; + struct list_head list; + struct list_head pages; +}; + +static struct desc_mapping *find_mapping(int fd) +{ + struct desc_mapping *desc; + struct list_head *ele; + + list_for_each(ele, &descriptor_mappings){ + desc = list_entry(ele, struct desc_mapping, list); + if(desc->fd == fd) + return(desc); + } + + return(NULL); +} + +static struct desc_mapping *descriptor_mapping(int fd) +{ + struct desc_mapping *desc; + + desc = find_mapping(fd); + if(desc != NULL) + return(desc); + + desc = kmalloc(sizeof(*desc), GFP_ATOMIC); + if(desc == NULL) + return(NULL); + + *desc = ((struct desc_mapping) + { .fd = fd, + .list = LIST_HEAD_INIT(desc->list), + .pages = LIST_HEAD_INIT(desc->pages) }); + list_add(&desc->list, &descriptor_mappings); + + return(desc); +} + +int physmem_subst_mapping(void *virt, int fd, __u64 offset, int w) +{ + struct desc_mapping *fd_maps; + struct phys_desc *desc; + unsigned long phys; + int err; + + fd_maps = descriptor_mapping(fd); + if(fd_maps == NULL) + return(-ENOMEM); + + phys = __pa(virt); + desc = find_phys_mapping(virt); + if(desc != NULL) + panic("Address 0x%p is already substituted\n", virt); + + err = -ENOMEM; + desc = kmalloc(sizeof(*desc), GFP_ATOMIC); + if(desc == NULL) + goto out; + + *desc = ((struct phys_desc) + { .fd = fd, + .offset = offset, + .virt = virt, + .phys = __pa(virt), + .list = LIST_HEAD_INIT(desc->list) }); + insert_phys_mapping(desc); + + list_add(&desc->list, &fd_maps->pages); + + virt = (void *) ((unsigned long) virt & PAGE_MASK); + err = os_map_memory(virt, fd, offset, PAGE_SIZE, 1, w, 0); + if(!err) + goto out; + + rb_erase(&desc->rb, &phys_mappings); + kfree(desc); + out: + return(err); +} + +static int physmem_fd = -1; + +static void remove_mapping(struct phys_desc *desc) +{ + void *virt = desc->virt; + int err; + + rb_erase(&desc->rb, &phys_mappings); + list_del(&desc->list); + kfree(desc); + + err = os_map_memory(virt, physmem_fd, __pa(virt), PAGE_SIZE, 1, 1, 0); + if(err) + panic("Failed to unmap block device page from physical memory, " + "errno = %d", -err); +} + +int physmem_remove_mapping(void *virt) +{ + struct phys_desc *desc; + + virt = (void *) ((unsigned long) virt & PAGE_MASK); + desc = find_phys_mapping(virt); + if(desc == NULL) + return(0); + + remove_mapping(desc); + return(1); +} + +void physmem_forget_descriptor(int fd) +{ + struct desc_mapping *desc; + struct phys_desc *page; + struct list_head *ele, *next; + __u64 offset; + void *addr; + int err; + + desc = find_mapping(fd); + if(desc == NULL) + return; + + list_for_each_safe(ele, next, &desc->pages){ + page = list_entry(ele, struct phys_desc, list); + offset = page->offset; + addr = page->virt; + remove_mapping(page); + err = os_seek_file(fd, offset); + if(err) + panic("physmem_forget_descriptor - failed to seek " + "to %lld in fd %d, error = %d\n", + offset, fd, -err); + err = os_read_file(fd, addr, PAGE_SIZE); + if(err < 0) + panic("physmem_forget_descriptor - failed to read " + "from fd %d to 0x%p, error = %d\n", + fd, addr, -err); + } + + list_del(&desc->list); + kfree(desc); +} + +EXPORT_SYMBOL(physmem_forget_descriptor); +EXPORT_SYMBOL(physmem_remove_mapping); +EXPORT_SYMBOL(physmem_subst_mapping); + +void arch_free_page(struct page *page, int order) +{ + void *virt; + int i; + + for(i = 0; i < (1 << order); i++){ + virt = __va(page_to_phys(page + i)); + physmem_remove_mapping(virt); + } +} + +int is_remapped(void *virt) +{ + struct phys_desc *desc = find_phys_mapping(virt); + + return(desc != NULL); +} + +/* Changed during early boot */ +unsigned long high_physmem; + +extern unsigned long physmem_size; + +void *to_virt(unsigned long phys) +{ + return((void *) uml_physmem + phys); +} + +unsigned long to_phys(void *virt) +{ + return(((unsigned long) virt) - uml_physmem); +} + +int init_maps(unsigned long physmem, unsigned long iomem, unsigned long highmem) +{ + struct page *p, *map; + unsigned long phys_len, phys_pages, highmem_len, highmem_pages; + unsigned long iomem_len, iomem_pages, total_len, total_pages; + int i; + + phys_pages = physmem >> PAGE_SHIFT; + phys_len = phys_pages * sizeof(struct page); + + iomem_pages = iomem >> PAGE_SHIFT; + iomem_len = iomem_pages * sizeof(struct page); + + highmem_pages = highmem >> PAGE_SHIFT; + highmem_len = highmem_pages * sizeof(struct page); + + total_pages = phys_pages + iomem_pages + highmem_pages; + total_len = phys_len + iomem_pages + highmem_len; + + if(kmalloc_ok){ + map = kmalloc(total_len, GFP_KERNEL); + if(map == NULL) + map = vmalloc(total_len); + } + else map = alloc_bootmem_low_pages(total_len); + + if(map == NULL) + return(-ENOMEM); + + for(i = 0; i < total_pages; i++){ + p = &map[i]; + set_page_count(p, 0); + SetPageReserved(p); + INIT_LIST_HEAD(&p->lru); + } + + max_mapnr = total_pages; + return(0); +} + +struct page *phys_to_page(const unsigned long phys) +{ + return(&mem_map[phys >> PAGE_SHIFT]); +} + +struct page *__virt_to_page(const unsigned long virt) +{ + return(&mem_map[__pa(virt) >> PAGE_SHIFT]); +} + +phys_t page_to_phys(struct page *page) +{ + return((page - mem_map) << PAGE_SHIFT); +} + +pte_t mk_pte(struct page *page, pgprot_t pgprot) +{ + pte_t pte; + + pte_set_val(pte, page_to_phys(page), pgprot); + if(pte_present(pte)) + pte_mknewprot(pte_mknewpage(pte)); + return(pte); +} + +/* Changed during early boot */ +static unsigned long kmem_top = 0; + +unsigned long get_kmem_end(void) +{ + if(kmem_top == 0) + kmem_top = CHOOSE_MODE(kmem_end_tt, kmem_end_skas); + return(kmem_top); +} + +void map_memory(unsigned long virt, unsigned long phys, unsigned long len, + int r, int w, int x) +{ + __u64 offset; + int fd, err; + + fd = phys_mapping(phys, &offset); + err = os_map_memory((void *) virt, fd, offset, len, r, w, x); + if(err) { + if(err == -ENOMEM) + printk("try increasing the host's " + "/proc/sys/vm/max_map_count to <physical " + "memory size>/4096\n"); + panic("map_memory(0x%lx, %d, 0x%llx, %ld, %d, %d, %d) failed, " + "err = %d\n", virt, fd, offset, len, r, w, x, err); + } +} + +#define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT) + +void setup_physmem(unsigned long start, unsigned long reserve_end, + unsigned long len, unsigned long highmem) +{ + unsigned long reserve = reserve_end - start; + int pfn = PFN_UP(__pa(reserve_end)); + int delta = (len - reserve) >> PAGE_SHIFT; + int err, offset, bootmap_size; + + physmem_fd = create_mem_file(len + highmem); + + offset = uml_reserved - uml_physmem; + err = os_map_memory((void *) uml_reserved, physmem_fd, offset, + len - offset, 1, 1, 0); + if(err < 0){ + os_print_error(err, "Mapping memory"); + exit(1); + } + + bootmap_size = init_bootmem(pfn, pfn + delta); + free_bootmem(__pa(reserve_end) + bootmap_size, + len - bootmap_size - reserve); +} + +int phys_mapping(unsigned long phys, __u64 *offset_out) +{ + struct phys_desc *desc = find_phys_mapping(__va(phys & PAGE_MASK)); + int fd = -1; + + if(desc != NULL){ + fd = desc->fd; + *offset_out = desc->offset; + } + else if(phys < physmem_size){ + fd = physmem_fd; + *offset_out = phys; + } + else if(phys < __pa(end_iomem)){ + struct iomem_region *region = iomem_regions; + + while(region != NULL){ + if((phys >= region->phys) && + (phys < region->phys + region->size)){ + fd = region->fd; + *offset_out = phys - region->phys; + break; + } + region = region->next; + } + } + else if(phys < __pa(end_iomem) + highmem){ + fd = physmem_fd; + *offset_out = phys - iomem_size; + } + + return(fd); +} + +static int __init uml_mem_setup(char *line, int *add) +{ + char *retptr; + physmem_size = memparse(line,&retptr); + return 0; +} +__uml_setup("mem=", uml_mem_setup, +"mem=<Amount of desired ram>\n" +" This controls how much \"physical\" memory the kernel allocates\n" +" for the system. The size is specified as a number followed by\n" +" one of 'k', 'K', 'm', 'M', which have the obvious meanings.\n" +" This is not related to the amount of memory in the host. It can\n" +" be more, and the excess, if it's ever used, will just be swapped out.\n" +" Example: mem=64M\n\n" +); + +unsigned long find_iomem(char *driver, unsigned long *len_out) +{ + struct iomem_region *region = iomem_regions; + + while(region != NULL){ + if(!strcmp(region->driver, driver)){ + *len_out = region->size; + return(region->virt); + } + } + + return(0); +} + +int setup_iomem(void) +{ + struct iomem_region *region = iomem_regions; + unsigned long iomem_start = high_physmem + PAGE_SIZE; + int err; + + while(region != NULL){ + err = os_map_memory((void *) iomem_start, region->fd, 0, + region->size, 1, 1, 0); + if(err) + printk("Mapping iomem region for driver '%s' failed, " + "errno = %d\n", region->driver, -err); + else { + region->virt = iomem_start; + region->phys = __pa(region->virt); + } + + iomem_start += region->size + PAGE_SIZE; + region = region->next; + } + + return(0); +} + +__initcall(setup_iomem); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c new file mode 100644 index 00000000000..f76a2692adc --- /dev/null +++ b/arch/um/kernel/process.c @@ -0,0 +1,423 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <signal.h> +#include <sched.h> +#include <errno.h> +#include <stdarg.h> +#include <stdlib.h> +#include <setjmp.h> +#include <sys/time.h> +#include <sys/wait.h> +#include <sys/mman.h> +#include <asm/unistd.h> +#include <asm/page.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "process.h" +#include "signal_kern.h" +#include "signal_user.h" +#include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" +#include "irq_user.h" +#include "ptrace_user.h" +#include "time_user.h" +#include "init.h" +#include "os.h" +#include "uml-config.h" +#include "ptrace_user.h" +#include "choose-mode.h" +#include "mode.h" +#ifdef UML_CONFIG_MODE_SKAS +#include "skas.h" +#include "skas_ptrace.h" +#include "registers.h" +#endif + +void init_new_thread_stack(void *sig_stack, void (*usr1_handler)(int)) +{ + int flags = 0, pages; + + if(sig_stack != NULL){ + pages = (1 << UML_CONFIG_KERNEL_STACK_ORDER); + set_sigstack(sig_stack, pages * page_size()); + flags = SA_ONSTACK; + } + if(usr1_handler) set_handler(SIGUSR1, usr1_handler, flags, -1); +} + +void init_new_thread_signals(int altstack) +{ + int flags = altstack ? SA_ONSTACK : 0; + + set_handler(SIGSEGV, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGTRAP, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGFPE, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGILL, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGBUS, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGWINCH, (__sighandler_t) sig_handler, flags, + SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + set_handler(SIGUSR2, (__sighandler_t) sig_handler, + flags, SIGUSR1, SIGIO, SIGWINCH, SIGALRM, SIGVTALRM, -1); + signal(SIGHUP, SIG_IGN); + + init_irq_signals(altstack); +} + +struct tramp { + int (*tramp)(void *); + void *tramp_data; + unsigned long temp_stack; + int flags; + int pid; +}; + +/* See above for why sigkill is here */ + +int sigkill = SIGKILL; + +int outer_tramp(void *arg) +{ + struct tramp *t; + int sig = sigkill; + + t = arg; + t->pid = clone(t->tramp, (void *) t->temp_stack + page_size()/2, + t->flags, t->tramp_data); + if(t->pid > 0) wait_for_stop(t->pid, SIGSTOP, PTRACE_CONT, NULL); + kill(os_getpid(), sig); + _exit(0); +} + +int start_fork_tramp(void *thread_arg, unsigned long temp_stack, + int clone_flags, int (*tramp)(void *)) +{ + struct tramp arg; + unsigned long sp; + int new_pid, status, err; + + /* The trampoline will run on the temporary stack */ + sp = stack_sp(temp_stack); + + clone_flags |= CLONE_FILES | SIGCHLD; + + arg.tramp = tramp; + arg.tramp_data = thread_arg; + arg.temp_stack = temp_stack; + arg.flags = clone_flags; + + /* Start the process and wait for it to kill itself */ + new_pid = clone(outer_tramp, (void *) sp, clone_flags, &arg); + if(new_pid < 0) + return(new_pid); + + CATCH_EINTR(err = waitpid(new_pid, &status, 0)); + if(err < 0) + panic("Waiting for outer trampoline failed - errno = %d", + errno); + + if(!WIFSIGNALED(status) || (WTERMSIG(status) != SIGKILL)) + panic("outer trampoline didn't exit with SIGKILL, " + "status = %d", status); + + return(arg.pid); +} + +static int ptrace_child(void *arg) +{ + int ret; + int pid = os_getpid(), ppid = getppid(); + int sc_result; + + if(ptrace(PTRACE_TRACEME, 0, 0, 0) < 0){ + perror("ptrace"); + os_kill_process(pid, 0); + } + os_stop_process(pid); + + /*This syscall will be intercepted by the parent. Don't call more than + * once, please.*/ + sc_result = os_getpid(); + + if (sc_result == pid) + ret = 1; /*Nothing modified by the parent, we are running + normally.*/ + else if (sc_result == ppid) + ret = 0; /*Expected in check_ptrace and check_sysemu when they + succeed in modifying the stack frame*/ + else + ret = 2; /*Serious trouble! This could be caused by a bug in + host 2.6 SKAS3/2.6 patch before release -V6, together + with a bug in the UML code itself.*/ + _exit(ret); +} + +static int start_ptraced_child(void **stack_out) +{ + void *stack; + unsigned long sp; + int pid, n, status; + + stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if(stack == MAP_FAILED) + panic("check_ptrace : mmap failed, errno = %d", errno); + sp = (unsigned long) stack + PAGE_SIZE - sizeof(void *); + pid = clone(ptrace_child, (void *) sp, SIGCHLD, NULL); + if(pid < 0) + panic("check_ptrace : clone failed, errno = %d", errno); + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("check_ptrace : wait failed, errno = %d", errno); + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP)) + panic("check_ptrace : expected SIGSTOP, got status = %d", + status); + + *stack_out = stack; + return(pid); +} + +/* When testing for SYSEMU support, if it is one of the broken versions, we must + * just avoid using sysemu, not panic, but only if SYSEMU features are broken. + * So only for SYSEMU features we test mustpanic, while normal host features + * must work anyway!*/ +static int stop_ptraced_child(int pid, void *stack, int exitcode, int mustpanic) +{ + int status, n, ret = 0; + + if(ptrace(PTRACE_CONT, pid, 0, 0) < 0) + panic("check_ptrace : ptrace failed, errno = %d", errno); + CATCH_EINTR(n = waitpid(pid, &status, 0)); + if(!WIFEXITED(status) || (WEXITSTATUS(status) != exitcode)) { + int exit_with = WEXITSTATUS(status); + if (exit_with == 2) + printk("check_ptrace : child exited with status 2. " + "Serious trouble happening! Try updating your " + "host skas patch!\nDisabling SYSEMU support."); + printk("check_ptrace : child exited with exitcode %d, while " + "expecting %d; status 0x%x", exit_with, + exitcode, status); + if (mustpanic) + panic("\n"); + else + printk("\n"); + ret = -1; + } + + if(munmap(stack, PAGE_SIZE) < 0) + panic("check_ptrace : munmap failed, errno = %d", errno); + return ret; +} + +static int force_sysemu_disabled = 0; + +static int __init nosysemu_cmd_param(char *str, int* add) +{ + force_sysemu_disabled = 1; + return 0; +} + +__uml_setup("nosysemu", nosysemu_cmd_param, + "nosysemu\n" + " Turns off syscall emulation patch for ptrace (SYSEMU) on.\n" + " SYSEMU is a performance-patch introduced by Laurent Vivier. It changes\n" + " behaviour of ptrace() and helps reducing host context switch rate.\n" + " To make it working, you need a kernel patch for your host, too.\n" + " See http://perso.wanadoo.fr/laurent.vivier/UML/ for further information.\n\n"); + +static void __init check_sysemu(void) +{ + void *stack; + int pid, syscall, n, status, count=0; + + printk("Checking syscall emulation patch for ptrace..."); + sysemu_supported = 0; + pid = start_ptraced_child(&stack); + + if(ptrace(PTRACE_SYSEMU, pid, 0, 0) < 0) + goto fail; + + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if (n < 0) + panic("check_sysemu : wait failed, errno = %d", errno); + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGTRAP)) + panic("check_sysemu : expected SIGTRAP, " + "got status = %d", status); + + n = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, + os_getpid()); + if(n < 0) + panic("check_sysemu : failed to modify system " + "call return, errno = %d", errno); + + if (stop_ptraced_child(pid, stack, 0, 0) < 0) + goto fail_stopped; + + sysemu_supported = 1; + printk("OK\n"); + set_using_sysemu(!force_sysemu_disabled); + + printk("Checking advanced syscall emulation patch for ptrace..."); + pid = start_ptraced_child(&stack); + while(1){ + count++; + if(ptrace(PTRACE_SYSEMU_SINGLESTEP, pid, 0, 0) < 0) + goto fail; + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("check_ptrace : wait failed, errno = %d", errno); + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGTRAP)) + panic("check_ptrace : expected (SIGTRAP|SYSCALL_TRAP), " + "got status = %d", status); + + syscall = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_NR_OFFSET, + 0); + if(syscall == __NR_getpid){ + if (!count) + panic("check_ptrace : SYSEMU_SINGLESTEP doesn't singlestep"); + n = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, + os_getpid()); + if(n < 0) + panic("check_sysemu : failed to modify system " + "call return, errno = %d", errno); + break; + } + } + if (stop_ptraced_child(pid, stack, 0, 0) < 0) + goto fail_stopped; + + sysemu_supported = 2; + printk("OK\n"); + + if ( !force_sysemu_disabled ) + set_using_sysemu(sysemu_supported); + return; + +fail: + stop_ptraced_child(pid, stack, 1, 0); +fail_stopped: + printk("missing\n"); +} + +void __init check_ptrace(void) +{ + void *stack; + int pid, syscall, n, status; + + printk("Checking that ptrace can change system call numbers..."); + pid = start_ptraced_child(&stack); + + if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) + panic("check_ptrace: PTRACE_SETOPTIONS failed, errno = %d", errno); + + while(1){ + if(ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0) + panic("check_ptrace : ptrace failed, errno = %d", + errno); + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("check_ptrace : wait failed, errno = %d", errno); + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGTRAP + 0x80)) + panic("check_ptrace : expected SIGTRAP + 0x80, " + "got status = %d", status); + + syscall = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_NR_OFFSET, + 0); + if(syscall == __NR_getpid){ + n = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, + __NR_getppid); + if(n < 0) + panic("check_ptrace : failed to modify system " + "call, errno = %d", errno); + break; + } + } + stop_ptraced_child(pid, stack, 0, 1); + printk("OK\n"); + check_sysemu(); +} + +int run_kernel_thread(int (*fn)(void *), void *arg, void **jmp_ptr) +{ + sigjmp_buf buf; + int n; + + *jmp_ptr = &buf; + n = sigsetjmp(buf, 1); + if(n != 0) + return(n); + (*fn)(arg); + return(0); +} + +void forward_pending_sigio(int target) +{ + sigset_t sigs; + + if(sigpending(&sigs)) + panic("forward_pending_sigio : sigpending failed"); + if(sigismember(&sigs, SIGIO)) + kill(target, SIGIO); +} + +#ifdef UML_CONFIG_MODE_SKAS +static inline int check_skas3_ptrace_support(void) +{ + struct ptrace_faultinfo fi; + void *stack; + int pid, n, ret = 1; + + printf("Checking for the skas3 patch in the host..."); + pid = start_ptraced_child(&stack); + + n = ptrace(PTRACE_FAULTINFO, pid, 0, &fi); + if (n < 0) { + if(errno == EIO) + printf("not found\n"); + else { + perror("not found"); + } + ret = 0; + } else { + printf("found\n"); + } + + init_registers(pid); + stop_ptraced_child(pid, stack, 1, 1); + + return(ret); +} + +int can_do_skas(void) +{ + int ret = 1; + + printf("Checking for /proc/mm..."); + if (os_access("/proc/mm", OS_ACC_W_OK) < 0) { + printf("not found\n"); + ret = 0; + goto out; + } else { + printf("found\n"); + } + + ret = check_skas3_ptrace_support(); +out: + return ret; +} +#else +int can_do_skas(void) +{ + return(0); +} +#endif diff --git a/arch/um/kernel/process_kern.c b/arch/um/kernel/process_kern.c new file mode 100644 index 00000000000..1d719d5b4bb --- /dev/null +++ b/arch/um/kernel/process_kern.c @@ -0,0 +1,500 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/kernel.h" +#include "linux/sched.h" +#include "linux/interrupt.h" +#include "linux/mm.h" +#include "linux/slab.h" +#include "linux/utsname.h" +#include "linux/fs.h" +#include "linux/utime.h" +#include "linux/smp_lock.h" +#include "linux/module.h" +#include "linux/init.h" +#include "linux/capability.h" +#include "linux/vmalloc.h" +#include "linux/spinlock.h" +#include "linux/proc_fs.h" +#include "linux/ptrace.h" +#include "linux/random.h" +#include "asm/unistd.h" +#include "asm/mman.h" +#include "asm/segment.h" +#include "asm/stat.h" +#include "asm/pgtable.h" +#include "asm/processor.h" +#include "asm/tlbflush.h" +#include "asm/uaccess.h" +#include "asm/user.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "signal_kern.h" +#include "signal_user.h" +#include "init.h" +#include "irq_user.h" +#include "mem_user.h" +#include "time_user.h" +#include "tlb.h" +#include "frame_kern.h" +#include "sigcontext.h" +#include "2_5compat.h" +#include "os.h" +#include "mode.h" +#include "mode_kern.h" +#include "choose-mode.h" + +/* This is a per-cpu array. A processor only modifies its entry and it only + * cares about its entry, so it's OK if another processor is modifying its + * entry. + */ +struct cpu_task cpu_tasks[NR_CPUS] = { [0 ... NR_CPUS - 1] = { -1, NULL } }; + +struct task_struct *get_task(int pid, int require) +{ + struct task_struct *ret; + + read_lock(&tasklist_lock); + ret = find_task_by_pid(pid); + read_unlock(&tasklist_lock); + + if(require && (ret == NULL)) panic("get_task couldn't find a task\n"); + return(ret); +} + +int external_pid(void *t) +{ + struct task_struct *task = t ? t : current; + + return(CHOOSE_MODE_PROC(external_pid_tt, external_pid_skas, task)); +} + +int pid_to_processor_id(int pid) +{ + int i; + + for(i = 0; i < ncpus; i++){ + if(cpu_tasks[i].pid == pid) return(i); + } + return(-1); +} + +void free_stack(unsigned long stack, int order) +{ + free_pages(stack, order); +} + +unsigned long alloc_stack(int order, int atomic) +{ + unsigned long page; + int flags = GFP_KERNEL; + + if(atomic) flags |= GFP_ATOMIC; + page = __get_free_pages(flags, order); + if(page == 0) + return(0); + stack_protections(page); + return(page); +} + +int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags) +{ + int pid; + + current->thread.request.u.thread.proc = fn; + current->thread.request.u.thread.arg = arg; + pid = do_fork(CLONE_VM | CLONE_UNTRACED | flags, 0, NULL, 0, NULL, + NULL); + if(pid < 0) + panic("do_fork failed in kernel_thread, errno = %d", pid); + return(pid); +} + +void switch_mm(struct mm_struct *prev, struct mm_struct *next, + struct task_struct *tsk) +{ + int cpu = smp_processor_id(); + + if (prev != next) + cpu_clear(cpu, prev->cpu_vm_mask); + cpu_set(cpu, next->cpu_vm_mask); +} + +void set_current(void *t) +{ + struct task_struct *task = t; + + cpu_tasks[task->thread_info->cpu] = ((struct cpu_task) + { external_pid(task), task }); +} + +void *_switch_to(void *prev, void *next, void *last) +{ + return(CHOOSE_MODE(switch_to_tt(prev, next), + switch_to_skas(prev, next))); +} + +void interrupt_end(void) +{ + if(need_resched()) schedule(); + if(test_tsk_thread_flag(current, TIF_SIGPENDING)) do_signal(); +} + +void release_thread(struct task_struct *task) +{ + CHOOSE_MODE(release_thread_tt(task), release_thread_skas(task)); +} + +void exit_thread(void) +{ + CHOOSE_MODE(exit_thread_tt(), exit_thread_skas()); + unprotect_stack((unsigned long) current_thread); +} + +void *get_current(void) +{ + return(current); +} + +void prepare_to_copy(struct task_struct *tsk) +{ +} + +int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct * p, + struct pt_regs *regs) +{ + p->thread = (struct thread_struct) INIT_THREAD; + return(CHOOSE_MODE_PROC(copy_thread_tt, copy_thread_skas, nr, + clone_flags, sp, stack_top, p, regs)); +} + +void initial_thread_cb(void (*proc)(void *), void *arg) +{ + int save_kmalloc_ok = kmalloc_ok; + + kmalloc_ok = 0; + CHOOSE_MODE_PROC(initial_thread_cb_tt, initial_thread_cb_skas, proc, + arg); + kmalloc_ok = save_kmalloc_ok; +} + +unsigned long stack_sp(unsigned long page) +{ + return(page + PAGE_SIZE - sizeof(void *)); +} + +int current_pid(void) +{ + return(current->pid); +} + +void default_idle(void) +{ + uml_idle_timer(); + + atomic_inc(&init_mm.mm_count); + current->mm = &init_mm; + current->active_mm = &init_mm; + + while(1){ + /* endless idle loop with no priority at all */ + SET_PRI(current); + + /* + * although we are an idle CPU, we do not want to + * get into the scheduler unnecessarily. + */ + if(need_resched()) + schedule(); + + idle_sleep(10); + } +} + +void cpu_idle(void) +{ + CHOOSE_MODE(init_idle_tt(), init_idle_skas()); +} + +int page_size(void) +{ + return(PAGE_SIZE); +} + +unsigned long page_mask(void) +{ + return(PAGE_MASK); +} + +void *um_virt_to_phys(struct task_struct *task, unsigned long addr, + pte_t *pte_out) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + + if(task->mm == NULL) + return(ERR_PTR(-EINVAL)); + pgd = pgd_offset(task->mm, addr); + if(!pgd_present(*pgd)) + return(ERR_PTR(-EINVAL)); + + pud = pud_offset(pgd, addr); + if(!pud_present(*pud)) + return(ERR_PTR(-EINVAL)); + + pmd = pmd_offset(pud, addr); + if(!pmd_present(*pmd)) + return(ERR_PTR(-EINVAL)); + + pte = pte_offset_kernel(pmd, addr); + if(!pte_present(*pte)) + return(ERR_PTR(-EINVAL)); + + if(pte_out != NULL) + *pte_out = *pte; + return((void *) (pte_val(*pte) & PAGE_MASK) + (addr & ~PAGE_MASK)); +} + +char *current_cmd(void) +{ +#if defined(CONFIG_SMP) || defined(CONFIG_HIGHMEM) + return("(Unknown)"); +#else + void *addr = um_virt_to_phys(current, current->mm->arg_start, NULL); + return IS_ERR(addr) ? "(Unknown)": __va((unsigned long) addr); +#endif +} + +void force_sigbus(void) +{ + printk(KERN_ERR "Killing pid %d because of a lack of memory\n", + current->pid); + lock_kernel(); + sigaddset(¤t->pending.signal, SIGBUS); + recalc_sigpending(); + current->flags |= PF_SIGNALED; + do_exit(SIGBUS | 0x80); +} + +void dump_thread(struct pt_regs *regs, struct user *u) +{ +} + +void enable_hlt(void) +{ + panic("enable_hlt"); +} + +EXPORT_SYMBOL(enable_hlt); + +void disable_hlt(void) +{ + panic("disable_hlt"); +} + +EXPORT_SYMBOL(disable_hlt); + +void *um_kmalloc(int size) +{ + return(kmalloc(size, GFP_KERNEL)); +} + +void *um_kmalloc_atomic(int size) +{ + return(kmalloc(size, GFP_ATOMIC)); +} + +void *um_vmalloc(int size) +{ + return(vmalloc(size)); +} + +unsigned long get_fault_addr(void) +{ + return((unsigned long) current->thread.fault_addr); +} + +EXPORT_SYMBOL(get_fault_addr); + +void not_implemented(void) +{ + printk(KERN_DEBUG "Something isn't implemented in here\n"); +} + +EXPORT_SYMBOL(not_implemented); + +int user_context(unsigned long sp) +{ + unsigned long stack; + + stack = sp & (PAGE_MASK << CONFIG_KERNEL_STACK_ORDER); + return(stack != (unsigned long) current_thread); +} + +extern void remove_umid_dir(void); + +__uml_exitcall(remove_umid_dir); + +extern exitcall_t __uml_exitcall_begin, __uml_exitcall_end; + +void do_uml_exitcalls(void) +{ + exitcall_t *call; + + call = &__uml_exitcall_end; + while (--call >= &__uml_exitcall_begin) + (*call)(); +} + +char *uml_strdup(char *string) +{ + char *new; + + new = kmalloc(strlen(string) + 1, GFP_KERNEL); + if(new == NULL) return(NULL); + strcpy(new, string); + return(new); +} + +void *get_init_task(void) +{ + return(&init_thread_union.thread_info.task); +} + +int copy_to_user_proc(void __user *to, void *from, int size) +{ + return(copy_to_user(to, from, size)); +} + +int copy_from_user_proc(void *to, void __user *from, int size) +{ + return(copy_from_user(to, from, size)); +} + +int clear_user_proc(void __user *buf, int size) +{ + return(clear_user(buf, size)); +} + +int strlen_user_proc(char __user *str) +{ + return(strlen_user(str)); +} + +int smp_sigio_handler(void) +{ +#ifdef CONFIG_SMP + int cpu = current_thread->cpu; + IPI_handler(cpu); + if(cpu != 0) + return(1); +#endif + return(0); +} + +int um_in_interrupt(void) +{ + return(in_interrupt()); +} + +int cpu(void) +{ + return(current_thread->cpu); +} + +static atomic_t using_sysemu = ATOMIC_INIT(0); +int sysemu_supported; + +void set_using_sysemu(int value) +{ + if (value > sysemu_supported) + return; + atomic_set(&using_sysemu, value); +} + +int get_using_sysemu(void) +{ + return atomic_read(&using_sysemu); +} + +static int proc_read_sysemu(char *buf, char **start, off_t offset, int size,int *eof, void *data) +{ + if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/ + *eof = 1; + + return strlen(buf); +} + +static int proc_write_sysemu(struct file *file,const char *buf, unsigned long count,void *data) +{ + char tmp[2]; + + if (copy_from_user(tmp, buf, 1)) + return -EFAULT; + + if (tmp[0] >= '0' && tmp[0] <= '2') + set_using_sysemu(tmp[0] - '0'); + return count; /*We use the first char, but pretend to write everything*/ +} + +int __init make_proc_sysemu(void) +{ + struct proc_dir_entry *ent; + if (!sysemu_supported) + return 0; + + ent = create_proc_entry("sysemu", 0600, &proc_root); + + if (ent == NULL) + { + printk("Failed to register /proc/sysemu\n"); + return(0); + } + + ent->read_proc = proc_read_sysemu; + ent->write_proc = proc_write_sysemu; + + return 0; +} + +late_initcall(make_proc_sysemu); + +int singlestepping(void * t) +{ + struct task_struct *task = t ? t : current; + + if ( ! (task->ptrace & PT_DTRACE) ) + return(0); + + if (task->thread.singlestep_syscall) + return(1); + + return 2; +} + +unsigned long arch_align_stack(unsigned long sp) +{ + if (randomize_va_space) + sp -= get_random_int() % 8192; + return sp & ~0xf; +} + + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c new file mode 100644 index 00000000000..3a99ee6d94e --- /dev/null +++ b/arch/um/kernel/ptrace.c @@ -0,0 +1,388 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/mm.h" +#include "linux/errno.h" +#include "linux/smp_lock.h" +#include "linux/security.h" +#include "linux/ptrace.h" +#include "linux/audit.h" +#ifdef CONFIG_PROC_MM +#include "linux/proc_mm.h" +#endif +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "kern_util.h" +#include "skas_ptrace.h" +#include "sysdep/ptrace.h" + +/* + * Called by kernel/ptrace.c when detaching.. + */ +void ptrace_disable(struct task_struct *child) +{ + child->ptrace &= ~PT_DTRACE; + child->thread.singlestep_syscall = 0; +} + +long sys_ptrace(long request, long pid, long addr, long data) +{ + struct task_struct *child; + int i, ret; + + lock_kernel(); + ret = -EPERM; + if (request == PTRACE_TRACEME) { + /* are we already being traced? */ + if (current->ptrace & PT_PTRACED) + goto out; + + ret = security_ptrace(current->parent, current); + if (ret) + goto out; + + /* set the ptrace bit in the process flags. */ + current->ptrace |= PT_PTRACED; + ret = 0; + goto out; + } + ret = -ESRCH; + read_lock(&tasklist_lock); + child = find_task_by_pid(pid); + if (child) + get_task_struct(child); + read_unlock(&tasklist_lock); + if (!child) + goto out; + + ret = -EPERM; + if (pid == 1) /* you may not mess with init */ + goto out_tsk; + + if (request == PTRACE_ATTACH) { + ret = ptrace_attach(child); + goto out_tsk; + } + + ret = ptrace_check_attach(child, request == PTRACE_KILL); + if (ret < 0) + goto out_tsk; + + switch (request) { + /* when I and D space are separate, these will need to be fixed. */ + case PTRACE_PEEKTEXT: /* read word at location addr. */ + case PTRACE_PEEKDATA: { + unsigned long tmp; + int copied; + + ret = -EIO; + copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0); + if (copied != sizeof(tmp)) + break; + ret = put_user(tmp, (unsigned long __user *) data); + break; + } + + /* read the word at location addr in the USER area. */ + case PTRACE_PEEKUSR: { + unsigned long tmp; + + ret = -EIO; + if ((addr & 3) || addr < 0) + break; + + tmp = 0; /* Default return condition */ + if(addr < MAX_REG_OFFSET){ + tmp = getreg(child, addr); + } + else if((addr >= offsetof(struct user, u_debugreg[0])) && + (addr <= offsetof(struct user, u_debugreg[7]))){ + addr -= offsetof(struct user, u_debugreg[0]); + addr = addr >> 2; + tmp = child->thread.arch.debugregs[addr]; + } + ret = put_user(tmp, (unsigned long __user *) data); + break; + } + + /* when I and D space are separate, this will have to be fixed. */ + case PTRACE_POKETEXT: /* write the word at location addr. */ + case PTRACE_POKEDATA: + ret = -EIO; + if (access_process_vm(child, addr, &data, sizeof(data), + 1) != sizeof(data)) + break; + ret = 0; + break; + + case PTRACE_POKEUSR: /* write the word at location addr in the USER area */ + ret = -EIO; + if ((addr & 3) || addr < 0) + break; + + if (addr < MAX_REG_OFFSET) { + ret = putreg(child, addr, data); + break; + } +#if 0 /* XXX x86_64 */ + else if((addr >= offsetof(struct user, u_debugreg[0])) && + (addr <= offsetof(struct user, u_debugreg[7]))){ + addr -= offsetof(struct user, u_debugreg[0]); + addr = addr >> 2; + if((addr == 4) || (addr == 5)) break; + child->thread.arch.debugregs[addr] = data; + ret = 0; + } +#endif + + break; + + case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ + case PTRACE_CONT: { /* restart after signal. */ + ret = -EIO; + if ((unsigned long) data > _NSIG) + break; + + child->ptrace &= ~PT_DTRACE; + child->thread.singlestep_syscall = 0; + if (request == PTRACE_SYSCALL) { + set_tsk_thread_flag(child, TIF_SYSCALL_TRACE); + } + else { + clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); + } + child->exit_code = data; + wake_up_process(child); + ret = 0; + break; + } + +/* + * make the child exit. Best I can do is send it a sigkill. + * perhaps it should be put in the status that it wants to + * exit. + */ + case PTRACE_KILL: { + ret = 0; + if (child->exit_state == EXIT_ZOMBIE) /* already dead */ + break; + + child->ptrace &= ~PT_DTRACE; + child->thread.singlestep_syscall = 0; + child->exit_code = SIGKILL; + wake_up_process(child); + break; + } + + case PTRACE_SINGLESTEP: { /* set the trap flag. */ + ret = -EIO; + if ((unsigned long) data > _NSIG) + break; + clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); + child->ptrace |= PT_DTRACE; + child->thread.singlestep_syscall = 0; + child->exit_code = data; + /* give it a chance to run. */ + wake_up_process(child); + ret = 0; + break; + } + + case PTRACE_DETACH: + /* detach a process that was attached. */ + ret = ptrace_detach(child, data); + break; + +#ifdef PTRACE_GETREGS + case PTRACE_GETREGS: { /* Get all gp regs from the child. */ + if (!access_ok(VERIFY_WRITE, (unsigned long *)data, + MAX_REG_OFFSET)) { + ret = -EIO; + break; + } + for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) { + __put_user(getreg(child, i), + (unsigned long __user *) data); + data += sizeof(long); + } + ret = 0; + break; + } +#endif +#ifdef PTRACE_SETREGS + case PTRACE_SETREGS: { /* Set all gp regs in the child. */ + unsigned long tmp = 0; + if (!access_ok(VERIFY_READ, (unsigned *)data, + MAX_REG_OFFSET)) { + ret = -EIO; + break; + } + for ( i = 0; i < MAX_REG_OFFSET; i += sizeof(long) ) { + __get_user(tmp, (unsigned long __user *) data); + putreg(child, i, tmp); + data += sizeof(long); + } + ret = 0; + break; + } +#endif +#ifdef PTRACE_GETFPREGS + case PTRACE_GETFPREGS: /* Get the child FPU state. */ + ret = get_fpregs(data, child); + break; +#endif +#ifdef PTRACE_SETFPREGS + case PTRACE_SETFPREGS: /* Set the child FPU state. */ + ret = set_fpregs(data, child); + break; +#endif +#ifdef PTRACE_GETFPXREGS + case PTRACE_GETFPXREGS: /* Get the child FPU state. */ + ret = get_fpxregs(data, child); + break; +#endif +#ifdef PTRACE_SETFPXREGS + case PTRACE_SETFPXREGS: /* Set the child FPU state. */ + ret = set_fpxregs(data, child); + break; +#endif + case PTRACE_FAULTINFO: { + struct ptrace_faultinfo fault; + + fault = ((struct ptrace_faultinfo) + { .is_write = child->thread.err, + .addr = child->thread.cr2 }); + ret = copy_to_user((unsigned long __user *) data, &fault, + sizeof(fault)); + if(ret) + break; + break; + } + case PTRACE_SIGPENDING: + ret = copy_to_user((unsigned long __user *) data, + &child->pending.signal, + sizeof(child->pending.signal)); + break; + + case PTRACE_LDT: { + struct ptrace_ldt ldt; + + if(copy_from_user(&ldt, (unsigned long __user *) data, + sizeof(ldt))){ + ret = -EIO; + break; + } + + /* This one is confusing, so just punt and return -EIO for + * now + */ + ret = -EIO; + break; + } +#ifdef CONFIG_PROC_MM + case PTRACE_SWITCH_MM: { + struct mm_struct *old = child->mm; + struct mm_struct *new = proc_mm_get_mm(data); + + if(IS_ERR(new)){ + ret = PTR_ERR(new); + break; + } + + atomic_inc(&new->mm_users); + child->mm = new; + child->active_mm = new; + mmput(old); + ret = 0; + break; + } +#endif + default: + ret = ptrace_request(child, request, addr, data); + break; + } + out_tsk: + put_task_struct(child); + out: + unlock_kernel(); + return ret; +} + +void send_sigtrap(struct task_struct *tsk, union uml_pt_regs *regs, + int error_code) +{ + struct siginfo info; + + memset(&info, 0, sizeof(info)); + info.si_signo = SIGTRAP; + info.si_code = TRAP_BRKPT; + + /* User-mode eip? */ + info.si_addr = UPT_IS_USER(regs) ? (void __user *) UPT_IP(regs) : NULL; + + /* Send us the fakey SIGTRAP */ + force_sig_info(SIGTRAP, &info, tsk); +} + +/* XXX Check PT_DTRACE vs TIF_SINGLESTEP for singlestepping check and + * PT_PTRACED vs TIF_SYSCALL_TRACE for syscall tracing check + */ +void syscall_trace(union uml_pt_regs *regs, int entryexit) +{ + int is_singlestep = (current->ptrace & PT_DTRACE) && entryexit; + int tracesysgood; + + if (unlikely(current->audit_context)) { + if (!entryexit) + audit_syscall_entry(current, + UPT_SYSCALL_NR(®s->regs), + UPT_SYSCALL_ARG1(®s->regs), + UPT_SYSCALL_ARG2(®s->regs), + UPT_SYSCALL_ARG3(®s->regs), + UPT_SYSCALL_ARG4(®s->regs)); + else + audit_syscall_exit(current, + UPT_SYSCALL_RET(®s->regs)); + } + + /* Fake a debug trap */ + if (is_singlestep) + send_sigtrap(current, regs, 0); + + if (!test_thread_flag(TIF_SYSCALL_TRACE)) + return; + + if (!(current->ptrace & PT_PTRACED)) + return; + + /* the 0x80 provides a way for the tracing parent to distinguish + between a syscall stop and SIGTRAP delivery */ + tracesysgood = (current->ptrace & PT_TRACESYSGOOD); + ptrace_notify(SIGTRAP | (tracesysgood ? 0x80 : 0)); + + if (entryexit) /* force do_signal() --> is_syscall() */ + set_thread_flag(TIF_SIGPENDING); + + /* this isn't the same as continuing with a signal, but it will do + * for normal use. strace only continues with a signal if the + * stopping signal is not SIGTRAP. -brl + */ + if (current->exit_code) { + send_sig(current->exit_code, current, 1); + current->exit_code = 0; + } +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c new file mode 100644 index 00000000000..207f89d7490 --- /dev/null +++ b/arch/um/kernel/reboot.c @@ -0,0 +1,79 @@ +/* + * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/module.h" +#include "linux/sched.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "os.h" +#include "mode.h" +#include "choose-mode.h" + +#ifdef CONFIG_SMP +static void kill_idlers(int me) +{ +#ifdef CONFIG_MODE_TT + struct task_struct *p; + int i; + + for(i = 0; i < sizeof(idle_threads)/sizeof(idle_threads[0]); i++){ + p = idle_threads[i]; + if((p != NULL) && (p->thread.mode.tt.extern_pid != me)) + os_kill_process(p->thread.mode.tt.extern_pid, 0); + } +#endif +} +#endif + +static void kill_off_processes(void) +{ + CHOOSE_MODE(kill_off_processes_tt(), kill_off_processes_skas()); +#ifdef CONFIG_SMP + kill_idlers(os_getpid()); +#endif +} + +void uml_cleanup(void) +{ + kill_off_processes(); + do_uml_exitcalls(); +} + +void machine_restart(char * __unused) +{ + do_uml_exitcalls(); + kill_off_processes(); + CHOOSE_MODE(reboot_tt(), reboot_skas()); +} + +EXPORT_SYMBOL(machine_restart); + +void machine_power_off(void) +{ + do_uml_exitcalls(); + kill_off_processes(); + CHOOSE_MODE(halt_tt(), halt_skas()); +} + +EXPORT_SYMBOL(machine_power_off); + +void machine_halt(void) +{ + machine_power_off(); +} + +EXPORT_SYMBOL(machine_halt); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/resource.c b/arch/um/kernel/resource.c new file mode 100644 index 00000000000..32188e12e8a --- /dev/null +++ b/arch/um/kernel/resource.c @@ -0,0 +1,23 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/pci.h" + +unsigned long resource_fixup(struct pci_dev * dev, struct resource * res, + unsigned long start, unsigned long size) +{ + return start; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/sigio_kern.c b/arch/um/kernel/sigio_kern.c new file mode 100644 index 00000000000..229988463c4 --- /dev/null +++ b/arch/um/kernel/sigio_kern.c @@ -0,0 +1,63 @@ +/* + * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/list.h" +#include "linux/slab.h" +#include "linux/signal.h" +#include "linux/interrupt.h" +#include "init.h" +#include "sigio.h" +#include "irq_user.h" +#include "irq_kern.h" + +/* Protected by sigio_lock() called from write_sigio_workaround */ +static int sigio_irq_fd = -1; + +static irqreturn_t sigio_interrupt(int irq, void *data, struct pt_regs *unused) +{ + read_sigio_fd(sigio_irq_fd); + reactivate_fd(sigio_irq_fd, SIGIO_WRITE_IRQ); + return(IRQ_HANDLED); +} + +int write_sigio_irq(int fd) +{ + int err; + + err = um_request_irq(SIGIO_WRITE_IRQ, fd, IRQ_READ, sigio_interrupt, + SA_INTERRUPT | SA_SAMPLE_RANDOM, "write sigio", + NULL); + if(err){ + printk("write_sigio_irq : um_request_irq failed, err = %d\n", + err); + return(-1); + } + sigio_irq_fd = fd; + return(0); +} + +static DEFINE_SPINLOCK(sigio_spinlock); + +void sigio_lock(void) +{ + spin_lock(&sigio_spinlock); +} + +void sigio_unlock(void) +{ + spin_unlock(&sigio_spinlock); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/sigio_user.c b/arch/um/kernel/sigio_user.c new file mode 100644 index 00000000000..668df13d8c9 --- /dev/null +++ b/arch/um/kernel/sigio_user.c @@ -0,0 +1,431 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <stdlib.h> +#include <termios.h> +#include <pty.h> +#include <signal.h> +#include <errno.h> +#include <string.h> +#include <sched.h> +#include <sys/socket.h> +#include <sys/poll.h> +#include "init.h" +#include "user.h" +#include "kern_util.h" +#include "user_util.h" +#include "sigio.h" +#include "helper.h" +#include "os.h" + +/* Changed during early boot */ +int pty_output_sigio = 0; +int pty_close_sigio = 0; + +/* Used as a flag during SIGIO testing early in boot */ +static volatile int got_sigio = 0; + +void __init handler(int sig) +{ + got_sigio = 1; +} + +struct openpty_arg { + int master; + int slave; + int err; +}; + +static void openpty_cb(void *arg) +{ + struct openpty_arg *info = arg; + + info->err = 0; + if(openpty(&info->master, &info->slave, NULL, NULL, NULL)) + info->err = -errno; +} + +void __init check_one_sigio(void (*proc)(int, int)) +{ + struct sigaction old, new; + struct openpty_arg pty = { .master = -1, .slave = -1 }; + int master, slave, err; + + initial_thread_cb(openpty_cb, &pty); + if(pty.err){ + printk("openpty failed, errno = %d\n", -pty.err); + return; + } + + master = pty.master; + slave = pty.slave; + + if((master == -1) || (slave == -1)){ + printk("openpty failed to allocate a pty\n"); + return; + } + + /* Not now, but complain so we now where we failed. */ + err = raw(master); + if (err < 0) + panic("check_sigio : __raw failed, errno = %d\n", -err); + + err = os_sigio_async(master, slave); + if(err < 0) + panic("tty_fds : sigio_async failed, err = %d\n", -err); + + if(sigaction(SIGIO, NULL, &old) < 0) + panic("check_sigio : sigaction 1 failed, errno = %d\n", errno); + new = old; + new.sa_handler = handler; + if(sigaction(SIGIO, &new, NULL) < 0) + panic("check_sigio : sigaction 2 failed, errno = %d\n", errno); + + got_sigio = 0; + (*proc)(master, slave); + + os_close_file(master); + os_close_file(slave); + + if(sigaction(SIGIO, &old, NULL) < 0) + panic("check_sigio : sigaction 3 failed, errno = %d\n", errno); +} + +static void tty_output(int master, int slave) +{ + int n; + char buf[512]; + + printk("Checking that host ptys support output SIGIO..."); + + memset(buf, 0, sizeof(buf)); + + while(os_write_file(master, buf, sizeof(buf)) > 0) ; + if(errno != EAGAIN) + panic("check_sigio : write failed, errno = %d\n", errno); + while(((n = os_read_file(slave, buf, sizeof(buf))) > 0) && !got_sigio) ; + + if (got_sigio) { + printk("Yes\n"); + pty_output_sigio = 1; + } else if (n == -EAGAIN) { + printk("No, enabling workaround\n"); + } else { + panic("check_sigio : read failed, err = %d\n", n); + } +} + +static void tty_close(int master, int slave) +{ + printk("Checking that host ptys support SIGIO on close..."); + + os_close_file(slave); + if(got_sigio){ + printk("Yes\n"); + pty_close_sigio = 1; + } + else printk("No, enabling workaround\n"); +} + +void __init check_sigio(void) +{ + if((os_access("/dev/ptmx", OS_ACC_R_OK) < 0) && + (os_access("/dev/ptyp0", OS_ACC_R_OK) < 0)){ + printk("No pseudo-terminals available - skipping pty SIGIO " + "check\n"); + return; + } + check_one_sigio(tty_output); + check_one_sigio(tty_close); +} + +/* Protected by sigio_lock(), also used by sigio_cleanup, which is an + * exitcall. + */ +static int write_sigio_pid = -1; + +/* These arrays are initialized before the sigio thread is started, and + * the descriptors closed after it is killed. So, it can't see them change. + * On the UML side, they are changed under the sigio_lock. + */ +static int write_sigio_fds[2] = { -1, -1 }; +static int sigio_private[2] = { -1, -1 }; + +struct pollfds { + struct pollfd *poll; + int size; + int used; +}; + +/* Protected by sigio_lock(). Used by the sigio thread, but the UML thread + * synchronizes with it. + */ +struct pollfds current_poll = { + .poll = NULL, + .size = 0, + .used = 0 +}; + +struct pollfds next_poll = { + .poll = NULL, + .size = 0, + .used = 0 +}; + +static int write_sigio_thread(void *unused) +{ + struct pollfds *fds, tmp; + struct pollfd *p; + int i, n, respond_fd; + char c; + + fds = ¤t_poll; + while(1){ + n = poll(fds->poll, fds->used, -1); + if(n < 0){ + if(errno == EINTR) continue; + printk("write_sigio_thread : poll returned %d, " + "errno = %d\n", n, errno); + } + for(i = 0; i < fds->used; i++){ + p = &fds->poll[i]; + if(p->revents == 0) continue; + if(p->fd == sigio_private[1]){ + n = os_read_file(sigio_private[1], &c, sizeof(c)); + if(n != sizeof(c)) + printk("write_sigio_thread : " + "read failed, err = %d\n", -n); + tmp = current_poll; + current_poll = next_poll; + next_poll = tmp; + respond_fd = sigio_private[1]; + } + else { + respond_fd = write_sigio_fds[1]; + fds->used--; + memmove(&fds->poll[i], &fds->poll[i + 1], + (fds->used - i) * sizeof(*fds->poll)); + } + + n = os_write_file(respond_fd, &c, sizeof(c)); + if(n != sizeof(c)) + printk("write_sigio_thread : write failed, " + "err = %d\n", -n); + } + } +} + +static int need_poll(int n) +{ + if(n <= next_poll.size){ + next_poll.used = n; + return(0); + } + if(next_poll.poll != NULL) kfree(next_poll.poll); + next_poll.poll = um_kmalloc_atomic(n * sizeof(struct pollfd)); + if(next_poll.poll == NULL){ + printk("need_poll : failed to allocate new pollfds\n"); + next_poll.size = 0; + next_poll.used = 0; + return(-1); + } + next_poll.size = n; + next_poll.used = n; + return(0); +} + +/* Must be called with sigio_lock held, because it's needed by the marked + * critical section. */ +static void update_thread(void) +{ + unsigned long flags; + int n; + char c; + + flags = set_signals(0); + n = os_write_file(sigio_private[0], &c, sizeof(c)); + if(n != sizeof(c)){ + printk("update_thread : write failed, err = %d\n", -n); + goto fail; + } + + n = os_read_file(sigio_private[0], &c, sizeof(c)); + if(n != sizeof(c)){ + printk("update_thread : read failed, err = %d\n", -n); + goto fail; + } + + set_signals(flags); + return; + fail: + /* Critical section start */ + if(write_sigio_pid != -1) + os_kill_process(write_sigio_pid, 1); + write_sigio_pid = -1; + os_close_file(sigio_private[0]); + os_close_file(sigio_private[1]); + os_close_file(write_sigio_fds[0]); + os_close_file(write_sigio_fds[1]); + /* Critical section end */ + set_signals(flags); +} + +int add_sigio_fd(int fd, int read) +{ + int err = 0, i, n, events; + + sigio_lock(); + for(i = 0; i < current_poll.used; i++){ + if(current_poll.poll[i].fd == fd) + goto out; + } + + n = current_poll.used + 1; + err = need_poll(n); + if(err) + goto out; + + for(i = 0; i < current_poll.used; i++) + next_poll.poll[i] = current_poll.poll[i]; + + if(read) events = POLLIN; + else events = POLLOUT; + + next_poll.poll[n - 1] = ((struct pollfd) { .fd = fd, + .events = events, + .revents = 0 }); + update_thread(); + out: + sigio_unlock(); + return(err); +} + +int ignore_sigio_fd(int fd) +{ + struct pollfd *p; + int err = 0, i, n = 0; + + sigio_lock(); + for(i = 0; i < current_poll.used; i++){ + if(current_poll.poll[i].fd == fd) break; + } + if(i == current_poll.used) + goto out; + + err = need_poll(current_poll.used - 1); + if(err) + goto out; + + for(i = 0; i < current_poll.used; i++){ + p = ¤t_poll.poll[i]; + if(p->fd != fd) next_poll.poll[n++] = current_poll.poll[i]; + } + if(n == i){ + printk("ignore_sigio_fd : fd %d not found\n", fd); + err = -1; + goto out; + } + + update_thread(); + out: + sigio_unlock(); + return(err); +} + +static int setup_initial_poll(int fd) +{ + struct pollfd *p; + + p = um_kmalloc(sizeof(struct pollfd)); + if(p == NULL){ + printk("setup_initial_poll : failed to allocate poll\n"); + return(-1); + } + *p = ((struct pollfd) { .fd = fd, + .events = POLLIN, + .revents = 0 }); + current_poll = ((struct pollfds) { .poll = p, + .used = 1, + .size = 1 }); + return(0); +} + +void write_sigio_workaround(void) +{ + unsigned long stack; + int err; + + sigio_lock(); + if(write_sigio_pid != -1) + goto out; + + err = os_pipe(write_sigio_fds, 1, 1); + if(err < 0){ + printk("write_sigio_workaround - os_pipe 1 failed, " + "err = %d\n", -err); + goto out; + } + err = os_pipe(sigio_private, 1, 1); + if(err < 0){ + printk("write_sigio_workaround - os_pipe 2 failed, " + "err = %d\n", -err); + goto out_close1; + } + if(setup_initial_poll(sigio_private[1])) + goto out_close2; + + write_sigio_pid = run_helper_thread(write_sigio_thread, NULL, + CLONE_FILES | CLONE_VM, &stack, 0); + + if(write_sigio_pid < 0) goto out_close2; + + if(write_sigio_irq(write_sigio_fds[0])) + goto out_kill; + + out: + sigio_unlock(); + return; + + out_kill: + os_kill_process(write_sigio_pid, 1); + write_sigio_pid = -1; + out_close2: + os_close_file(sigio_private[0]); + os_close_file(sigio_private[1]); + out_close1: + os_close_file(write_sigio_fds[0]); + os_close_file(write_sigio_fds[1]); + sigio_unlock(); +} + +int read_sigio_fd(int fd) +{ + int n; + char c; + + n = os_read_file(fd, &c, sizeof(c)); + if(n != sizeof(c)){ + if(n < 0) { + printk("read_sigio_fd - read failed, err = %d\n", -n); + return(n); + } + else { + printk("read_sigio_fd - short read, bytes = %d\n", n); + return(-EIO); + } + } + return(n); +} + +static void sigio_cleanup(void) +{ + if (write_sigio_pid != -1) { + os_kill_process(write_sigio_pid, 1); + write_sigio_pid = -1; + } +} + +__uml_exitcall(sigio_cleanup); diff --git a/arch/um/kernel/signal_kern.c b/arch/um/kernel/signal_kern.c new file mode 100644 index 00000000000..7807a3e8c42 --- /dev/null +++ b/arch/um/kernel/signal_kern.c @@ -0,0 +1,213 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/stddef.h" +#include "linux/sys.h" +#include "linux/sched.h" +#include "linux/wait.h" +#include "linux/kernel.h" +#include "linux/smp_lock.h" +#include "linux/module.h" +#include "linux/slab.h" +#include "linux/tty.h" +#include "linux/binfmts.h" +#include "linux/ptrace.h" +#include "asm/signal.h" +#include "asm/uaccess.h" +#include "asm/unistd.h" +#include "user_util.h" +#include "asm/ucontext.h" +#include "kern_util.h" +#include "signal_kern.h" +#include "signal_user.h" +#include "kern.h" +#include "frame_kern.h" +#include "sigcontext.h" +#include "mode.h" + +EXPORT_SYMBOL(block_signals); +EXPORT_SYMBOL(unblock_signals); + +#define _S(nr) (1<<((nr)-1)) + +#define _BLOCKABLE (~(_S(SIGKILL) | _S(SIGSTOP))) + +/* + * OK, we're invoking a handler + */ +static int handle_signal(struct pt_regs *regs, unsigned long signr, + struct k_sigaction *ka, siginfo_t *info, + sigset_t *oldset) +{ + unsigned long sp; + int err; + + /* Always make any pending restarted system calls return -EINTR */ + current_thread_info()->restart_block.fn = do_no_restart_syscall; + + /* Did we come from a system call? */ + if(PT_REGS_SYSCALL_NR(regs) >= 0){ + /* If so, check system call restarting.. */ + switch(PT_REGS_SYSCALL_RET(regs)){ + case -ERESTART_RESTARTBLOCK: + case -ERESTARTNOHAND: + PT_REGS_SYSCALL_RET(regs) = -EINTR; + break; + + case -ERESTARTSYS: + if (!(ka->sa.sa_flags & SA_RESTART)) { + PT_REGS_SYSCALL_RET(regs) = -EINTR; + break; + } + /* fallthrough */ + case -ERESTARTNOINTR: + PT_REGS_RESTART_SYSCALL(regs); + PT_REGS_ORIG_SYSCALL(regs) = PT_REGS_SYSCALL_NR(regs); + break; + } + } + + sp = PT_REGS_SP(regs); + if((ka->sa.sa_flags & SA_ONSTACK) && (sas_ss_flags(sp) == 0)) + sp = current->sas_ss_sp + current->sas_ss_size; + +#ifdef CONFIG_ARCH_HAS_SC_SIGNALS + if(!(ka->sa.sa_flags & SA_SIGINFO)) + err = setup_signal_stack_sc(sp, signr, ka, regs, oldset); + else +#endif + err = setup_signal_stack_si(sp, signr, ka, regs, info, oldset); + + if(err){ + spin_lock_irq(¤t->sighand->siglock); + current->blocked = *oldset; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + force_sigsegv(signr, current); + } + else if(!(ka->sa.sa_flags & SA_NODEFER)){ + spin_lock_irq(¤t->sighand->siglock); + sigorsets(¤t->blocked, ¤t->blocked, + &ka->sa.sa_mask); + sigaddset(¤t->blocked, signr); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + } + + return err; +} + +static int kern_do_signal(struct pt_regs *regs, sigset_t *oldset) +{ + struct k_sigaction ka_copy; + siginfo_t info; + int sig, handled_sig = 0; + + while((sig = get_signal_to_deliver(&info, &ka_copy, regs, NULL)) > 0){ + handled_sig = 1; + /* Whee! Actually deliver the signal. */ + if(!handle_signal(regs, sig, &ka_copy, &info, oldset)) + break; + } + + /* Did we come from a system call? */ + if(!handled_sig && (PT_REGS_SYSCALL_NR(regs) >= 0)){ + /* Restart the system call - no handlers present */ + if(PT_REGS_SYSCALL_RET(regs) == -ERESTARTNOHAND || + PT_REGS_SYSCALL_RET(regs) == -ERESTARTSYS || + PT_REGS_SYSCALL_RET(regs) == -ERESTARTNOINTR){ + PT_REGS_ORIG_SYSCALL(regs) = PT_REGS_SYSCALL_NR(regs); + PT_REGS_RESTART_SYSCALL(regs); + } + else if(PT_REGS_SYSCALL_RET(regs) == -ERESTART_RESTARTBLOCK){ + PT_REGS_SYSCALL_RET(regs) = __NR_restart_syscall; + PT_REGS_RESTART_SYSCALL(regs); + } + } + + /* This closes a way to execute a system call on the host. If + * you set a breakpoint on a system call instruction and singlestep + * from it, the tracing thread used to PTRACE_SINGLESTEP the process + * rather than PTRACE_SYSCALL it, allowing the system call to execute + * on the host. The tracing thread will check this flag and + * PTRACE_SYSCALL if necessary. + */ + if(current->ptrace & PT_DTRACE) + current->thread.singlestep_syscall = + is_syscall(PT_REGS_IP(¤t->thread.regs)); + return(handled_sig); +} + +int do_signal(void) +{ + return(kern_do_signal(¤t->thread.regs, ¤t->blocked)); +} + +/* + * Atomically swap in the new signal mask, and wait for a signal. + */ +long sys_sigsuspend(int history0, int history1, old_sigset_t mask) +{ + sigset_t saveset; + + mask &= _BLOCKABLE; + spin_lock_irq(¤t->sighand->siglock); + saveset = current->blocked; + siginitset(¤t->blocked, mask); + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + PT_REGS_SYSCALL_RET(¤t->thread.regs) = -EINTR; + while (1) { + current->state = TASK_INTERRUPTIBLE; + schedule(); + if(kern_do_signal(¤t->thread.regs, &saveset)) + return(-EINTR); + } +} + +long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) +{ + sigset_t saveset, newset; + + /* XXX: Don't preclude handling different sized sigset_t's. */ + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + + if (copy_from_user(&newset, unewset, sizeof(newset))) + return -EFAULT; + sigdelsetmask(&newset, ~_BLOCKABLE); + + spin_lock_irq(¤t->sighand->siglock); + saveset = current->blocked; + current->blocked = newset; + recalc_sigpending(); + spin_unlock_irq(¤t->sighand->siglock); + + PT_REGS_SYSCALL_RET(¤t->thread.regs) = -EINTR; + while (1) { + current->state = TASK_INTERRUPTIBLE; + schedule(); + if (kern_do_signal(¤t->thread.regs, &saveset)) + return(-EINTR); + } +} + +long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss) +{ + return(do_sigaltstack(uss, uoss, PT_REGS_SP(¤t->thread.regs))); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/signal_user.c b/arch/um/kernel/signal_user.c new file mode 100644 index 00000000000..62f457835fb --- /dev/null +++ b/arch/um/kernel/signal_user.c @@ -0,0 +1,157 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <stdlib.h> +#include <signal.h> +#include <errno.h> +#include <stdarg.h> +#include <string.h> +#include <sys/mman.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "signal_user.h" +#include "signal_kern.h" +#include "sysdep/sigcontext.h" +#include "sigcontext.h" + +void set_sigstack(void *sig_stack, int size) +{ + stack_t stack = ((stack_t) { .ss_flags = 0, + .ss_sp = (__ptr_t) sig_stack, + .ss_size = size - sizeof(void *) }); + + if(sigaltstack(&stack, NULL) != 0) + panic("enabling signal stack failed, errno = %d\n", errno); +} + +void set_handler(int sig, void (*handler)(int), int flags, ...) +{ + struct sigaction action; + va_list ap; + int mask; + + va_start(ap, flags); + action.sa_handler = handler; + sigemptyset(&action.sa_mask); + while((mask = va_arg(ap, int)) != -1){ + sigaddset(&action.sa_mask, mask); + } + va_end(ap); + action.sa_flags = flags; + action.sa_restorer = NULL; + if(sigaction(sig, &action, NULL) < 0) + panic("sigaction failed"); +} + +int change_sig(int signal, int on) +{ + sigset_t sigset, old; + + sigemptyset(&sigset); + sigaddset(&sigset, signal); + sigprocmask(on ? SIG_UNBLOCK : SIG_BLOCK, &sigset, &old); + return(!sigismember(&old, signal)); +} + +/* Both here and in set/get_signal we don't touch SIGPROF, because we must not + * disable profiling; it's safe because the profiling code does not interact + * with the kernel code at all.*/ + +static void change_signals(int type) +{ + sigset_t mask; + + sigemptyset(&mask); + sigaddset(&mask, SIGVTALRM); + sigaddset(&mask, SIGALRM); + sigaddset(&mask, SIGIO); + if(sigprocmask(type, &mask, NULL) < 0) + panic("Failed to change signal mask - errno = %d", errno); +} + +void block_signals(void) +{ + change_signals(SIG_BLOCK); +} + +void unblock_signals(void) +{ + change_signals(SIG_UNBLOCK); +} + +/* These are the asynchronous signals. SIGVTALRM and SIGARLM are handled + * together under SIGVTALRM_BIT. SIGPROF is excluded because we want to + * be able to profile all of UML, not just the non-critical sections. If + * profiling is not thread-safe, then that is not my problem. We can disable + * profiling when SMP is enabled in that case. + */ +#define SIGIO_BIT 0 +#define SIGVTALRM_BIT 1 + +static int enable_mask(sigset_t *mask) +{ + int sigs; + + sigs = sigismember(mask, SIGIO) ? 0 : 1 << SIGIO_BIT; + sigs |= sigismember(mask, SIGVTALRM) ? 0 : 1 << SIGVTALRM_BIT; + sigs |= sigismember(mask, SIGALRM) ? 0 : 1 << SIGVTALRM_BIT; + return(sigs); +} + +int get_signals(void) +{ + sigset_t mask; + + if(sigprocmask(SIG_SETMASK, NULL, &mask) < 0) + panic("Failed to get signal mask"); + return(enable_mask(&mask)); +} + +int set_signals(int enable) +{ + sigset_t mask; + int ret; + + sigemptyset(&mask); + if(enable & (1 << SIGIO_BIT)) + sigaddset(&mask, SIGIO); + if(enable & (1 << SIGVTALRM_BIT)){ + sigaddset(&mask, SIGVTALRM); + sigaddset(&mask, SIGALRM); + } + + /* This is safe - sigprocmask is guaranteed to copy locally the + * value of new_set, do his work and then, at the end, write to + * old_set. + */ + if(sigprocmask(SIG_UNBLOCK, &mask, &mask) < 0) + panic("Failed to enable signals"); + ret = enable_mask(&mask); + sigemptyset(&mask); + if((enable & (1 << SIGIO_BIT)) == 0) + sigaddset(&mask, SIGIO); + if((enable & (1 << SIGVTALRM_BIT)) == 0){ + sigaddset(&mask, SIGVTALRM); + sigaddset(&mask, SIGALRM); + } + if(sigprocmask(SIG_BLOCK, &mask, NULL) < 0) + panic("Failed to block signals"); + + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/Makefile b/arch/um/kernel/skas/Makefile new file mode 100644 index 00000000000..d37d1bfcd6f --- /dev/null +++ b/arch/um/kernel/skas/Makefile @@ -0,0 +1,13 @@ +# +# Copyright (C) 2002 - 2004 Jeff Dike (jdike@addtoit.com) +# Licensed under the GPL +# + +obj-y := exec_kern.o mem.o mem_user.o mmu.o process.o process_kern.o \ + syscall_kern.o syscall_user.o time.o tlb.o trap_user.o uaccess.o \ + +subdir- := util + +USER_OBJS := process.o time.o + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/skas/exec_kern.c b/arch/um/kernel/skas/exec_kern.c new file mode 100644 index 00000000000..c6b4d5dba78 --- /dev/null +++ b/arch/um/kernel/skas/exec_kern.c @@ -0,0 +1,41 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "asm/current.h" +#include "asm/page.h" +#include "asm/signal.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "asm/mmu_context.h" +#include "tlb.h" +#include "skas.h" +#include "um_mmu.h" +#include "os.h" + +void flush_thread_skas(void) +{ + force_flush_all(); + switch_mm_skas(current->mm->context.skas.mm_fd); +} + +void start_thread_skas(struct pt_regs *regs, unsigned long eip, + unsigned long esp) +{ + set_fs(USER_DS); + PT_REGS_IP(regs) = eip; + PT_REGS_SP(regs) = esp; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/mmu-skas.h b/arch/um/kernel/skas/include/mmu-skas.h new file mode 100644 index 00000000000..4cd60d7213f --- /dev/null +++ b/arch/um/kernel/skas/include/mmu-skas.h @@ -0,0 +1,24 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_MMU_H +#define __SKAS_MMU_H + +struct mmu_context_skas { + int mm_fd; +}; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/mode-skas.h b/arch/um/kernel/skas/include/mode-skas.h new file mode 100644 index 00000000000..c1e33bd788d --- /dev/null +++ b/arch/um/kernel/skas/include/mode-skas.h @@ -0,0 +1,34 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MODE_SKAS_H__ +#define __MODE_SKAS_H__ + +#include <sysdep/ptrace.h> + +extern unsigned long exec_regs[]; +extern unsigned long exec_fp_regs[]; +extern unsigned long exec_fpx_regs[]; +extern int have_fpx_regs; + +extern void user_time_init_skas(void); +extern void sig_handler_common_skas(int sig, void *sc_ptr); +extern void halt_skas(void); +extern void reboot_skas(void); +extern void kill_off_processes_skas(void); +extern int is_skas_winch(int pid, int fd, void *data); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/mode_kern-skas.h b/arch/um/kernel/skas/include/mode_kern-skas.h new file mode 100644 index 00000000000..94c56496237 --- /dev/null +++ b/arch/um/kernel/skas/include/mode_kern-skas.h @@ -0,0 +1,53 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_MODE_KERN_H__ +#define __SKAS_MODE_KERN_H__ + +#include "linux/sched.h" +#include "asm/page.h" +#include "asm/ptrace.h" + +extern void flush_thread_skas(void); +extern void *switch_to_skas(void *prev, void *next); +extern void start_thread_skas(struct pt_regs *regs, unsigned long eip, + unsigned long esp); +extern int copy_thread_skas(int nr, unsigned long clone_flags, + unsigned long sp, unsigned long stack_top, + struct task_struct *p, struct pt_regs *regs); +extern void release_thread_skas(struct task_struct *task); +extern void exit_thread_skas(void); +extern void initial_thread_cb_skas(void (*proc)(void *), void *arg); +extern void init_idle_skas(void); +extern void flush_tlb_kernel_range_skas(unsigned long start, + unsigned long end); +extern void flush_tlb_kernel_vm_skas(void); +extern void __flush_tlb_one_skas(unsigned long addr); +extern void flush_tlb_range_skas(struct vm_area_struct *vma, + unsigned long start, unsigned long end); +extern void flush_tlb_mm_skas(struct mm_struct *mm); +extern void force_flush_all_skas(void); +extern long execute_syscall_skas(void *r); +extern void before_mem_skas(unsigned long unused); +extern unsigned long set_task_sizes_skas(int arg, unsigned long *host_size_out, + unsigned long *task_size_out); +extern int start_uml_skas(void); +extern int external_pid_skas(struct task_struct *task); +extern int thread_pid_skas(struct task_struct *task); + +#define kmem_end_skas (host_task_size - 1024 * 1024) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/proc_mm.h b/arch/um/kernel/skas/include/proc_mm.h new file mode 100644 index 00000000000..cce61a67905 --- /dev/null +++ b/arch/um/kernel/skas/include/proc_mm.h @@ -0,0 +1,55 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_PROC_MM_H +#define __SKAS_PROC_MM_H + +#define MM_MMAP 54 +#define MM_MUNMAP 55 +#define MM_MPROTECT 56 +#define MM_COPY_SEGMENTS 57 + +struct mm_mmap { + unsigned long addr; + unsigned long len; + unsigned long prot; + unsigned long flags; + unsigned long fd; + unsigned long offset; +}; + +struct mm_munmap { + unsigned long addr; + unsigned long len; +}; + +struct mm_mprotect { + unsigned long addr; + unsigned long len; + unsigned int prot; +}; + +struct proc_mm_op { + int op; + union { + struct mm_mmap mmap; + struct mm_munmap munmap; + struct mm_mprotect mprotect; + int copy_segments; + } u; +}; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/skas.h b/arch/um/kernel/skas/include/skas.h new file mode 100644 index 00000000000..f0702c2c720 --- /dev/null +++ b/arch/um/kernel/skas/include/skas.h @@ -0,0 +1,46 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_H +#define __SKAS_H + +#include "sysdep/ptrace.h" + +extern int userspace_pid[]; + +extern void switch_threads(void *me, void *next); +extern void thread_wait(void *sw, void *fb); +extern void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, + void (*handler)(int)); +extern int start_idle_thread(void *stack, void *switch_buf_ptr, + void **fork_buf_ptr); +extern int user_thread(unsigned long stack, int flags); +extern void userspace(union uml_pt_regs *regs); +extern void new_thread_proc(void *stack, void (*handler)(int sig)); +extern void remove_sigstack(void); +extern void new_thread_handler(int sig); +extern void handle_syscall(union uml_pt_regs *regs); +extern void map(int fd, unsigned long virt, unsigned long len, int r, int w, + int x, int phys_fd, unsigned long long offset); +extern int unmap(int fd, void *addr, unsigned long len); +extern int protect(int fd, unsigned long addr, unsigned long len, + int r, int w, int x); +extern void user_signal(int sig, union uml_pt_regs *regs); +extern int new_mm(int from); +extern void start_userspace(int cpu); +extern long execute_syscall_skas(void *r); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/include/uaccess-skas.h b/arch/um/kernel/skas/include/uaccess-skas.h new file mode 100644 index 00000000000..11986c9b9dd --- /dev/null +++ b/arch/um/kernel/skas/include/uaccess-skas.h @@ -0,0 +1,45 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __SKAS_UACCESS_H +#define __SKAS_UACCESS_H + +#include "asm/errno.h" +#include "asm/fixmap.h" + +#define access_ok_skas(type, addr, size) \ + ((segment_eq(get_fs(), KERNEL_DS)) || \ + (((unsigned long) (addr) < TASK_SIZE) && \ + ((unsigned long) (addr) + (size) <= TASK_SIZE)) || \ + ((type == VERIFY_READ ) && \ + ((unsigned long) (addr) >= FIXADDR_USER_START) && \ + ((unsigned long) (addr) + (size) <= FIXADDR_USER_END) && \ + ((unsigned long) (addr) + (size) >= (unsigned long)(addr)))) + +static inline int verify_area_skas(int type, const void * addr, + unsigned long size) +{ + return(access_ok_skas(type, addr, size) ? 0 : -EFAULT); +} + +extern int copy_from_user_skas(void *to, const void *from, int n); +extern int copy_to_user_skas(void *to, const void *from, int n); +extern int strncpy_from_user_skas(char *dst, const char *src, int count); +extern int __clear_user_skas(void *mem, int len); +extern int clear_user_skas(void *mem, int len); +extern int strnlen_user_skas(const void *str, int len); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/mem.c b/arch/um/kernel/skas/mem.c new file mode 100644 index 00000000000..438db2f4345 --- /dev/null +++ b/arch/um/kernel/skas/mem.c @@ -0,0 +1,35 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/mm.h" +#include "mem_user.h" + +unsigned long set_task_sizes_skas(int arg, unsigned long *host_size_out, + unsigned long *task_size_out) +{ + /* Round up to the nearest 4M */ + unsigned long top = ROUND_4M((unsigned long) &arg); + +#ifdef CONFIG_HOST_TASK_SIZE + *host_size_out = CONFIG_HOST_TASK_SIZE; + *task_size_out = CONFIG_HOST_TASK_SIZE; +#else + *host_size_out = top; + *task_size_out = top; +#endif + return(((unsigned long) set_task_sizes_skas) & ~0xffffff); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/mem_user.c b/arch/um/kernel/skas/mem_user.c new file mode 100644 index 00000000000..1310bf1e88d --- /dev/null +++ b/arch/um/kernel/skas/mem_user.c @@ -0,0 +1,102 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <errno.h> +#include <sys/mman.h> +#include "mem_user.h" +#include "mem.h" +#include "user.h" +#include "os.h" +#include "proc_mm.h" + +void map(int fd, unsigned long virt, unsigned long len, int r, int w, + int x, int phys_fd, unsigned long long offset) +{ + struct proc_mm_op map; + int prot, n; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + + map = ((struct proc_mm_op) { .op = MM_MMAP, + .u = + { .mmap = + { .addr = virt, + .len = len, + .prot = prot, + .flags = MAP_SHARED | + MAP_FIXED, + .fd = phys_fd, + .offset = offset + } } } ); + n = os_write_file(fd, &map, sizeof(map)); + if(n != sizeof(map)) + printk("map : /proc/mm map failed, err = %d\n", -n); +} + +int unmap(int fd, void *addr, unsigned long len) +{ + struct proc_mm_op unmap; + int n; + + unmap = ((struct proc_mm_op) { .op = MM_MUNMAP, + .u = + { .munmap = + { .addr = (unsigned long) addr, + .len = len } } } ); + n = os_write_file(fd, &unmap, sizeof(unmap)); + if(n != sizeof(unmap)) { + if(n < 0) + return(n); + else if(n > 0) + return(-EIO); + } + + return(0); +} + +int protect(int fd, unsigned long addr, unsigned long len, int r, int w, + int x, int must_succeed) +{ + struct proc_mm_op protect; + int prot, n; + + prot = (r ? PROT_READ : 0) | (w ? PROT_WRITE : 0) | + (x ? PROT_EXEC : 0); + + protect = ((struct proc_mm_op) { .op = MM_MPROTECT, + .u = + { .mprotect = + { .addr = (unsigned long) addr, + .len = len, + .prot = prot } } } ); + + n = os_write_file(fd, &protect, sizeof(protect)); + if(n != sizeof(protect)) { + if(n == 0) return(0); + + if(must_succeed) + panic("protect failed, err = %d", -n); + + return(-EIO); + } + + return(0); +} + +void before_mem_skas(unsigned long unused) +{ +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c new file mode 100644 index 00000000000..6cb9a6d028a --- /dev/null +++ b/arch/um/kernel/skas/mmu.c @@ -0,0 +1,48 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/list.h" +#include "linux/spinlock.h" +#include "linux/slab.h" +#include "asm/current.h" +#include "asm/segment.h" +#include "asm/mmu.h" +#include "os.h" +#include "skas.h" + +int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) +{ + int from; + + if((current->mm != NULL) && (current->mm != &init_mm)) + from = current->mm->context.skas.mm_fd; + else from = -1; + + mm->context.skas.mm_fd = new_mm(from); + if(mm->context.skas.mm_fd < 0){ + printk("init_new_context_skas - new_mm failed, errno = %d\n", + mm->context.skas.mm_fd); + return(mm->context.skas.mm_fd); + } + + return(0); +} + +void destroy_context_skas(struct mm_struct *mm) +{ + os_close_file(mm->context.skas.mm_fd); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/process.c b/arch/um/kernel/skas/process.c new file mode 100644 index 00000000000..b4ffaaa8124 --- /dev/null +++ b/arch/um/kernel/skas/process.c @@ -0,0 +1,339 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <unistd.h> +#include <errno.h> +#include <signal.h> +#include <setjmp.h> +#include <sched.h> +#include <sys/wait.h> +#include <sys/mman.h> +#include <sys/user.h> +#include <asm/unistd.h> +#include "user.h" +#include "ptrace_user.h" +#include "time_user.h" +#include "sysdep/ptrace.h" +#include "user_util.h" +#include "kern_util.h" +#include "skas.h" +#include "sysdep/sigcontext.h" +#include "os.h" +#include "proc_mm.h" +#include "skas_ptrace.h" +#include "chan_user.h" +#include "signal_user.h" +#include "registers.h" + +int is_skas_winch(int pid, int fd, void *data) +{ + if(pid != os_getpid()) + return(0); + + register_winch_irq(-1, fd, -1, data); + return(1); +} + +static void handle_segv(int pid) +{ + struct ptrace_faultinfo fault; + int err; + + err = ptrace(PTRACE_FAULTINFO, pid, 0, &fault); + if(err) + panic("handle_segv - PTRACE_FAULTINFO failed, errno = %d\n", + errno); + + segv(fault.addr, 0, FAULT_WRITE(fault.is_write), 1, NULL); +} + +/*To use the same value of using_sysemu as the caller, ask it that value (in local_using_sysemu)*/ +static void handle_trap(int pid, union uml_pt_regs *regs, int local_using_sysemu) +{ + int err, status; + + /* Mark this as a syscall */ + UPT_SYSCALL_NR(regs) = PT_SYSCALL_NR(regs->skas.regs); + + if (!local_using_sysemu) + { + err = ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, __NR_getpid); + if(err < 0) + panic("handle_trap - nullifying syscall failed errno = %d\n", + errno); + + err = ptrace(PTRACE_SYSCALL, pid, 0, 0); + if(err < 0) + panic("handle_trap - continuing to end of syscall failed, " + "errno = %d\n", errno); + + CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED)); + if((err < 0) || !WIFSTOPPED(status) || + (WSTOPSIG(status) != SIGTRAP + 0x80)) + panic("handle_trap - failed to wait at end of syscall, " + "errno = %d, status = %d\n", errno, status); + } + + handle_syscall(regs); +} + +static int userspace_tramp(void *arg) +{ + init_new_thread_signals(0); + enable_timer(); + ptrace(PTRACE_TRACEME, 0, 0, 0); + os_stop_process(os_getpid()); + return(0); +} + +/* Each element set once, and only accessed by a single processor anyway */ +#undef NR_CPUS +#define NR_CPUS 1 +int userspace_pid[NR_CPUS]; + +void start_userspace(int cpu) +{ + void *stack; + unsigned long sp; + int pid, status, n; + + stack = mmap(NULL, PAGE_SIZE, PROT_READ | PROT_WRITE | PROT_EXEC, + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); + if(stack == MAP_FAILED) + panic("start_userspace : mmap failed, errno = %d", errno); + sp = (unsigned long) stack + PAGE_SIZE - sizeof(void *); + + pid = clone(userspace_tramp, (void *) sp, + CLONE_FILES | CLONE_VM | SIGCHLD, NULL); + if(pid < 0) + panic("start_userspace : clone failed, errno = %d", errno); + + do { + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0) + panic("start_userspace : wait failed, errno = %d", + errno); + } while(WIFSTOPPED(status) && (WSTOPSIG(status) == SIGVTALRM)); + + if(!WIFSTOPPED(status) || (WSTOPSIG(status) != SIGSTOP)) + panic("start_userspace : expected SIGSTOP, got status = %d", + status); + + if (ptrace(PTRACE_OLDSETOPTIONS, pid, NULL, (void *)PTRACE_O_TRACESYSGOOD) < 0) + panic("start_userspace : PTRACE_SETOPTIONS failed, errno=%d\n", + errno); + + if(munmap(stack, PAGE_SIZE) < 0) + panic("start_userspace : munmap failed, errno = %d\n", errno); + + userspace_pid[cpu] = pid; +} + +void userspace(union uml_pt_regs *regs) +{ + int err, status, op, pid = userspace_pid[0]; + int local_using_sysemu; /*To prevent races if using_sysemu changes under us.*/ + + while(1){ + restore_registers(pid, regs); + + /* Now we set local_using_sysemu to be used for one loop */ + local_using_sysemu = get_using_sysemu(); + + op = SELECT_PTRACE_OPERATION(local_using_sysemu, singlestepping(NULL)); + + err = ptrace(op, pid, 0, 0); + if(err) + panic("userspace - could not resume userspace process, " + "pid=%d, ptrace operation = %d, errno = %d\n", + op, errno); + + CATCH_EINTR(err = waitpid(pid, &status, WUNTRACED)); + if(err < 0) + panic("userspace - waitpid failed, errno = %d\n", + errno); + + regs->skas.is_user = 1; + save_registers(pid, regs); + UPT_SYSCALL_NR(regs) = -1; /* Assume: It's not a syscall */ + + if(WIFSTOPPED(status)){ + switch(WSTOPSIG(status)){ + case SIGSEGV: + handle_segv(pid); + break; + case SIGTRAP + 0x80: + handle_trap(pid, regs, local_using_sysemu); + break; + case SIGTRAP: + relay_signal(SIGTRAP, regs); + break; + case SIGIO: + case SIGVTALRM: + case SIGILL: + case SIGBUS: + case SIGFPE: + case SIGWINCH: + user_signal(WSTOPSIG(status), regs); + break; + default: + printk("userspace - child stopped with signal " + "%d\n", WSTOPSIG(status)); + } + interrupt_end(); + + /* Avoid -ERESTARTSYS handling in host */ + PT_SYSCALL_NR(regs->skas.regs) = -1; + } + } +} + +void new_thread(void *stack, void **switch_buf_ptr, void **fork_buf_ptr, + void (*handler)(int)) +{ + unsigned long flags; + sigjmp_buf switch_buf, fork_buf; + + *switch_buf_ptr = &switch_buf; + *fork_buf_ptr = &fork_buf; + + /* Somewhat subtle - siglongjmp restores the signal mask before doing + * the longjmp. This means that when jumping from one stack to another + * when the target stack has interrupts enabled, an interrupt may occur + * on the source stack. This is bad when starting up a process because + * it's not supposed to get timer ticks until it has been scheduled. + * So, we disable interrupts around the sigsetjmp to ensure that + * they can't happen until we get back here where they are safe. + */ + flags = get_signals(); + block_signals(); + if(sigsetjmp(fork_buf, 1) == 0) + new_thread_proc(stack, handler); + + remove_sigstack(); + + set_signals(flags); +} + +void thread_wait(void *sw, void *fb) +{ + sigjmp_buf buf, **switch_buf = sw, *fork_buf; + + *switch_buf = &buf; + fork_buf = fb; + if(sigsetjmp(buf, 1) == 0) + siglongjmp(*fork_buf, 1); +} + +void switch_threads(void *me, void *next) +{ + sigjmp_buf my_buf, **me_ptr = me, *next_buf = next; + + *me_ptr = &my_buf; + if(sigsetjmp(my_buf, 1) == 0) + siglongjmp(*next_buf, 1); +} + +static sigjmp_buf initial_jmpbuf; + +/* XXX Make these percpu */ +static void (*cb_proc)(void *arg); +static void *cb_arg; +static sigjmp_buf *cb_back; + +int start_idle_thread(void *stack, void *switch_buf_ptr, void **fork_buf_ptr) +{ + sigjmp_buf **switch_buf = switch_buf_ptr; + int n; + + *fork_buf_ptr = &initial_jmpbuf; + n = sigsetjmp(initial_jmpbuf, 1); + if(n == 0) + new_thread_proc((void *) stack, new_thread_handler); + else if(n == 1) + remove_sigstack(); + else if(n == 2){ + (*cb_proc)(cb_arg); + siglongjmp(*cb_back, 1); + } + else if(n == 3){ + kmalloc_ok = 0; + return(0); + } + else if(n == 4){ + kmalloc_ok = 0; + return(1); + } + siglongjmp(**switch_buf, 1); +} + +void remove_sigstack(void) +{ + stack_t stack = ((stack_t) { .ss_flags = SS_DISABLE, + .ss_sp = NULL, + .ss_size = 0 }); + + if(sigaltstack(&stack, NULL) != 0) + panic("disabling signal stack failed, errno = %d\n", errno); +} + +void initial_thread_cb_skas(void (*proc)(void *), void *arg) +{ + sigjmp_buf here; + + cb_proc = proc; + cb_arg = arg; + cb_back = &here; + + block_signals(); + if(sigsetjmp(here, 1) == 0) + siglongjmp(initial_jmpbuf, 2); + unblock_signals(); + + cb_proc = NULL; + cb_arg = NULL; + cb_back = NULL; +} + +void halt_skas(void) +{ + block_signals(); + siglongjmp(initial_jmpbuf, 3); +} + +void reboot_skas(void) +{ + block_signals(); + siglongjmp(initial_jmpbuf, 4); +} + +void switch_mm_skas(int mm_fd) +{ + int err; + +#warning need cpu pid in switch_mm_skas + err = ptrace(PTRACE_SWITCH_MM, userspace_pid[0], 0, mm_fd); + if(err) + panic("switch_mm_skas - PTRACE_SWITCH_MM failed, errno = %d\n", + errno); +} + +void kill_off_processes_skas(void) +{ +#warning need to loop over userspace_pids in kill_off_processes_skas + os_kill_ptraced_process(userspace_pid[0], 1); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/process_kern.c b/arch/um/kernel/skas/process_kern.c new file mode 100644 index 00000000000..5d096ea63b9 --- /dev/null +++ b/arch/um/kernel/skas/process_kern.c @@ -0,0 +1,213 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/slab.h" +#include "linux/ptrace.h" +#include "linux/proc_fs.h" +#include "linux/file.h" +#include "linux/errno.h" +#include "linux/init.h" +#include "asm/uaccess.h" +#include "asm/atomic.h" +#include "kern_util.h" +#include "time_user.h" +#include "signal_user.h" +#include "skas.h" +#include "os.h" +#include "user_util.h" +#include "tlb.h" +#include "kern.h" +#include "mode.h" +#include "proc_mm.h" +#include "registers.h" + +void *switch_to_skas(void *prev, void *next) +{ + struct task_struct *from, *to; + + from = prev; + to = next; + + /* XXX need to check runqueues[cpu].idle */ + if(current->pid == 0) + switch_timers(0); + + to->thread.prev_sched = from; + set_current(to); + + switch_threads(&from->thread.mode.skas.switch_buf, + to->thread.mode.skas.switch_buf); + + if(current->pid == 0) + switch_timers(1); + + return(current->thread.prev_sched); +} + +extern void schedule_tail(struct task_struct *prev); + +void new_thread_handler(int sig) +{ + int (*fn)(void *), n; + void *arg; + + fn = current->thread.request.u.thread.proc; + arg = current->thread.request.u.thread.arg; + change_sig(SIGUSR1, 1); + thread_wait(¤t->thread.mode.skas.switch_buf, + current->thread.mode.skas.fork_buf); + + if(current->thread.prev_sched != NULL) + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; + + /* The return value is 1 if the kernel thread execs a process, + * 0 if it just exits + */ + n = run_kernel_thread(fn, arg, ¤t->thread.exec_buf); + if(n == 1) + userspace(¤t->thread.regs.regs); + else do_exit(0); +} + +void new_thread_proc(void *stack, void (*handler)(int sig)) +{ + init_new_thread_stack(stack, handler); + os_usr1_process(os_getpid()); +} + +void release_thread_skas(struct task_struct *task) +{ +} + +void exit_thread_skas(void) +{ +} + +void fork_handler(int sig) +{ + change_sig(SIGUSR1, 1); + thread_wait(¤t->thread.mode.skas.switch_buf, + current->thread.mode.skas.fork_buf); + + force_flush_all(); + if(current->thread.prev_sched == NULL) + panic("blech"); + + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; + + userspace(¤t->thread.regs.regs); +} + +int copy_thread_skas(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct * p, + struct pt_regs *regs) +{ + void (*handler)(int); + + if(current->thread.forking){ + memcpy(&p->thread.regs.regs.skas, + ¤t->thread.regs.regs.skas, + sizeof(p->thread.regs.regs.skas)); + REGS_SET_SYSCALL_RETURN(p->thread.regs.regs.skas.regs, 0); + if(sp != 0) REGS_SP(p->thread.regs.regs.skas.regs) = sp; + + handler = fork_handler; + } + else { + init_thread_registers(&p->thread.regs.regs); + p->thread.request.u.thread = current->thread.request.u.thread; + handler = new_thread_handler; + } + + new_thread(p->thread_info, &p->thread.mode.skas.switch_buf, + &p->thread.mode.skas.fork_buf, handler); + return(0); +} + +int new_mm(int from) +{ + struct proc_mm_op copy; + int n, fd; + + fd = os_open_file("/proc/mm", of_cloexec(of_write(OPENFLAGS())), 0); + if(fd < 0) + return(fd); + + if(from != -1){ + copy = ((struct proc_mm_op) { .op = MM_COPY_SEGMENTS, + .u = + { .copy_segments = from } } ); + n = os_write_file(fd, ©, sizeof(copy)); + if(n != sizeof(copy)) + printk("new_mm : /proc/mm copy_segments failed, " + "err = %d\n", -n); + } + + return(fd); +} + +void init_idle_skas(void) +{ + cpu_tasks[current_thread->cpu].pid = os_getpid(); + default_idle(); +} + +extern void start_kernel(void); + +static int start_kernel_proc(void *unused) +{ + int pid; + + block_signals(); + pid = os_getpid(); + + cpu_tasks[0].pid = pid; + cpu_tasks[0].task = current; +#ifdef CONFIG_SMP + cpu_online_map = cpumask_of_cpu(0); +#endif + start_kernel(); + return(0); +} + +int start_uml_skas(void) +{ + start_userspace(0); + + init_new_thread_signals(1); + uml_idle_timer(); + + init_task.thread.request.u.thread.proc = start_kernel_proc; + init_task.thread.request.u.thread.arg = NULL; + return(start_idle_thread(init_task.thread_info, + &init_task.thread.mode.skas.switch_buf, + &init_task.thread.mode.skas.fork_buf)); +} + +int external_pid_skas(struct task_struct *task) +{ +#warning Need to look up userspace_pid by cpu + return(userspace_pid[0]); +} + +int thread_pid_skas(struct task_struct *task) +{ +#warning Need to look up userspace_pid by cpu + return(userspace_pid[0]); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/syscall_kern.c b/arch/um/kernel/skas/syscall_kern.c new file mode 100644 index 00000000000..bdf040ce5b8 --- /dev/null +++ b/arch/um/kernel/skas/syscall_kern.c @@ -0,0 +1,43 @@ +/* + * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/sys.h" +#include "linux/ptrace.h" +#include "asm/errno.h" +#include "asm/unistd.h" +#include "asm/ptrace.h" +#include "asm/current.h" +#include "sysdep/syscalls.h" +#include "kern_util.h" + +extern syscall_handler_t *sys_call_table[]; + +long execute_syscall_skas(void *r) +{ + struct pt_regs *regs = r; + long res; + int syscall; + + current->thread.nsyscalls++; + nsyscalls++; + syscall = UPT_SYSCALL_NR(®s->regs); + + if((syscall >= NR_syscalls) || (syscall < 0)) + res = -ENOSYS; + else res = EXECUTE_SYSCALL(syscall, regs); + + return(res); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/syscall_user.c b/arch/um/kernel/skas/syscall_user.c new file mode 100644 index 00000000000..2828e6e3772 --- /dev/null +++ b/arch/um/kernel/skas/syscall_user.c @@ -0,0 +1,44 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <signal.h> +#include "kern_util.h" +#include "uml-config.h" +#include "syscall_user.h" +#include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" +#include "skas.h" + +void handle_syscall(union uml_pt_regs *regs) +{ + long result; +#if UML_CONFIG_SYSCALL_DEBUG + int index; + + index = record_syscall_start(UPT_SYSCALL_NR(regs)); +#endif + + syscall_trace(regs, 0); + result = execute_syscall_skas(regs); + + REGS_SET_SYSCALL_RETURN(regs->skas.regs, result); + + syscall_trace(regs, 1); +#if UML_CONFIG_SYSCALL_DEBUG + record_syscall_end(index, result); +#endif +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/time.c b/arch/um/kernel/skas/time.c new file mode 100644 index 00000000000..98091494b89 --- /dev/null +++ b/arch/um/kernel/skas/time.c @@ -0,0 +1,30 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <sys/signal.h> +#include <sys/time.h> +#include "time_user.h" +#include "process.h" +#include "user.h" + +void user_time_init_skas(void) +{ + if(signal(SIGALRM, (__sighandler_t) alarm_handler) == SIG_ERR) + panic("Couldn't set SIGALRM handler"); + if(signal(SIGVTALRM, (__sighandler_t) alarm_handler) == SIG_ERR) + panic("Couldn't set SIGVTALRM handler"); + set_interval(ITIMER_VIRTUAL); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/tlb.c b/arch/um/kernel/skas/tlb.c new file mode 100644 index 00000000000..b8c5e71763d --- /dev/null +++ b/arch/um/kernel/skas/tlb.c @@ -0,0 +1,85 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/sched.h" +#include "linux/mm.h" +#include "asm/page.h" +#include "asm/pgtable.h" +#include "asm/mmu.h" +#include "user_util.h" +#include "mem_user.h" +#include "mem.h" +#include "skas.h" +#include "os.h" +#include "tlb.h" + +static void do_ops(int fd, struct host_vm_op *ops, int last) +{ + struct host_vm_op *op; + int i; + + for(i = 0; i <= last; i++){ + op = &ops[i]; + switch(op->type){ + case MMAP: + map(fd, op->u.mmap.addr, op->u.mmap.len, + op->u.mmap.r, op->u.mmap.w, op->u.mmap.x, + op->u.mmap.fd, op->u.mmap.offset); + break; + case MUNMAP: + unmap(fd, (void *) op->u.munmap.addr, + op->u.munmap.len); + break; + case MPROTECT: + protect(fd, op->u.mprotect.addr, op->u.mprotect.len, + op->u.mprotect.r, op->u.mprotect.w, + op->u.mprotect.x); + break; + default: + printk("Unknown op type %d in do_ops\n", op->type); + break; + } + } +} + +static void fix_range(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force) +{ + int fd = mm->context.skas.mm_fd; + + fix_range_common(mm, start_addr, end_addr, force, fd, do_ops); +} + +void __flush_tlb_one_skas(unsigned long addr) +{ + flush_tlb_kernel_range_common(addr, addr + PAGE_SIZE); +} + +void flush_tlb_range_skas(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + if(vma->vm_mm == NULL) + flush_tlb_kernel_range_common(start, end); + else fix_range(vma->vm_mm, start, end, 0); +} + +void flush_tlb_mm_skas(struct mm_struct *mm) +{ + /* Don't bother flushing if this address space is about to be + * destroyed. + */ + if(atomic_read(&mm->mm_users) == 0) + return; + + fix_range(mm, 0, host_task_size, 0); + flush_tlb_kernel_range_common(start_vm, end_vm); +} + +void force_flush_all_skas(void) +{ + fix_range(current->mm, 0, host_task_size, 1); +} diff --git a/arch/um/kernel/skas/trap_user.c b/arch/um/kernel/skas/trap_user.c new file mode 100644 index 00000000000..8e9b46d4702 --- /dev/null +++ b/arch/um/kernel/skas/trap_user.c @@ -0,0 +1,71 @@ +/* + * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include <signal.h> +#include <errno.h> +#include "sysdep/ptrace.h" +#include "signal_user.h" +#include "user_util.h" +#include "kern_util.h" +#include "task.h" +#include "sigcontext.h" + +void sig_handler_common_skas(int sig, void *sc_ptr) +{ + struct sigcontext *sc = sc_ptr; + struct skas_regs *r; + struct signal_info *info; + int save_errno = errno; + int save_user; + + /* This is done because to allow SIGSEGV to be delivered inside a SEGV + * handler. This can happen in copy_user, and if SEGV is disabled, + * the process will die. + * XXX Figure out why this is better than SA_NODEFER + */ + if(sig == SIGSEGV) + change_sig(SIGSEGV, 1); + + r = &TASK_REGS(get_current())->skas; + save_user = r->is_user; + r->is_user = 0; + r->fault_addr = SC_FAULT_ADDR(sc); + r->fault_type = SC_FAULT_TYPE(sc); + r->trap_type = SC_TRAP_TYPE(sc); + + change_sig(SIGUSR1, 1); + info = &sig_info[sig]; + if(!info->is_irq) unblock_signals(); + + (*info->handler)(sig, (union uml_pt_regs *) r); + + errno = save_errno; + r->is_user = save_user; +} + +void user_signal(int sig, union uml_pt_regs *regs) +{ + struct signal_info *info; + + regs->skas.is_user = 1; + regs->skas.fault_addr = 0; + regs->skas.fault_type = 0; + regs->skas.trap_type = 0; + info = &sig_info[sig]; + (*info->handler)(sig, regs); + + unblock_signals(); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c new file mode 100644 index 00000000000..7575ec489b6 --- /dev/null +++ b/arch/um/kernel/skas/uaccess.c @@ -0,0 +1,259 @@ +/* + * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/kernel.h" +#include "linux/string.h" +#include "linux/fs.h" +#include "linux/highmem.h" +#include "asm/page.h" +#include "asm/pgtable.h" +#include "asm/uaccess.h" +#include "kern_util.h" +#include "user_util.h" + +extern void *um_virt_to_phys(struct task_struct *task, unsigned long addr, + pte_t *pte_out); + +static unsigned long maybe_map(unsigned long virt, int is_write) +{ + pte_t pte; + int err; + + void *phys = um_virt_to_phys(current, virt, &pte); + int dummy_code; + + if(IS_ERR(phys) || (is_write && !pte_write(pte))){ + err = handle_page_fault(virt, 0, is_write, 1, &dummy_code); + if(err) + return(0); + phys = um_virt_to_phys(current, virt, NULL); + } + return((unsigned long) phys); +} + +static int do_op(unsigned long addr, int len, int is_write, + int (*op)(unsigned long addr, int len, void *arg), void *arg) +{ + struct page *page; + int n; + + addr = maybe_map(addr, is_write); + if(addr == -1) + return(-1); + + page = phys_to_page(addr); + addr = (unsigned long) kmap(page) + (addr & ~PAGE_MASK); + n = (*op)(addr, len, arg); + kunmap(page); + + return(n); +} + +static void do_buffer_op(void *jmpbuf, void *arg_ptr) +{ + va_list args; + unsigned long addr; + int len, is_write, size, remain, n; + int (*op)(unsigned long, int, void *); + void *arg; + int *res; + + /* Some old gccs recognize __va_copy, but not va_copy */ + __va_copy(args, *(va_list *)arg_ptr); + addr = va_arg(args, unsigned long); + len = va_arg(args, int); + is_write = va_arg(args, int); + op = va_arg(args, void *); + arg = va_arg(args, void *); + res = va_arg(args, int *); + va_end(args); + size = min(PAGE_ALIGN(addr) - addr, (unsigned long) len); + remain = len; + + current->thread.fault_catcher = jmpbuf; + n = do_op(addr, size, is_write, op, arg); + if(n != 0){ + *res = (n < 0 ? remain : 0); + goto out; + } + + addr += size; + remain -= size; + if(remain == 0){ + *res = 0; + goto out; + } + + while(addr < ((addr + remain) & PAGE_MASK)){ + n = do_op(addr, PAGE_SIZE, is_write, op, arg); + if(n != 0){ + *res = (n < 0 ? remain : 0); + goto out; + } + + addr += PAGE_SIZE; + remain -= PAGE_SIZE; + } + if(remain == 0){ + *res = 0; + goto out; + } + + n = do_op(addr, remain, is_write, op, arg); + if(n != 0) + *res = (n < 0 ? remain : 0); + else *res = 0; + out: + current->thread.fault_catcher = NULL; +} + +static int buffer_op(unsigned long addr, int len, int is_write, + int (*op)(unsigned long addr, int len, void *arg), + void *arg) +{ + int faulted, res; + + faulted = setjmp_wrapper(do_buffer_op, addr, len, is_write, op, arg, + &res); + if(!faulted) + return(res); + + return(addr + len - (unsigned long) current->thread.fault_addr); +} + +static int copy_chunk_from_user(unsigned long from, int len, void *arg) +{ + unsigned long *to_ptr = arg, to = *to_ptr; + + memcpy((void *) to, (void *) from, len); + *to_ptr += len; + return(0); +} + +int copy_from_user_skas(void *to, const void __user *from, int n) +{ + if(segment_eq(get_fs(), KERNEL_DS)){ + memcpy(to, (__force void*)from, n); + return(0); + } + + return(access_ok_skas(VERIFY_READ, from, n) ? + buffer_op((unsigned long) from, n, 0, copy_chunk_from_user, &to): + n); +} + +static int copy_chunk_to_user(unsigned long to, int len, void *arg) +{ + unsigned long *from_ptr = arg, from = *from_ptr; + + memcpy((void *) to, (void *) from, len); + *from_ptr += len; + return(0); +} + +int copy_to_user_skas(void __user *to, const void *from, int n) +{ + if(segment_eq(get_fs(), KERNEL_DS)){ + memcpy((__force void*)to, from, n); + return(0); + } + + return(access_ok_skas(VERIFY_WRITE, to, n) ? + buffer_op((unsigned long) to, n, 1, copy_chunk_to_user, &from) : + n); +} + +static int strncpy_chunk_from_user(unsigned long from, int len, void *arg) +{ + char **to_ptr = arg, *to = *to_ptr; + int n; + + strncpy(to, (void *) from, len); + n = strnlen(to, len); + *to_ptr += n; + + if(n < len) + return(1); + return(0); +} + +int strncpy_from_user_skas(char *dst, const char __user *src, int count) +{ + int n; + char *ptr = dst; + + if(segment_eq(get_fs(), KERNEL_DS)){ + strncpy(dst, (__force void*)src, count); + return(strnlen(dst, count)); + } + + if(!access_ok_skas(VERIFY_READ, src, 1)) + return(-EFAULT); + + n = buffer_op((unsigned long) src, count, 0, strncpy_chunk_from_user, + &ptr); + if(n != 0) + return(-EFAULT); + return(strnlen(dst, count)); +} + +static int clear_chunk(unsigned long addr, int len, void *unused) +{ + memset((void *) addr, 0, len); + return(0); +} + +int __clear_user_skas(void __user *mem, int len) +{ + return(buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL)); +} + +int clear_user_skas(void __user *mem, int len) +{ + if(segment_eq(get_fs(), KERNEL_DS)){ + memset((__force void*)mem, 0, len); + return(0); + } + + return(access_ok_skas(VERIFY_WRITE, mem, len) ? + buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL) : len); +} + +static int strnlen_chunk(unsigned long str, int len, void *arg) +{ + int *len_ptr = arg, n; + + n = strnlen((void *) str, len); + *len_ptr += n; + + if(n < len) + return(1); + return(0); +} + +int strnlen_user_skas(const void __user *str, int len) +{ + int count = 0, n; + + if(segment_eq(get_fs(), KERNEL_DS)) + return(strnlen((__force char*)str, len) + 1); + + n = buffer_op((unsigned long) str, len, 0, strnlen_chunk, &count); + if(n == 0) + return(count + 1); + return(-EFAULT); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/util/Makefile b/arch/um/kernel/skas/util/Makefile new file mode 100644 index 00000000000..17f5909d60f --- /dev/null +++ b/arch/um/kernel/skas/util/Makefile @@ -0,0 +1,4 @@ +hostprogs-y := mk_ptregs +always := $(hostprogs-y) + +mk_ptregs-objs := mk_ptregs-$(SUBARCH).o diff --git a/arch/um/kernel/skas/util/mk_ptregs-i386.c b/arch/um/kernel/skas/util/mk_ptregs-i386.c new file mode 100644 index 00000000000..0788dd05bca --- /dev/null +++ b/arch/um/kernel/skas/util/mk_ptregs-i386.c @@ -0,0 +1,51 @@ +#include <stdio.h> +#include <asm/ptrace.h> +#include <asm/user.h> + +#define PRINT_REG(name, val) printf("#define HOST_%s %d\n", (name), (val)) + +int main(int argc, char **argv) +{ + printf("/* Automatically generated by " + "arch/um/kernel/skas/util/mk_ptregs */\n"); + printf("\n"); + printf("#ifndef __SKAS_PT_REGS_\n"); + printf("#define __SKAS_PT_REGS_\n"); + printf("\n"); + printf("#define HOST_FRAME_SIZE %d\n", FRAME_SIZE); + printf("#define HOST_FP_SIZE %d\n", + sizeof(struct user_i387_struct) / sizeof(unsigned long)); + printf("#define HOST_XFP_SIZE %d\n", + sizeof(struct user_fxsr_struct) / sizeof(unsigned long)); + + PRINT_REG("IP", EIP); + PRINT_REG("SP", UESP); + PRINT_REG("EFLAGS", EFL); + PRINT_REG("EAX", EAX); + PRINT_REG("EBX", EBX); + PRINT_REG("ECX", ECX); + PRINT_REG("EDX", EDX); + PRINT_REG("ESI", ESI); + PRINT_REG("EDI", EDI); + PRINT_REG("EBP", EBP); + PRINT_REG("CS", CS); + PRINT_REG("SS", SS); + PRINT_REG("DS", DS); + PRINT_REG("FS", FS); + PRINT_REG("ES", ES); + PRINT_REG("GS", GS); + printf("\n"); + printf("#endif\n"); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/skas/util/mk_ptregs-x86_64.c b/arch/um/kernel/skas/util/mk_ptregs-x86_64.c new file mode 100644 index 00000000000..67aee92a70e --- /dev/null +++ b/arch/um/kernel/skas/util/mk_ptregs-x86_64.c @@ -0,0 +1,68 @@ +/* + * Copyright 2003 PathScale, Inc. + * + * Licensed under the GPL + */ + +#include <stdio.h> +#define __FRAME_OFFSETS +#include <asm/ptrace.h> + +#define PRINT_REG(name, val) \ + printf("#define HOST_%s (%d / sizeof(unsigned long))\n", (name), (val)) + +int main(int argc, char **argv) +{ + printf("/* Automatically generated by " + "arch/um/kernel/skas/util/mk_ptregs */\n"); + printf("\n"); + printf("#ifndef __SKAS_PT_REGS_\n"); + printf("#define __SKAS_PT_REGS_\n"); + printf("#define HOST_FRAME_SIZE (%d / sizeof(unsigned long))\n", + FRAME_SIZE); + PRINT_REG("RBX", RBX); + PRINT_REG("RCX", RCX); + PRINT_REG("RDI", RDI); + PRINT_REG("RSI", RSI); + PRINT_REG("RDX", RDX); + PRINT_REG("RBP", RBP); + PRINT_REG("RAX", RAX); + PRINT_REG("R8", R8); + PRINT_REG("R9", R9); + PRINT_REG("R10", R10); + PRINT_REG("R11", R11); + PRINT_REG("R12", R12); + PRINT_REG("R13", R13); + PRINT_REG("R14", R14); + PRINT_REG("R15", R15); + PRINT_REG("ORIG_RAX", ORIG_RAX); + PRINT_REG("CS", CS); + PRINT_REG("SS", SS); + PRINT_REG("EFLAGS", EFLAGS); +#if 0 + PRINT_REG("FS", FS); + PRINT_REG("GS", GS); + PRINT_REG("DS", DS); + PRINT_REG("ES", ES); +#endif + + PRINT_REG("IP", RIP); + PRINT_REG("SP", RSP); + printf("#define HOST_FP_SIZE 0\n"); + printf("#define HOST_XFP_SIZE 0\n"); + printf("\n"); + printf("\n"); + printf("#endif\n"); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/smp.c b/arch/um/kernel/smp.c new file mode 100644 index 00000000000..72113b0a96e --- /dev/null +++ b/arch/um/kernel/smp.c @@ -0,0 +1,269 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/percpu.h" +#include "asm/pgalloc.h" +#include "asm/tlb.h" + +/* For some reason, mmu_gathers are referenced when CONFIG_SMP is off. */ +DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); + +#ifdef CONFIG_SMP + +#include "linux/sched.h" +#include "linux/module.h" +#include "linux/threads.h" +#include "linux/interrupt.h" +#include "linux/err.h" +#include "linux/hardirq.h" +#include "asm/smp.h" +#include "asm/processor.h" +#include "asm/spinlock.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "irq_user.h" +#include "os.h" + +/* CPU online map, set by smp_boot_cpus */ +cpumask_t cpu_online_map = CPU_MASK_NONE; +cpumask_t cpu_possible_map = CPU_MASK_NONE; + +EXPORT_SYMBOL(cpu_online_map); +EXPORT_SYMBOL(cpu_possible_map); + +/* Per CPU bogomips and other parameters + * The only piece used here is the ipi pipe, which is set before SMP is + * started and never changed. + */ +struct cpuinfo_um cpu_data[NR_CPUS]; + +/* A statistic, can be a little off */ +int num_reschedules_sent = 0; + +/* Not changed after boot */ +struct task_struct *idle_threads[NR_CPUS]; + +void smp_send_reschedule(int cpu) +{ + os_write_file(cpu_data[cpu].ipi_pipe[1], "R", 1); + num_reschedules_sent++; +} + +void smp_send_stop(void) +{ + int i; + + printk(KERN_INFO "Stopping all CPUs..."); + for(i = 0; i < num_online_cpus(); i++){ + if(i == current_thread->cpu) + continue; + os_write_file(cpu_data[i].ipi_pipe[1], "S", 1); + } + printk("done\n"); +} + +static cpumask_t smp_commenced_mask = CPU_MASK_NONE; +static cpumask_t cpu_callin_map = CPU_MASK_NONE; + +static int idle_proc(void *cpup) +{ + int cpu = (int) cpup, err; + + err = os_pipe(cpu_data[cpu].ipi_pipe, 1, 1); + if(err < 0) + panic("CPU#%d failed to create IPI pipe, err = %d", cpu, -err); + + activate_ipi(cpu_data[cpu].ipi_pipe[0], + current->thread.mode.tt.extern_pid); + + wmb(); + if (cpu_test_and_set(cpu, cpu_callin_map)) { + printk("huh, CPU#%d already present??\n", cpu); + BUG(); + } + + while (!cpu_isset(cpu, smp_commenced_mask)) + cpu_relax(); + + cpu_set(cpu, cpu_online_map); + default_idle(); + return(0); +} + +static struct task_struct *idle_thread(int cpu) +{ + struct task_struct *new_task; + unsigned char c; + + current->thread.request.u.thread.proc = idle_proc; + current->thread.request.u.thread.arg = (void *) cpu; + new_task = fork_idle(cpu); + if(IS_ERR(new_task)) + panic("copy_process failed in idle_thread, error = %ld", + PTR_ERR(new_task)); + + cpu_tasks[cpu] = ((struct cpu_task) + { .pid = new_task->thread.mode.tt.extern_pid, + .task = new_task } ); + idle_threads[cpu] = new_task; + CHOOSE_MODE(os_write_file(new_task->thread.mode.tt.switch_pipe[1], &c, + sizeof(c)), + ({ panic("skas mode doesn't support SMP"); })); + return(new_task); +} + +void smp_prepare_cpus(unsigned int maxcpus) +{ + struct task_struct *idle; + unsigned long waittime; + int err, cpu, me = smp_processor_id(); + int i; + + for (i = 0; i < ncpus; ++i) + cpu_set(i, cpu_possible_map); + + cpu_clear(me, cpu_online_map); + cpu_set(me, cpu_online_map); + cpu_set(me, cpu_callin_map); + + err = os_pipe(cpu_data[me].ipi_pipe, 1, 1); + if(err < 0) + panic("CPU#0 failed to create IPI pipe, errno = %d", -err); + + activate_ipi(cpu_data[me].ipi_pipe[0], + current->thread.mode.tt.extern_pid); + + for(cpu = 1; cpu < ncpus; cpu++){ + printk("Booting processor %d...\n", cpu); + + idle = idle_thread(cpu); + + init_idle(idle, cpu); + unhash_process(idle); + + waittime = 200000000; + while (waittime-- && !cpu_isset(cpu, cpu_callin_map)) + cpu_relax(); + + if (cpu_isset(cpu, cpu_callin_map)) + printk("done\n"); + else printk("failed\n"); + } +} + +void smp_prepare_boot_cpu(void) +{ + cpu_set(smp_processor_id(), cpu_online_map); +} + +int __cpu_up(unsigned int cpu) +{ + cpu_set(cpu, smp_commenced_mask); + while (!cpu_isset(cpu, cpu_online_map)) + mb(); + return(0); +} + +int setup_profiling_timer(unsigned int multiplier) +{ + printk(KERN_INFO "setup_profiling_timer\n"); + return(0); +} + +void smp_call_function_slave(int cpu); + +void IPI_handler(int cpu) +{ + unsigned char c; + int fd; + + fd = cpu_data[cpu].ipi_pipe[0]; + while (os_read_file(fd, &c, 1) == 1) { + switch (c) { + case 'C': + smp_call_function_slave(cpu); + break; + + case 'R': + set_tsk_need_resched(current); + break; + + case 'S': + printk("CPU#%d stopping\n", cpu); + while(1) + pause(); + break; + + default: + printk("CPU#%d received unknown IPI [%c]!\n", cpu, c); + break; + } + } +} + +int hard_smp_processor_id(void) +{ + return(pid_to_processor_id(os_getpid())); +} + +static DEFINE_SPINLOCK(call_lock); +static atomic_t scf_started; +static atomic_t scf_finished; +static void (*func)(void *info); +static void *info; + +void smp_call_function_slave(int cpu) +{ + atomic_inc(&scf_started); + (*func)(info); + atomic_inc(&scf_finished); +} + +int smp_call_function(void (*_func)(void *info), void *_info, int nonatomic, + int wait) +{ + int cpus = num_online_cpus() - 1; + int i; + + if (!cpus) + return 0; + + /* Can deadlock when called with interrupts disabled */ + WARN_ON(irqs_disabled()); + + spin_lock_bh(&call_lock); + atomic_set(&scf_started, 0); + atomic_set(&scf_finished, 0); + func = _func; + info = _info; + + for_each_online_cpu(i) + os_write_file(cpu_data[i].ipi_pipe[1], "C", 1); + + while (atomic_read(&scf_started) != cpus) + barrier(); + + if (wait) + while (atomic_read(&scf_finished) != cpus) + barrier(); + + spin_unlock_bh(&call_lock); + return 0; +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/sys_call_table.c b/arch/um/kernel/sys_call_table.c new file mode 100644 index 00000000000..7fc06c85b29 --- /dev/null +++ b/arch/um/kernel/sys_call_table.c @@ -0,0 +1,276 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/unistd.h" +#include "linux/sys.h" +#include "linux/swap.h" +#include "linux/syscalls.h" +#include "linux/sysctl.h" +#include "asm/signal.h" +#include "sysdep/syscalls.h" +#include "kern_util.h" + +#ifdef CONFIG_NFSD +#define NFSSERVCTL sys_nfsservctl +#else +#define NFSSERVCTL sys_ni_syscall +#endif + +#define LAST_GENERIC_SYSCALL __NR_keyctl + +#if LAST_GENERIC_SYSCALL > LAST_ARCH_SYSCALL +#define LAST_SYSCALL LAST_GENERIC_SYSCALL +#else +#define LAST_SYSCALL LAST_ARCH_SYSCALL +#endif + +extern syscall_handler_t sys_fork; +extern syscall_handler_t sys_execve; +extern syscall_handler_t um_time; +extern syscall_handler_t um_stime; +extern syscall_handler_t sys_pipe; +extern syscall_handler_t sys_olduname; +extern syscall_handler_t sys_sigaction; +extern syscall_handler_t sys_sigsuspend; +extern syscall_handler_t old_readdir; +extern syscall_handler_t sys_uname; +extern syscall_handler_t sys_ipc; +extern syscall_handler_t sys_sigreturn; +extern syscall_handler_t sys_clone; +extern syscall_handler_t sys_rt_sigreturn; +extern syscall_handler_t sys_sigaltstack; +extern syscall_handler_t sys_vfork; +extern syscall_handler_t old_select; +extern syscall_handler_t sys_modify_ldt; +extern syscall_handler_t sys_rt_sigsuspend; +extern syscall_handler_t sys_mbind; +extern syscall_handler_t sys_get_mempolicy; +extern syscall_handler_t sys_set_mempolicy; +extern syscall_handler_t sys_sys_setaltroot; + +syscall_handler_t *sys_call_table[] = { + [ __NR_restart_syscall ] = (syscall_handler_t *) sys_restart_syscall, + [ __NR_exit ] = (syscall_handler_t *) sys_exit, + [ __NR_fork ] = (syscall_handler_t *) sys_fork, + [ __NR_read ] = (syscall_handler_t *) sys_read, + [ __NR_write ] = (syscall_handler_t *) sys_write, + + /* These three are declared differently in asm/unistd.h */ + [ __NR_open ] = (syscall_handler_t *) sys_open, + [ __NR_close ] = (syscall_handler_t *) sys_close, + [ __NR_creat ] = (syscall_handler_t *) sys_creat, + [ __NR_link ] = (syscall_handler_t *) sys_link, + [ __NR_unlink ] = (syscall_handler_t *) sys_unlink, + [ __NR_execve ] = (syscall_handler_t *) sys_execve, + + /* declared differently in kern_util.h */ + [ __NR_chdir ] = (syscall_handler_t *) sys_chdir, + [ __NR_time ] = um_time, + [ __NR_mknod ] = (syscall_handler_t *) sys_mknod, + [ __NR_chmod ] = (syscall_handler_t *) sys_chmod, + [ __NR_lchown ] = (syscall_handler_t *) sys_lchown16, + [ __NR_lseek ] = (syscall_handler_t *) sys_lseek, + [ __NR_getpid ] = (syscall_handler_t *) sys_getpid, + [ __NR_mount ] = (syscall_handler_t *) sys_mount, + [ __NR_setuid ] = (syscall_handler_t *) sys_setuid16, + [ __NR_getuid ] = (syscall_handler_t *) sys_getuid16, + [ __NR_ptrace ] = (syscall_handler_t *) sys_ptrace, + [ __NR_alarm ] = (syscall_handler_t *) sys_alarm, + [ __NR_pause ] = (syscall_handler_t *) sys_pause, + [ __NR_utime ] = (syscall_handler_t *) sys_utime, + [ __NR_access ] = (syscall_handler_t *) sys_access, + [ __NR_sync ] = (syscall_handler_t *) sys_sync, + [ __NR_kill ] = (syscall_handler_t *) sys_kill, + [ __NR_rename ] = (syscall_handler_t *) sys_rename, + [ __NR_mkdir ] = (syscall_handler_t *) sys_mkdir, + [ __NR_rmdir ] = (syscall_handler_t *) sys_rmdir, + + /* Declared differently in asm/unistd.h */ + [ __NR_dup ] = (syscall_handler_t *) sys_dup, + [ __NR_pipe ] = (syscall_handler_t *) sys_pipe, + [ __NR_times ] = (syscall_handler_t *) sys_times, + [ __NR_brk ] = (syscall_handler_t *) sys_brk, + [ __NR_setgid ] = (syscall_handler_t *) sys_setgid16, + [ __NR_getgid ] = (syscall_handler_t *) sys_getgid16, + [ __NR_geteuid ] = (syscall_handler_t *) sys_geteuid16, + [ __NR_getegid ] = (syscall_handler_t *) sys_getegid16, + [ __NR_acct ] = (syscall_handler_t *) sys_acct, + [ __NR_umount2 ] = (syscall_handler_t *) sys_umount, + [ __NR_ioctl ] = (syscall_handler_t *) sys_ioctl, + [ __NR_fcntl ] = (syscall_handler_t *) sys_fcntl, + [ __NR_setpgid ] = (syscall_handler_t *) sys_setpgid, + [ __NR_umask ] = (syscall_handler_t *) sys_umask, + [ __NR_chroot ] = (syscall_handler_t *) sys_chroot, + [ __NR_ustat ] = (syscall_handler_t *) sys_ustat, + [ __NR_dup2 ] = (syscall_handler_t *) sys_dup2, + [ __NR_getppid ] = (syscall_handler_t *) sys_getppid, + [ __NR_getpgrp ] = (syscall_handler_t *) sys_getpgrp, + [ __NR_setsid ] = (syscall_handler_t *) sys_setsid, + [ __NR_setreuid ] = (syscall_handler_t *) sys_setreuid16, + [ __NR_setregid ] = (syscall_handler_t *) sys_setregid16, + [ __NR_sethostname ] = (syscall_handler_t *) sys_sethostname, + [ __NR_setrlimit ] = (syscall_handler_t *) sys_setrlimit, + [ __NR_getrlimit ] = (syscall_handler_t *) sys_old_getrlimit, + [ __NR_getrusage ] = (syscall_handler_t *) sys_getrusage, + [ __NR_gettimeofday ] = (syscall_handler_t *) sys_gettimeofday, + [ __NR_settimeofday ] = (syscall_handler_t *) sys_settimeofday, + [ __NR_getgroups ] = (syscall_handler_t *) sys_getgroups16, + [ __NR_setgroups ] = (syscall_handler_t *) sys_setgroups16, + [ __NR_symlink ] = (syscall_handler_t *) sys_symlink, + [ __NR_readlink ] = (syscall_handler_t *) sys_readlink, + [ __NR_uselib ] = (syscall_handler_t *) sys_uselib, + [ __NR_swapon ] = (syscall_handler_t *) sys_swapon, + [ __NR_reboot ] = (syscall_handler_t *) sys_reboot, + [ __NR_munmap ] = (syscall_handler_t *) sys_munmap, + [ __NR_truncate ] = (syscall_handler_t *) sys_truncate, + [ __NR_ftruncate ] = (syscall_handler_t *) sys_ftruncate, + [ __NR_fchmod ] = (syscall_handler_t *) sys_fchmod, + [ __NR_fchown ] = (syscall_handler_t *) sys_fchown16, + [ __NR_getpriority ] = (syscall_handler_t *) sys_getpriority, + [ __NR_setpriority ] = (syscall_handler_t *) sys_setpriority, + [ __NR_statfs ] = (syscall_handler_t *) sys_statfs, + [ __NR_fstatfs ] = (syscall_handler_t *) sys_fstatfs, + [ __NR_ioperm ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_syslog ] = (syscall_handler_t *) sys_syslog, + [ __NR_setitimer ] = (syscall_handler_t *) sys_setitimer, + [ __NR_getitimer ] = (syscall_handler_t *) sys_getitimer, + [ __NR_stat ] = (syscall_handler_t *) sys_newstat, + [ __NR_lstat ] = (syscall_handler_t *) sys_newlstat, + [ __NR_fstat ] = (syscall_handler_t *) sys_newfstat, + [ __NR_vhangup ] = (syscall_handler_t *) sys_vhangup, + [ __NR_wait4 ] = (syscall_handler_t *) sys_wait4, + [ __NR_swapoff ] = (syscall_handler_t *) sys_swapoff, + [ __NR_sysinfo ] = (syscall_handler_t *) sys_sysinfo, + [ __NR_fsync ] = (syscall_handler_t *) sys_fsync, + [ __NR_clone ] = (syscall_handler_t *) sys_clone, + [ __NR_setdomainname ] = (syscall_handler_t *) sys_setdomainname, + [ __NR_uname ] = (syscall_handler_t *) sys_newuname, + [ __NR_adjtimex ] = (syscall_handler_t *) sys_adjtimex, + [ __NR_mprotect ] = (syscall_handler_t *) sys_mprotect, + [ __NR_create_module ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_init_module ] = (syscall_handler_t *) sys_init_module, + [ __NR_delete_module ] = (syscall_handler_t *) sys_delete_module, + [ __NR_get_kernel_syms ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_quotactl ] = (syscall_handler_t *) sys_quotactl, + [ __NR_getpgid ] = (syscall_handler_t *) sys_getpgid, + [ __NR_fchdir ] = (syscall_handler_t *) sys_fchdir, + [ __NR_sysfs ] = (syscall_handler_t *) sys_sysfs, + [ __NR_personality ] = (syscall_handler_t *) sys_personality, + [ __NR_afs_syscall ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_setfsuid ] = (syscall_handler_t *) sys_setfsuid16, + [ __NR_setfsgid ] = (syscall_handler_t *) sys_setfsgid16, + [ __NR_getdents ] = (syscall_handler_t *) sys_getdents, + [ __NR_flock ] = (syscall_handler_t *) sys_flock, + [ __NR_msync ] = (syscall_handler_t *) sys_msync, + [ __NR_readv ] = (syscall_handler_t *) sys_readv, + [ __NR_writev ] = (syscall_handler_t *) sys_writev, + [ __NR_getsid ] = (syscall_handler_t *) sys_getsid, + [ __NR_fdatasync ] = (syscall_handler_t *) sys_fdatasync, + [ __NR__sysctl ] = (syscall_handler_t *) sys_sysctl, + [ __NR_mlock ] = (syscall_handler_t *) sys_mlock, + [ __NR_munlock ] = (syscall_handler_t *) sys_munlock, + [ __NR_mlockall ] = (syscall_handler_t *) sys_mlockall, + [ __NR_munlockall ] = (syscall_handler_t *) sys_munlockall, + [ __NR_sched_setparam ] = (syscall_handler_t *) sys_sched_setparam, + [ __NR_sched_getparam ] = (syscall_handler_t *) sys_sched_getparam, + [ __NR_sched_setscheduler ] = (syscall_handler_t *) sys_sched_setscheduler, + [ __NR_sched_getscheduler ] = (syscall_handler_t *) sys_sched_getscheduler, + [ __NR_sched_yield ] = (syscall_handler_t *) yield, + [ __NR_sched_get_priority_max ] = (syscall_handler_t *) sys_sched_get_priority_max, + [ __NR_sched_get_priority_min ] = (syscall_handler_t *) sys_sched_get_priority_min, + [ __NR_sched_rr_get_interval ] = (syscall_handler_t *) sys_sched_rr_get_interval, + [ __NR_nanosleep ] = (syscall_handler_t *) sys_nanosleep, + [ __NR_mremap ] = (syscall_handler_t *) sys_mremap, + [ __NR_setresuid ] = (syscall_handler_t *) sys_setresuid16, + [ __NR_getresuid ] = (syscall_handler_t *) sys_getresuid16, + [ __NR_query_module ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_poll ] = (syscall_handler_t *) sys_poll, + [ __NR_nfsservctl ] = (syscall_handler_t *) NFSSERVCTL, + [ __NR_setresgid ] = (syscall_handler_t *) sys_setresgid16, + [ __NR_getresgid ] = (syscall_handler_t *) sys_getresgid16, + [ __NR_prctl ] = (syscall_handler_t *) sys_prctl, + [ __NR_rt_sigreturn ] = (syscall_handler_t *) sys_rt_sigreturn, + [ __NR_rt_sigaction ] = (syscall_handler_t *) sys_rt_sigaction, + [ __NR_rt_sigprocmask ] = (syscall_handler_t *) sys_rt_sigprocmask, + [ __NR_rt_sigpending ] = (syscall_handler_t *) sys_rt_sigpending, + [ __NR_rt_sigtimedwait ] = (syscall_handler_t *) sys_rt_sigtimedwait, + [ __NR_rt_sigqueueinfo ] = (syscall_handler_t *) sys_rt_sigqueueinfo, + [ __NR_rt_sigsuspend ] = (syscall_handler_t *) sys_rt_sigsuspend, + [ __NR_pread64 ] = (syscall_handler_t *) sys_pread64, + [ __NR_pwrite64 ] = (syscall_handler_t *) sys_pwrite64, + [ __NR_chown ] = (syscall_handler_t *) sys_chown16, + [ __NR_getcwd ] = (syscall_handler_t *) sys_getcwd, + [ __NR_capget ] = (syscall_handler_t *) sys_capget, + [ __NR_capset ] = (syscall_handler_t *) sys_capset, + [ __NR_sigaltstack ] = (syscall_handler_t *) sys_sigaltstack, + [ __NR_sendfile ] = (syscall_handler_t *) sys_sendfile, + [ __NR_getpmsg ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_putpmsg ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_vfork ] = (syscall_handler_t *) sys_vfork, + [ __NR_getdents64 ] = (syscall_handler_t *) sys_getdents64, + [ __NR_gettid ] = (syscall_handler_t *) sys_gettid, + [ __NR_readahead ] = (syscall_handler_t *) sys_readahead, + [ __NR_setxattr ] = (syscall_handler_t *) sys_setxattr, + [ __NR_lsetxattr ] = (syscall_handler_t *) sys_lsetxattr, + [ __NR_fsetxattr ] = (syscall_handler_t *) sys_fsetxattr, + [ __NR_getxattr ] = (syscall_handler_t *) sys_getxattr, + [ __NR_lgetxattr ] = (syscall_handler_t *) sys_lgetxattr, + [ __NR_fgetxattr ] = (syscall_handler_t *) sys_fgetxattr, + [ __NR_listxattr ] = (syscall_handler_t *) sys_listxattr, + [ __NR_llistxattr ] = (syscall_handler_t *) sys_llistxattr, + [ __NR_flistxattr ] = (syscall_handler_t *) sys_flistxattr, + [ __NR_removexattr ] = (syscall_handler_t *) sys_removexattr, + [ __NR_lremovexattr ] = (syscall_handler_t *) sys_lremovexattr, + [ __NR_fremovexattr ] = (syscall_handler_t *) sys_fremovexattr, + [ __NR_tkill ] = (syscall_handler_t *) sys_tkill, + [ __NR_futex ] = (syscall_handler_t *) sys_futex, + [ __NR_sched_setaffinity ] = (syscall_handler_t *) sys_sched_setaffinity, + [ __NR_sched_getaffinity ] = (syscall_handler_t *) sys_sched_getaffinity, + [ __NR_io_setup ] = (syscall_handler_t *) sys_io_setup, + [ __NR_io_destroy ] = (syscall_handler_t *) sys_io_destroy, + [ __NR_io_getevents ] = (syscall_handler_t *) sys_io_getevents, + [ __NR_io_submit ] = (syscall_handler_t *) sys_io_submit, + [ __NR_io_cancel ] = (syscall_handler_t *) sys_io_cancel, + [ __NR_exit_group ] = (syscall_handler_t *) sys_exit_group, + [ __NR_lookup_dcookie ] = (syscall_handler_t *) sys_lookup_dcookie, + [ __NR_epoll_create ] = (syscall_handler_t *) sys_epoll_create, + [ __NR_epoll_ctl ] = (syscall_handler_t *) sys_epoll_ctl, + [ __NR_epoll_wait ] = (syscall_handler_t *) sys_epoll_wait, + [ __NR_remap_file_pages ] = (syscall_handler_t *) sys_remap_file_pages, + [ __NR_set_tid_address ] = (syscall_handler_t *) sys_set_tid_address, + [ __NR_timer_create ] = (syscall_handler_t *) sys_timer_create, + [ __NR_timer_settime ] = (syscall_handler_t *) sys_timer_settime, + [ __NR_timer_gettime ] = (syscall_handler_t *) sys_timer_gettime, + [ __NR_timer_getoverrun ] = (syscall_handler_t *) sys_timer_getoverrun, + [ __NR_timer_delete ] = (syscall_handler_t *) sys_timer_delete, + [ __NR_clock_settime ] = (syscall_handler_t *) sys_clock_settime, + [ __NR_clock_gettime ] = (syscall_handler_t *) sys_clock_gettime, + [ __NR_clock_getres ] = (syscall_handler_t *) sys_clock_getres, + [ __NR_clock_nanosleep ] = (syscall_handler_t *) sys_clock_nanosleep, + [ __NR_tgkill ] = (syscall_handler_t *) sys_tgkill, + [ __NR_utimes ] = (syscall_handler_t *) sys_utimes, + [ __NR_fadvise64 ] = (syscall_handler_t *) sys_fadvise64, + [ __NR_vserver ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_mbind ] = (syscall_handler_t *) sys_mbind, + [ __NR_get_mempolicy ] = (syscall_handler_t *) sys_get_mempolicy, + [ __NR_set_mempolicy ] = (syscall_handler_t *) sys_set_mempolicy, + [ __NR_mq_open ] = (syscall_handler_t *) sys_mq_open, + [ __NR_mq_unlink ] = (syscall_handler_t *) sys_mq_unlink, + [ __NR_mq_timedsend ] = (syscall_handler_t *) sys_mq_timedsend, + [ __NR_mq_timedreceive ] = (syscall_handler_t *) sys_mq_timedreceive, + [ __NR_mq_notify ] = (syscall_handler_t *) sys_mq_notify, + [ __NR_mq_getsetattr ] = (syscall_handler_t *) sys_mq_getsetattr, + [ __NR_kexec_load ] = (syscall_handler_t *) sys_ni_syscall, + [ __NR_waitid ] = (syscall_handler_t *) sys_waitid, + [ __NR_add_key ] = (syscall_handler_t *) sys_add_key, + [ __NR_request_key ] = (syscall_handler_t *) sys_request_key, + [ __NR_keyctl ] = (syscall_handler_t *) sys_keyctl, + + ARCH_SYSCALLS + [ LAST_SYSCALL + 1 ... NR_syscalls ] = + (syscall_handler_t *) sys_ni_syscall +}; diff --git a/arch/um/kernel/syscall_kern.c b/arch/um/kernel/syscall_kern.c new file mode 100644 index 00000000000..42731e04f50 --- /dev/null +++ b/arch/um/kernel/syscall_kern.c @@ -0,0 +1,176 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/file.h" +#include "linux/smp_lock.h" +#include "linux/mm.h" +#include "linux/utsname.h" +#include "linux/msg.h" +#include "linux/shm.h" +#include "linux/sys.h" +#include "linux/syscalls.h" +#include "linux/unistd.h" +#include "linux/slab.h" +#include "linux/utime.h" +#include "asm/mman.h" +#include "asm/uaccess.h" +#include "asm/ipc.h" +#include "kern_util.h" +#include "user_util.h" +#include "sysdep/syscalls.h" +#include "mode_kern.h" +#include "choose-mode.h" + +/* Unlocked, I don't care if this is a bit off */ +int nsyscalls = 0; + +long sys_fork(void) +{ + long ret; + + current->thread.forking = 1; + ret = do_fork(SIGCHLD, 0, NULL, 0, NULL, NULL); + current->thread.forking = 0; + return(ret); +} + +long sys_vfork(void) +{ + long ret; + + current->thread.forking = 1; + ret = do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, 0, NULL, 0, NULL, + NULL); + current->thread.forking = 0; + return(ret); +} + +/* common code for old and new mmaps */ +long sys_mmap2(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long pgoff) +{ + long error = -EBADF; + struct file * file = NULL; + + flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); + if (!(flags & MAP_ANONYMOUS)) { + file = fget(fd); + if (!file) + goto out; + } + + down_write(¤t->mm->mmap_sem); + error = do_mmap_pgoff(file, addr, len, prot, flags, pgoff); + up_write(¤t->mm->mmap_sem); + + if (file) + fput(file); + out: + return error; +} + +long old_mmap(unsigned long addr, unsigned long len, + unsigned long prot, unsigned long flags, + unsigned long fd, unsigned long offset) +{ + long err = -EINVAL; + if (offset & ~PAGE_MASK) + goto out; + + err = sys_mmap2(addr, len, prot, flags, fd, offset >> PAGE_SHIFT); + out: + return err; +} +/* + * sys_pipe() is the normal C calling standard for creating + * a pipe. It's not the way unix traditionally does this, though. + */ +long sys_pipe(unsigned long __user * fildes) +{ + int fd[2]; + long error; + + error = do_pipe(fd); + if (!error) { + if (copy_to_user(fildes, fd, sizeof(fd))) + error = -EFAULT; + } + return error; +} + + +long sys_uname(struct old_utsname * name) +{ + long err; + if (!name) + return -EFAULT; + down_read(&uts_sem); + err=copy_to_user(name, &system_utsname, sizeof (*name)); + up_read(&uts_sem); + return err?-EFAULT:0; +} + +long sys_olduname(struct oldold_utsname * name) +{ + long error; + + if (!name) + return -EFAULT; + if (!access_ok(VERIFY_WRITE,name,sizeof(struct oldold_utsname))) + return -EFAULT; + + down_read(&uts_sem); + + error = __copy_to_user(&name->sysname,&system_utsname.sysname, + __OLD_UTS_LEN); + error |= __put_user(0,name->sysname+__OLD_UTS_LEN); + error |= __copy_to_user(&name->nodename,&system_utsname.nodename, + __OLD_UTS_LEN); + error |= __put_user(0,name->nodename+__OLD_UTS_LEN); + error |= __copy_to_user(&name->release,&system_utsname.release, + __OLD_UTS_LEN); + error |= __put_user(0,name->release+__OLD_UTS_LEN); + error |= __copy_to_user(&name->version,&system_utsname.version, + __OLD_UTS_LEN); + error |= __put_user(0,name->version+__OLD_UTS_LEN); + error |= __copy_to_user(&name->machine,&system_utsname.machine, + __OLD_UTS_LEN); + error |= __put_user(0,name->machine+__OLD_UTS_LEN); + + up_read(&uts_sem); + + error = error ? -EFAULT : 0; + + return error; +} + +DEFINE_SPINLOCK(syscall_lock); + +static int syscall_index = 0; + +int next_syscall_index(int limit) +{ + int ret; + + spin_lock(&syscall_lock); + ret = syscall_index; + if(++syscall_index == limit) + syscall_index = 0; + spin_unlock(&syscall_lock); + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/syscall_user.c b/arch/um/kernel/syscall_user.c new file mode 100644 index 00000000000..01b711e00a8 --- /dev/null +++ b/arch/um/kernel/syscall_user.c @@ -0,0 +1,48 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <sys/time.h> +#include "kern_util.h" +#include "syscall_user.h" + +struct { + int syscall; + int pid; + long result; + struct timeval start; + struct timeval end; +} syscall_record[1024]; + +int record_syscall_start(int syscall) +{ + int max, index; + + max = sizeof(syscall_record)/sizeof(syscall_record[0]); + index = next_syscall_index(max); + + syscall_record[index].syscall = syscall; + syscall_record[index].pid = current_pid(); + syscall_record[index].result = 0xdeadbeef; + gettimeofday(&syscall_record[index].start, NULL); + return(index); +} + +void record_syscall_end(int index, long result) +{ + syscall_record[index].result = result; + gettimeofday(&syscall_record[index].end, NULL); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/sysrq.c b/arch/um/kernel/sysrq.c new file mode 100644 index 00000000000..e630438f9e7 --- /dev/null +++ b/arch/um/kernel/sysrq.c @@ -0,0 +1,81 @@ +/* + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/kernel.h" +#include "linux/module.h" +#include "linux/kallsyms.h" +#include "asm/page.h" +#include "asm/processor.h" +#include "sysrq.h" +#include "user_util.h" + +void show_trace(unsigned long * stack) +{ + /* XXX: Copy the CONFIG_FRAME_POINTER stack-walking backtrace from + * arch/i386/kernel/traps.c, and then move this to sys-i386/sysrq.c.*/ + unsigned long addr; + + if (!stack) { + stack = (unsigned long*) &stack; + WARN_ON(1); + } + + printk("Call Trace: \n"); + while (((long) stack & (THREAD_SIZE-1)) != 0) { + addr = *stack; + if (__kernel_text_address(addr)) { + printk("%08lx: [<%08lx>]", (unsigned long) stack, addr); + print_symbol(" %s", addr); + printk("\n"); + } + stack++; + } + printk("\n"); +} + +/* + * stack dumps generator - this is used by arch-independent code. + * And this is identical to i386 currently. + */ +void dump_stack(void) +{ + unsigned long stack; + + show_trace(&stack); +} +EXPORT_SYMBOL(dump_stack); + +/*Stolen from arch/i386/kernel/traps.c */ +static int kstack_depth_to_print = 24; + +/* This recently started being used in arch-independent code too, as in + * kernel/sched.c.*/ +void show_stack(struct task_struct *task, unsigned long *esp) +{ + unsigned long *stack; + int i; + + if (esp == NULL) { + if (task != current) { + esp = (unsigned long *) KSTK_ESP(task); + /* Which one? No actual difference - just coding style.*/ + //esp = (unsigned long *) PT_REGS_IP(&task->thread.regs); + } else { + esp = (unsigned long *) &esp; + } + } + + stack = esp; + for(i = 0; i < kstack_depth_to_print; i++) { + if (kstack_end(stack)) + break; + if (i && ((i % 8) == 0)) + printk("\n "); + printk("%08lx ", *stack++); + } + + show_trace(esp); +} diff --git a/arch/um/kernel/tempfile.c b/arch/um/kernel/tempfile.c new file mode 100644 index 00000000000..b1674bc1395 --- /dev/null +++ b/arch/um/kernel/tempfile.c @@ -0,0 +1,82 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <string.h> +#include <errno.h> +#include <sys/param.h> +#include "init.h" + +/* Modified from create_mem_file and start_debugger */ +static char *tempdir = NULL; + +static void __init find_tempdir(void) +{ + char *dirs[] = { "TMP", "TEMP", "TMPDIR", NULL }; + int i; + char *dir = NULL; + + if(tempdir != NULL) return; /* We've already been called */ + for(i = 0; dirs[i]; i++){ + dir = getenv(dirs[i]); + if((dir != NULL) && (*dir != '\0')) + break; + } + if((dir == NULL) || (*dir == '\0')) + dir = "/tmp"; + + tempdir = malloc(strlen(dir) + 2); + if(tempdir == NULL){ + fprintf(stderr, "Failed to malloc tempdir, " + "errno = %d\n", errno); + return; + } + strcpy(tempdir, dir); + strcat(tempdir, "/"); +} + +int make_tempfile(const char *template, char **out_tempname, int do_unlink) +{ + char tempname[MAXPATHLEN]; + int fd; + + find_tempdir(); + if (*template != '/') + strcpy(tempname, tempdir); + else + *tempname = 0; + strcat(tempname, template); + fd = mkstemp(tempname); + if(fd < 0){ + fprintf(stderr, "open - cannot create %s: %s\n", tempname, + strerror(errno)); + return -1; + } + if(do_unlink && (unlink(tempname) < 0)){ + perror("unlink"); + return -1; + } + if(out_tempname){ + *out_tempname = strdup(tempname); + if(*out_tempname == NULL){ + perror("strdup"); + return -1; + } + } + return(fd); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c new file mode 100644 index 00000000000..c40c86a3f91 --- /dev/null +++ b/arch/um/kernel/time.c @@ -0,0 +1,167 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <time.h> +#include <sys/time.h> +#include <signal.h> +#include <errno.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "process.h" +#include "signal_user.h" +#include "time_user.h" +#include "kern_constants.h" + +/* XXX This really needs to be declared and initialized in a kernel file since + * it's in <linux/time.h> + */ +extern struct timespec wall_to_monotonic; + +extern struct timeval xtime; + +struct timeval local_offset = { 0, 0 }; + +void timer(void) +{ + gettimeofday(&xtime, NULL); + timeradd(&xtime, &local_offset, &xtime); +} + +void set_interval(int timer_type) +{ + int usec = 1000000/hz(); + struct itimerval interval = ((struct itimerval) { { 0, usec }, + { 0, usec } }); + + if(setitimer(timer_type, &interval, NULL) == -1) + panic("setitimer failed - errno = %d\n", errno); +} + +void enable_timer(void) +{ + int usec = 1000000/hz(); + struct itimerval enable = ((struct itimerval) { { 0, usec }, + { 0, usec }}); + if(setitimer(ITIMER_VIRTUAL, &enable, NULL)) + printk("enable_timer - setitimer failed, errno = %d\n", + errno); +} + +void disable_timer(void) +{ + struct itimerval disable = ((struct itimerval) { { 0, 0 }, { 0, 0 }}); + if((setitimer(ITIMER_VIRTUAL, &disable, NULL) < 0) || + (setitimer(ITIMER_REAL, &disable, NULL) < 0)) + printk("disnable_timer - setitimer failed, errno = %d\n", + errno); + /* If there are signals already queued, after unblocking ignore them */ + set_handler(SIGALRM, SIG_IGN, 0, -1); + set_handler(SIGVTALRM, SIG_IGN, 0, -1); +} + +void switch_timers(int to_real) +{ + struct itimerval disable = ((struct itimerval) { { 0, 0 }, { 0, 0 }}); + struct itimerval enable = ((struct itimerval) { { 0, 1000000/hz() }, + { 0, 1000000/hz() }}); + int old, new; + + if(to_real){ + old = ITIMER_VIRTUAL; + new = ITIMER_REAL; + } + else { + old = ITIMER_REAL; + new = ITIMER_VIRTUAL; + } + + if((setitimer(old, &disable, NULL) < 0) || + (setitimer(new, &enable, NULL))) + printk("switch_timers - setitimer failed, errno = %d\n", + errno); +} + +void uml_idle_timer(void) +{ + if(signal(SIGVTALRM, SIG_IGN) == SIG_ERR) + panic("Couldn't unset SIGVTALRM handler"); + + set_handler(SIGALRM, (__sighandler_t) alarm_handler, + SA_RESTART, SIGUSR1, SIGIO, SIGWINCH, SIGVTALRM, -1); + set_interval(ITIMER_REAL); +} + +extern int do_posix_clock_monotonic_gettime(struct timespec *tp); + +void time_init(void) +{ + struct timespec now; + + if(signal(SIGVTALRM, boot_timer_handler) == SIG_ERR) + panic("Couldn't set SIGVTALRM handler"); + set_interval(ITIMER_VIRTUAL); + + do_posix_clock_monotonic_gettime(&now); + wall_to_monotonic.tv_sec = -now.tv_sec; + wall_to_monotonic.tv_nsec = -now.tv_nsec; +} + +/* Declared in linux/time.h, which can't be included here */ +extern void clock_was_set(void); + +void do_gettimeofday(struct timeval *tv) +{ + unsigned long flags; + + flags = time_lock(); + gettimeofday(tv, NULL); + timeradd(tv, &local_offset, tv); + time_unlock(flags); + clock_was_set(); +} + +int do_settimeofday(struct timespec *tv) +{ + struct timeval now; + unsigned long flags; + struct timeval tv_in; + + if ((unsigned long) tv->tv_nsec >= UM_NSEC_PER_SEC) + return -EINVAL; + + tv_in.tv_sec = tv->tv_sec; + tv_in.tv_usec = tv->tv_nsec / 1000; + + flags = time_lock(); + gettimeofday(&now, NULL); + timersub(&tv_in, &now, &local_offset); + time_unlock(flags); + + return(0); +} + +void idle_sleep(int secs) +{ + struct timespec ts; + + ts.tv_sec = secs; + ts.tv_nsec = 0; + nanosleep(&ts, NULL); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/time_kern.c b/arch/um/kernel/time_kern.c new file mode 100644 index 00000000000..2461cd73ca8 --- /dev/null +++ b/arch/um/kernel/time_kern.c @@ -0,0 +1,203 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/module.h" +#include "linux/unistd.h" +#include "linux/stddef.h" +#include "linux/spinlock.h" +#include "linux/time.h" +#include "linux/sched.h" +#include "linux/interrupt.h" +#include "linux/init.h" +#include "linux/delay.h" +#include "asm/irq.h" +#include "asm/param.h" +#include "asm/current.h" +#include "kern_util.h" +#include "user_util.h" +#include "time_user.h" +#include "mode.h" +#include "os.h" + +u64 jiffies_64 = INITIAL_JIFFIES; + +EXPORT_SYMBOL(jiffies_64); + +int hz(void) +{ + return(HZ); +} + +/* + * Scheduler clock - returns current time in nanosec units. + */ +unsigned long long sched_clock(void) +{ + return (unsigned long long)jiffies_64 * (1000000000 / HZ); +} + +/* Changed at early boot */ +int timer_irq_inited = 0; + +static int first_tick; +static unsigned long long prev_usecs; +#ifdef CONFIG_UML_REAL_TIME_CLOCK +static long long delta; /* Deviation per interval */ +#endif + +#define MILLION 1000000 + +void timer_irq(union uml_pt_regs *regs) +{ + unsigned long long ticks = 0; + + if(!timer_irq_inited){ + /* This is to ensure that ticks don't pile up when + * the timer handler is suspended */ + first_tick = 0; + return; + } + + if(first_tick){ +#ifdef CONFIG_UML_REAL_TIME_CLOCK + /* We've had 1 tick */ + unsigned long long usecs = os_usecs(); + + delta += usecs - prev_usecs; + prev_usecs = usecs; + + /* Protect against the host clock being set backwards */ + if(delta < 0) + delta = 0; + + ticks += (delta * HZ) / MILLION; + delta -= (ticks * MILLION) / HZ; +#else + ticks = 1; +#endif + } + else { + prev_usecs = os_usecs(); + first_tick = 1; + } + + while(ticks > 0){ + do_IRQ(TIMER_IRQ, regs); + ticks--; + } +} + +void boot_timer_handler(int sig) +{ + struct pt_regs regs; + + CHOOSE_MODE((void) + (UPT_SC(®s.regs) = (struct sigcontext *) (&sig + 1)), + (void) (regs.regs.skas.is_user = 0)); + do_timer(®s); +} + +irqreturn_t um_timer(int irq, void *dev, struct pt_regs *regs) +{ + unsigned long flags; + + do_timer(regs); + write_seqlock_irqsave(&xtime_lock, flags); + timer(); + write_sequnlock_irqrestore(&xtime_lock, flags); + return(IRQ_HANDLED); +} + +long um_time(int __user *tloc) +{ + struct timeval now; + + do_gettimeofday(&now); + if (tloc) { + if (put_user(now.tv_sec, tloc)) + now.tv_sec = -EFAULT; + } + return now.tv_sec; +} + +long um_stime(int __user *tptr) +{ + int value; + struct timespec new; + + if (get_user(value, tptr)) + return -EFAULT; + new.tv_sec = value; + new.tv_nsec = 0; + do_settimeofday(&new); + return 0; +} + +void __udelay(unsigned long usecs) +{ + int i, n; + + n = (loops_per_jiffy * HZ * usecs) / MILLION; + for(i=0;i<n;i++) ; +} + +void __const_udelay(unsigned long usecs) +{ + int i, n; + + n = (loops_per_jiffy * HZ * usecs) / MILLION; + for(i=0;i<n;i++) ; +} + +void timer_handler(int sig, union uml_pt_regs *regs) +{ + local_irq_disable(); + update_process_times(CHOOSE_MODE(user_context(UPT_SP(regs)), (regs)->skas.is_user)); + local_irq_enable(); + if(current_thread->cpu == 0) + timer_irq(regs); +} + +static DEFINE_SPINLOCK(timer_spinlock); + +unsigned long time_lock(void) +{ + unsigned long flags; + + spin_lock_irqsave(&timer_spinlock, flags); + return(flags); +} + +void time_unlock(unsigned long flags) +{ + spin_unlock_irqrestore(&timer_spinlock, flags); +} + +int __init timer_init(void) +{ + int err; + + CHOOSE_MODE(user_time_init_tt(), user_time_init_skas()); + err = request_irq(TIMER_IRQ, um_timer, SA_INTERRUPT, "timer", NULL); + if(err != 0) + printk(KERN_ERR "timer_init : request_irq failed - " + "errno = %d\n", -err); + timer_irq_inited = 1; + return(0); +} + +__initcall(timer_init); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c new file mode 100644 index 00000000000..eda477edfdf --- /dev/null +++ b/arch/um/kernel/tlb.c @@ -0,0 +1,369 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/mm.h" +#include "asm/page.h" +#include "asm/pgalloc.h" +#include "asm/tlbflush.h" +#include "choose-mode.h" +#include "mode_kern.h" +#include "user_util.h" +#include "tlb.h" +#include "mem.h" +#include "mem_user.h" +#include "os.h" + +#define ADD_ROUND(n, inc) (((n) + (inc)) & ~((inc) - 1)) + +void fix_range_common(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force, int data, + void (*do_ops)(int, struct host_vm_op *, int)) +{ + pgd_t *npgd; + pud_t *npud; + pmd_t *npmd; + pte_t *npte; + unsigned long addr, end; + int r, w, x; + struct host_vm_op ops[16]; + int op_index = -1, last_op = sizeof(ops) / sizeof(ops[0]) - 1; + + if(mm == NULL) return; + + for(addr = start_addr; addr < end_addr;){ + npgd = pgd_offset(mm, addr); + if(!pgd_present(*npgd)){ + end = ADD_ROUND(addr, PGDIR_SIZE); + if(end > end_addr) + end = end_addr; + if(force || pgd_newpage(*npgd)){ + op_index = add_munmap(addr, end - addr, ops, + op_index, last_op, data, + do_ops); + pgd_mkuptodate(*npgd); + } + addr = end; + continue; + } + + npud = pud_offset(npgd, addr); + if(!pud_present(*npud)){ + end = ADD_ROUND(addr, PUD_SIZE); + if(end > end_addr) + end = end_addr; + if(force || pud_newpage(*npud)){ + op_index = add_munmap(addr, end - addr, ops, + op_index, last_op, data, + do_ops); + pud_mkuptodate(*npud); + } + addr = end; + continue; + } + + npmd = pmd_offset(npud, addr); + if(!pmd_present(*npmd)){ + end = ADD_ROUND(addr, PMD_SIZE); + if(end > end_addr) + end = end_addr; + if(force || pmd_newpage(*npmd)){ + op_index = add_munmap(addr, end - addr, ops, + op_index, last_op, data, + do_ops); + pmd_mkuptodate(*npmd); + } + addr = end; + continue; + } + + npte = pte_offset_kernel(npmd, addr); + r = pte_read(*npte); + w = pte_write(*npte); + x = pte_exec(*npte); + if(!pte_dirty(*npte)) + w = 0; + if(!pte_young(*npte)){ + r = 0; + w = 0; + } + if(force || pte_newpage(*npte)){ + if(pte_present(*npte)) + op_index = add_mmap(addr, + pte_val(*npte) & PAGE_MASK, + PAGE_SIZE, r, w, x, ops, + op_index, last_op, data, + do_ops); + else op_index = add_munmap(addr, PAGE_SIZE, ops, + op_index, last_op, data, + do_ops); + } + else if(pte_newprot(*npte)) + op_index = add_mprotect(addr, PAGE_SIZE, r, w, x, ops, + op_index, last_op, data, + do_ops); + + *npte = pte_mkuptodate(*npte); + addr += PAGE_SIZE; + } + (*do_ops)(data, ops, op_index); +} + +int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) +{ + struct mm_struct *mm; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long addr, last; + int updated = 0, err; + + mm = &init_mm; + for(addr = start; addr < end;){ + pgd = pgd_offset(mm, addr); + if(!pgd_present(*pgd)){ + last = ADD_ROUND(addr, PGDIR_SIZE); + if(last > end) + last = end; + if(pgd_newpage(*pgd)){ + updated = 1; + err = os_unmap_memory((void *) addr, + last - addr); + if(err < 0) + panic("munmap failed, errno = %d\n", + -err); + } + addr = last; + continue; + } + + pud = pud_offset(pgd, addr); + if(!pud_present(*pud)){ + last = ADD_ROUND(addr, PUD_SIZE); + if(last > end) + last = end; + if(pud_newpage(*pud)){ + updated = 1; + err = os_unmap_memory((void *) addr, + last - addr); + if(err < 0) + panic("munmap failed, errno = %d\n", + -err); + } + addr = last; + continue; + } + + pmd = pmd_offset(pud, addr); + if(!pmd_present(*pmd)){ + last = ADD_ROUND(addr, PMD_SIZE); + if(last > end) + last = end; + if(pmd_newpage(*pmd)){ + updated = 1; + err = os_unmap_memory((void *) addr, + last - addr); + if(err < 0) + panic("munmap failed, errno = %d\n", + -err); + } + addr = last; + continue; + } + + pte = pte_offset_kernel(pmd, addr); + if(!pte_present(*pte) || pte_newpage(*pte)){ + updated = 1; + err = os_unmap_memory((void *) addr, + PAGE_SIZE); + if(err < 0) + panic("munmap failed, errno = %d\n", + -err); + if(pte_present(*pte)) + map_memory(addr, + pte_val(*pte) & PAGE_MASK, + PAGE_SIZE, 1, 1, 1); + } + else if(pte_newprot(*pte)){ + updated = 1; + protect_memory(addr, PAGE_SIZE, 1, 1, 1, 1); + } + addr += PAGE_SIZE; + } + return(updated); +} + +void flush_tlb_page(struct vm_area_struct *vma, unsigned long address) +{ + address &= PAGE_MASK; + flush_tlb_range(vma, address, address + PAGE_SIZE); +} + +void flush_tlb_all(void) +{ + flush_tlb_mm(current->mm); +} + +void flush_tlb_kernel_range(unsigned long start, unsigned long end) +{ + CHOOSE_MODE_PROC(flush_tlb_kernel_range_tt, + flush_tlb_kernel_range_common, start, end); +} + +void flush_tlb_kernel_vm(void) +{ + CHOOSE_MODE(flush_tlb_kernel_vm_tt(), + flush_tlb_kernel_range_common(start_vm, end_vm)); +} + +void __flush_tlb_one(unsigned long addr) +{ + CHOOSE_MODE_PROC(__flush_tlb_one_tt, __flush_tlb_one_skas, addr); +} + +void flush_tlb_range(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + CHOOSE_MODE_PROC(flush_tlb_range_tt, flush_tlb_range_skas, vma, start, + end); +} + +void flush_tlb_mm(struct mm_struct *mm) +{ + CHOOSE_MODE_PROC(flush_tlb_mm_tt, flush_tlb_mm_skas, mm); +} + +void force_flush_all(void) +{ + CHOOSE_MODE(force_flush_all_tt(), force_flush_all_skas()); +} + +pgd_t *pgd_offset_proc(struct mm_struct *mm, unsigned long address) +{ + return(pgd_offset(mm, address)); +} + +pud_t *pud_offset_proc(pgd_t *pgd, unsigned long address) +{ + return(pud_offset(pgd, address)); +} + +pmd_t *pmd_offset_proc(pud_t *pud, unsigned long address) +{ + return(pmd_offset(pud, address)); +} + +pte_t *pte_offset_proc(pmd_t *pmd, unsigned long address) +{ + return(pte_offset_kernel(pmd, address)); +} + +pte_t *addr_pte(struct task_struct *task, unsigned long addr) +{ + pgd_t *pgd = pgd_offset(task->mm, addr); + pud_t *pud = pud_offset(pgd, addr); + pmd_t *pmd = pmd_offset(pud, addr); + + return(pte_offset_map(pmd, addr)); +} + +int add_mmap(unsigned long virt, unsigned long phys, unsigned long len, + int r, int w, int x, struct host_vm_op *ops, int index, + int last_filled, int data, + void (*do_ops)(int, struct host_vm_op *, int)) +{ + __u64 offset; + struct host_vm_op *last; + int fd; + + fd = phys_mapping(phys, &offset); + if(index != -1){ + last = &ops[index]; + if((last->type == MMAP) && + (last->u.mmap.addr + last->u.mmap.len == virt) && + (last->u.mmap.r == r) && (last->u.mmap.w == w) && + (last->u.mmap.x == x) && (last->u.mmap.fd == fd) && + (last->u.mmap.offset + last->u.mmap.len == offset)){ + last->u.mmap.len += len; + return(index); + } + } + + if(index == last_filled){ + (*do_ops)(data, ops, last_filled); + index = -1; + } + + ops[++index] = ((struct host_vm_op) { .type = MMAP, + .u = { .mmap = { + .addr = virt, + .len = len, + .r = r, + .w = w, + .x = x, + .fd = fd, + .offset = offset } + } }); + return(index); +} + +int add_munmap(unsigned long addr, unsigned long len, struct host_vm_op *ops, + int index, int last_filled, int data, + void (*do_ops)(int, struct host_vm_op *, int)) +{ + struct host_vm_op *last; + + if(index != -1){ + last = &ops[index]; + if((last->type == MUNMAP) && + (last->u.munmap.addr + last->u.mmap.len == addr)){ + last->u.munmap.len += len; + return(index); + } + } + + if(index == last_filled){ + (*do_ops)(data, ops, last_filled); + index = -1; + } + + ops[++index] = ((struct host_vm_op) { .type = MUNMAP, + .u = { .munmap = { + .addr = addr, + .len = len } } }); + return(index); +} + +int add_mprotect(unsigned long addr, unsigned long len, int r, int w, int x, + struct host_vm_op *ops, int index, int last_filled, int data, + void (*do_ops)(int, struct host_vm_op *, int)) +{ + struct host_vm_op *last; + + if(index != -1){ + last = &ops[index]; + if((last->type == MPROTECT) && + (last->u.mprotect.addr + last->u.mprotect.len == addr) && + (last->u.mprotect.r == r) && (last->u.mprotect.w == w) && + (last->u.mprotect.x == x)){ + last->u.mprotect.len += len; + return(index); + } + } + + if(index == last_filled){ + (*do_ops)(data, ops, last_filled); + index = -1; + } + + ops[++index] = ((struct host_vm_op) { .type = MPROTECT, + .u = { .mprotect = { + .addr = addr, + .len = len, + .r = r, + .w = w, + .x = x } } }); + return(index); +} diff --git a/arch/um/kernel/trap_kern.c b/arch/um/kernel/trap_kern.c new file mode 100644 index 00000000000..47e766e6ba1 --- /dev/null +++ b/arch/um/kernel/trap_kern.c @@ -0,0 +1,251 @@ +/* + * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "asm/errno.h" +#include "linux/sched.h" +#include "linux/mm.h" +#include "linux/spinlock.h" +#include "linux/config.h" +#include "linux/init.h" +#include "linux/ptrace.h" +#include "asm/semaphore.h" +#include "asm/pgtable.h" +#include "asm/pgalloc.h" +#include "asm/tlbflush.h" +#include "asm/a.out.h" +#include "asm/current.h" +#include "asm/irq.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "chan_kern.h" +#include "mconsole_kern.h" +#include "2_5compat.h" +#include "mem.h" +#include "mem_kern.h" + +int handle_page_fault(unsigned long address, unsigned long ip, + int is_write, int is_user, int *code_out) +{ + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long page; + int err = -EFAULT; + + *code_out = SEGV_MAPERR; + down_read(&mm->mmap_sem); + vma = find_vma(mm, address); + if(!vma) + goto out; + else if(vma->vm_start <= address) + goto good_area; + else if(!(vma->vm_flags & VM_GROWSDOWN)) + goto out; + else if(!ARCH_IS_STACKGROW(address)) + goto out; + else if(expand_stack(vma, address)) + goto out; + + good_area: + *code_out = SEGV_ACCERR; + if(is_write && !(vma->vm_flags & VM_WRITE)) + goto out; + page = address & PAGE_MASK; + pgd = pgd_offset(mm, page); + pud = pud_offset(pgd, page); + pmd = pmd_offset(pud, page); + do { + survive: + switch (handle_mm_fault(mm, vma, address, is_write)){ + case VM_FAULT_MINOR: + current->min_flt++; + break; + case VM_FAULT_MAJOR: + current->maj_flt++; + break; + case VM_FAULT_SIGBUS: + err = -EACCES; + goto out; + case VM_FAULT_OOM: + err = -ENOMEM; + goto out_of_memory; + default: + BUG(); + } + pgd = pgd_offset(mm, page); + pud = pud_offset(pgd, page); + pmd = pmd_offset(pud, page); + pte = pte_offset_kernel(pmd, page); + } while(!pte_present(*pte)); + err = 0; + *pte = pte_mkyoung(*pte); + if(pte_write(*pte)) *pte = pte_mkdirty(*pte); + flush_tlb_page(vma, page); + out: + up_read(&mm->mmap_sem); + return(err); + +/* + * We ran out of memory, or some other thing happened to us that made + * us unable to handle the page fault gracefully. + */ +out_of_memory: + if (current->pid == 1) { + up_read(&mm->mmap_sem); + yield(); + down_read(&mm->mmap_sem); + goto survive; + } + goto out; +} + +LIST_HEAD(physmem_remappers); + +void register_remapper(struct remapper *info) +{ + list_add(&info->list, &physmem_remappers); +} + +static int check_remapped_addr(unsigned long address, int is_write) +{ + struct remapper *remapper; + struct list_head *ele; + __u64 offset; + int fd; + + fd = phys_mapping(__pa(address), &offset); + if(fd == -1) + return(0); + + list_for_each(ele, &physmem_remappers){ + remapper = list_entry(ele, struct remapper, list); + if((*remapper->proc)(fd, address, is_write, offset)) + return(1); + } + + return(0); +} + +unsigned long segv(unsigned long address, unsigned long ip, int is_write, + int is_user, void *sc) +{ + struct siginfo si; + void *catcher; + int err; + + if(!is_user && (address >= start_vm) && (address < end_vm)){ + flush_tlb_kernel_vm(); + return(0); + } + else if(check_remapped_addr(address & PAGE_MASK, is_write)) + return(0); + else if(current->mm == NULL) + panic("Segfault with no mm"); + err = handle_page_fault(address, ip, is_write, is_user, &si.si_code); + + catcher = current->thread.fault_catcher; + if(!err) + return(0); + else if(catcher != NULL){ + current->thread.fault_addr = (void *) address; + do_longjmp(catcher, 1); + } + else if(current->thread.fault_addr != NULL) + panic("fault_addr set but no fault catcher"); + else if(arch_fixup(ip, sc)) + return(0); + + if(!is_user) + panic("Kernel mode fault at addr 0x%lx, ip 0x%lx", + address, ip); + + if(err == -EACCES){ + si.si_signo = SIGBUS; + si.si_errno = 0; + si.si_code = BUS_ADRERR; + si.si_addr = (void *)address; + force_sig_info(SIGBUS, &si, current); + } + else if(err == -ENOMEM){ + printk("VM: killing process %s\n", current->comm); + do_exit(SIGKILL); + } + else { + si.si_signo = SIGSEGV; + si.si_addr = (void *) address; + current->thread.cr2 = address; + current->thread.err = is_write; + force_sig_info(SIGSEGV, &si, current); + } + return(0); +} + +void bad_segv(unsigned long address, unsigned long ip, int is_write) +{ + struct siginfo si; + + si.si_signo = SIGSEGV; + si.si_code = SEGV_ACCERR; + si.si_addr = (void *) address; + current->thread.cr2 = address; + current->thread.err = is_write; + force_sig_info(SIGSEGV, &si, current); +} + +void relay_signal(int sig, union uml_pt_regs *regs) +{ + if(arch_handle_signal(sig, regs)) return; + if(!UPT_IS_USER(regs)) + panic("Kernel mode signal %d", sig); + force_sig(sig, current); +} + +void bus_handler(int sig, union uml_pt_regs *regs) +{ + if(current->thread.fault_catcher != NULL) + do_longjmp(current->thread.fault_catcher, 1); + else relay_signal(sig, regs); +} + +void winch(int sig, union uml_pt_regs *regs) +{ + do_IRQ(WINCH_IRQ, regs); +} + +void trap_init(void) +{ +} + +DEFINE_SPINLOCK(trap_lock); + +static int trap_index = 0; + +int next_trap_index(int limit) +{ + int ret; + + spin_lock(&trap_lock); + ret = trap_index; + if(++trap_index == limit) + trap_index = 0; + spin_unlock(&trap_lock); + return(ret); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/trap_user.c b/arch/um/kernel/trap_user.c new file mode 100644 index 00000000000..50a4042a509 --- /dev/null +++ b/arch/um/kernel/trap_user.c @@ -0,0 +1,120 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <errno.h> +#include <setjmp.h> +#include <signal.h> +#include <sys/time.h> +#include <sys/wait.h> +#include <asm/page.h> +#include <asm/unistd.h> +#include <asm/ptrace.h> +#include "init.h" +#include "sysdep/ptrace.h" +#include "sigcontext.h" +#include "sysdep/sigcontext.h" +#include "irq_user.h" +#include "signal_user.h" +#include "time_user.h" +#include "task.h" +#include "mode.h" +#include "choose-mode.h" +#include "kern_util.h" +#include "user_util.h" +#include "os.h" + +void kill_child_dead(int pid) +{ + kill(pid, SIGKILL); + kill(pid, SIGCONT); + do { + int n; + CATCH_EINTR(n = waitpid(pid, NULL, 0)); + if (n > 0) + kill(pid, SIGCONT); + else + break; + } while(1); +} + +/* Unlocked - don't care if this is a bit off */ +int nsegfaults = 0; + +struct { + unsigned long address; + int is_write; + int pid; + unsigned long sp; + int is_user; +} segfault_record[1024]; + +void segv_handler(int sig, union uml_pt_regs *regs) +{ + int index, max; + + if(UPT_IS_USER(regs) && !UPT_SEGV_IS_FIXABLE(regs)){ + bad_segv(UPT_FAULT_ADDR(regs), UPT_IP(regs), + UPT_FAULT_WRITE(regs)); + return; + } + max = sizeof(segfault_record)/sizeof(segfault_record[0]); + index = next_trap_index(max); + + nsegfaults++; + segfault_record[index].address = UPT_FAULT_ADDR(regs); + segfault_record[index].pid = os_getpid(); + segfault_record[index].is_write = UPT_FAULT_WRITE(regs); + segfault_record[index].sp = UPT_SP(regs); + segfault_record[index].is_user = UPT_IS_USER(regs); + segv(UPT_FAULT_ADDR(regs), UPT_IP(regs), UPT_FAULT_WRITE(regs), + UPT_IS_USER(regs), regs); +} + +void usr2_handler(int sig, union uml_pt_regs *regs) +{ + CHOOSE_MODE(syscall_handler_tt(sig, regs), (void) 0); +} + +struct signal_info sig_info[] = { + [ SIGTRAP ] { .handler = relay_signal, + .is_irq = 0 }, + [ SIGFPE ] { .handler = relay_signal, + .is_irq = 0 }, + [ SIGILL ] { .handler = relay_signal, + .is_irq = 0 }, + [ SIGWINCH ] { .handler = winch, + .is_irq = 1 }, + [ SIGBUS ] { .handler = bus_handler, + .is_irq = 0 }, + [ SIGSEGV] { .handler = segv_handler, + .is_irq = 0 }, + [ SIGIO ] { .handler = sigio_handler, + .is_irq = 1 }, + [ SIGVTALRM ] { .handler = timer_handler, + .is_irq = 1 }, + [ SIGALRM ] { .handler = timer_handler, + .is_irq = 1 }, + [ SIGUSR2 ] { .handler = usr2_handler, + .is_irq = 0 }, +}; + +void do_longjmp(void *b, int val) +{ + sigjmp_buf *buf = b; + + siglongjmp(*buf, val); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/Makefile b/arch/um/kernel/tt/Makefile new file mode 100644 index 00000000000..3d5177df350 --- /dev/null +++ b/arch/um/kernel/tt/Makefile @@ -0,0 +1,28 @@ +# +# Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) +# Licensed under the GPL +# + +extra-y := unmap_fin.o +clean-files := unmap_tmp.o + +obj-y = exec_kern.o exec_user.o gdb.o ksyms.o mem.o mem_user.o process_kern.o \ + syscall_kern.o syscall_user.o time.o tlb.o tracer.o trap_user.o \ + uaccess.o uaccess_user.o + +obj-$(CONFIG_PT_PROXY) += gdb_kern.o ptproxy/ + +USER_OBJS := gdb.o time.o tracer.o + +include arch/um/scripts/Makefile.rules + +UNMAP_CFLAGS := $(patsubst -pg -DPROFILING,,$(USER_CFLAGS)) +UNMAP_CFLAGS := $(patsubst -fprofile-arcs -ftest-coverage,,$(UNMAP_CFLAGS)) + +#XXX: partially copied from arch/um/scripts/Makefile.rules +$(obj)/unmap.o: c_flags = -Wp,-MD,$(depfile) $(UNMAP_CFLAGS) + +$(obj)/unmap_fin.o : $(obj)/unmap.o + $(LD) -r -o $(obj)/unmap_tmp.o $< $(shell $(CC) -print-file-name=libc.a) + $(OBJCOPY) $(obj)/unmap_tmp.o $@ -G switcheroo + diff --git a/arch/um/kernel/tt/exec_kern.c b/arch/um/kernel/tt/exec_kern.c new file mode 100644 index 00000000000..065b504a653 --- /dev/null +++ b/arch/um/kernel/tt/exec_kern.c @@ -0,0 +1,87 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/kernel.h" +#include "linux/mm.h" +#include "asm/signal.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "asm/pgalloc.h" +#include "asm/tlbflush.h" +#include "user_util.h" +#include "kern_util.h" +#include "irq_user.h" +#include "time_user.h" +#include "signal_user.h" +#include "mem_user.h" +#include "os.h" +#include "tlb.h" +#include "mode.h" + +static int exec_tramp(void *sig_stack) +{ + init_new_thread_stack(sig_stack, NULL); + init_new_thread_signals(1); + os_stop_process(os_getpid()); + return(0); +} + +void flush_thread_tt(void) +{ + unsigned long stack; + int new_pid; + + stack = alloc_stack(0, 0); + if(stack == 0){ + printk(KERN_ERR + "flush_thread : failed to allocate temporary stack\n"); + do_exit(SIGKILL); + } + + new_pid = start_fork_tramp(current->thread_info, stack, 0, exec_tramp); + if(new_pid < 0){ + printk(KERN_ERR + "flush_thread : new thread failed, errno = %d\n", + -new_pid); + do_exit(SIGKILL); + } + + if(current_thread->cpu == 0) + forward_interrupts(new_pid); + current->thread.request.op = OP_EXEC; + current->thread.request.u.exec.pid = new_pid; + unprotect_stack((unsigned long) current_thread); + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); + + change_sig(SIGUSR1, 0); + enable_timer(); + free_page(stack); + protect_memory(uml_reserved, high_physmem - uml_reserved, 1, 1, 0, 1); + task_protections((unsigned long) current_thread); + force_flush_all(); + unblock_signals(); +} + +void start_thread_tt(struct pt_regs *regs, unsigned long eip, + unsigned long esp) +{ + set_fs(USER_DS); + flush_tlb_mm(current->mm); + PT_REGS_IP(regs) = eip; + PT_REGS_SP(regs) = esp; + PT_FIX_EXEC_STACK(esp); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/exec_user.c b/arch/um/kernel/tt/exec_user.c new file mode 100644 index 00000000000..a92c02ff2ce --- /dev/null +++ b/arch/um/kernel/tt/exec_user.c @@ -0,0 +1,57 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <stdlib.h> +#include <sched.h> +#include <errno.h> +#include <sys/wait.h> +#include <signal.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "ptrace_user.h" +#include "os.h" + +void do_exec(int old_pid, int new_pid) +{ + unsigned long regs[FRAME_SIZE]; + int err; + + if((ptrace(PTRACE_ATTACH, new_pid, 0, 0) < 0) || + (ptrace(PTRACE_CONT, new_pid, 0, 0) < 0)) + tracer_panic("do_exec failed to attach proc - errno = %d", + errno); + + CATCH_EINTR(err = waitpid(new_pid, 0, WUNTRACED)); + if (err < 0) + tracer_panic("do_exec failed to attach proc in waitpid - errno = %d", + errno); + + if(ptrace_getregs(old_pid, regs) < 0) + tracer_panic("do_exec failed to get registers - errno = %d", + errno); + + os_kill_ptraced_process(old_pid, 0); + + if (ptrace(PTRACE_OLDSETOPTIONS, new_pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) + tracer_panic("do_exec: PTRACE_SETOPTIONS failed, errno = %d", errno); + + if(ptrace_setregs(new_pid, regs) < 0) + tracer_panic("do_exec failed to start new proc - errno = %d", + errno); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/gdb.c b/arch/um/kernel/tt/gdb.c new file mode 100644 index 00000000000..19a0ad7b35b --- /dev/null +++ b/arch/um/kernel/tt/gdb.c @@ -0,0 +1,278 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <errno.h> +#include <string.h> +#include <signal.h> +#include <sys/types.h> +#include "ptrace_user.h" +#include "uml-config.h" +#include "kern_constants.h" +#include "chan_user.h" +#include "init.h" +#include "user.h" +#include "debug.h" +#include "kern_util.h" +#include "user_util.h" +#include "tt.h" +#include "sysdep/thread.h" + +extern int debugger_pid; +extern int debugger_fd; +extern int debugger_parent; + +int detach(int pid, int sig) +{ + return(ptrace(PTRACE_DETACH, pid, 0, sig)); +} + +int attach(int pid) +{ + int err; + + err = ptrace(PTRACE_ATTACH, pid, 0, 0); + if(err < 0) return(-errno); + else return(err); +} + +int cont(int pid) +{ + return(ptrace(PTRACE_CONT, pid, 0, 0)); +} + +#ifdef UML_CONFIG_PT_PROXY + +int debugger_signal(int status, pid_t pid) +{ + return(debugger_proxy(status, pid)); +} + +void child_signal(pid_t pid, int status) +{ + child_proxy(pid, status); +} + +static void gdb_announce(char *dev_name, int dev) +{ + printf("gdb assigned device '%s'\n", dev_name); +} + +static struct chan_opts opts = { + .announce = gdb_announce, + .xterm_title = "UML kernel debugger", + .raw = 0, + .tramp_stack = 0, + .in_kernel = 0, +}; + +/* Accessed by the tracing thread, which automatically serializes access */ +static void *xterm_data; +static int xterm_fd; + +extern void *xterm_init(char *, int, struct chan_opts *); +extern int xterm_open(int, int, int, void *, char **); +extern void xterm_close(int, void *); + +int open_gdb_chan(void) +{ + char stack[UM_KERN_PAGE_SIZE], *dummy; + + opts.tramp_stack = (unsigned long) stack; + xterm_data = xterm_init("", 0, &opts); + xterm_fd = xterm_open(1, 1, 1, xterm_data, &dummy); + return(xterm_fd); +} + +static void exit_debugger_cb(void *unused) +{ + if(debugger_pid != -1){ + if(gdb_pid != -1){ + fake_child_exit(); + gdb_pid = -1; + } + else kill_child_dead(debugger_pid); + debugger_pid = -1; + if(debugger_parent != -1) + detach(debugger_parent, SIGINT); + } + if(xterm_data != NULL) xterm_close(xterm_fd, xterm_data); +} + +static void exit_debugger(void) +{ + initial_thread_cb(exit_debugger_cb, NULL); +} + +__uml_exitcall(exit_debugger); + +struct gdb_data { + char *str; + int err; +}; + +static void config_gdb_cb(void *arg) +{ + struct gdb_data *data = arg; + void *task; + int pid; + + data->err = -1; + if(debugger_pid != -1) exit_debugger_cb(NULL); + if(!strncmp(data->str, "pid,", strlen("pid,"))){ + data->str += strlen("pid,"); + pid = strtoul(data->str, NULL, 0); + task = cpu_tasks[0].task; + debugger_pid = attach_debugger(TASK_EXTERN_PID(task), pid, 0); + if(debugger_pid != -1){ + data->err = 0; + gdb_pid = pid; + } + return; + } + data->err = 0; + debugger_pid = start_debugger(linux_prog, 0, 0, &debugger_fd); + init_proxy(debugger_pid, 0, 0); +} + +int gdb_config(char *str) +{ + struct gdb_data data; + + if(*str++ != '=') return(-1); + data.str = str; + initial_thread_cb(config_gdb_cb, &data); + return(data.err); +} + +void remove_gdb_cb(void *unused) +{ + exit_debugger_cb(NULL); +} + +int gdb_remove(char *unused) +{ + initial_thread_cb(remove_gdb_cb, NULL); + return(0); +} + +void signal_usr1(int sig) +{ + if(debugger_pid != -1){ + printf("The debugger is already running\n"); + return; + } + debugger_pid = start_debugger(linux_prog, 0, 0, &debugger_fd); + init_proxy(debugger_pid, 0, 0); +} + +int init_ptrace_proxy(int idle_pid, int startup, int stop) +{ + int pid, status; + + pid = start_debugger(linux_prog, startup, stop, &debugger_fd); + status = wait_for_stop(idle_pid, SIGSTOP, PTRACE_CONT, NULL); + if(pid < 0){ + cont(idle_pid); + return(-1); + } + init_proxy(pid, 1, status); + return(pid); +} + +int attach_debugger(int idle_pid, int pid, int stop) +{ + int status = 0, err; + + err = attach(pid); + if(err < 0){ + printf("Failed to attach pid %d, errno = %d\n", pid, -err); + return(-1); + } + if(stop) status = wait_for_stop(idle_pid, SIGSTOP, PTRACE_CONT, NULL); + init_proxy(pid, 1, status); + return(pid); +} + +#ifdef notdef /* Put this back in when it does something useful */ +static int __init uml_gdb_init_setup(char *line, int *add) +{ + gdb_init = uml_strdup(line); + return 0; +} + +__uml_setup("gdb=", uml_gdb_init_setup, +"gdb=<channel description>\n\n" +); +#endif + +static int __init uml_gdb_pid_setup(char *line, int *add) +{ + gdb_pid = strtoul(line, NULL, 0); + *add = 0; + return 0; +} + +__uml_setup("gdb-pid=", uml_gdb_pid_setup, +"gdb-pid=<pid>\n" +" gdb-pid is used to attach an external debugger to UML. This may be\n" +" an already-running gdb or a debugger-like process like strace.\n\n" +); + +#else + +int debugger_signal(int status, pid_t pid){ return(0); } +void child_signal(pid_t pid, int status){ } +int init_ptrace_proxy(int idle_pid, int startup, int stop) +{ + printf("debug requested when CONFIG_PT_PROXY is off\n"); + kill_child_dead(idle_pid); + exit(1); +} + +void signal_usr1(int sig) +{ + printf("debug requested when CONFIG_PT_PROXY is off\n"); +} + +int attach_debugger(int idle_pid, int pid, int stop) +{ + printf("attach_debugger called when CONFIG_PT_PROXY " + "is off\n"); + return(-1); +} + +int config_gdb(char *str) +{ + return(-1); +} + +int remove_gdb(void) +{ + return(-1); +} + +int init_parent_proxy(int pid) +{ + return(-1); +} + +void debugger_parent_signal(int status, int pid) +{ +} + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/gdb_kern.c b/arch/um/kernel/tt/gdb_kern.c new file mode 100644 index 00000000000..93fb121f86a --- /dev/null +++ b/arch/um/kernel/tt/gdb_kern.c @@ -0,0 +1,40 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/init.h" +#include "linux/config.h" +#include "mconsole_kern.h" + +#ifdef CONFIG_MCONSOLE + +extern int gdb_config(char *str); +extern int gdb_remove(char *unused); + +static struct mc_device gdb_mc = { + .name = "gdb", + .config = gdb_config, + .remove = gdb_remove, +}; + +int gdb_mc_init(void) +{ + mconsole_register_dev(&gdb_mc); + return(0); +} + +__initcall(gdb_mc_init); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/debug.h b/arch/um/kernel/tt/include/debug.h new file mode 100644 index 00000000000..8eff674107c --- /dev/null +++ b/arch/um/kernel/tt/include/debug.h @@ -0,0 +1,29 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) and + * Lars Brinkhoff. + * Licensed under the GPL + */ + +#ifndef __DEBUG_H +#define __DEBUG_H + +extern int debugger_proxy(int status, pid_t pid); +extern void child_proxy(pid_t pid, int status); +extern void init_proxy (pid_t pid, int waiting, int status); +extern int start_debugger(char *prog, int startup, int stop, int *debugger_fd); +extern void fake_child_exit(void); +extern int gdb_config(char *str); +extern int gdb_remove(char *unused); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/mmu-tt.h b/arch/um/kernel/tt/include/mmu-tt.h new file mode 100644 index 00000000000..0440510ab3f --- /dev/null +++ b/arch/um/kernel/tt/include/mmu-tt.h @@ -0,0 +1,23 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_MMU_H +#define __TT_MMU_H + +struct mmu_context_tt { +}; + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/mode-tt.h b/arch/um/kernel/tt/include/mode-tt.h new file mode 100644 index 00000000000..efe46201906 --- /dev/null +++ b/arch/um/kernel/tt/include/mode-tt.h @@ -0,0 +1,35 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __MODE_TT_H__ +#define __MODE_TT_H__ + +#include "sysdep/ptrace.h" + +enum { OP_NONE, OP_EXEC, OP_FORK, OP_TRACE_ON, OP_REBOOT, OP_HALT, OP_CB }; + +extern int tracing_pid; + +extern int tracer(int (*init_proc)(void *), void *sp); +extern void user_time_init_tt(void); +extern void sig_handler_common_tt(int sig, void *sc); +extern void syscall_handler_tt(int sig, union uml_pt_regs *regs); +extern void reboot_tt(void); +extern void halt_tt(void); +extern int is_tracer_winch(int pid, int fd, void *data); +extern void kill_off_processes_tt(void); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/mode_kern-tt.h b/arch/um/kernel/tt/include/mode_kern-tt.h new file mode 100644 index 00000000000..28aaab3448f --- /dev/null +++ b/arch/um/kernel/tt/include/mode_kern-tt.h @@ -0,0 +1,53 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_MODE_KERN_H__ +#define __TT_MODE_KERN_H__ + +#include "linux/sched.h" +#include "asm/page.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" + +extern void *switch_to_tt(void *prev, void *next); +extern void flush_thread_tt(void); +extern void start_thread_tt(struct pt_regs *regs, unsigned long eip, + unsigned long esp); +extern int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct *p, + struct pt_regs *regs); +extern void release_thread_tt(struct task_struct *task); +extern void exit_thread_tt(void); +extern void initial_thread_cb_tt(void (*proc)(void *), void *arg); +extern void init_idle_tt(void); +extern void flush_tlb_kernel_range_tt(unsigned long start, unsigned long end); +extern void flush_tlb_kernel_vm_tt(void); +extern void __flush_tlb_one_tt(unsigned long addr); +extern void flush_tlb_range_tt(struct vm_area_struct *vma, + unsigned long start, unsigned long end); +extern void flush_tlb_mm_tt(struct mm_struct *mm); +extern void force_flush_all_tt(void); +extern long execute_syscall_tt(void *r); +extern void before_mem_tt(unsigned long brk_start); +extern unsigned long set_task_sizes_tt(int arg, unsigned long *host_size_out, + unsigned long *task_size_out); +extern int start_uml_tt(void); +extern int external_pid_tt(struct task_struct *task); +extern int thread_pid_tt(struct task_struct *task); + +#define kmem_end_tt (host_task_size - ABOVE_KMEM) + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/tt.h b/arch/um/kernel/tt/include/tt.h new file mode 100644 index 00000000000..c667b67af40 --- /dev/null +++ b/arch/um/kernel/tt/include/tt.h @@ -0,0 +1,46 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#ifndef __TT_H__ +#define __TT_H__ + +#include "sysdep/ptrace.h" + +extern int gdb_pid; +extern int debug; +extern int debug_stop; +extern int debug_trace; + +extern int honeypot; + +extern int fork_tramp(void *sig_stack); +extern int do_proc_op(void *t, int proc_id); +extern int tracer(int (*init_proc)(void *), void *sp); +extern void attach_process(int pid); +extern void tracer_panic(char *format, ...); +extern void set_init_pid(int pid); +extern int set_user_mode(void *task); +extern void set_tracing(void *t, int tracing); +extern int is_tracing(void *task); +extern void syscall_handler(int sig, union uml_pt_regs *regs); +extern void exit_kernel(int pid, void *task); +extern void do_syscall(void *task, int pid, int local_using_sysemu); +extern void do_sigtrap(void *task); +extern int is_valid_pid(int pid); +extern void remap_data(void *segment_start, void *segment_end, int w); +extern long execute_syscall_tt(void *r); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/include/uaccess-tt.h b/arch/um/kernel/tt/include/uaccess-tt.h new file mode 100644 index 00000000000..f0bad010ceb --- /dev/null +++ b/arch/um/kernel/tt/include/uaccess-tt.h @@ -0,0 +1,71 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#ifndef __TT_UACCESS_H +#define __TT_UACCESS_H + +#include "linux/string.h" +#include "linux/sched.h" +#include "asm/processor.h" +#include "asm/errno.h" +#include "asm/current.h" +#include "asm/a.out.h" +#include "uml_uaccess.h" + +#define ABOVE_KMEM (16 * 1024 * 1024) + +extern unsigned long end_vm; +extern unsigned long uml_physmem; + +#define under_task_size(addr, size) \ + (((unsigned long) (addr) < TASK_SIZE) && \ + (((unsigned long) (addr) + (size)) < TASK_SIZE)) + +#define is_stack(addr, size) \ + (((unsigned long) (addr) < STACK_TOP) && \ + ((unsigned long) (addr) >= STACK_TOP - ABOVE_KMEM) && \ + (((unsigned long) (addr) + (size)) <= STACK_TOP)) + +#define access_ok_tt(type, addr, size) \ + ((type == VERIFY_READ) || (segment_eq(get_fs(), KERNEL_DS)) || \ + (((unsigned long) (addr) <= ((unsigned long) (addr) + (size))) && \ + (under_task_size(addr, size) || is_stack(addr, size)))) + +static inline int verify_area_tt(int type, const void * addr, + unsigned long size) +{ + return(access_ok_tt(type, addr, size) ? 0 : -EFAULT); +} + +extern unsigned long get_fault_addr(void); + +extern int __do_copy_from_user(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher); +extern int __do_strncpy_from_user(char *dst, const char *src, size_t n, + void **fault_addr, void **fault_catcher); +extern int __do_clear_user(void *mem, size_t len, void **fault_addr, + void **fault_catcher); +extern int __do_strnlen_user(const char *str, unsigned long n, + void **fault_addr, void **fault_catcher); + +extern int copy_from_user_tt(void *to, const void *from, int n); +extern int copy_to_user_tt(void *to, const void *from, int n); +extern int strncpy_from_user_tt(char *dst, const char *src, int count); +extern int __clear_user_tt(void *mem, int len); +extern int clear_user_tt(void *mem, int len); +extern int strnlen_user_tt(const void *str, int len); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ksyms.c b/arch/um/kernel/tt/ksyms.c new file mode 100644 index 00000000000..92ec85d67c7 --- /dev/null +++ b/arch/um/kernel/tt/ksyms.c @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/module.h" +#include "asm/uaccess.h" +#include "mode.h" + +EXPORT_SYMBOL(__do_copy_from_user); +EXPORT_SYMBOL(__do_copy_to_user); +EXPORT_SYMBOL(__do_strncpy_from_user); +EXPORT_SYMBOL(__do_strnlen_user); +EXPORT_SYMBOL(__do_clear_user); + +EXPORT_SYMBOL(tracing_pid); +EXPORT_SYMBOL(honeypot); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/mem.c b/arch/um/kernel/tt/mem.c new file mode 100644 index 00000000000..74346a04a2b --- /dev/null +++ b/arch/um/kernel/tt/mem.c @@ -0,0 +1,51 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/config.h" +#include "linux/mm.h" +#include "asm/uaccess.h" +#include "mem_user.h" +#include "kern_util.h" +#include "user_util.h" +#include "kern.h" +#include "tt.h" + +void before_mem_tt(unsigned long brk_start) +{ + if(debug) + remap_data(UML_ROUND_DOWN(&_stext), UML_ROUND_UP(&_etext), 1); + remap_data(UML_ROUND_DOWN(&_sdata), UML_ROUND_UP(&_edata), 1); + remap_data(UML_ROUND_DOWN(&__bss_start), UML_ROUND_UP(&_end), 1); +} + +#ifdef CONFIG_HOST_2G_2G +#define TOP 0x80000000 +#else +#define TOP 0xc0000000 +#endif + +#define SIZE ((CONFIG_NEST_LEVEL + CONFIG_KERNEL_HALF_GIGS) * 0x20000000) +#define START (TOP - SIZE) + +unsigned long set_task_sizes_tt(int arg, unsigned long *host_size_out, + unsigned long *task_size_out) +{ + /* Round up to the nearest 4M */ + *host_size_out = ROUND_4M((unsigned long) &arg); + *task_size_out = START; + return(START); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/mem_user.c b/arch/um/kernel/tt/mem_user.c new file mode 100644 index 00000000000..3085267459b --- /dev/null +++ b/arch/um/kernel/tt/mem_user.c @@ -0,0 +1,49 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <stdio.h> +#include <unistd.h> +#include <string.h> +#include <errno.h> +#include <sys/mman.h> +#include "tt.h" +#include "mem_user.h" +#include "user_util.h" + +void remap_data(void *segment_start, void *segment_end, int w) +{ + void *addr; + unsigned long size; + int data, prot; + + if(w) prot = PROT_WRITE; + else prot = 0; + prot |= PROT_READ | PROT_EXEC; + size = (unsigned long) segment_end - + (unsigned long) segment_start; + data = create_mem_file(size); + addr = mmap(NULL, size, PROT_WRITE | PROT_READ, MAP_SHARED, data, 0); + if(addr == MAP_FAILED){ + perror("mapping new data segment"); + exit(1); + } + memcpy(addr, segment_start, size); + if(switcheroo(data, prot, addr, segment_start, size) < 0){ + printf("switcheroo failed\n"); + exit(1); + } +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/process_kern.c b/arch/um/kernel/tt/process_kern.c new file mode 100644 index 00000000000..f19f7c18feb --- /dev/null +++ b/arch/um/kernel/tt/process_kern.c @@ -0,0 +1,476 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "linux/signal.h" +#include "linux/kernel.h" +#include "linux/interrupt.h" +#include "linux/ptrace.h" +#include "asm/system.h" +#include "asm/pgalloc.h" +#include "asm/ptrace.h" +#include "asm/tlbflush.h" +#include "irq_user.h" +#include "signal_user.h" +#include "kern_util.h" +#include "user_util.h" +#include "os.h" +#include "kern.h" +#include "sigcontext.h" +#include "time_user.h" +#include "mem_user.h" +#include "tlb.h" +#include "mode.h" +#include "init.h" +#include "tt.h" + +void *switch_to_tt(void *prev, void *next, void *last) +{ + struct task_struct *from, *to, *prev_sched; + unsigned long flags; + int err, vtalrm, alrm, prof, cpu; + char c; + /* jailing and SMP are incompatible, so this doesn't need to be + * made per-cpu + */ + static int reading; + + from = prev; + to = next; + + to->thread.prev_sched = from; + + cpu = from->thread_info->cpu; + if(cpu == 0) + forward_interrupts(to->thread.mode.tt.extern_pid); +#ifdef CONFIG_SMP + forward_ipi(cpu_data[cpu].ipi_pipe[0], to->thread.mode.tt.extern_pid); +#endif + local_irq_save(flags); + + vtalrm = change_sig(SIGVTALRM, 0); + alrm = change_sig(SIGALRM, 0); + prof = change_sig(SIGPROF, 0); + + forward_pending_sigio(to->thread.mode.tt.extern_pid); + + c = 0; + set_current(to); + + reading = 0; + err = os_write_file(to->thread.mode.tt.switch_pipe[1], &c, sizeof(c)); + if(err != sizeof(c)) + panic("write of switch_pipe failed, err = %d", -err); + + reading = 1; + if((from->exit_state == EXIT_ZOMBIE) || + (from->exit_state == EXIT_DEAD)) + os_kill_process(os_getpid(), 0); + + err = os_read_file(from->thread.mode.tt.switch_pipe[0], &c, sizeof(c)); + if(err != sizeof(c)) + panic("read of switch_pipe failed, errno = %d", -err); + + /* If the process that we have just scheduled away from has exited, + * then it needs to be killed here. The reason is that, even though + * it will kill itself when it next runs, that may be too late. Its + * stack will be freed, possibly before then, and if that happens, + * we have a use-after-free situation. So, it gets killed here + * in case it has not already killed itself. + */ + prev_sched = current->thread.prev_sched; + if((prev_sched->exit_state == EXIT_ZOMBIE) || + (prev_sched->exit_state == EXIT_DEAD)) + os_kill_process(prev_sched->thread.mode.tt.extern_pid, 1); + + change_sig(SIGVTALRM, vtalrm); + change_sig(SIGALRM, alrm); + change_sig(SIGPROF, prof); + + arch_switch(); + + flush_tlb_all(); + local_irq_restore(flags); + + return(current->thread.prev_sched); +} + +void release_thread_tt(struct task_struct *task) +{ + int pid = task->thread.mode.tt.extern_pid; + + if(os_getpid() != pid) + os_kill_process(pid, 0); +} + +void exit_thread_tt(void) +{ + os_close_file(current->thread.mode.tt.switch_pipe[0]); + os_close_file(current->thread.mode.tt.switch_pipe[1]); +} + +void suspend_new_thread(int fd) +{ + int err; + char c; + + os_stop_process(os_getpid()); + err = os_read_file(fd, &c, sizeof(c)); + if(err != sizeof(c)) + panic("read failed in suspend_new_thread, err = %d", -err); +} + +void schedule_tail(task_t *prev); + +static void new_thread_handler(int sig) +{ + unsigned long disable; + int (*fn)(void *); + void *arg; + + fn = current->thread.request.u.thread.proc; + arg = current->thread.request.u.thread.arg; + + UPT_SC(¤t->thread.regs.regs) = (void *) (&sig + 1); + disable = (1 << (SIGVTALRM - 1)) | (1 << (SIGALRM - 1)) | + (1 << (SIGIO - 1)) | (1 << (SIGPROF - 1)); + SC_SIGMASK(UPT_SC(¤t->thread.regs.regs)) &= ~disable; + + suspend_new_thread(current->thread.mode.tt.switch_pipe[0]); + + force_flush_all(); + if(current->thread.prev_sched != NULL) + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; + + init_new_thread_signals(1); + enable_timer(); + free_page(current->thread.temp_stack); + set_cmdline("(kernel thread)"); + + change_sig(SIGUSR1, 1); + change_sig(SIGVTALRM, 1); + change_sig(SIGPROF, 1); + local_irq_enable(); + if(!run_kernel_thread(fn, arg, ¤t->thread.exec_buf)) + do_exit(0); + + /* XXX No set_user_mode here because a newly execed process will + * immediately segfault on its non-existent IP, coming straight back + * to the signal handler, which will call set_user_mode on its way + * out. This should probably change since it's confusing. + */ +} + +static int new_thread_proc(void *stack) +{ + /* local_irq_disable is needed to block out signals until this thread is + * properly scheduled. Otherwise, the tracing thread will get mighty + * upset about any signals that arrive before that. + * This has the complication that it sets the saved signal mask in + * the sigcontext to block signals. This gets restored when this + * thread (or a descendant, since they get a copy of this sigcontext) + * returns to userspace. + * So, this is compensated for elsewhere. + * XXX There is still a small window until local_irq_disable() actually + * finishes where signals are possible - shouldn't be a problem in + * practice since SIGIO hasn't been forwarded here yet, and the + * local_irq_disable should finish before a SIGVTALRM has time to be + * delivered. + */ + + local_irq_disable(); + init_new_thread_stack(stack, new_thread_handler); + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); + return(0); +} + +/* Signal masking - signals are blocked at the start of fork_tramp. They + * are re-enabled when finish_fork_handler is entered by fork_tramp hitting + * itself with a SIGUSR1. set_user_mode has to be run with SIGUSR1 off, + * so it is blocked before it's called. They are re-enabled on sigreturn + * despite the fact that they were blocked when the SIGUSR1 was issued because + * copy_thread copies the parent's sigcontext, including the signal mask + * onto the signal frame. + */ + +void finish_fork_handler(int sig) +{ + UPT_SC(¤t->thread.regs.regs) = (void *) (&sig + 1); + suspend_new_thread(current->thread.mode.tt.switch_pipe[0]); + + force_flush_all(); + if(current->thread.prev_sched != NULL) + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; + + enable_timer(); + change_sig(SIGVTALRM, 1); + local_irq_enable(); + if(current->mm != current->parent->mm) + protect_memory(uml_reserved, high_physmem - uml_reserved, 1, + 1, 0, 1); + task_protections((unsigned long) current_thread); + + free_page(current->thread.temp_stack); + local_irq_disable(); + change_sig(SIGUSR1, 0); + set_user_mode(current); +} + +int fork_tramp(void *stack) +{ + local_irq_disable(); + arch_init_thread(); + init_new_thread_stack(stack, finish_fork_handler); + + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); + return(0); +} + +int copy_thread_tt(int nr, unsigned long clone_flags, unsigned long sp, + unsigned long stack_top, struct task_struct * p, + struct pt_regs *regs) +{ + int (*tramp)(void *); + int new_pid, err; + unsigned long stack; + + if(current->thread.forking) + tramp = fork_tramp; + else { + tramp = new_thread_proc; + p->thread.request.u.thread = current->thread.request.u.thread; + } + + err = os_pipe(p->thread.mode.tt.switch_pipe, 1, 1); + if(err < 0){ + printk("copy_thread : pipe failed, err = %d\n", -err); + return(err); + } + + stack = alloc_stack(0, 0); + if(stack == 0){ + printk(KERN_ERR "copy_thread : failed to allocate " + "temporary stack\n"); + return(-ENOMEM); + } + + clone_flags &= CLONE_VM; + p->thread.temp_stack = stack; + new_pid = start_fork_tramp(p->thread_info, stack, clone_flags, tramp); + if(new_pid < 0){ + printk(KERN_ERR "copy_thread : clone failed - errno = %d\n", + -new_pid); + return(new_pid); + } + + if(current->thread.forking){ + sc_to_sc(UPT_SC(&p->thread.regs.regs), + UPT_SC(¤t->thread.regs.regs)); + SC_SET_SYSCALL_RETURN(UPT_SC(&p->thread.regs.regs), 0); + if(sp != 0) SC_SP(UPT_SC(&p->thread.regs.regs)) = sp; + } + p->thread.mode.tt.extern_pid = new_pid; + + current->thread.request.op = OP_FORK; + current->thread.request.u.fork.pid = new_pid; + os_usr1_process(os_getpid()); + + /* Enable the signal and then disable it to ensure that it is handled + * here, and nowhere else. + */ + change_sig(SIGUSR1, 1); + + change_sig(SIGUSR1, 0); + err = 0; + return(err); +} + +void reboot_tt(void) +{ + current->thread.request.op = OP_REBOOT; + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); +} + +void halt_tt(void) +{ + current->thread.request.op = OP_HALT; + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); +} + +void kill_off_processes_tt(void) +{ + struct task_struct *p; + int me; + + me = os_getpid(); + for_each_process(p){ + if(p->thread.mode.tt.extern_pid != me) + os_kill_process(p->thread.mode.tt.extern_pid, 0); + } + if(init_task.thread.mode.tt.extern_pid != me) + os_kill_process(init_task.thread.mode.tt.extern_pid, 0); +} + +void initial_thread_cb_tt(void (*proc)(void *), void *arg) +{ + if(os_getpid() == tracing_pid){ + (*proc)(arg); + } + else { + current->thread.request.op = OP_CB; + current->thread.request.u.cb.proc = proc; + current->thread.request.u.cb.arg = arg; + os_usr1_process(os_getpid()); + change_sig(SIGUSR1, 1); + + change_sig(SIGUSR1, 0); + } +} + +int do_proc_op(void *t, int proc_id) +{ + struct task_struct *task; + struct thread_struct *thread; + int op, pid; + + task = t; + thread = &task->thread; + op = thread->request.op; + switch(op){ + case OP_NONE: + case OP_TRACE_ON: + break; + case OP_EXEC: + pid = thread->request.u.exec.pid; + do_exec(thread->mode.tt.extern_pid, pid); + thread->mode.tt.extern_pid = pid; + cpu_tasks[task->thread_info->cpu].pid = pid; + break; + case OP_FORK: + attach_process(thread->request.u.fork.pid); + break; + case OP_CB: + (*thread->request.u.cb.proc)(thread->request.u.cb.arg); + break; + case OP_REBOOT: + case OP_HALT: + break; + default: + tracer_panic("Bad op in do_proc_op"); + break; + } + thread->request.op = OP_NONE; + return(op); +} + +void init_idle_tt(void) +{ + default_idle(); +} + +extern void start_kernel(void); + +static int start_kernel_proc(void *unused) +{ + int pid; + + block_signals(); + pid = os_getpid(); + + cpu_tasks[0].pid = pid; + cpu_tasks[0].task = current; +#ifdef CONFIG_SMP + cpu_online_map = cpumask_of_cpu(0); +#endif + if(debug) os_stop_process(pid); + start_kernel(); + return(0); +} + +void set_tracing(void *task, int tracing) +{ + ((struct task_struct *) task)->thread.mode.tt.tracing = tracing; +} + +int is_tracing(void *t) +{ + return (((struct task_struct *) t)->thread.mode.tt.tracing); +} + +int set_user_mode(void *t) +{ + struct task_struct *task; + + task = t ? t : current; + if(task->thread.mode.tt.tracing) + return(1); + task->thread.request.op = OP_TRACE_ON; + os_usr1_process(os_getpid()); + return(0); +} + +void set_init_pid(int pid) +{ + int err; + + init_task.thread.mode.tt.extern_pid = pid; + err = os_pipe(init_task.thread.mode.tt.switch_pipe, 1, 1); + if(err) + panic("Can't create switch pipe for init_task, errno = %d", + -err); +} + +int start_uml_tt(void) +{ + void *sp; + int pages; + + pages = (1 << CONFIG_KERNEL_STACK_ORDER); + sp = (void *) ((unsigned long) init_task.thread_info) + + pages * PAGE_SIZE - sizeof(unsigned long); + return(tracer(start_kernel_proc, sp)); +} + +int external_pid_tt(struct task_struct *task) +{ + return(task->thread.mode.tt.extern_pid); +} + +int thread_pid_tt(struct task_struct *task) +{ + return(task->thread.mode.tt.extern_pid); +} + +int is_valid_pid(int pid) +{ + struct task_struct *task; + + read_lock(&tasklist_lock); + for_each_process(task){ + if(task->thread.mode.tt.extern_pid == pid){ + read_unlock(&tasklist_lock); + return(1); + } + } + read_unlock(&tasklist_lock); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/Makefile b/arch/um/kernel/tt/ptproxy/Makefile new file mode 100644 index 00000000000..3ad5b774de5 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/Makefile @@ -0,0 +1,10 @@ +# +# Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +# Licensed under the GPL +# + +obj-y = proxy.o ptrace.o sysdep.o wait.o + +USER_OBJS := $(obj-y) + +include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/tt/ptproxy/proxy.c b/arch/um/kernel/tt/ptproxy/proxy.c new file mode 100644 index 00000000000..58800c50b10 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/proxy.c @@ -0,0 +1,377 @@ +/********************************************************************** +proxy.c + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. + +Jeff Dike (jdike@karaya.com) : Modified for integration into uml +**********************************************************************/ + +/* XXX This file shouldn't refer to CONFIG_* */ + +#include <errno.h> +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <signal.h> +#include <string.h> +#include <termios.h> +#include <sys/wait.h> +#include <sys/types.h> +#include <sys/ioctl.h> +#include <asm/unistd.h> +#include "ptrace_user.h" + +#include "ptproxy.h" +#include "sysdep.h" +#include "wait.h" + +#include "user_util.h" +#include "user.h" +#include "os.h" +#include "tempfile.h" + +static int debugger_wait(debugger_state *debugger, int *status, int options, + int (*syscall)(debugger_state *debugger, pid_t child), + int (*normal_return)(debugger_state *debugger, + pid_t unused), + int (*wait_return)(debugger_state *debugger, + pid_t unused)) +{ + if(debugger->real_wait){ + debugger->handle_trace = normal_return; + syscall_continue(debugger->pid); + debugger->real_wait = 0; + return(1); + } + debugger->wait_status_ptr = status; + debugger->wait_options = options; + if((debugger->debugee != NULL) && debugger->debugee->event){ + syscall_continue(debugger->pid); + wait_for_stop(debugger->pid, SIGTRAP, PTRACE_SYSCALL, + NULL); + (*wait_return)(debugger, -1); + return(0); + } + else if(debugger->wait_options & WNOHANG){ + syscall_cancel(debugger->pid, 0); + debugger->handle_trace = syscall; + return(0); + } + else { + syscall_pause(debugger->pid); + debugger->handle_trace = wait_return; + debugger->waiting = 1; + } + return(1); +} + +/* + * Handle debugger trap, i.e. syscall. + */ + +int debugger_syscall(debugger_state *debugger, pid_t child) +{ + long arg1, arg2, arg3, arg4, arg5, result; + int syscall, ret = 0; + + syscall = get_syscall(debugger->pid, &arg1, &arg2, &arg3, &arg4, + &arg5); + + switch(syscall){ + case __NR_execve: + /* execve never returns */ + debugger->handle_trace = debugger_syscall; + break; + + case __NR_ptrace: + if(debugger->debugee->pid != 0) arg2 = debugger->debugee->pid; + if(!debugger->debugee->in_context) + child = debugger->debugee->pid; + result = proxy_ptrace(debugger, arg1, arg2, arg3, arg4, child, + &ret); + syscall_cancel(debugger->pid, result); + debugger->handle_trace = debugger_syscall; + return(ret); + +#ifdef __NR_waitpid + case __NR_waitpid: +#endif + case __NR_wait4: + if(!debugger_wait(debugger, (int *) arg2, arg3, + debugger_syscall, debugger_normal_return, + proxy_wait_return)) + return(0); + break; + + case __NR_kill: + if(!debugger->debugee->in_context) + child = debugger->debugee->pid; + if(arg1 == debugger->debugee->pid){ + result = kill(child, arg2); + syscall_cancel(debugger->pid, result); + debugger->handle_trace = debugger_syscall; + return(0); + } + else debugger->handle_trace = debugger_normal_return; + break; + + default: + debugger->handle_trace = debugger_normal_return; + } + + syscall_continue(debugger->pid); + return(0); +} + +/* Used by the tracing thread */ +static debugger_state parent; +static int parent_syscall(debugger_state *debugger, int pid); + +int init_parent_proxy(int pid) +{ + parent = ((debugger_state) { .pid = pid, + .wait_options = 0, + .wait_status_ptr = NULL, + .waiting = 0, + .real_wait = 0, + .expecting_child = 0, + .handle_trace = parent_syscall, + .debugee = NULL } ); + return(0); +} + +int parent_normal_return(debugger_state *debugger, pid_t unused) +{ + debugger->handle_trace = parent_syscall; + syscall_continue(debugger->pid); + return(0); +} + +static int parent_syscall(debugger_state *debugger, int pid) +{ + long arg1, arg2, arg3, arg4, arg5; + int syscall; + + syscall = get_syscall(pid, &arg1, &arg2, &arg3, &arg4, &arg5); + + if((syscall == __NR_wait4) +#ifdef __NR_waitpid + || (syscall == __NR_waitpid) +#endif + ){ + debugger_wait(&parent, (int *) arg2, arg3, parent_syscall, + parent_normal_return, parent_wait_return); + } + else ptrace(PTRACE_SYSCALL, pid, 0, 0); + return(0); +} + +int debugger_normal_return(debugger_state *debugger, pid_t unused) +{ + debugger->handle_trace = debugger_syscall; + syscall_continue(debugger->pid); + return(0); +} + +void debugger_cancelled_return(debugger_state *debugger, int result) +{ + debugger->handle_trace = debugger_syscall; + syscall_set_result(debugger->pid, result); + syscall_continue(debugger->pid); +} + +/* Used by the tracing thread */ +static debugger_state debugger; +static debugee_state debugee; + +void init_proxy (pid_t debugger_pid, int stopped, int status) +{ + debugger.pid = debugger_pid; + debugger.handle_trace = debugger_syscall; + debugger.debugee = &debugee; + debugger.waiting = 0; + debugger.real_wait = 0; + debugger.expecting_child = 0; + + debugee.pid = 0; + debugee.traced = 0; + debugee.stopped = stopped; + debugee.event = 0; + debugee.zombie = 0; + debugee.died = 0; + debugee.wait_status = status; + debugee.in_context = 1; +} + +int debugger_proxy(int status, int pid) +{ + int ret = 0, sig; + + if(WIFSTOPPED(status)){ + sig = WSTOPSIG(status); + if (sig == SIGTRAP) + ret = (*debugger.handle_trace)(&debugger, pid); + + else if(sig == SIGCHLD){ + if(debugger.expecting_child){ + ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig); + debugger.expecting_child = 0; + } + else if(debugger.waiting) + real_wait_return(&debugger); + else { + ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig); + debugger.real_wait = 1; + } + } + else ptrace(PTRACE_SYSCALL, debugger.pid, 0, sig); + } + else if(WIFEXITED(status)){ + tracer_panic("debugger (pid %d) exited with status %d", + debugger.pid, WEXITSTATUS(status)); + } + else if(WIFSIGNALED(status)){ + tracer_panic("debugger (pid %d) exited with signal %d", + debugger.pid, WTERMSIG(status)); + } + else { + tracer_panic("proxy got unknown status (0x%x) on debugger " + "(pid %d)", status, debugger.pid); + } + return(ret); +} + +void child_proxy(pid_t pid, int status) +{ + debugee.event = 1; + debugee.wait_status = status; + + if(WIFSTOPPED(status)){ + debugee.stopped = 1; + debugger.expecting_child = 1; + kill(debugger.pid, SIGCHLD); + } + else if(WIFEXITED(status) || WIFSIGNALED(status)){ + debugee.zombie = 1; + debugger.expecting_child = 1; + kill(debugger.pid, SIGCHLD); + } + else panic("proxy got unknown status (0x%x) on child (pid %d)", + status, pid); +} + +void debugger_parent_signal(int status, int pid) +{ + int sig; + + if(WIFSTOPPED(status)){ + sig = WSTOPSIG(status); + if(sig == SIGTRAP) (*parent.handle_trace)(&parent, pid); + else ptrace(PTRACE_SYSCALL, pid, 0, sig); + } +} + +void fake_child_exit(void) +{ + int status, pid; + + child_proxy(1, W_EXITCODE(0, 0)); + while(debugger.waiting == 1){ + CATCH_EINTR(pid = waitpid(debugger.pid, &status, WUNTRACED)); + if(pid != debugger.pid){ + printk("fake_child_exit - waitpid failed, " + "errno = %d\n", errno); + return; + } + debugger_proxy(status, debugger.pid); + } + CATCH_EINTR(pid = waitpid(debugger.pid, &status, WUNTRACED)); + if(pid != debugger.pid){ + printk("fake_child_exit - waitpid failed, " + "errno = %d\n", errno); + return; + } + if(ptrace(PTRACE_DETACH, debugger.pid, 0, SIGCONT) < 0) + printk("fake_child_exit - PTRACE_DETACH failed, errno = %d\n", + errno); +} + +char gdb_init_string[] = +"att 1 \n\ +b panic \n\ +b stop \n\ +handle SIGWINCH nostop noprint pass \n\ +"; + +int start_debugger(char *prog, int startup, int stop, int *fd_out) +{ + int slave, child; + + slave = open_gdb_chan(); + child = fork(); + if(child == 0){ + char *tempname = NULL; + int fd; + + if(setsid() < 0) perror("setsid"); + if((dup2(slave, 0) < 0) || (dup2(slave, 1) < 0) || + (dup2(slave, 2) < 0)){ + printk("start_debugger : dup2 failed, errno = %d\n", + errno); + exit(1); + } + if(ioctl(0, TIOCSCTTY, 0) < 0){ + printk("start_debugger : TIOCSCTTY failed, " + "errno = %d\n", errno); + exit(1); + } + if(tcsetpgrp (1, os_getpid()) < 0){ + printk("start_debugger : tcsetpgrp failed, " + "errno = %d\n", errno); +#ifdef notdef + exit(1); +#endif + } + fd = make_tempfile("/tmp/gdb_init-XXXXXX", &tempname, 0); + if(fd < 0){ + printk("start_debugger : make_tempfile failed," + "err = %d\n", -fd); + exit(1); + } + os_write_file(fd, gdb_init_string, sizeof(gdb_init_string) - 1); + if(startup){ + if(stop){ + os_write_file(fd, "b start_kernel\n", + strlen("b start_kernel\n")); + } + os_write_file(fd, "c\n", strlen("c\n")); + } + if(ptrace(PTRACE_TRACEME, 0, 0, 0) < 0){ + printk("start_debugger : PTRACE_TRACEME failed, " + "errno = %d\n", errno); + exit(1); + } + execlp("gdb", "gdb", "--command", tempname, prog, NULL); + printk("start_debugger : exec of gdb failed, errno = %d\n", + errno); + } + if(child < 0){ + printk("start_debugger : fork for gdb failed, errno = %d\n", + errno); + return(-1); + } + *fd_out = slave; + return(child); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/ptproxy.h b/arch/um/kernel/tt/ptproxy/ptproxy.h new file mode 100644 index 00000000000..5eb0285b196 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/ptproxy.h @@ -0,0 +1,61 @@ +/********************************************************************** +ptproxy.h + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. +**********************************************************************/ + +#ifndef __PTPROXY_H +#define __PTPROXY_H + +#include <sys/types.h> + +typedef struct debugger debugger_state; +typedef struct debugee debugee_state; + +struct debugger +{ + pid_t pid; + int wait_options; + int *wait_status_ptr; + unsigned int waiting : 1; + unsigned int real_wait : 1; + unsigned int expecting_child : 1; + int (*handle_trace) (debugger_state *, pid_t); + + debugee_state *debugee; +}; + +struct debugee +{ + pid_t pid; + int wait_status; + unsigned int died : 1; + unsigned int event : 1; + unsigned int stopped : 1; + unsigned int trace_singlestep : 1; + unsigned int trace_syscall : 1; + unsigned int traced : 1; + unsigned int zombie : 1; + unsigned int in_context : 1; +}; + +extern int debugger_syscall(debugger_state *debugger, pid_t pid); +extern int debugger_normal_return (debugger_state *debugger, pid_t unused); + +extern long proxy_ptrace (struct debugger *, int, pid_t, long, long, pid_t, + int *strace_out); +extern void debugger_cancelled_return(debugger_state *debugger, int result); + +#endif + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/ptrace.c b/arch/um/kernel/tt/ptproxy/ptrace.c new file mode 100644 index 00000000000..528a5fc8d88 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/ptrace.c @@ -0,0 +1,237 @@ +/********************************************************************** +ptrace.c + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. + +Jeff Dike (jdike@karaya.com) : Modified for integration into uml +**********************************************************************/ + +#include <errno.h> +#include <unistd.h> +#include <signal.h> +#include <sys/types.h> +#include <sys/time.h> +#include <sys/wait.h> + +#include "ptproxy.h" +#include "debug.h" +#include "user_util.h" +#include "kern_util.h" +#include "ptrace_user.h" +#include "tt.h" + +long proxy_ptrace(struct debugger *debugger, int arg1, pid_t arg2, + long arg3, long arg4, pid_t child, int *ret) +{ + sigset_t relay; + long result; + int status; + + *ret = 0; + if(debugger->debugee->died) return(-ESRCH); + + switch(arg1){ + case PTRACE_ATTACH: + if(debugger->debugee->traced) return(-EPERM); + + debugger->debugee->pid = arg2; + debugger->debugee->traced = 1; + + if(is_valid_pid(arg2) && (arg2 != child)){ + debugger->debugee->in_context = 0; + kill(arg2, SIGSTOP); + debugger->debugee->event = 1; + debugger->debugee->wait_status = W_STOPCODE(SIGSTOP); + } + else { + debugger->debugee->in_context = 1; + if(debugger->debugee->stopped) + child_proxy(child, W_STOPCODE(SIGSTOP)); + else kill(child, SIGSTOP); + } + + return(0); + + case PTRACE_DETACH: + if(!debugger->debugee->traced) return(-EPERM); + + debugger->debugee->traced = 0; + debugger->debugee->pid = 0; + if(!debugger->debugee->in_context) + kill(child, SIGCONT); + + return(0); + + case PTRACE_CONT: + if(!debugger->debugee->in_context) return(-EPERM); + *ret = PTRACE_CONT; + return(ptrace(PTRACE_CONT, child, arg3, arg4)); + +#ifdef UM_HAVE_GETFPREGS + case PTRACE_GETFPREGS: + { + long regs[FP_FRAME_SIZE]; + int i, result; + + result = ptrace(PTRACE_GETFPREGS, child, 0, regs); + if(result == -1) return(-errno); + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + ptrace(PTRACE_POKEDATA, debugger->pid, arg4 + 4 * i, + regs[i]); + return(result); + } +#endif + +#ifdef UM_HAVE_GETFPXREGS + case PTRACE_GETFPXREGS: + { + long regs[FPX_FRAME_SIZE]; + int i, result; + + result = ptrace(PTRACE_GETFPXREGS, child, 0, regs); + if(result == -1) return(-errno); + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + ptrace(PTRACE_POKEDATA, debugger->pid, arg4 + 4 * i, + regs[i]); + return(result); + } +#endif + +#ifdef UM_HAVE_GETREGS + case PTRACE_GETREGS: + { + long regs[FRAME_SIZE]; + int i, result; + + result = ptrace(PTRACE_GETREGS, child, 0, regs); + if(result == -1) return(-errno); + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + ptrace (PTRACE_POKEDATA, debugger->pid, + arg4 + 4 * i, regs[i]); + return(result); + } + break; +#endif + + case PTRACE_KILL: + result = ptrace(PTRACE_KILL, child, arg3, arg4); + if(result == -1) return(-errno); + + return(result); + + case PTRACE_PEEKDATA: + case PTRACE_PEEKTEXT: + case PTRACE_PEEKUSR: + /* The value being read out could be -1, so we have to + * check errno to see if there's an error, and zero it + * beforehand so we're not faked out by an old error + */ + + errno = 0; + result = ptrace(arg1, child, arg3, 0); + if((result == -1) && (errno != 0)) return(-errno); + + result = ptrace(PTRACE_POKEDATA, debugger->pid, arg4, result); + if(result == -1) return(-errno); + + return(result); + + case PTRACE_POKEDATA: + case PTRACE_POKETEXT: + case PTRACE_POKEUSR: + result = ptrace(arg1, child, arg3, arg4); + if(result == -1) return(-errno); + + if(arg1 == PTRACE_POKEUSR) ptrace_pokeuser(arg3, arg4); + return(result); + +#ifdef UM_HAVE_SETFPREGS + case PTRACE_SETFPREGS: + { + long regs[FP_FRAME_SIZE]; + int i; + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + regs[i] = ptrace (PTRACE_PEEKDATA, debugger->pid, + arg4 + 4 * i, 0); + result = ptrace(PTRACE_SETFPREGS, child, 0, regs); + if(result == -1) return(-errno); + + return(result); + } +#endif + +#ifdef UM_HAVE_SETFPXREGS + case PTRACE_SETFPXREGS: + { + long regs[FPX_FRAME_SIZE]; + int i; + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + regs[i] = ptrace (PTRACE_PEEKDATA, debugger->pid, + arg4 + 4 * i, 0); + result = ptrace(PTRACE_SETFPXREGS, child, 0, regs); + if(result == -1) return(-errno); + + return(result); + } +#endif + +#ifdef UM_HAVE_SETREGS + case PTRACE_SETREGS: + { + long regs[FRAME_SIZE]; + int i; + + for (i = 0; i < sizeof(regs)/sizeof(regs[0]); i++) + regs[i] = ptrace(PTRACE_PEEKDATA, debugger->pid, + arg4 + 4 * i, 0); + result = ptrace(PTRACE_SETREGS, child, 0, regs); + if(result == -1) return(-errno); + + return(result); + } +#endif + + case PTRACE_SINGLESTEP: + if(!debugger->debugee->in_context) return(-EPERM); + sigemptyset(&relay); + sigaddset(&relay, SIGSEGV); + sigaddset(&relay, SIGILL); + sigaddset(&relay, SIGBUS); + result = ptrace(PTRACE_SINGLESTEP, child, arg3, arg4); + if(result == -1) return(-errno); + + status = wait_for_stop(child, SIGTRAP, PTRACE_SINGLESTEP, + &relay); + child_proxy(child, status); + return(result); + + case PTRACE_SYSCALL: + if(!debugger->debugee->in_context) return(-EPERM); + result = ptrace(PTRACE_SYSCALL, child, arg3, arg4); + if(result == -1) return(-errno); + + *ret = PTRACE_SYSCALL; + return(result); + + case PTRACE_TRACEME: + default: + return(-EINVAL); + } +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/sysdep.c b/arch/um/kernel/tt/ptproxy/sysdep.c new file mode 100644 index 00000000000..a5f0e01e214 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/sysdep.c @@ -0,0 +1,70 @@ +/********************************************************************** +sysdep.c + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. +**********************************************************************/ + +#include <stdio.h> +#include <string.h> +#include <stdlib.h> +#include <signal.h> +#include <errno.h> +#include <sys/types.h> +#include <linux/unistd.h> +#include "ptrace_user.h" +#include "user_util.h" +#include "user.h" + +int get_syscall(pid_t pid, long *arg1, long *arg2, long *arg3, long *arg4, + long *arg5) +{ + *arg1 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG1_OFFSET, 0); + *arg2 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG2_OFFSET, 0); + *arg3 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG3_OFFSET, 0); + *arg4 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG4_OFFSET, 0); + *arg5 = ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_ARG5_OFFSET, 0); + return(ptrace(PTRACE_PEEKUSR, pid, PT_SYSCALL_NR_OFFSET, 0)); +} + +void syscall_cancel(pid_t pid, int result) +{ + if((ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, + __NR_getpid) < 0) || + (ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0) || + (wait_for_stop(pid, SIGTRAP, PTRACE_SYSCALL, NULL) < 0) || + (ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, result) < 0) || + (ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0)) + printk("ptproxy: couldn't cancel syscall: errno = %d\n", + errno); +} + +void syscall_set_result(pid_t pid, long result) +{ + ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_RET_OFFSET, result); +} + +void syscall_continue(pid_t pid) +{ + ptrace(PTRACE_SYSCALL, pid, 0, 0); +} + +int syscall_pause(pid_t pid) +{ + if(ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, __NR_pause) < 0){ + printk("syscall_change - ptrace failed, errno = %d\n", errno); + return(-1); + } + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/sysdep.h b/arch/um/kernel/tt/ptproxy/sysdep.h new file mode 100644 index 00000000000..735f488049a --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/sysdep.h @@ -0,0 +1,25 @@ +/********************************************************************** +sysdep.h + +Copyright (C) 1999 Lars Brinkhoff. +Copyright (C) 2001 Jeff Dike (jdike@karaya.com) +See the file COPYING for licensing terms and conditions. +**********************************************************************/ + +extern int get_syscall(pid_t pid, long *arg1, long *arg2, long *arg3, + long *arg4, long *arg5); +extern void syscall_cancel (pid_t pid, long result); +extern void syscall_set_result (pid_t pid, long result); +extern void syscall_continue (pid_t pid); +extern int syscall_pause(pid_t pid); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/wait.c b/arch/um/kernel/tt/ptproxy/wait.c new file mode 100644 index 00000000000..12f6319d8d7 --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/wait.c @@ -0,0 +1,86 @@ +/********************************************************************** +wait.c + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. + +**********************************************************************/ + +#include <errno.h> +#include <signal.h> +#include <sys/wait.h> + +#include "ptproxy.h" +#include "sysdep.h" +#include "wait.h" +#include "user_util.h" +#include "ptrace_user.h" +#include "sysdep/ptrace.h" +#include "sysdep/sigcontext.h" + +int proxy_wait_return(struct debugger *debugger, pid_t unused) +{ + debugger->waiting = 0; + + if(debugger->debugee->died || (debugger->wait_options & __WCLONE)){ + debugger_cancelled_return(debugger, -ECHILD); + return(0); + } + + if(debugger->debugee->zombie && debugger->debugee->event) + debugger->debugee->died = 1; + + if(debugger->debugee->event){ + debugger->debugee->event = 0; + ptrace(PTRACE_POKEDATA, debugger->pid, + debugger->wait_status_ptr, + debugger->debugee->wait_status); + /* if (wait4) + ptrace (PTRACE_POKEDATA, pid, rusage_ptr, ...); */ + debugger_cancelled_return(debugger, debugger->debugee->pid); + return(0); + } + + /* pause will return -EINTR, which happens to be right for wait */ + debugger_normal_return(debugger, -1); + return(0); +} + +int parent_wait_return(struct debugger *debugger, pid_t unused) +{ + return(debugger_normal_return(debugger, -1)); +} + +int real_wait_return(struct debugger *debugger) +{ + unsigned long ip; + int pid; + + pid = debugger->pid; + + ip = ptrace(PTRACE_PEEKUSR, pid, PT_IP_OFFSET, 0); + IP_RESTART_SYSCALL(ip); + + if(ptrace(PTRACE_POKEUSR, pid, PT_IP_OFFSET, ip) < 0) + tracer_panic("real_wait_return : Failed to restart system " + "call, errno = %d\n", errno); + + if((ptrace(PTRACE_SYSCALL, debugger->pid, 0, SIGCHLD) < 0) || + (ptrace(PTRACE_SYSCALL, debugger->pid, 0, 0) < 0) || + (ptrace(PTRACE_SYSCALL, debugger->pid, 0, 0) < 0) || + debugger_normal_return(debugger, -1)) + tracer_panic("real_wait_return : gdb failed to wait, " + "errno = %d\n", errno); + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/ptproxy/wait.h b/arch/um/kernel/tt/ptproxy/wait.h new file mode 100644 index 00000000000..542e73ee2ce --- /dev/null +++ b/arch/um/kernel/tt/ptproxy/wait.h @@ -0,0 +1,15 @@ +/********************************************************************** +wait.h + +Copyright (C) 1999 Lars Brinkhoff. See the file COPYING for licensing +terms and conditions. +**********************************************************************/ + +#ifndef __PTPROXY_WAIT_H +#define __PTPROXY_WAIT_H + +extern int proxy_wait_return(struct debugger *debugger, pid_t unused); +extern int real_wait_return(struct debugger *debugger); +extern int parent_wait_return(struct debugger *debugger, pid_t unused); + +#endif diff --git a/arch/um/kernel/tt/syscall_kern.c b/arch/um/kernel/tt/syscall_kern.c new file mode 100644 index 00000000000..2650a628719 --- /dev/null +++ b/arch/um/kernel/tt/syscall_kern.c @@ -0,0 +1,47 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/types.h" +#include "linux/utime.h" +#include "linux/sys.h" +#include "linux/ptrace.h" +#include "asm/unistd.h" +#include "asm/ptrace.h" +#include "asm/uaccess.h" +#include "asm/stat.h" +#include "sysdep/syscalls.h" +#include "kern_util.h" + +extern syscall_handler_t *sys_call_table[]; + +long execute_syscall_tt(void *r) +{ + struct pt_regs *regs = r; + long res; + int syscall; + +#ifdef CONFIG_SYSCALL_DEBUG + current->thread.nsyscalls++; + nsyscalls++; +#endif + syscall = UPT_SYSCALL_NR(®s->regs); + + if((syscall >= NR_syscalls) || (syscall < 0)) + res = -ENOSYS; + else res = EXECUTE_SYSCALL(syscall, regs); + + return(res); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/syscall_user.c b/arch/um/kernel/tt/syscall_user.c new file mode 100644 index 00000000000..e4e7e9c2224 --- /dev/null +++ b/arch/um/kernel/tt/syscall_user.c @@ -0,0 +1,90 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <unistd.h> +#include <signal.h> +#include <errno.h> +#include <asm/unistd.h> +#include "sysdep/ptrace.h" +#include "sigcontext.h" +#include "ptrace_user.h" +#include "task.h" +#include "user_util.h" +#include "kern_util.h" +#include "syscall_user.h" +#include "tt.h" + + +void syscall_handler_tt(int sig, union uml_pt_regs *regs) +{ + void *sc; + long result; + int syscall; +#ifdef UML_CONFIG_DEBUG_SYSCALL + int index; +#endif + + syscall = UPT_SYSCALL_NR(regs); + sc = UPT_SC(regs); + SC_START_SYSCALL(sc); + +#ifdef UML_CONFIG_DEBUG_SYSCALL + index = record_syscall_start(syscall); +#endif + syscall_trace(regs, 0); + result = execute_syscall_tt(regs); + + /* regs->sc may have changed while the system call ran (there may + * have been an interrupt or segfault), so it needs to be refreshed. + */ + UPT_SC(regs) = sc; + + SC_SET_SYSCALL_RETURN(sc, result); + + syscall_trace(regs, 1); +#ifdef UML_CONFIG_DEBUG_SYSCALL + record_syscall_end(index, result); +#endif +} + +void do_sigtrap(void *task) +{ + UPT_SYSCALL_NR(TASK_REGS(task)) = -1; +} + +void do_syscall(void *task, int pid, int local_using_sysemu) +{ + unsigned long proc_regs[FRAME_SIZE]; + + if(ptrace_getregs(pid, proc_regs) < 0) + tracer_panic("Couldn't read registers"); + + UPT_SYSCALL_NR(TASK_REGS(task)) = PT_SYSCALL_NR(proc_regs); + + if(((unsigned long *) PT_IP(proc_regs) >= &_stext) && + ((unsigned long *) PT_IP(proc_regs) <= &_etext)) + tracer_panic("I'm tracing myself and I can't get out"); + + /* advanced sysemu mode set syscall number to -1 automatically */ + if (local_using_sysemu==2) + return; + + /* syscall number -1 in sysemu skips syscall restarting in host */ + if(ptrace(PTRACE_POKEUSR, pid, PT_SYSCALL_NR_OFFSET, + local_using_sysemu ? -1 : __NR_getpid) < 0) + tracer_panic("do_syscall : Nullifying syscall failed, " + "errno = %d", errno); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/time.c b/arch/um/kernel/tt/time.c new file mode 100644 index 00000000000..8565b71b07c --- /dev/null +++ b/arch/um/kernel/tt/time.c @@ -0,0 +1,28 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <signal.h> +#include <sys/time.h> +#include <time_user.h> +#include "process.h" +#include "user.h" + +void user_time_init_tt(void) +{ + if(signal(SIGVTALRM, (__sighandler_t) alarm_handler) == SIG_ERR) + panic("Couldn't set SIGVTALRM handler"); + set_interval(ITIMER_VIRTUAL); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/tlb.c b/arch/um/kernel/tt/tlb.c new file mode 100644 index 00000000000..203216ad86f --- /dev/null +++ b/arch/um/kernel/tt/tlb.c @@ -0,0 +1,149 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright 2003 PathScale, Inc. + * Licensed under the GPL + */ + +#include "linux/stddef.h" +#include "linux/kernel.h" +#include "linux/sched.h" +#include "linux/mm.h" +#include "asm/page.h" +#include "asm/pgtable.h" +#include "asm/uaccess.h" +#include "asm/tlbflush.h" +#include "user_util.h" +#include "mem_user.h" +#include "os.h" +#include "tlb.h" + +static void do_ops(int unused, struct host_vm_op *ops, int last) +{ + struct host_vm_op *op; + int i; + + for(i = 0; i <= last; i++){ + op = &ops[i]; + switch(op->type){ + case MMAP: + os_map_memory((void *) op->u.mmap.addr, op->u.mmap.fd, + op->u.mmap.offset, op->u.mmap.len, + op->u.mmap.r, op->u.mmap.w, + op->u.mmap.x); + break; + case MUNMAP: + os_unmap_memory((void *) op->u.munmap.addr, + op->u.munmap.len); + break; + case MPROTECT: + protect_memory(op->u.mprotect.addr, op->u.munmap.len, + op->u.mprotect.r, op->u.mprotect.w, + op->u.mprotect.x, 1); + break; + default: + printk("Unknown op type %d in do_ops\n", op->type); + break; + } + } +} + +static void fix_range(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force) +{ + if((current->thread.mode.tt.extern_pid != -1) && + (current->thread.mode.tt.extern_pid != os_getpid())) + panic("fix_range fixing wrong address space, current = 0x%p", + current); + + fix_range_common(mm, start_addr, end_addr, force, 0, do_ops); +} + +atomic_t vmchange_seq = ATOMIC_INIT(1); + +void flush_tlb_kernel_range_tt(unsigned long start, unsigned long end) +{ + if(flush_tlb_kernel_range_common(start, end)) + atomic_inc(&vmchange_seq); +} + +static void protect_vm_page(unsigned long addr, int w, int must_succeed) +{ + int err; + + err = protect_memory(addr, PAGE_SIZE, 1, w, 1, must_succeed); + if(err == 0) return; + else if((err == -EFAULT) || (err == -ENOMEM)){ + flush_tlb_kernel_range(addr, addr + PAGE_SIZE); + protect_vm_page(addr, w, 1); + } + else panic("protect_vm_page : protect failed, errno = %d\n", err); +} + +void mprotect_kernel_vm(int w) +{ + struct mm_struct *mm; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + unsigned long addr; + + mm = &init_mm; + for(addr = start_vm; addr < end_vm;){ + pgd = pgd_offset(mm, addr); + pud = pud_offset(pgd, addr); + pmd = pmd_offset(pud, addr); + if(pmd_present(*pmd)){ + pte = pte_offset_kernel(pmd, addr); + if(pte_present(*pte)) protect_vm_page(addr, w, 0); + addr += PAGE_SIZE; + } + else addr += PMD_SIZE; + } +} + +void flush_tlb_kernel_vm_tt(void) +{ + flush_tlb_kernel_range(start_vm, end_vm); +} + +void __flush_tlb_one_tt(unsigned long addr) +{ + flush_tlb_kernel_range(addr, addr + PAGE_SIZE); +} + +void flush_tlb_range_tt(struct vm_area_struct *vma, unsigned long start, + unsigned long end) +{ + if(vma->vm_mm != current->mm) return; + + /* Assumes that the range start ... end is entirely within + * either process memory or kernel vm + */ + if((start >= start_vm) && (start < end_vm)){ + if(flush_tlb_kernel_range_common(start, end)) + atomic_inc(&vmchange_seq); + } + else fix_range(vma->vm_mm, start, end, 0); +} + +void flush_tlb_mm_tt(struct mm_struct *mm) +{ + unsigned long seq; + + if(mm != current->mm) return; + + fix_range(mm, 0, STACK_TOP, 0); + + seq = atomic_read(&vmchange_seq); + if(current->thread.mode.tt.vm_seq == seq) + return; + current->thread.mode.tt.vm_seq = seq; + flush_tlb_kernel_range_common(start_vm, end_vm); +} + +void force_flush_all_tt(void) +{ + fix_range(current->mm, 0, STACK_TOP, 1); + flush_tlb_kernel_range_common(start_vm, end_vm); +} diff --git a/arch/um/kernel/tt/tracer.c b/arch/um/kernel/tt/tracer.c new file mode 100644 index 00000000000..7b5d937e595 --- /dev/null +++ b/arch/um/kernel/tt/tracer.c @@ -0,0 +1,480 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <stdarg.h> +#include <unistd.h> +#include <signal.h> +#include <errno.h> +#include <sched.h> +#include <string.h> +#include <sys/mman.h> +#include <sys/time.h> +#include <sys/wait.h> +#include "user.h" +#include "sysdep/ptrace.h" +#include "sigcontext.h" +#include "sysdep/sigcontext.h" +#include "os.h" +#include "signal_user.h" +#include "user_util.h" +#include "mem_user.h" +#include "process.h" +#include "kern_util.h" +#include "chan_user.h" +#include "ptrace_user.h" +#include "mode.h" +#include "tt.h" + +static int tracer_winch[2]; + +int is_tracer_winch(int pid, int fd, void *data) +{ + if(pid != tracing_pid) + return(0); + + register_winch_irq(tracer_winch[0], fd, -1, data); + return(1); +} + +static void tracer_winch_handler(int sig) +{ + int n; + char c = 1; + + n = os_write_file(tracer_winch[1], &c, sizeof(c)); + if(n != sizeof(c)) + printk("tracer_winch_handler - write failed, err = %d\n", -n); +} + +/* Called only by the tracing thread during initialization */ + +static void setup_tracer_winch(void) +{ + int err; + + err = os_pipe(tracer_winch, 1, 1); + if(err < 0){ + printk("setup_tracer_winch : os_pipe failed, err = %d\n", -err); + return; + } + signal(SIGWINCH, tracer_winch_handler); +} + +void attach_process(int pid) +{ + if((ptrace(PTRACE_ATTACH, pid, 0, 0) < 0) || + (ptrace(PTRACE_CONT, pid, 0, 0) < 0)) + tracer_panic("OP_FORK failed to attach pid"); + wait_for_stop(pid, SIGSTOP, PTRACE_CONT, NULL); + if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) + tracer_panic("OP_FORK: PTRACE_SETOPTIONS failed, errno = %d", errno); + if(ptrace(PTRACE_CONT, pid, 0, 0) < 0) + tracer_panic("OP_FORK failed to continue process"); +} + +void tracer_panic(char *format, ...) +{ + va_list ap; + + va_start(ap, format); + vprintf(format, ap); + va_end(ap); + printf("\n"); + while(1) pause(); +} + +static void tracer_segv(int sig, struct sigcontext sc) +{ + printf("Tracing thread segfault at address 0x%lx, ip 0x%lx\n", + SC_FAULT_ADDR(&sc), SC_IP(&sc)); + while(1) + pause(); +} + +/* Changed early in boot, and then only read */ +int debug = 0; +int debug_stop = 1; +int debug_parent = 0; +int honeypot = 0; + +static int signal_tramp(void *arg) +{ + int (*proc)(void *); + + if(honeypot && munmap((void *) (host_task_size - 0x10000000), + 0x10000000)) + panic("Unmapping stack failed"); + if(ptrace(PTRACE_TRACEME, 0, 0, 0) < 0) + panic("ptrace PTRACE_TRACEME failed"); + os_stop_process(os_getpid()); + change_sig(SIGWINCH, 0); + signal(SIGUSR1, SIG_IGN); + change_sig(SIGCHLD, 0); + signal(SIGSEGV, (__sighandler_t) sig_handler); + set_cmdline("(idle thread)"); + set_init_pid(os_getpid()); + proc = arg; + return((*proc)(NULL)); +} + +static void sleeping_process_signal(int pid, int sig) +{ + switch(sig){ + /* These two result from UML being ^Z-ed and bg-ed. PTRACE_CONT is + * right because the process must be in the kernel already. + */ + case SIGCONT: + case SIGTSTP: + if(ptrace(PTRACE_CONT, pid, 0, sig) < 0) + tracer_panic("sleeping_process_signal : Failed to " + "continue pid %d, signal = %d, " + "errno = %d\n", pid, sig, errno); + break; + + /* This happens when the debugger (e.g. strace) is doing system call + * tracing on the kernel. During a context switch, the current task + * will be set to the incoming process and the outgoing process will + * hop into write and then read. Since it's not the current process + * any more, the trace of those will land here. So, we need to just + * PTRACE_SYSCALL it. + */ + case (SIGTRAP + 0x80): + if(ptrace(PTRACE_SYSCALL, pid, 0, 0) < 0) + tracer_panic("sleeping_process_signal : Failed to " + "PTRACE_SYSCALL pid %d, errno = %d\n", + pid, errno); + break; + case SIGSTOP: + break; + default: + tracer_panic("sleeping process %d got unexpected " + "signal : %d\n", pid, sig); + break; + } +} + +/* Accessed only by the tracing thread */ +int debugger_pid = -1; +int debugger_parent = -1; +int debugger_fd = -1; +int gdb_pid = -1; + +struct { + int pid; + int signal; + unsigned long addr; + struct timeval time; +} signal_record[1024][32]; + +int signal_index[32]; +int nsignals = 0; +int debug_trace = 0; +extern int io_nsignals, io_count, intr_count; + +extern void signal_usr1(int sig); + +int tracing_pid = -1; + +int tracer(int (*init_proc)(void *), void *sp) +{ + void *task = NULL; + int status, pid = 0, sig = 0, cont_type, tracing = 0, op = 0; + int proc_id = 0, n, err, old_tracing = 0, strace = 0; + int local_using_sysemu = 0; +#ifdef UML_CONFIG_SYSCALL_DEBUG + unsigned long eip = 0; + int last_index; +#endif + signal(SIGPIPE, SIG_IGN); + setup_tracer_winch(); + tracing_pid = os_getpid(); + printf("tracing thread pid = %d\n", tracing_pid); + + pid = clone(signal_tramp, sp, CLONE_FILES | SIGCHLD, init_proc); + CATCH_EINTR(n = waitpid(pid, &status, WUNTRACED)); + if(n < 0){ + printf("waitpid on idle thread failed, errno = %d\n", errno); + exit(1); + } + if (ptrace(PTRACE_OLDSETOPTIONS, pid, 0, (void *)PTRACE_O_TRACESYSGOOD) < 0) { + printf("Failed to PTRACE_SETOPTIONS for idle thread, errno = %d\n", errno); + exit(1); + } + if((ptrace(PTRACE_CONT, pid, 0, 0) < 0)){ + printf("Failed to continue idle thread, errno = %d\n", errno); + exit(1); + } + + signal(SIGSEGV, (sighandler_t) tracer_segv); + signal(SIGUSR1, signal_usr1); + if(debug_trace){ + printf("Tracing thread pausing to be attached\n"); + stop(); + } + if(debug){ + if(gdb_pid != -1) + debugger_pid = attach_debugger(pid, gdb_pid, 1); + else debugger_pid = init_ptrace_proxy(pid, 1, debug_stop); + if(debug_parent){ + debugger_parent = os_process_parent(debugger_pid); + init_parent_proxy(debugger_parent); + err = attach(debugger_parent); + if(err){ + printf("Failed to attach debugger parent %d, " + "errno = %d\n", debugger_parent, -err); + debugger_parent = -1; + } + else { + if(ptrace(PTRACE_SYSCALL, debugger_parent, + 0, 0) < 0){ + printf("Failed to continue debugger " + "parent, errno = %d\n", errno); + debugger_parent = -1; + } + } + } + } + set_cmdline("(tracing thread)"); + while(1){ + CATCH_EINTR(pid = waitpid(-1, &status, WUNTRACED)); + if(pid <= 0){ + if(errno != ECHILD){ + printf("wait failed - errno = %d\n", errno); + } + continue; + } + if(pid == debugger_pid){ + int cont = 0; + + if(WIFEXITED(status) || WIFSIGNALED(status)) + debugger_pid = -1; + /* XXX Figure out how to deal with gdb and SMP */ + else cont = debugger_signal(status, cpu_tasks[0].pid); + if(cont == PTRACE_SYSCALL) strace = 1; + continue; + } + else if(pid == debugger_parent){ + debugger_parent_signal(status, pid); + continue; + } + nsignals++; + if(WIFEXITED(status)) ; +#ifdef notdef + { + printf("Child %d exited with status %d\n", pid, + WEXITSTATUS(status)); + } +#endif + else if(WIFSIGNALED(status)){ + sig = WTERMSIG(status); + if(sig != 9){ + printf("Child %d exited with signal %d\n", pid, + sig); + } + } + else if(WIFSTOPPED(status)){ + proc_id = pid_to_processor_id(pid); + sig = WSTOPSIG(status); +#ifdef UML_CONFIG_SYSCALL_DEBUG + if(signal_index[proc_id] == 1024){ + signal_index[proc_id] = 0; + last_index = 1023; + } + else last_index = signal_index[proc_id] - 1; + if(((sig == SIGPROF) || (sig == SIGVTALRM) || + (sig == SIGALRM)) && + (signal_record[proc_id][last_index].signal == sig)&& + (signal_record[proc_id][last_index].pid == pid)) + signal_index[proc_id] = last_index; + signal_record[proc_id][signal_index[proc_id]].pid = pid; + gettimeofday(&signal_record[proc_id][signal_index[proc_id]].time, NULL); + eip = ptrace(PTRACE_PEEKUSR, pid, PT_IP_OFFSET, 0); + signal_record[proc_id][signal_index[proc_id]].addr = eip; + signal_record[proc_id][signal_index[proc_id]++].signal = sig; +#endif + if(proc_id == -1){ + sleeping_process_signal(pid, sig); + continue; + } + + task = cpu_tasks[proc_id].task; + tracing = is_tracing(task); + old_tracing = tracing; + + /* Assume: no syscall, when coming from user */ + if ( tracing ) + do_sigtrap(task); + + switch(sig){ + case SIGUSR1: + sig = 0; + op = do_proc_op(task, proc_id); + switch(op){ + /* + * This is called when entering user mode; after + * this, we start intercepting syscalls. + * + * In fact, a process is started in kernel mode, + * so with is_tracing() == 0 (and that is reset + * when executing syscalls, since UML kernel has + * the right to do syscalls); + */ + case OP_TRACE_ON: + arch_leave_kernel(task, pid); + tracing = 1; + break; + case OP_REBOOT: + case OP_HALT: + unmap_physmem(); + kmalloc_ok = 0; + os_kill_ptraced_process(pid, 0); + /* Now let's reap remaining zombies */ + errno = 0; + do { + waitpid(-1, &status, + WUNTRACED); + } while (errno != ECHILD); + return(op == OP_REBOOT); + case OP_NONE: + printf("Detaching pid %d\n", pid); + detach(pid, SIGSTOP); + continue; + default: + break; + } + /* OP_EXEC switches host processes on us, + * we want to continue the new one. + */ + pid = cpu_tasks[proc_id].pid; + break; + case (SIGTRAP + 0x80): + if(!tracing && (debugger_pid != -1)){ + child_signal(pid, status & 0x7fff); + continue; + } + tracing = 0; + /* local_using_sysemu has been already set + * below, since if we are here, is_tracing() on + * the traced task was 1, i.e. the process had + * already run through one iteration of the + * loop which executed a OP_TRACE_ON request.*/ + do_syscall(task, pid, local_using_sysemu); + sig = SIGUSR2; + break; + case SIGTRAP: + if(!tracing && (debugger_pid != -1)){ + child_signal(pid, status); + continue; + } + tracing = 0; + break; + case SIGPROF: + if(tracing) sig = 0; + break; + case SIGCHLD: + case SIGHUP: + sig = 0; + break; + case SIGSEGV: + case SIGIO: + case SIGALRM: + case SIGVTALRM: + case SIGFPE: + case SIGBUS: + case SIGILL: + case SIGWINCH: + + default: + tracing = 0; + break; + } + set_tracing(task, tracing); + + if(!tracing && old_tracing) + arch_enter_kernel(task, pid); + + if(!tracing && (debugger_pid != -1) && (sig != 0) && + (sig != SIGALRM) && (sig != SIGVTALRM) && + (sig != SIGSEGV) && (sig != SIGTRAP) && + (sig != SIGUSR2) && (sig != SIGIO) && + (sig != SIGFPE)){ + child_signal(pid, status); + continue; + } + + local_using_sysemu = get_using_sysemu(); + + if(tracing) + cont_type = SELECT_PTRACE_OPERATION(local_using_sysemu, + singlestepping(task)); + else if((debugger_pid != -1) && strace) + cont_type = PTRACE_SYSCALL; + else + cont_type = PTRACE_CONT; + + if(ptrace(cont_type, pid, 0, sig) != 0){ + tracer_panic("ptrace failed to continue " + "process - errno = %d\n", + errno); + } + } + } + return(0); +} + +static int __init uml_debug_setup(char *line, int *add) +{ + char *next; + + debug = 1; + *add = 0; + if(*line != '=') return(0); + line++; + + while(line != NULL){ + next = strchr(line, ','); + if(next) *next++ = '\0'; + + if(!strcmp(line, "go")) debug_stop = 0; + else if(!strcmp(line, "parent")) debug_parent = 1; + else printf("Unknown debug option : '%s'\n", line); + + line = next; + } + return(0); +} + +__uml_setup("debug", uml_debug_setup, +"debug\n" +" Starts up the kernel under the control of gdb. See the \n" +" kernel debugging tutorial and the debugging session pages\n" +" at http://user-mode-linux.sourceforge.net/ for more information.\n\n" +); + +static int __init uml_debugtrace_setup(char *line, int *add) +{ + debug_trace = 1; + return 0; +} +__uml_setup("debugtrace", uml_debugtrace_setup, +"debugtrace\n" +" Causes the tracing thread to pause until it is attached by a\n" +" debugger and continued. This is mostly for debugging crashes\n" +" early during boot, and should be pretty much obsoleted by\n" +" the debug switch.\n\n" +); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/trap_user.c b/arch/um/kernel/tt/trap_user.c new file mode 100644 index 00000000000..92a3820ca54 --- /dev/null +++ b/arch/um/kernel/tt/trap_user.c @@ -0,0 +1,60 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdlib.h> +#include <errno.h> +#include <signal.h> +#include "sysdep/ptrace.h" +#include "signal_user.h" +#include "user_util.h" +#include "kern_util.h" +#include "task.h" +#include "tt.h" + +void sig_handler_common_tt(int sig, void *sc_ptr) +{ + struct sigcontext *sc = sc_ptr; + struct tt_regs save_regs, *r; + struct signal_info *info; + int save_errno = errno, is_user; + + /* This is done because to allow SIGSEGV to be delivered inside a SEGV + * handler. This can happen in copy_user, and if SEGV is disabled, + * the process will die. + */ + if(sig == SIGSEGV) + change_sig(SIGSEGV, 1); + + r = &TASK_REGS(get_current())->tt; + save_regs = *r; + is_user = user_context(SC_SP(sc)); + r->sc = sc; + if(sig != SIGUSR2) + r->syscall = -1; + + info = &sig_info[sig]; + if(!info->is_irq) unblock_signals(); + + (*info->handler)(sig, (union uml_pt_regs *) r); + + if(is_user){ + interrupt_end(); + block_signals(); + set_user_mode(NULL); + } + *r = save_regs; + errno = save_errno; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/uaccess.c b/arch/um/kernel/tt/uaccess.c new file mode 100644 index 00000000000..a72aa632972 --- /dev/null +++ b/arch/um/kernel/tt/uaccess.c @@ -0,0 +1,73 @@ +/* + * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Licensed under the GPL + */ + +#include "linux/sched.h" +#include "asm/uaccess.h" + +int copy_from_user_tt(void *to, const void __user *from, int n) +{ + if(!access_ok_tt(VERIFY_READ, from, n)) + return(n); + + return(__do_copy_from_user(to, from, n, ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +int copy_to_user_tt(void __user *to, const void *from, int n) +{ + if(!access_ok_tt(VERIFY_WRITE, to, n)) + return(n); + + return(__do_copy_to_user(to, from, n, ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +int strncpy_from_user_tt(char *dst, const char __user *src, int count) +{ + int n; + + if(!access_ok_tt(VERIFY_READ, src, 1)) + return(-EFAULT); + + n = __do_strncpy_from_user(dst, src, count, + ¤t->thread.fault_addr, + ¤t->thread.fault_catcher); + if(n < 0) return(-EFAULT); + return(n); +} + +int __clear_user_tt(void __user *mem, int len) +{ + return(__do_clear_user(mem, len, + ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +int clear_user_tt(void __user *mem, int len) +{ + if(!access_ok_tt(VERIFY_WRITE, mem, len)) + return(len); + + return(__do_clear_user(mem, len, ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +int strnlen_user_tt(const void __user *str, int len) +{ + return(__do_strnlen_user(str, len, + ¤t->thread.fault_addr, + ¤t->thread.fault_catcher)); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/uaccess_user.c b/arch/um/kernel/tt/uaccess_user.c new file mode 100644 index 00000000000..f01475512ec --- /dev/null +++ b/arch/um/kernel/tt/uaccess_user.c @@ -0,0 +1,98 @@ +/* + * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk) + * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <setjmp.h> +#include <string.h> +#include "user_util.h" +#include "uml_uaccess.h" +#include "task.h" +#include "kern_util.h" + +int __do_copy_from_user(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher) +{ + struct tt_regs save = TASK_REGS(get_current())->tt; + unsigned long fault; + int faulted; + + fault = __do_user_copy(to, from, n, fault_addr, fault_catcher, + __do_copy, &faulted); + TASK_REGS(get_current())->tt = save; + + if(!faulted) return(0); + else return(n - (fault - (unsigned long) from)); +} + +static void __do_strncpy(void *dst, const void *src, int count) +{ + strncpy(dst, src, count); +} + +int __do_strncpy_from_user(char *dst, const char *src, unsigned long count, + void **fault_addr, void **fault_catcher) +{ + struct tt_regs save = TASK_REGS(get_current())->tt; + unsigned long fault; + int faulted; + + fault = __do_user_copy(dst, src, count, fault_addr, fault_catcher, + __do_strncpy, &faulted); + TASK_REGS(get_current())->tt = save; + + if(!faulted) return(strlen(dst)); + else return(-1); +} + +static void __do_clear(void *to, const void *from, int n) +{ + memset(to, 0, n); +} + +int __do_clear_user(void *mem, unsigned long len, + void **fault_addr, void **fault_catcher) +{ + struct tt_regs save = TASK_REGS(get_current())->tt; + unsigned long fault; + int faulted; + + fault = __do_user_copy(mem, NULL, len, fault_addr, fault_catcher, + __do_clear, &faulted); + TASK_REGS(get_current())->tt = save; + + if(!faulted) return(0); + else return(len - (fault - (unsigned long) mem)); +} + +int __do_strnlen_user(const char *str, unsigned long n, + void **fault_addr, void **fault_catcher) +{ + struct tt_regs save = TASK_REGS(get_current())->tt; + int ret; + unsigned long *faddrp = (unsigned long *)fault_addr; + sigjmp_buf jbuf; + + *fault_catcher = &jbuf; + if(sigsetjmp(jbuf, 1) == 0) + ret = strlen(str) + 1; + else ret = *faddrp - (unsigned long) str; + + *fault_addr = NULL; + *fault_catcher = NULL; + + TASK_REGS(get_current())->tt = save; + return ret; +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tt/unmap.c b/arch/um/kernel/tt/unmap.c new file mode 100644 index 00000000000..3f7aecdbe53 --- /dev/null +++ b/arch/um/kernel/tt/unmap.c @@ -0,0 +1,31 @@ +/* + * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <sys/mman.h> + +int switcheroo(int fd, int prot, void *from, void *to, int size) +{ + if(munmap(to, size) < 0){ + return(-1); + } + if(mmap(to, size, prot, MAP_SHARED | MAP_FIXED, fd, 0) != to){ + return(-1); + } + if(munmap(from, size) < 0){ + return(-1); + } + return(0); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/tty_log.c b/arch/um/kernel/tty_log.c new file mode 100644 index 00000000000..9ada656f68c --- /dev/null +++ b/arch/um/kernel/tty_log.c @@ -0,0 +1,230 @@ +/* + * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) and + * geoffrey hing <ghing@net.ohio-state.edu> + * Licensed under the GPL + */ + +#include <errno.h> +#include <string.h> +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <sys/time.h> +#include "init.h" +#include "user.h" +#include "kern_util.h" +#include "os.h" + +#define TTY_LOG_DIR "./" + +/* Set early in boot and then unchanged */ +static char *tty_log_dir = TTY_LOG_DIR; +static int tty_log_fd = -1; + +#define TTY_LOG_OPEN 1 +#define TTY_LOG_CLOSE 2 +#define TTY_LOG_WRITE 3 +#define TTY_LOG_EXEC 4 + +#define TTY_READ 1 +#define TTY_WRITE 2 + +struct tty_log_buf { + int what; + unsigned long tty; + int len; + int direction; + unsigned long sec; + unsigned long usec; +}; + +int open_tty_log(void *tty, void *current_tty) +{ + struct timeval tv; + struct tty_log_buf data; + char buf[strlen(tty_log_dir) + sizeof("01234567890-01234567\0")]; + int fd; + + gettimeofday(&tv, NULL); + if(tty_log_fd != -1){ + data = ((struct tty_log_buf) { .what = TTY_LOG_OPEN, + .tty = (unsigned long) tty, + .len = sizeof(current_tty), + .direction = 0, + .sec = tv.tv_sec, + .usec = tv.tv_usec } ); + os_write_file(tty_log_fd, &data, sizeof(data)); + os_write_file(tty_log_fd, ¤t_tty, data.len); + return(tty_log_fd); + } + + sprintf(buf, "%s/%0u-%0u", tty_log_dir, (unsigned int) tv.tv_sec, + (unsigned int) tv.tv_usec); + + fd = os_open_file(buf, of_append(of_create(of_rdwr(OPENFLAGS()))), + 0644); + if(fd < 0){ + printk("open_tty_log : couldn't open '%s', errno = %d\n", + buf, -fd); + } + return(fd); +} + +void close_tty_log(int fd, void *tty) +{ + struct tty_log_buf data; + struct timeval tv; + + if(tty_log_fd != -1){ + gettimeofday(&tv, NULL); + data = ((struct tty_log_buf) { .what = TTY_LOG_CLOSE, + .tty = (unsigned long) tty, + .len = 0, + .direction = 0, + .sec = tv.tv_sec, + .usec = tv.tv_usec } ); + os_write_file(tty_log_fd, &data, sizeof(data)); + return; + } + os_close_file(fd); +} + +static int log_chunk(int fd, const char *buf, int len) +{ + int total = 0, try, missed, n; + char chunk[64]; + + while(len > 0){ + try = (len > sizeof(chunk)) ? sizeof(chunk) : len; + missed = copy_from_user_proc(chunk, (char *) buf, try); + try -= missed; + n = os_write_file(fd, chunk, try); + if(n != try) { + if(n < 0) + return(n); + return(-EIO); + } + if(missed != 0) + return(-EFAULT); + + len -= try; + total += try; + buf += try; + } + + return(total); +} + +int write_tty_log(int fd, const char *buf, int len, void *tty, int is_read) +{ + struct timeval tv; + struct tty_log_buf data; + int direction; + + if(fd == tty_log_fd){ + gettimeofday(&tv, NULL); + direction = is_read ? TTY_READ : TTY_WRITE; + data = ((struct tty_log_buf) { .what = TTY_LOG_WRITE, + .tty = (unsigned long) tty, + .len = len, + .direction = direction, + .sec = tv.tv_sec, + .usec = tv.tv_usec } ); + os_write_file(tty_log_fd, &data, sizeof(data)); + } + + return(log_chunk(fd, buf, len)); +} + +void log_exec(char **argv, void *tty) +{ + struct timeval tv; + struct tty_log_buf data; + char **ptr,*arg; + int len; + + if(tty_log_fd == -1) return; + + gettimeofday(&tv, NULL); + + len = 0; + for(ptr = argv; ; ptr++){ + if(copy_from_user_proc(&arg, ptr, sizeof(arg))) + return; + if(arg == NULL) break; + len += strlen_user_proc(arg); + } + + data = ((struct tty_log_buf) { .what = TTY_LOG_EXEC, + .tty = (unsigned long) tty, + .len = len, + .direction = 0, + .sec = tv.tv_sec, + .usec = tv.tv_usec } ); + os_write_file(tty_log_fd, &data, sizeof(data)); + + for(ptr = argv; ; ptr++){ + if(copy_from_user_proc(&arg, ptr, sizeof(arg))) + return; + if(arg == NULL) break; + log_chunk(tty_log_fd, arg, strlen_user_proc(arg)); + } +} + +extern void register_tty_logger(int (*opener)(void *, void *), + int (*writer)(int, const char *, int, + void *, int), + void (*closer)(int, void *)); + +static int register_logger(void) +{ + register_tty_logger(open_tty_log, write_tty_log, close_tty_log); + return(0); +} + +__uml_initcall(register_logger); + +static int __init set_tty_log_dir(char *name, int *add) +{ + tty_log_dir = name; + return 0; +} + +__uml_setup("tty_log_dir=", set_tty_log_dir, +"tty_log_dir=<directory>\n" +" This is used to specify the directory where the logs of all pty\n" +" data from this UML machine will be written.\n\n" +); + +static int __init set_tty_log_fd(char *name, int *add) +{ + char *end; + + tty_log_fd = strtoul(name, &end, 0); + if((*end != '\0') || (end == name)){ + printf("set_tty_log_fd - strtoul failed on '%s'\n", name); + tty_log_fd = -1; + } + + *add = 0; + return 0; +} + +__uml_setup("tty_log_fd=", set_tty_log_fd, +"tty_log_fd=<fd>\n" +" This is used to specify a preconfigured file descriptor to which all\n" +" tty data will be written. Preconfigure the descriptor with something\n" +" like '10>tty_log tty_log_fd=10'.\n\n" +); + + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/uaccess_user.c b/arch/um/kernel/uaccess_user.c new file mode 100644 index 00000000000..d035257ed0a --- /dev/null +++ b/arch/um/kernel/uaccess_user.c @@ -0,0 +1,64 @@ +/* + * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk) + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <setjmp.h> +#include <string.h> + +/* These are here rather than tt/uaccess.c because skas mode needs them in + * order to do SIGBUS recovery when a tmpfs mount runs out of room. + */ + +unsigned long __do_user_copy(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher, + void (*op)(void *to, const void *from, + int n), int *faulted_out) +{ + unsigned long *faddrp = (unsigned long *) fault_addr, ret; + + sigjmp_buf jbuf; + *fault_catcher = &jbuf; + if(sigsetjmp(jbuf, 1) == 0){ + (*op)(to, from, n); + ret = 0; + *faulted_out = 0; + } + else { + ret = *faddrp; + *faulted_out = 1; + } + *fault_addr = NULL; + *fault_catcher = NULL; + return ret; +} + +void __do_copy(void *to, const void *from, int n) +{ + memcpy(to, from, n); +} + + +int __do_copy_to_user(void *to, const void *from, int n, + void **fault_addr, void **fault_catcher) +{ + unsigned long fault; + int faulted; + + fault = __do_user_copy(to, from, n, fault_addr, fault_catcher, + __do_copy, &faulted); + if(!faulted) return(0); + else return(n - (fault - (unsigned long) to)); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c new file mode 100644 index 00000000000..5c49d88eed3 --- /dev/null +++ b/arch/um/kernel/um_arch.c @@ -0,0 +1,467 @@ +/* + * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include "linux/config.h" +#include "linux/kernel.h" +#include "linux/sched.h" +#include "linux/notifier.h" +#include "linux/mm.h" +#include "linux/types.h" +#include "linux/tty.h" +#include "linux/init.h" +#include "linux/bootmem.h" +#include "linux/spinlock.h" +#include "linux/utsname.h" +#include "linux/sysrq.h" +#include "linux/seq_file.h" +#include "linux/delay.h" +#include "linux/module.h" +#include "asm/page.h" +#include "asm/pgtable.h" +#include "asm/ptrace.h" +#include "asm/elf.h" +#include "asm/user.h" +#include "ubd_user.h" +#include "asm/current.h" +#include "asm/setup.h" +#include "user_util.h" +#include "kern_util.h" +#include "kern.h" +#include "mem_user.h" +#include "mem.h" +#include "umid.h" +#include "initrd.h" +#include "init.h" +#include "os.h" +#include "choose-mode.h" +#include "mode_kern.h" +#include "mode.h" + +#define DEFAULT_COMMAND_LINE "root=98:0" + +/* Changed in linux_main and setup_arch, which run before SMP is started */ +char command_line[COMMAND_LINE_SIZE] = { 0 }; + +void add_arg(char *arg) +{ + if (strlen(command_line) + strlen(arg) + 1 > COMMAND_LINE_SIZE) { + printf("add_arg: Too many command line arguments!\n"); + exit(1); + } + if(strlen(command_line) > 0) + strcat(command_line, " "); + strcat(command_line, arg); +} + +struct cpuinfo_um boot_cpu_data = { + .loops_per_jiffy = 0, + .ipi_pipe = { -1, -1 } +}; + +unsigned long thread_saved_pc(struct task_struct *task) +{ + return(os_process_pc(CHOOSE_MODE_PROC(thread_pid_tt, thread_pid_skas, + task))); +} + +static int show_cpuinfo(struct seq_file *m, void *v) +{ + int index = 0; + +#ifdef CONFIG_SMP + index = (struct cpuinfo_um *) v - cpu_data; + if (!cpu_online(index)) + return 0; +#endif + + seq_printf(m, "processor\t: %d\n", index); + seq_printf(m, "vendor_id\t: User Mode Linux\n"); + seq_printf(m, "model name\t: UML\n"); + seq_printf(m, "mode\t\t: %s\n", CHOOSE_MODE("tt", "skas")); + seq_printf(m, "host\t\t: %s\n", host_info); + seq_printf(m, "bogomips\t: %lu.%02lu\n\n", + loops_per_jiffy/(500000/HZ), + (loops_per_jiffy/(5000/HZ)) % 100); + + return(0); +} + +static void *c_start(struct seq_file *m, loff_t *pos) +{ + return *pos < NR_CPUS ? cpu_data + *pos : NULL; +} + +static void *c_next(struct seq_file *m, void *v, loff_t *pos) +{ + ++*pos; + return c_start(m, pos); +} + +static void c_stop(struct seq_file *m, void *v) +{ +} + +struct seq_operations cpuinfo_op = { + .start = c_start, + .next = c_next, + .stop = c_stop, + .show = show_cpuinfo, +}; + +pte_t * __bad_pagetable(void) +{ + panic("Someone should implement __bad_pagetable"); + return(NULL); +} + +/* Set in linux_main */ +unsigned long host_task_size; +unsigned long task_size; + +unsigned long uml_start; + +/* Set in early boot */ +unsigned long uml_physmem; +unsigned long uml_reserved; +unsigned long start_vm; +unsigned long end_vm; +int ncpus = 1; + +#ifdef CONFIG_MODE_TT +/* Pointer set in linux_main, the array itself is private to each thread, + * and changed at address space creation time so this poses no concurrency + * problems. + */ +static char *argv1_begin = NULL; +static char *argv1_end = NULL; +#endif + +/* Set in early boot */ +static int have_root __initdata = 0; +long physmem_size = 32 * 1024 * 1024; + +void set_cmdline(char *cmd) +{ +#ifdef CONFIG_MODE_TT + char *umid, *ptr; + + if(CHOOSE_MODE(honeypot, 0)) return; + + umid = get_umid(1); + if(umid != NULL){ + snprintf(argv1_begin, + (argv1_end - argv1_begin) * sizeof(*ptr), + "(%s) ", umid); + ptr = &argv1_begin[strlen(argv1_begin)]; + } + else ptr = argv1_begin; + + snprintf(ptr, (argv1_end - ptr) * sizeof(*ptr), "[%s]", cmd); + memset(argv1_begin + strlen(argv1_begin), '\0', + argv1_end - argv1_begin - strlen(argv1_begin)); +#endif +} + +static char *usage_string = +"User Mode Linux v%s\n" +" available at http://user-mode-linux.sourceforge.net/\n\n"; + +static int __init uml_version_setup(char *line, int *add) +{ + printf("%s\n", system_utsname.release); + exit(0); + + return 0; +} + +__uml_setup("--version", uml_version_setup, +"--version\n" +" Prints the version number of the kernel.\n\n" +); + +static int __init uml_root_setup(char *line, int *add) +{ + have_root = 1; + return 0; +} + +__uml_setup("root=", uml_root_setup, +"root=<file containing the root fs>\n" +" This is actually used by the generic kernel in exactly the same\n" +" way as in any other kernel. If you configure a number of block\n" +" devices and want to boot off something other than ubd0, you \n" +" would use something like:\n" +" root=/dev/ubd5\n\n" +); + +#ifdef CONFIG_SMP +static int __init uml_ncpus_setup(char *line, int *add) +{ + if (!sscanf(line, "%d", &ncpus)) { + printf("Couldn't parse [%s]\n", line); + return -1; + } + + return 0; +} + +__uml_setup("ncpus=", uml_ncpus_setup, +"ncpus=<# of desired CPUs>\n" +" This tells an SMP kernel how many virtual processors to start.\n\n" +); +#endif + +static int force_tt = 0; + +#if defined(CONFIG_MODE_TT) && defined(CONFIG_MODE_SKAS) +#define DEFAULT_TT 0 + +static int __init mode_tt_setup(char *line, int *add) +{ + force_tt = 1; + return(0); +} + +#else +#ifdef CONFIG_MODE_SKAS + +#define DEFAULT_TT 0 + +static int __init mode_tt_setup(char *line, int *add) +{ + printf("CONFIG_MODE_TT disabled - 'mode=tt' ignored\n"); + return(0); +} + +#else +#ifdef CONFIG_MODE_TT + +#define DEFAULT_TT 1 + +static int __init mode_tt_setup(char *line, int *add) +{ + printf("CONFIG_MODE_SKAS disabled - 'mode=tt' redundant\n"); + return(0); +} + +#else + +#error Either CONFIG_MODE_TT or CONFIG_MODE_SKAS must be enabled + +#endif +#endif +#endif + +__uml_setup("mode=tt", mode_tt_setup, +"mode=tt\n" +" When both CONFIG_MODE_TT and CONFIG_MODE_SKAS are enabled, this option\n" +" forces UML to run in tt (tracing thread) mode. It is not the default\n" +" because it's slower and less secure than skas mode.\n\n" +); + +int mode_tt = DEFAULT_TT; + +static int __init Usage(char *line, int *add) +{ + const char **p; + + printf(usage_string, system_utsname.release); + p = &__uml_help_start; + while (p < &__uml_help_end) { + printf("%s", *p); + p++; + } + exit(0); + + return 0; +} + +__uml_setup("--help", Usage, +"--help\n" +" Prints this message.\n\n" +); + +static int __init uml_checksetup(char *line, int *add) +{ + struct uml_param *p; + + p = &__uml_setup_start; + while(p < &__uml_setup_end) { + int n; + + n = strlen(p->str); + if(!strncmp(line, p->str, n)){ + if (p->setup_func(line + n, add)) return 1; + } + p++; + } + return 0; +} + +static void __init uml_postsetup(void) +{ + initcall_t *p; + + p = &__uml_postsetup_start; + while(p < &__uml_postsetup_end){ + (*p)(); + p++; + } + return; +} + +/* Set during early boot */ +unsigned long brk_start; +unsigned long end_iomem; +EXPORT_SYMBOL(end_iomem); + +#define MIN_VMALLOC (32 * 1024 * 1024) + +int linux_main(int argc, char **argv) +{ + unsigned long avail, diff; + unsigned long virtmem_size, max_physmem; + unsigned int i, add; + + for (i = 1; i < argc; i++){ + if((i == 1) && (argv[i][0] == ' ')) continue; + add = 1; + uml_checksetup(argv[i], &add); + if (add) + add_arg(argv[i]); + } + if(have_root == 0) + add_arg(DEFAULT_COMMAND_LINE); + + mode_tt = force_tt ? 1 : !can_do_skas(); +#ifndef CONFIG_MODE_TT + if (mode_tt) { + /*Since CONFIG_MODE_TT is #undef'ed, force_tt cannot be 1. So, + * can_do_skas() returned 0, and the message is correct. */ + printf("Support for TT mode is disabled, and no SKAS support is present on the host.\n"); + exit(1); + } +#endif + uml_start = CHOOSE_MODE_PROC(set_task_sizes_tt, set_task_sizes_skas, 0, + &host_task_size, &task_size); + + /* Need to check this early because mmapping happens before the + * kernel is running. + */ + check_tmpexec(); + + brk_start = (unsigned long) sbrk(0); + CHOOSE_MODE_PROC(before_mem_tt, before_mem_skas, brk_start); + /* Increase physical memory size for exec-shield users + so they actually get what they asked for. This should + add zero for non-exec shield users */ + + diff = UML_ROUND_UP(brk_start) - UML_ROUND_UP(&_end); + if(diff > 1024 * 1024){ + printf("Adding %ld bytes to physical memory to account for " + "exec-shield gap\n", diff); + physmem_size += UML_ROUND_UP(brk_start) - UML_ROUND_UP(&_end); + } + + uml_physmem = uml_start; + + /* Reserve up to 4M after the current brk */ + uml_reserved = ROUND_4M(brk_start) + (1 << 22); + + setup_machinename(system_utsname.machine); + +#ifdef CONFIG_MODE_TT + argv1_begin = argv[1]; + argv1_end = &argv[1][strlen(argv[1])]; +#endif + + highmem = 0; + iomem_size = (iomem_size + PAGE_SIZE - 1) & PAGE_MASK; + max_physmem = get_kmem_end() - uml_physmem - iomem_size - MIN_VMALLOC; + + /* Zones have to begin on a 1 << MAX_ORDER page boundary, + * so this makes sure that's true for highmem + */ + max_physmem &= ~((1 << (PAGE_SHIFT + MAX_ORDER)) - 1); + if(physmem_size + iomem_size > max_physmem){ + highmem = physmem_size + iomem_size - max_physmem; + physmem_size -= highmem; +#ifndef CONFIG_HIGHMEM + highmem = 0; + printf("CONFIG_HIGHMEM not enabled - physical memory shrunk " + "to %ld bytes\n", physmem_size); +#endif + } + + high_physmem = uml_physmem + physmem_size; + end_iomem = high_physmem + iomem_size; + high_memory = (void *) end_iomem; + + start_vm = VMALLOC_START; + + setup_physmem(uml_physmem, uml_reserved, physmem_size, highmem); + if(init_maps(physmem_size, iomem_size, highmem)){ + printf("Failed to allocate mem_map for %ld bytes of physical " + "memory and %ld bytes of highmem\n", physmem_size, + highmem); + exit(1); + } + + virtmem_size = physmem_size; + avail = get_kmem_end() - start_vm; + if(physmem_size > avail) virtmem_size = avail; + end_vm = start_vm + virtmem_size; + + if(virtmem_size < physmem_size) + printf("Kernel virtual memory size shrunk to %ld bytes\n", + virtmem_size); + + uml_postsetup(); + + task_protections((unsigned long) &init_thread_info); + os_flush_stdout(); + + return(CHOOSE_MODE(start_uml_tt(), start_uml_skas())); +} + +extern int uml_exitcode; + +static int panic_exit(struct notifier_block *self, unsigned long unused1, + void *unused2) +{ + bust_spinlocks(1); + show_regs(&(current->thread.regs)); + bust_spinlocks(0); + uml_exitcode = 1; + machine_halt(); + return(0); +} + +static struct notifier_block panic_exit_notifier = { + .notifier_call = panic_exit, + .next = NULL, + .priority = 0 +}; + +void __init setup_arch(char **cmdline_p) +{ + notifier_chain_register(&panic_notifier_list, &panic_exit_notifier); + paging_init(); + strlcpy(saved_command_line, command_line, COMMAND_LINE_SIZE); + *cmdline_p = command_line; + setup_hostinfo(); +} + +void __init check_bugs(void) +{ + arch_check_bugs(); + check_ptrace(); + check_sigio(); + check_devanon(); +} + +void apply_alternatives(void *start, void *end) +{ +} diff --git a/arch/um/kernel/umid.c b/arch/um/kernel/umid.c new file mode 100644 index 00000000000..186c2888501 --- /dev/null +++ b/arch/um/kernel/umid.c @@ -0,0 +1,325 @@ +/* + * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <unistd.h> +#include <errno.h> +#include <string.h> +#include <stdlib.h> +#include <dirent.h> +#include <signal.h> +#include <sys/stat.h> +#include <sys/param.h> +#include "user.h" +#include "umid.h" +#include "init.h" +#include "os.h" +#include "user_util.h" +#include "choose-mode.h" + +#define UMID_LEN 64 +#define UML_DIR "~/.uml/" + +/* Changed by set_umid and make_umid, which are run early in boot */ +static char umid[UMID_LEN] = { 0 }; + +/* Changed by set_uml_dir and make_uml_dir, which are run early in boot */ +static char *uml_dir = UML_DIR; + +/* Changed by set_umid */ +static int umid_is_random = 1; +static int umid_inited = 0; + +static int make_umid(int (*printer)(const char *fmt, ...)); + +static int __init set_umid(char *name, int is_random, + int (*printer)(const char *fmt, ...)) +{ + if(umid_inited){ + (*printer)("Unique machine name can't be set twice\n"); + return(-1); + } + + if(strlen(name) > UMID_LEN - 1) + (*printer)("Unique machine name is being truncated to %d " + "characters\n", UMID_LEN); + strlcpy(umid, name, sizeof(umid)); + + umid_is_random = is_random; + umid_inited = 1; + return 0; +} + +static int __init set_umid_arg(char *name, int *add) +{ + *add = 0; + return(set_umid(name, 0, printf)); +} + +__uml_setup("umid=", set_umid_arg, +"umid=<name>\n" +" This is used to assign a unique identity to this UML machine and\n" +" is used for naming the pid file and management console socket.\n\n" +); + +int __init umid_file_name(char *name, char *buf, int len) +{ + int n; + + if(!umid_inited && make_umid(printk)) return(-1); + + n = strlen(uml_dir) + strlen(umid) + strlen(name) + 1; + if(n > len){ + printk("umid_file_name : buffer too short\n"); + return(-1); + } + + sprintf(buf, "%s%s/%s", uml_dir, umid, name); + return(0); +} + +extern int tracing_pid; + +static int __init create_pid_file(void) +{ + char file[strlen(uml_dir) + UMID_LEN + sizeof("/pid\0")]; + char pid[sizeof("nnnnn\0")]; + int fd, n; + + if(umid_file_name("pid", file, sizeof(file))) return 0; + + fd = os_open_file(file, of_create(of_excl(of_rdwr(OPENFLAGS()))), + 0644); + if(fd < 0){ + printf("Open of machine pid file \"%s\" failed: %s\n", + file, strerror(-fd)); + return 0; + } + + sprintf(pid, "%d\n", os_getpid()); + n = os_write_file(fd, pid, strlen(pid)); + if(n != strlen(pid)) + printf("Write of pid file failed - err = %d\n", -n); + os_close_file(fd); + return 0; +} + +static int actually_do_remove(char *dir) +{ + DIR *directory; + struct dirent *ent; + int len; + char file[256]; + + directory = opendir(dir); + if(directory == NULL){ + printk("actually_do_remove : couldn't open directory '%s', " + "errno = %d\n", dir, errno); + return(1); + } + while((ent = readdir(directory)) != NULL){ + if(!strcmp(ent->d_name, ".") || !strcmp(ent->d_name, "..")) + continue; + len = strlen(dir) + sizeof("/") + strlen(ent->d_name) + 1; + if(len > sizeof(file)){ + printk("Not deleting '%s' from '%s' - name too long\n", + ent->d_name, dir); + continue; + } + sprintf(file, "%s/%s", dir, ent->d_name); + if(unlink(file) < 0){ + printk("actually_do_remove : couldn't remove '%s' " + "from '%s', errno = %d\n", ent->d_name, dir, + errno); + return(1); + } + } + if(rmdir(dir) < 0){ + printk("actually_do_remove : couldn't rmdir '%s', " + "errno = %d\n", dir, errno); + return(1); + } + return(0); +} + +void remove_umid_dir(void) +{ + char dir[strlen(uml_dir) + UMID_LEN + 1]; + if(!umid_inited) return; + + sprintf(dir, "%s%s", uml_dir, umid); + actually_do_remove(dir); +} + +char *get_umid(int only_if_set) +{ + if(only_if_set && umid_is_random) return(NULL); + return(umid); +} + +int not_dead_yet(char *dir) +{ + char file[strlen(uml_dir) + UMID_LEN + sizeof("/pid\0")]; + char pid[sizeof("nnnnn\0")], *end; + int dead, fd, p, n; + + sprintf(file, "%s/pid", dir); + dead = 0; + fd = os_open_file(file, of_read(OPENFLAGS()), 0); + if(fd < 0){ + if(fd != -ENOENT){ + printk("not_dead_yet : couldn't open pid file '%s', " + "err = %d\n", file, -fd); + return(1); + } + dead = 1; + } + if(fd > 0){ + n = os_read_file(fd, pid, sizeof(pid)); + if(n < 0){ + printk("not_dead_yet : couldn't read pid file '%s', " + "err = %d\n", file, -n); + return(1); + } + p = strtoul(pid, &end, 0); + if(end == pid){ + printk("not_dead_yet : couldn't parse pid file '%s', " + "errno = %d\n", file, errno); + dead = 1; + } + if(((kill(p, 0) < 0) && (errno == ESRCH)) || + (p == CHOOSE_MODE(tracing_pid, os_getpid()))) + dead = 1; + } + if(!dead) return(1); + return(actually_do_remove(dir)); +} + +static int __init set_uml_dir(char *name, int *add) +{ + if((strlen(name) > 0) && (name[strlen(name) - 1] != '/')){ + uml_dir = malloc(strlen(name) + 2); + if(uml_dir == NULL){ + printf("Failed to malloc uml_dir - error = %d\n", + errno); + uml_dir = name; + /* Return 0 here because do_initcalls doesn't look at + * the return value. + */ + return(0); + } + sprintf(uml_dir, "%s/", name); + } + else uml_dir = name; + return(0); +} + +static int __init make_uml_dir(void) +{ + char dir[MAXPATHLEN + 1] = { '\0' }; + int len; + + if(*uml_dir == '~'){ + char *home = getenv("HOME"); + + if(home == NULL){ + printf("make_uml_dir : no value in environment for " + "$HOME\n"); + exit(1); + } + strlcpy(dir, home, sizeof(dir)); + uml_dir++; + } + len = strlen(dir); + strncat(dir, uml_dir, sizeof(dir) - len); + len = strlen(dir); + if((len > 0) && (len < sizeof(dir) - 1) && (dir[len - 1] != '/')){ + dir[len] = '/'; + dir[len + 1] = '\0'; + } + + uml_dir = malloc(strlen(dir) + 1); + if(uml_dir == NULL){ + printf("make_uml_dir : malloc failed, errno = %d\n", errno); + exit(1); + } + strcpy(uml_dir, dir); + + if((mkdir(uml_dir, 0777) < 0) && (errno != EEXIST)){ + printf("Failed to mkdir %s: %s\n", uml_dir, strerror(errno)); + return(-1); + } + return 0; +} + +static int __init make_umid(int (*printer)(const char *fmt, ...)) +{ + int fd, err; + char tmp[strlen(uml_dir) + UMID_LEN + 1]; + + strlcpy(tmp, uml_dir, sizeof(tmp)); + + if(!umid_inited){ + strcat(tmp, "XXXXXX"); + fd = mkstemp(tmp); + if(fd < 0){ + (*printer)("make_umid - mkstemp(%s) failed: %s\n", + tmp,strerror(errno)); + return(1); + } + + os_close_file(fd); + /* There's a nice tiny little race between this unlink and + * the mkdir below. It'd be nice if there were a mkstemp + * for directories. + */ + unlink(tmp); + set_umid(&tmp[strlen(uml_dir)], 1, printer); + } + + sprintf(tmp, "%s%s", uml_dir, umid); + + err = mkdir(tmp, 0777); + if(err < 0){ + if(errno == EEXIST){ + if(not_dead_yet(tmp)){ + (*printer)("umid '%s' is in use\n", umid); + return(-1); + } + err = mkdir(tmp, 0777); + } + } + if(err < 0){ + (*printer)("Failed to create %s - errno = %d\n", umid, errno); + return(-1); + } + + return(0); +} + +__uml_setup("uml_dir=", set_uml_dir, +"uml_dir=<directory>\n" +" The location to place the pid and umid files.\n\n" +); + +static int __init make_umid_setup(void) +{ + /* one function with the ordering we need ... */ + make_uml_dir(); + make_umid(printf); + return create_pid_file(); +} +__uml_postsetup(make_umid_setup); + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S new file mode 100644 index 00000000000..76eadb30918 --- /dev/null +++ b/arch/um/kernel/uml.lds.S @@ -0,0 +1,106 @@ +#include <asm-generic/vmlinux.lds.h> + +OUTPUT_FORMAT(ELF_FORMAT) +OUTPUT_ARCH(ELF_ARCH) +ENTRY(_start) +jiffies = jiffies_64; + +SECTIONS +{ + /*This must contain the right address - not quite the default ELF one.*/ + PROVIDE (__executable_start = START); + . = START + SIZEOF_HEADERS; + + /* Used in arch/um/kernel/mem.c. Any memory between START and __binary_start + * is remapped.*/ + __binary_start = .; +#ifdef MODE_TT + .thread_private : { + __start_thread_private = .; + errno = .; + . += 4; + arch/um/kernel/tt/unmap_fin.o (.data) + __end_thread_private = .; + } + . = ALIGN(4096); + .remap : { arch/um/kernel/tt/unmap_fin.o (.text) } + + /* We want it only if we are in MODE_TT. In both cases, however, when MODE_TT + * is off the resulting binary segfaults.*/ + + . = ALIGN(4096); /* Init code and data */ +#endif + + _stext = .; + __init_begin = .; + .init.text : { + _sinittext = .; + *(.init.text) + _einittext = .; + } + . = ALIGN(4096); + .text : + { + *(.text) + SCHED_TEXT + LOCK_TEXT + *(.fixup) + /* .gnu.warning sections are handled specially by elf32.em. */ + *(.gnu.warning) + *(.gnu.linkonce.t*) + } + + #include "asm/common.lds.S" + + init.data : { *(init.data) } + .data : + { + . = ALIGN(KERNEL_STACK_SIZE); /* init_task */ + *(.data.init_task) + *(.data) + *(.gnu.linkonce.d*) + CONSTRUCTORS + } + .data1 : { *(.data1) } + .ctors : + { + *(.ctors) + } + .dtors : + { + *(.dtors) + } + + .got : { *(.got.plt) *(.got) } + .dynamic : { *(.dynamic) } + /* We want the small data sections together, so single-instruction offsets + can access them all, and initialized data all before uninitialized, so + we can shorten the on-disk segment size. */ + .sdata : { *(.sdata) } + _edata = .; + PROVIDE (edata = .); + . = ALIGN(0x1000); + .sbss : + { + __bss_start = .; + PROVIDE(_bss_start = .); + *(.sbss) + *(.scommon) + } + .bss : + { + *(.dynbss) + *(.bss) + *(COMMON) + } + _end = . ; + PROVIDE (end = .); + /* Stabs debugging sections. */ + .stab 0 : { *(.stab) } + .stabstr 0 : { *(.stabstr) } + .stab.excl 0 : { *(.stab.excl) } + .stab.exclstr 0 : { *(.stab.exclstr) } + .stab.index 0 : { *(.stab.index) } + .stab.indexstr 0 : { *(.stab.indexstr) } + .comment 0 : { *(.comment) } +} diff --git a/arch/um/kernel/user_util.c b/arch/um/kernel/user_util.c new file mode 100644 index 00000000000..954ff67cc8b --- /dev/null +++ b/arch/um/kernel/user_util.c @@ -0,0 +1,173 @@ +/* + * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Licensed under the GPL + */ + +#include <stdio.h> +#include <stdlib.h> +#include <unistd.h> +#include <limits.h> +#include <setjmp.h> +#include <sys/mman.h> +#include <sys/stat.h> +#include <sys/utsname.h> +#include <sys/param.h> +#include <sys/time.h> +#include "asm/types.h" +#include <ctype.h> +#include <signal.h> +#include <wait.h> +#include <errno.h> +#include <stdarg.h> +#include <sched.h> +#include <termios.h> +#include <string.h> +#include "user_util.h" +#include "kern_util.h" +#include "user.h" +#include "mem_user.h" +#include "init.h" +#include "helper.h" +#include "ptrace_user.h" +#include "uml-config.h" + +void stop(void) +{ + while(1) sleep(1000000); +} + +void stack_protections(unsigned long address) +{ + int prot = PROT_READ | PROT_WRITE | PROT_EXEC; + + if(mprotect((void *) address, page_size(), prot) < 0) + panic("protecting stack failed, errno = %d", errno); +} + +void task_protections(unsigned long address) +{ + unsigned long guard = address + page_size(); + unsigned long stack = guard + page_size(); + int prot = 0, pages; + +#ifdef notdef + if(mprotect((void *) stack, page_size(), prot) < 0) + panic("protecting guard page failed, errno = %d", errno); +#endif + pages = (1 << UML_CONFIG_KERNEL_STACK_ORDER) - 2; + prot = PROT_READ | PROT_WRITE | PROT_EXEC; + if(mprotect((void *) stack, pages * page_size(), prot) < 0) + panic("protecting stack failed, errno = %d", errno); +} + +int wait_for_stop(int pid, int sig, int cont_type, void *relay) +{ + sigset_t *relay_signals = relay; + int status, ret; + + while(1){ + CATCH_EINTR(ret = waitpid(pid, &status, WUNTRACED)); + if((ret < 0) || + !WIFSTOPPED(status) || (WSTOPSIG(status) != sig)){ + if(ret < 0){ + printk("wait failed, errno = %d\n", + errno); + } + else if(WIFEXITED(status)) + printk("process %d exited with status %d\n", + pid, WEXITSTATUS(status)); + else if(WIFSIGNALED(status)) + printk("process %d exited with signal %d\n", + pid, WTERMSIG(status)); + else if((WSTOPSIG(status) == SIGVTALRM) || + (WSTOPSIG(status) == SIGALRM) || + (WSTOPSIG(status) == SIGIO) || + (WSTOPSIG(status) == SIGPROF) || + (WSTOPSIG(status) == SIGCHLD) || + (WSTOPSIG(status) == SIGWINCH) || + (WSTOPSIG(status) == SIGINT)){ + ptrace(cont_type, pid, 0, WSTOPSIG(status)); + continue; + } + else if((relay_signals != NULL) && + sigismember(relay_signals, WSTOPSIG(status))){ + ptrace(cont_type, pid, 0, WSTOPSIG(status)); + continue; + } + else printk("process %d stopped with signal %d\n", + pid, WSTOPSIG(status)); + panic("wait_for_stop failed to wait for %d to stop " + "with %d\n", pid, sig); + } + return(status); + } +} + +int raw(int fd) +{ + struct termios tt; + int err; + + CATCH_EINTR(err = tcgetattr(fd, &tt)); + if (err < 0) { + printk("tcgetattr failed, errno = %d\n", errno); + return(-errno); + } + + cfmakeraw(&tt); + + CATCH_EINTR(err = tcsetattr(fd, TCSADRAIN, &tt)); + if (err < 0) { + printk("tcsetattr failed, errno = %d\n", errno); + return(-errno); + } + + /* XXX tcsetattr could have applied only some changes + * (and cfmakeraw() is a set of changes) */ + return(0); +} + +void setup_machinename(char *machine_out) +{ + struct utsname host; + + uname(&host); + strcpy(machine_out, host.machine); +} + +char host_info[(_UTSNAME_LENGTH + 1) * 4 + _UTSNAME_NODENAME_LENGTH + 1]; + +void setup_hostinfo(void) +{ + struct utsname host; + + uname(&host); + sprintf(host_info, "%s %s %s %s %s", host.sysname, host.nodename, + host.release, host.version, host.machine); +} + +int setjmp_wrapper(void (*proc)(void *, void *), ...) +{ + va_list args; + sigjmp_buf buf; + int n; + + n = sigsetjmp(buf, 1); + if(n == 0){ + va_start(args, proc); + (*proc)(&buf, &args); + } + va_end(args); + return(n); +} + +/* + * Overrides for Emacs so that we follow Linus's tabbing style. + * Emacs will notice this stuff at the end of the file and automatically + * adjust the settings for this buffer only. This must remain at the end + * of the file. + * --------------------------------------------------------------------------- + * Local variables: + * c-file-style: "linux" + * End: + */ |