diff options
Diffstat (limited to 'arch/um/kernel')
68 files changed, 2252 insertions, 6295 deletions
diff --git a/arch/um/kernel/Makefile b/arch/um/kernel/Makefile index 6fa63a2a89e..d8b78a03855 100644 --- a/arch/um/kernel/Makefile +++ b/arch/um/kernel/Makefile @@ -1,22 +1,24 @@ # -# Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +# Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux,intel}.com) # Licensed under the GPL # +CPPFLAGS_vmlinux.lds := -DSTART=$(LDS_START) \ + -DELF_ARCH=$(LDS_ELF_ARCH) \ + -DELF_FORMAT=$(LDS_ELF_FORMAT) \ + $(LDS_EXTRA) extra-y := vmlinux.lds clean-files := -obj-y = config.o exec.o exitcode.o init_task.o irq.o ksyms.o mem.o \ - physmem.o process.o ptrace.o reboot.o resource.o sigio.o \ - signal.o smp.o syscall.o sysrq.o time.o tlb.o trap.o uaccess.o \ - um_arch.o umid.o +obj-y = config.o exec.o exitcode.o irq.o ksyms.o mem.o \ + physmem.o process.o ptrace.o reboot.o sigio.o \ + signal.o smp.o syscall.o sysrq.o time.o tlb.o trap.o \ + um_arch.o umid.o maccess.o skas/ obj-$(CONFIG_BLK_DEV_INITRD) += initrd.o obj-$(CONFIG_GPROF) += gprof_syms.o obj-$(CONFIG_GCOV) += gmon_syms.o - -obj-$(CONFIG_MODE_TT) += tt/ -obj-$(CONFIG_MODE_SKAS) += skas/ +obj-$(CONFIG_EARLY_PRINTK) += early_printk.o USER_OBJS := config.o @@ -31,7 +33,7 @@ $(obj)/config.tmp: $(objtree)/.config FORCE $(call if_changed,quote1) quiet_cmd_quote1 = QUOTE $@ - cmd_quote1 = sed -e 's/"/\\"/g' -e 's/^/"/' -e 's/$$/\\n"/' \ + cmd_quote1 = sed -e 's/"/\\"/g' -e 's/^/"/' -e 's/$$/\\n",/' \ $< > $@ $(obj)/config.c: $(src)/config.c.in $(obj)/config.tmp FORCE @@ -39,9 +41,9 @@ $(obj)/config.c: $(src)/config.c.in $(obj)/config.tmp FORCE quiet_cmd_quote2 = QUOTE $@ cmd_quote2 = sed -e '/CONFIG/{' \ - -e 's/"CONFIG"\;/""/' \ + -e 's/"CONFIG"//' \ -e 'r $(obj)/config.tmp' \ -e 'a \' \ - -e '""\;' \ + -e '""' \ -e '}' \ $< > $@ diff --git a/arch/um/kernel/asm-offsets.c b/arch/um/kernel/asm-offsets.c index 91ea538e161..1fb12235ab9 100644 --- a/arch/um/kernel/asm-offsets.c +++ b/arch/um/kernel/asm-offsets.c @@ -1 +1 @@ -#include "sysdep/kernel-offsets.h" +#include <sysdep/kernel-offsets.h> diff --git a/arch/um/kernel/config.c.in b/arch/um/kernel/config.c.in index c062cbfe386..972bf165956 100644 --- a/arch/um/kernel/config.c.in +++ b/arch/um/kernel/config.c.in @@ -5,13 +5,17 @@ #include <stdio.h> #include <stdlib.h> -#include "init.h" +#include <init.h> -static __initdata char *config = "CONFIG"; +static __initdata const char *config[] = { +"CONFIG" +}; static int __init print_config(char *line, int *add) { - printf("%s", config); + int i; + for (i = 0; i < sizeof(config)/sizeof(config[0]); i++) + printf("%s", config[i]); exit(0); } @@ -20,13 +24,3 @@ __uml_setup("--showconfig", print_config, " 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 index e36f92b463c..adde088aeef 100644 --- a/arch/um/kernel/dyn.lds.S +++ b/arch/um/kernel/dyn.lds.S @@ -1,4 +1,5 @@ #include <asm-generic/vmlinux.lds.h> +#include <asm/page.h> OUTPUT_FORMAT(ELF_FORMAT) OUTPUT_ARCH(ELF_ARCH) @@ -10,20 +11,12 @@ 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 */ _text = .; - _stext = .; - __init_begin = .; - .init.text : { - _sinittext = .; - *(.init.text) - _einittext = .; - } + INIT_TEXT_SECTION(PAGE_SIZE) - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE); /* Read-only sections, merged into text segment: */ .hash : { *(.hash) } @@ -55,14 +48,25 @@ SECTIONS .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) } + .rel.plt : { + *(.rel.plt) + PROVIDE_HIDDEN(__rel_iplt_start = .); + *(.rel.iplt) + PROVIDE_HIDDEN(__rel_iplt_end = .); + } + .rela.plt : { + *(.rela.plt) + PROVIDE_HIDDEN(__rela_iplt_start = .); + *(.rela.iplt) + PROVIDE_HIDDEN(__rela_iplt_end = .); + } .init : { KEEP (*(.init)) } =0x90909090 .plt : { *(.plt) } .text : { - *(.text) + _stext = .; + TEXT_TEXT SCHED_TEXT LOCK_TEXT *(.fixup) @@ -70,21 +74,25 @@ SECTIONS /* .gnu.warning sections are handled specially by elf32.em. */ *(.gnu.warning) - . = ALIGN(4096); - __syscall_stub_start = .; - *(.__syscall_stub*) - __syscall_stub_end = .; - . = ALIGN(4096); + . = ALIGN(PAGE_SIZE); } =0x90909090 + . = ALIGN(PAGE_SIZE); + .syscall_stub : { + __syscall_stub_start = .; + *(.__syscall_stub*) + __syscall_stub_end = .; + } .fini : { KEEP (*(.fini)) } =0x90909090 .kstrtab : { *(.kstrtab) } - #include "asm/common.lds.S" + #include <asm/common.lds.S> - init.data : { *(.init.data) } + __init_begin = .; + init.data : { INIT_DATA } + __init_end = .; /* Ensure the __preinit_array_start label is properly aligned. We could instead move the label definition inside the section, but @@ -95,9 +103,11 @@ SECTIONS .init_array : { *(.init_array) } .fini_array : { *(.fini_array) } .data : { - . = ALIGN(KERNEL_STACK_SIZE); /* init_task */ - *(.data.init_task) - *(.data .data.* .gnu.linkonce.d.*) + INIT_TASK_DATA(KERNEL_STACK_SIZE) + . = ALIGN(KERNEL_STACK_SIZE); + *(.data..init_irqstack) + DATA_DATA + *(.data.* .gnu.linkonce.d.*) SORT(CONSTRUCTORS) } .data1 : { *(.data1) } @@ -135,8 +145,8 @@ SECTIONS .got : { *(.got.plt) *(.got) } _edata = .; PROVIDE (edata = .); - __bss_start = .; .bss : { + __bss_start = .; *(.dynbss) *(.bss .bss.* .gnu.linkonce.b.*) *(COMMON) @@ -146,10 +156,13 @@ SECTIONS . = ALIGN(32 / 8); . = ALIGN(32 / 8); } + __bss_stop = .; _end = .; PROVIDE (end = .); STABS_DEBUG DWARF_DEBUG + + DISCARDS } diff --git a/arch/um/kernel/early_printk.c b/arch/um/kernel/early_printk.c new file mode 100644 index 00000000000..4a0800bc37b --- /dev/null +++ b/arch/um/kernel/early_printk.c @@ -0,0 +1,35 @@ +/* + * Copyright (C) 2011 Richard Weinberger <richrd@nod.at> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/kernel.h> +#include <linux/console.h> +#include <linux/init.h> +#include <os.h> + +static void early_console_write(struct console *con, const char *s, unsigned int n) +{ + um_early_printk(s, n); +} + +static struct console early_console_dev = { + .name = "earlycon", + .write = early_console_write, + .flags = CON_BOOT, + .index = -1, +}; + +static int __init setup_early_printk(char *buf) +{ + if (!early_console) { + early_console = &early_console_dev; + register_console(&early_console_dev); + } + return 0; +} + +early_param("earlyprintk", setup_early_printk); diff --git a/arch/um/kernel/exec.c b/arch/um/kernel/exec.c index 8d56ec6cca7..0d7103c9eff 100644 --- a/arch/um/kernel/exec.c +++ b/arch/um/kernel/exec.c @@ -1,89 +1,50 @@ /* - * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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 "os.h" -#include "choose-mode.h" -#include "mode_kern.h" +#include <linux/stddef.h> +#include <linux/module.h> +#include <linux/fs.h> +#include <linux/ptrace.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <asm/current.h> +#include <asm/processor.h> +#include <asm/uaccess.h> +#include <as-layout.h> +#include <mem_user.h> +#include <skas.h> +#include <os.h> void flush_thread(void) { + void *data = NULL; + int ret; + arch_flush_thread(¤t->thread.arch); - 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); + ret = unmap(¤t->mm->context.id, 0, STUB_START, 0, &data); + ret = ret || unmap(¤t->mm->context.id, STUB_END, + host_task_size - STUB_END, 1, &data); + if (ret) { + printk(KERN_ERR "flush_thread - clearing address space failed, " + "err = %d\n", ret); + force_sig(SIGKILL, current); + } + get_safe_registers(current_pt_regs()->regs.gp, + current_pt_regs()->regs.fp); + + __switch_mm(¤t->mm->context.id); } -#ifdef CONFIG_TTY_LOG -extern void log_exec(char **argv, void *tty); -#endif - -static long execve1(char *file, char __user * __user *argv, - char __user *__user *env) +void start_thread(struct pt_regs *regs, unsigned long eip, unsigned long esp) { - long error; - struct tty_struct *tty; - -#ifdef CONFIG_TTY_LOG - mutex_lock(&tty_mutex); - tty = get_current_tty(); - if (tty) - log_exec(argv, tty); - mutex_unlock(&tty_mutex); -#endif - error = do_execve(file, argv, env, ¤t->thread.regs); - if (error == 0){ - task_lock(current); - current->ptrace &= ~PT_DTRACE; + PT_REGS_IP(regs) = eip; + PT_REGS_SP(regs) = esp; + current->ptrace &= ~PT_DTRACE; #ifdef SUBARCH_EXECVE1 - SUBARCH_EXECVE1(¤t->thread.regs.regs); + SUBARCH_EXECVE1(regs->regs); #endif - 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 __user *file, char __user *__user *argv, - char __user *__user *env) -{ - long error; - char *filename; - - lock_kernel(); - filename = getname(file); - error = PTR_ERR(filename); - if (IS_ERR(filename)) goto out; - error = execve1(filename, argv, env); - putname(filename); - out: - unlock_kernel(); - return(error); } +EXPORT_SYMBOL(start_thread); diff --git a/arch/um/kernel/exitcode.c b/arch/um/kernel/exitcode.c index 8b7f2cdedf9..41ebbfebb33 100644 --- a/arch/um/kernel/exitcode.c +++ b/arch/um/kernel/exitcode.c @@ -1,77 +1,80 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/init.h" -#include "linux/ctype.h" -#include "linux/proc_fs.h" -#include "asm/uaccess.h" +#include <linux/ctype.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/types.h> +#include <asm/uaccess.h> -/* If read and write race, the read will still atomically read a valid +/* + * 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) +static int exitcode_proc_show(struct seq_file *m, void *v) { - int len, val; + int val; - /* Save uml_exitcode in a local so that we don't need to guarantee + /* + * Save uml_exitcode in a local so that we don't need to guarantee * that sprintf accesses it atomically. */ val = uml_exitcode; - len = sprintf(page, "%d\n", val); - len -= off; - if(len <= off+count) *eof = 1; - *start = page + off; - if(len > count) len = count; - if(len < 0) len = 0; - return(len); + seq_printf(m, "%d\n", val); + return 0; +} + +static int exitcode_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, exitcode_proc_show, NULL); } -static int write_proc_exitcode(struct file *file, const char __user *buffer, - unsigned long count, void *data) +static ssize_t exitcode_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *pos) { char *end, buf[sizeof("nnnnn\0")]; + size_t size; int tmp; - if(copy_from_user(buf, buffer, count)) - return(-EFAULT); + size = min(count, sizeof(buf)); + if (copy_from_user(buf, buffer, size)) + return -EFAULT; + tmp = simple_strtol(buf, &end, 0); - if((*end != '\0') && !isspace(*end)) - return(-EINVAL); + if ((*end != '\0') && !isspace(*end)) + return -EINVAL; + uml_exitcode = tmp; - return(count); + return count; } +static const struct file_operations exitcode_proc_fops = { + .owner = THIS_MODULE, + .open = exitcode_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = exitcode_proc_write, +}; + static int make_proc_exitcode(void) { struct proc_dir_entry *ent; - ent = create_proc_entry("exitcode", 0600, &proc_root); - if(ent == NULL){ + ent = proc_create("exitcode", 0600, NULL, &exitcode_proc_fops); + if (ent == NULL) { printk(KERN_WARNING "make_proc_exitcode : Failed to register " "/proc/exitcode\n"); - return(0); + return 0; } - - ent->read_proc = read_proc_exitcode; - ent->write_proc = write_proc_exitcode; - - return(0); + 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 index 13aa115cd1b..1bf61266da8 100644 --- a/arch/um/kernel/gmon_syms.c +++ b/arch/um/kernel/gmon_syms.c @@ -1,23 +1,9 @@ -/* - * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/module.h" +#include <linux/module.h> extern void __bb_init_func(void *) __attribute__((weak)); 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); diff --git a/arch/um/kernel/gprof_syms.c b/arch/um/kernel/gprof_syms.c index 9244f018d44..74ddb44288a 100644 --- a/arch/um/kernel/gprof_syms.c +++ b/arch/um/kernel/gprof_syms.c @@ -1,20 +1,9 @@ /* - * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/module.h" +#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/init_task.c b/arch/um/kernel/init_task.c deleted file mode 100644 index 8cde431348c..00000000000 --- a/arch/um/kernel/init_task.c +++ /dev/null @@ -1,61 +0,0 @@ -/* - * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#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" -#include "os.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) -{ - os_protect_memory((void *) stack, (1 << CONFIG_KERNEL_STACK_ORDER) * PAGE_SIZE, - 1, 1, 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/initrd.c b/arch/um/kernel/initrd.c index 82ecf904b09..55cead809b1 100644 --- a/arch/um/kernel/initrd.c +++ b/arch/um/kernel/initrd.c @@ -1,20 +1,18 @@ /* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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" +#include <linux/init.h> +#include <linux/bootmem.h> +#include <linux/initrd.h> +#include <asm/types.h> +#include <init.h> +#include <os.h> /* Changed by uml_initrd_setup, which is a setup */ static char *initrd __initdata = NULL; +static int load_initrd(char *filename, void *buf, int size); static int __init read_initrd(void) { @@ -22,12 +20,29 @@ static int __init read_initrd(void) long long size; int err; - if(initrd == NULL) return 0; + if (initrd == NULL) + return 0; + err = os_file_size(initrd, &size); - if(err) return 0; + if (err) + return 0; + + /* + * This is necessary because alloc_bootmem craps out if you + * ask for no memory. + */ + if (size == 0) { + printk(KERN_ERR "\"%s\" is a zero-size initrd\n", initrd); + return 0; + } + area = alloc_bootmem(size); - if(area == NULL) return 0; - if(load_initrd(initrd, area, size) == -1) return 0; + 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; @@ -47,32 +62,24 @@ __uml_setup("initrd=", uml_initrd_setup, " name of the file containing the image.\n\n" ); -int load_initrd(char *filename, void *buf, int size) +static 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); + if (fd < 0) { + printk(KERN_ERR "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, + if (n != size) { + printk(KERN_ERR "Read of %d bytes from '%s' failed, " + "err = %d\n", size, filename, -n); - return(-1); + return -1; } os_close_file(fd); - return(0); + 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 index 5c1e611f628..1d8505b1e29 100644 --- a/arch/um/kernel/irq.c +++ b/arch/um/kernel/irq.c @@ -1,90 +1,36 @@ -/* - * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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/kernel.h" -#include "linux/module.h" -#include "linux/smp.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" -#include "os.h" -#include "sigio.h" -#include "um_malloc.h" -#include "misc_constants.h" +#include <linux/cpumask.h> +#include <linux/hardirq.h> +#include <linux/interrupt.h> +#include <linux/kernel_stat.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <linux/seq_file.h> +#include <linux/slab.h> +#include <as-layout.h> +#include <kern_util.h> +#include <os.h> /* - * Generic, controller-independent functions: + * This list is accessed under irq_lock, except in sigio_handler, + * where it is safe from being modified. IRQ handlers won't change it - + * if an IRQ source has vanished, it will be freed by free_irqs just + * before returning from sigio_handler. That will process a separate + * list of irqs to free, with its own locking, coming back here to + * remove list elements, taking the irq_lock to do so. */ - -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].chip->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; -} - -struct irq_fd *active_fds = NULL; +static struct irq_fd *active_fds = NULL; static struct irq_fd **last_irq_ptr = &active_fds; extern void free_irqs(void); -void sigio_handler(int sig, union uml_pt_regs *regs) +void sigio_handler(int sig, struct siginfo *unused_si, struct uml_pt_regs *regs) { struct irq_fd *irq_fd; int n; @@ -95,11 +41,13 @@ void sigio_handler(int sig, union uml_pt_regs *regs) while (1) { n = os_waiting_for_events(active_fds); if (n <= 0) { - if(n == -EINTR) continue; + if (n == -EINTR) + continue; else break; } - for (irq_fd = active_fds; irq_fd != NULL; irq_fd = irq_fd->next) { + for (irq_fd = active_fds; irq_fd != NULL; + irq_fd = irq_fd->next) { if (irq_fd->current_events != 0) { irq_fd->current_events = 0; do_IRQ(irq_fd->irq, regs); @@ -112,42 +60,41 @@ void sigio_handler(int sig, union uml_pt_regs *regs) static DEFINE_SPINLOCK(irq_lock); -int activate_fd(int irq, int fd, int type, void *dev_id) +static 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; + int events, err, n; - pid = os_getpid(); - err = os_set_fd_async(fd, pid); + err = os_set_fd_async(fd); if (err < 0) goto out; - new_fd = um_kmalloc(sizeof(*new_fd)); err = -ENOMEM; + new_fd = kmalloc(sizeof(struct irq_fd), GFP_KERNEL); if (new_fd == NULL) goto out; if (type == IRQ_READ) events = UM_POLLIN | UM_POLLPRI; - else - events = UM_POLLOUT; + else events = UM_POLLOUT; *new_fd = ((struct irq_fd) { .next = NULL, .id = dev_id, .fd = fd, .type = type, .irq = irq, - .pid = pid, .events = events, .current_events = 0 } ); + err = -EBUSY; spin_lock_irqsave(&irq_lock, flags); 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%p, 0x%p\n", irq_fd->id, dev_id); + printk(KERN_ERR "Registering fd %d twice\n", fd); + printk(KERN_ERR "Irqs : %d, %d\n", irq_fd->irq, irq); + printk(KERN_ERR "Ids : 0x%p, 0x%p\n", irq_fd->id, + dev_id); goto out_unlock; } } @@ -163,7 +110,8 @@ int activate_fd(int irq, int fd, int type, void *dev_id) if (n == 0) break; - /* n > 0 + /* + * n > 0 * It means we couldn't put new pollfd to current pollfds * and tmp_fds is NULL or too small for new pollfds array. * Needed size is equal to n as minimum. @@ -176,9 +124,8 @@ int activate_fd(int irq, int fd, int type, void *dev_id) */ spin_unlock_irqrestore(&irq_lock, flags); kfree(tmp_pfd); - tmp_pfd = NULL; - tmp_pfd = um_kmalloc(n); + tmp_pfd = kmalloc(n, GFP_KERNEL); if (tmp_pfd == NULL) goto out_kfree; @@ -190,7 +137,8 @@ int activate_fd(int irq, int fd, int type, void *dev_id) spin_unlock_irqrestore(&irq_lock, flags); - /* This calls activate_fd, so it has to be outside the critical + /* + * This calls activate_fd, so it has to be outside the critical * section. */ maybe_sigio_broken(fd, (type == IRQ_READ)); @@ -226,7 +174,7 @@ static int same_irq_and_dev(struct irq_fd *irq, void *d) return ((irq->irq == data->irq) && (irq->id == data->dev)); } -void free_irq_by_irq_and_dev(unsigned int irq, void *dev) +static 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 }); @@ -244,6 +192,7 @@ void free_irq_by_fd(int fd) free_irq_by_cb(same_fd, &fd); } +/* Must be called with irq_lock held */ static struct irq_fd *find_irq_by_fd(int fd, int irqnum, int *index_out) { struct irq_fd *irq; @@ -256,13 +205,14 @@ static struct irq_fd *find_irq_by_fd(int fd, int irqnum, int *index_out) i++; } if (irq == NULL) { - printk("find_irq_by_fd doesn't have descriptor %d\n", fd); + printk(KERN_ERR "find_irq_by_fd doesn't have descriptor %d\n", + fd); goto out; } fdi = os_get_pollfd(i); if ((fdi != -1) && (fdi != fd)) { - printk("find_irq_by_fd - mismatch between active_fds and " - "pollfds, fd %d vs %d, need %d\n", irq->fd, + printk(KERN_ERR "find_irq_by_fd - mismatch between active_fds " + "and pollfds, fd %d vs %d, need %d\n", irq->fd, fdi, fd); irq = NULL; goto out; @@ -298,7 +248,7 @@ void deactivate_fd(int fd, int irqnum) spin_lock_irqsave(&irq_lock, flags); irq = find_irq_by_fd(fd, irqnum, &i); - if(irq == NULL){ + if (irq == NULL) { spin_unlock_irqrestore(&irq_lock, flags); return; } @@ -308,7 +258,14 @@ void deactivate_fd(int fd, int irqnum) ignore_sigio_fd(fd); } +EXPORT_SYMBOL(deactivate_fd); +/* + * Called just before shutdown in order to provide a clean exec + * environment in case the system is rebooting. No locking because + * that would cause a pointless shutdown hang if something hadn't + * released the lock. + */ int deactivate_all_fds(void) { struct irq_fd *irq; @@ -325,45 +282,28 @@ int deactivate_all_fds(void) return 0; } -#ifdef CONFIG_MODE_TT -void forward_interrupts(int pid) -{ - struct irq_fd *irq; - unsigned long flags; - int err; - - spin_lock_irqsave(&irq_lock, flags); - 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; - } - spin_unlock_irqrestore(&irq_lock, flags); -} -#endif - /* - * do_IRQ handles all normal device IRQ's (the special + * do_IRQ handles all normal device IRQs (the special * SMP cross-CPU interrupts have their own specific * handlers). */ -unsigned int do_IRQ(int irq, union uml_pt_regs *regs) +unsigned int do_IRQ(int irq, struct uml_pt_regs *regs) { struct pt_regs *old_regs = set_irq_regs((struct pt_regs *)regs); irq_enter(); - __do_IRQ(irq); + generic_handle_irq(irq); irq_exit(); set_irq_regs(old_regs); return 1; } +void um_free_irq(unsigned int irq, void *dev) +{ + free_irq_by_irq_and_dev(irq, dev); + free_irq(irq, dev); +} +EXPORT_SYMBOL(um_free_irq); + int um_request_irq(unsigned int irq, int fd, int type, irq_handler_t handler, unsigned long irqflags, const char * devname, @@ -371,86 +311,163 @@ int um_request_irq(unsigned int irq, int fd, int type, { int err; - err = request_irq(irq, handler, irqflags, devname, dev_id); - if (err) - return err; - - if (fd != -1) + if (fd != -1) { err = activate_fd(irq, fd, type, dev_id); - return err; + if (err) + return err; + } + + return request_irq(irq, handler, irqflags, devname, dev_id); } + EXPORT_SYMBOL(um_request_irq); EXPORT_SYMBOL(reactivate_fd); -/* hw_interrupt_type must define (startup || enable) && - * (shutdown || disable) && end */ -static void dummy(unsigned int irq) +/* + * irq_chip must define at least enable/disable and ack when + * the edge handler is used. + */ +static void dummy(struct irq_data *d) { } /* This is used for everything else than the timer. */ -static struct hw_interrupt_type normal_irq_type = { - .typename = "SIGIO", - .release = free_irq_by_irq_and_dev, - .disable = dummy, - .enable = dummy, - .ack = dummy, - .end = dummy +static struct irq_chip normal_irq_type = { + .name = "SIGIO", + .irq_disable = dummy, + .irq_enable = dummy, + .irq_ack = dummy, + .irq_mask = dummy, + .irq_unmask = dummy, }; -static struct hw_interrupt_type SIGVTALRM_irq_type = { - .typename = "SIGVTALRM", - .release = free_irq_by_irq_and_dev, - .shutdown = dummy, /* never called */ - .disable = dummy, - .enable = dummy, - .ack = dummy, - .end = dummy +static struct irq_chip SIGVTALRM_irq_type = { + .name = "SIGVTALRM", + .irq_disable = dummy, + .irq_enable = dummy, + .irq_ack = dummy, + .irq_mask = dummy, + .irq_unmask = 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].chip = &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].chip = &normal_irq_type; - enable_irq(i); - } + irq_set_chip_and_handler(TIMER_IRQ, &SIGVTALRM_irq_type, handle_edge_irq); + + for (i = 1; i < NR_IRQS; i++) + irq_set_chip_and_handler(i, &normal_irq_type, handle_edge_irq); } -int init_aio_irq(int irq, char *name, irq_handler_t handler) -{ - int fds[2], err; +/* + * IRQ stack entry and exit: + * + * Unlike i386, UML doesn't receive IRQs on the normal kernel stack + * and switch over to the IRQ stack after some preparation. We use + * sigaltstack to receive signals on a separate stack from the start. + * These two functions make sure the rest of the kernel won't be too + * upset by being on a different stack. The IRQ stack has a + * thread_info structure at the bottom so that current et al continue + * to work. + * + * to_irq_stack copies the current task's thread_info to the IRQ stack + * thread_info and sets the tasks's stack to point to the IRQ stack. + * + * from_irq_stack copies the thread_info struct back (flags may have + * been modified) and resets the task's stack pointer. + * + * Tricky bits - + * + * What happens when two signals race each other? UML doesn't block + * signals with sigprocmask, SA_DEFER, or sa_mask, so a second signal + * could arrive while a previous one is still setting up the + * thread_info. + * + * There are three cases - + * The first interrupt on the stack - sets up the thread_info and + * handles the interrupt + * A nested interrupt interrupting the copying of the thread_info - + * can't handle the interrupt, as the stack is in an unknown state + * A nested interrupt not interrupting the copying of the + * thread_info - doesn't do any setup, just handles the interrupt + * + * The first job is to figure out whether we interrupted stack setup. + * This is done by xchging the signal mask with thread_info->pending. + * If the value that comes back is zero, then there is no setup in + * progress, and the interrupt can be handled. If the value is + * non-zero, then there is stack setup in progress. In order to have + * the interrupt handled, we leave our signal in the mask, and it will + * be handled by the upper handler after it has set up the stack. + * + * Next is to figure out whether we are the outer handler or a nested + * one. As part of setting up the stack, thread_info->real_thread is + * set to non-NULL (and is reset to NULL on exit). This is the + * nesting indicator. If it is non-NULL, then the stack is already + * set up and the handler can run. + */ - err = os_pipe(fds, 1, 1); - if (err) { - printk("init_aio_irq - os_pipe failed, err = %d\n", -err); - goto out; +static unsigned long pending_mask; + +unsigned long to_irq_stack(unsigned long *mask_out) +{ + struct thread_info *ti; + unsigned long mask, old; + int nested; + + mask = xchg(&pending_mask, *mask_out); + if (mask != 0) { + /* + * If any interrupts come in at this point, we want to + * make sure that their bits aren't lost by our + * putting our bit in. So, this loop accumulates bits + * until xchg returns the same value that we put in. + * When that happens, there were no new interrupts, + * and pending_mask contains a bit for each interrupt + * that came in. + */ + old = *mask_out; + do { + old |= mask; + mask = xchg(&pending_mask, old); + } while (mask != old); + return 1; } - err = um_request_irq(irq, fds[0], IRQ_READ, handler, - IRQF_DISABLED | IRQF_SAMPLE_RANDOM, name, - (void *) (long) fds[0]); - if (err) { - printk("init_aio_irq - : um_request_irq failed, err = %d\n", - err); - goto out_close; + ti = current_thread_info(); + nested = (ti->real_thread != NULL); + if (!nested) { + struct task_struct *task; + struct thread_info *tti; + + task = cpu_tasks[ti->cpu].task; + tti = task_thread_info(task); + + *ti = *tti; + ti->real_thread = tti; + task->stack = ti; } - err = fds[1]; - goto out; + mask = xchg(&pending_mask, 0); + *mask_out |= mask | nested; + return 0; +} - out_close: - os_close_file(fds[0]); - os_close_file(fds[1]); - out: - return err; +unsigned long from_irq_stack(int nested) +{ + struct thread_info *ti, *to; + unsigned long mask; + + ti = current_thread_info(); + + pending_mask = 1; + + to = ti->real_thread; + current->stack = to; + ti->real_thread = NULL; + *to = *ti; + + mask = xchg(&pending_mask, 0); + return mask & ~1; } + diff --git a/arch/um/kernel/ksyms.c b/arch/um/kernel/ksyms.c index 0e00cf93f90..543c0475693 100644 --- a/arch/um/kernel/ksyms.c +++ b/arch/um/kernel/ksyms.c @@ -1,64 +1,17 @@ /* - * Copyright (C) 2001 - 2004 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#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/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 <linux/module.h> +#include <os.h> -EXPORT_SYMBOL(uml_physmem); EXPORT_SYMBOL(set_signals); EXPORT_SYMBOL(get_signals); -EXPORT_SYMBOL(kernel_thread); -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(high_physmem); -EXPORT_SYMBOL(empty_zero_page); -EXPORT_SYMBOL(um_virt_to_phys); -EXPORT_SYMBOL(mode_tt); -EXPORT_SYMBOL(handle_page_fault); -EXPORT_SYMBOL(find_iomem); - -#ifdef CONFIG_MODE_TT -EXPORT_SYMBOL(stop); -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(strnlen_user_skas); -EXPORT_SYMBOL(strncpy_from_user_skas); -EXPORT_SYMBOL(copy_to_user_skas); -EXPORT_SYMBOL(copy_from_user_skas); -EXPORT_SYMBOL(clear_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); @@ -82,20 +35,10 @@ 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); - -#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); +EXPORT_SYMBOL(os_major); +EXPORT_SYMBOL(os_minor); +EXPORT_SYMBOL(os_makedev); -#endif +EXPORT_SYMBOL(add_sigio_fd); +EXPORT_SYMBOL(ignore_sigio_fd); +EXPORT_SYMBOL(sigio_broken); diff --git a/arch/um/kernel/maccess.c b/arch/um/kernel/maccess.c new file mode 100644 index 00000000000..1f3d5c4910d --- /dev/null +++ b/arch/um/kernel/maccess.c @@ -0,0 +1,24 @@ +/* + * Copyright (C) 2013 Richard Weinberger <richrd@nod.at> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#include <linux/uaccess.h> +#include <linux/kernel.h> +#include <os.h> + +long probe_kernel_read(void *dst, const void *src, size_t size) +{ + void *psrc = (void *)rounddown((unsigned long)src, PAGE_SIZE); + + if ((unsigned long)src < PAGE_SIZE || size <= 0) + return -EFAULT; + + if (os_mincore(psrc, size + src - psrc) <= 0) + return -EFAULT; + + return __probe_kernel_read(dst, src, size); +} diff --git a/arch/um/kernel/mem.c b/arch/um/kernel/mem.c index c95855ba6ab..8636e905426 100644 --- a/arch/um/kernel/mem.c +++ b/arch/um/kernel/mem.c @@ -1,97 +1,78 @@ /* - * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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" -#include "linux/types.h" -#include "linux/string.h" -#include "init.h" -#include "kern_constants.h" - -/* Changed during early boot */ +#include <linux/stddef.h> +#include <linux/module.h> +#include <linux/bootmem.h> +#include <linux/highmem.h> +#include <linux/mm.h> +#include <linux/swap.h> +#include <linux/slab.h> +#include <asm/fixmap.h> +#include <asm/page.h> +#include <as-layout.h> +#include <init.h> +#include <kern.h> +#include <kern_util.h> +#include <mem_user.h> +#include <os.h> + +/* allocated in paging_init, zeroed in mem_init, and unchanged thereafter */ unsigned long *empty_zero_page = NULL; -unsigned long *empty_bad_page = NULL; +EXPORT_SYMBOL(empty_zero_page); +/* allocated in paging_init and unchanged thereafter */ +static unsigned long *empty_bad_page = NULL; + +/* + * Initialized during boot, and readonly for initializing page tables + * afterwards + */ pgd_t swapper_pg_dir[PTRS_PER_PGD]; + +/* Initialized at boot time, and readonly after that */ unsigned long long highmem; int kmalloc_ok = 0; +/* Used during early boot */ 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); - init_page_count(page); - __free_page(page); - } + for (i = 0; i < highmem_len >> PAGE_SHIFT; i++) + free_highmem_page(&mem_map[highmem_pfn + i]); } #endif -void mem_init(void) +void __init mem_init(void) { - max_low_pfn = (high_physmem - uml_physmem) >> PAGE_SHIFT; - - /* clear the zero-page */ - memset((void *) empty_zero_page, 0, PAGE_SIZE); + /* clear the zero-page */ + memset(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); + map_memory(brk_end, __pa(brk_end), uml_reserved - brk_end, 1, 1, 0); free_bootmem(__pa(brk_end), uml_reserved - brk_end); uml_reserved = brk_end; /* this will put all low memory onto the freelists */ - totalram_pages = free_all_bootmem(); + free_all_bootmem(); + max_low_pfn = totalram_pages; #ifdef CONFIG_HIGHMEM - totalhigh_pages = highmem >> PAGE_SHIFT; - totalram_pages += totalhigh_pages; + setup_highmem(end_iomem, highmem); #endif - num_physpages = totalram_pages; max_pfn = totalram_pages; - printk(KERN_INFO "Memory: %luk available\n", - (unsigned long) nr_free_pages() << (PAGE_SHIFT-10)); + mem_init_print_info(NULL); kmalloc_ok = 1; - -#ifdef CONFIG_HIGHMEM - setup_highmem(end_iomem, highmem); -#endif } /* @@ -119,7 +100,7 @@ static void __init one_md_table_init(pud_t *pud) #endif } -static void __init fixrange_init(unsigned long start, unsigned long end, +static void __init fixrange_init(unsigned long start, unsigned long end, pgd_t *pgd_base) { pgd_t *pgd; @@ -138,7 +119,7 @@ static void __init fixrange_init(unsigned long start, unsigned long end, if (pud_none(*pud)) one_md_table_init(pud); pmd = pmd_offset(pud, vaddr); - for (; (j < PTRS_PER_PMD) && (vaddr != end); pmd++, j++) { + for (; (j < PTRS_PER_PMD) && (vaddr < end); pmd++, j++) { one_page_table_init(pmd); vaddr += PMD_SIZE; } @@ -152,7 +133,7 @@ pgprot_t kmap_prot; #define kmap_get_fixmap_pte(vaddr) \ pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), (vaddr)),\ - (vaddr)), (vaddr)) + (vaddr)), (vaddr)) static void __init kmap_init(void) { @@ -165,7 +146,7 @@ static void __init kmap_init(void) kmap_prot = PAGE_KERNEL; } -static void init_highmem(void) +static void __init init_highmem(void) { pgd_t *pgd; pud_t *pud; @@ -197,33 +178,35 @@ static void __init fixaddr_user_init( void) pud_t *pud; pmd_t *pmd; pte_t *pte; - unsigned long paddr, vaddr = FIXADDR_USER_START; + phys_t p; + unsigned long v, vaddr = FIXADDR_USER_START; - if ( ! size ) + 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){ + v = (unsigned long) alloc_bootmem_low_pages(size); + memcpy((void *) v , (void *) FIXADDR_USER_START, size); + p = __pa(v); + for ( ; size > 0; size -= PAGE_SIZE, vaddr += PAGE_SIZE, + p += 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); + pte_set_val(*pte, p, PAGE_READONLY); } #endif } -void paging_init(void) +void __init 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 < ARRAY_SIZE(zones_size); i++) + for (i = 0; i < ARRAY_SIZE(zones_size); i++) zones_size[i] = 0; zones_size[ZONE_NORMAL] = (end_iomem >> PAGE_SHIFT) - @@ -247,34 +230,8 @@ void paging_init(void) #endif } -struct page *arch_validate(struct page *page, gfp_t 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 +/* + * This can't do anything because nothing in the kernel image can be freed * since it's not in kernel physical memory. */ @@ -283,55 +240,13 @@ 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)); - init_page_count(virt_to_page(start)); - free_page(start); - totalram_pages++; - } + free_reserved_area((void *)start, (void *)end, -1, "initrd"); } - #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. - */ +/* Allocate and free page tables. */ pgd_t *pgd_alloc(struct mm_struct *mm) { @@ -339,14 +254,14 @@ pgd_t *pgd_alloc(struct mm_struct *mm) 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, + 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) +void pgd_free(struct mm_struct *mm, pgd_t *pgd) { free_page((unsigned long) pgd); } @@ -359,31 +274,33 @@ pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) return pte; } -struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address) +pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address) { struct page *pte; - + pte = alloc_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO); + if (!pte) + return NULL; + if (!pgtable_page_ctor(pte)) { + __free_page(pte); + return NULL; + } return pte; } -struct iomem_region *iomem_regions = NULL; -int iomem_size = 0; +#ifdef CONFIG_3_LEVEL_PGTABLES +pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long address) +{ + pmd_t *pmd = (pmd_t *) __get_free_page(GFP_KERNEL); -extern int parse_iomem(char *str, int *add) __init; + if (pmd) + memset(pmd, 0, PAGE_SIZE); -__uml_setup("iomem=", parse_iomem, -"iomem=<name>,<file>\n" -" Configure <file> as an IO memory region named <name>.\n\n" -); + return pmd; +} +#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: - */ +void *uml_kmalloc(int size, int flags) +{ + return kmalloc(size, flags); +} diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c index abafa64b872..30fdd5d0067 100644 --- a/arch/um/kernel/physmem.c +++ b/arch/um/kernel/physmem.c @@ -1,255 +1,29 @@ /* - * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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 "linux/pfn.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, rb_parent(*n), 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); -} +#include <linux/module.h> +#include <linux/bootmem.h> +#include <linux/mm.h> +#include <linux/pfn.h> +#include <asm/page.h> +#include <as-layout.h> +#include <init.h> +#include <kern.h> +#include <mem_user.h> +#include <os.h> 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; +EXPORT_SYMBOL(high_physmem); extern unsigned long long physmem_size; -int init_maps(unsigned long physmem, unsigned long iomem, unsigned long highmem) +int __init 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; @@ -268,17 +42,11 @@ int init_maps(unsigned long physmem, unsigned long iomem, unsigned long highmem) total_pages = phys_pages + iomem_pages + highmem_pages; total_len = phys_len + iomem_len + 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); + map = alloc_bootmem_low_pages(total_len); + if (map == NULL) + return -ENOMEM; - for(i = 0; i < total_pages; i++){ + for (i = 0; i < total_pages; i++) { p = &map[i]; memset(p, 0, sizeof(struct page)); SetPageReserved(p); @@ -286,17 +54,7 @@ int init_maps(unsigned long physmem, unsigned long iomem, unsigned long highmem) } max_mapnr = total_pages; - return(0); -} - -/* 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); + return 0; } void map_memory(unsigned long virt, unsigned long phys, unsigned long len, @@ -307,9 +65,9 @@ void map_memory(unsigned long virt, unsigned long phys, unsigned long len, 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 " + if (err) { + if (err == -ENOMEM) + printk(KERN_ERR "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, " @@ -319,8 +77,8 @@ void map_memory(unsigned long virt, unsigned long phys, unsigned long len, extern int __syscall_stub_start; -void setup_physmem(unsigned long start, unsigned long reserve_end, - unsigned long len, unsigned long long highmem) +void __init setup_physmem(unsigned long start, unsigned long reserve_end, + unsigned long len, unsigned long long highmem) { unsigned long reserve = reserve_end - start; int pfn = PFN_UP(__pa(reserve_end)); @@ -331,42 +89,41 @@ void setup_physmem(unsigned long start, unsigned long reserve_end, 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"); + len - offset, 1, 1, 1); + if (err < 0) { + printf("setup_physmem - mapping %ld bytes of memory at 0x%p " + "failed - errno = %d\n", len - offset, + (void *) uml_reserved, err); exit(1); } - /* Special kludge - This page will be mapped in to userspace processes + /* + * Special kludge - This page will be mapped in to userspace processes * from physmem_fd, so it needs to be written out there. */ os_seek_file(physmem_fd, __pa(&__syscall_stub_start)); os_write_file(physmem_fd, &__syscall_stub_start, PAGE_SIZE); + os_fsync_file(physmem_fd); 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) +int phys_mapping(unsigned long phys, unsigned long long *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){ + if (phys < physmem_size) { fd = physmem_fd; *offset_out = phys; } - else if(phys < __pa(end_iomem)){ + else if (phys < __pa(end_iomem)) { struct iomem_region *region = iomem_regions; - while(region != NULL){ - if((phys >= region->phys) && - (phys < region->phys + region->size)){ + while (region != NULL) { + if ((phys >= region->phys) && + (phys < region->phys + region->size)) { fd = region->fd; *offset_out = phys - region->phys; break; @@ -374,12 +131,12 @@ int phys_mapping(unsigned long phys, __u64 *offset_out) region = region->next; } } - else if(phys < __pa(end_iomem) + highmem){ + else if (phys < __pa(end_iomem) + highmem) { fd = physmem_fd; *offset_out = phys - iomem_size; } - return(fd); + return fd; } static int __init uml_mem_setup(char *line, int *add) @@ -398,34 +155,52 @@ __uml_setup("mem=", uml_mem_setup, " Example: mem=64M\n\n" ); +extern int __init parse_iomem(char *str, int *add); + +__uml_setup("iomem=", parse_iomem, +"iomem=<name>,<file>\n" +" Configure <file> as an IO memory region named <name>.\n\n" +); + +/* + * This list is constructed in parse_iomem and addresses filled in in + * setup_iomem, both of which run during early boot. Afterwards, it's + * unchanged. + */ +struct iomem_region *iomem_regions; + +/* Initialized in parse_iomem and unchanged thereafter */ +int iomem_size; + unsigned long find_iomem(char *driver, unsigned long *len_out) { struct iomem_region *region = iomem_regions; - while(region != NULL){ - if(!strcmp(region->driver, driver)){ + while (region != NULL) { + if (!strcmp(region->driver, driver)) { *len_out = region->size; - return(region->virt); + return region->virt; } region = region->next; } - return(0); + return 0; } +EXPORT_SYMBOL(find_iomem); -int setup_iomem(void) +static int setup_iomem(void) { struct iomem_region *region = iomem_regions; unsigned long iomem_start = high_physmem + PAGE_SIZE; int err; - while(region != NULL){ + 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); + if (err) + printk(KERN_ERR "Mapping iomem region for driver '%s' " + "failed, errno = %d\n", region->driver, -err); else { region->virt = iomem_start; region->phys = __pa(region->virt); @@ -435,18 +210,7 @@ int setup_iomem(void) region = region->next; } - return(0); + 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 index 348b272bb76..f17bca8ed2c 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -1,74 +1,55 @@ /* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Copyright 2003 PathScale, Inc. * Licensed under the GPL */ -#include "linux/kernel.h" -#include "linux/sched.h" -#include "linux/interrupt.h" -#include "linux/string.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 "linux/personality.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 "init.h" -#include "irq_user.h" -#include "mem_user.h" -#include "tlb.h" -#include "frame_kern.h" -#include "sigcontext.h" -#include "os.h" -#include "mode.h" -#include "mode_kern.h" -#include "choose-mode.h" -#include "um_malloc.h" - -/* This is a per-cpu array. A processor only modifies its entry and it only +#include <linux/stddef.h> +#include <linux/err.h> +#include <linux/hardirq.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/personality.h> +#include <linux/proc_fs.h> +#include <linux/ptrace.h> +#include <linux/random.h> +#include <linux/slab.h> +#include <linux/sched.h> +#include <linux/seq_file.h> +#include <linux/tick.h> +#include <linux/threads.h> +#include <linux/tracehook.h> +#include <asm/current.h> +#include <asm/pgtable.h> +#include <asm/mmu_context.h> +#include <asm/uaccess.h> +#include <as-layout.h> +#include <kern_util.h> +#include <os.h> +#include <skas.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 } }; -int external_pid(void *t) +static inline int external_pid(void) { - struct task_struct *task = t ? t : current; - - return(CHOOSE_MODE_PROC(external_pid_tt, external_pid_skas, task)); + /* FIXME: Need to look up userspace_pid by cpu */ + return userspace_pid[0]; } int pid_to_processor_id(int pid) { int i; - for(i = 0; i < ncpus; i++){ - if(cpu_tasks[i].pid == pid) return(i); + for (i = 0; i < ncpus; i++) { + if (cpu_tasks[i].pid == pid) + return i; } - return(-1); + return -1; } void free_stack(unsigned long stack, int order) @@ -84,238 +65,148 @@ unsigned long alloc_stack(int order, int atomic) 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, - ¤t->thread.regs, 0, NULL, NULL); - if(pid < 0) - panic("do_fork failed in kernel_thread, errno = %d", pid); - return(pid); + return page; } -void set_current(void *t) +static inline void set_current(struct task_struct *task) { - struct task_struct *task = t; - cpu_tasks[task_thread_info(task)->cpu] = ((struct cpu_task) - { external_pid(task), task }); + { external_pid(), task }); } -void *_switch_to(void *prev, void *next, void *last) -{ - struct task_struct *from = prev; - struct task_struct *to= next; +extern void arch_switch_to(struct task_struct *to); +void *__switch_to(struct task_struct *from, struct task_struct *to) +{ to->thread.prev_sched = from; set_current(to); - do { - current->thread.saved_task = NULL ; - CHOOSE_MODE_PROC(switch_to_tt, switch_to_skas, prev, next); - if(current->thread.saved_task) - show_regs(&(current->thread.regs)); - next= current->thread.saved_task; - prev= current; - } while(current->thread.saved_task); - - return(current->thread.prev_sched); + switch_threads(&from->thread.switch_buf, &to->thread.switch_buf); + arch_switch_to(current); + return current->thread.prev_sched; } 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)); + if (need_resched()) + schedule(); + if (test_thread_flag(TIF_SIGPENDING)) + do_signal(); + if (test_and_clear_thread_flag(TIF_NOTIFY_RESUME)) + tracehook_notify_resume(¤t->thread.regs); } void exit_thread(void) { - unprotect_stack((unsigned long) current_thread); } -void *get_current(void) +int get_current_pid(void) { - return(current); + return task_pid_nr(current); } -int copy_thread(int nr, unsigned long clone_flags, unsigned long sp, - unsigned long stack_top, struct task_struct * p, - struct pt_regs *regs) +/* + * This is called magically, by its address being stuffed in a jmp_buf + * and being longjmp-d to. + */ +void new_thread_handler(void) { - int ret; - - p->thread = (struct thread_struct) INIT_THREAD; - ret = CHOOSE_MODE_PROC(copy_thread_tt, copy_thread_skas, nr, - clone_flags, sp, stack_top, p, regs); + int (*fn)(void *), n; + void *arg; - if (ret || !current->thread.forking) - goto out; + if (current->thread.prev_sched != NULL) + schedule_tail(current->thread.prev_sched); + current->thread.prev_sched = NULL; - clear_flushed_tls(p); + fn = current->thread.request.u.thread.proc; + arg = current->thread.request.u.thread.arg; /* - * Set a new TLS for the child thread? + * callback returns only if the kernel thread execs a process */ - if (clone_flags & CLONE_SETTLS) - ret = arch_copy_tls(p); - -out: - return ret; -} - -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; + n = fn(arg); + userspace(¤t->thread.regs.regs); } -unsigned long stack_sp(unsigned long page) +/* Called magically, see new_thread_handler above */ +void fork_handler(void) { - return(page + PAGE_SIZE - sizeof(void *)); -} - -int current_pid(void) -{ - return(current->pid); -} - -void default_idle(void) -{ - CHOOSE_MODE(uml_idle_timer(), (void) 0); - - while(1){ - /* endless idle loop with no priority at all */ + force_flush_all(); - /* - * although we are an idle CPU, we do not want to - * get into the scheduler unnecessarily. - */ - if(need_resched()) - schedule(); + schedule_tail(current->thread.prev_sched); - idle_sleep(10); - } -} + /* + * XXX: if interrupt_end() calls schedule, this call to + * arch_switch_to isn't needed. We could want to apply this to + * improve performance. -bb + */ + arch_switch_to(current); -void cpu_idle(void) -{ - CHOOSE_MODE(init_idle_tt(), init_idle_skas()); -} + current->thread.prev_sched = NULL; -int page_size(void) -{ - return(PAGE_SIZE); + userspace(¤t->thread.regs.regs); } -void *um_virt_to_phys(struct task_struct *task, unsigned long addr, - pte_t *pte_out) +int copy_thread(unsigned long clone_flags, unsigned long sp, + unsigned long arg, struct task_struct * p) { - pgd_t *pgd; - pud_t *pud; - pmd_t *pmd; - pte_t *pte; - pte_t ptent; - - 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); - ptent = *pte; - if(!pte_present(ptent)) - return(ERR_PTR(-EINVAL)); - - if(pte_out != NULL) - *pte_out = ptent; - return((void *) (pte_val(ptent) & PAGE_MASK) + (addr & ~PAGE_MASK)); -} + void (*handler)(void); + int kthread = current->flags & PF_KTHREAD; + int ret = 0; -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 -} + p->thread = (struct thread_struct) INIT_THREAD; -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); -} + if (!kthread) { + memcpy(&p->thread.regs.regs, current_pt_regs(), + sizeof(p->thread.regs.regs)); + PT_REGS_SET_SYSCALL_RETURN(&p->thread.regs, 0); + if (sp != 0) + REGS_SP(p->thread.regs.regs.gp) = sp; + + handler = fork_handler; + + arch_copy_thread(¤t->thread.arch, &p->thread.arch); + } else { + get_safe_registers(p->thread.regs.regs.gp, p->thread.regs.regs.fp); + p->thread.request.u.thread.proc = (int (*)(void *))sp; + p->thread.request.u.thread.arg = (void *)arg; + handler = new_thread_handler; + } -void dump_thread(struct pt_regs *regs, struct user *u) -{ -} + new_thread(task_stack_page(p), &p->thread.switch_buf, handler); -void enable_hlt(void) -{ - panic("enable_hlt"); -} + if (!kthread) { + clear_flushed_tls(p); -EXPORT_SYMBOL(enable_hlt); + /* + * Set a new TLS for the child thread? + */ + if (clone_flags & CLONE_SETTLS) + ret = arch_copy_tls(p); + } -void disable_hlt(void) -{ - panic("disable_hlt"); + return ret; } -EXPORT_SYMBOL(disable_hlt); - -void *um_kmalloc(int size) +void initial_thread_cb(void (*proc)(void *), void *arg) { - return kmalloc(size, GFP_KERNEL); -} + int save_kmalloc_ok = kmalloc_ok; -void *um_kmalloc_atomic(int size) -{ - return kmalloc(size, GFP_ATOMIC); + kmalloc_ok = 0; + initial_thread_cb_skas(proc, arg); + kmalloc_ok = save_kmalloc_ok; } -void *um_vmalloc(int size) +void arch_cpu_idle(void) { - return vmalloc(size); -} + unsigned long long nsecs; -void *um_vmalloc_atomic(int size) -{ - return __vmalloc(size, GFP_ATOMIC | __GFP_HIGHMEM, PAGE_KERNEL); + cpu_tasks[current_thread_info()->cpu].pid = os_getpid(); + nsecs = disable_timer(); + idle_sleep(nsecs); + local_irq_enable(); } int __cant_sleep(void) { @@ -323,26 +214,12 @@ int __cant_sleep(void) { /* Is in_interrupt() really needed? */ } -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); + return stack != (unsigned long) current_thread_info(); } extern exitcall_t __uml_exitcall_begin, __uml_exitcall_end; @@ -356,45 +233,46 @@ void do_uml_exitcalls(void) (*call)(); } -char *uml_strdup(char *string) +char *uml_strdup(const char *string) { return kstrdup(string, GFP_KERNEL); } +EXPORT_SYMBOL(uml_strdup); int copy_to_user_proc(void __user *to, void *from, int size) { - return(copy_to_user(to, from, 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)); + return copy_from_user(to, from, size); } int clear_user_proc(void __user *buf, int size) { - return(clear_user(buf, size)); + return clear_user(buf, size); } int strlen_user_proc(char __user *str) { - return(strlen_user(str)); + return strlen_user(str); } int smp_sigio_handler(void) { #ifdef CONFIG_SMP - int cpu = current_thread->cpu; + int cpu = current_thread_info()->cpu; IPI_handler(cpu); - if(cpu != 0) - return(1); + if (cpu != 0) + return 1; #endif - return(0); + return 0; } int cpu(void) { - return(current_thread->cpu); + return current_thread_info()->cpu; } static atomic_t using_sysemu = ATOMIC_INIT(0); @@ -412,15 +290,19 @@ 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) +static int sysemu_proc_show(struct seq_file *m, void *v) { - if (snprintf(buf, size, "%d\n", get_using_sysemu()) < size) /*No overflow*/ - *eof = 1; + seq_printf(m, "%d\n", get_using_sysemu()); + return 0; +} - return strlen(buf); +static int sysemu_proc_open(struct inode *inode, struct file *file) +{ + return single_open(file, sysemu_proc_show, NULL); } -static int proc_write_sysemu(struct file *file,const char __user *buf, unsigned long count,void *data) +static ssize_t sysemu_proc_write(struct file *file, const char __user *buf, + size_t count, loff_t *pos) { char tmp[2]; @@ -429,26 +311,33 @@ static int proc_write_sysemu(struct file *file,const char __user *buf, unsigned 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*/ + /* We use the first char, but pretend to write everything */ + return count; } +static const struct file_operations sysemu_proc_fops = { + .owner = THIS_MODULE, + .open = sysemu_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = sysemu_proc_write, +}; + int __init make_proc_sysemu(void) { struct proc_dir_entry *ent; if (!sysemu_supported) return 0; - ent = create_proc_entry("sysemu", 0600, &proc_root); + ent = proc_create("sysemu", 0600, NULL, &sysemu_proc_fops); if (ent == NULL) { printk(KERN_WARNING "Failed to register /proc/sysemu\n"); - return(0); + return 0; } - ent->read_proc = proc_read_sysemu; - ent->write_proc = proc_write_sysemu; - return 0; } @@ -458,11 +347,11 @@ int singlestepping(void * t) { struct task_struct *task = t ? t : current; - if ( ! (task->ptrace & PT_DTRACE) ) - return(0); + if (!(task->ptrace & PT_DTRACE)) + return 0; if (task->thread.singlestep_syscall) - return(1); + return 1; return 2; } @@ -470,7 +359,7 @@ int singlestepping(void * t) /* * Only x86 and x86_64 have an arch_align_stack(). * All other arches have "#define arch_align_stack(x) (x)" - * in their asm/system.h + * in their asm/exec.h * As this is included in UML from asm-um/system-generic.h, * we can use it to behave as the subarch does. */ @@ -482,3 +371,46 @@ unsigned long arch_align_stack(unsigned long sp) return sp & ~0xf; } #endif + +unsigned long get_wchan(struct task_struct *p) +{ + unsigned long stack_page, sp, ip; + bool seen_sched = 0; + + if ((p == NULL) || (p == current) || (p->state == TASK_RUNNING)) + return 0; + + stack_page = (unsigned long) task_stack_page(p); + /* Bail if the process has no kernel stack for some reason */ + if (stack_page == 0) + return 0; + + sp = p->thread.switch_buf->JB_SP; + /* + * Bail if the stack pointer is below the bottom of the kernel + * stack for some reason + */ + if (sp < stack_page) + return 0; + + while (sp < stack_page + THREAD_SIZE) { + ip = *((unsigned long *) sp); + if (in_sched_functions(ip)) + /* Ignore everything until we're above the scheduler */ + seen_sched = 1; + else if (kernel_text_address(ip) && seen_sched) + return ip; + + sp += sizeof(unsigned long); + } + + return 0; +} + +int elf_core_copy_fpregs(struct task_struct *t, elf_fpregset_t *fpu) +{ + int cpu = current_thread_info()->cpu; + + return save_fp_registers(userspace_pid[cpu], (unsigned long *) fpu); +} + diff --git a/arch/um/kernel/ptrace.c b/arch/um/kernel/ptrace.c index 9a77fb3c269..694d551c889 100644 --- a/arch/um/kernel/ptrace.c +++ b/arch/um/kernel/ptrace.c @@ -1,34 +1,34 @@ -/* - * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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" +#include <linux/audit.h> +#include <linux/ptrace.h> +#include <linux/sched.h> +#include <linux/tracehook.h> +#include <asm/uaccess.h> +#include <skas_ptrace.h> + + + +void user_enable_single_step(struct task_struct *child) +{ + child->ptrace |= PT_DTRACE; + child->thread.singlestep_syscall = 0; + +#ifdef SUBARCH_SET_SINGLESTEPPING + SUBARCH_SET_SINGLESTEPPING(child, 1); #endif -#include "asm/ptrace.h" -#include "asm/uaccess.h" -#include "kern_util.h" -#include "skas_ptrace.h" -#include "sysdep/ptrace.h" +} -static inline void set_singlestepping(struct task_struct *child, int on) +void user_disable_single_step(struct task_struct *child) { - if (on) - child->ptrace |= PT_DTRACE; - else - child->ptrace &= ~PT_DTRACE; - child->thread.singlestep_syscall = 0; + child->ptrace &= ~PT_DTRACE; + child->thread.singlestep_syscall = 0; #ifdef SUBARCH_SET_SINGLESTEPPING - SUBARCH_SET_SINGLESTEPPING(child, on); + SUBARCH_SET_SINGLESTEPPING(child, 0); #endif } @@ -36,104 +36,35 @@ static inline void set_singlestepping(struct task_struct *child, int on) * Called by kernel/ptrace.c when detaching.. */ void ptrace_disable(struct task_struct *child) -{ - set_singlestepping(child,0); +{ + user_disable_single_step(child); } extern int peek_user(struct task_struct * child, long addr, long data); extern int poke_user(struct task_struct * child, long addr, long data); -long arch_ptrace(struct task_struct *child, long request, long addr, long data) +long arch_ptrace(struct task_struct *child, long request, + unsigned long addr, unsigned long data) { int i, ret; - unsigned long __user *p = (void __user *)(unsigned long)data; + unsigned long __user *p = (void __user *)data; + void __user *vp = p; 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, p); - break; - } - /* read the word at location addr in the USER area. */ - case PTRACE_PEEKUSR: - ret = peek_user(child, addr, 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 = poke_user(child, addr, data); - break; - - case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */ - case PTRACE_CONT: { /* restart after signal. */ - ret = -EIO; - if (!valid_signal(data)) - break; - - set_singlestepping(child, 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; + case PTRACE_PEEKUSR: + ret = peek_user(child, addr, data); 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; - - set_singlestepping(child, 0); - child->exit_code = SIGKILL; - wake_up_process(child); + /* write the word at location addr in the USER area */ + case PTRACE_POKEUSR: + ret = poke_user(child, addr, data); break; - } - case PTRACE_SINGLESTEP: { /* set the trap flag. */ + case PTRACE_SYSEMU: + case PTRACE_SYSEMU_SINGLESTEP: ret = -EIO; - if (!valid_signal(data)) - break; - clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE); - set_singlestepping(child, 1); - 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. */ @@ -165,45 +96,23 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) 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_GET_THREAD_AREA: - ret = ptrace_get_thread_area(child, addr, - (struct user_desc __user *) data); + ret = ptrace_get_thread_area(child, addr, vp); break; case PTRACE_SET_THREAD_AREA: - ret = ptrace_set_thread_area(child, addr, - (struct user_desc __user *) data); + ret = ptrace_set_thread_area(child, addr, vp); break; case PTRACE_FAULTINFO: { - /* Take the info from thread->arch->faultinfo, + /* + * Take the info from thread->arch->faultinfo, * but transfer max. sizeof(struct ptrace_faultinfo). * On i386, ptrace_faultinfo is smaller! */ ret = copy_to_user(p, &child->thread.arch.faultinfo, - sizeof(struct ptrace_faultinfo)); - if(ret) - break; + sizeof(struct ptrace_faultinfo)) ? + -EIO : 0; break; } @@ -211,45 +120,30 @@ long arch_ptrace(struct task_struct *child, long request, long addr, long data) case PTRACE_LDT: { struct ptrace_ldt ldt; - if(copy_from_user(&ldt, p, sizeof(ldt))){ + if (copy_from_user(&ldt, p, sizeof(ldt))) { ret = -EIO; break; } - /* This one is confusing, so just punt and return -EIO for + /* + * This one is confusing, so just punt and return -EIO for * now */ ret = -EIO; break; } #endif -#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); + if (ret == -EIO) + ret = subarch_ptrace(child, request, addr, data); break; } return ret; } -void send_sigtrap(struct task_struct *tsk, union uml_pt_regs *regs, +static void send_sigtrap(struct task_struct *tsk, struct uml_pt_regs *regs, int error_code) { struct siginfo info; @@ -261,54 +155,44 @@ void send_sigtrap(struct task_struct *tsk, union uml_pt_regs *regs, /* User-mode eip? */ info.si_addr = UPT_IS_USER(regs) ? (void __user *) UPT_IP(regs) : NULL; - /* Send us the fakey SIGTRAP */ + /* Send us the fake SIGTRAP */ force_sig_info(SIGTRAP, &info, tsk); } -/* XXX Check PT_DTRACE vs TIF_SINGLESTEP for singlestepping check and +/* + * 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) +void syscall_trace_enter(struct pt_regs *regs) { - int is_singlestep = (current->ptrace & PT_DTRACE) && entryexit; - int tracesysgood; - - if (unlikely(current->audit_context)) { - if (!entryexit) - audit_syscall_entry(HOST_AUDIT_ARCH, - UPT_SYSCALL_NR(regs), - UPT_SYSCALL_ARG1(regs), - UPT_SYSCALL_ARG2(regs), - UPT_SYSCALL_ARG3(regs), - UPT_SYSCALL_ARG4(regs)); - else audit_syscall_exit(AUDITSC_RESULT(UPT_SYSCALL_RET(regs)), - UPT_SYSCALL_RET(regs)); - } - - /* Fake a debug trap */ - if (is_singlestep) - send_sigtrap(current, regs, 0); + audit_syscall_entry(HOST_AUDIT_ARCH, + 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)); if (!test_thread_flag(TIF_SYSCALL_TRACE)) return; - if (!(current->ptrace & PT_PTRACED)) - return; + tracehook_report_syscall_entry(regs); +} - /* 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)); +void syscall_trace_leave(struct pt_regs *regs) +{ + int ptraced = current->ptrace; - if (entryexit) /* force do_signal() --> is_syscall() */ - set_thread_flag(TIF_SIGPENDING); + audit_syscall_exit(regs); - /* 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; - } + /* Fake a debug trap */ + if (ptraced & PT_DTRACE) + send_sigtrap(current, ®s->regs, 0); + + if (!test_thread_flag(TIF_SYSCALL_TRACE)) + return; + + tracehook_report_syscall_exit(regs, 0); + /* force do_signal() --> is_syscall() */ + if (ptraced & PT_PTRACED) + set_thread_flag(TIF_SIGPENDING); } diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c index f602623644a..ced8903921a 100644 --- a/arch/um/kernel/reboot.c +++ b/arch/um/kernel/reboot.c @@ -1,61 +1,61 @@ /* - * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/module.h" -#include "linux/sched.h" -#include "asm/smp.h" -#include "user_util.h" -#include "kern_util.h" -#include "kern.h" -#include "os.h" -#include "mode.h" -#include "choose-mode.h" +#include <linux/sched.h> +#include <linux/spinlock.h> +#include <linux/slab.h> +#include <linux/oom.h> +#include <kern_util.h> +#include <os.h> +#include <skas.h> void (*pm_power_off)(void); -#ifdef CONFIG_SMP -static void kill_idlers(int me) +static void kill_off_processes(void) { -#ifdef CONFIG_MODE_TT - struct task_struct *p; - int i; + if (proc_mm) + /* + * FIXME: need to loop over userspace_pids + */ + os_kill_ptraced_process(userspace_pid[0], 1); + else { + struct task_struct *p; + int pid; - for(i = 0; i < ARRAY_SIZE(idle_threads); 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 + read_lock(&tasklist_lock); + for_each_process(p) { + struct task_struct *t; -static void kill_off_processes(void) -{ - CHOOSE_MODE(kill_off_processes_tt(), kill_off_processes_skas()); -#ifdef CONFIG_SMP - kill_idlers(os_getpid()); -#endif + t = find_lock_task_mm(p); + if (!t) + continue; + pid = t->mm->context.id.u.pid; + task_unlock(t); + os_kill_ptraced_process(pid, 1); + } + read_unlock(&tasklist_lock); + } } void uml_cleanup(void) { - kmalloc_ok = 0; + kmalloc_ok = 0; do_uml_exitcalls(); kill_off_processes(); } void machine_restart(char * __unused) { - uml_cleanup(); - CHOOSE_MODE(reboot_tt(), reboot_skas()); + uml_cleanup(); + reboot_skas(); } void machine_power_off(void) { - uml_cleanup(); - CHOOSE_MODE(halt_tt(), halt_skas()); + uml_cleanup(); + halt_skas(); } void machine_halt(void) diff --git a/arch/um/kernel/resource.c b/arch/um/kernel/resource.c deleted file mode 100644 index 32188e12e8a..00000000000 --- a/arch/um/kernel/resource.c +++ /dev/null @@ -1,23 +0,0 @@ -/* - * 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.c b/arch/um/kernel/sigio.c index 2b0ab438301..b5e0cbb3438 100644 --- a/arch/um/kernel/sigio.c +++ b/arch/um/kernel/sigio.c @@ -1,18 +1,12 @@ /* - * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{linux.intel,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" -#include "os.h" +#include <linux/interrupt.h> +#include <irq_kern.h> +#include <os.h> +#include <sigio.h> /* Protected by sigio_lock() called from write_sigio_workaround */ static int sigio_irq_fd = -1; @@ -23,7 +17,7 @@ static irqreturn_t sigio_interrupt(int irq, void *data) os_read_file(sigio_irq_fd, &c, sizeof(c)); reactivate_fd(sigio_irq_fd, SIGIO_WRITE_IRQ); - return(IRQ_HANDLED); + return IRQ_HANDLED; } int write_sigio_irq(int fd) @@ -31,17 +25,17 @@ int write_sigio_irq(int fd) int err; err = um_request_irq(SIGIO_WRITE_IRQ, fd, IRQ_READ, sigio_interrupt, - IRQF_DISABLED|IRQF_SAMPLE_RANDOM, "write sigio", - NULL); - if(err){ - printk("write_sigio_irq : um_request_irq failed, err = %d\n", - err); - return(-1); + 0, "write sigio", NULL); + if (err) { + printk(KERN_ERR "write_sigio_irq : um_request_irq failed, " + "err = %d\n", err); + return -1; } sigio_irq_fd = fd; - return(0); + return 0; } +/* These are called from os-Linux/sigio.c to protect its pollfds arrays. */ static DEFINE_SPINLOCK(sigio_spinlock); void sigio_lock(void) diff --git a/arch/um/kernel/signal.c b/arch/um/kernel/signal.c index 2a32e5e8e9c..f57e02e7910 100644 --- a/arch/um/kernel/signal.c +++ b/arch/um/kernel/signal.c @@ -1,55 +1,38 @@ /* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#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 "kern.h" -#include "frame_kern.h" -#include "sigcontext.h" -#include "mode.h" +#include <linux/module.h> +#include <linux/ptrace.h> +#include <linux/sched.h> +#include <asm/siginfo.h> +#include <asm/signal.h> +#include <asm/unistd.h> +#include <frame_kern.h> +#include <kern_util.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) +static void handle_signal(struct pt_regs *regs, unsigned long signr, + struct k_sigaction *ka, struct siginfo *info) { + sigset_t *oldset = sigmask_to_save(); + int singlestep = 0; unsigned long sp; int err; - /* Always make any pending restarted system calls return -EINTR */ - current_thread_info()->restart_block.fn = do_no_restart_syscall; + if ((current->ptrace & PT_DTRACE) && (current->ptrace & PT_PTRACED)) + singlestep = 1; /* Did we come from a system call? */ - if(PT_REGS_SYSCALL_NR(regs) >= 0){ + if (PT_REGS_SYSCALL_NR(regs) >= 0) { /* If so, check system call restarting.. */ - switch(PT_REGS_SYSCALL_RET(regs)){ + switch (PT_REGS_SYSCALL_RET(regs)) { case -ERESTART_RESTARTBLOCK: case -ERESTARTNOHAND: PT_REGS_SYSCALL_RET(regs) = -EINTR; @@ -69,65 +52,38 @@ static int handle_signal(struct pt_regs *regs, unsigned long signr, } sp = PT_REGS_SP(regs); - if((ka->sa.sa_flags & SA_ONSTACK) && (sas_ss_flags(sp) == 0)) + 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)) + 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); + if (err) force_sigsegv(signr, current); - } else { - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked, ¤t->blocked, - &ka->sa.sa_mask); - if(!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked, signr); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - } - - return err; + else + signal_delivered(signr, info, ka, regs, singlestep); } static int kern_do_signal(struct pt_regs *regs) { struct k_sigaction ka_copy; - siginfo_t info; - sigset_t *oldset; + struct siginfo info; int sig, handled_sig = 0; - if (test_thread_flag(TIF_RESTORE_SIGMASK)) - oldset = ¤t->saved_sigmask; - else - oldset = ¤t->blocked; - - while((sig = get_signal_to_deliver(&info, &ka_copy, regs, NULL)) > 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)){ - /* a signal was successfully delivered; the saved - * sigmask will have been stored in the signal frame, - * and will be restored by sigreturn, so we can simply - * clear the TIF_RESTORE_SIGMASK flag */ - if (test_thread_flag(TIF_RESTORE_SIGMASK)) - clear_thread_flag(TIF_RESTORE_SIGMASK); - break; - } + handle_signal(regs, sig, &ka_copy, &info); } /* Did we come from a system call? */ - if(!handled_sig && (PT_REGS_SYSCALL_NR(regs) >= 0)){ + if (!handled_sig && (PT_REGS_SYSCALL_NR(regs) >= 0)) { /* Restart the system call - no handlers present */ - switch(PT_REGS_SYSCALL_RET(regs)){ + switch (PT_REGS_SYSCALL_RET(regs)) { case -ERESTARTNOHAND: case -ERESTARTSYS: case -ERESTARTNOINTR: @@ -138,53 +94,31 @@ static int kern_do_signal(struct pt_regs *regs) PT_REGS_ORIG_SYSCALL(regs) = __NR_restart_syscall; PT_REGS_RESTART_SYSCALL(regs); break; - } + } } - /* This closes a way to execute a system call on the host. If + /* + * 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) + if (current->ptrace & PT_DTRACE) current->thread.singlestep_syscall = is_syscall(PT_REGS_IP(¤t->thread.regs)); - /* if there's no signal to deliver, we just put the saved sigmask - * back */ - if (!handled_sig && test_thread_flag(TIF_RESTORE_SIGMASK)) { - clear_thread_flag(TIF_RESTORE_SIGMASK); - sigprocmask(SIG_SETMASK, ¤t->saved_sigmask, NULL); - } - return(handled_sig); + /* + * if there's no signal to deliver, we just put the saved sigmask + * back + */ + if (!handled_sig) + restore_saved_sigmask(); + return handled_sig; } int do_signal(void) { - return(kern_do_signal(¤t->thread.regs)); -} - -/* - * Atomically swap in the new signal mask, and wait for a signal. - */ -long sys_sigsuspend(int history0, int history1, old_sigset_t mask) -{ - mask &= _BLOCKABLE; - spin_lock_irq(¤t->sighand->siglock); - current->saved_sigmask = current->blocked; - siginitset(¤t->blocked, mask); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); - - current->state = TASK_INTERRUPTIBLE; - schedule(); - set_thread_flag(TIF_RESTORE_SIGMASK); - return -ERESTARTNOHAND; -} - -long sys_sigaltstack(const stack_t __user *uss, stack_t __user *uoss) -{ - return(do_sigaltstack(uss, uoss, PT_REGS_SP(¤t->thread.regs))); + return kern_do_signal(¤t->thread.regs); } diff --git a/arch/um/kernel/skas/Makefile b/arch/um/kernel/skas/Makefile index 3e3fa7e7e3c..0b76d8869c9 100644 --- a/arch/um/kernel/skas/Makefile +++ b/arch/um/kernel/skas/Makefile @@ -1,9 +1,9 @@ # -# Copyright (C) 2002 - 2004 Jeff Dike (jdike@addtoit.com) +# Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) # Licensed under the GPL # -obj-y := clone.o exec.o mem.o mmu.o process.o syscall.o tlb.o uaccess.o +obj-y := clone.o mmu.o process.o syscall.o uaccess.o # clone.o is in the stub, so it can't be built with profiling # GCC hardened also auto-enables -fpic, but we need %ebx so it can't work -> diff --git a/arch/um/kernel/skas/clone.c b/arch/um/kernel/skas/clone.c index 47b812b3bca..289771dadf8 100644 --- a/arch/um/kernel/skas/clone.c +++ b/arch/um/kernel/skas/clone.c @@ -1,17 +1,19 @@ -#include <sched.h> +/* + * Copyright (C) 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) + * Licensed under the GPL + */ + #include <signal.h> -#include <sys/mman.h> -#include <sys/time.h> +#include <sched.h> #include <asm/unistd.h> -#include <asm/page.h> -#include "ptrace_user.h" -#include "skas.h" -#include "stub-data.h" -#include "uml-config.h" -#include "sysdep/stub.h" -#include "kern_constants.h" - -/* This is in a separate file because it needs to be compiled with any +#include <sys/time.h> +#include <as-layout.h> +#include <ptrace_user.h> +#include <stub-data.h> +#include <sysdep/stub.h> + +/* + * This is in a separate file because it needs to be compiled with any * extraneous gcc flags (-pg, -fprofile-arcs, -ftest-coverage) disabled * * Use UM_KERN_PAGE_SIZE instead of PAGE_SIZE because that calls getpagesize @@ -21,31 +23,31 @@ void __attribute__ ((__section__ (".__syscall_stub"))) stub_clone_handler(void) { - struct stub_data *data = (struct stub_data *) UML_CONFIG_STUB_DATA; + struct stub_data *data = (struct stub_data *) STUB_DATA; long err; err = stub_syscall2(__NR_clone, CLONE_PARENT | CLONE_FILES | SIGCHLD, - UML_CONFIG_STUB_DATA + UM_KERN_PAGE_SIZE / 2 - - sizeof(void *)); - if(err != 0) + STUB_DATA + UM_KERN_PAGE_SIZE / 2 - sizeof(void *)); + if (err != 0) goto out; err = stub_syscall4(__NR_ptrace, PTRACE_TRACEME, 0, 0, 0); - if(err) + if (err) goto out; - err = stub_syscall3(__NR_setitimer, ITIMER_VIRTUAL, + err = stub_syscall3(__NR_setitimer, ITIMER_VIRTUAL, (long) &data->timer, 0); - if(err) + if (err) goto out; remap_stack(data->fd, data->offset); goto done; out: - /* save current result. - * Parent: pid; - * child: retcode of mmap already saved and it jumps around this + /* + * save current result. + * Parent: pid; + * child: retcode of mmap already saved and it jumps around this * assignment */ data->err = err; diff --git a/arch/um/kernel/skas/exec.c b/arch/um/kernel/skas/exec.c deleted file mode 100644 index 54b79595137..00000000000 --- a/arch/um/kernel/skas/exec.c +++ /dev/null @@ -1,30 +0,0 @@ -/* - * 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(¤t->mm->context.skas.id); -} - -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; -} diff --git a/arch/um/kernel/skas/mem.c b/arch/um/kernel/skas/mem.c deleted file mode 100644 index 0d2cce62113..00000000000 --- a/arch/um/kernel/skas/mem.c +++ /dev/null @@ -1,26 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include "linux/mm.h" -#include "asm/pgtable.h" -#include "mem_user.h" -#include "skas.h" - -unsigned long set_task_sizes_skas(unsigned long *task_size_out) -{ - /* Round up to the nearest 4M */ - unsigned long host_task_size = ROUND_4M((unsigned long) - &host_task_size); - -#ifdef CONFIG_HOST_TASK_SIZE - *host_size_out = ROUND_4M(CONFIG_HOST_TASK_SIZE); - *task_size_out = CONFIG_HOST_TASK_SIZE; -#else - if (!skas_needs_stub) - *task_size_out = host_task_size; - else *task_size_out = CONFIG_STUB_START & PGDIR_MASK; -#endif - return host_task_size; -} diff --git a/arch/um/kernel/skas/mmu.c b/arch/um/kernel/skas/mmu.c index 2c6d090a2e8..007d5503f49 100644 --- a/arch/um/kernel/skas/mmu.c +++ b/arch/um/kernel/skas/mmu.c @@ -1,22 +1,16 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/sched.h" -#include "linux/list.h" -#include "linux/spinlock.h" -#include "linux/slab.h" -#include "linux/errno.h" -#include "linux/mm.h" -#include "asm/current.h" -#include "asm/segment.h" -#include "asm/mmu.h" -#include "asm/pgalloc.h" -#include "asm/pgtable.h" -#include "asm/ldt.h" -#include "os.h" -#include "skas.h" +#include <linux/mm.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <asm/pgalloc.h> +#include <asm/pgtable.h> +#include <as-layout.h> +#include <os.h> +#include <skas.h> extern int __syscall_stub_start; @@ -37,93 +31,63 @@ static int init_stub_pte(struct mm_struct *mm, unsigned long proc, if (!pmd) goto out_pmd; - pte = pte_alloc_map(mm, pmd, proc); + pte = pte_alloc_map(mm, NULL, pmd, proc); if (!pte) goto out_pte; - /* There's an interaction between the skas0 stub pages, stack - * randomization, and the BUG at the end of exit_mmap. exit_mmap - * checks that the number of page tables freed is the same as had - * been allocated. If the stack is on the last page table page, - * then the stack pte page will be freed, and if not, it won't. To - * avoid having to know where the stack is, or if the process mapped - * something at the top of its address space for some other reason, - * we set TASK_SIZE to end at the start of the last page table. - * This keeps exit_mmap off the last page, but introduces a leak - * of that page. So, we hang onto it here and free it in - * destroy_context_skas. - */ - - mm->context.skas.last_page_table = pmd_page_vaddr(*pmd); -#ifdef CONFIG_3_LEVEL_PGTABLES - mm->context.skas.last_pmd = (unsigned long) __va(pud_val(*pud)); -#endif - *pte = mk_pte(virt_to_page(kernel), __pgprot(_PAGE_PRESENT)); *pte = pte_mkread(*pte); - return(0); + return 0; - out_pmd: - pud_free(pud); out_pte: - pmd_free(pmd); + pmd_free(mm, pmd); + out_pmd: + pud_free(mm, pud); out: - return(-ENOMEM); + return -ENOMEM; } -int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) +int init_new_context(struct task_struct *task, struct mm_struct *mm) { - struct mmu_context_skas *from_mm = NULL; - struct mmu_context_skas *to_mm = &mm->context.skas; + struct mm_context *from_mm = NULL; + struct mm_context *to_mm = &mm->context; unsigned long stack = 0; int ret = -ENOMEM; - if(skas_needs_stub){ + if (skas_needs_stub) { stack = get_zeroed_page(GFP_KERNEL); - if(stack == 0) + if (stack == 0) goto out; - - /* This zeros the entry that pgd_alloc didn't, needed since - * we are about to reinitialize it, and want mm.nr_ptes to - * be accurate. - */ - mm->pgd[USER_PTRS_PER_PGD] = __pgd(0); - - ret = init_stub_pte(mm, CONFIG_STUB_CODE, - (unsigned long) &__syscall_stub_start); - if(ret) - goto out_free; - - ret = init_stub_pte(mm, CONFIG_STUB_DATA, stack); - if(ret) - goto out_free; - - mm->nr_ptes--; } to_mm->id.stack = stack; - if(current->mm != NULL && current->mm != &init_mm) - from_mm = ¤t->mm->context.skas; + if (current->mm != NULL && current->mm != &init_mm) + from_mm = ¤t->mm->context; - if(proc_mm){ + if (proc_mm) { ret = new_mm(stack); - if(ret < 0){ - printk("init_new_context_skas - new_mm failed, " - "errno = %d\n", ret); + if (ret < 0) { + printk(KERN_ERR "init_new_context_skas - " + "new_mm failed, errno = %d\n", ret); goto out_free; } to_mm->id.u.mm_fd = ret; } else { - if(from_mm) + if (from_mm) to_mm->id.u.pid = copy_context_skas0(stack, from_mm->id.u.pid); else to_mm->id.u.pid = start_userspace(stack); + + if (to_mm->id.u.pid < 0) { + ret = to_mm->id.u.pid; + goto out_free; + } } ret = init_new_ldt(to_mm, from_mm); - if(ret < 0){ - printk("init_new_context_skas - init_ldt" + if (ret < 0) { + printk(KERN_ERR "init_new_context_skas - init_ldt" " failed, errno = %d\n", ret); goto out_free; } @@ -131,28 +95,84 @@ int init_new_context_skas(struct task_struct *task, struct mm_struct *mm) return 0; out_free: - if(to_mm->id.stack != 0) + if (to_mm->id.stack != 0) free_page(to_mm->id.stack); out: return ret; } -void destroy_context_skas(struct mm_struct *mm) +void uml_setup_stubs(struct mm_struct *mm) +{ + int err, ret; + + if (!skas_needs_stub) + return; + + ret = init_stub_pte(mm, STUB_CODE, + (unsigned long) &__syscall_stub_start); + if (ret) + goto out; + + ret = init_stub_pte(mm, STUB_DATA, mm->context.id.stack); + if (ret) + goto out; + + mm->context.stub_pages[0] = virt_to_page(&__syscall_stub_start); + mm->context.stub_pages[1] = virt_to_page(mm->context.id.stack); + + /* dup_mmap already holds mmap_sem */ + err = install_special_mapping(mm, STUB_START, STUB_END - STUB_START, + VM_READ | VM_MAYREAD | VM_EXEC | + VM_MAYEXEC | VM_DONTCOPY | VM_PFNMAP, + mm->context.stub_pages); + if (err) { + printk(KERN_ERR "install_special_mapping returned %d\n", err); + goto out; + } + return; + +out: + force_sigsegv(SIGSEGV, current); +} + +void arch_exit_mmap(struct mm_struct *mm) +{ + pte_t *pte; + + pte = virt_to_pte(mm, STUB_CODE); + if (pte != NULL) + pte_clear(mm, STUB_CODE, pte); + + pte = virt_to_pte(mm, STUB_DATA); + if (pte == NULL) + return; + + pte_clear(mm, STUB_DATA, pte); +} + +void destroy_context(struct mm_struct *mm) { - struct mmu_context_skas *mmu = &mm->context.skas; + struct mm_context *mmu = &mm->context; - if(proc_mm) + if (proc_mm) os_close_file(mmu->id.u.mm_fd); - else + else { + /* + * If init_new_context wasn't called, this will be + * zero, resulting in a kill(0), which will result in the + * whole UML suddenly dying. Also, cover negative and + * 1 cases, since they shouldn't happen either. + */ + if (mmu->id.u.pid < 2) { + printk(KERN_ERR "corrupt mm_context - pid = %d\n", + mmu->id.u.pid); + return; + } os_kill_ptraced_process(mmu->id.u.pid, 1); + } - if(!proc_mm || !ptrace_faultinfo){ + if (skas_needs_stub) free_page(mmu->id.stack); - pte_lock_deinit(virt_to_page(mmu->last_page_table)); - pte_free_kernel((pte_t *) mmu->last_page_table); - dec_zone_page_state(virt_to_page(mmu->last_page_table), NR_PAGETABLE); -#ifdef CONFIG_3_LEVEL_PGTABLES - pmd_free((pmd_t *) mmu->last_pmd); -#endif - } + + free_ldt(mmu); } diff --git a/arch/um/kernel/skas/process.c b/arch/um/kernel/skas/process.c index ae4fa71d3b8..4da11b3c8dd 100644 --- a/arch/um/kernel/skas/process.c +++ b/arch/um/kernel/skas/process.c @@ -1,151 +1,37 @@ /* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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 "skas.h" -#include "os.h" -#include "user_util.h" -#include "tlb.h" -#include "kern.h" -#include "mode.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); - - switch_threads(&from->thread.mode.skas.switch_buf, - &to->thread.mode.skas.switch_buf); - - arch_switch_to_skas(current->thread.prev_sched, current); - - if(current->pid == 0) - switch_timers(1); -} - -extern void schedule_tail(struct task_struct *prev); - -/* This is called magically, by its address being stuffed in a jmp_buf - * and being longjmp-d to. - */ -void new_thread_handler(void) -{ - int (*fn)(void *), n; - void *arg; - - if(current->thread.prev_sched != NULL) - schedule_tail(current->thread.prev_sched); - current->thread.prev_sched = NULL; - - fn = current->thread.request.u.thread.proc; - arg = current->thread.request.u.thread.arg; - - /* 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){ - /* Handle any immediate reschedules or signals */ - interrupt_end(); - userspace(¤t->thread.regs.regs); - } - else do_exit(0); -} - -void release_thread_skas(struct task_struct *task) -{ -} - -/* Called magically, see new_thread_handler above */ -void fork_handler(void) -{ - force_flush_all(); - if(current->thread.prev_sched == NULL) - panic("blech"); - - schedule_tail(current->thread.prev_sched); - - /* XXX: if interrupt_end() calls schedule, this call to - * arch_switch_to_skas isn't needed. We could want to apply this to - * improve performance. -bb */ - arch_switch_to_skas(current->thread.prev_sched, current); - - current->thread.prev_sched = NULL; - -/* Handle any immediate reschedules or signals */ - interrupt_end(); - - 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)(void); - - if(current->thread.forking){ - memcpy(&p->thread.regs.regs.skas, ®s->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; - - arch_copy_thread(¤t->thread.arch, &p->thread.arch); - } - else { - init_thread_registers(&p->thread.regs.regs); - p->thread.request.u.thread = current->thread.request.u.thread; - handler = new_thread_handler; - } - - new_thread(task_stack_page(p), &p->thread.mode.skas.switch_buf, - handler); - return(0); -} +#include <linux/init.h> +#include <linux/sched.h> +#include <as-layout.h> +#include <kern.h> +#include <os.h> +#include <skas.h> int new_mm(unsigned long stack) { - int fd; + int fd, err; fd = os_open_file("/proc/mm", of_cloexec(of_write(OPENFLAGS())), 0); - if(fd < 0) - return(fd); - - if(skas_needs_stub) - map_stub_pages(fd, CONFIG_STUB_CODE, CONFIG_STUB_DATA, stack); - - return(fd); -} + if (fd < 0) + return fd; + + if (skas_needs_stub) { + err = map_stub_pages(fd, STUB_CODE, STUB_DATA, stack); + if (err) { + os_close_file(fd); + return err; + } + } -void init_idle_skas(void) -{ - cpu_tasks[current_thread->cpu].pid = os_getpid(); - default_idle(); + return fd; } extern void start_kernel(void); -static int start_kernel_proc(void *unused) +static int __init start_kernel_proc(void *unused) { int pid; @@ -155,63 +41,41 @@ static int start_kernel_proc(void *unused) cpu_tasks[0].pid = pid; cpu_tasks[0].task = current; #ifdef CONFIG_SMP - cpu_online_map = cpumask_of_cpu(0); + init_cpu_online(get_cpu_mask(0)); #endif start_kernel(); - return(0); + return 0; } extern int userspace_pid[]; -int start_uml_skas(void) +extern char cpu0_irqstack[]; + +int __init start_uml(void) { - if(proc_mm) + stack_protections((unsigned long) &cpu0_irqstack); + set_sigstack(cpu0_irqstack, THREAD_SIZE); + if (proc_mm) { userspace_pid[0] = start_userspace(0); + if (userspace_pid[0] < 0) { + printf("start_uml - start_userspace returned %d\n", + userspace_pid[0]); + exit(1); + } + } init_new_thread_signals(); init_task.thread.request.u.thread.proc = start_kernel_proc; init_task.thread.request.u.thread.arg = NULL; - return(start_idle_thread(task_stack_page(&init_task), - &init_task.thread.mode.skas.switch_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]); -} - -void kill_off_processes_skas(void) -{ - if(proc_mm) -#warning need to loop over userspace_pids in kill_off_processes_skas - os_kill_ptraced_process(userspace_pid[0], 1); - else { - struct task_struct *p; - int pid, me; - - me = os_getpid(); - for_each_process(p){ - if(p->mm == NULL) - continue; - - pid = p->mm->context.skas.id.u.pid; - os_kill_ptraced_process(pid, 1); - } - } + return start_idle_thread(task_stack_page(&init_task), + &init_task.thread.switch_buf); } unsigned long current_stub_stack(void) { - if(current->mm == NULL) - return(0); + if (current->mm == NULL) + return 0; - return(current->mm->context.skas.id.stack); + return current->mm->context.id.stack; } diff --git a/arch/um/kernel/skas/syscall.c b/arch/um/kernel/skas/syscall.c index 0ae4eea21be..c0681e09743 100644 --- a/arch/um/kernel/skas/syscall.c +++ b/arch/um/kernel/skas/syscall.c @@ -1,30 +1,27 @@ /* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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" -#include "syscall.h" +#include <linux/kernel.h> +#include <linux/ptrace.h> +#include <kern_util.h> +#include <sysdep/ptrace.h> +#include <sysdep/syscalls.h> -void handle_syscall(union uml_pt_regs *r) +extern int syscall_table_size; +#define NR_SYSCALLS (syscall_table_size / sizeof(void *)) + +void handle_syscall(struct uml_pt_regs *r) { struct pt_regs *regs = container_of(r, struct pt_regs, regs); long result; int syscall; - syscall_trace(r, 0); - - current->thread.nsyscalls++; - nsyscalls++; + syscall_trace_enter(regs); - /* This should go in the declaration of syscall, but when I do that, + /* + * This should go in the declaration of syscall, but when I do that, * strace -f -c bash -c 'ls ; ls' breaks, sometimes not tracing * children at all, sometimes hanging when bash doesn't see the first * ls exit. @@ -33,11 +30,11 @@ void handle_syscall(union uml_pt_regs *r) * in case it's a compiler bug. */ syscall = UPT_SYSCALL_NR(r); - if((syscall >= NR_syscalls) || (syscall < 0)) + if ((syscall >= NR_SYSCALLS) || (syscall < 0)) result = -ENOSYS; else result = EXECUTE_SYSCALL(syscall, regs); - REGS_SET_SYSCALL_RETURN(r->skas.regs, result); + PT_REGS_SET_SYSCALL_RETURN(regs, result); - syscall_trace(r, 1); + syscall_trace_leave(regs); } diff --git a/arch/um/kernel/skas/tlb.c b/arch/um/kernel/skas/tlb.c deleted file mode 100644 index 27eb29ce666..00000000000 --- a/arch/um/kernel/skas/tlb.c +++ /dev/null @@ -1,97 +0,0 @@ -/* - * 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 int do_ops(union mm_context *mmu, struct host_vm_op *ops, int last, - int finished, void **flush) -{ - struct host_vm_op *op; - int i, ret = 0; - - for(i = 0; i <= last && !ret; i++){ - op = &ops[i]; - switch(op->type){ - case MMAP: - ret = map(&mmu->skas.id, 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, finished, flush); - break; - case MUNMAP: - ret = unmap(&mmu->skas.id, - (void *) op->u.munmap.addr, - op->u.munmap.len, finished, flush); - break; - case MPROTECT: - ret = protect(&mmu->skas.id, op->u.mprotect.addr, - op->u.mprotect.len, op->u.mprotect.r, - op->u.mprotect.w, op->u.mprotect.x, - finished, flush); - break; - default: - printk("Unknown op type %d in do_ops\n", op->type); - break; - } - } - - return ret; -} - -extern int proc_mm; - -static void fix_range(struct mm_struct *mm, unsigned long start_addr, - unsigned long end_addr, int force) -{ - if(!proc_mm && (end_addr > CONFIG_STUB_START)) - end_addr = CONFIG_STUB_START; - - fix_range_common(mm, start_addr, end_addr, force, 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) -{ - unsigned long end; - - /* Don't bother flushing if this address space is about to be - * destroyed. - */ - if(atomic_read(&mm->mm_users) == 0) - return; - - end = proc_mm ? task_size : CONFIG_STUB_START; - fix_range(mm, 0, end, 0); -} - -void force_flush_all_skas(void) -{ - unsigned long end = proc_mm ? task_size : CONFIG_STUB_START; - fix_range(current->mm, 0, end, 1); -} diff --git a/arch/um/kernel/skas/uaccess.c b/arch/um/kernel/skas/uaccess.c index 8912cec0fe4..4ffb644d6c0 100644 --- a/arch/um/kernel/skas/uaccess.c +++ b/arch/um/kernel/skas/uaccess.c @@ -1,133 +1,133 @@ /* - * Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2002 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "linux/compiler.h" -#include "linux/stddef.h" -#include "linux/kernel.h" -#include "linux/string.h" -#include "linux/fs.h" -#include "linux/hardirq.h" -#include "linux/highmem.h" -#include "asm/page.h" -#include "asm/pgtable.h" -#include "asm/uaccess.h" -#include "kern_util.h" -#include "os.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) +#include <linux/err.h> +#include <linux/highmem.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <asm/current.h> +#include <asm/page.h> +#include <asm/pgtable.h> +#include <kern_util.h> +#include <os.h> + +pte_t *virt_to_pte(struct mm_struct *mm, unsigned long addr) { - pte_t pte; - int err; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; - void *phys = um_virt_to_phys(current, virt, &pte); - int dummy_code; + if (mm == NULL) + return NULL; - if(IS_ERR(phys) || (is_write && !pte_write(pte))){ + pgd = pgd_offset(mm, addr); + if (!pgd_present(*pgd)) + return NULL; + + pud = pud_offset(pgd, addr); + if (!pud_present(*pud)) + return NULL; + + pmd = pmd_offset(pud, addr); + if (!pmd_present(*pmd)) + return NULL; + + return pte_offset_kernel(pmd, addr); +} + +static pte_t *maybe_map(unsigned long virt, int is_write) +{ + pte_t *pte = virt_to_pte(current->mm, virt); + int err, dummy_code; + + if ((pte == NULL) || !pte_present(*pte) || + (is_write && !pte_write(*pte))) { err = handle_page_fault(virt, 0, is_write, 1, &dummy_code); - if(err) - return(-1UL); - phys = um_virt_to_phys(current, virt, NULL); + if (err) + return NULL; + pte = virt_to_pte(current->mm, virt); } - if(IS_ERR(phys)) - phys = (void *) -1; + if (!pte_present(*pte)) + pte = NULL; - return((unsigned long) phys); + return pte; } static int do_op_one_page(unsigned long addr, int len, int is_write, int (*op)(unsigned long addr, int len, void *arg), void *arg) { + jmp_buf buf; struct page *page; - int n; + pte_t *pte; + int n, faulted; + + pte = maybe_map(addr, is_write); + if (pte == NULL) + return -1; - addr = maybe_map(addr, is_write); - if(addr == -1UL) - return(-1); + page = pte_page(*pte); + addr = (unsigned long) kmap_atomic(page) + + (addr & ~PAGE_MASK); - page = phys_to_page(addr); - addr = (unsigned long) kmap_atomic(page, KM_UML_USERCOPY) + (addr & ~PAGE_MASK); + current->thread.fault_catcher = &buf; - n = (*op)(addr, len, arg); + faulted = UML_SETJMP(&buf); + if (faulted == 0) + n = (*op)(addr, len, arg); + else + n = -1; + + current->thread.fault_catcher = NULL; - kunmap_atomic(page, KM_UML_USERCOPY); + kunmap_atomic((void *)addr); - return(n); + return n; } -static void do_buffer_op(void *jmpbuf, void *arg_ptr) +static int buffer_op(unsigned long addr, int len, int is_write, + int (*op)(unsigned long, int, void *), void *arg) { - va_list args; - unsigned long addr; - int len, is_write, size, remain, n; - int (*op)(unsigned long, int, void *); - void *arg; - int *res; - - 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); + int size, remain, n; + size = min(PAGE_ALIGN(addr) - addr, (unsigned long) len); remain = len; - current->thread.fault_catcher = jmpbuf; n = do_op_one_page(addr, size, is_write, op, arg); - if(n != 0){ - *res = (n < 0 ? remain : 0); + if (n != 0) { + remain = (n < 0 ? remain : 0); goto out; } addr += size; remain -= size; - if(remain == 0){ - *res = 0; + if (remain == 0) goto out; - } - while(addr < ((addr + remain) & PAGE_MASK)){ + while (addr < ((addr + remain) & PAGE_MASK)) { n = do_op_one_page(addr, PAGE_SIZE, is_write, op, arg); - if(n != 0){ - *res = (n < 0 ? remain : 0); + if (n != 0) { + remain = (n < 0 ? remain : 0); goto out; } addr += PAGE_SIZE; remain -= PAGE_SIZE; } - if(remain == 0){ - *res = 0; + if (remain == 0) goto out; - } n = do_op_one_page(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); + if (n != 0) { + remain = (n < 0 ? remain : 0); + goto out; + } - return(addr + len - (unsigned long) current->thread.fault_addr); + return 0; + out: + return remain; } static int copy_chunk_from_user(unsigned long from, int len, void *arg) @@ -136,20 +136,21 @@ static int copy_chunk_from_user(unsigned long from, int len, void *arg) memcpy((void *) to, (void *) from, len); *to_ptr += len; - return(0); + return 0; } -int copy_from_user_skas(void *to, const void __user *from, int n) +int copy_from_user(void *to, const void __user *from, int n) { - if(segment_eq(get_fs(), KERNEL_DS)){ + if (segment_eq(get_fs(), KERNEL_DS)) { memcpy(to, (__force void*)from, n); - return(0); + return 0; } - return(access_ok(VERIFY_READ, from, n) ? + return access_ok(VERIFY_READ, from, n) ? buffer_op((unsigned long) from, n, 0, copy_chunk_from_user, &to): - n); + n; } +EXPORT_SYMBOL(copy_from_user); static int copy_chunk_to_user(unsigned long to, int len, void *arg) { @@ -157,20 +158,21 @@ static int copy_chunk_to_user(unsigned long to, int len, void *arg) memcpy((void *) to, (void *) from, len); *from_ptr += len; - return(0); + return 0; } -int copy_to_user_skas(void __user *to, const void *from, int n) +int copy_to_user(void __user *to, const void *from, int n) { - if(segment_eq(get_fs(), KERNEL_DS)){ - memcpy((__force void*)to, from, n); - return(0); + if (segment_eq(get_fs(), KERNEL_DS)) { + memcpy((__force void *) to, from, n); + return 0; } - return(access_ok(VERIFY_WRITE, to, n) ? + return access_ok(VERIFY_WRITE, to, n) ? buffer_op((unsigned long) to, n, 1, copy_chunk_to_user, &from) : - n); + n; } +EXPORT_SYMBOL(copy_to_user); static int strncpy_chunk_from_user(unsigned long from, int len, void *arg) { @@ -181,52 +183,54 @@ static int strncpy_chunk_from_user(unsigned long from, int len, void *arg) n = strnlen(to, len); *to_ptr += n; - if(n < len) - return(1); - return(0); + if (n < len) + return 1; + return 0; } -int strncpy_from_user_skas(char *dst, const char __user *src, int count) +int strncpy_from_user(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 (segment_eq(get_fs(), KERNEL_DS)) { + strncpy(dst, (__force void *) src, count); + return strnlen(dst, count); } - if(!access_ok(VERIFY_READ, src, 1)) - return(-EFAULT); + if (!access_ok(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)); + if (n != 0) + return -EFAULT; + return strnlen(dst, count); } +EXPORT_SYMBOL(strncpy_from_user); static int clear_chunk(unsigned long addr, int len, void *unused) { memset((void *) addr, 0, len); - return(0); + return 0; } -int __clear_user_skas(void __user *mem, int len) +int __clear_user(void __user *mem, int len) { - return(buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL)); + return buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL); } -int clear_user_skas(void __user *mem, int len) +int clear_user(void __user *mem, int len) { - if(segment_eq(get_fs(), KERNEL_DS)){ + if (segment_eq(get_fs(), KERNEL_DS)) { memset((__force void*)mem, 0, len); - return(0); + return 0; } - return(access_ok(VERIFY_WRITE, mem, len) ? - buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL) : len); + return access_ok(VERIFY_WRITE, mem, len) ? + buffer_op((unsigned long) mem, len, 1, clear_chunk, NULL) : len; } +EXPORT_SYMBOL(clear_user); static int strnlen_chunk(unsigned long str, int len, void *arg) { @@ -235,31 +239,21 @@ static int strnlen_chunk(unsigned long str, int len, void *arg) n = strnlen((void *) str, len); *len_ptr += n; - if(n < len) - return(1); - return(0); + if (n < len) + return 1; + return 0; } -int strnlen_user_skas(const void __user *str, int len) +int strnlen_user(const void __user *str, int len) { int count = 0, n; - if(segment_eq(get_fs(), KERNEL_DS)) - return(strnlen((__force char*)str, len) + 1); + 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); + if (n == 0) + return count + 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: - */ +EXPORT_SYMBOL(strnlen_user); diff --git a/arch/um/kernel/smp.c b/arch/um/kernel/smp.c index 759b0705316..5c8c3ea7db7 100644 --- a/arch/um/kernel/smp.c +++ b/arch/um/kernel/smp.c @@ -1,38 +1,26 @@ /* - * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#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); +#include <linux/percpu.h> +#include <asm/pgalloc.h> +#include <asm/tlb.h> #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); +#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 <kern.h> +#include <irq_user.h> +#include <os.h> /* Per CPU bogomips and other parameters * The only piece used here is the ipi pipe, which is set before SMP is @@ -57,12 +45,12 @@ 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) + 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"); + printk(KERN_CONT "done\n"); } static cpumask_t smp_commenced_mask = CPU_MASK_NONE; @@ -73,35 +61,34 @@ static int idle_proc(void *cpup) int cpu = (int) cpup, err; err = os_pipe(cpu_data[cpu].ipi_pipe, 1, 1); - if(err < 0) + if (err < 0) panic("CPU#%d failed to create IPI pipe, err = %d", cpu, -err); - os_set_fd_async(cpu_data[cpu].ipi_pipe[0], - current->thread.mode.tt.extern_pid); + os_set_fd_async(cpu_data[cpu].ipi_pipe[0]); wmb(); if (cpu_test_and_set(cpu, cpu_callin_map)) { - printk("huh, CPU#%d already present??\n", cpu); + printk(KERN_ERR "huh, CPU#%d already present??\n", cpu); BUG(); } while (!cpu_isset(cpu, smp_commenced_mask)) cpu_relax(); - cpu_set(cpu, cpu_online_map); + notify_cpu_starting(cpu); + set_cpu_online(cpu, true); default_idle(); - return(0); + 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; + 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)) + if (IS_ERR(new_task)) panic("copy_process failed in idle_thread, error = %ld", PTR_ERR(new_task)); @@ -109,10 +96,8 @@ static struct task_struct *idle_thread(int cpu) { .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); + panic("skas mode doesn't support SMP"); + return new_task; } void smp_prepare_cpus(unsigned int maxcpus) @@ -123,21 +108,19 @@ void smp_prepare_cpus(unsigned int maxcpus) int i; for (i = 0; i < ncpus; ++i) - cpu_set(i, cpu_possible_map); + set_cpu_possible(i, true); - cpu_clear(me, cpu_online_map); - cpu_set(me, cpu_online_map); + set_cpu_online(me, true); cpu_set(me, cpu_callin_map); err = os_pipe(cpu_data[me].ipi_pipe, 1, 1); - if(err < 0) + if (err < 0) panic("CPU#0 failed to create IPI pipe, errno = %d", -err); - os_set_fd_async(cpu_data[me].ipi_pipe[0], - current->thread.mode.tt.extern_pid); + os_set_fd_async(cpu_data[me].ipi_pipe[0]); - for(cpu = 1; cpu < ncpus; cpu++){ - printk("Booting processor %d...\n", cpu); + for (cpu = 1; cpu < ncpus; cpu++) { + printk(KERN_INFO "Booting processor %d...\n", cpu); idle = idle_thread(cpu); @@ -147,29 +130,28 @@ void smp_prepare_cpus(unsigned int maxcpus) while (waittime-- && !cpu_isset(cpu, cpu_callin_map)) cpu_relax(); - if (cpu_isset(cpu, cpu_callin_map)) - printk("done\n"); - else printk("failed\n"); + printk(KERN_INFO "%s\n", + cpu_isset(cpu, cpu_calling_map) ? "done" : "failed"); } } void smp_prepare_boot_cpu(void) { - cpu_set(smp_processor_id(), cpu_online_map); + set_cpu_online(smp_processor_id(), true); } -int __cpu_up(unsigned int cpu) +int __cpu_up(unsigned int cpu, struct task_struct *tidle) { cpu_set(cpu, smp_commenced_mask); - while (!cpu_isset(cpu, cpu_online_map)) + while (!cpu_online(cpu)) mb(); - return(0); + return 0; } int setup_profiling_timer(unsigned int multiplier) { printk(KERN_INFO "setup_profiling_timer\n"); - return(0); + return 0; } void smp_call_function_slave(int cpu); @@ -187,17 +169,18 @@ void IPI_handler(int cpu) break; case 'R': - set_tsk_need_resched(current); + scheduler_ipi(); break; case 'S': - printk("CPU#%d stopping\n", cpu); - while(1) + printk(KERN_INFO "CPU#%d stopping\n", cpu); + while (1) pause(); break; default: - printk("CPU#%d received unknown IPI [%c]!\n", cpu, c); + printk(KERN_ERR "CPU#%d received unknown IPI [%c]!\n", + cpu, c); break; } } @@ -205,7 +188,7 @@ void IPI_handler(int cpu) int hard_smp_processor_id(void) { - return(pid_to_processor_id(os_getpid())); + return pid_to_processor_id(os_getpid()); } static DEFINE_SPINLOCK(call_lock); @@ -221,8 +204,7 @@ void smp_call_function_slave(int cpu) atomic_inc(&scf_finished); } -int smp_call_function(void (*_func)(void *info), void *_info, int nonatomic, - int wait) +int smp_call_function(void (*_func)(void *info), void *_info, int wait) { int cpus = num_online_cpus() - 1; int i; @@ -254,14 +236,3 @@ int smp_call_function(void (*_func)(void *info), void *_info, int nonatomic, } #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/syscall.c b/arch/um/kernel/syscall.c index f5ed8624648..c1d0ae069b5 100644 --- a/arch/um/kernel/syscall.c +++ b/arch/um/kernel/syscall.c @@ -1,78 +1,18 @@ /* - * Copyright (C) 2000 - 2003 Jeff Dike (jdike@addtoit.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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 "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, UPT_SP(¤t->thread.regs.regs), - ¤t->thread.regs, 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, - UPT_SP(¤t->thread.regs.regs), - ¤t->thread.regs, 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; -} +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/mm.h> +#include <linux/sched.h> +#include <linux/utsname.h> +#include <linux/syscalls.h> +#include <asm/current.h> +#include <asm/mman.h> +#include <asm/uaccess.h> +#include <asm/unistd.h> long old_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, @@ -82,98 +22,7 @@ long old_mmap(unsigned long addr, unsigned long len, if (offset & ~PAGE_MASK) goto out; - err = sys_mmap2(addr, len, prot, flags, fd, offset >> PAGE_SHIFT); + err = sys_mmap_pgoff(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 __user * name) -{ - long err; - if (!name) - return -EFAULT; - down_read(&uts_sem); - err = copy_to_user(name, utsname(), sizeof (*name)); - up_read(&uts_sem); - return err?-EFAULT:0; -} - -long sys_olduname(struct oldold_utsname __user * 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, &utsname()->sysname, - __OLD_UTS_LEN); - error |= __put_user(0, name->sysname + __OLD_UTS_LEN); - error |= __copy_to_user(&name->nodename, &utsname()->nodename, - __OLD_UTS_LEN); - error |= __put_user(0, name->nodename + __OLD_UTS_LEN); - error |= __copy_to_user(&name->release, &utsname()->release, - __OLD_UTS_LEN); - error |= __put_user(0, name->release + __OLD_UTS_LEN); - error |= __copy_to_user(&name->version, &utsname()->version, - __OLD_UTS_LEN); - error |= __put_user(0, name->version + __OLD_UTS_LEN); - error |= __copy_to_user(&name->machine, &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); -} - -int kernel_execve(const char *filename, char *const argv[], char *const envp[]) -{ - mm_segment_t fs; - int ret; - - fs = get_fs(); - set_fs(KERNEL_DS); - ret = um_execve(filename, argv, envp); - set_fs(fs); - - return ret; -} diff --git a/arch/um/kernel/sysrq.c b/arch/um/kernel/sysrq.c index 239c98054de..799d7e413bf 100644 --- a/arch/um/kernel/sysrq.c +++ b/arch/um/kernel/sysrq.c @@ -1,81 +1,98 @@ -/* - * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL +/* + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) + * Copyright (C) 2013 Richard Weinberger <richrd@nod.at> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. */ -#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" +#include <linux/kallsyms.h> +#include <linux/kernel.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <asm/sysrq.h> +#include <os.h> -/* Catch non-i386 SUBARCH's. */ -#if !defined(CONFIG_UML_X86) || defined(CONFIG_64BIT) -void show_trace(struct task_struct *task, unsigned long * stack) -{ - unsigned long addr; +struct stack_frame { + struct stack_frame *next_frame; + unsigned long return_address; +}; - if (!stack) { - stack = (unsigned long*) &stack; - WARN_ON(1); - } +static void do_stack_trace(unsigned long *sp, unsigned long bp) +{ + int reliable; + unsigned long addr; + struct stack_frame *frame = (struct stack_frame *)bp; - printk("Call Trace: \n"); - while (((long) stack & (THREAD_SIZE-1)) != 0) { - addr = *stack; + printk(KERN_INFO "Call Trace:\n"); + while (((long) sp & (THREAD_SIZE-1)) != 0) { + addr = *sp; if (__kernel_text_address(addr)) { - printk("%08lx: [<%08lx>]", (unsigned long) stack, addr); - print_symbol(" %s", addr); - printk("\n"); - } - stack++; - } - printk("\n"); + reliable = 0; + if ((unsigned long) sp == bp + sizeof(long)) { + frame = frame ? frame->next_frame : NULL; + bp = (unsigned long)frame; + reliable = 1; + } + + printk(KERN_INFO " [<%08lx>]", addr); + printk(KERN_CONT " %s", reliable ? "" : "? "); + print_symbol(KERN_CONT "%s", addr); + printk(KERN_CONT "\n"); + } + sp++; + } + printk(KERN_INFO "\n"); } -#endif -/* - * stack dumps generator - this is used by arch-independent code. - * And this is identical to i386 currently. - */ -void dump_stack(void) +static unsigned long get_frame_pointer(struct task_struct *task, + struct pt_regs *segv_regs) { - unsigned long stack; - - show_trace(current, &stack); + if (!task || task == current) + return segv_regs ? PT_REGS_BP(segv_regs) : current_bp(); + else + return KSTK_EBP(task); } -EXPORT_SYMBOL(dump_stack); -/*Stolen from arch/i386/kernel/traps.c */ -static int kstack_depth_to_print = 24; +static unsigned long *get_stack_pointer(struct task_struct *task, + struct pt_regs *segv_regs) +{ + if (!task || task == current) + return segv_regs ? (unsigned long *)PT_REGS_SP(segv_regs) : current_sp(); + else + return (unsigned long *)KSTK_ESP(task); +} -/* 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) +void show_stack(struct task_struct *task, unsigned long *stack) { - unsigned long *stack; + unsigned long *sp = stack, bp = 0; + struct pt_regs *segv_regs = current->thread.segv_regs; int i; - if (esp == NULL) { - if (task != current && task != NULL) { - esp = (unsigned long *) KSTK_ESP(task); - } else { - esp = (unsigned long *) &esp; - } + if (!segv_regs && os_is_signal_stack()) { + printk(KERN_ERR "Received SIGSEGV in SIGSEGV handler," + " aborting stack trace!\n"); + return; } - stack = esp; - for(i = 0; i < kstack_depth_to_print; i++) { +#ifdef CONFIG_FRAME_POINTER + bp = get_frame_pointer(task, segv_regs); +#endif + + if (!stack) + sp = get_stack_pointer(task, segv_regs); + + printk(KERN_INFO "Stack:\n"); + stack = sp; + for (i = 0; i < 3 * STACKSLOTS_PER_LINE; i++) { if (kstack_end(stack)) break; - if (i && ((i % 8) == 0)) - printk("\n "); - printk("%08lx ", *stack++); + if (i && ((i % STACKSLOTS_PER_LINE) == 0)) + printk(KERN_CONT "\n"); + printk(KERN_CONT " %08lx", *stack++); } + printk(KERN_CONT "\n"); - printk("Call Trace: \n"); - show_trace(task, esp); + do_stack_trace(sp, bp); } diff --git a/arch/um/kernel/time.c b/arch/um/kernel/time.c index 2e354b3ca06..117568d4f64 100644 --- a/arch/um/kernel/time.c +++ b/arch/um/kernel/time.c @@ -1,180 +1,115 @@ /* - * Copyright (C) 2000 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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 "linux/hrtimer.h" -#include "asm/irq.h" -#include "asm/param.h" -#include "asm/current.h" -#include "kern_util.h" -#include "user_util.h" -#include "mode.h" -#include "os.h" - -int hz(void) +#include <linux/clockchips.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/jiffies.h> +#include <linux/threads.h> +#include <asm/irq.h> +#include <asm/param.h> +#include <kern_util.h> +#include <os.h> + +void timer_handler(int sig, struct siginfo *unused_si, struct uml_pt_regs *regs) { - return(HZ); -} + unsigned long flags; -/* - * Scheduler clock - returns current time in nanosec units. - */ -unsigned long long sched_clock(void) -{ - return (unsigned long long)jiffies_64 * (1000000000 / HZ); + local_irq_save(flags); + do_IRQ(TIMER_IRQ, regs); + local_irq_restore(flags); } -static unsigned long long prev_nsecs; -#ifdef CONFIG_UML_REAL_TIME_CLOCK -static long long delta; /* Deviation per interval */ -#endif - -void timer_irq(union uml_pt_regs *regs) +static void itimer_set_mode(enum clock_event_mode mode, + struct clock_event_device *evt) { - unsigned long long ticks = 0; - -#ifdef CONFIG_UML_REAL_TIME_CLOCK - if(prev_nsecs){ - /* We've had 1 tick */ - unsigned long long nsecs = os_nsecs(); - - delta += nsecs - prev_nsecs; - prev_nsecs = nsecs; - - /* Protect against the host clock being set backwards */ - if(delta < 0) - delta = 0; - - ticks += (delta * HZ) / BILLION; - delta -= (ticks * BILLION) / HZ; - } - else prev_nsecs = os_nsecs(); -#else - ticks = 1; -#endif - while(ticks > 0){ - do_IRQ(TIMER_IRQ, regs); - ticks--; + switch (mode) { + case CLOCK_EVT_MODE_PERIODIC: + set_interval(); + break; + + case CLOCK_EVT_MODE_SHUTDOWN: + case CLOCK_EVT_MODE_UNUSED: + case CLOCK_EVT_MODE_ONESHOT: + disable_timer(); + break; + + case CLOCK_EVT_MODE_RESUME: + break; } } -static DEFINE_SPINLOCK(timer_spinlock); - -static unsigned long long local_offset = 0; - -static inline unsigned long long get_time(void) +static int itimer_next_event(unsigned long delta, + struct clock_event_device *evt) { - unsigned long long nsecs; - unsigned long flags; - - spin_lock_irqsave(&timer_spinlock, flags); - nsecs = os_nsecs(); - nsecs += local_offset; - spin_unlock_irqrestore(&timer_spinlock, flags); - - return nsecs; + return timer_one_shot(delta + 1); } -irqreturn_t um_timer(int irq, void *dev) +static struct clock_event_device itimer_clockevent = { + .name = "itimer", + .rating = 250, + .cpumask = cpu_all_mask, + .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT, + .set_mode = itimer_set_mode, + .set_next_event = itimer_next_event, + .shift = 32, + .irq = 0, +}; + +static irqreturn_t um_timer(int irq, void *dev) { - unsigned long long nsecs; - unsigned long flags; - - write_seqlock_irqsave(&xtime_lock, flags); - - do_timer(1); - - nsecs = get_time(); - xtime.tv_sec = nsecs / NSEC_PER_SEC; - xtime.tv_nsec = nsecs - xtime.tv_sec * NSEC_PER_SEC; - - write_sequnlock_irqrestore(&xtime_lock, flags); + (*itimer_clockevent.event_handler)(&itimer_clockevent); return IRQ_HANDLED; } -static void register_timer(void) +static cycle_t itimer_read(struct clocksource *cs) { - int err; - - err = request_irq(TIMER_IRQ, um_timer, IRQF_DISABLED, "timer", NULL); - if(err != 0) - printk(KERN_ERR "register_timer : request_irq failed - " - "errno = %d\n", -err); - - err = set_interval(1); - if(err != 0) - printk(KERN_ERR "register_timer : set_interval failed - " - "errno = %d\n", -err); + return os_nsecs() / 1000; } -extern void (*late_time_init)(void); +static struct clocksource itimer_clocksource = { + .name = "itimer", + .rating = 300, + .read = itimer_read, + .mask = CLOCKSOURCE_MASK(64), + .flags = CLOCK_SOURCE_IS_CONTINUOUS, +}; -void time_init(void) +static void __init setup_itimer(void) { - long long nsecs; - - nsecs = os_nsecs(); - set_normalized_timespec(&wall_to_monotonic, -nsecs / BILLION, - -nsecs % BILLION); - late_time_init = register_timer; -} - -void do_gettimeofday(struct timeval *tv) -{ - unsigned long long nsecs = get_time(); - - tv->tv_sec = nsecs / NSEC_PER_SEC; - /* Careful about calculations here - this was originally done as - * (nsecs - tv->tv_sec * NSEC_PER_SEC) / NSEC_PER_USEC - * which gave bogus (> 1000000) values. Dunno why, suspect gcc - * (4.0.0) miscompiled it, or there's a subtle 64/32-bit conversion - * problem that I missed. - */ - nsecs -= tv->tv_sec * NSEC_PER_SEC; - tv->tv_usec = (unsigned long) nsecs / NSEC_PER_USEC; -} - -static inline void set_time(unsigned long long nsecs) -{ - unsigned long long now; - unsigned long flags; + int err; - spin_lock_irqsave(&timer_spinlock, flags); - now = os_nsecs(); - local_offset = nsecs - now; - spin_unlock_irqrestore(&timer_spinlock, flags); + err = request_irq(TIMER_IRQ, um_timer, 0, "timer", NULL); + if (err != 0) + printk(KERN_ERR "register_timer : request_irq failed - " + "errno = %d\n", -err); - clock_was_set(); + itimer_clockevent.mult = div_sc(HZ, NSEC_PER_SEC, 32); + itimer_clockevent.max_delta_ns = + clockevent_delta2ns(60 * HZ, &itimer_clockevent); + itimer_clockevent.min_delta_ns = + clockevent_delta2ns(1, &itimer_clockevent); + err = clocksource_register_hz(&itimer_clocksource, USEC_PER_SEC); + if (err) { + printk(KERN_ERR "clocksource_register_hz returned %d\n", err); + return; + } + clockevents_register_device(&itimer_clockevent); } -int do_settimeofday(struct timespec *tv) +void read_persistent_clock(struct timespec *ts) { - set_time((unsigned long long) tv->tv_sec * NSEC_PER_SEC + tv->tv_nsec); + long long nsecs = os_nsecs(); - return 0; + set_normalized_timespec(ts, nsecs / NSEC_PER_SEC, + nsecs % NSEC_PER_SEC); } -void timer_handler(int sig, union uml_pt_regs *regs) +void __init time_init(void) { - local_irq_disable(); - irq_enter(); - update_process_times(CHOOSE_MODE( - (UPT_SC(regs) && user_context(UPT_SP(regs))), - (regs)->skas.is_user)); - irq_exit(); - local_irq_enable(); - if(current_thread->cpu == 0) - timer_irq(regs); + timer_init(); + late_time_init = setup_itimer; } diff --git a/arch/um/kernel/tlb.c b/arch/um/kernel/tlb.c index 54a5ff25645..f1b3eb14b85 100644 --- a/arch/um/kernel/tlb.c +++ b/arch/um/kernel/tlb.c @@ -1,235 +1,301 @@ /* - * Copyright (C) 2000, 2001, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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" +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/sched.h> +#include <asm/pgtable.h> +#include <asm/tlbflush.h> +#include <as-layout.h> +#include <mem_user.h> +#include <os.h> +#include <skas.h> +#include <kern_util.h> + +struct host_vm_change { + struct host_vm_op { + enum { NONE, MMAP, MUNMAP, MPROTECT } type; + union { + struct { + unsigned long addr; + unsigned long len; + unsigned int prot; + int fd; + __u64 offset; + } mmap; + struct { + unsigned long addr; + unsigned long len; + } munmap; + struct { + unsigned long addr; + unsigned long len; + unsigned int prot; + } mprotect; + } u; + } ops[1]; + int index; + struct mm_id *id; + void *data; + int force; +}; + +#define INIT_HVC(mm, force) \ + ((struct host_vm_change) \ + { .ops = { { .type = NONE } }, \ + .id = &mm->context.id, \ + .data = NULL, \ + .index = 0, \ + .force = force }) + +static int do_ops(struct host_vm_change *hvc, int end, + int finished) +{ + struct host_vm_op *op; + int i, ret = 0; + + for (i = 0; i < end && !ret; i++) { + op = &hvc->ops[i]; + switch (op->type) { + case MMAP: + ret = map(hvc->id, op->u.mmap.addr, op->u.mmap.len, + op->u.mmap.prot, op->u.mmap.fd, + op->u.mmap.offset, finished, &hvc->data); + break; + case MUNMAP: + ret = unmap(hvc->id, op->u.munmap.addr, + op->u.munmap.len, finished, &hvc->data); + break; + case MPROTECT: + ret = protect(hvc->id, op->u.mprotect.addr, + op->u.mprotect.len, op->u.mprotect.prot, + finished, &hvc->data); + break; + default: + printk(KERN_ERR "Unknown op type %d in do_ops\n", + op->type); + BUG(); + break; + } + } + + return ret; +} static 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, union mm_context *mmu, void **flush, - int (*do_ops)(union mm_context *, struct host_vm_op *, - int, int, void **)) + unsigned int prot, struct host_vm_change *hvc) { __u64 offset; struct host_vm_op *last; int fd, ret = 0; fd = phys_mapping(phys, &offset); - if(*index != -1){ - last = &ops[*index]; - if((last->type == MMAP) && + if (hvc->index != 0) { + last = &hvc->ops[hvc->index - 1]; + 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.prot == prot) && (last->u.mmap.fd == fd) && + (last->u.mmap.offset + last->u.mmap.len == offset)) { last->u.mmap.len += len; return 0; } } - if(*index == last_filled){ - ret = (*do_ops)(mmu, ops, last_filled, 0, flush); - *index = -1; + if (hvc->index == ARRAY_SIZE(hvc->ops)) { + ret = do_ops(hvc, ARRAY_SIZE(hvc->ops), 0); + hvc->index = 0; } - ops[++*index] = ((struct host_vm_op) { .type = MMAP, - .u = { .mmap = { - .addr = virt, - .len = len, - .r = r, - .w = w, - .x = x, - .fd = fd, - .offset = offset } + hvc->ops[hvc->index++] = ((struct host_vm_op) + { .type = MMAP, + .u = { .mmap = { .addr = virt, + .len = len, + .prot = prot, + .fd = fd, + .offset = offset } } }); return ret; } static int add_munmap(unsigned long addr, unsigned long len, - struct host_vm_op *ops, int *index, int last_filled, - union mm_context *mmu, void **flush, - int (*do_ops)(union mm_context *, struct host_vm_op *, - int, int, void **)) + struct host_vm_change *hvc) { struct host_vm_op *last; int ret = 0; - if(*index != -1){ - last = &ops[*index]; - if((last->type == MUNMAP) && - (last->u.munmap.addr + last->u.mmap.len == addr)){ + if ((addr >= STUB_START) && (addr < STUB_END)) + return -EINVAL; + + if (hvc->index != 0) { + last = &hvc->ops[hvc->index - 1]; + if ((last->type == MUNMAP) && + (last->u.munmap.addr + last->u.mmap.len == addr)) { last->u.munmap.len += len; return 0; } } - if(*index == last_filled){ - ret = (*do_ops)(mmu, ops, last_filled, 0, flush); - *index = -1; + if (hvc->index == ARRAY_SIZE(hvc->ops)) { + ret = do_ops(hvc, ARRAY_SIZE(hvc->ops), 0); + hvc->index = 0; } - ops[++*index] = ((struct host_vm_op) { .type = MUNMAP, - .u = { .munmap = { - .addr = addr, - .len = len } } }); + hvc->ops[hvc->index++] = ((struct host_vm_op) + { .type = MUNMAP, + .u = { .munmap = { .addr = addr, + .len = len } } }); return ret; } -static 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, union mm_context *mmu, void **flush, - int (*do_ops)(union mm_context *, struct host_vm_op *, - int, int, void **)) +static int add_mprotect(unsigned long addr, unsigned long len, + unsigned int prot, struct host_vm_change *hvc) { struct host_vm_op *last; int ret = 0; - if(*index != -1){ - last = &ops[*index]; - if((last->type == MPROTECT) && + if (hvc->index != 0) { + last = &hvc->ops[hvc->index - 1]; + 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.prot == prot)) { last->u.mprotect.len += len; return 0; } } - if(*index == last_filled){ - ret = (*do_ops)(mmu, ops, last_filled, 0, flush); - *index = -1; + if (hvc->index == ARRAY_SIZE(hvc->ops)) { + ret = do_ops(hvc, ARRAY_SIZE(hvc->ops), 0); + hvc->index = 0; } - ops[++*index] = ((struct host_vm_op) { .type = MPROTECT, - .u = { .mprotect = { - .addr = addr, - .len = len, - .r = r, - .w = w, - .x = x } } }); + hvc->ops[hvc->index++] = ((struct host_vm_op) + { .type = MPROTECT, + .u = { .mprotect = { .addr = addr, + .len = len, + .prot = prot } } }); return ret; } #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 (*do_ops)(union mm_context *, struct host_vm_op *, - int, int, void **)) +static inline int update_pte_range(pmd_t *pmd, unsigned long addr, + unsigned long end, + struct host_vm_change *hvc) { - pgd_t *npgd; - pud_t *npud; - pmd_t *npmd; - pte_t *npte; - union mm_context *mmu = &mm->context; - unsigned long addr, end; - int r, w, x; - struct host_vm_op ops[1]; - void *flush = NULL; - int op_index = -1, last_op = ARRAY_SIZE(ops) - 1; - int ret = 0; - - if(mm == NULL) - return; - - ops[0].type = NONE; - for(addr = start_addr; addr < end_addr && !ret;){ - 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)){ - ret = add_munmap(addr, end - addr, ops, - &op_index, last_op, mmu, - &flush, 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)){ - ret = add_munmap(addr, end - addr, ops, - &op_index, last_op, mmu, - &flush, do_ops); - pud_mkuptodate(*npud); - } - addr = end; - continue; - } + pte_t *pte; + int r, w, x, prot, ret = 0; - 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)){ - ret = add_munmap(addr, end - addr, ops, - &op_index, last_op, mmu, - &flush, do_ops); - pmd_mkuptodate(*npmd); - } - addr = end; + pte = pte_offset_kernel(pmd, addr); + do { + if ((addr >= STUB_START) && (addr < STUB_END)) continue; - } - npte = pte_offset_kernel(npmd, addr); - r = pte_read(*npte); - w = pte_write(*npte); - x = pte_exec(*npte); - if (!pte_young(*npte)) { + r = pte_read(*pte); + w = pte_write(*pte); + x = pte_exec(*pte); + if (!pte_young(*pte)) { r = 0; w = 0; - } else if (!pte_dirty(*npte)) { + } else if (!pte_dirty(*pte)) w = 0; + + prot = ((r ? UM_PROT_READ : 0) | (w ? UM_PROT_WRITE : 0) | + (x ? UM_PROT_EXEC : 0)); + if (hvc->force || pte_newpage(*pte)) { + if (pte_present(*pte)) + ret = add_mmap(addr, pte_val(*pte) & PAGE_MASK, + PAGE_SIZE, prot, hvc); + else + ret = add_munmap(addr, PAGE_SIZE, hvc); + } else if (pte_newprot(*pte)) + ret = add_mprotect(addr, PAGE_SIZE, prot, hvc); + *pte = pte_mkuptodate(*pte); + } while (pte++, addr += PAGE_SIZE, ((addr < end) && !ret)); + return ret; +} + +static inline int update_pmd_range(pud_t *pud, unsigned long addr, + unsigned long end, + struct host_vm_change *hvc) +{ + pmd_t *pmd; + unsigned long next; + int ret = 0; + + pmd = pmd_offset(pud, addr); + do { + next = pmd_addr_end(addr, end); + if (!pmd_present(*pmd)) { + if (hvc->force || pmd_newpage(*pmd)) { + ret = add_munmap(addr, next - addr, hvc); + pmd_mkuptodate(*pmd); + } } - if(force || pte_newpage(*npte)){ - if(pte_present(*npte)) - ret = add_mmap(addr, - pte_val(*npte) & PAGE_MASK, - PAGE_SIZE, r, w, x, ops, - &op_index, last_op, mmu, - &flush, do_ops); - else ret = add_munmap(addr, PAGE_SIZE, ops, - &op_index, last_op, mmu, - &flush, do_ops); + else ret = update_pte_range(pmd, addr, next, hvc); + } while (pmd++, addr = next, ((addr < end) && !ret)); + return ret; +} + +static inline int update_pud_range(pgd_t *pgd, unsigned long addr, + unsigned long end, + struct host_vm_change *hvc) +{ + pud_t *pud; + unsigned long next; + int ret = 0; + + pud = pud_offset(pgd, addr); + do { + next = pud_addr_end(addr, end); + if (!pud_present(*pud)) { + if (hvc->force || pud_newpage(*pud)) { + ret = add_munmap(addr, next - addr, hvc); + pud_mkuptodate(*pud); + } } - else if(pte_newprot(*npte)) - ret = add_mprotect(addr, PAGE_SIZE, r, w, x, ops, - &op_index, last_op, mmu, - &flush, do_ops); + else ret = update_pmd_range(pud, addr, next, hvc); + } while (pud++, addr = next, ((addr < end) && !ret)); + return ret; +} - *npte = pte_mkuptodate(*npte); - addr += PAGE_SIZE; - } - if(!ret) - ret = (*do_ops)(mmu, ops, op_index, 1, &flush); +void fix_range_common(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force) +{ + pgd_t *pgd; + struct host_vm_change hvc; + unsigned long addr = start_addr, next; + int ret = 0; -/* This is not an else because ret is modified above */ - if(ret) { - printk("fix_range_common: failed, killing current process\n"); + hvc = INIT_HVC(mm, force); + pgd = pgd_offset(mm, addr); + do { + next = pgd_addr_end(addr, end_addr); + if (!pgd_present(*pgd)) { + if (force || pgd_newpage(*pgd)) { + ret = add_munmap(addr, next - addr, &hvc); + pgd_mkuptodate(*pgd); + } + } + else ret = update_pud_range(pgd, addr, next, &hvc); + } while (pgd++, addr = next, ((addr < end_addr) && !ret)); + + if (!ret) + ret = do_ops(&hvc, hvc.index, 1); + + /* This is not an else because ret is modified above */ + if (ret) { + printk(KERN_ERR "fix_range_common: failed, killing current " + "process: %d\n", task_tgid_vnr(current)); + /* We are under mmap_sem, release it such that current can terminate */ + up_write(¤t->mm->mmap_sem); force_sig(SIGKILL, current); + do_signal(); } } -int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) +static int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) { struct mm_struct *mm; pgd_t *pgd; @@ -240,17 +306,17 @@ int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) int updated = 0, err; mm = &init_mm; - for(addr = start; addr < end;){ + for (addr = start; addr < end;) { pgd = pgd_offset(mm, addr); - if(!pgd_present(*pgd)){ + if (!pgd_present(*pgd)) { last = ADD_ROUND(addr, PGDIR_SIZE); - if(last > end) + if (last > end) last = end; - if(pgd_newpage(*pgd)){ + if (pgd_newpage(*pgd)) { updated = 1; err = os_unmap_memory((void *) addr, last - addr); - if(err < 0) + if (err < 0) panic("munmap failed, errno = %d\n", -err); } @@ -259,15 +325,15 @@ int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) } pud = pud_offset(pgd, addr); - if(!pud_present(*pud)){ + if (!pud_present(*pud)) { last = ADD_ROUND(addr, PUD_SIZE); - if(last > end) + if (last > end) last = end; - if(pud_newpage(*pud)){ + if (pud_newpage(*pud)) { updated = 1; err = os_unmap_memory((void *) addr, last - addr); - if(err < 0) + if (err < 0) panic("munmap failed, errno = %d\n", -err); } @@ -276,15 +342,15 @@ int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) } pmd = pmd_offset(pud, addr); - if(!pmd_present(*pmd)){ + if (!pmd_present(*pmd)) { last = ADD_ROUND(addr, PMD_SIZE); - if(last > end) + if (last > end) last = end; - if(pmd_newpage(*pmd)){ + if (pmd_newpage(*pmd)) { updated = 1; err = os_unmap_memory((void *) addr, last - addr); - if(err < 0) + if (err < 0) panic("munmap failed, errno = %d\n", -err); } @@ -293,45 +359,110 @@ int flush_tlb_kernel_range_common(unsigned long start, unsigned long end) } pte = pte_offset_kernel(pmd, addr); - if(!pte_present(*pte) || pte_newpage(*pte)){ + if (!pte_present(*pte) || pte_newpage(*pte)) { updated = 1; err = os_unmap_memory((void *) addr, PAGE_SIZE); - if(err < 0) + if (err < 0) panic("munmap failed, errno = %d\n", -err); - if(pte_present(*pte)) + if (pte_present(*pte)) map_memory(addr, pte_val(*pte) & PAGE_MASK, PAGE_SIZE, 1, 1, 1); } - else if(pte_newprot(*pte)){ + else if (pte_newprot(*pte)) { updated = 1; os_protect_memory((void *) addr, PAGE_SIZE, 1, 1, 1); } addr += PAGE_SIZE; } - return(updated); + return updated; +} + +void flush_tlb_page(struct vm_area_struct *vma, unsigned long address) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + struct mm_struct *mm = vma->vm_mm; + void *flush = NULL; + int r, w, x, prot, err = 0; + struct mm_id *mm_id; + + address &= PAGE_MASK; + pgd = pgd_offset(mm, address); + if (!pgd_present(*pgd)) + goto kill; + + pud = pud_offset(pgd, address); + if (!pud_present(*pud)) + goto kill; + + pmd = pmd_offset(pud, address); + if (!pmd_present(*pmd)) + goto kill; + + pte = pte_offset_kernel(pmd, address); + + r = pte_read(*pte); + w = pte_write(*pte); + x = pte_exec(*pte); + if (!pte_young(*pte)) { + r = 0; + w = 0; + } else if (!pte_dirty(*pte)) { + w = 0; + } + + mm_id = &mm->context.id; + prot = ((r ? UM_PROT_READ : 0) | (w ? UM_PROT_WRITE : 0) | + (x ? UM_PROT_EXEC : 0)); + if (pte_newpage(*pte)) { + if (pte_present(*pte)) { + unsigned long long offset; + int fd; + + fd = phys_mapping(pte_val(*pte) & PAGE_MASK, &offset); + err = map(mm_id, address, PAGE_SIZE, prot, fd, offset, + 1, &flush); + } + else err = unmap(mm_id, address, PAGE_SIZE, 1, &flush); + } + else if (pte_newprot(*pte)) + err = protect(mm_id, address, PAGE_SIZE, prot, 1, &flush); + + if (err) + goto kill; + + *pte = pte_mkuptodate(*pte); + + return; + +kill: + printk(KERN_ERR "Failed to flush page for address 0x%lx\n", address); + force_sig(SIGKILL, current); } pgd_t *pgd_offset_proc(struct mm_struct *mm, unsigned long address) { - return(pgd_offset(mm, address)); + return pgd_offset(mm, address); } pud_t *pud_offset_proc(pgd_t *pgd, unsigned long address) { - return(pud_offset(pgd, address)); + return pud_offset(pgd, address); } pmd_t *pmd_offset_proc(pud_t *pud, unsigned long address) { - return(pmd_offset(pud, address)); + return pmd_offset(pud, address); } pte_t *pte_offset_proc(pmd_t *pmd, unsigned long address) { - return(pte_offset_kernel(pmd, address)); + return pte_offset_kernel(pmd, address); } pte_t *addr_pte(struct task_struct *task, unsigned long addr) @@ -340,13 +471,7 @@ pte_t *addr_pte(struct task_struct *task, unsigned long addr) pud_t *pud = pud_offset(pgd, addr); pmd_t *pmd = pmd_offset(pud, addr); - return(pte_offset_map(pmd, addr)); -} - -void flush_tlb_page(struct vm_area_struct *vma, unsigned long address) -{ - address &= PAGE_MASK; - flush_tlb_range(vma, address, address + PAGE_SIZE); + return pte_offset_map(pmd, addr); } void flush_tlb_all(void) @@ -356,35 +481,64 @@ void flush_tlb_all(void) 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); + 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)); + 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); + flush_tlb_kernel_range_common(addr, addr + PAGE_SIZE); +} + +static void fix_range(struct mm_struct *mm, unsigned long start_addr, + unsigned long end_addr, int force) +{ + fix_range_common(mm, start_addr, end_addr, force); } 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); + if (vma->vm_mm == NULL) + flush_tlb_kernel_range_common(start, end); + else fix_range(vma->vm_mm, start, end, 0); +} +EXPORT_SYMBOL(flush_tlb_range); + +void flush_tlb_mm_range(struct mm_struct *mm, unsigned long start, + unsigned long end) +{ + /* + * Don't bother flushing if this address space is about to be + * destroyed. + */ + if (atomic_read(&mm->mm_users) == 0) + return; + + fix_range(mm, start, end, 0); } void flush_tlb_mm(struct mm_struct *mm) { - CHOOSE_MODE_PROC(flush_tlb_mm_tt, flush_tlb_mm_skas, mm); + struct vm_area_struct *vma = mm->mmap; + + while (vma != NULL) { + fix_range(mm, vma->vm_start, vma->vm_end, 0); + vma = vma->vm_next; + } } void force_flush_all(void) { - CHOOSE_MODE(force_flush_all_tt(), force_flush_all_skas()); -} + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma = mm->mmap; + while (vma != NULL) { + fix_range(mm, vma->vm_start, vma->vm_end, 1); + vma = vma->vm_next; + } +} diff --git a/arch/um/kernel/trap.c b/arch/um/kernel/trap.c index b5f124a2f6a..5678c3571e7 100644 --- a/arch/um/kernel/trap.c +++ b/arch/um/kernel/trap.c @@ -1,39 +1,25 @@ /* - * Copyright (C) 2000, 2001 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.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/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 "sysdep/sigcontext.h" -#include "user_util.h" -#include "kern_util.h" -#include "kern.h" -#include "chan_kern.h" -#include "mconsole_kern.h" -#include "mem.h" -#include "mem_kern.h" -#include "sysdep/sigcontext.h" -#include "sysdep/ptrace.h" -#include "os.h" -#ifdef CONFIG_MODE_SKAS -#include "skas.h" -#endif -#include "os.h" +#include <linux/mm.h> +#include <linux/sched.h> +#include <linux/hardirq.h> +#include <linux/module.h> +#include <asm/current.h> +#include <asm/pgtable.h> +#include <asm/tlbflush.h> +#include <arch.h> +#include <as-layout.h> +#include <kern_util.h> +#include <os.h> +#include <skas.h> -/* Note this is constrained to return 0, -EFAULT, -EACCESS, -ENOMEM by segv(). */ +/* + * Note this is constrained to return 0, -EFAULT, -EACCESS, -ENOMEM by + * segv(). + */ int handle_page_fault(unsigned long address, unsigned long ip, int is_write, int is_user, int *code_out) { @@ -44,61 +30,83 @@ int handle_page_fault(unsigned long address, unsigned long ip, pmd_t *pmd; pte_t *pte; int err = -EFAULT; + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE; *code_out = SEGV_MAPERR; - /* If the fault was during atomic operation, don't take the fault, just - * fail. */ + /* + * If the fault was during atomic operation, don't take the fault, just + * fail. + */ if (in_atomic()) goto out_nosemaphore; + if (is_user) + flags |= FAULT_FLAG_USER; +retry: down_read(&mm->mmap_sem); vma = find_vma(mm, address); - if(!vma) + if (!vma) goto out; - else if(vma->vm_start <= address) + else if (vma->vm_start <= address) goto good_area; - else if(!(vma->vm_flags & VM_GROWSDOWN)) + else if (!(vma->vm_flags & VM_GROWSDOWN)) goto out; - else if(is_user && !ARCH_IS_STACKGROW(address)) + else if (is_user && !ARCH_IS_STACKGROW(address)) goto out; - else if(expand_stack(vma, address)) + else if (expand_stack(vma, address)) goto out; good_area: *code_out = SEGV_ACCERR; - if(is_write && !(vma->vm_flags & VM_WRITE)) - goto out; - - /* Don't require VM_READ|VM_EXEC for write faults! */ - if(!is_write && !(vma->vm_flags & (VM_READ | VM_EXEC))) - goto out; + if (is_write) { + if (!(vma->vm_flags & VM_WRITE)) + goto out; + flags |= FAULT_FLAG_WRITE; + } else { + /* Don't require VM_READ|VM_EXEC for write faults! */ + if (!(vma->vm_flags & (VM_READ | VM_EXEC))) + goto out; + } 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: + int fault; + + fault = handle_mm_fault(mm, vma, address, flags); + + if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) + goto out_nosemaphore; + + if (unlikely(fault & VM_FAULT_ERROR)) { + if (fault & VM_FAULT_OOM) { + goto out_of_memory; + } else if (fault & VM_FAULT_SIGBUS) { + err = -EACCES; + goto out; + } BUG(); } + if (flags & FAULT_FLAG_ALLOW_RETRY) { + if (fault & VM_FAULT_MAJOR) + current->maj_flt++; + else + current->min_flt++; + if (fault & VM_FAULT_RETRY) { + flags &= ~FAULT_FLAG_ALLOW_RETRY; + flags |= FAULT_FLAG_TRIED; + + goto retry; + } + } + pgd = pgd_offset(mm, address); pud = pud_offset(pgd, address); pmd = pmd_offset(pud, address); pte = pte_offset_kernel(pmd, address); - } while(!pte_present(*pte)); + } while (!pte_present(*pte)); err = 0; - /* The below warning was added in place of + /* + * The below warning was added in place of * pte_mkyoung(); if (is_write) pte_mkdirty(); * If it's triggered, we'd see normally a hang here (a clean pte is * marked read-only to emulate the dirty bit). @@ -112,27 +120,71 @@ survive: out: up_read(&mm->mmap_sem); out_nosemaphore: - return(err); + 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 (is_init(current)) { - up_read(&mm->mmap_sem); - yield(); - down_read(&mm->mmap_sem); - goto survive; - } - goto out; + /* + * We ran out of memory, call the OOM killer, and return the userspace + * (which will retry the fault, or kill us if we got oom-killed). + */ + up_read(&mm->mmap_sem); + if (!is_user) + goto out_nosemaphore; + pagefault_out_of_memory(); + return 0; +} +EXPORT_SYMBOL(handle_page_fault); + +static void show_segv_info(struct uml_pt_regs *regs) +{ + struct task_struct *tsk = current; + struct faultinfo *fi = UPT_FAULTINFO(regs); + + if (!unhandled_signal(tsk, SIGSEGV)) + return; + + if (!printk_ratelimit()) + return; + + printk("%s%s[%d]: segfault at %lx ip %p sp %p error %x", + task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG, + tsk->comm, task_pid_nr(tsk), FAULT_ADDRESS(*fi), + (void *)UPT_IP(regs), (void *)UPT_SP(regs), + fi->error_code); + + print_vma_addr(KERN_CONT " in ", UPT_IP(regs)); + printk(KERN_CONT "\n"); +} + +static void bad_segv(struct faultinfo fi, unsigned long ip) +{ + struct siginfo si; + + si.si_signo = SIGSEGV; + si.si_code = SEGV_ACCERR; + si.si_addr = (void __user *) FAULT_ADDRESS(fi); + current->thread.arch.faultinfo = fi; + force_sig_info(SIGSEGV, &si, current); +} + +void fatal_sigsegv(void) +{ + force_sigsegv(SIGSEGV, current); + do_signal(); + /* + * This is to tell gcc that we're not returning - do_signal + * can, in general, return, but in this case, it's not, since + * we just got a fatal SIGSEGV queued. + */ + os_dump_core(); } -void segv_handler(int sig, union uml_pt_regs *regs) +void segv_handler(int sig, struct siginfo *unused_si, struct uml_pt_regs *regs) { struct faultinfo * fi = UPT_FAULTINFO(regs); - if(UPT_IS_USER(regs) && !SEGV_IS_FIXABLE(fi)){ + if (UPT_IS_USER(regs) && !SEGV_IS_FIXABLE(fi)) { + show_segv_info(regs); bad_segv(*fi, UPT_IP(regs)); return; } @@ -145,114 +197,134 @@ void segv_handler(int sig, union uml_pt_regs *regs) * the info in the regs. A pointer to the info then would * give us bad data! */ -unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, void *sc) +unsigned long segv(struct faultinfo fi, unsigned long ip, int is_user, + struct uml_pt_regs *regs) { struct siginfo si; - void *catcher; + jmp_buf *catcher; int err; - int is_write = FAULT_WRITE(fi); - unsigned long address = FAULT_ADDRESS(fi); - - if(!is_user && (address >= start_vm) && (address < end_vm)){ - flush_tlb_kernel_vm(); - return(0); - } - else if(current->mm == NULL) + int is_write = FAULT_WRITE(fi); + unsigned long address = FAULT_ADDRESS(fi); + + if (!is_user && regs) + current->thread.segv_regs = container_of(regs, struct pt_regs, regs); + + if (!is_user && (address >= start_vm) && (address < end_vm)) { + flush_tlb_kernel_vm(); + goto out; + } + else if (current->mm == NULL) { + show_regs(container_of(regs, struct pt_regs, regs)); panic("Segfault with no mm"); + } if (SEGV_IS_FIXABLE(&fi) || SEGV_MAYBE_FIXABLE(&fi)) - err = handle_page_fault(address, ip, is_write, is_user, &si.si_code); + err = handle_page_fault(address, ip, is_write, is_user, + &si.si_code); else { err = -EFAULT; - /* A thread accessed NULL, we get a fault, but CR2 is invalid. - * This code is used in __do_copy_from_user() of TT mode. */ + /* + * A thread accessed NULL, we get a fault, but CR2 is invalid. + * This code is used in __do_copy_from_user() of TT mode. + * XXX tt mode is gone, so maybe this isn't needed any more + */ address = 0; } catcher = current->thread.fault_catcher; - if(!err) - return(0); - else if(catcher != NULL){ + if (!err) + goto out; + else if (catcher != NULL) { current->thread.fault_addr = (void *) address; - do_longjmp(catcher, 1); + UML_LONGJMP(catcher, 1); } - else if(current->thread.fault_addr != NULL) + else if (current->thread.fault_addr != NULL) panic("fault_addr set but no fault catcher"); - else if(!is_user && arch_fixup(ip, sc)) - return(0); + else if (!is_user && arch_fixup(ip, regs)) + goto out; - if(!is_user) + if (!is_user) { + show_regs(container_of(regs, struct pt_regs, regs)); panic("Kernel mode fault at addr 0x%lx, ip 0x%lx", address, ip); + } + + show_segv_info(regs); if (err == -EACCES) { si.si_signo = SIGBUS; si.si_errno = 0; si.si_code = BUS_ADRERR; si.si_addr = (void __user *)address; - current->thread.arch.faultinfo = fi; + current->thread.arch.faultinfo = fi; force_sig_info(SIGBUS, &si, current); - } else if (err == -ENOMEM) { - printk("VM: killing process %s\n", current->comm); - do_exit(SIGKILL); } else { BUG_ON(err != -EFAULT); si.si_signo = SIGSEGV; si.si_addr = (void __user *) address; - current->thread.arch.faultinfo = fi; + current->thread.arch.faultinfo = fi; force_sig_info(SIGSEGV, &si, current); } - return(0); -} -void bad_segv(struct faultinfo fi, unsigned long ip) -{ - struct siginfo si; +out: + if (regs) + current->thread.segv_regs = NULL; - si.si_signo = SIGSEGV; - si.si_code = SEGV_ACCERR; - si.si_addr = (void __user *) FAULT_ADDRESS(fi); - current->thread.arch.faultinfo = fi; - force_sig_info(SIGSEGV, &si, current); + return 0; } -void relay_signal(int sig, union uml_pt_regs *regs) +void relay_signal(int sig, struct siginfo *si, struct uml_pt_regs *regs) { - if(arch_handle_signal(sig, regs)) - return; + struct faultinfo *fi; + struct siginfo clean_si; - if(!UPT_IS_USER(regs)){ - if(sig == SIGBUS) - printk("Bus error - the /dev/shm or /tmp mount likely " - "just ran out of space\n"); + if (!UPT_IS_USER(regs)) { + if (sig == SIGBUS) + printk(KERN_ERR "Bus error - the host /dev/shm or /tmp " + "mount likely just ran out of space\n"); panic("Kernel mode signal %d", sig); } - current->thread.arch.faultinfo = *UPT_FAULTINFO(regs); - force_sig(sig, current); + arch_examine_signal(sig, regs); + + memset(&clean_si, 0, sizeof(clean_si)); + clean_si.si_signo = si->si_signo; + clean_si.si_errno = si->si_errno; + clean_si.si_code = si->si_code; + switch (sig) { + case SIGILL: + case SIGFPE: + case SIGSEGV: + case SIGBUS: + case SIGTRAP: + fi = UPT_FAULTINFO(regs); + clean_si.si_addr = (void __user *) FAULT_ADDRESS(*fi); + current->thread.arch.faultinfo = *fi; +#ifdef __ARCH_SI_TRAPNO + clean_si.si_trapno = si->si_trapno; +#endif + break; + default: + printk(KERN_ERR "Attempted to relay unknown signal %d (si_code = %d)\n", + sig, si->si_code); + } + + force_sig_info(sig, &clean_si, current); } -void bus_handler(int sig, union uml_pt_regs *regs) +void bus_handler(int sig, struct siginfo *si, struct uml_pt_regs *regs) { - if(current->thread.fault_catcher != NULL) - do_longjmp(current->thread.fault_catcher, 1); - else relay_signal(sig, regs); + if (current->thread.fault_catcher != NULL) + UML_LONGJMP(current->thread.fault_catcher, 1); + else + relay_signal(sig, si, regs); } -void winch(int sig, union uml_pt_regs *regs) +void winch(int sig, struct siginfo *unused_si, struct uml_pt_regs *regs) { do_IRQ(WINCH_IRQ, regs); } -const struct kern_handlers handlinfo_kern = { - .relay_signal = relay_signal, - .winch = winch, - .bus_handler = bus_handler, - .page_fault = segv_handler, - .sigio_handler = sigio_handler, - .timer_handler = timer_handler -}; - void trap_init(void) { } diff --git a/arch/um/kernel/tt/Makefile b/arch/um/kernel/tt/Makefile deleted file mode 100644 index 6939e5af847..00000000000 --- a/arch/um/kernel/tt/Makefile +++ /dev/null @@ -1,14 +0,0 @@ -# -# Copyright (C) 2002 - 2003 Jeff Dike (jdike@addtoit.com) -# Licensed under the GPL -# - -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 tlb.o tracer.o trap_user.o \ - uaccess.o uaccess_user.o - -obj-$(CONFIG_PT_PROXY) += gdb_kern.o ptproxy/ - -USER_OBJS := gdb.o tracer.o - -include arch/um/scripts/Makefile.rules diff --git a/arch/um/kernel/tt/exec_kern.c b/arch/um/kernel/tt/exec_kern.c deleted file mode 100644 index ad66df17d9d..00000000000 --- a/arch/um/kernel/tt/exec_kern.c +++ /dev/null @@ -1,85 +0,0 @@ -/* - * 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 "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(); - 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(task_stack_page(current), 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 deleted file mode 100644 index a92c02ff2ce..00000000000 --- a/arch/um/kernel/tt/exec_user.c +++ /dev/null @@ -1,57 +0,0 @@ -/* - * 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 deleted file mode 100644 index 786e4edd86c..00000000000 --- a/arch/um/kernel/tt/gdb.c +++ /dev/null @@ -1,279 +0,0 @@ -/* - * 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" -#include "os.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(int 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 deleted file mode 100644 index 68e1bf63cd0..00000000000 --- a/arch/um/kernel/tt/gdb_kern.c +++ /dev/null @@ -1,39 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include "linux/init.h" -#include "mconsole_kern.h" - -#ifdef CONFIG_MCONSOLE - -extern int gdb_config(char *str); -extern int gdb_remove(int n); - -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/mode-tt.h b/arch/um/kernel/tt/include/mode-tt.h deleted file mode 100644 index e171e15fead..00000000000 --- a/arch/um/kernel/tt/include/mode-tt.h +++ /dev/null @@ -1,34 +0,0 @@ -/* - * 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 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 deleted file mode 100644 index 2a35b15c5fe..00000000000 --- a/arch/um/kernel/tt/include/mode_kern-tt.h +++ /dev/null @@ -1,52 +0,0 @@ -/* - * 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 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/ksyms.c b/arch/um/kernel/tt/ksyms.c deleted file mode 100644 index 84a9385a8fe..00000000000 --- a/arch/um/kernel/tt/ksyms.c +++ /dev/null @@ -1,29 +0,0 @@ -/* - * 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(clear_user_tt); - -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 deleted file mode 100644 index 4d1929dfa28..00000000000 --- a/arch/um/kernel/tt/mem.c +++ /dev/null @@ -1,35 +0,0 @@ -/* - * Copyright (C) 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include "linux/stddef.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); -} - -#define SIZE ((CONFIG_NEST_LEVEL + CONFIG_KERNEL_HALF_GIGS) * 0x20000000) -#define START (CONFIG_TOP_ADDR - SIZE) - -unsigned long set_task_sizes_tt(unsigned long *task_size_out) -{ - unsigned long host_task_size; - - /* Round up to the nearest 4M */ - host_task_size = ROUND_4M((unsigned long) &host_task_size); - *task_size_out = START; - - return host_task_size; -} diff --git a/arch/um/kernel/tt/mem_user.c b/arch/um/kernel/tt/mem_user.c deleted file mode 100644 index 03e58989538..00000000000 --- a/arch/um/kernel/tt/mem_user.c +++ /dev/null @@ -1,50 +0,0 @@ -/* - * 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" -#include "os.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 deleted file mode 100644 index 1e86f0bfef7..00000000000 --- a/arch/um/kernel/tt/process_kern.c +++ /dev/null @@ -1,461 +0,0 @@ -/* - * 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 "kern_util.h" -#include "user_util.h" -#include "os.h" -#include "kern.h" -#include "sigcontext.h" -#include "mem_user.h" -#include "tlb.h" -#include "mode.h" -#include "mode_kern.h" -#include "init.h" -#include "tt.h" - -void switch_to_tt(void *prev, void *next) -{ - struct task_struct *from, *to, *prev_sched; - unsigned long flags; - int err, vtalrm, alrm, prof, cpu; - char c; - - from = prev; - to = next; - - cpu = task_thread_info(from)->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; - - /* Notice that here we "up" the semaphore on which "to" is waiting, and - * below (the read) we wait on this semaphore (which is implemented by - * switch_pipe) and go sleeping. Thus, after that, we have resumed in - * "to", and can't use any more the value of "from" (which is outdated), - * nor the value in "to" (since it was the task which stole us the CPU, - * which we don't care about). */ - - 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); - - if(from->thread.mode.tt.switch_pipe[0] == -1) - 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->thread.mode.tt.switch_pipe[0] == -1) - 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_to_tt(prev_sched, current); - - flush_tlb_all(); - local_irq_restore(flags); -} - -void release_thread_tt(struct task_struct *task) -{ - int pid = task->thread.mode.tt.extern_pid; - - /* - * We first have to kill the other process, before - * closing its switch_pipe. Else it might wake up - * and receive "EOF" before we could kill it. - */ - if(os_getpid() != pid) - os_kill_process(pid, 0); - - os_close_file(task->thread.mode.tt.switch_pipe[0]); - os_close_file(task->thread.mode.tt.switch_pipe[1]); - /* use switch_pipe as flag: thread is released */ - task->thread.mode.tt.switch_pipe[0] = -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(struct task_struct *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(); - enable_timer(); - free_page(current->thread.temp_stack); - set_cmdline("(kernel thread)"); - - change_sig(SIGUSR1, 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(task_stack_page(p), 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(®s->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(task)->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 = task_stack_page(&init_task) + - 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); -} diff --git a/arch/um/kernel/tt/ptproxy/Makefile b/arch/um/kernel/tt/ptproxy/Makefile deleted file mode 100644 index 3ad5b774de5..00000000000 --- a/arch/um/kernel/tt/ptproxy/Makefile +++ /dev/null @@ -1,10 +0,0 @@ -# -# 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 deleted file mode 100644 index 58800c50b10..00000000000 --- a/arch/um/kernel/tt/ptproxy/proxy.c +++ /dev/null @@ -1,377 +0,0 @@ -/********************************************************************** -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 deleted file mode 100644 index 5eb0285b196..00000000000 --- a/arch/um/kernel/tt/ptproxy/ptproxy.h +++ /dev/null @@ -1,61 +0,0 @@ -/********************************************************************** -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 deleted file mode 100644 index 03774427d46..00000000000 --- a/arch/um/kernel/tt/ptproxy/ptrace.c +++ /dev/null @@ -1,238 +0,0 @@ -/********************************************************************** -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" -#include "os.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 deleted file mode 100644 index 99f178319d0..00000000000 --- a/arch/um/kernel/tt/ptproxy/sysdep.c +++ /dev/null @@ -1,71 +0,0 @@ -/********************************************************************** -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" -#include "os.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 deleted file mode 100644 index 735f488049a..00000000000 --- a/arch/um/kernel/tt/ptproxy/sysdep.h +++ /dev/null @@ -1,25 +0,0 @@ -/********************************************************************** -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 deleted file mode 100644 index 12f6319d8d7..00000000000 --- a/arch/um/kernel/tt/ptproxy/wait.c +++ /dev/null @@ -1,86 +0,0 @@ -/********************************************************************** -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 deleted file mode 100644 index 542e73ee2ce..00000000000 --- a/arch/um/kernel/tt/ptproxy/wait.h +++ /dev/null @@ -1,15 +0,0 @@ -/********************************************************************** -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 deleted file mode 100644 index 293caa6d0c2..00000000000 --- a/arch/um/kernel/tt/syscall_kern.c +++ /dev/null @@ -1,46 +0,0 @@ -/* - * 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 "sysdep/sigcontext.h" -#include "kern_util.h" -#include "syscall.h" - -void syscall_handler_tt(int sig, struct pt_regs *regs) -{ - void *sc; - long result; - int syscall; - - sc = UPT_SC(®s->regs); - SC_START_SYSCALL(sc); - - syscall = UPT_SYSCALL_NR(®s->regs); - syscall_trace(®s->regs, 0); - - current->thread.nsyscalls++; - nsyscalls++; - - if((syscall >= NR_syscalls) || (syscall < 0)) - result = -ENOSYS; - else result = EXECUTE_SYSCALL(syscall, 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(®s->regs) = sc; - - SC_SET_SYSCALL_RETURN(sc, result); - - syscall_trace(®s->regs, 1); -} diff --git a/arch/um/kernel/tt/syscall_user.c b/arch/um/kernel/tt/syscall_user.c deleted file mode 100644 index 902987bf379..00000000000 --- a/arch/um/kernel/tt/syscall_user.c +++ /dev/null @@ -1,61 +0,0 @@ -/* - * 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.h" -#include "tt.h" - -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); - -#ifdef UPT_ORIGGPR2 - UPT_ORIGGPR2(TASK_REGS(task)) = REGS_ORIGGPR2(proc_regs); -#endif - - 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/tlb.c b/arch/um/kernel/tt/tlb.c deleted file mode 100644 index ae6217c8613..00000000000 --- a/arch/um/kernel/tt/tlb.c +++ /dev/null @@ -1,121 +0,0 @@ -/* - * 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 int do_ops(union mm_context *mmu, struct host_vm_op *ops, int last, - int finished, void **flush) -{ - struct host_vm_op *op; - int i, ret=0; - - for(i = 0; i <= last && !ret; i++){ - op = &ops[i]; - switch(op->type){ - case MMAP: - ret = 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: - ret = os_unmap_memory((void *) op->u.munmap.addr, - op->u.munmap.len); - break; - case MPROTECT: - ret = protect_memory(op->u.mprotect.addr, - op->u.munmap.len, - op->u.mprotect.r, - op->u.mprotect.w, - op->u.mprotect.x, 1); - 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; - } - } - - return ret; -} - -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, 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); -} - -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 deleted file mode 100644 index b9195355075..00000000000 --- a/arch/um/kernel/tt/tracer.c +++ /dev/null @@ -1,462 +0,0 @@ -/* - * 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 "user_util.h" -#include "mem_user.h" -#include "process.h" -#include "kern_util.h" -#include "chan_user.h" -#include "ptrace_user.h" -#include "irq_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 != os_getpgrp()) - 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) -{ - struct faultinfo fi; - GET_FAULTINFO_FROM_SC(fi, &sc); - printf("Tracing thread segfault at address 0x%lx, ip 0x%lx\n", - FAULT_ADDRESS(fi), 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()); - init_irq_signals(0); - 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 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; - - 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); - 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 deleted file mode 100644 index b5d9d64d91e..00000000000 --- a/arch/um/kernel/tt/trap_user.c +++ /dev/null @@ -1,71 +0,0 @@ -/* - * 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 "sysdep/sigcontext.h" -#include "user_util.h" -#include "kern_util.h" -#include "task.h" -#include "tt.h" -#include "os.h" - -void sig_handler_common_tt(int sig, void *sc_ptr) -{ - struct sigcontext *sc = sc_ptr; - struct tt_regs save_regs, *r; - int save_errno = errno, is_user = 0; - void (*handler)(int, union uml_pt_regs *); - - /* 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; - if ( sig == SIGFPE || sig == SIGSEGV || - sig == SIGBUS || sig == SIGILL || - sig == SIGTRAP ) { - GET_FAULTINFO_FROM_SC(r->faultinfo, sc); - } - save_regs = *r; - if (sc) - is_user = user_context(SC_SP(sc)); - r->sc = sc; - if(sig != SIGUSR2) - r->syscall = -1; - - handler = sig_info[sig]; - - /* unblock SIGALRM, SIGVTALRM, SIGIO if sig isn't IRQ signal */ - if (sig != SIGIO && sig != SIGWINCH && - sig != SIGVTALRM && sig != SIGALRM) - unblock_signals(); - - 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 deleted file mode 100644 index 1cb60726567..00000000000 --- a/arch/um/kernel/tt/uaccess.c +++ /dev/null @@ -1,73 +0,0 @@ -/* - * 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(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(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(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(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 deleted file mode 100644 index ed1abcf4d05..00000000000 --- a/arch/um/kernel/tt/uaccess_user.c +++ /dev/null @@ -1,106 +0,0 @@ -/* - * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk) - * Copyright (C) 2001 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -#include <string.h> -#include "user_util.h" -#include "uml_uaccess.h" -#include "task.h" -#include "kern_util.h" -#include "os.h" -#include "longjmp.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 if (fault) - return n - (fault - (unsigned long) from); - else - /* In case of a general protection fault, we don't have the - * fault address, so NULL is used instead. Pretend we didn't - * copy anything. */ - return n; -} - -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; - jmp_buf jbuf; - - *fault_catcher = &jbuf; - if(UML_SETJMP(&jbuf) == 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/uaccess.c b/arch/um/kernel/uaccess.c deleted file mode 100644 index 054e3de0784..00000000000 --- a/arch/um/kernel/uaccess.c +++ /dev/null @@ -1,30 +0,0 @@ -/* - * Copyright (C) 2001 Chris Emerson (cemerson@chiark.greenend.org.uk) - * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) - * Licensed under the GPL - */ - -/* 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. - */ - -#include <linux/string.h> -#include "os.h" - -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)); -} diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c index 66f43c90682..016adf0985d 100644 --- a/arch/um/kernel/um_arch.c +++ b/arch/um/kernel/um_arch.c @@ -1,73 +1,67 @@ /* - * Copyright (C) 2000, 2002 Jeff Dike (jdike@karaya.com) + * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#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 "asm/setup.h" -#include "ubd_user.h" -#include "asm/current.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" -#ifdef UML_CONFIG_MODE_SKAS -#include "skas.h" -#endif +#include <linux/delay.h> +#include <linux/init.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/seq_file.h> +#include <linux/string.h> +#include <linux/utsname.h> +#include <linux/sched.h> +#include <asm/pgtable.h> +#include <asm/processor.h> +#include <asm/sections.h> +#include <asm/setup.h> +#include <as-layout.h> +#include <arch.h> +#include <init.h> +#include <kern.h> +#include <kern_util.h> +#include <mem_user.h> +#include <os.h> #define DEFAULT_COMMAND_LINE "root=98:0" -/* Changed in linux_main and setup_arch, which run before SMP is started */ -static char command_line[COMMAND_LINE_SIZE] = { 0 }; +/* Changed in add_arg and setup_arch, which run before SMP is started */ +static char __initdata command_line[COMMAND_LINE_SIZE] = { 0 }; -static void add_arg(char *arg) +static void __init 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) + if (strlen(command_line) > 0) strcat(command_line, " "); strcat(command_line, arg); } -struct cpuinfo_um boot_cpu_data = { +/* + * These fields are initialized at boot time and not changed. + * XXX This structure is used only in the non-SMP case. Maybe this + * should be moved to smp.c. + */ +struct cpuinfo_um boot_cpu_data = { .loops_per_jiffy = 0, .ipi_pipe = { -1, -1 } }; +union thread_union cpu0_irqstack + __attribute__((__section__(".data..init_irqstack"))) = + { INIT_THREAD_INFO(init_task) }; + unsigned long thread_saved_pc(struct task_struct *task) { - return(os_process_pc(CHOOSE_MODE_PROC(thread_pid_tt, thread_pid_skas, - task))); + /* FIXME: Need to look up userspace_pid by cpu */ + return os_process_pc(userspace_pid[0]); } +/* Changed in setup_arch, which is called in early boot */ +static char host_info[(__NEW_UTS_LEN + 1) * 5]; + static int show_cpuinfo(struct seq_file *m, void *v) { int index = 0; @@ -81,13 +75,13 @@ static int show_cpuinfo(struct seq_file *m, void *v) 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, "mode\t\t: skas\n"); 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); + return 0; } static void *c_start(struct seq_file *m, loff_t *pos) @@ -113,54 +107,23 @@ const struct seq_operations cpuinfo_op = { }; /* 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; +EXPORT_SYMBOL(uml_physmem); + +unsigned long uml_reserved; /* Also modified in mem_init */ unsigned long start_vm; unsigned long end_vm; -int ncpus = 1; -#ifdef CONFIG_CMDLINE_ON_HOST -/* 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 uml_ncpus_setup */ +int ncpus = 1; /* Set in early boot */ static int have_root __initdata = 0; -long long physmem_size = 32 * 1024 * 1024; -void set_cmdline(char *cmd) -{ -#ifdef CONFIG_CMDLINE_ON_HOST - char *umid, *ptr; - - if(CHOOSE_MODE(honeypot, 0)) return; - - umid = get_umid(); - if(*umid != '\0'){ - 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 -} +/* Set in uml_mem_setup and modified in linux_main */ +long long physmem_size = 32 * 1024 * 1024; -static char *usage_string = +static const char *usage_string = "User Mode Linux v%s\n" " available at http://user-mode-linux.sourceforge.net/\n\n"; @@ -192,13 +155,10 @@ __uml_setup("root=", uml_root_setup, " root=/dev/ubd5\n\n" ); -#ifndef CONFIG_MODE_TT - static int __init no_skas_debug_setup(char *line, int *add) { printf("'debug' is not necessary to gdb UML in skas mode - run \n"); - printf("'gdb linux' and disable CONFIG_CMDLINE_ON_HOST if gdb \n"); - printf("doesn't work as expected\n"); + printf("'gdb linux'\n"); return 0; } @@ -208,83 +168,34 @@ __uml_setup("debug", no_skas_debug_setup, " this flag is not needed to run gdb on UML in skas mode\n\n" ); -#endif - #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; - } + if (!sscanf(line, "%d", &ncpus)) { + printf("Couldn't parse [%s]\n", line); + return -1; + } - return 0; + 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" +" 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); -} - -#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; + const char **p; printf(usage_string, init_utsname()->release); - p = &__uml_help_start; - while (p < &__uml_help_end) { - printf("%s", *p); - p++; - } + p = &__uml_help_start; + while (p < &__uml_help_end) { + printf("%s", *p); + p++; + } exit(0); - return 0; } @@ -293,21 +204,19 @@ __uml_setup("--help", Usage, " Prints this message.\n\n" ); -static int __init uml_checksetup(char *line, int *add) +static void __init uml_checksetup(char *line, int *add) { struct uml_param *p; p = &__uml_setup_start; - while(p < &__uml_setup_end) { - int n; + while (p < &__uml_setup_end) { + size_t n; n = strlen(p->str); - if(!strncmp(line, p->str, n)){ - if (p->setup_func(line + n, add)) return 1; - } + if (!strncmp(line, p->str, n) && p->setup_func(line + n, add)) + return; p++; } - return 0; } static void __init uml_postsetup(void) @@ -315,14 +224,35 @@ static void __init uml_postsetup(void) initcall_t *p; p = &__uml_postsetup_start; - while(p < &__uml_postsetup_end){ + while (p < &__uml_postsetup_end) { (*p)(); p++; } return; } +static int panic_exit(struct notifier_block *self, unsigned long unused1, + void *unused2) +{ + bust_spinlocks(1); + bust_spinlocks(0); + uml_exitcode = 1; + os_dump_core(); + return 0; +} + +static struct notifier_block panic_exit_notifier = { + .notifier_call = panic_exit, + .next = NULL, + .priority = 0 +}; + /* Set during early boot */ +unsigned long task_size; +EXPORT_SYMBOL(task_size); + +unsigned long host_task_size; + unsigned long brk_start; unsigned long end_iomem; EXPORT_SYMBOL(end_iomem); @@ -331,93 +261,77 @@ EXPORT_SYMBOL(end_iomem); extern char __binary_start; -int linux_main(int argc, char **argv) +int __init linux_main(int argc, char **argv) { unsigned long avail, diff; unsigned long virtmem_size, max_physmem; - unsigned int i, add; + unsigned long stack; + unsigned int i; + int add; char * mode; - for (i = 1; i < argc; i++){ - if((i == 1) && (argv[i][0] == ' ')) continue; + 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) + if (have_root == 0) add_arg(DEFAULT_COMMAND_LINE); + host_task_size = os_get_top_address(); + /* + * TASK_SIZE needs to be PGDIR_SIZE aligned or else exit_mmap craps + * out + */ + task_size = host_task_size & PGDIR_MASK; + + /* OS sanity checks that need to happen before the kernel runs */ os_early_checks(); - if (force_tt) - clear_can_do_skas(); - 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 -#ifndef CONFIG_MODE_SKAS - mode = "TT"; -#else - /* Show to the user the result of selection */ - if (mode_tt) - mode = "TT"; - else if (proc_mm && ptrace_faultinfo) + can_do_skas(); + + if (proc_mm && ptrace_faultinfo) mode = "SKAS3"; else mode = "SKAS0"; -#endif printf("UML running in %s mode\n", mode); - uml_start = (unsigned long) &__binary_start; - host_task_size = CHOOSE_MODE_PROC(set_task_sizes_tt, - set_task_sizes_skas, &task_size); + brk_start = (unsigned long) sbrk(0); /* - * Setting up handlers to 'sig_info' struct - */ - os_fill_handlinfo(handlinfo_kern); - - 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 */ + * 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){ + 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 & PAGE_MASK; + uml_physmem = (unsigned long) &__binary_start & PAGE_MASK; /* Reserve up to 4M after the current brk */ uml_reserved = ROUND_4M(brk_start) + (1 << 22); setup_machinename(init_utsname()->machine); -#ifdef CONFIG_CMDLINE_ON_HOST - 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; + max_physmem = TASK_SIZE - uml_physmem - iomem_size - MIN_VMALLOC; - /* Zones have to begin on a 1 << MAX_ORDER page boundary, + /* + * 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){ + if (physmem_size + iomem_size > max_physmem) { highmem = physmem_size + iomem_size - max_physmem; physmem_size -= highmem; #ifndef CONFIG_HIGHMEM @@ -434,7 +348,7 @@ int linux_main(int argc, char **argv) start_vm = VMALLOC_START; setup_physmem(uml_physmem, uml_reserved, physmem_size, highmem); - if(init_maps(physmem_size, iomem_size, highmem)){ + if (init_maps(physmem_size, iomem_size, highmem)) { printf("Failed to allocate mem_map for %Lu bytes of physical " "memory and %Lu bytes of highmem\n", physmem_size, highmem); @@ -442,55 +356,40 @@ int linux_main(int argc, char **argv) } virtmem_size = physmem_size; - avail = get_kmem_end() - start_vm; - if(physmem_size > avail) virtmem_size = avail; + stack = (unsigned long) argv; + stack &= ~(1024 * 1024 - 1); + avail = stack - start_vm; + if (physmem_size > avail) + virtmem_size = avail; end_vm = start_vm + virtmem_size; - if(virtmem_size < physmem_size) + if (virtmem_size < physmem_size) printf("Kernel virtual memory size shrunk to %lu bytes\n", virtmem_size); - uml_postsetup(); - - task_protections((unsigned long) &init_thread_info); - os_flush_stdout(); + atomic_notifier_chain_register(&panic_notifier_list, + &panic_exit_notifier); - return(CHOOSE_MODE(start_uml_tt(), start_uml_skas())); -} + uml_postsetup(); -extern int uml_exitcode; + stack_protections((unsigned long) &init_thread_info); + os_flush_stdout(); -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); + return start_uml(); } -static struct notifier_block panic_exit_notifier = { - .notifier_call = panic_exit, - .next = NULL, - .priority = 0 -}; - void __init setup_arch(char **cmdline_p) { - atomic_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(); + strlcpy(boot_command_line, command_line, COMMAND_LINE_SIZE); + *cmdline_p = command_line; + setup_hostinfo(host_info, sizeof host_info); } void __init check_bugs(void) { arch_check_bugs(); - os_check_bugs(); + os_check_bugs(); } void apply_alternatives(struct alt_instr *start, struct alt_instr *end) diff --git a/arch/um/kernel/umid.c b/arch/um/kernel/umid.c index 4eaee823bfd..f6cc3bd6178 100644 --- a/arch/um/kernel/umid.c +++ b/arch/um/kernel/umid.c @@ -1,13 +1,12 @@ -/* - * Copyright (C) 2001, 2002 Jeff Dike (jdike@karaya.com) +/* + * Copyright (C) 2001 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) * Licensed under the GPL */ -#include "asm/errno.h" -#include "init.h" -#include "os.h" -#include "kern.h" -#include "linux/kernel.h" +#include <asm/errno.h> +#include <init.h> +#include <kern.h> +#include <os.h> /* Changed by set_umid_arg */ static int umid_inited = 0; @@ -16,14 +15,16 @@ static int __init set_umid_arg(char *name, int *add) { int err; - if(umid_inited) + if (umid_inited) { + printf("umid already set\n"); return 0; + } *add = 0; err = set_umid(name); - if(err == -EEXIST) + if (err == -EEXIST) printf("umid '%s' already in use\n", name); - else if(!err) + else if (!err) umid_inited = 1; return 0; diff --git a/arch/um/kernel/uml.lds.S b/arch/um/kernel/uml.lds.S index f6301274cf3..6899195602b 100644 --- a/arch/um/kernel/uml.lds.S +++ b/arch/um/kernel/uml.lds.S @@ -1,4 +1,5 @@ #include <asm-generic/vmlinux.lds.h> +#include <asm/page.h> OUTPUT_FORMAT(ELF_FORMAT) OUTPUT_ARCH(ELF_ARCH) @@ -18,48 +19,58 @@ SECTIONS . = START + SIZEOF_HEADERS; -#ifdef MODE_TT - .remap_data : { UNMAP_PATH (.data .bss) } - .remap : { UNMAP_PATH (.text) } - - . = ALIGN(4096); /* Init code and data */ -#endif - _text = .; - _stext = .; - __init_begin = .; - .init.text : { - _sinittext = .; - *(.init.text) - _einittext = .; - } - . = ALIGN(4096); + INIT_TEXT_SECTION(0) + . = ALIGN(PAGE_SIZE); .text : { - *(.text) + _stext = .; + TEXT_TEXT SCHED_TEXT LOCK_TEXT *(.fixup) /* .gnu.warning sections are handled specially by elf32.em. */ *(.gnu.warning) *(.gnu.linkonce.t*) + } + + . = ALIGN(PAGE_SIZE); + .syscall_stub : { + __syscall_stub_start = .; + *(.__syscall_stub*) + __syscall_stub_end = .; + } - . = ALIGN(4096); - __syscall_stub_start = .; - *(.__syscall_stub*) - __syscall_stub_end = .; - . = ALIGN(4096); + /* + * These are needed even in a static link, even if they wind up being empty. + * Newer glibc needs these __rel{,a}_iplt_{start,end} symbols. + */ + .rel.plt : { + *(.rel.plt) + PROVIDE_HIDDEN(__rel_iplt_start = .); + *(.rel.iplt) + PROVIDE_HIDDEN(__rel_iplt_end = .); } + .rela.plt : { + *(.rela.plt) + PROVIDE_HIDDEN(__rela_iplt_start = .); + *(.rela.iplt) + PROVIDE_HIDDEN(__rela_iplt_end = .); + } + + #include <asm/common.lds.S> - #include "asm/common.lds.S" + __init_begin = .; + init.data : { INIT_DATA } + __init_end = .; - init.data : { *(init.data) } .data : { - . = ALIGN(KERNEL_STACK_SIZE); /* init_task */ - *(.data.init_task) - *(.data) + INIT_TASK_DATA(KERNEL_STACK_SIZE) + . = ALIGN(KERNEL_STACK_SIZE); + *(.data..init_irqstack) + DATA_DATA *(.gnu.linkonce.d*) CONSTRUCTORS } @@ -83,24 +94,18 @@ SECTIONS .sdata : { *(.sdata) } _edata = .; PROVIDE (edata = .); - . = ALIGN(0x1000); - .sbss : - { - __bss_start = .; - PROVIDE(_bss_start = .); - *(.sbss) - *(.scommon) - } - .bss : - { - *(.dynbss) - *(.bss) - *(COMMON) - } + . = ALIGN(PAGE_SIZE); + __bss_start = .; + PROVIDE(_bss_start = .); + SBSS(0) + BSS(0) + __bss_stop = .; _end = .; PROVIDE (end = .); STABS_DEBUG DWARF_DEBUG + + DISCARDS } diff --git a/arch/um/kernel/vmlinux.lds.S b/arch/um/kernel/vmlinux.lds.S index f8aeb448aab..16e49bfa2b4 100644 --- a/arch/um/kernel/vmlinux.lds.S +++ b/arch/um/kernel/vmlinux.lds.S @@ -1,3 +1,6 @@ + +KERNEL_STACK_SIZE = 4096 * (1 << CONFIG_KERNEL_STACK_ORDER); + #ifdef CONFIG_LD_SCRIPT_STATIC #include "uml.lds.S" #else |