linux/arch/x86/kernel/vm86_32.c

842 lines
22 KiB
C
Raw Normal View History

/*
* Copyright (C) 1994 Linus Torvalds
*
* 29 dec 2001 - Fixed oopses caused by unchecked access to the vm86
* stack - Manfred Spraul <manfred@colorfullife.com>
*
* 22 mar 2002 - Manfred detected the stackfaults, but didn't handle
* them correctly. Now the emulation will be in a
* consistent state after stackfaults - Kasper Dupont
* <kasperd@daimi.au.dk>
*
* 22 mar 2002 - Added missing clear_IF in set_vflags_* Kasper Dupont
* <kasperd@daimi.au.dk>
*
* ?? ??? 2002 - Fixed premature returns from handle_vm86_fault
* caused by Kasper Dupont's changes - Stas Sergeev
*
* 4 apr 2002 - Fixed CHECK_IF_IN_TRAP broken by Stas' changes.
* Kasper Dupont <kasperd@daimi.au.dk>
*
* 9 apr 2002 - Changed syntax of macros in handle_vm86_fault.
* Kasper Dupont <kasperd@daimi.au.dk>
*
* 9 apr 2002 - Changed stack access macros to jump to a label
* instead of returning to userspace. This simplifies
* do_int, and is needed by handle_vm6_fault. Kasper
* Dupont <kasperd@daimi.au.dk>
*
*/
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/capability.h>
#include <linux/errno.h>
#include <linux/interrupt.h>
#include <linux/syscalls.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/signal.h>
#include <linux/string.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/highmem.h>
#include <linux/ptrace.h>
[PATCH] make vm86 call audit_syscall_exit hi, The motivation behind the patch below was to address messages in /var/log/messages such as: Jan 31 10:54:15 mets kernel: audit(:0): major=252 name_count=0: freeing multiple contexts (1) Jan 31 10:54:15 mets kernel: audit(:0): major=113 name_count=0: freeing multiple contexts (2) I can reproduce by running 'get-edid' from: http://john.fremlin.de/programs/linux/read-edid/. These messages come about in the log b/c the vm86 calls do not exit via the normal system call exit paths and thus do not call 'audit_syscall_exit'. The next system call will then free the context for itself and for the vm86 context, thus generating the above messages. This patch addresses the issue by simply adding a call to 'audit_syscall_exit' from the vm86 code. Besides fixing the above error messages the patch also now allows vm86 system calls to become auditable. This is useful since strace does not appear to properly record the return values from sys_vm86. I think this patch is also a step in the right direction in terms of cleaning up some core auditing code. If we can correct any other paths that do not properly call the audit exit and entries points, then we can also eliminate the notion of context chaining. I've tested this patch by verifying that the log messages no longer appear, and that the audit records for sys_vm86 appear to be correct. Also, 'read_edid' produces itentical output. thanks, -Jason Signed-off-by: Jason Baron <jbaron@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-01-31 21:56:28 +00:00
#include <linux/audit.h>
#include <linux/stddef.h>
#include <asm/uaccess.h>
#include <asm/io.h>
#include <asm/tlbflush.h>
#include <asm/irq.h>
/*
* Known problems:
*
* Interrupt handling is not guaranteed:
* - a real x86 will disable all interrupts for one instruction
* after a "mov ss,xx" to make stack handling atomic even without
* the 'lss' instruction. We can't guarantee this in v86 mode,
* as the next instruction might result in a page fault or similar.
* - a real x86 will have interrupts disabled for one instruction
* past the 'sti' that enables them. We don't bother with all the
* details yet.
*
* Let's hope these problems do not actually matter for anything.
*/
#define KVM86 ((struct kernel_vm86_struct *)regs)
#define VMPI KVM86->vm86plus
/*
* 8- and 16-bit register defines..
*/
#define AL(regs) (((unsigned char *)&((regs)->pt.ax))[0])
#define AH(regs) (((unsigned char *)&((regs)->pt.ax))[1])
#define IP(regs) (*(unsigned short *)&((regs)->pt.ip))
#define SP(regs) (*(unsigned short *)&((regs)->pt.sp))
/*
* virtual flags (16 and 32-bit versions)
*/
#define VFLAGS (*(unsigned short *)&(current->thread.v86flags))
#define VEFLAGS (current->thread.v86flags)
#define set_flags(X, new, mask) \
((X) = ((X) & ~(mask)) | ((new) & (mask)))
#define SAFE_MASK (0xDD5)
#define RETURN_MASK (0xDFF)
/* convert kernel_vm86_regs to vm86_regs */
static int copy_vm86_regs_to_user(struct vm86_regs __user *user,
const struct kernel_vm86_regs *regs)
{
int ret = 0;
/*
* kernel_vm86_regs is missing gs, so copy everything up to
* (but not including) orig_eax, and then rest including orig_eax.
*/
ret += copy_to_user(user, regs, offsetof(struct kernel_vm86_regs, pt.orig_ax));
ret += copy_to_user(&user->orig_eax, &regs->pt.orig_ax,
sizeof(struct kernel_vm86_regs) -
offsetof(struct kernel_vm86_regs, pt.orig_ax));
return ret;
}
/* convert vm86_regs to kernel_vm86_regs */
static int copy_vm86_regs_from_user(struct kernel_vm86_regs *regs,
const struct vm86_regs __user *user,
unsigned extra)
{
int ret = 0;
/* copy ax-fs inclusive */
ret += copy_from_user(regs, user, offsetof(struct kernel_vm86_regs, pt.orig_ax));
/* copy orig_ax-__gsh+extra */
ret += copy_from_user(&regs->pt.orig_ax, &user->orig_eax,
sizeof(struct kernel_vm86_regs) -
offsetof(struct kernel_vm86_regs, pt.orig_ax) +
extra);
return ret;
}
struct pt_regs *save_v86_state(struct kernel_vm86_regs *regs)
{
struct tss_struct *tss;
struct pt_regs *ret;
unsigned long tmp;
/*
* This gets called from entry.S with interrupts disabled, but
* from process context. Enable interrupts here, before trying
* to access user space.
*/
local_irq_enable();
if (!current->thread.vm86_info) {
pr_alert("no vm86_info: BAD\n");
do_exit(SIGSEGV);
}
set_flags(regs->pt.flags, VEFLAGS, X86_EFLAGS_VIF | current->thread.v86mask);
tmp = copy_vm86_regs_to_user(&current->thread.vm86_info->regs, regs);
tmp += put_user(current->thread.screen_bitmap, &current->thread.vm86_info->screen_bitmap);
if (tmp) {
pr_alert("could not access userspace vm86_info\n");
do_exit(SIGSEGV);
}
tss = &per_cpu(init_tss, get_cpu());
current->thread.sp0 = current->thread.saved_sp0;
current->thread.sysenter_cs = __KERNEL_CS;
load_sp0(tss, &current->thread);
current->thread.saved_sp0 = 0;
put_cpu();
ret = KVM86->regs32;
ret->fs = current->thread.saved_fs;
set_user_gs(ret, current->thread.saved_gs);
return ret;
}
static void mark_screen_rdonly(struct mm_struct *mm)
{
pgd_t *pgd;
pud_t *pud;
pmd_t *pmd;
pte_t *pte;
spinlock_t *ptl;
int i;
mm: thp: fix pmd_bad() triggering in code paths holding mmap_sem read mode In some cases it may happen that pmd_none_or_clear_bad() is called with the mmap_sem hold in read mode. In those cases the huge page faults can allocate hugepmds under pmd_none_or_clear_bad() and that can trigger a false positive from pmd_bad() that will not like to see a pmd materializing as trans huge. It's not khugepaged causing the problem, khugepaged holds the mmap_sem in write mode (and all those sites must hold the mmap_sem in read mode to prevent pagetables to go away from under them, during code review it seems vm86 mode on 32bit kernels requires that too unless it's restricted to 1 thread per process or UP builds). The race is only with the huge pagefaults that can convert a pmd_none() into a pmd_trans_huge(). Effectively all these pmd_none_or_clear_bad() sites running with mmap_sem in read mode are somewhat speculative with the page faults, and the result is always undefined when they run simultaneously. This is probably why it wasn't common to run into this. For example if the madvise(MADV_DONTNEED) runs zap_page_range() shortly before the page fault, the hugepage will not be zapped, if the page fault runs first it will be zapped. Altering pmd_bad() not to error out if it finds hugepmds won't be enough to fix this, because zap_pmd_range would then proceed to call zap_pte_range (which would be incorrect if the pmd become a pmd_trans_huge()). The simplest way to fix this is to read the pmd in the local stack (regardless of what we read, no need of actual CPU barriers, only compiler barrier needed), and be sure it is not changing under the code that computes its value. Even if the real pmd is changing under the value we hold on the stack, we don't care. If we actually end up in zap_pte_range it means the pmd was not none already and it was not huge, and it can't become huge from under us (khugepaged locking explained above). All we need is to enforce that there is no way anymore that in a code path like below, pmd_trans_huge can be false, but pmd_none_or_clear_bad can run into a hugepmd. The overhead of a barrier() is just a compiler tweak and should not be measurable (I only added it for THP builds). I don't exclude different compiler versions may have prevented the race too by caching the value of *pmd on the stack (that hasn't been verified, but it wouldn't be impossible considering pmd_none_or_clear_bad, pmd_bad, pmd_trans_huge, pmd_none are all inlines and there's no external function called in between pmd_trans_huge and pmd_none_or_clear_bad). if (pmd_trans_huge(*pmd)) { if (next-addr != HPAGE_PMD_SIZE) { VM_BUG_ON(!rwsem_is_locked(&tlb->mm->mmap_sem)); split_huge_page_pmd(vma->vm_mm, pmd); } else if (zap_huge_pmd(tlb, vma, pmd, addr)) continue; /* fall through */ } if (pmd_none_or_clear_bad(pmd)) Because this race condition could be exercised without special privileges this was reported in CVE-2012-1179. The race was identified and fully explained by Ulrich who debugged it. I'm quoting his accurate explanation below, for reference. ====== start quote ======= mapcount 0 page_mapcount 1 kernel BUG at mm/huge_memory.c:1384! At some point prior to the panic, a "bad pmd ..." message similar to the following is logged on the console: mm/memory.c:145: bad pmd ffff8800376e1f98(80000000314000e7). The "bad pmd ..." message is logged by pmd_clear_bad() before it clears the page's PMD table entry. 143 void pmd_clear_bad(pmd_t *pmd) 144 { -> 145 pmd_ERROR(*pmd); 146 pmd_clear(pmd); 147 } After the PMD table entry has been cleared, there is an inconsistency between the actual number of PMD table entries that are mapping the page and the page's map count (_mapcount field in struct page). When the page is subsequently reclaimed, __split_huge_page() detects this inconsistency. 1381 if (mapcount != page_mapcount(page)) 1382 printk(KERN_ERR "mapcount %d page_mapcount %d\n", 1383 mapcount, page_mapcount(page)); -> 1384 BUG_ON(mapcount != page_mapcount(page)); The root cause of the problem is a race of two threads in a multithreaded process. Thread B incurs a page fault on a virtual address that has never been accessed (PMD entry is zero) while Thread A is executing an madvise() system call on a virtual address within the same 2 MB (huge page) range. virtual address space .---------------------. | | | | .-|---------------------| | | | | | |<-- B(fault) | | | 2 MB | |/////////////////////|-. huge < |/////////////////////| > A(range) page | |/////////////////////|-' | | | | | | '-|---------------------| | | | | '---------------------' - Thread A is executing an madvise(..., MADV_DONTNEED) system call on the virtual address range "A(range)" shown in the picture. sys_madvise // Acquire the semaphore in shared mode. down_read(&current->mm->mmap_sem) ... madvise_vma switch (behavior) case MADV_DONTNEED: madvise_dontneed zap_page_range unmap_vmas unmap_page_range zap_pud_range zap_pmd_range // // Assume that this huge page has never been accessed. // I.e. content of the PMD entry is zero (not mapped). // if (pmd_trans_huge(*pmd)) { // We don't get here due to the above assumption. } // // Assume that Thread B incurred a page fault and .---------> // sneaks in here as shown below. | // | if (pmd_none_or_clear_bad(pmd)) | { | if (unlikely(pmd_bad(*pmd))) | pmd_clear_bad | { | pmd_ERROR | // Log "bad pmd ..." message here. | pmd_clear | // Clear the page's PMD entry. | // Thread B incremented the map count | // in page_add_new_anon_rmap(), but | // now the page is no longer mapped | // by a PMD entry (-> inconsistency). | } | } | v - Thread B is handling a page fault on virtual address "B(fault)" shown in the picture. ... do_page_fault __do_page_fault // Acquire the semaphore in shared mode. down_read_trylock(&mm->mmap_sem) ... handle_mm_fault if (pmd_none(*pmd) && transparent_hugepage_enabled(vma)) // We get here due to the above assumption (PMD entry is zero). do_huge_pmd_anonymous_page alloc_hugepage_vma // Allocate a new transparent huge page here. ... __do_huge_pmd_anonymous_page ... spin_lock(&mm->page_table_lock) ... page_add_new_anon_rmap // Here we increment the page's map count (starts at -1). atomic_set(&page->_mapcount, 0) set_pmd_at // Here we set the page's PMD entry which will be cleared // when Thread A calls pmd_clear_bad(). ... spin_unlock(&mm->page_table_lock) The mmap_sem does not prevent the race because both threads are acquiring it in shared mode (down_read). Thread B holds the page_table_lock while the page's map count and PMD table entry are updated. However, Thread A does not synchronize on that lock. ====== end quote ======= [akpm@linux-foundation.org: checkpatch fixes] Reported-by: Ulrich Obergfell <uobergfe@redhat.com> Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Hugh Dickins <hughd@google.com> Cc: Dave Jones <davej@redhat.com> Acked-by: Larry Woodman <lwoodman@redhat.com> Acked-by: Rik van Riel <riel@redhat.com> Cc: <stable@vger.kernel.org> [2.6.38+] Cc: Mark Salter <msalter@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-03-21 23:33:42 +00:00
down_write(&mm->mmap_sem);
pgd = pgd_offset(mm, 0xA0000);
if (pgd_none_or_clear_bad(pgd))
goto out;
pud = pud_offset(pgd, 0xA0000);
if (pud_none_or_clear_bad(pud))
goto out;
pmd = pmd_offset(pud, 0xA0000);
split_huge_page_pmd_mm(mm, 0xA0000, pmd);
if (pmd_none_or_clear_bad(pmd))
goto out;
pte = pte_offset_map_lock(mm, pmd, 0xA0000, &ptl);
for (i = 0; i < 32; i++) {
if (pte_present(*pte))
set_pte(pte, pte_wrprotect(*pte));
pte++;
}
pte_unmap_unlock(pte, ptl);
out:
mm: thp: fix pmd_bad() triggering in code paths holding mmap_sem read mode In some cases it may happen that pmd_none_or_clear_bad() is called with the mmap_sem hold in read mode. In those cases the huge page faults can allocate hugepmds under pmd_none_or_clear_bad() and that can trigger a false positive from pmd_bad() that will not like to see a pmd materializing as trans huge. It's not khugepaged causing the problem, khugepaged holds the mmap_sem in write mode (and all those sites must hold the mmap_sem in read mode to prevent pagetables to go away from under them, during code review it seems vm86 mode on 32bit kernels requires that too unless it's restricted to 1 thread per process or UP builds). The race is only with the huge pagefaults that can convert a pmd_none() into a pmd_trans_huge(). Effectively all these pmd_none_or_clear_bad() sites running with mmap_sem in read mode are somewhat speculative with the page faults, and the result is always undefined when they run simultaneously. This is probably why it wasn't common to run into this. For example if the madvise(MADV_DONTNEED) runs zap_page_range() shortly before the page fault, the hugepage will not be zapped, if the page fault runs first it will be zapped. Altering pmd_bad() not to error out if it finds hugepmds won't be enough to fix this, because zap_pmd_range would then proceed to call zap_pte_range (which would be incorrect if the pmd become a pmd_trans_huge()). The simplest way to fix this is to read the pmd in the local stack (regardless of what we read, no need of actual CPU barriers, only compiler barrier needed), and be sure it is not changing under the code that computes its value. Even if the real pmd is changing under the value we hold on the stack, we don't care. If we actually end up in zap_pte_range it means the pmd was not none already and it was not huge, and it can't become huge from under us (khugepaged locking explained above). All we need is to enforce that there is no way anymore that in a code path like below, pmd_trans_huge can be false, but pmd_none_or_clear_bad can run into a hugepmd. The overhead of a barrier() is just a compiler tweak and should not be measurable (I only added it for THP builds). I don't exclude different compiler versions may have prevented the race too by caching the value of *pmd on the stack (that hasn't been verified, but it wouldn't be impossible considering pmd_none_or_clear_bad, pmd_bad, pmd_trans_huge, pmd_none are all inlines and there's no external function called in between pmd_trans_huge and pmd_none_or_clear_bad). if (pmd_trans_huge(*pmd)) { if (next-addr != HPAGE_PMD_SIZE) { VM_BUG_ON(!rwsem_is_locked(&tlb->mm->mmap_sem)); split_huge_page_pmd(vma->vm_mm, pmd); } else if (zap_huge_pmd(tlb, vma, pmd, addr)) continue; /* fall through */ } if (pmd_none_or_clear_bad(pmd)) Because this race condition could be exercised without special privileges this was reported in CVE-2012-1179. The race was identified and fully explained by Ulrich who debugged it. I'm quoting his accurate explanation below, for reference. ====== start quote ======= mapcount 0 page_mapcount 1 kernel BUG at mm/huge_memory.c:1384! At some point prior to the panic, a "bad pmd ..." message similar to the following is logged on the console: mm/memory.c:145: bad pmd ffff8800376e1f98(80000000314000e7). The "bad pmd ..." message is logged by pmd_clear_bad() before it clears the page's PMD table entry. 143 void pmd_clear_bad(pmd_t *pmd) 144 { -> 145 pmd_ERROR(*pmd); 146 pmd_clear(pmd); 147 } After the PMD table entry has been cleared, there is an inconsistency between the actual number of PMD table entries that are mapping the page and the page's map count (_mapcount field in struct page). When the page is subsequently reclaimed, __split_huge_page() detects this inconsistency. 1381 if (mapcount != page_mapcount(page)) 1382 printk(KERN_ERR "mapcount %d page_mapcount %d\n", 1383 mapcount, page_mapcount(page)); -> 1384 BUG_ON(mapcount != page_mapcount(page)); The root cause of the problem is a race of two threads in a multithreaded process. Thread B incurs a page fault on a virtual address that has never been accessed (PMD entry is zero) while Thread A is executing an madvise() system call on a virtual address within the same 2 MB (huge page) range. virtual address space .---------------------. | | | | .-|---------------------| | | | | | |<-- B(fault) | | | 2 MB | |/////////////////////|-. huge < |/////////////////////| > A(range) page | |/////////////////////|-' | | | | | | '-|---------------------| | | | | '---------------------' - Thread A is executing an madvise(..., MADV_DONTNEED) system call on the virtual address range "A(range)" shown in the picture. sys_madvise // Acquire the semaphore in shared mode. down_read(&current->mm->mmap_sem) ... madvise_vma switch (behavior) case MADV_DONTNEED: madvise_dontneed zap_page_range unmap_vmas unmap_page_range zap_pud_range zap_pmd_range // // Assume that this huge page has never been accessed. // I.e. content of the PMD entry is zero (not mapped). // if (pmd_trans_huge(*pmd)) { // We don't get here due to the above assumption. } // // Assume that Thread B incurred a page fault and .---------> // sneaks in here as shown below. | // | if (pmd_none_or_clear_bad(pmd)) | { | if (unlikely(pmd_bad(*pmd))) | pmd_clear_bad | { | pmd_ERROR | // Log "bad pmd ..." message here. | pmd_clear | // Clear the page's PMD entry. | // Thread B incremented the map count | // in page_add_new_anon_rmap(), but | // now the page is no longer mapped | // by a PMD entry (-> inconsistency). | } | } | v - Thread B is handling a page fault on virtual address "B(fault)" shown in the picture. ... do_page_fault __do_page_fault // Acquire the semaphore in shared mode. down_read_trylock(&mm->mmap_sem) ... handle_mm_fault if (pmd_none(*pmd) && transparent_hugepage_enabled(vma)) // We get here due to the above assumption (PMD entry is zero). do_huge_pmd_anonymous_page alloc_hugepage_vma // Allocate a new transparent huge page here. ... __do_huge_pmd_anonymous_page ... spin_lock(&mm->page_table_lock) ... page_add_new_anon_rmap // Here we increment the page's map count (starts at -1). atomic_set(&page->_mapcount, 0) set_pmd_at // Here we set the page's PMD entry which will be cleared // when Thread A calls pmd_clear_bad(). ... spin_unlock(&mm->page_table_lock) The mmap_sem does not prevent the race because both threads are acquiring it in shared mode (down_read). Thread B holds the page_table_lock while the page's map count and PMD table entry are updated. However, Thread A does not synchronize on that lock. ====== end quote ======= [akpm@linux-foundation.org: checkpatch fixes] Reported-by: Ulrich Obergfell <uobergfe@redhat.com> Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Acked-by: Johannes Weiner <hannes@cmpxchg.org> Cc: Mel Gorman <mgorman@suse.de> Cc: Hugh Dickins <hughd@google.com> Cc: Dave Jones <davej@redhat.com> Acked-by: Larry Woodman <lwoodman@redhat.com> Acked-by: Rik van Riel <riel@redhat.com> Cc: <stable@vger.kernel.org> [2.6.38+] Cc: Mark Salter <msalter@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-03-21 23:33:42 +00:00
up_write(&mm->mmap_sem);
flush_tlb();
}
static int do_vm86_irq_handling(int subfunction, int irqnumber);
static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk);
SYSCALL_DEFINE1(vm86old, struct vm86_struct __user *, v86)
{
struct kernel_vm86_struct info; /* declare this _on top_,
* this avoids wasting of stack space.
* This remains on the stack until we
* return to 32 bit user space.
*/
struct task_struct *tsk = current;
int tmp;
if (tsk->thread.saved_sp0)
return -EPERM;
tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs,
offsetof(struct kernel_vm86_struct, vm86plus) -
sizeof(info.regs));
if (tmp)
return -EFAULT;
memset(&info.vm86plus, 0, (int)&info.regs32 - (int)&info.vm86plus);
info.regs32 = current_pt_regs();
tsk->thread.vm86_info = v86;
do_sys_vm86(&info, tsk);
return 0; /* we never return here */
}
SYSCALL_DEFINE2(vm86, unsigned long, cmd, unsigned long, arg)
{
struct kernel_vm86_struct info; /* declare this _on top_,
* this avoids wasting of stack space.
* This remains on the stack until we
* return to 32 bit user space.
*/
struct task_struct *tsk;
int tmp;
struct vm86plus_struct __user *v86;
tsk = current;
switch (cmd) {
case VM86_REQUEST_IRQ:
case VM86_FREE_IRQ:
case VM86_GET_IRQ_BITS:
case VM86_GET_AND_RESET_IRQ:
return do_vm86_irq_handling(cmd, (int)arg);
case VM86_PLUS_INSTALL_CHECK:
/*
* NOTE: on old vm86 stuff this will return the error
* from access_ok(), because the subfunction is
* interpreted as (invalid) address to vm86_struct.
* So the installation check works.
*/
return 0;
}
/* we come here only for functions VM86_ENTER, VM86_ENTER_NO_BYPASS */
if (tsk->thread.saved_sp0)
return -EPERM;
v86 = (struct vm86plus_struct __user *)arg;
tmp = copy_vm86_regs_from_user(&info.regs, &v86->regs,
offsetof(struct kernel_vm86_struct, regs32) -
sizeof(info.regs));
if (tmp)
return -EFAULT;
info.regs32 = current_pt_regs();
info.vm86plus.is_vm86pus = 1;
tsk->thread.vm86_info = (struct vm86_struct __user *)v86;
do_sys_vm86(&info, tsk);
return 0; /* we never return here */
}
static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk)
{
struct tss_struct *tss;
/*
* make sure the vm86() system call doesn't try to do anything silly
*/
info->regs.pt.ds = 0;
info->regs.pt.es = 0;
info->regs.pt.fs = 0;
#ifndef CONFIG_X86_32_LAZY_GS
info->regs.pt.gs = 0;
#endif
/*
* The flags register is also special: we cannot trust that the user
* has set it up safely, so this makes sure interrupt etc flags are
* inherited from protected mode.
*/
VEFLAGS = info->regs.pt.flags;
info->regs.pt.flags &= SAFE_MASK;
info->regs.pt.flags |= info->regs32->flags & ~SAFE_MASK;
info->regs.pt.flags |= X86_VM_MASK;
switch (info->cpu_type) {
case CPU_286:
tsk->thread.v86mask = 0;
break;
case CPU_386:
tsk->thread.v86mask = X86_EFLAGS_NT | X86_EFLAGS_IOPL;
break;
case CPU_486:
tsk->thread.v86mask = X86_EFLAGS_AC | X86_EFLAGS_NT | X86_EFLAGS_IOPL;
break;
default:
tsk->thread.v86mask = X86_EFLAGS_ID | X86_EFLAGS_AC | X86_EFLAGS_NT | X86_EFLAGS_IOPL;
break;
}
/*
* Save old state, set default return value (%ax) to 0 (VM86_SIGNAL)
*/
info->regs32->ax = VM86_SIGNAL;
tsk->thread.saved_sp0 = tsk->thread.sp0;
tsk->thread.saved_fs = info->regs32->fs;
tsk->thread.saved_gs = get_user_gs(info->regs32);
tss = &per_cpu(init_tss, get_cpu());
tsk->thread.sp0 = (unsigned long) &info->VM86_TSS_ESP0;
if (cpu_has_sep)
tsk->thread.sysenter_cs = 0;
load_sp0(tss, &tsk->thread);
put_cpu();
tsk->thread.screen_bitmap = info->screen_bitmap;
if (info->flags & VM86_SCREEN_BITMAP)
mark_screen_rdonly(tsk->mm);
[PATCH] make vm86 call audit_syscall_exit hi, The motivation behind the patch below was to address messages in /var/log/messages such as: Jan 31 10:54:15 mets kernel: audit(:0): major=252 name_count=0: freeing multiple contexts (1) Jan 31 10:54:15 mets kernel: audit(:0): major=113 name_count=0: freeing multiple contexts (2) I can reproduce by running 'get-edid' from: http://john.fremlin.de/programs/linux/read-edid/. These messages come about in the log b/c the vm86 calls do not exit via the normal system call exit paths and thus do not call 'audit_syscall_exit'. The next system call will then free the context for itself and for the vm86 context, thus generating the above messages. This patch addresses the issue by simply adding a call to 'audit_syscall_exit' from the vm86 code. Besides fixing the above error messages the patch also now allows vm86 system calls to become auditable. This is useful since strace does not appear to properly record the return values from sys_vm86. I think this patch is also a step in the right direction in terms of cleaning up some core auditing code. If we can correct any other paths that do not properly call the audit exit and entries points, then we can also eliminate the notion of context chaining. I've tested this patch by verifying that the log messages no longer appear, and that the audit records for sys_vm86 appear to be correct. Also, 'read_edid' produces itentical output. thanks, -Jason Signed-off-by: Jason Baron <jbaron@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-01-31 21:56:28 +00:00
Audit: push audit success and retcode into arch ptrace.h The audit system previously expected arches calling to audit_syscall_exit to supply as arguments if the syscall was a success and what the return code was. Audit also provides a helper AUDITSC_RESULT which was supposed to simplify things by converting from negative retcodes to an audit internal magic value stating success or failure. This helper was wrong and could indicate that a valid pointer returned to userspace was a failed syscall. The fix is to fix the layering foolishness. We now pass audit_syscall_exit a struct pt_reg and it in turns calls back into arch code to collect the return value and to determine if the syscall was a success or failure. We also define a generic is_syscall_success() macro which determines success/failure based on if the value is < -MAX_ERRNO. This works for arches like x86 which do not use a separate mechanism to indicate syscall failure. We make both the is_syscall_success() and regs_return_value() static inlines instead of macros. The reason is because the audit function must take a void* for the regs. (uml calls theirs struct uml_pt_regs instead of just struct pt_regs so audit_syscall_exit can't take a struct pt_regs). Since the audit function takes a void* we need to use static inlines to cast it back to the arch correct structure to dereference it. The other major change is that on some arches, like ia64, MIPS and ppc, we change regs_return_value() to give us the negative value on syscall failure. THE only other user of this macro, kretprobe_example.c, won't notice and it makes the value signed consistently for the audit functions across all archs. In arch/sh/kernel/ptrace_64.c I see that we were using regs[9] in the old audit code as the return value. But the ptrace_64.h code defined the macro regs_return_value() as regs[3]. I have no idea which one is correct, but this patch now uses the regs_return_value() function, so it now uses regs[3]. For powerpc we previously used regs->result but now use the regs_return_value() function which uses regs->gprs[3]. regs->gprs[3] is always positive so the regs_return_value(), much like ia64 makes it negative before calling the audit code when appropriate. Signed-off-by: Eric Paris <eparis@redhat.com> Acked-by: H. Peter Anvin <hpa@zytor.com> [for x86 portion] Acked-by: Tony Luck <tony.luck@intel.com> [for ia64] Acked-by: Richard Weinberger <richard@nod.at> [for uml] Acked-by: David S. Miller <davem@davemloft.net> [for sparc] Acked-by: Ralf Baechle <ralf@linux-mips.org> [for mips] Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> [for ppc]
2012-01-03 19:23:06 +00:00
/*call __audit_syscall_exit since we do not exit via the normal paths */
#ifdef CONFIG_AUDITSYSCALL
[PATCH] make vm86 call audit_syscall_exit hi, The motivation behind the patch below was to address messages in /var/log/messages such as: Jan 31 10:54:15 mets kernel: audit(:0): major=252 name_count=0: freeing multiple contexts (1) Jan 31 10:54:15 mets kernel: audit(:0): major=113 name_count=0: freeing multiple contexts (2) I can reproduce by running 'get-edid' from: http://john.fremlin.de/programs/linux/read-edid/. These messages come about in the log b/c the vm86 calls do not exit via the normal system call exit paths and thus do not call 'audit_syscall_exit'. The next system call will then free the context for itself and for the vm86 context, thus generating the above messages. This patch addresses the issue by simply adding a call to 'audit_syscall_exit' from the vm86 code. Besides fixing the above error messages the patch also now allows vm86 system calls to become auditable. This is useful since strace does not appear to properly record the return values from sys_vm86. I think this patch is also a step in the right direction in terms of cleaning up some core auditing code. If we can correct any other paths that do not properly call the audit exit and entries points, then we can also eliminate the notion of context chaining. I've tested this patch by verifying that the log messages no longer appear, and that the audit records for sys_vm86 appear to be correct. Also, 'read_edid' produces itentical output. thanks, -Jason Signed-off-by: Jason Baron <jbaron@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-01-31 21:56:28 +00:00
if (unlikely(current->audit_context))
Audit: push audit success and retcode into arch ptrace.h The audit system previously expected arches calling to audit_syscall_exit to supply as arguments if the syscall was a success and what the return code was. Audit also provides a helper AUDITSC_RESULT which was supposed to simplify things by converting from negative retcodes to an audit internal magic value stating success or failure. This helper was wrong and could indicate that a valid pointer returned to userspace was a failed syscall. The fix is to fix the layering foolishness. We now pass audit_syscall_exit a struct pt_reg and it in turns calls back into arch code to collect the return value and to determine if the syscall was a success or failure. We also define a generic is_syscall_success() macro which determines success/failure based on if the value is < -MAX_ERRNO. This works for arches like x86 which do not use a separate mechanism to indicate syscall failure. We make both the is_syscall_success() and regs_return_value() static inlines instead of macros. The reason is because the audit function must take a void* for the regs. (uml calls theirs struct uml_pt_regs instead of just struct pt_regs so audit_syscall_exit can't take a struct pt_regs). Since the audit function takes a void* we need to use static inlines to cast it back to the arch correct structure to dereference it. The other major change is that on some arches, like ia64, MIPS and ppc, we change regs_return_value() to give us the negative value on syscall failure. THE only other user of this macro, kretprobe_example.c, won't notice and it makes the value signed consistently for the audit functions across all archs. In arch/sh/kernel/ptrace_64.c I see that we were using regs[9] in the old audit code as the return value. But the ptrace_64.h code defined the macro regs_return_value() as regs[3]. I have no idea which one is correct, but this patch now uses the regs_return_value() function, so it now uses regs[3]. For powerpc we previously used regs->result but now use the regs_return_value() function which uses regs->gprs[3]. regs->gprs[3] is always positive so the regs_return_value(), much like ia64 makes it negative before calling the audit code when appropriate. Signed-off-by: Eric Paris <eparis@redhat.com> Acked-by: H. Peter Anvin <hpa@zytor.com> [for x86 portion] Acked-by: Tony Luck <tony.luck@intel.com> [for ia64] Acked-by: Richard Weinberger <richard@nod.at> [for uml] Acked-by: David S. Miller <davem@davemloft.net> [for sparc] Acked-by: Ralf Baechle <ralf@linux-mips.org> [for mips] Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> [for ppc]
2012-01-03 19:23:06 +00:00
__audit_syscall_exit(1, 0);
#endif
[PATCH] make vm86 call audit_syscall_exit hi, The motivation behind the patch below was to address messages in /var/log/messages such as: Jan 31 10:54:15 mets kernel: audit(:0): major=252 name_count=0: freeing multiple contexts (1) Jan 31 10:54:15 mets kernel: audit(:0): major=113 name_count=0: freeing multiple contexts (2) I can reproduce by running 'get-edid' from: http://john.fremlin.de/programs/linux/read-edid/. These messages come about in the log b/c the vm86 calls do not exit via the normal system call exit paths and thus do not call 'audit_syscall_exit'. The next system call will then free the context for itself and for the vm86 context, thus generating the above messages. This patch addresses the issue by simply adding a call to 'audit_syscall_exit' from the vm86 code. Besides fixing the above error messages the patch also now allows vm86 system calls to become auditable. This is useful since strace does not appear to properly record the return values from sys_vm86. I think this patch is also a step in the right direction in terms of cleaning up some core auditing code. If we can correct any other paths that do not properly call the audit exit and entries points, then we can also eliminate the notion of context chaining. I've tested this patch by verifying that the log messages no longer appear, and that the audit records for sys_vm86 appear to be correct. Also, 'read_edid' produces itentical output. thanks, -Jason Signed-off-by: Jason Baron <jbaron@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2006-01-31 21:56:28 +00:00
__asm__ __volatile__(
"movl %0,%%esp\n\t"
"movl %1,%%ebp\n\t"
#ifdef CONFIG_X86_32_LAZY_GS
"mov %2, %%gs\n\t"
#endif
"jmp resume_userspace"
: /* no outputs */
:"r" (&info->regs), "r" (task_thread_info(tsk)), "r" (0));
/* we never return here */
}
static inline void return_to_32bit(struct kernel_vm86_regs *regs16, int retval)
{
struct pt_regs *regs32;
regs32 = save_v86_state(regs16);
regs32->ax = retval;
__asm__ __volatile__("movl %0,%%esp\n\t"
"movl %1,%%ebp\n\t"
"jmp resume_userspace"
: : "r" (regs32), "r" (current_thread_info()));
}
static inline void set_IF(struct kernel_vm86_regs *regs)
{
VEFLAGS |= X86_EFLAGS_VIF;
if (VEFLAGS & X86_EFLAGS_VIP)
return_to_32bit(regs, VM86_STI);
}
static inline void clear_IF(struct kernel_vm86_regs *regs)
{
VEFLAGS &= ~X86_EFLAGS_VIF;
}
static inline void clear_TF(struct kernel_vm86_regs *regs)
{
regs->pt.flags &= ~X86_EFLAGS_TF;
}
static inline void clear_AC(struct kernel_vm86_regs *regs)
{
regs->pt.flags &= ~X86_EFLAGS_AC;
}
/*
* It is correct to call set_IF(regs) from the set_vflags_*
* functions. However someone forgot to call clear_IF(regs)
* in the opposite case.
* After the command sequence CLI PUSHF STI POPF you should
* end up with interrupts disabled, but you ended up with
* interrupts enabled.
* ( I was testing my own changes, but the only bug I
* could find was in a function I had not changed. )
* [KD]
*/
static inline void set_vflags_long(unsigned long flags, struct kernel_vm86_regs *regs)
{
set_flags(VEFLAGS, flags, current->thread.v86mask);
set_flags(regs->pt.flags, flags, SAFE_MASK);
if (flags & X86_EFLAGS_IF)
set_IF(regs);
else
clear_IF(regs);
}
static inline void set_vflags_short(unsigned short flags, struct kernel_vm86_regs *regs)
{
set_flags(VFLAGS, flags, current->thread.v86mask);
set_flags(regs->pt.flags, flags, SAFE_MASK);
if (flags & X86_EFLAGS_IF)
set_IF(regs);
else
clear_IF(regs);
}
static inline unsigned long get_vflags(struct kernel_vm86_regs *regs)
{
unsigned long flags = regs->pt.flags & RETURN_MASK;
if (VEFLAGS & X86_EFLAGS_VIF)
flags |= X86_EFLAGS_IF;
flags |= X86_EFLAGS_IOPL;
return flags | (VEFLAGS & current->thread.v86mask);
}
static inline int is_revectored(int nr, struct revectored_struct *bitmap)
{
__asm__ __volatile__("btl %2,%1\n\tsbbl %0,%0"
:"=r" (nr)
:"m" (*bitmap), "r" (nr));
return nr;
}
#define val_byte(val, n) (((__u8 *)&val)[n])
#define pushb(base, ptr, val, err_label) \
do { \
__u8 __val = val; \
ptr--; \
if (put_user(__val, base + ptr) < 0) \
goto err_label; \
} while (0)
#define pushw(base, ptr, val, err_label) \
do { \
__u16 __val = val; \
ptr--; \
if (put_user(val_byte(__val, 1), base + ptr) < 0) \
goto err_label; \
ptr--; \
if (put_user(val_byte(__val, 0), base + ptr) < 0) \
goto err_label; \
} while (0)
#define pushl(base, ptr, val, err_label) \
do { \
__u32 __val = val; \
ptr--; \
if (put_user(val_byte(__val, 3), base + ptr) < 0) \
goto err_label; \
ptr--; \
if (put_user(val_byte(__val, 2), base + ptr) < 0) \
goto err_label; \
ptr--; \
if (put_user(val_byte(__val, 1), base + ptr) < 0) \
goto err_label; \
ptr--; \
if (put_user(val_byte(__val, 0), base + ptr) < 0) \
goto err_label; \
} while (0)
#define popb(base, ptr, err_label) \
({ \
__u8 __res; \
if (get_user(__res, base + ptr) < 0) \
goto err_label; \
ptr++; \
__res; \
})
#define popw(base, ptr, err_label) \
({ \
__u16 __res; \
if (get_user(val_byte(__res, 0), base + ptr) < 0) \
goto err_label; \
ptr++; \
if (get_user(val_byte(__res, 1), base + ptr) < 0) \
goto err_label; \
ptr++; \
__res; \
})
#define popl(base, ptr, err_label) \
({ \
__u32 __res; \
if (get_user(val_byte(__res, 0), base + ptr) < 0) \
goto err_label; \
ptr++; \
if (get_user(val_byte(__res, 1), base + ptr) < 0) \
goto err_label; \
ptr++; \
if (get_user(val_byte(__res, 2), base + ptr) < 0) \
goto err_label; \
ptr++; \
if (get_user(val_byte(__res, 3), base + ptr) < 0) \
goto err_label; \
ptr++; \
__res; \
})
/* There are so many possible reasons for this function to return
* VM86_INTx, so adding another doesn't bother me. We can expect
* userspace programs to be able to handle it. (Getting a problem
* in userspace is always better than an Oops anyway.) [KD]
*/
static void do_int(struct kernel_vm86_regs *regs, int i,
unsigned char __user *ssp, unsigned short sp)
{
unsigned long __user *intr_ptr;
unsigned long segoffs;
if (regs->pt.cs == BIOSSEG)
goto cannot_handle;
if (is_revectored(i, &KVM86->int_revectored))
goto cannot_handle;
if (i == 0x21 && is_revectored(AH(regs), &KVM86->int21_revectored))
goto cannot_handle;
intr_ptr = (unsigned long __user *) (i << 2);
if (get_user(segoffs, intr_ptr))
goto cannot_handle;
if ((segoffs >> 16) == BIOSSEG)
goto cannot_handle;
pushw(ssp, sp, get_vflags(regs), cannot_handle);
pushw(ssp, sp, regs->pt.cs, cannot_handle);
pushw(ssp, sp, IP(regs), cannot_handle);
regs->pt.cs = segoffs >> 16;
SP(regs) -= 6;
IP(regs) = segoffs & 0xffff;
clear_TF(regs);
clear_IF(regs);
clear_AC(regs);
return;
cannot_handle:
return_to_32bit(regs, VM86_INTx + (i << 8));
}
int handle_vm86_trap(struct kernel_vm86_regs *regs, long error_code, int trapno)
{
if (VMPI.is_vm86pus) {
x86, vm86: Fix preemption bug for int1 debug and int3 breakpoint handlers. Impact: fix kernel bug such as: BUG: scheduling while atomic: dosemu.bin/19680/0x00000004 See also Ubuntu bug 455067 at https://bugs.launchpad.net/ubuntu/+source/linux/+bug/455067 Commits 4915a35e35a037254550a2ba9f367a812bc37d40 ("Use preempt_conditional_sti/cli in do_int3, like on x86_64.") and 3d2a71a596bd9c761c8487a2178e95f8a61da083 ("x86, traps: converge do_debug handlers") started disabling preemption in int1 and int3 handlers on i386. The problem with vm86 is that the call to handle_vm86_trap() may jump straight to entry_32.S and never returns so preempt is never enabled again, and there is an imbalance in the preempt count. Commit be716615fe596ee117292dc615e95f707fb67fd1 ("x86, vm86: fix preemption bug"), which was later (accidentally?) reverted by commit 08d68323d1f0c34452e614263b212ca556dae47f ("hw-breakpoints: modifying generic debug exception to use thread-specific debug registers") fixed the problem for debug exceptions but not for breakpoints. There are three solutions to this problem. 1. Reenable preemption before calling handle_vm86_trap(). This was the approach that was later reverted. 2. Do not disable preemption for i386 in breakpoint and debug handlers. This was the situation before October 2008. As far as I understand preemption only needs to be disabled on x86_64 because a seperate stack is used, but it's nice to have things work the same way on i386 and x86_64. 3. Let handle_vm86_trap() return instead of jumping to assembly code. By setting a flag in _TIF_WORK_MASK, either TIF_IRET or TIF_NOTIFY_RESUME, the code in entry_32.S is instructed to return to 32 bit mode from V86 mode. The logic in entry_32.S was already present to handle signals. (I chose TIF_IRET because it's slightly more efficient in do_notify_resume() in signal.c, but in fact TIF_IRET can probably be replaced by TIF_NOTIFY_RESUME everywhere.) I'm submitting approach 3, because I believe it is the most elegant and prevents future confusion. Still, an obvious preempt_conditional_cli(regs); is necessary in traps.c to correct the bug. [ hpa: This is technically a regression, but because: 1. the regression is so old, 2. the patch seems relatively high risk, justifying more testing, and 3. we're late in the 2.6.36-rc cycle, I'm queuing it up for the 2.6.37 merge window. It might, however, justify as a -stable backport at a latter time, hence Cc: stable. ] Signed-off-by: Bart Oldeman <bartoldeman@users.sourceforge.net> LKML-Reference: <alpine.DEB.2.00.1009231312330.4732@localhost.localdomain> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: K.Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Alexander van Heukelum <heukelum@fastmail.fm> Cc: <stable@kernel.org> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2010-09-23 17:16:58 +00:00
if ((trapno == 3) || (trapno == 1)) {
KVM86->regs32->ax = VM86_TRAP + (trapno << 8);
/* setting this flag forces the code in entry_32.S to
the path where we call save_v86_state() and change
the stack pointer to KVM86->regs32 */
set_thread_flag(TIF_NOTIFY_RESUME);
x86, vm86: Fix preemption bug for int1 debug and int3 breakpoint handlers. Impact: fix kernel bug such as: BUG: scheduling while atomic: dosemu.bin/19680/0x00000004 See also Ubuntu bug 455067 at https://bugs.launchpad.net/ubuntu/+source/linux/+bug/455067 Commits 4915a35e35a037254550a2ba9f367a812bc37d40 ("Use preempt_conditional_sti/cli in do_int3, like on x86_64.") and 3d2a71a596bd9c761c8487a2178e95f8a61da083 ("x86, traps: converge do_debug handlers") started disabling preemption in int1 and int3 handlers on i386. The problem with vm86 is that the call to handle_vm86_trap() may jump straight to entry_32.S and never returns so preempt is never enabled again, and there is an imbalance in the preempt count. Commit be716615fe596ee117292dc615e95f707fb67fd1 ("x86, vm86: fix preemption bug"), which was later (accidentally?) reverted by commit 08d68323d1f0c34452e614263b212ca556dae47f ("hw-breakpoints: modifying generic debug exception to use thread-specific debug registers") fixed the problem for debug exceptions but not for breakpoints. There are three solutions to this problem. 1. Reenable preemption before calling handle_vm86_trap(). This was the approach that was later reverted. 2. Do not disable preemption for i386 in breakpoint and debug handlers. This was the situation before October 2008. As far as I understand preemption only needs to be disabled on x86_64 because a seperate stack is used, but it's nice to have things work the same way on i386 and x86_64. 3. Let handle_vm86_trap() return instead of jumping to assembly code. By setting a flag in _TIF_WORK_MASK, either TIF_IRET or TIF_NOTIFY_RESUME, the code in entry_32.S is instructed to return to 32 bit mode from V86 mode. The logic in entry_32.S was already present to handle signals. (I chose TIF_IRET because it's slightly more efficient in do_notify_resume() in signal.c, but in fact TIF_IRET can probably be replaced by TIF_NOTIFY_RESUME everywhere.) I'm submitting approach 3, because I believe it is the most elegant and prevents future confusion. Still, an obvious preempt_conditional_cli(regs); is necessary in traps.c to correct the bug. [ hpa: This is technically a regression, but because: 1. the regression is so old, 2. the patch seems relatively high risk, justifying more testing, and 3. we're late in the 2.6.36-rc cycle, I'm queuing it up for the 2.6.37 merge window. It might, however, justify as a -stable backport at a latter time, hence Cc: stable. ] Signed-off-by: Bart Oldeman <bartoldeman@users.sourceforge.net> LKML-Reference: <alpine.DEB.2.00.1009231312330.4732@localhost.localdomain> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: K.Prasad <prasad@linux.vnet.ibm.com> Cc: Alan Stern <stern@rowland.harvard.edu> Cc: Alexander van Heukelum <heukelum@fastmail.fm> Cc: <stable@kernel.org> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2010-09-23 17:16:58 +00:00
return 0;
}
do_int(regs, trapno, (unsigned char __user *) (regs->pt.ss << 4), SP(regs));
return 0;
}
if (trapno != 1)
return 1; /* we let this handle by the calling routine */
current->thread.trap_nr = trapno;
current->thread.error_code = error_code;
force_sig(SIGTRAP, current);
return 0;
}
void handle_vm86_fault(struct kernel_vm86_regs *regs, long error_code)
{
unsigned char opcode;
unsigned char __user *csp;
unsigned char __user *ssp;
unsigned short ip, sp, orig_flags;
int data32, pref_done;
#define CHECK_IF_IN_TRAP \
if (VMPI.vm86dbg_active && VMPI.vm86dbg_TFpendig) \
newflags |= X86_EFLAGS_TF
#define VM86_FAULT_RETURN do { \
if (VMPI.force_return_for_pic && (VEFLAGS & (X86_EFLAGS_IF | X86_EFLAGS_VIF))) \
return_to_32bit(regs, VM86_PICRETURN); \
if (orig_flags & X86_EFLAGS_TF) \
handle_vm86_trap(regs, 0, 1); \
return; } while (0)
orig_flags = *(unsigned short *)&regs->pt.flags;
csp = (unsigned char __user *) (regs->pt.cs << 4);
ssp = (unsigned char __user *) (regs->pt.ss << 4);
sp = SP(regs);
ip = IP(regs);
data32 = 0;
pref_done = 0;
do {
switch (opcode = popb(csp, ip, simulate_sigsegv)) {
case 0x66: /* 32-bit data */ data32 = 1; break;
case 0x67: /* 32-bit address */ break;
case 0x2e: /* CS */ break;
case 0x3e: /* DS */ break;
case 0x26: /* ES */ break;
case 0x36: /* SS */ break;
case 0x65: /* GS */ break;
case 0x64: /* FS */ break;
case 0xf2: /* repnz */ break;
case 0xf3: /* rep */ break;
default: pref_done = 1;
}
} while (!pref_done);
switch (opcode) {
/* pushf */
case 0x9c:
if (data32) {
pushl(ssp, sp, get_vflags(regs), simulate_sigsegv);
SP(regs) -= 4;
} else {
pushw(ssp, sp, get_vflags(regs), simulate_sigsegv);
SP(regs) -= 2;
}
IP(regs) = ip;
VM86_FAULT_RETURN;
/* popf */
case 0x9d:
{
unsigned long newflags;
if (data32) {
newflags = popl(ssp, sp, simulate_sigsegv);
SP(regs) += 4;
} else {
newflags = popw(ssp, sp, simulate_sigsegv);
SP(regs) += 2;
}
IP(regs) = ip;
CHECK_IF_IN_TRAP;
if (data32)
set_vflags_long(newflags, regs);
else
set_vflags_short(newflags, regs);
VM86_FAULT_RETURN;
}
/* int xx */
case 0xcd: {
int intno = popb(csp, ip, simulate_sigsegv);
IP(regs) = ip;
if (VMPI.vm86dbg_active) {
if ((1 << (intno & 7)) & VMPI.vm86dbg_intxxtab[intno >> 3])
return_to_32bit(regs, VM86_INTx + (intno << 8));
}
do_int(regs, intno, ssp, sp);
return;
}
/* iret */
case 0xcf:
{
unsigned long newip;
unsigned long newcs;
unsigned long newflags;
if (data32) {
newip = popl(ssp, sp, simulate_sigsegv);
newcs = popl(ssp, sp, simulate_sigsegv);
newflags = popl(ssp, sp, simulate_sigsegv);
SP(regs) += 12;
} else {
newip = popw(ssp, sp, simulate_sigsegv);
newcs = popw(ssp, sp, simulate_sigsegv);
newflags = popw(ssp, sp, simulate_sigsegv);
SP(regs) += 6;
}
IP(regs) = newip;
regs->pt.cs = newcs;
CHECK_IF_IN_TRAP;
if (data32) {
set_vflags_long(newflags, regs);
} else {
set_vflags_short(newflags, regs);
}
VM86_FAULT_RETURN;
}
/* cli */
case 0xfa:
IP(regs) = ip;
clear_IF(regs);
VM86_FAULT_RETURN;
/* sti */
/*
* Damn. This is incorrect: the 'sti' instruction should actually
* enable interrupts after the /next/ instruction. Not good.
*
* Probably needs some horsing around with the TF flag. Aiee..
*/
case 0xfb:
IP(regs) = ip;
set_IF(regs);
VM86_FAULT_RETURN;
default:
return_to_32bit(regs, VM86_UNKNOWN);
}
return;
simulate_sigsegv:
/* FIXME: After a long discussion with Stas we finally
* agreed, that this is wrong. Here we should
* really send a SIGSEGV to the user program.
* But how do we create the correct context? We
* are inside a general protection fault handler
* and has just returned from a page fault handler.
* The correct context for the signal handler
* should be a mixture of the two, but how do we
* get the information? [KD]
*/
return_to_32bit(regs, VM86_UNKNOWN);
}
/* ---------------- vm86 special IRQ passing stuff ----------------- */
#define VM86_IRQNAME "vm86irq"
static struct vm86_irqs {
struct task_struct *tsk;
int sig;
} vm86_irqs[16];
static DEFINE_SPINLOCK(irqbits_lock);
static int irqbits;
#define ALLOWED_SIGS (1 /* 0 = don't send a signal */ \
| (1 << SIGUSR1) | (1 << SIGUSR2) | (1 << SIGIO) | (1 << SIGURG) \
| (1 << SIGUNUSED))
IRQ: Maintain regs pointer globally rather than passing to IRQ handlers Maintain a per-CPU global "struct pt_regs *" variable which can be used instead of passing regs around manually through all ~1800 interrupt handlers in the Linux kernel. The regs pointer is used in few places, but it potentially costs both stack space and code to pass it around. On the FRV arch, removing the regs parameter from all the genirq function results in a 20% speed up of the IRQ exit path (ie: from leaving timer_interrupt() to leaving do_IRQ()). Where appropriate, an arch may override the generic storage facility and do something different with the variable. On FRV, for instance, the address is maintained in GR28 at all times inside the kernel as part of general exception handling. Having looked over the code, it appears that the parameter may be handed down through up to twenty or so layers of functions. Consider a USB character device attached to a USB hub, attached to a USB controller that posts its interrupts through a cascaded auxiliary interrupt controller. A character device driver may want to pass regs to the sysrq handler through the input layer which adds another few layers of parameter passing. I've build this code with allyesconfig for x86_64 and i386. I've runtested the main part of the code on FRV and i386, though I can't test most of the drivers. I've also done partial conversion for powerpc and MIPS - these at least compile with minimal configurations. This will affect all archs. Mostly the changes should be relatively easy. Take do_IRQ(), store the regs pointer at the beginning, saving the old one: struct pt_regs *old_regs = set_irq_regs(regs); And put the old one back at the end: set_irq_regs(old_regs); Don't pass regs through to generic_handle_irq() or __do_IRQ(). In timer_interrupt(), this sort of change will be necessary: - update_process_times(user_mode(regs)); - profile_tick(CPU_PROFILING, regs); + update_process_times(user_mode(get_irq_regs())); + profile_tick(CPU_PROFILING); I'd like to move update_process_times()'s use of get_irq_regs() into itself, except that i386, alone of the archs, uses something other than user_mode(). Some notes on the interrupt handling in the drivers: (*) input_dev() is now gone entirely. The regs pointer is no longer stored in the input_dev struct. (*) finish_unlinks() in drivers/usb/host/ohci-q.c needs checking. It does something different depending on whether it's been supplied with a regs pointer or not. (*) Various IRQ handler function pointers have been moved to type irq_handler_t. Signed-Off-By: David Howells <dhowells@redhat.com> (cherry picked from 1b16e7ac850969f38b375e511e3fa2f474a33867 commit)
2006-10-05 13:55:46 +00:00
static irqreturn_t irq_handler(int intno, void *dev_id)
{
int irq_bit;
unsigned long flags;
spin_lock_irqsave(&irqbits_lock, flags);
irq_bit = 1 << intno;
if ((irqbits & irq_bit) || !vm86_irqs[intno].tsk)
goto out;
irqbits |= irq_bit;
if (vm86_irqs[intno].sig)
send_sig(vm86_irqs[intno].sig, vm86_irqs[intno].tsk, 1);
/*
* IRQ will be re-enabled when user asks for the irq (whether
* polling or as a result of the signal)
*/
disable_irq_nosync(intno);
spin_unlock_irqrestore(&irqbits_lock, flags);
return IRQ_HANDLED;
out:
spin_unlock_irqrestore(&irqbits_lock, flags);
return IRQ_NONE;
}
static inline void free_vm86_irq(int irqnumber)
{
unsigned long flags;
free_irq(irqnumber, NULL);
vm86_irqs[irqnumber].tsk = NULL;
spin_lock_irqsave(&irqbits_lock, flags);
irqbits &= ~(1 << irqnumber);
spin_unlock_irqrestore(&irqbits_lock, flags);
}
void release_vm86_irqs(struct task_struct *task)
{
int i;
for (i = FIRST_VM86_IRQ ; i <= LAST_VM86_IRQ; i++)
if (vm86_irqs[i].tsk == task)
free_vm86_irq(i);
}
static inline int get_and_reset_irq(int irqnumber)
{
int bit;
unsigned long flags;
int ret = 0;
if (invalid_vm86_irq(irqnumber)) return 0;
if (vm86_irqs[irqnumber].tsk != current) return 0;
spin_lock_irqsave(&irqbits_lock, flags);
bit = irqbits & (1 << irqnumber);
irqbits &= ~bit;
if (bit) {
enable_irq(irqnumber);
ret = 1;
}
spin_unlock_irqrestore(&irqbits_lock, flags);
return ret;
}
static int do_vm86_irq_handling(int subfunction, int irqnumber)
{
int ret;
switch (subfunction) {
case VM86_GET_AND_RESET_IRQ: {
return get_and_reset_irq(irqnumber);
}
case VM86_GET_IRQ_BITS: {
return irqbits;
}
case VM86_REQUEST_IRQ: {
int sig = irqnumber >> 8;
int irq = irqnumber & 255;
if (!capable(CAP_SYS_ADMIN)) return -EPERM;
if (!((1 << sig) & ALLOWED_SIGS)) return -EPERM;
if (invalid_vm86_irq(irq)) return -EPERM;
if (vm86_irqs[irq].tsk) return -EPERM;
ret = request_irq(irq, &irq_handler, 0, VM86_IRQNAME, NULL);
if (ret) return ret;
vm86_irqs[irq].sig = sig;
vm86_irqs[irq].tsk = current;
return irq;
}
case VM86_FREE_IRQ: {
if (invalid_vm86_irq(irqnumber)) return -EPERM;
if (!vm86_irqs[irqnumber].tsk) return 0;
if (vm86_irqs[irqnumber].tsk != current) return -EPERM;
free_vm86_irq(irqnumber);
return 0;
}
}
return -EINVAL;
}