From: Andy Lutomirski <luto@kernel.org>
To: x86@kernel.org, linux-kernel@vger.kernel.org
Cc: "Frédéric Weisbecker" <fweisbec@gmail.com>,
"Rik van Riel" <riel@redhat.com>,
"Oleg Nesterov" <oleg@redhat.com>,
"Denys Vlasenko" <vda.linux@googlemail.com>,
"Borislav Petkov" <bp@alien8.de>,
"Kees Cook" <keescook@chromium.org>,
"Brian Gerst" <brgerst@gmail.com>,
"Linus Torvalds" <torvalds@linux-foundation.org>,
"Andy Lutomirski" <luto@kernel.org>
Subject: [RFC/PATCH 5/7] x86/vm86: Teach handle_vm86_trap to return to 32bit mode directly
Date: Wed, 8 Jul 2015 12:24:02 -0700 [thread overview]
Message-ID: <c3bfa8ab98867fe07521a55765b85b3d4b582579.1436383168.git.luto@kernel.org> (raw)
In-Reply-To: <cover.1436383168.git.luto@kernel.org>
In-Reply-To: <cover.1436383168.git.luto@kernel.org>
The TIF_NOTIFY_RESUME hack it was using was buggy and unsupportable.
vm86 mode was completely broken under ptrace, for example, because
we'd never make it to v8086 mode.
This code is still a huge, scary mess, but at least it's no longer
tangled with the exit-to-userspace loop.
Signed-off-by: Andy Lutomirski <luto@kernel.org>
---
arch/x86/entry/entry_32.S | 17 -----------------
arch/x86/kernel/traps.c | 12 ++++++++++++
arch/x86/kernel/vm86_32.c | 8 ++------
3 files changed, 14 insertions(+), 23 deletions(-)
diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index 66ff9c4055d7..3afd201bce55 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -481,29 +481,12 @@ work_resched:
work_notifysig: # deal with pending signals and
# notify-resume requests
-#ifdef CONFIG_VM86
- testl $X86_EFLAGS_VM, PT_EFLAGS(%esp)
- movl %esp, %eax
- jnz work_notifysig_v86 # special case for v86
-1:
-#else
movl %esp, %eax
-#endif
TRACE_IRQS_ON
ENABLE_INTERRUPTS(CLBR_NONE)
xorl %edx, %edx
call do_notify_resume
jmp resume_userspace
-
-#ifdef CONFIG_VM86
- ALIGN
-work_notifysig_v86:
- pushl %ecx # save ti_flags for do_notify_resume
- call save_v86_state # %eax contains pt_regs pointer
- popl %ecx
- movl %eax, %esp
- jmp 1b
-#endif
END(work_pending)
# perform syscall exit tracing
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 8e65d8a9b8db..f86172d4f5db 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -190,6 +190,13 @@ do_trap_no_signal(struct task_struct *tsk, int trapnr, char *str,
* On nmi (interrupt 2), do_trap should not be called.
*/
if (trapnr < X86_TRAP_UD) {
+ /*
+ * handle_vm86_trap may not return. If that
+ * happens, then there is no debug stack counter
+ * (it's a 32-bit kernel) and handle_vm86_trap
+ * will clear the preempt counter.
+ */
+
if (!handle_vm86_trap((struct kernel_vm86_regs *) regs,
error_code, trapnr))
return 0;
@@ -650,6 +657,11 @@ dotraplinkage void do_debug(struct pt_regs *regs, long error_code)
preempt_conditional_sti(regs);
if (v8086_mode(regs)) {
+ /*
+ * handle_vm86_trap may not return. If that happens, then
+ * there is no debug stack counter (it's a 32-bit kernel)
+ * and handle_vm86_trap will clear the preempt counter.
+ */
handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code,
X86_TRAP_DB);
preempt_conditional_cli(regs);
diff --git a/arch/x86/kernel/vm86_32.c b/arch/x86/kernel/vm86_32.c
index fc9db6ef2a95..c526ee34c22d 100644
--- a/arch/x86/kernel/vm86_32.c
+++ b/arch/x86/kernel/vm86_32.c
@@ -351,6 +351,7 @@ static inline void return_to_32bit(struct kernel_vm86_regs *regs16, int retval)
{
struct pt_regs *regs32;
+ preempt_count_set(0);
regs32 = save_v86_state(regs16);
regs32->ax = retval;
__asm__ __volatile__("movl %0,%%esp\n\t"
@@ -549,12 +550,7 @@ int handle_vm86_trap(struct kernel_vm86_regs *regs, long error_code, int trapno)
{
if (VMPI.is_vm86pus) {
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);
- return 0;
+ return_to_32bit(regs, VM86_TRAP + (trapno << 8));
}
do_int(regs, trapno, (unsigned char __user *) (regs->pt.ss << 4), SP(regs));
return 0;
--
2.4.3
next prev parent reply other threads:[~2015-07-08 19:25 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-07-08 19:23 [RFC/PATCH 0/7] x86_32: Migrate to new entry/exit paths Andy Lutomirski
2015-07-08 19:23 ` [RFC/PATCH 1/7] x86/entry/32: Remove 32-bit syscall audit optimizations Andy Lutomirski
2015-07-08 19:23 ` [RFC/PATCH 2/7] x86/entry/32: Fix an incorrect comment for work_notifysig_v86 Andy Lutomirski
2015-07-08 19:24 ` [RFC/PATCH 3/7] [TEMPORARY] x86/entry/32: Sanity check for work_notifysig Andy Lutomirski
2015-07-08 19:25 ` Andy Lutomirski
2015-07-08 19:24 ` [RFC/PATCH 4/7] x86/entry/32: Finish removing bogus kernel-mode check Andy Lutomirski
2015-07-08 19:25 ` Andy Lutomirski
2015-07-08 19:24 ` Andy Lutomirski [this message]
2015-07-09 22:41 ` [RFC/PATCH 5/7] x86/vm86: Teach handle_vm86_trap to return to 32bit mode directly Andy Lutomirski
2015-07-10 1:33 ` Andy Lutomirski
2015-07-10 15:27 ` Brian Gerst
2015-07-08 19:24 ` [RFC/PATCH 6/7] x86/entry/32: Use prepare_exit_to_usermode and syscall_return_slowpath Andy Lutomirski
2015-07-08 19:24 ` [RFC/PATCH 7/7] x86/entry: Remove do_notify_resume, syscall_trace_leave, and their TIF masks Andy Lutomirski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=c3bfa8ab98867fe07521a55765b85b3d4b582579.1436383168.git.luto@kernel.org \
--to=luto@kernel.org \
--cc=bp@alien8.de \
--cc=brgerst@gmail.com \
--cc=fweisbec@gmail.com \
--cc=keescook@chromium.org \
--cc=linux-kernel@vger.kernel.org \
--cc=oleg@redhat.com \
--cc=riel@redhat.com \
--cc=torvalds@linux-foundation.org \
--cc=vda.linux@googlemail.com \
--cc=x86@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).