From: Willy Tarreau <w@1wt.eu>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Andy Lutomirski <luto@amacapital.net>,
Thomas Gleixner <tglx@linutronix.de>,
Linus Torvalds <torvalds@linux-foundation.org>,
Willy Tarreau <w@1wt.eu>
Subject: [ 14/25] x86_64, traps: Fix the espfix64 #DF fixup and rewrite it in C
Date: Sat, 06 Dec 2014 18:42:02 +0100 [thread overview]
Message-ID: <20141206174149.021474424@1wt.eu> (raw)
In-Reply-To: <2a26e912d2438674771c36169c190830@local>
2.6.32-longterm review patch. If anyone has any objections, please let me know.
------------------
From: Andy Lutomirski <luto@amacapital.net>
There's nothing special enough about the espfix64 double fault fixup to
justify writing it in assembly. Move it to C.
This also fixes a bug: if the double fault came from an IST stack, the
old asm code would return to a partially uninitialized stack frame.
Fixes: 3891a04aafd668686239349ea58f3314ea2af86b
Signed-off-by: Andy Lutomirski <luto@amacapital.net>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@vger.kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
(cherry picked from commit af726f21ed8af2cdaa4e93098dc211521218ae65)
[wt: backport notes for 2.6.32 :
- Adaptations to entry_64.S in declaration of do_double_fault.
- no exception_enter() in 2.6.32. Seems to be only for context tracking.
/wt]
Signed-off-by: Willy Tarreau <w@1wt.eu>
---
arch/x86/kernel/entry_64.S | 34 ++--------------------------------
arch/x86/kernel/traps.c | 24 ++++++++++++++++++++++++
2 files changed, 26 insertions(+), 32 deletions(-)
diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S
index 8e2f14a..c862780 100644
--- a/arch/x86/kernel/entry_64.S
+++ b/arch/x86/kernel/entry_64.S
@@ -871,6 +871,7 @@ ENTRY(native_iret)
jnz native_irq_return_ldt
#endif
+.global native_irq_return_iret
native_irq_return_iret:
iretq
@@ -972,37 +973,6 @@ ENTRY(retint_kernel)
CFI_ENDPROC
END(common_interrupt)
- /*
- * If IRET takes a fault on the espfix stack, then we
- * end up promoting it to a doublefault. In that case,
- * modify the stack to make it look like we just entered
- * the #GP handler from user space, similar to bad_iret.
- */
-#ifdef CONFIG_X86_ESPFIX64
- ALIGN
-__do_double_fault:
- XCPT_FRAME 1 RDI+8
- movq RSP(%rdi),%rax /* Trap on the espfix stack? */
- sarq $PGDIR_SHIFT,%rax
- cmpl $ESPFIX_PGD_ENTRY,%eax
- jne do_double_fault /* No, just deliver the fault */
- cmpl $__KERNEL_CS,CS(%rdi)
- jne do_double_fault
- movq RIP(%rdi),%rax
- cmpq $native_irq_return_iret,%rax
- jne do_double_fault /* This shouldn't happen... */
- movq PER_CPU_VAR(kernel_stack),%rax
- subq $(6*8-KERNEL_STACK_OFFSET),%rax /* Reset to original stack */
- movq %rax,RSP(%rdi)
- movq $0,(%rax) /* Missing (lost) #GP error code */
- movq $general_protection,RIP(%rdi)
- retq
- CFI_ENDPROC
-END(__do_double_fault)
-#else
-# define __do_double_fault do_double_fault
-#endif
-
/*
* APIC interrupts.
*/
@@ -1179,7 +1149,7 @@ zeroentry overflow do_overflow
zeroentry bounds do_bounds
zeroentry invalid_op do_invalid_op
zeroentry device_not_available do_device_not_available
-paranoiderrorentry double_fault __do_double_fault
+paranoiderrorentry double_fault do_double_fault
zeroentry coprocessor_segment_overrun do_coprocessor_segment_overrun
errorentry invalid_TSS do_invalid_TSS
errorentry segment_not_present do_segment_not_present
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index c4cc05a..03563a4 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -230,6 +230,30 @@ dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
static const char str[] = "double fault";
struct task_struct *tsk = current;
+#ifdef CONFIG_X86_ESPFIX64
+ extern unsigned char native_irq_return_iret[];
+
+ /*
+ * If IRET takes a non-IST fault on the espfix64 stack, then we
+ * end up promoting it to a doublefault. In that case, modify
+ * the stack to make it look like we just entered the #GP
+ * handler from user space, similar to bad_iret.
+ */
+ if (((long)regs->sp >> PGDIR_SHIFT) == ESPFIX_PGD_ENTRY &&
+ regs->cs == __KERNEL_CS &&
+ regs->ip == (unsigned long)native_irq_return_iret)
+ {
+ struct pt_regs *normal_regs = task_pt_regs(current);
+
+ /* Fake a #GP(0) from userspace. */
+ memmove(&normal_regs->ip, (void *)regs->sp, 5*8);
+ normal_regs->orig_ax = 0; /* Missing (lost) #GP error code */
+ regs->ip = (unsigned long)general_protection;
+ regs->sp = (unsigned long)&normal_regs->orig_ax;
+ return;
+ }
+#endif
+
/* Return not checked because double check cannot be ignored */
notify_die(DIE_TRAP, str, regs, error_code, 8, SIGSEGV);
--
1.7.12.2.21.g234cd45.dirty
next prev parent reply other threads:[~2014-12-06 17:55 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-12-06 17:41 [ 01/25] net: sendmsg: fix failed backport of "fix NULL pointer dereference" Willy Tarreau
2014-12-06 17:41 ` [ 00/25] 2.6.32.65-longterm review Willy Tarreau
2014-12-08 0:58 ` Willy Tarreau
2014-12-06 17:41 ` [ 01/25] net: sendmsg: fix failed backport of "fix NULL pointer dereference" Willy Tarreau
2014-12-06 17:41 ` [ 02/25] x86, 64-bit: Move K8 B step iret fixup to fault entry asm Willy Tarreau
2014-12-06 17:41 ` [ 03/25] x86-64: Adjust frame type at paranoid_exit: Willy Tarreau
2014-12-06 17:41 ` [ 04/25] x86-64, modify_ldt: Ban 16-bit segments on 64-bit kernels Willy Tarreau
2014-12-06 17:41 ` [ 05/25] x86-32, espfix: Remove filter for espfix32 due to race Willy Tarreau
2014-12-06 17:41 ` [ 06/25] x86-64, espfix: Dont leak bits 31:16 of %esp returning to 16-bit stack Willy Tarreau
2014-12-06 17:41 ` [ 07/25] x86, espfix: Move espfix definitions into a separate header file Willy Tarreau
2014-12-06 17:41 ` [ 08/25] x86, espfix: Fix broken header guard Willy Tarreau
2014-12-06 17:41 ` [ 09/25] x86, espfix: Make espfix64 a Kconfig option, fix UML Willy Tarreau
2014-12-06 17:41 ` [ 10/25] x86, espfix: Make it possible to disable 16-bit support Willy Tarreau
2014-12-08 2:58 ` Ben Hutchings
2014-12-08 7:11 ` Willy Tarreau
2014-12-06 17:41 ` [ 11/25] x86_64/entry/xen: Do not invoke espfix64 on Xen Willy Tarreau
2014-12-06 17:42 ` [ 12/25] x86/espfix/xen: Fix allocation of pages for paravirt page tables Willy Tarreau
2014-12-06 17:42 ` [ 13/25] x86_64, traps: Stop using IST for #SS Willy Tarreau
2014-12-06 17:42 ` Willy Tarreau [this message]
2014-12-06 17:42 ` [ 15/25] x86_64, traps: Rework bad_iret Willy Tarreau
2014-12-06 17:42 ` [ 16/25] net/l2tp: dont fall back on UDP [get|set]sockopt Willy Tarreau
2014-12-06 17:42 ` [ 17/25] ALSA: control: Dont access controls outside of protected regions Willy Tarreau
2014-12-06 17:42 ` [ 18/25] ALSA: control: Fix replacing user controls Willy Tarreau
2014-12-06 17:42 ` [ 19/25] USB: whiteheat: Added bounds checking for bulk command response Willy Tarreau
2014-12-06 17:42 ` [ 20/25] net: sctp: fix panic on duplicate ASCONF chunks Willy Tarreau
2014-12-06 17:42 ` [ 21/25] net: sctp: fix remote memory pressure from excessive queueing Willy Tarreau
2014-12-06 17:42 ` [ 22/25] udf: Avoid infinite loop when processing indirect ICBs Willy Tarreau
2014-12-06 17:42 ` [ 23/25] net: sctp: fix NULL pointer dereference in af->from_addr_param on malformed packet Willy Tarreau
2014-12-06 17:42 ` [ 24/25] mac80211: fix fragmentation code, particularly for encryption Willy Tarreau
2014-12-06 17:42 ` [ 25/25] ttusb-dec: buffer overflow in ioctl Willy Tarreau
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=20141206174149.021474424@1wt.eu \
--to=w@1wt.eu \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@amacapital.net \
--cc=stable@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.