linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [tip: objtool/core] objtool: Support stack-swizzle
@ 2021-02-10 20:01 tip-bot2 for Peter Zijlstra
  2021-02-18 16:21 ` [PATCH] objtool: Fix stack-swizzle for FRAME_POINTER=y Peter Zijlstra
  0 siblings, 1 reply; 5+ messages in thread
From: tip-bot2 for Peter Zijlstra @ 2021-02-10 20:01 UTC (permalink / raw)
  To: linux-tip-commits
  Cc: Peter Zijlstra (Intel),
	Miroslav Benes, Josh Poimboeuf, x86, linux-kernel

The following commit has been merged into the objtool/core branch of tip:

Commit-ID:     aafeb14e9da29e323b0605f8f1bae0d45d5f3acf
Gitweb:        https://git.kernel.org/tip/aafeb14e9da29e323b0605f8f1bae0d45d5f3acf
Author:        Peter Zijlstra <peterz@infradead.org>
AuthorDate:    Wed, 03 Feb 2021 12:02:17 +01:00
Committer:     Peter Zijlstra <peterz@infradead.org>
CommitterDate: Wed, 10 Feb 2021 20:53:52 +01:00

objtool: Support stack-swizzle

Natively support the stack swizzle pattern:

	mov %rsp, (%[tos])
	mov %[tos], %rsp
	...
	pop %rsp

It uses the vals[] array to link the first two stack-ops, and detect
the SP to SP_INDIRECT swizzle.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Miroslav Benes <mbenes@suse.cz>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
---
 tools/objtool/check.c | 45 ++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 45 insertions(+)

diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index 5f056dd..62cd211 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -1945,6 +1945,38 @@ static int update_cfi_state(struct instruction *insn, struct cfi_state *cfi,
 					cfa->offset = -cfi->vals[op->src.reg].offset;
 					cfi->stack_size = cfa->offset;
 
+				} else if (cfa->base == CFI_SP &&
+					   cfi->vals[op->src.reg].base == CFI_SP_INDIRECT &&
+					   cfi->vals[op->src.reg].offset == cfa->offset) {
+
+					/*
+					 * Stack swizzle:
+					 *
+					 * 1: mov %rsp, (%[tos])
+					 * 2: mov %[tos], %rsp
+					 *    ...
+					 * 3: pop %rsp
+					 *
+					 * Where:
+					 *
+					 * 1 - places a pointer to the previous
+					 *     stack at the Top-of-Stack of the
+					 *     new stack.
+					 *
+					 * 2 - switches to the new stack.
+					 *
+					 * 3 - pops the Top-of-Stack to restore
+					 *     the original stack.
+					 *
+					 * Note: we set base to SP_INDIRECT
+					 * here and preserve offset. Therefore
+					 * when the unwinder reaches ToS it
+					 * will dereference SP and then add the
+					 * offset to find the next frame, IOW:
+					 * (%rsp) + offset.
+					 */
+					cfa->base = CFI_SP_INDIRECT;
+
 				} else {
 					cfa->base = CFI_UNDEFINED;
 					cfa->offset = 0;
@@ -2047,6 +2079,13 @@ static int update_cfi_state(struct instruction *insn, struct cfi_state *cfi,
 
 		case OP_SRC_POP:
 		case OP_SRC_POPF:
+			if (op->dest.reg == CFI_SP && cfa->base == CFI_SP_INDIRECT) {
+
+				/* pop %rsp; # restore from a stack swizzle */
+				cfa->base = CFI_SP;
+				break;
+			}
+
 			if (!cfi->drap && op->dest.reg == cfa->base) {
 
 				/* pop %rbp */
@@ -2193,6 +2232,12 @@ static int update_cfi_state(struct instruction *insn, struct cfi_state *cfi,
 			/* mov reg, disp(%rsp) */
 			save_reg(cfi, op->src.reg, CFI_CFA,
 				 op->dest.offset - cfi->stack_size);
+
+		} else if (op->src.reg == CFI_SP && op->dest.offset == 0) {
+
+			/* mov %rsp, (%reg); # setup a stack swizzle. */
+			cfi->vals[op->dest.reg].base = CFI_SP_INDIRECT;
+			cfi->vals[op->dest.reg].offset = cfa->offset;
 		}
 
 		break;

^ permalink raw reply related	[flat|nested] 5+ messages in thread
* [PATCH] objtool: Fix stack-swizzle for FRAME_POINTER=y
  2021-02-10 20:01 [tip: objtool/core] objtool: Support stack-swizzle tip-bot2 for Peter Zijlstra
@ 2021-02-18 16:21 ` Peter Zijlstra
  2021-02-18 17:53   ` Josh Poimboeuf
                     ` (2 more replies)
  0 siblings, 3 replies; 5+ messages in thread
From: Peter Zijlstra @ 2021-02-18 16:21 UTC (permalink / raw)
  To: linux-kernel; +Cc: linux-tip-commits, Miroslav Benes, Josh Poimboeuf, x86

On Wed, Feb 10, 2021 at 08:01:23PM -0000, tip-bot2 for Peter Zijlstra wrote:
> objtool: Support stack-swizzle

---
Subject: objtool: Fix stack-swizzle for FRAME_POINTER=y
From: Peter Zijlstra <peterz@infradead.org>
Date: Thu Feb 18 17:14:10 CET 2021

When objtool encounters the stack-swizzle:

	mov %rsp, (%[tos])
	mov %[tos], %rsp
	...
	pop %rsp

Inside a FRAME_POINTER=y build, things go a little screwy because
clearly we're not adjusting the cfa->base. This then results in the
pop %rsp not being detected as a restore of cfa->base so it will turn
into a regular POP and offset the stack, resulting in:

  kernel/softirq.o: warning: objtool: do_softirq()+0xdb: return with modified stack frame

Therefore, have "mov %[tos], %rsp" act like a PUSH (it sorta is
anyway) to balance the things out. We're not too concerned with the
actual stack_size for frame-pointer builds, since we don't generate
ORC data for them anyway.

Fixes: aafeb14e9da2 ("objtool: Support stack-swizzle")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
 tools/objtool/check.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -1996,6 +1996,20 @@ static int update_cfi_state(struct instr
 				}
 			}
 
+			else if (op->dest.reg == CFI_SP &&
+				 cfi->vals[op->src.reg].base == CFI_SP_INDIRECT &&
+				 cfi->vals[op->src.reg].offset == cfa->offset) {
+
+				/*
+				 * The same stack swizzle case 2) as above. But
+				 * because we can't change cfa->base, case 3)
+				 * will become a regular POP. Pretend we're a
+				 * PUSH so things don't go unbalanced.
+				 */
+				cfi->stack_size += 8;
+			}
+
+
 			break;
 
 		case OP_SRC_ADD:

^ permalink raw reply	[flat|nested] 5+ messages in thread
* Re: [PATCH] objtool: Fix stack-swizzle for FRAME_POINTER=y
  2021-02-18 16:21 ` [PATCH] objtool: Fix stack-swizzle for FRAME_POINTER=y Peter Zijlstra
@ 2021-02-18 17:53   ` Josh Poimboeuf
  2021-02-22 11:08   ` [tip: objtool/core] " tip-bot2 for Peter Zijlstra
  2021-02-22 19:00   ` [tip: x86/entry] " tip-bot2 for Peter Zijlstra
  2 siblings, 0 replies; 5+ messages in thread
From: Josh Poimboeuf @ 2021-02-18 17:53 UTC (permalink / raw)
  To: Peter Zijlstra; +Cc: linux-kernel, linux-tip-commits, Miroslav Benes, x86

On Thu, Feb 18, 2021 at 05:21:31PM +0100, Peter Zijlstra wrote:
> On Wed, Feb 10, 2021 at 08:01:23PM -0000, tip-bot2 for Peter Zijlstra wrote:
> > objtool: Support stack-swizzle
> 
> ---
> Subject: objtool: Fix stack-swizzle for FRAME_POINTER=y
> From: Peter Zijlstra <peterz@infradead.org>
> Date: Thu Feb 18 17:14:10 CET 2021
> 
> When objtool encounters the stack-swizzle:
> 
> 	mov %rsp, (%[tos])
> 	mov %[tos], %rsp
> 	...
> 	pop %rsp
> 
> Inside a FRAME_POINTER=y build, things go a little screwy because
> clearly we're not adjusting the cfa->base. This then results in the
> pop %rsp not being detected as a restore of cfa->base so it will turn
> into a regular POP and offset the stack, resulting in:
> 
>   kernel/softirq.o: warning: objtool: do_softirq()+0xdb: return with modified stack frame
> 
> Therefore, have "mov %[tos], %rsp" act like a PUSH (it sorta is
> anyway) to balance the things out. We're not too concerned with the
> actual stack_size for frame-pointer builds, since we don't generate
> ORC data for them anyway.
> 
> Fixes: aafeb14e9da2 ("objtool: Support stack-swizzle")
> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>

-- 
Josh


^ permalink raw reply	[flat|nested] 5+ messages in thread
* [tip: objtool/core] objtool: Fix stack-swizzle for FRAME_POINTER=y
  2021-02-18 16:21 ` [PATCH] objtool: Fix stack-swizzle for FRAME_POINTER=y Peter Zijlstra
  2021-02-18 17:53   ` Josh Poimboeuf
@ 2021-02-22 11:08   ` tip-bot2 for Peter Zijlstra
  2021-02-22 19:00   ` [tip: x86/entry] " tip-bot2 for Peter Zijlstra
  2 siblings, 0 replies; 5+ messages in thread
From: tip-bot2 for Peter Zijlstra @ 2021-02-22 11:08 UTC (permalink / raw)
  To: linux-tip-commits
  Cc: kernel test robot, Peter Zijlstra (Intel),
	Josh Poimboeuf, x86, linux-kernel

The following commit has been merged into the objtool/core branch of tip:

Commit-ID:     23e34c5988088b8bb4c55905973ca76114cb33ee
Gitweb:        https://git.kernel.org/tip/23e34c5988088b8bb4c55905973ca76114cb33ee
Author:        Peter Zijlstra <peterz@infradead.org>
AuthorDate:    Thu, 18 Feb 2021 17:14:10 +01:00
Committer:     Peter Zijlstra <peterz@infradead.org>
CommitterDate: Mon, 22 Feb 2021 12:05:18 +01:00

objtool: Fix stack-swizzle for FRAME_POINTER=y

When objtool encounters the stack-swizzle:

	mov %rsp, (%[tos])
	mov %[tos], %rsp
	...
	pop %rsp

Inside a FRAME_POINTER=y build, things go a little screwy because
clearly we're not adjusting the cfa->base. This then results in the
pop %rsp not being detected as a restore of cfa->base so it will turn
into a regular POP and offset the stack, resulting in:

  kernel/softirq.o: warning: objtool: do_softirq()+0xdb: return with modified stack frame

Therefore, have "mov %[tos], %rsp" act like a PUSH (it sorta is
anyway) to balance the things out. We're not too concerned with the
actual stack_size for frame-pointer builds, since we don't generate
ORC data for them anyway.

Fixes: aafeb14e9da2 ("objtool: Support stack-swizzle")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/YC6UC+rc9KKmQrkd@hirez.programming.kicks-ass.net
---
 tools/objtool/check.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index 62cd211..d7f1496 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -1983,6 +1983,20 @@ static int update_cfi_state(struct instruction *insn, struct cfi_state *cfi,
 				}
 			}
 
+			else if (op->dest.reg == CFI_SP &&
+				 cfi->vals[op->src.reg].base == CFI_SP_INDIRECT &&
+				 cfi->vals[op->src.reg].offset == cfa->offset) {
+
+				/*
+				 * The same stack swizzle case 2) as above. But
+				 * because we can't change cfa->base, case 3)
+				 * will become a regular POP. Pretend we're a
+				 * PUSH so things don't go unbalanced.
+				 */
+				cfi->stack_size += 8;
+			}
+
+
 			break;
 
 		case OP_SRC_ADD:

^ permalink raw reply related	[flat|nested] 5+ messages in thread
* [tip: x86/entry] objtool: Fix stack-swizzle for FRAME_POINTER=y
  2021-02-18 16:21 ` [PATCH] objtool: Fix stack-swizzle for FRAME_POINTER=y Peter Zijlstra
  2021-02-18 17:53   ` Josh Poimboeuf
  2021-02-22 11:08   ` [tip: objtool/core] " tip-bot2 for Peter Zijlstra
@ 2021-02-22 19:00   ` tip-bot2 for Peter Zijlstra
  2 siblings, 0 replies; 5+ messages in thread
From: tip-bot2 for Peter Zijlstra @ 2021-02-22 19:00 UTC (permalink / raw)
  To: linux-tip-commits
  Cc: kernel test robot, Peter Zijlstra (Intel),
	Thomas Gleixner, Josh Poimboeuf, x86, linux-kernel

The following commit has been merged into the x86/entry branch of tip:

Commit-ID:     724c8a23d589d8a002d2e39633c2f9a5a429616f
Gitweb:        https://git.kernel.org/tip/724c8a23d589d8a002d2e39633c2f9a5a429616f
Author:        Peter Zijlstra <peterz@infradead.org>
AuthorDate:    Thu, 18 Feb 2021 17:14:10 +01:00
Committer:     Thomas Gleixner <tglx@linutronix.de>
CommitterDate: Mon, 22 Feb 2021 19:54:09 +01:00

objtool: Fix stack-swizzle for FRAME_POINTER=y

When objtool encounters the stack-swizzle:

	mov %rsp, (%[tos])
	mov %[tos], %rsp
	...
	pop %rsp

Inside a FRAME_POINTER=y build, things go a little screwy because
clearly we're not adjusting the cfa->base. This then results in the
pop %rsp not being detected as a restore of cfa->base so it will turn
into a regular POP and offset the stack, resulting in:

  kernel/softirq.o: warning: objtool: do_softirq()+0xdb: return with modified stack frame

Therefore, have "mov %[tos], %rsp" act like a PUSH (it sorta is
anyway) to balance the things out. We're not too concerned with the
actual stack_size for frame-pointer builds, since we don't generate
ORC data for them anyway.

Fixes: aafeb14e9da2 ("objtool: Support stack-swizzle")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Josh Poimboeuf <jpoimboe@redhat.com>
Link: https://lkml.kernel.org/r/YC6UC+rc9KKmQrkd@hirez.programming.kicks-ass.net
---
 tools/objtool/check.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

diff --git a/tools/objtool/check.c b/tools/objtool/check.c
index 8e74210..2087974 100644
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -1983,6 +1983,20 @@ static int update_cfi_state(struct instruction *insn, struct cfi_state *cfi,
 				}
 			}
 
+			else if (op->dest.reg == CFI_SP &&
+				 cfi->vals[op->src.reg].base == CFI_SP_INDIRECT &&
+				 cfi->vals[op->src.reg].offset == cfa->offset) {
+
+				/*
+				 * The same stack swizzle case 2) as above. But
+				 * because we can't change cfa->base, case 3)
+				 * will become a regular POP. Pretend we're a
+				 * PUSH so things don't go unbalanced.
+				 */
+				cfi->stack_size += 8;
+			}
+
+
 			break;
 
 		case OP_SRC_ADD:

^ permalink raw reply related	[flat|nested] 5+ messages in thread
end of thread, other threads:[~2021-02-22 19:02 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-02-10 20:01 [tip: objtool/core] objtool: Support stack-swizzle tip-bot2 for Peter Zijlstra
2021-02-18 16:21 ` [PATCH] objtool: Fix stack-swizzle for FRAME_POINTER=y Peter Zijlstra
2021-02-18 17:53   ` Josh Poimboeuf
2021-02-22 11:08   ` [tip: objtool/core] " tip-bot2 for Peter Zijlstra
2021-02-22 19:00   ` [tip: x86/entry] " tip-bot2 for Peter Zijlstra

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).