Skip to contentSkip to conversation

Correct PowerPC VDSO call frame info

PATCH In review Alan Modra 4 participants 5 messages Sep 13, 2018 – Sep 20, 2018 PowerPC

lore
Share & tools
lore

Alan Modra

author
#0 lore raw

PATCH 1/1 Correct PowerPC VDSO call frame info

There is control flow in __kernel_clock_gettime that reaches label 99 without saving lr in r12. CFI info however is interpreted by the unwinder without reference to control flow: It's a simple matter of "Execute all the CFI opcodes up to the current address". That means the unwinder thinks r12 contains the return address at label 99. Disabuse it of that notion by resetting CFI for the return address at label 99.

Note that the ".cfi_restore lr" could have gone anywhere from the "mtlr r12" a few instructions earlier to the instruction at label 99. I put the CFI as late as possible, because in general that's best practice (and if possible grouped with other CFI in order to reduce the number of CFI opcodes executed when unwinding). Using r12 as the return address is perfectly fine after the "mtlr r12" since r12 on that code path still contains the return address.

__get_datapage also has a CFI error. That function temporarily saves lr in r0, and reflects that fact with ".cfi_register lr,r0". A later use of r0 means the CFI at that point isn't correct, as r0 no longer contains the return address. Fix that too.

arch/powerpc/kernel/vdso32/datapage.S +1 -0
index 3745113fcc65..2a7eb5452aba 100644
@@ -37,6 +37,7 @@ data_page_branch:
 	mtlr	r0
 	addi	r3, r3, __kernel_datapage_offset-data_page_branch
 	lwz	r0,0(r3)
+  .cfi_restore lr
 	add	r3,r0,r3
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso32/gettimeofday.S +1 -0
index 769c2624e0a6..1e0bc5955a40 100644
@@ -139,6 +139,7 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 	 */
 99:
 	li	r0,__NR_clock_gettime
+  .cfi_restore lr
 	sc
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso64/datapage.S +1 -0
index abf17feffe40..bf9668691511 100644
@@ -37,6 +37,7 @@ data_page_branch:
 	mtlr	r0
 	addi	r3, r3, __kernel_datapage_offset-data_page_branch
 	lwz	r0,0(r3)
+  .cfi_restore lr
 	add	r3,r0,r3
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso64/gettimeofday.S +1 -0
index c002adcc694c..a4ed9edfd5f0 100644
@@ -169,6 +169,7 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 	 */
 99:
 	li	r0,__NR_clock_gettime
+  .cfi_restore lr
 	sc
 	blr
   .cfi_endproc
Signature · 2 lines
Alan Modra
Australia Development Lab, IBM
#1 lore raw

Alan,

Thanks for the patch... A few minor comments...

Quoted 1 line

Re: [PATCH] Correct PowerPC VDSO call frame info

Our convention is to add powerpc: to the start. ie

[PATCH] powerpc: Correct VDSO call frame info

or even:=20

[PATCH] powerpc/vdso: Correct call frame info

Quoted 7 lines · Alan Modra

There is control flow in __kernel_clock_gettime that reaches label 99 without saving lr in r12. CFI info however is interpreted by the unwinder without reference to control flow: It's a simple matter of "Execute all the CFI opcodes up to the current address". That means the unwinder thinks r12 contains the return address at label 99. Disabuse it of that notion by resetting CFI for the return address at label 99.

Can you expand on CFI =3D=3D Call Frame Information in the commit message. =
Not a
common term for us mere kernel developers?

Quoted 4 lines

Note that the ".cfi_restore lr" could have gone anywhere from the "mtlr r12" a few instructions earlier to the instruction at label 99. I put the CFI as late as possible, because in general that's best practice (and if possible grouped with other CFI in order to reduce

Quoted 8 lines

the number of CFI opcodes executed when unwinding). Using r12 as the return address is perfectly fine after the "mtlr r12" since r12 on that code path still contains the return address.
=20
__get_datapage also has a CFI error. That function temporarily saves lr in r0, and reflects that fact with ".cfi_register lr,r0". A later use of r0 means the CFI at that point isn't correct, as r0 no longer contains the return address. Fix that too.

Can you describe the problem this fixes at a high level? ie This fixes doi= ng a
stack unwind with gdb when in __kernel_clock_gettime.

Mikey

Quoted 55 lines

=20

arch/powerpc/kernel/vdso32/datapage.S +1 -0
index 3745113fcc65..2a7eb5452aba 100644
@@ -37,6 +37,7 @@ data_page_branch:
 	mtlr	r0
 	addi	r3, r3, __kernel_datapage_offset-data_page_branch
 	lwz	r0,0(r3)
+  .cfi_restore lr
 	add	r3,r0,r3
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso32/gettimeofday.S +1 -0
index 769c2624e0a6..1e0bc5955a40 100644
@@ -139,6 +139,7 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 	 */
 99:
 	li	r0,__NR_clock_gettime
+  .cfi_restore lr
 	sc
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso64/datapage.S +1 -0
index abf17feffe40..bf9668691511 100644
@@ -37,6 +37,7 @@ data_page_branch:
 	mtlr	r0
 	addi	r3, r3, __kernel_datapage_offset-data_page_branch
 	lwz	r0,0(r3)
+  .cfi_restore lr
 	add	r3,r0,r3
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso64/gettimeofday.S +1 -0
index c002adcc694c..a4ed9edfd5f0 100644
@@ -169,6 +169,7 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 	 */
 99:
 	li	r0,__NR_clock_gettime
+  .cfi_restore lr
 	sc
 	blr
   .cfi_endproc

=20

Tested-by on v1 1/1
#4 lore raw
In reply to Alan Modra
Quoted 7 lines · Alan Modra

There is control flow in __kernel_clock_gettime that reaches label 99 without saving lr in r12. CFI info however is interpreted by the unwinder without reference to control flow: It's a simple matter of "Execute all the CFI opcodes up to the current address". That means the unwinder thinks r12 contains the return address at label 99. Disabuse it of that notion by resetting CFI for the return address at label 99.

Thanks for this! It looks like v2 will just be a commit log change, so feel free to carry over my

Signature · 1 line
Reza Arbab
author
#2 lore raw

PATCH 1/1 PowerPC/VDSO: Correct call frame information

In reply to Michael Neuling

Call Frame Information is used by gdb for back-traces and inserting breakpoints on function return for the "finish" command. This failed when inside __kernel_clock_gettime. More concerning than difficulty debugging is that CFI is also used by stack frame unwinding code to implement exceptions. If you have an app that needs to handle asynchronous exceptions for some reason, and you are unlucky enough to get one inside the VDSO time functions, your app will crash.

What's wrong: There is control flow in __kernel_clock_gettime that reaches label 99 without saving lr in r12. CFI info however is interpreted by the unwinder without reference to control flow: It's a simple matter of "Execute all the CFI opcodes up to the current address". That means the unwinder thinks r12 contains the return address at label 99. Disabuse it of that notion by resetting CFI for the return address at label 99.

Note that the ".cfi_restore lr" could have gone anywhere from the "mtlr r12" a few instructions earlier to the instruction at label 99. I put the CFI as late as possible, because in general that's best practice (and if possible grouped with other CFI in order to reduce the number of CFI opcodes executed when unwinding). Using r12 as the return address is perfectly fine after the "mtlr r12" since r12 on that code path still contains the return address.

__get_datapage also has a CFI error. That function temporarily saves lr in r0, and reflects that fact with ".cfi_register lr,r0". A later use of r0 means the CFI at that point isn't correct, as r0 no longer contains the return address. Fix that too.

arch/powerpc/kernel/vdso32/datapage.S +1 -0
index 3745113fcc65..2a7eb5452aba 100644
@@ -37,6 +37,7 @@ data_page_branch:
 	mtlr	r0
 	addi	r3, r3, __kernel_datapage_offset-data_page_branch
 	lwz	r0,0(r3)
+  .cfi_restore lr
 	add	r3,r0,r3
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso32/gettimeofday.S +1 -0
index 769c2624e0a6..1e0bc5955a40 100644
@@ -139,6 +139,7 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 	 */
 99:
 	li	r0,__NR_clock_gettime
+  .cfi_restore lr
 	sc
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso64/datapage.S +1 -0
index abf17feffe40..bf9668691511 100644
@@ -37,6 +37,7 @@ data_page_branch:
 	mtlr	r0
 	addi	r3, r3, __kernel_datapage_offset-data_page_branch
 	lwz	r0,0(r3)
+  .cfi_restore lr
 	add	r3,r0,r3
 	blr
   .cfi_endproc
arch/powerpc/kernel/vdso64/gettimeofday.S +1 -0
index c002adcc694c..a4ed9edfd5f0 100644
@@ -169,6 +169,7 @@ V_FUNCTION_BEGIN(__kernel_clock_gettime)
 	 */
 99:
 	li	r0,__NR_clock_gettime
+  .cfi_restore lr
 	sc
 	blr
   .cfi_endproc
Signature · 2 lines
Alan Modra
Australia Development Lab, IBM
#3 lore raw

Re: PowerPC/VDSO: Correct call frame information

Quoted 24 lines · Alan Modra

Call Frame Information is used by gdb for back-traces and inserting breakpoints on function return for the "finish" command. This failed when inside __kernel_clock_gettime. More concerning than difficulty debugging is that CFI is also used by stack frame unwinding code to implement exceptions. If you have an app that needs to handle asynchronous exceptions for some reason, and you are unlucky enough to get one inside the VDSO time functions, your app will crash.

What's wrong: There is control flow in __kernel_clock_gettime that reaches label 99 without saving lr in r12. CFI info however is interpreted by the unwinder without reference to control flow: It's a simple matter of "Execute all the CFI opcodes up to the current address". That means the unwinder thinks r12 contains the return address at label 99. Disabuse it of that notion by resetting CFI for the return address at label 99.

Note that the ".cfi_restore lr" could have gone anywhere from the "mtlr r12" a few instructions earlier to the instruction at label 99. I put the CFI as late as possible, because in general that's best practice (and if possible grouped with other CFI in order to reduce the number of CFI opcodes executed when unwinding). Using r12 as the return address is perfectly fine after the "mtlr r12" since r12 on that code path still contains the return address.

Quoted 7 lines

__get_datapage also has a CFI error. That function temporarily saves lr in r0, and reflects that fact with ".cfi_register lr,r0". A later use of r0 means the CFI at that point isn't correct, as r0 no longer contains the return address. Fix that too.

Applied to powerpc next, thanks.

https://git.kernel.org/powerpc/c/56d20861c027498b5a1112b4f9f05b

cheers