[PATCH] D40894: [XRay][compiler-rt] Update XRay trampoline CFI and saving/restoring semantics

[Dean Michael Berris via Phabricator via llvm-commits]

dberris created this revision.

This change updates the CFI directives used by the XRay trampoline
implementations for x86_64 as well as expand the amount of registers
stashed by the __xray_CustomEvent trampoline. We've found that since the
__xray_CustomEvent trampoline calls can show up in situations where the
scratch registers are being used, and since we don't typically want to
affect the code-gen around the disabled `__xray_customevent(...)`
intrinsic calls, that we need to save and restore the state of even the
scratch registers in the handling of these custom events.

The CFI directive changes are useful for helping the debugger(s)
determine where the frame of the handlers begin, and when the frame of
the trampolines end.


https://reviews.llvm.org/D40894

Files:
  compiler-rt/lib/xray/xray_trampoline_x86_64.S


Index: compiler-rt/lib/xray/xray_trampoline_x86_64.S
===================================================================
--- compiler-rt/lib/xray/xray_trampoline_x86_64.S
+++ compiler-rt/lib/xray/xray_trampoline_x86_64.S
@@ -20,7 +20,7 @@
 
 .macro SAVE_REGISTERS
 	subq $192, %rsp
-	CFI_DEF_CFA_OFFSET(200)
+	CFI_ADJUST_CFA_OFFSET(192)
 	// At this point, the stack pointer should be aligned to an 8-byte boundary,
 	// because any call instructions that come after this will add another 8
 	// bytes and therefore align it to 16-bytes.
@@ -60,7 +60,7 @@
 	movq	8(%rsp), %r8
 	movq	0(%rsp), %r9
 	addq	$192, %rsp
-	CFI_DEF_CFA_OFFSET(8)
+	CFI_ADJUST_CFA_OFFSET(-192)
 .endm
 
 .macro ALIGNED_CALL_RAX
@@ -71,10 +71,13 @@
 	// pointer, we can always look -8 bytes from the current position to get
 	// either of the values we've stashed in the first place.
 	pushq %rsp
+	CFI_ADJUST_CFA_OFFSET(8)
 	pushq (%rsp)
+	CFI_ADJUST_CFA_OFFSET(8)
 	andq $-0x10, %rsp
   callq *%rax
 	movq 8(%rsp), %rsp
+	CFI_ADJUST_CFA_OFFSET(-16)
 .endm
 
 	.text
@@ -100,7 +103,7 @@
 	testq	%rax, %rax
 	je	.Ltmp0
 
-	// The patched function prolog puts its xray_instr_map index into %r10d.
+	// The patched function prologue puts its xray_instr_map index into %r10d.
 	movl	%r10d, %edi
 	xor	%esi,%esi
 	ALIGNED_CALL_RAX
@@ -122,7 +125,7 @@
 	// function is only jumped into, we only preserve the registers for
 	// returning.
 	subq	$56, %rsp
-	CFI_DEF_CFA_OFFSET(64)
+	CFI_ADJUST_CFA_OFFSET(56)
 	movq  %rbp, 48(%rsp)
 	movupd	%xmm0, 32(%rsp)
 	movupd	%xmm1, 16(%rsp)
@@ -144,7 +147,7 @@
 	movq	8(%rsp), %rax
 	movq	0(%rsp), %rdx
 	addq	$56, %rsp
-	CFI_DEF_CFA_OFFSET(8)
+	CFI_ADJUST_CFA_OFFSET(-56)
 	retq
 	ASM_SIZE(__xray_FunctionExit)
 	CFI_ENDPROC
@@ -219,6 +222,15 @@
 	CFI_STARTPROC
 	SAVE_REGISTERS
 
+	// Because calls to __xray_CustomEvent can occur in the middle of a function
+	// which may already be using scratch registers, we preserve more information
+	// in this trampoline than when we are handling entry/exit events.
+	subq $24, %rsp
+	CFI_ADJUST_CFA_OFFSET(24)
+	movq %r10, 16(%rsp)
+	movq %r11, 8(%rsp)
+	movq %r12, 0(%rsp)
+
 	// We take two arguments to this trampoline, which should be in rdi	and rsi
 	// already. We also make sure that we stash %rax because we use that register
 	// to call the logging handler.
@@ -229,6 +241,12 @@
 	ALIGNED_CALL_RAX
 
 .LcustomEventCleanup:
+
+	movq 16(%rsp), %r10
+	movq 8(%rsp), %r11
+	movq 0(%rsp), %r12
+	addq $24, %rsp
+	CFI_ADJUST_CFA_OFFSET(-24)
 	RESTORE_REGISTERS
 	retq
 	ASM_SIZE(__xray_CustomEvent)


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D40894.125719.patch
Type: text/x-patch
Size: 2556 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20171206/218c730c/attachment.bin>