PR15868 fix.

Introduction:
In case when stack alignment is 8 and GPRs parameter part size is not N*8:
we add padding to GPRs part, so part's last byte must be recovered at
address K*8-1.
We need to do it, since remained (stack) part of parameter starts from
address K*8, and we need to "attach" "GPRs head" without gaps to it:

Stack:
|---- 8 bytes block ----| |---- 8 bytes block ----| |---- 8 bytes...
[ [padding] [GPRs head] ] [ ------ Tail passed via stack  ------ ...

FIX:
Note, once we added padding we need to correct *all* Arg offsets that are going
after padded one. That's why we need this fix: Arg offsets were never corrected
before this patch. See new test-cases included in patch.

We also don't need to insert padding for byval parameters that are stored in GPRs
only. We need pad only last byval parameter and only in case it outsides GPRs
and stack alignment = 8.
Though, stack area, allocated for recovered byval params, must satisfy
"Size mod 8 = 0" restriction.

This patch reduces stack usage for some cases:
We can reduce ArgRegsSaveArea since inner N*4 bytes sized byval params my be
"packed" with alignment 4 in some cases.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182237 91177308-0d34-0410-b5e6-96231b3b80d8

3 files changed, 64 insertions, 8 deletions

diff --git a/test/CodeGen/ARM/2013-04-05-Small-ByVal-Structs-PR15293.ll b/test/CodeGen/ARM/2013-04-05-Small-ByVal-Structs-PR15293.ll
index 4a5ca9db0e..80b9d28a4b 100644
--- a/test/CodeGen/ARM/2013-04-05-Small-ByVal-Structs-PR15293.ll
+++ b/test/CodeGen/ARM/2013-04-05-Small-ByVal-Structs-PR15293.ll
@@ -4,24 +4,24 @@
 ;CHECK: foo:
 ;CHECK: 	sub	sp, sp, #8
 ;CHECK: 	push	{r11, lr}
-;CHECK: 	str	r0, [sp, #12]
-;CHECK: 	add	r0, sp, #12
+;CHECK: 	str	r0, [sp, #8]
+;CHECK: 	add	r0, sp, #8
 ;CHECK: 	bl	fooUseParam
 ;CHECK: 	pop	{r11, lr}
 ;CHECK: 	add	sp, sp, #8
 ;CHECK: 	mov	pc, lr
 
 ;CHECK: foo2:
-;CHECK: 	sub	sp, sp, #16
+;CHECK: 	sub	sp, sp, #8
 ;CHECK: 	push	{r11, lr}
-;CHECK: 	str	r0, [sp, #12]
-;CHECK: 	add	r0, sp, #12
-;CHECK: 	str	r2, [sp, #16]
+;CHECK: 	str	r0, [sp, #8]
+;CHECK: 	add	r0, sp, #8
+;CHECK: 	str	r2, [sp, #12]
 ;CHECK: 	bl	fooUseParam
-;CHECK: 	add	r0, sp, #16
+;CHECK: 	add	r0, sp, #12
 ;CHECK: 	bl	fooUseParam
 ;CHECK: 	pop	{r11, lr}
-;CHECK: 	add	sp, sp, #16
+;CHECK: 	add	sp, sp, #8
 ;CHECK: 	mov	pc, lr
 
 ;CHECK: doFoo:
diff --git a/test/CodeGen/ARM/2013-05-13-AAPCS-byval-padding.ll b/test/CodeGen/ARM/2013-05-13-AAPCS-byval-padding.ll
new file mode 100644
index 0000000000..7bf03a16c6
--- /dev/null
+++ b/test/CodeGen/ARM/2013-05-13-AAPCS-byval-padding.ll
@@ -0,0 +1,31 @@
+;PR15293: ARM codegen ice - expected larger existing stack allocation
+;RUN: llc -mtriple=arm-linux-gnueabihf < %s | FileCheck %s
+
+%struct.S227 = type { [49 x i32], i32 }
+
+define void @check227(
+                      i32 %b,                              
+                      %struct.S227* byval nocapture %arg0,
+                      %struct.S227* %arg1) {
+; b --> R0
+; arg0 --> [R1, R2, R3, SP+0 .. SP+188)
+; arg1 --> SP+188
+
+entry:
+
+;CHECK:  sub   sp, sp, #16
+;CHECK:  push  {r11, lr}
+;CHECK:  add   r0, sp, #12
+;CHECK:  stm   r0, {r1, r2, r3}
+;CHECK:  ldr   r0, [sp, #212]
+;CHECK:  bl    useInt
+;CHECK:  pop   {r11, lr}
+;CHECK:  add   sp, sp, #16
+
+  %0 = ptrtoint %struct.S227* %arg1 to i32
+  tail call void @useInt(i32 %0)
+  ret void
+}
+
+declare void @useInt(i32)
+
diff --git a/test/CodeGen/ARM/2013-05-13-AAPCS-byval-padding2.ll b/test/CodeGen/ARM/2013-05-13-AAPCS-byval-padding2.ll
new file mode 100644
index 0000000000..438b021a04
--- /dev/null
+++ b/test/CodeGen/ARM/2013-05-13-AAPCS-byval-padding2.ll
@@ -0,0 +1,25 @@
+;PR15293: ARM codegen ice - expected larger existing stack allocation
+;RUN: llc -mtriple=arm-linux-gnueabihf < %s | FileCheck %s
+
+%struct4bytes = type { i32 }
+%struct20bytes = type { i32, i32, i32, i32, i32 }
+
+define void @foo(%struct4bytes* byval %p0, ; --> R0
+                 %struct20bytes* byval %p1 ; --> R1,R2,R3, [SP+0 .. SP+8)
+) {
+;CHECK:  sub  sp, sp, #16
+;CHECK:  push  {r11, lr}
+;CHECK:  add  r11, sp, #8
+;CHECK:  stm  r11, {r0, r1, r2, r3}
+;CHECK:  add  r0, sp, #12
+;CHECK:  bl  useInt
+;CHECK:  pop  {r11, lr}
+;CHECK:  add  sp, sp, #16
+
+  %1 = ptrtoint %struct20bytes* %p1 to i32
+  tail call void @useInt(i32 %1)
+  ret void
+}
+
+declare void @useInt(i32)
+