maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/llvm/test/CodeGen/X86/segmented-stacks.ll

711 lines
20 KiB
LLVM
Raw Normal View History

Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
; RUN: llc < %s -mcpu=generic -mtriple=i686-linux -verify-machineinstrs | FileCheck %s -check-prefix=X32-Linux
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-linux -verify-machineinstrs | FileCheck %s -check-prefix=X64-Linux
x86_64: Fix calls to __morestack under the large code model. Under the large code model, we cannot assume that __morestack lives within 2^31 bytes of the call site, so we cannot use pc-relative addressing. We cannot perform the call via a temporary register, as the rax register may be used to store the static chain, and all other suitable registers may be either callee-save or used for parameter passing. We cannot use the stack at this point either because __morestack manipulates the stack directly. To avoid these issues, perform an indirect call via a read-only memory location containing the address. This solution is not perfect, as it assumes that the .rodata section is laid out within 2^31 bytes of each function body, but this seems to be sufficient for JIT. Differential Revision: http://reviews.llvm.org/D6787 llvm-svn: 225003
2014-12-31 04:05:19 +08:00
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-linux -code-model=large -verify-machineinstrs | FileCheck %s -check-prefix=X64-Linux-Large
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-linux-gnux32 -verify-machineinstrs | FileCheck %s -check-prefix=X32ABI
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
; RUN: llc < %s -mcpu=generic -mtriple=i686-darwin -verify-machineinstrs | FileCheck %s -check-prefix=X32-Darwin
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-darwin -verify-machineinstrs | FileCheck %s -check-prefix=X64-Darwin
; RUN: llc < %s -mcpu=generic -mtriple=i686-mingw32 -verify-machineinstrs | FileCheck %s -check-prefix=X32-MinGW
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-freebsd -verify-machineinstrs | FileCheck %s -check-prefix=X64-FreeBSD
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; RUN: llc < %s -mcpu=generic -mtriple=i686-dragonfly -verify-machineinstrs | FileCheck %s -check-prefix=X32-DFlyBSD
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-dragonfly -verify-machineinstrs | FileCheck %s -check-prefix=X64-DFlyBSD
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-mingw32 -verify-machineinstrs | FileCheck %s -check-prefix=X64-MinGW
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Explicitly set the scale to 1 on some segstack prologue instrs. Patch by Brian Anderson. llvm-svn: 147952
2012-01-12 02:14:03 +08:00
; We used to crash with filetype=obj
Avoid spewing binary to stdout in some filetype=obj tests llvm-svn: 234627
2015-04-11 03:36:55 +08:00
; RUN: llc < %s -mcpu=generic -mtriple=i686-linux -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-linux -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-linux-gnux32 -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=i686-darwin -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-darwin -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=i686-mingw32 -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-freebsd -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=i686-dragonfly -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-dragonfly -filetype=obj -o /dev/null
; RUN: llc < %s -mcpu=generic -mtriple=x86_64-mingw32 -filetype=obj -o /dev/null
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
; RUN: not llc < %s -mcpu=generic -mtriple=x86_64-solaris 2> %t.log
Add error-reporting tests for platforms that don't support segmented stacks. Patch by Brian Anderson. llvm-svn: 148042
2012-01-13 04:26:13 +08:00
; RUN: FileCheck %s -input-file=%t.log -check-prefix=X64-Solaris
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
; RUN: not llc < %s -mcpu=generic -mtriple=i686-freebsd 2> %t.log
Add error-reporting tests for platforms that don't support segmented stacks. Patch by Brian Anderson. llvm-svn: 148042
2012-01-13 04:26:13 +08:00
; RUN: FileCheck %s -input-file=%t.log -check-prefix=X32-FreeBSD
; X64-Solaris: Segmented stacks not supported on this platform
; X32-FreeBSD: Segmented stacks not supported on FreeBSD i386
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
; Just to prevent the alloca from being optimized away
declare void @dummy_use(i32*, i32)
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
define void @test_basic() #0 {
Split segmented stacks tests into tests for static- and dynamic-size frames. Patch by Brian Anderson. llvm-svn: 147959
2012-01-12 02:51:03 +08:00
%mem = alloca i32, i32 10
call void @dummy_use (i32* %mem, i32 10)
ret void
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-Linux-LABEL: test_basic:
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: cmpl %gs:48, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Linux-NEXT: jbe .LBB0_1
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-Linux-NEXT: pushl $44
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux-NEXT: calll __morestack
; X32-Linux-NEXT: ret
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-Linux-LABEL: test_basic:
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux: cmpq %fs:112, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Linux-NEXT: jbe .LBB0_1
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux: movabsq $40, %r10
; X64-Linux-NEXT: movabsq $0, %r11
; X64-Linux-NEXT: callq __morestack
; X64-Linux-NEXT: ret
x86_64: Fix calls to __morestack under the large code model. Under the large code model, we cannot assume that __morestack lives within 2^31 bytes of the call site, so we cannot use pc-relative addressing. We cannot perform the call via a temporary register, as the rax register may be used to store the static chain, and all other suitable registers may be either callee-save or used for parameter passing. We cannot use the stack at this point either because __morestack manipulates the stack directly. To avoid these issues, perform an indirect call via a read-only memory location containing the address. This solution is not perfect, as it assumes that the .rodata section is laid out within 2^31 bytes of each function body, but this seems to be sufficient for JIT. Differential Revision: http://reviews.llvm.org/D6787 llvm-svn: 225003
2014-12-31 04:05:19 +08:00
; X64-Linux-Large-LABEL: test_basic:
; X64-Linux-Large: cmpq %fs:112, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Linux-Large-NEXT: jbe .LBB0_1
x86_64: Fix calls to __morestack under the large code model. Under the large code model, we cannot assume that __morestack lives within 2^31 bytes of the call site, so we cannot use pc-relative addressing. We cannot perform the call via a temporary register, as the rax register may be used to store the static chain, and all other suitable registers may be either callee-save or used for parameter passing. We cannot use the stack at this point either because __morestack manipulates the stack directly. To avoid these issues, perform an indirect call via a read-only memory location containing the address. This solution is not perfect, as it assumes that the .rodata section is laid out within 2^31 bytes of each function body, but this seems to be sufficient for JIT. Differential Revision: http://reviews.llvm.org/D6787 llvm-svn: 225003
2014-12-31 04:05:19 +08:00
; X64-Linux-Large: movabsq $40, %r10
; X64-Linux-Large-NEXT: movabsq $0, %r11
; X64-Linux-Large-NEXT: callq *__morestack_addr(%rip)
; X64-Linux-Large-NEXT: ret
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI-LABEL: test_basic:
; X32ABI: cmpl %fs:64, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32ABI-NEXT: jbe .LBB0_1
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI: movl $40, %r10d
; X32ABI-NEXT: movl $0, %r11d
; X32ABI-NEXT: callq __morestack
; X32ABI-NEXT: ret
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-Darwin-LABEL: test_basic:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: movl $432, %ecx
; X32-Darwin-NEXT: cmpl %gs:(%ecx), %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Darwin-NEXT: jbe LBB0_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: pushl $0
; X32-Darwin-NEXT: pushl $60
; X32-Darwin-NEXT: calll ___morestack
; X32-Darwin-NEXT: ret
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-Darwin-LABEL: test_basic:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: cmpq %gs:816, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Darwin-NEXT: jbe LBB0_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: movabsq $40, %r10
; X64-Darwin-NEXT: movabsq $0, %r11
; X64-Darwin-NEXT: callq ___morestack
; X64-Darwin-NEXT: ret
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-MinGW-LABEL: test_basic:
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW: cmpl %fs:20, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-MinGW-NEXT: jbe LBB0_1
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-MinGW-NEXT: pushl $40
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW-NEXT: calll ___morestack
; X32-MinGW-NEXT: ret
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW-LABEL: test_basic:
; X64-MinGW: cmpq %gs:40, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-MinGW-NEXT: jbe .LBB0_1
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW: movabsq $72, %r10
; X64-MinGW-NEXT: movabsq $32, %r11
; X64-MinGW-NEXT: callq __morestack
; X64-MinGW-NEXT: retq
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-FreeBSD-LABEL: test_basic:
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: cmpq %fs:24, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-FreeBSD-NEXT: jbe .LBB0_1
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: movabsq $40, %r10
; X64-FreeBSD-NEXT: movabsq $0, %r11
; X64-FreeBSD-NEXT: callq __morestack
; X64-FreeBSD-NEXT: ret
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-LABEL: test_basic:
; X32-DFlyBSD: cmpl %fs:16, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-DFlyBSD-NEXT: jbe .LBB0_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-DFlyBSD-NEXT: pushl $40
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-NEXT: calll __morestack
; X32-DFlyBSD-NEXT: ret
; X64-DFlyBSD-LABEL: test_basic:
; X64-DFlyBSD: cmpq %fs:32, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-DFlyBSD-NEXT: jbe .LBB0_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X64-DFlyBSD: movabsq $40, %r10
; X64-DFlyBSD-NEXT: movabsq $0, %r11
; X64-DFlyBSD-NEXT: callq __morestack
; X64-DFlyBSD-NEXT: ret
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
}
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
define i32 @test_nested(i32 * nest %closure, i32 %other) #0 {
[opaque pointer type] Add textual IR support for explicit type parameter to load instruction Essentially the same as the GEP change in r230786. A similar migration script can be used to update test cases, though a few more test case improvements/changes were required this time around: (r229269-r229278) import fileinput import sys import re pat = re.compile(r"((?:=|:|^)\s*load (?:atomic )?(?:volatile )?(.*?))(| addrspace\(\d+\) *)\*($| *(?:%|@|null|undef|blockaddress|getelementptr|addrspacecast|bitcast|inttoptr|\[\[[a-zA-Z]|\{\{).*$)") for line in sys.stdin: sys.stdout.write(re.sub(pat, r"\1, \2\3*\4", line)) Reviewers: rafael, dexonsmith, grosser Differential Revision: http://reviews.llvm.org/D7649 llvm-svn: 230794
2015-02-28 05:17:42 +08:00
%addend = load i32 , i32 * %closure
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
%result = add i32 %other, %addend
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
%mem = alloca i32, i32 10
call void @dummy_use (i32* %mem, i32 10)
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
ret i32 %result
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: cmpl %gs:48, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Linux-NEXT: jbe .LBB1_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: pushl $4
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-Linux-NEXT: pushl $44
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux-NEXT: calll __morestack
; X32-Linux-NEXT: ret
; X64-Linux: cmpq %fs:112, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Linux-NEXT: jbe .LBB1_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux: movq %r10, %rax
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
; X64-Linux-NEXT: movabsq $56, %r10
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux-NEXT: movabsq $0, %r11
; X64-Linux-NEXT: callq __morestack
; X64-Linux-NEXT: ret
; X64-Linux-NEXT: movq %rax, %r10
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI: cmpl %fs:64, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32ABI-NEXT: jbe .LBB1_1
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI: movl %r10d, %eax
; X32ABI-NEXT: movl $56, %r10d
; X32ABI-NEXT: movl $0, %r11d
; X32ABI-NEXT: callq __morestack
; X32ABI-NEXT: ret
; X32ABI-NEXT: movq %rax, %r10
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: movl $432, %edx
; X32-Darwin-NEXT: cmpl %gs:(%edx), %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Darwin-NEXT: jbe LBB1_1
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: pushl $4
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
; X32-Darwin-NEXT: pushl $60
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin-NEXT: calll ___morestack
; X32-Darwin-NEXT: ret
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: cmpq %gs:816, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Darwin-NEXT: jbe LBB1_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: movq %r10, %rax
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
; X64-Darwin-NEXT: movabsq $56, %r10
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin-NEXT: movabsq $0, %r11
; X64-Darwin-NEXT: callq ___morestack
; X64-Darwin-NEXT: ret
; X64-Darwin-NEXT: movq %rax, %r10
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW: cmpl %fs:20, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-MinGW-NEXT: jbe LBB1_1
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW: pushl $4
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-MinGW-NEXT: pushl $44
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW-NEXT: calll ___morestack
; X32-MinGW-NEXT: ret
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW-LABEL: test_nested:
; X64-MinGW: cmpq %gs:40, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-MinGW-NEXT: jbe .LBB1_1
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW: movq %r10, %rax
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
; X64-MinGW-NEXT: movabsq $88, %r10
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW-NEXT: movabsq $32, %r11
; X64-MinGW-NEXT: callq __morestack
; X64-MinGW-NEXT: retq
; X64-MinGW-NEXT: movq %rax, %r10
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: cmpq %fs:24, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-FreeBSD-NEXT: jbe .LBB1_1
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: movq %r10, %rax
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
; X64-FreeBSD-NEXT: movabsq $56, %r10
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD-NEXT: movabsq $0, %r11
; X64-FreeBSD-NEXT: callq __morestack
; X64-FreeBSD-NEXT: ret
; X64-FreeBSD-NEXT: movq %rax, %r10
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD: cmpl %fs:16, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-DFlyBSD-NEXT: jbe .LBB1_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD: pushl $4
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-DFlyBSD-NEXT: pushl $44
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-NEXT: calll __morestack
; X32-DFlyBSD-NEXT: ret
; X64-DFlyBSD: cmpq %fs:32, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-DFlyBSD-NEXT: jbe .LBB1_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X64-DFlyBSD: movq %r10, %rax
; X64-DFlyBSD-NEXT: movabsq $56, %r10
; X64-DFlyBSD-NEXT: movabsq $0, %r11
; X64-DFlyBSD-NEXT: callq __morestack
; X64-DFlyBSD-NEXT: ret
; X64-DFlyBSD-NEXT: movq %rax, %r10
Some test code to check if correct code is being generated. Patch by Sanjoy Das. llvm-svn: 138820
2011-08-31 03:51:29 +08:00
}
Check for stack space more intelligently. libgcc sets the stack limit field in TCB to 256 bytes above the actual allocated stack limit. This means if the function's stack frame needs less than 256 bytes, we can just compare the stack pointer with the stack limit. This should result in lesser calls to __morestack. llvm-svn: 145766
2011-12-03 17:32:07 +08:00
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
define void @test_large() #0 {
Check for stack space more intelligently. libgcc sets the stack limit field in TCB to 256 bytes above the actual allocated stack limit. This means if the function's stack frame needs less than 256 bytes, we can just compare the stack pointer with the stack limit. This should result in lesser calls to __morestack. llvm-svn: 145766
2011-12-03 17:32:07 +08:00
%mem = alloca i32, i32 10000
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
call void @dummy_use (i32* %mem, i32 3)
Check for stack space more intelligently. libgcc sets the stack limit field in TCB to 256 bytes above the actual allocated stack limit. This means if the function's stack frame needs less than 256 bytes, we can just compare the stack pointer with the stack limit. This should result in lesser calls to __morestack. llvm-svn: 145766
2011-12-03 17:32:07 +08:00
ret void
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Linux-LABEL: test_large:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: leal -40012(%esp), %ecx
; X32-Linux-NEXT: cmpl %gs:48, %ecx
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Linux-NEXT: jbe .LBB2_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: pushl $0
; X32-Linux-NEXT: pushl $40012
; X32-Linux-NEXT: calll __morestack
; X32-Linux-NEXT: ret
; X64-Linux: leaq -40008(%rsp), %r11
; X64-Linux-NEXT: cmpq %fs:112, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Linux-NEXT: jbe .LBB2_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux: movabsq $40008, %r10
; X64-Linux-NEXT: movabsq $0, %r11
; X64-Linux-NEXT: callq __morestack
; X64-Linux-NEXT: ret
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI: leal -40008(%rsp), %r11d
; X32ABI-NEXT: cmpl %fs:64, %r11d
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32ABI-NEXT: jbe .LBB2_1
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI: movl $40008, %r10d
; X32ABI-NEXT: movl $0, %r11d
; X32ABI-NEXT: callq __morestack
; X32ABI-NEXT: ret
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: leal -40012(%esp), %ecx
; X32-Darwin-NEXT: movl $432, %eax
; X32-Darwin-NEXT: cmpl %gs:(%eax), %ecx
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Darwin-NEXT: jbe LBB2_1
Check for stack space more intelligently. libgcc sets the stack limit field in TCB to 256 bytes above the actual allocated stack limit. This means if the function's stack frame needs less than 256 bytes, we can just compare the stack pointer with the stack limit. This should result in lesser calls to __morestack. llvm-svn: 145766
2011-12-03 17:32:07 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: pushl $0
; X32-Darwin-NEXT: pushl $40012
; X32-Darwin-NEXT: calll ___morestack
; X32-Darwin-NEXT: ret
Check for stack space more intelligently. libgcc sets the stack limit field in TCB to 256 bytes above the actual allocated stack limit. This means if the function's stack frame needs less than 256 bytes, we can just compare the stack pointer with the stack limit. This should result in lesser calls to __morestack. llvm-svn: 145766
2011-12-03 17:32:07 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: leaq -40008(%rsp), %r11
; X64-Darwin-NEXT: cmpq %gs:816, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Darwin-NEXT: jbe LBB2_1
Check for stack space more intelligently. libgcc sets the stack limit field in TCB to 256 bytes above the actual allocated stack limit. This means if the function's stack frame needs less than 256 bytes, we can just compare the stack pointer with the stack limit. This should result in lesser calls to __morestack. llvm-svn: 145766
2011-12-03 17:32:07 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: movabsq $40008, %r10
; X64-Darwin-NEXT: movabsq $0, %r11
; X64-Darwin-NEXT: callq ___morestack
; X64-Darwin-NEXT: ret
Check for stack space more intelligently. libgcc sets the stack limit field in TCB to 256 bytes above the actual allocated stack limit. This means if the function's stack frame needs less than 256 bytes, we can just compare the stack pointer with the stack limit. This should result in lesser calls to __morestack. llvm-svn: 145766
2011-12-03 17:32:07 +08:00
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-MinGW: leal -40000(%esp), %ecx
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW-NEXT: cmpl %fs:20, %ecx
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-MinGW-NEXT: jbe LBB2_1
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-MinGW-NEXT: pushl $40000
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW-NEXT: calll ___morestack
; X32-MinGW-NEXT: ret
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW-LABEL: test_large:
; X64-MinGW: leaq -40040(%rsp), %r11
; X64-MinGW-NEXT: cmpq %gs:40, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-MinGW-NEXT: jbe .LBB2_1
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW: movabsq $40040, %r10
; X64-MinGW-NEXT: movabsq $32, %r11
; X64-MinGW-NEXT: callq __morestack
; X64-MinGW-NEXT: retq
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: leaq -40008(%rsp), %r11
; X64-FreeBSD-NEXT: cmpq %fs:24, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-FreeBSD-NEXT: jbe .LBB2_1
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: movabsq $40008, %r10
; X64-FreeBSD-NEXT: movabsq $0, %r11
; X64-FreeBSD-NEXT: callq __morestack
; X64-FreeBSD-NEXT: ret
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-DFlyBSD: leal -40000(%esp), %ecx
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-NEXT: cmpl %fs:16, %ecx
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-DFlyBSD-NEXT: jbe .LBB2_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-DFlyBSD-NEXT: pushl $40000
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-NEXT: calll __morestack
; X32-DFlyBSD-NEXT: ret
; X64-DFlyBSD: leaq -40008(%rsp), %r11
; X64-DFlyBSD-NEXT: cmpq %fs:32, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-DFlyBSD-NEXT: jbe .LBB2_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X64-DFlyBSD: movabsq $40008, %r10
; X64-DFlyBSD-NEXT: movabsq $0, %r11
; X64-DFlyBSD-NEXT: callq __morestack
; X64-DFlyBSD-NEXT: ret
Check for stack space more intelligently. libgcc sets the stack limit field in TCB to 256 bytes above the actual allocated stack limit. This means if the function's stack frame needs less than 256 bytes, we can just compare the stack pointer with the stack limit. This should result in lesser calls to __morestack. llvm-svn: 145766
2011-12-03 17:32:07 +08:00
}
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
define fastcc void @test_fastcc() #0 {
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
%mem = alloca i32, i32 10
call void @dummy_use (i32* %mem, i32 10)
ret void
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-Linux-LABEL: test_fastcc:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: cmpl %gs:48, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Linux-NEXT: jbe .LBB3_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-Linux-NEXT: pushl $44
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux-NEXT: calll __morestack
; X32-Linux-NEXT: ret
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-Linux-LABEL: test_fastcc:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux: cmpq %fs:112, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Linux-NEXT: jbe .LBB3_1
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux: movabsq $40, %r10
; X64-Linux-NEXT: movabsq $0, %r11
; X64-Linux-NEXT: callq __morestack
; X64-Linux-NEXT: ret
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI-LABEL: test_fastcc:
; X32ABI: cmpl %fs:64, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32ABI-NEXT: jbe .LBB3_1
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI: movl $40, %r10d
; X32ABI-NEXT: movl $0, %r11d
; X32ABI-NEXT: callq __morestack
; X32ABI-NEXT: ret
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-Darwin-LABEL: test_fastcc:
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: movl $432, %eax
; X32-Darwin-NEXT: cmpl %gs:(%eax), %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Darwin-NEXT: jbe LBB3_1
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: pushl $0
; X32-Darwin-NEXT: pushl $60
; X32-Darwin-NEXT: calll ___morestack
; X32-Darwin-NEXT: ret
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-Darwin-LABEL: test_fastcc:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: cmpq %gs:816, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Darwin-NEXT: jbe LBB3_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: movabsq $40, %r10
; X64-Darwin-NEXT: movabsq $0, %r11
; X64-Darwin-NEXT: callq ___morestack
; X64-Darwin-NEXT: ret
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-MinGW-LABEL: test_fastcc:
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW: cmpl %fs:20, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-MinGW-NEXT: jbe LBB3_1
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-MinGW-NEXT: pushl $40
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW-NEXT: calll ___morestack
; X32-MinGW-NEXT: ret
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW-LABEL: test_fastcc:
; X64-MinGW: cmpq %gs:40, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-MinGW-NEXT: jbe .LBB3_1
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW: movabsq $72, %r10
; X64-MinGW-NEXT: movabsq $32, %r11
; X64-MinGW-NEXT: callq __morestack
; X64-MinGW-NEXT: retq
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-FreeBSD-LABEL: test_fastcc:
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: cmpq %fs:24, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-FreeBSD-NEXT: jbe .LBB3_1
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: movabsq $40, %r10
; X64-FreeBSD-NEXT: movabsq $0, %r11
; X64-FreeBSD-NEXT: callq __morestack
; X64-FreeBSD-NEXT: ret
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-LABEL: test_fastcc:
; X32-DFlyBSD: cmpl %fs:16, %esp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-DFlyBSD-NEXT: jbe .LBB3_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-DFlyBSD-NEXT: pushl $40
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-NEXT: calll __morestack
; X32-DFlyBSD-NEXT: ret
; X64-DFlyBSD-LABEL: test_fastcc:
; X64-DFlyBSD: cmpq %fs:32, %rsp
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-DFlyBSD-NEXT: jbe .LBB3_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X64-DFlyBSD: movabsq $40, %r10
; X64-DFlyBSD-NEXT: movabsq $0, %r11
; X64-DFlyBSD-NEXT: callq __morestack
; X64-DFlyBSD-NEXT: ret
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
}
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
define fastcc void @test_fastcc_large() #0 {
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
%mem = alloca i32, i32 10000
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
call void @dummy_use (i32* %mem, i32 3)
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
ret void
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-Linux-LABEL: test_fastcc_large:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: leal -40012(%esp), %eax
; X32-Linux-NEXT: cmpl %gs:48, %eax
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Linux-NEXT: jbe .LBB4_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Linux: pushl $0
; X32-Linux-NEXT: pushl $40012
; X32-Linux-NEXT: calll __morestack
; X32-Linux-NEXT: ret
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-Linux-LABEL: test_fastcc_large:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux: leaq -40008(%rsp), %r11
; X64-Linux-NEXT: cmpq %fs:112, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Linux-NEXT: jbe .LBB4_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Linux: movabsq $40008, %r10
; X64-Linux-NEXT: movabsq $0, %r11
; X64-Linux-NEXT: callq __morestack
; X64-Linux-NEXT: ret
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI-LABEL: test_fastcc_large:
; X32ABI: leal -40008(%rsp), %r11d
; X32ABI-NEXT: cmpl %fs:64, %r11d
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32ABI-NEXT: jbe .LBB4_1
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI: movl $40008, %r10d
; X32ABI-NEXT: movl $0, %r11d
; X32ABI-NEXT: callq __morestack
; X32ABI-NEXT: ret
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-Darwin-LABEL: test_fastcc_large:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: leal -40012(%esp), %eax
; X32-Darwin-NEXT: movl $432, %ecx
; X32-Darwin-NEXT: cmpl %gs:(%ecx), %eax
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Darwin-NEXT: jbe LBB4_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: pushl $0
; X32-Darwin-NEXT: pushl $40012
; X32-Darwin-NEXT: calll ___morestack
; X32-Darwin-NEXT: ret
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-Darwin-LABEL: test_fastcc_large:
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: leaq -40008(%rsp), %r11
; X64-Darwin-NEXT: cmpq %gs:816, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-Darwin-NEXT: jbe LBB4_1
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X64-Darwin: movabsq $40008, %r10
; X64-Darwin-NEXT: movabsq $0, %r11
; X64-Darwin-NEXT: callq ___morestack
; X64-Darwin-NEXT: ret
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-MinGW-LABEL: test_fastcc_large:
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-MinGW: leal -40000(%esp), %eax
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW-NEXT: cmpl %fs:20, %eax
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-MinGW-NEXT: jbe LBB4_1
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-MinGW-NEXT: pushl $40000
Support segmented stacks on win32. Uses the pvArbitrary slot of the TIB, which is reserved for applications. We only support frames with a static size. llvm-svn: 148040
2012-01-13 04:22:08 +08:00
; X32-MinGW-NEXT: calll ___morestack
; X32-MinGW-NEXT: ret
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW-LABEL: test_fastcc_large:
; X64-MinGW: leaq -40040(%rsp), %r11
; X64-MinGW-NEXT: cmpq %gs:40, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-MinGW-NEXT: jbe .LBB4_1
Support segmented stacks on Win64 Identical to Win32 method except the GS segment register is used for TLS instead of FS and pvArbitrary is at TEB offset 0x28 instead of 0x14. llvm-svn: 205342
2014-04-02 02:34:21 +08:00
; X64-MinGW: movabsq $40040, %r10
; X64-MinGW-NEXT: movabsq $32, %r11
; X64-MinGW-NEXT: callq __morestack
; X64-MinGW-NEXT: retq
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X64-FreeBSD-LABEL: test_fastcc_large:
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: leaq -40008(%rsp), %r11
; X64-FreeBSD-NEXT: cmpq %fs:24, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-FreeBSD-NEXT: jbe .LBB4_1
Support segmented stacks on 64-bit FreeBSD. This patch uses tcb_spare field in the tcb structure to store info. Patch by Jyun-Yan You. llvm-svn: 148041
2012-01-13 04:24:30 +08:00
; X64-FreeBSD: movabsq $40008, %r10
; X64-FreeBSD-NEXT: movabsq $0, %r11
; X64-FreeBSD-NEXT: callq __morestack
; X64-FreeBSD-NEXT: ret
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-LABEL: test_fastcc_large:
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-DFlyBSD: leal -40000(%esp), %eax
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-NEXT: cmpl %fs:16, %eax
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-DFlyBSD-NEXT: jbe .LBB4_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD: pushl $0
[X86] Enable call frame optimization ("mov to push") not only for optsize (PR26325) The size savings are significant, and from what I can tell, both ICC and GCC do this. Differential Revision: http://reviews.llvm.org/D18573 llvm-svn: 264966
2016-03-31 07:38:01 +08:00
; X32-DFlyBSD-NEXT: pushl $40000
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-NEXT: calll __morestack
; X32-DFlyBSD-NEXT: ret
; X64-DFlyBSD-LABEL: test_fastcc_large:
; X64-DFlyBSD: leaq -40008(%rsp), %r11
; X64-DFlyBSD-NEXT: cmpq %fs:32, %r11
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X64-DFlyBSD-NEXT: jbe .LBB4_1
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X64-DFlyBSD: movabsq $40008, %r10
; X64-DFlyBSD-NEXT: movabsq $0, %r11
; X64-DFlyBSD-NEXT: callq __morestack
; X64-DFlyBSD-NEXT: ret
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
}
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
define fastcc void @test_fastcc_large_with_ecx_arg(i32 %a) #0 {
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
%mem = alloca i32, i32 10000
call void @dummy_use (i32* %mem, i32 %a)
ret void
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; This is testing that the Mac implementation preserves ecx
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Mass update to CodeGen tests to use CHECK-LABEL for labels corresponding to function definitions for more informative error messages. No functionality change and all updated tests passed locally. This update was done with the following bash script: find test/CodeGen -name "*.ll" | \ while read NAME; do echo "$NAME" if ! grep -q "^; *RUN: *llc.*debug" $NAME; then TEMP=`mktemp -t temp` cp $NAME $TEMP sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \ while read FUNC; do sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP done sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP mv $TEMP $NAME fi done llvm-svn: 186280
2013-07-14 14:24:09 +08:00
; X32-Darwin-LABEL: test_fastcc_large_with_ecx_arg:
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: leal -40012(%esp), %eax
; X32-Darwin-NEXT: pushl %ecx
; X32-Darwin-NEXT: movl $432, %ecx
; X32-Darwin-NEXT: cmpl %gs:(%ecx), %eax
; X32-Darwin-NEXT: popl %ecx
[X86] improve split-stack machine BB placement Summary: The conditional branch created to support -fsplit-stack for X86 is left unbiased/unhinted, resulting in less than ideal block placement: the __morestack call block is kept on the main hot path. Bias the branch to insure that the stack allocation block is treated as a "cold" block during machine basic block placement. Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D54123 llvm-svn: 346336
2018-11-08 01:41:57 +08:00
; X32-Darwin-NEXT: jbe LBB5_1
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
Support segmented stacks on mac. This uses TLS slot 90, which actually belongs to JavaScriptCore. We only support frames with static size Patch by Brian Anderson. llvm-svn: 147960
2012-01-12 03:00:37 +08:00
; X32-Darwin: pushl $0
; X32-Darwin-NEXT: pushl $40012
; X32-Darwin-NEXT: calll ___morestack
; X32-Darwin-NEXT: ret
Generate the segmented stack prologue for fastcc too. Patch by Brian Anderson. llvm-svn: 147958
2012-01-12 02:41:19 +08:00
}
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
define void @test_nostack() #0 {
ret void
; X32-Linux-LABEL: test_nostack:
; X32-Linux-NOT: calll __morestack
; X64-Linux-LABEL: test_nostack:
; X32-Linux-NOT: callq __morestack
[x32] Fix segmented stacks support Summary: Update segmented-stacks*.ll tests with x32 target case and make corresponding changes to make them pass. Test Plan: tests updated with x32 target Reviewers: nadav, rafael, dschuff Subscribers: llvm-commits, zinovy.nis Differential Revision: http://reviews.llvm.org/D5245 llvm-svn: 218247
2014-09-22 21:11:35 +08:00
; X32ABI-LABEL: test_nostack:
; X32ABI-NOT: callq __morestack
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
; X32-Darwin-LABEL: test_nostack:
; X32-Darwin-NOT: calll __morestack
; X64-Darwin-LABEL: test_nostack:
; X64-Darwin-NOT: callq __morestack
; X32-MinGW-LABEL: test_nostack:
; X32-MinGW-NOT: calll __morestack
; X64-MinGW-LABEL: test_nostack:
; X64-MinGW-NOT: callq __morestack
; X64-FreeBSD-LABEL: test_nostack:
; X64-FreeBSD-NOT: callq __morestack
Add segmented stack support for DragonFlyBSD. Patch by Michael Neumann. llvm-svn: 224936
2014-12-29 23:47:28 +08:00
; X32-DFlyBSD-LABEL: test_nostack:
; X32-DFlyBSD-NOT: calll __morestack
; X64-DFlyBSD-LABEL: test_nostack:
; X64-DFlyBSD-NOT: callq __morestack
Segmented stacks: omit __morestack call when there's no frame. Patch by Florian Zeitz llvm-svn: 209436
2014-05-22 21:03:43 +08:00
}
[CodeGen] Emit necessary .note sections for -fsplit-stack Summary: According to https://gcc.gnu.org/wiki/SplitStacks, the linker expects a zero-sized .note.GNU-split-stack section if split-stack is used (and also .note.GNU-no-split-stack section if it also contains non-split-stack functions), so it can handle the cases where a split-stack function calls non-split-stack function. This change adds the sections if needed. Fixes PR #34670. Reviewers: thanm, rnk, luqmana Reviewed By: rnk Subscribers: llvm-commits Patch by Cherry Zhang <cherryyz@google.com> Differential Revision: https://reviews.llvm.org/D38051 llvm-svn: 314335
2017-09-28 03:34:00 +08:00
define void @test_nosplitstck() {
ret void
}
[X86,ARM] Retain split-stack prolog check for sibling calls Summary: If a routine with no stack frame makes a sibling call, we need to preserve the stack space check even if the local stack frame is empty, since the call target could be a "no-split" function (in which case the linker needs to be able to fix up the prolog sequence in order to switch to a larger stack). This fixes PR37807. Reviewers: cherry, javed.absar Subscribers: srhines, llvm-commits Differential Revision: https://reviews.llvm.org/D48444 llvm-svn: 335604
2018-06-26 22:11:30 +08:00
; Test to make sure that a morestack call is generated if there is a
; sibling call, even if the function in question has no stack frame
; (PR37807).
declare i32 @callee(i32)
define i32 @test_sibling_call_empty_frame(i32 %x) #0 {
%call = tail call i32 @callee(i32 %x) #0
ret i32 %call
; X32-Linux-LABEL: test_sibling_call_empty_frame:
; X32-Linux: calll __morestack
; X64-Linux-LABEL: test_sibling_call_empty_frame:
; X64-Linux: callq __morestack
; X64-Linux-Large-LABEL: test_sibling_call_empty_frame:
; X64-Linux-Large: callq *__morestack_addr(%rip)
; X32ABI-LABEL: test_sibling_call_empty_frame:
; X32ABI: callq __morestack
; X32-Darwin-LABEL: test_sibling_call_empty_frame:
; X32-Darwin: calll ___morestack
; X64-Darwin-LABEL: test_sibling_call_empty_frame:
; X64-Darwin: callq ___morestack
; X32-MinGW-LABEL: test_sibling_call_empty_frame:
; X32-MinGW: calll ___morestack
; X64-MinGW-LABEL: test_sibling_call_empty_frame:
; X64-MinGW: callq __morestack
; X64-FreeBSD-LABEL: test_sibling_call_empty_frame:
; X64-FreeBSD: callq __morestack
; X32-DFlyBSD-LABEL: test_sibling_call_empty_frame:
; X32-DFlyBSD: calll __morestack
; X32-DFlyBSD-NEXT: ret
; X64-DFlyBSD-LABEL: test_sibling_call_empty_frame:
; X64-DFlyBSD: callq __morestack
}
Move the segmented stack switch to a function attribute This removes the -segmented-stacks command line flag in favor of a per-function "split-stack" attribute. Patch by Luqman Aden and Alex Crichton! llvm-svn: 205997
2014-04-11 06:58:43 +08:00
attributes #0 = { "split-stack" }
x86_64: Fix calls to __morestack under the large code model. Under the large code model, we cannot assume that __morestack lives within 2^31 bytes of the call site, so we cannot use pc-relative addressing. We cannot perform the call via a temporary register, as the rax register may be used to store the static chain, and all other suitable registers may be either callee-save or used for parameter passing. We cannot use the stack at this point either because __morestack manipulates the stack directly. To avoid these issues, perform an indirect call via a read-only memory location containing the address. This solution is not perfect, as it assumes that the .rodata section is laid out within 2^31 bytes of each function body, but this seems to be sufficient for JIT. Differential Revision: http://reviews.llvm.org/D6787 llvm-svn: 225003
2014-12-31 04:05:19 +08:00
; X64-Linux-Large: .rodata
; X64-Linux-Large-NEXT: __morestack_addr:
; X64-Linux-Large-NEXT: .quad __morestack
[CodeGen] Emit necessary .note sections for -fsplit-stack Summary: According to https://gcc.gnu.org/wiki/SplitStacks, the linker expects a zero-sized .note.GNU-split-stack section if split-stack is used (and also .note.GNU-no-split-stack section if it also contains non-split-stack functions), so it can handle the cases where a split-stack function calls non-split-stack function. This change adds the sections if needed. Fixes PR #34670. Reviewers: thanm, rnk, luqmana Reviewed By: rnk Subscribers: llvm-commits Patch by Cherry Zhang <cherryyz@google.com> Differential Revision: https://reviews.llvm.org/D38051 llvm-svn: 314335
2017-09-28 03:34:00 +08:00
; X32-Linux: .section ".note.GNU-split-stack","",@progbits
; X32-Linux: .section ".note.GNU-no-split-stack","",@progbits
; X64-Linux: .section ".note.GNU-split-stack","",@progbits
; X64-Linux: .section ".note.GNU-no-split-stack","",@progbits
; X64-FreeBSD: .section ".note.GNU-split-stack","",@progbits
; X64-FreeBSD: .section ".note.GNU-no-split-stack","",@progbits
; X32-DFlyBSD: .section ".note.GNU-split-stack","",@progbits
; X32-DFlyBSD: .section ".note.GNU-no-split-stack","",@progbits
; X64-DFlyBSD: .section ".note.GNU-split-stack","",@progbits
; X64-DFlyBSD: .section ".note.GNU-no-split-stack","",@progbits