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/ARM/fast-isel-ext.ll

138 lines
3.3 KiB
LLVM
Raw Normal View History

ARM FastISel integer sext/zext improvements My recent ARM FastISel patch exposed this bug: http://llvm.org/bugs/show_bug.cgi?id=16178 The root cause is that it can't select integer sext/zext pre-ARMv6 and asserts out. The current integer sext/zext code doesn't handle other cases gracefully either, so this patch makes it handle all sext and zext from i1/i8/i16 to i8/i16/i32, with and without ARMv6, both in Thumb and ARM mode. This should fix the bug as well as make FastISel faster because it bails to SelectionDAG less often. See fastisel-ext.patch for this. fastisel-ext-tests.patch changes current tests to always use reg-imm AND for 8-bit zext instead of UXTB. This simplifies code since it is supported on ARMv4t and later, and at least on A15 both should perform exactly the same (both have exec 1 uop 1, type I). 2013-05-31-char-shift-crash.ll is a bitcode version of the above bug 16178 repro. fast-isel-ext.ll tests all sext/zext combinations that ARM FastISel should now handle. Note that my ARM FastISel enabling patch was reverted due to a separate failure when dealing with MCJIT, I'll fix this second failure and then turn FastISel on again for non-iOS ARM targets. I've tested "make check-all" on my x86 box, and "lnt test-suite" on A15 hardware. llvm-svn: 183551
2013-06-08 04:10:37 +08:00
; RUN: llc < %s -O0 -fast-isel-abort -mtriple=armv7-apple-ios | FileCheck %s --check-prefix=v7
Enable FastISel on ARM for Linux and NaCl, not MCJIT This is a resubmit of r182877, which was reverted because it broken MCJIT tests on ARM. The patch leaves MCJIT on ARM as it was before: only enabled for iOS. I've CC'ed people from the original review and revert. FastISel was only enabled for iOS ARM and Thumb2, this patch enables it for ARM (not Thumb2) on Linux and NaCl, but not MCJIT. Thumb2 support needs a bit more work, mainly around register class restrictions. The patch punts to SelectionDAG when doing TLS relocation on non-Darwin targets. I will fix this and other FastISel-to-SelectionDAG failures in a separate patch. The patch also forces FastISel to retain frame pointers: iOS always keeps them for backtracking (so emitted code won't change because of this), but Linux was getting much worse code that was incorrect when using big frames (such as test-suite's lencod). I'll also fix this in a later patch, it will probably require a peephole so that FastISel doesn't rematerialize frame pointers back-to-back. The test changes are straightforward, similar to: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20130513/174279.html They also add a vararg test that got dropped in that change. I ran all of lnt test-suite on A15 hardware with --optimize-option=-O0 and all the tests pass. All the tests also pass on x86 make check-all. I also re-ran the check-all tests that failed on ARM, and they all seem to pass. llvm-svn: 183966
2013-06-14 10:49:43 +08:00
; RUN: llc < %s -O0 -fast-isel-abort -mtriple=armv7-linux-gnueabi | FileCheck %s --check-prefix=v7
ARM FastISel integer sext/zext improvements My recent ARM FastISel patch exposed this bug: http://llvm.org/bugs/show_bug.cgi?id=16178 The root cause is that it can't select integer sext/zext pre-ARMv6 and asserts out. The current integer sext/zext code doesn't handle other cases gracefully either, so this patch makes it handle all sext and zext from i1/i8/i16 to i8/i16/i32, with and without ARMv6, both in Thumb and ARM mode. This should fix the bug as well as make FastISel faster because it bails to SelectionDAG less often. See fastisel-ext.patch for this. fastisel-ext-tests.patch changes current tests to always use reg-imm AND for 8-bit zext instead of UXTB. This simplifies code since it is supported on ARMv4t and later, and at least on A15 both should perform exactly the same (both have exec 1 uop 1, type I). 2013-05-31-char-shift-crash.ll is a bitcode version of the above bug 16178 repro. fast-isel-ext.ll tests all sext/zext combinations that ARM FastISel should now handle. Note that my ARM FastISel enabling patch was reverted due to a separate failure when dealing with MCJIT, I'll fix this second failure and then turn FastISel on again for non-iOS ARM targets. I've tested "make check-all" on my x86 box, and "lnt test-suite" on A15 hardware. llvm-svn: 183551
2013-06-08 04:10:37 +08:00
; RUN: llc < %s -O0 -fast-isel-abort -mtriple=armv4t-apple-ios | FileCheck %s --check-prefix=prev6
Enable FastISel on ARM for Linux and NaCl, not MCJIT This is a resubmit of r182877, which was reverted because it broken MCJIT tests on ARM. The patch leaves MCJIT on ARM as it was before: only enabled for iOS. I've CC'ed people from the original review and revert. FastISel was only enabled for iOS ARM and Thumb2, this patch enables it for ARM (not Thumb2) on Linux and NaCl, but not MCJIT. Thumb2 support needs a bit more work, mainly around register class restrictions. The patch punts to SelectionDAG when doing TLS relocation on non-Darwin targets. I will fix this and other FastISel-to-SelectionDAG failures in a separate patch. The patch also forces FastISel to retain frame pointers: iOS always keeps them for backtracking (so emitted code won't change because of this), but Linux was getting much worse code that was incorrect when using big frames (such as test-suite's lencod). I'll also fix this in a later patch, it will probably require a peephole so that FastISel doesn't rematerialize frame pointers back-to-back. The test changes are straightforward, similar to: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20130513/174279.html They also add a vararg test that got dropped in that change. I ran all of lnt test-suite on A15 hardware with --optimize-option=-O0 and all the tests pass. All the tests also pass on x86 make check-all. I also re-ran the check-all tests that failed on ARM, and they all seem to pass. llvm-svn: 183966
2013-06-14 10:49:43 +08:00
; RUN: llc < %s -O0 -fast-isel-abort -mtriple=armv4t-linux-gnueabi | FileCheck %s --check-prefix=prev6
ARM FastISel integer sext/zext improvements My recent ARM FastISel patch exposed this bug: http://llvm.org/bugs/show_bug.cgi?id=16178 The root cause is that it can't select integer sext/zext pre-ARMv6 and asserts out. The current integer sext/zext code doesn't handle other cases gracefully either, so this patch makes it handle all sext and zext from i1/i8/i16 to i8/i16/i32, with and without ARMv6, both in Thumb and ARM mode. This should fix the bug as well as make FastISel faster because it bails to SelectionDAG less often. See fastisel-ext.patch for this. fastisel-ext-tests.patch changes current tests to always use reg-imm AND for 8-bit zext instead of UXTB. This simplifies code since it is supported on ARMv4t and later, and at least on A15 both should perform exactly the same (both have exec 1 uop 1, type I). 2013-05-31-char-shift-crash.ll is a bitcode version of the above bug 16178 repro. fast-isel-ext.ll tests all sext/zext combinations that ARM FastISel should now handle. Note that my ARM FastISel enabling patch was reverted due to a separate failure when dealing with MCJIT, I'll fix this second failure and then turn FastISel on again for non-iOS ARM targets. I've tested "make check-all" on my x86 box, and "lnt test-suite" on A15 hardware. llvm-svn: 183551
2013-06-08 04:10:37 +08:00
; RUN: llc < %s -O0 -fast-isel-abort -mtriple=armv5-apple-ios | FileCheck %s --check-prefix=prev6
Enable FastISel on ARM for Linux and NaCl, not MCJIT This is a resubmit of r182877, which was reverted because it broken MCJIT tests on ARM. The patch leaves MCJIT on ARM as it was before: only enabled for iOS. I've CC'ed people from the original review and revert. FastISel was only enabled for iOS ARM and Thumb2, this patch enables it for ARM (not Thumb2) on Linux and NaCl, but not MCJIT. Thumb2 support needs a bit more work, mainly around register class restrictions. The patch punts to SelectionDAG when doing TLS relocation on non-Darwin targets. I will fix this and other FastISel-to-SelectionDAG failures in a separate patch. The patch also forces FastISel to retain frame pointers: iOS always keeps them for backtracking (so emitted code won't change because of this), but Linux was getting much worse code that was incorrect when using big frames (such as test-suite's lencod). I'll also fix this in a later patch, it will probably require a peephole so that FastISel doesn't rematerialize frame pointers back-to-back. The test changes are straightforward, similar to: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20130513/174279.html They also add a vararg test that got dropped in that change. I ran all of lnt test-suite on A15 hardware with --optimize-option=-O0 and all the tests pass. All the tests also pass on x86 make check-all. I also re-ran the check-all tests that failed on ARM, and they all seem to pass. llvm-svn: 183966
2013-06-14 10:49:43 +08:00
; RUN: llc < %s -O0 -fast-isel-abort -mtriple=armv5-linux-gnueabi | FileCheck %s --check-prefix=prev6
ARM FastISel integer sext/zext improvements My recent ARM FastISel patch exposed this bug: http://llvm.org/bugs/show_bug.cgi?id=16178 The root cause is that it can't select integer sext/zext pre-ARMv6 and asserts out. The current integer sext/zext code doesn't handle other cases gracefully either, so this patch makes it handle all sext and zext from i1/i8/i16 to i8/i16/i32, with and without ARMv6, both in Thumb and ARM mode. This should fix the bug as well as make FastISel faster because it bails to SelectionDAG less often. See fastisel-ext.patch for this. fastisel-ext-tests.patch changes current tests to always use reg-imm AND for 8-bit zext instead of UXTB. This simplifies code since it is supported on ARMv4t and later, and at least on A15 both should perform exactly the same (both have exec 1 uop 1, type I). 2013-05-31-char-shift-crash.ll is a bitcode version of the above bug 16178 repro. fast-isel-ext.ll tests all sext/zext combinations that ARM FastISel should now handle. Note that my ARM FastISel enabling patch was reverted due to a separate failure when dealing with MCJIT, I'll fix this second failure and then turn FastISel on again for non-iOS ARM targets. I've tested "make check-all" on my x86 box, and "lnt test-suite" on A15 hardware. llvm-svn: 183551
2013-06-08 04:10:37 +08:00
; RUN: llc < %s -O0 -fast-isel-abort -mtriple=thumbv7-apple-ios | FileCheck %s --check-prefix=v7
; Can't test pre-ARMv6 Thumb because ARM FastISel currently only supports
; Thumb2. The ARMFastISel::ARMEmitIntExt code should work for Thumb by always
; using two shifts.
; Note that lsl, asr and lsr in Thumb are all encoded as 16-bit instructions
; and therefore must set flags. {{s?}} below denotes this, instead of
; duplicating tests.
; zext
define i8 @zext_1_8(i1 %a) nounwind ssp {
; v7: zext_1_8:
; v7: and r0, r0, #1
; prev6: zext_1_8:
; prev6: and r0, r0, #1
%r = zext i1 %a to i8
ret i8 %r
}
define i16 @zext_1_16(i1 %a) nounwind ssp {
; v7: zext_1_16:
; v7: and r0, r0, #1
; prev6: zext_1_16:
; prev6: and r0, r0, #1
%r = zext i1 %a to i16
ret i16 %r
}
define i32 @zext_1_32(i1 %a) nounwind ssp {
; v7: zext_1_32:
; v7: and r0, r0, #1
; prev6: zext_1_32:
; prev6: and r0, r0, #1
%r = zext i1 %a to i32
ret i32 %r
}
define i16 @zext_8_16(i8 %a) nounwind ssp {
; v7: zext_8_16:
; v7: and r0, r0, #255
; prev6: zext_8_16:
; prev6: and r0, r0, #255
%r = zext i8 %a to i16
ret i16 %r
}
define i32 @zext_8_32(i8 %a) nounwind ssp {
; v7: zext_8_32:
; v7: and r0, r0, #255
; prev6: zext_8_32:
; prev6: and r0, r0, #255
%r = zext i8 %a to i32
ret i32 %r
}
define i32 @zext_16_32(i16 %a) nounwind ssp {
; v7: zext_16_32:
; v7: uxth r0, r0
; prev6: zext_16_32:
; prev6: lsl{{s?}} r0, r0, #16
; prev6: lsr{{s?}} r0, r0, #16
%r = zext i16 %a to i32
ret i32 %r
}
; sext
define i8 @sext_1_8(i1 %a) nounwind ssp {
; v7: sext_1_8:
; v7: lsl{{s?}} r0, r0, #31
; v7: asr{{s?}} r0, r0, #31
; prev6: sext_1_8:
; prev6: lsl{{s?}} r0, r0, #31
; prev6: asr{{s?}} r0, r0, #31
%r = sext i1 %a to i8
ret i8 %r
}
define i16 @sext_1_16(i1 %a) nounwind ssp {
; v7: sext_1_16:
; v7: lsl{{s?}} r0, r0, #31
; v7: asr{{s?}} r0, r0, #31
; prev6: sext_1_16:
; prev6: lsl{{s?}} r0, r0, #31
; prev6: asr{{s?}} r0, r0, #31
%r = sext i1 %a to i16
ret i16 %r
}
define i32 @sext_1_32(i1 %a) nounwind ssp {
; v7: sext_1_32:
; v7: lsl{{s?}} r0, r0, #31
; v7: asr{{s?}} r0, r0, #31
; prev6: sext_1_32:
; prev6: lsl{{s?}} r0, r0, #31
; prev6: asr{{s?}} r0, r0, #31
%r = sext i1 %a to i32
ret i32 %r
}
define i16 @sext_8_16(i8 %a) nounwind ssp {
; v7: sext_8_16:
; v7: sxtb r0, r0
; prev6: sext_8_16:
; prev6: lsl{{s?}} r0, r0, #24
; prev6: asr{{s?}} r0, r0, #24
%r = sext i8 %a to i16
ret i16 %r
}
define i32 @sext_8_32(i8 %a) nounwind ssp {
; v7: sext_8_32:
; v7: sxtb r0, r0
; prev6: sext_8_32:
; prev6: lsl{{s?}} r0, r0, #24
; prev6: asr{{s?}} r0, r0, #24
%r = sext i8 %a to i32
ret i32 %r
}
define i32 @sext_16_32(i16 %a) nounwind ssp {
; v7: sext_16_32:
; v7: sxth r0, r0
; prev6: sext_16_32:
; prev6: lsl{{s?}} r0, r0, #16
; prev6: asr{{s?}} r0, r0, #16
%r = sext i16 %a to i32
ret i32 %r
}