[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
This revision was automatically updated to reflect the committed changes.
Closed by commit rL348930: [CUDA][OPENMP][NVPTX]Improve logic of the debug info
support. (authored by ABataev, committed by ).
Herald added a subscriber: llvm-commits.
Repository:
rL LLVM
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D51554/new/
https://reviews.llvm.org/D51554
Files:
cfe/trunk/include/clang/Driver/ToolChain.h
cfe/trunk/lib/Driver/ToolChains/Clang.cpp
cfe/trunk/lib/Driver/ToolChains/Cuda.cpp
cfe/trunk/lib/Driver/ToolChains/Cuda.h
cfe/trunk/test/Driver/cuda-dwarf-2.cu
cfe/trunk/test/Driver/openmp-offload-gpu.c
Index: cfe/trunk/include/clang/Driver/ToolChain.h
===
--- cfe/trunk/include/clang/Driver/ToolChain.h
+++ cfe/trunk/include/clang/Driver/ToolChain.h
@@ -10,6 +10,7 @@
#ifndef LLVM_CLANG_DRIVER_TOOLCHAIN_H
#define LLVM_CLANG_DRIVER_TOOLCHAIN_H
+#include "clang/Basic/DebugInfoOptions.h"
#include "clang/Basic/LLVM.h"
#include "clang/Basic/Sanitizers.h"
#include "clang/Basic/DebugInfoOptions.h"
@@ -435,6 +436,10 @@
return true;
}
+ /// Adjust debug information kind considering all passed options.
+ virtual void adjustDebugInfoKind(codegenoptions::DebugInfoKind ,
+ const llvm::opt::ArgList ) const {}
+
/// GetExceptionModel - Return the tool chain exception model.
virtual llvm::ExceptionHandling
GetExceptionModel(const llvm::opt::ArgList ) const;
Index: cfe/trunk/test/Driver/cuda-dwarf-2.cu
===
--- cfe/trunk/test/Driver/cuda-dwarf-2.cu
+++ cfe/trunk/test/Driver/cuda-dwarf-2.cu
@@ -1,25 +1,28 @@
// REQUIRES: clang-driver
//
-// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g -O0 --no-cuda-noopt-device-debug 2>&1 | \
-// RUN: FileCheck %s -check-prefix NO_DEBUG
+// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g -O1 --no-cuda-noopt-device-debug 2>&1 | \
+// RUN: FileCheck %s -check-prefix DEBUG_DIRECTIVES
// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g -O3 2>&1 | \
-// RUN: FileCheck %s -check-prefix NO_DEBUG
+// RUN: FileCheck %s -check-prefix DEBUG_DIRECTIVES
// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g -O3 --no-cuda-noopt-device-debug 2>&1 | \
-// RUN: FileCheck %s -check-prefix NO_DEBUG
+// RUN: FileCheck %s -check-prefix DEBUG_DIRECTIVES
// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g0 2>&1 | \
// RUN: FileCheck %s -check-prefix NO_DEBUG
// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -ggdb0 -O3 --cuda-noopt-device-debug 2>&1 | \
// RUN: FileCheck %s -check-prefix NO_DEBUG
-// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -ggdb1 2>&1 | \
-// RUN: FileCheck %s -check-prefix NO_DEBUG -check-prefix LINE_TABLE
-// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -gline-tables-only -O2 --cuda-noopt-device-debug 2>&1 | \
-// RUN: FileCheck %s -check-prefix NO_DEBUG -check-prefix LINE_TABLE
+// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -gline-directives-only -O2 --cuda-noopt-device-debug 2>&1 | \
+// RUN: FileCheck %s -check-prefix DEBUG_DIRECTIVES
// NO_DEBUG-NOT: warning: debug
-// LINE_TABLE-NOT: warning: debug
+// DEBUG_DIRECTIVES-NOT: warning: debug
+// NO_DEBUG: "-fcuda-is-device"
+// NO_DEBUG-NOT: "-debug-info-kind=
// NO_DEBUG: ptxas
// NO_DEBUG-NOT: "-g"
-// LINE_TABLE: "-lineinfo"
+// DEBUG_DIRECTIVES: "-fcuda-is-device"
+// DEBUG_DIRECTIVES-SAME: "-debug-info-kind=line-directives-only"
+// DEBUG_DIRECTIVES: ptxas
+// DEBUG_DIRECTIVES-SAME: "-lineinfo"
// NO_DEBUG: fatbinary
// NO_DEBUG-NOT: "-g"
@@ -27,6 +30,8 @@
// RUN: FileCheck %s -check-prefix HAS_DEBUG
// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g -O0 --cuda-noopt-device-debug 2>&1 | \
// RUN: FileCheck %s -check-prefix HAS_DEBUG
+// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g -O0 --no-cuda-noopt-device-debug 2>&1 | \
+// RUN: FileCheck %s -check-prefix HAS_DEBUG
// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g -O3 --cuda-noopt-device-debug 2>&1 | \
// RUN: FileCheck %s -check-prefix HAS_DEBUG
// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -g2 2>&1 | \
@@ -37,9 +42,14 @@
// RUN: FileCheck %s -check-prefix HAS_DEBUG
// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -ggdb3 -O3 --cuda-noopt-device-debug 2>&1 | \
// RUN: FileCheck %s -check-prefix HAS_DEBUG
+// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -ggdb1 2>&1 | \
+// RUN: FileCheck %s -check-prefix HAS_DEBUG
+// RUN: %clang -### -target x86_64-linux-gnu -c --cuda-gpu-arch=sm_20 %s -gline-tables-only -O2 --cuda-noopt-device-debug 2>&1 | \
+//
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
echristo accepted this revision. echristo added a comment. LGTM. I'm quite a bit happier with this now. Thanks for going through the back and forth. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D51554/new/ https://reviews.llvm.org/D51554 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
ABataev updated this revision to Diff 176993. ABataev marked an inline comment as done. ABataev added a comment. Reworked according to the latest comments from Eric. Repository: rC Clang CHANGES SINCE LAST ACTION https://reviews.llvm.org/D51554/new/ https://reviews.llvm.org/D51554 Files: include/clang/Driver/ToolChain.h lib/Driver/ToolChains/Clang.cpp lib/Driver/ToolChains/Cuda.cpp lib/Driver/ToolChains/Cuda.h test/Driver/cuda-dwarf-2.cu test/Driver/openmp-offload-gpu.c Index: test/Driver/openmp-offload-gpu.c === --- test/Driver/openmp-offload-gpu.c +++ test/Driver/openmp-offload-gpu.c @@ -190,29 +190,35 @@ // CHK-BCLIB-WARN: No library 'libomptarget-nvptx-sm_20.bc' found in the default clang lib directory or in LIBRARY_PATH. Expect degraded performance due to no inlining of runtime functions on target devices. /// Check that debug info is emitted in dwarf-2 -// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --no-cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O1 --no-cuda-noopt-device-debug 2>&1 \ +// RUN: | FileCheck -check-prefix=DEBUG_DIRECTIVES %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O3 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: | FileCheck -check-prefix=DEBUG_DIRECTIVES %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O3 --no-cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: | FileCheck -check-prefix=DEBUG_DIRECTIVES %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g0 2>&1 \ // RUN: | FileCheck -check-prefix=NO_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb0 -O3 --cuda-noopt-device-debug 2>&1 \ // RUN: | FileCheck -check-prefix=NO_DEBUG %s -// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -gline-tables-only 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG -check-prefix=LINE_TABLE %s -// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb1 -O2 --cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG -check-prefix=LINE_TABLE %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -gline-directives-only 2>&1 \ +// RUN: | FileCheck -check-prefix=DEBUG_DIRECTIVES %s -// LINE_TABLE-NOT: warning: debug +// DEBUG_DIRECTIVES-NOT: warning: debug // NO_DEBUG-NOT: warning: debug +// NO_DEBUG: "-fopenmp-is-device" +// NO_DEBUG-NOT: "-debug-info-kind= // NO_DEBUG: ptxas -// LINE_TABLE: "-lineinfo" +// DEBUG_DIRECTIVES: "-triple" "nvptx64-nvidia-cuda" +// DEBUG_DIRECTIVES-SAME: "-debug-info-kind=line-directives-only" +// DEBUG_DIRECTIVES-SAME: "-fopenmp-is-device" +// DEBUG_DIRECTIVES: ptxas +// DEBUG_DIRECTIVES: "-lineinfo" // NO_DEBUG-NOT: "-g" // NO_DEBUG: nvlink // NO_DEBUG-NOT: "-g" +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --no-cuda-noopt-device-debug 2>&1 \ +// RUN: | FileCheck -check-prefix=HAS_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g 2>&1 \ // RUN: | FileCheck -check-prefix=HAS_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --cuda-noopt-device-debug 2>&1 \ @@ -227,9 +233,14 @@ // RUN: | FileCheck -check-prefix=HAS_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb3 -O2 --cuda-noopt-device-debug 2>&1 \ // RUN: | FileCheck -check-prefix=HAS_DEBUG %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -gline-tables-only 2>&1 \ +// RUN: | FileCheck -check-prefix=HAS_DEBUG %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb1 -O2 --cuda-noopt-device-debug 2>&1 \ +// RUN: | FileCheck -check-prefix=HAS_DEBUG %s // HAS_DEBUG-NOT: warning: debug // HAS_DEBUG: "-triple" "nvptx64-nvidia-cuda" +//
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
ABataev marked 4 inline comments as done.
ABataev added inline comments.
Comment at: lib/Driver/ToolChains/Cuda.cpp:282-285
enum DebugInfoKind {
- NoDebug, /// No debug info.
- LineTableOnly, /// Line tables only.
- FullDebug /// Full debug info.
+ NoDebug, /// No debug info.
+ DebugDirectivesOnly, /// Line tables only.
+ FullDebug, /// Full debug info.
echristo wrote:
> This enum doesn't appear to be complete? Either way can you make it match the
> other and document what each thing means a bit more?
No, it is complete, but probably has some wrong names. I reworked it. Actually,
this enum is intended to track the debug info emitted for the device. It may be
disabled, debug directives only or same debug info as for the host.
Comment at: lib/Driver/ToolChains/Cuda.cpp:289
static DebugInfoKind mustEmitDebugInfo(const ArgList ) {
+ const Arg *A = Args.getLastArg(options::OPT_O_Group);
echristo wrote:
> Please document this routine in prose.
Added description.
Comment at: lib/Driver/ToolChains/Cuda.cpp:706-708
+void CudaToolChain::adjustDebugInfoKind(
+codegenoptions::DebugInfoKind , const ArgList ) const {
+ switch (mustEmitDebugInfo(Args)) {
echristo wrote:
> Is this really doing anything?
Yes, actually it does. Currently, when we need to emit the code for the device,
we use the same debug info level just like for the host. But in some
situations, we need to disable it or emit only debug directives for the device,
while keeping the original debug info for the host. This function allows us to
change the debug info level for the device and force clang to emit required
debug info data during codegen for the NVPTX devices.
Repository:
rC Clang
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D51554/new/
https://reviews.llvm.org/D51554
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
echristo added inline comments.
Comment at: lib/Driver/ToolChains/Cuda.cpp:282-285
enum DebugInfoKind {
- NoDebug, /// No debug info.
- LineTableOnly, /// Line tables only.
- FullDebug /// Full debug info.
+ NoDebug, /// No debug info.
+ DebugDirectivesOnly, /// Line tables only.
+ FullDebug, /// Full debug info.
This enum doesn't appear to be complete? Either way can you make it match the
other and document what each thing means a bit more?
Comment at: lib/Driver/ToolChains/Cuda.cpp:289
static DebugInfoKind mustEmitDebugInfo(const ArgList ) {
+ const Arg *A = Args.getLastArg(options::OPT_O_Group);
Please document this routine in prose.
Comment at: lib/Driver/ToolChains/Cuda.cpp:292
+ bool IsDebugEnabled = !A || A->getOption().matches(options::OPT_O0) ||
+Args.hasFlag(options::OPT_cuda_noopt_device_debug,
+ options::OPT_no_cuda_noopt_device_debug,
ABataev wrote:
> echristo wrote:
> > Is this an nvcc compatibility flag?
> No, nvcc uses different set of flags. It uses `-g` for the debug info for the
> host code and `-G` for the device code. I'm not the original author of this
> option. clang uses it to control emission of the debug info for the device.
> The bad thing about nvcc that it disables optimizations when `-G` is used.
> Using this option we can use LLVM optimizations and disable the optimizations
> only when we call `ptxas` tool.
OK.
Comment at: lib/Driver/ToolChains/Cuda.cpp:706-708
+void CudaToolChain::adjustDebugInfoKind(
+codegenoptions::DebugInfoKind , const ArgList ) const {
+ switch (mustEmitDebugInfo(Args)) {
Is this really doing anything?
Repository:
rC Clang
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D51554/new/
https://reviews.llvm.org/D51554
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
ABataev added inline comments. Comment at: lib/Driver/ToolChains/Cuda.cpp:292 + bool IsDebugEnabled = !A || A->getOption().matches(options::OPT_O0) || +Args.hasFlag(options::OPT_cuda_noopt_device_debug, + options::OPT_no_cuda_noopt_device_debug, echristo wrote: > Is this an nvcc compatibility flag? No, nvcc uses different set of flags. It uses `-g` for the debug info for the host code and `-G` for the device code. I'm not the original author of this option. clang uses it to control emission of the debug info for the device. The bad thing about nvcc that it disables optimizations when `-G` is used. Using this option we can use LLVM optimizations and disable the optimizations only when we call `ptxas` tool. Repository: rC Clang https://reviews.llvm.org/D51554 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
echristo added a comment. The llvm backend patch here has discussion around debug info kinds that we should iron out first. Comment at: lib/Driver/ToolChains/Cuda.cpp:292 + bool IsDebugEnabled = !A || A->getOption().matches(options::OPT_O0) || +Args.hasFlag(options::OPT_cuda_noopt_device_debug, + options::OPT_no_cuda_noopt_device_debug, Is this an nvcc compatibility flag? Repository: rC Clang https://reviews.llvm.org/D51554 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
ABataev updated this revision to Diff 172003. ABataev added a comment. Updated processing of the debug options. -g1 (line-tables) is considered as the regular debug directive, which may emit some useful debug info. Repository: rC Clang https://reviews.llvm.org/D51554 Files: include/clang/Driver/ToolChain.h lib/Driver/ToolChains/Clang.cpp lib/Driver/ToolChains/Cuda.cpp lib/Driver/ToolChains/Cuda.h test/Driver/cuda-dwarf-2.cu test/Driver/openmp-offload-gpu.c Index: test/Driver/openmp-offload-gpu.c === --- test/Driver/openmp-offload-gpu.c +++ test/Driver/openmp-offload-gpu.c @@ -190,29 +190,35 @@ // CHK-BCLIB-WARN: No library 'libomptarget-nvptx-sm_20.bc' found in the default clang lib directory or in LIBRARY_PATH. Expect degraded performance due to no inlining of runtime functions on target devices. /// Check that debug info is emitted in dwarf-2 -// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --no-cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O1 --no-cuda-noopt-device-debug 2>&1 \ +// RUN: | FileCheck -check-prefix=DEBUG_DIRECTIVES %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O3 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: | FileCheck -check-prefix=DEBUG_DIRECTIVES %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O3 --no-cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: | FileCheck -check-prefix=DEBUG_DIRECTIVES %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g0 2>&1 \ // RUN: | FileCheck -check-prefix=NO_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb0 -O3 --cuda-noopt-device-debug 2>&1 \ // RUN: | FileCheck -check-prefix=NO_DEBUG %s -// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -gline-tables-only 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG -check-prefix=LINE_TABLE %s -// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb1 -O2 --cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG -check-prefix=LINE_TABLE %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -gline-directives-only 2>&1 \ +// RUN: | FileCheck -check-prefix=DEBUG_DIRECTIVES %s -// LINE_TABLE-NOT: warning: debug +// DEBUG_DIRECTIVES-NOT: warning: debug // NO_DEBUG-NOT: warning: debug +// NO_DEBUG: "-fopenmp-is-device" +// NO_DEBUG-NOT: "-debug-info-kind= // NO_DEBUG: ptxas -// LINE_TABLE: "-lineinfo" +// DEBUG_DIRECTIVES: "-triple" "nvptx64-nvidia-cuda" +// DEBUG_DIRECTIVES-SAME: "-debug-info-kind=line-directives-only" +// DEBUG_DIRECTIVES-SAME: "-fopenmp-is-device" +// DEBUG_DIRECTIVES: ptxas +// DEBUG_DIRECTIVES: "-lineinfo" // NO_DEBUG-NOT: "-g" // NO_DEBUG: nvlink // NO_DEBUG-NOT: "-g" +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --no-cuda-noopt-device-debug 2>&1 \ +// RUN: | FileCheck -check-prefix=HAS_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g 2>&1 \ // RUN: | FileCheck -check-prefix=HAS_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --cuda-noopt-device-debug 2>&1 \ @@ -227,9 +233,14 @@ // RUN: | FileCheck -check-prefix=HAS_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb3 -O2 --cuda-noopt-device-debug 2>&1 \ // RUN: | FileCheck -check-prefix=HAS_DEBUG %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -gline-tables-only 2>&1 \ +// RUN: | FileCheck -check-prefix=HAS_DEBUG %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb1 -O2 --cuda-noopt-device-debug 2>&1 \ +// RUN: | FileCheck -check-prefix=HAS_DEBUG %s // HAS_DEBUG-NOT: warning: debug // HAS_DEBUG: "-triple" "nvptx64-nvidia-cuda" +// HAS_DEBUG-SAME:
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
ABataev added a comment. Ping! Repository: rC Clang https://reviews.llvm.org/D51554 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
ABataev added a comment. In https://reviews.llvm.org/D51554#1224049, @echristo wrote: > The change in name here from "line tables" to "directives only" feels a bit > confusing. "Limited" seems to be a bit more clear, or even remaining line > tables only. Can you explain where you were going with this particular set of > changes in a bit more detail please? > > Thanks! > > -eric CUDA/NVPTX supports only 3 types of the debug info: limited/full, debug directives and no debug info at all. It does not support debug tables, so I just convert this into debug directives only. The main idea is to mimic what nvcc does. It behaves absolutely the same way. If the opt level is O0, we can use full debug info. if opt level is >O0, we can use only lineinfo(debug directives) or no debug info. If we enabling debug info for the device code using `--cuda-noopt-device-debug`, the opt level for the device code is lowered to O0 and we enable full debug info. The host code will be optimized still. Repository: rC Clang https://reviews.llvm.org/D51554 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
dblaikie added a comment. In https://reviews.llvm.org/D51554#1224049, @echristo wrote: > The change in name here from "line tables" to "directives only" feels a bit > confusing. "Limited" seems to be a bit more clear, or even remaining line > tables only. Can you explain where you were going with this particular set of > changes in a bit more detail please? Can't say I have much of an informed opinion about the parts that are only in the CUDA code. The "line directives only" terminology did come from a suggestion I made in one of the other reviews I can't seem to find right now.. ah, here: https://reviews.llvm.org/D51177 - whether or not that matches up with the use in the CUDA ToolChain code, I'm not sure. Repository: rC Clang https://reviews.llvm.org/D51554 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
echristo added a reviewer: dblaikie. echristo added a comment. The change in name here from "line tables" to "directives only" feels a bit confusing. "Limited" seems to be a bit more clear, or even remaining line tables only. Can you explain where you were going with this particular set of changes in a bit more detail please? Thanks! -eric Repository: rC Clang https://reviews.llvm.org/D51554 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
tra accepted this revision. tra added a comment. This revision is now accepted and ready to land. Nice. So, in effect, for optimized builds we'll generate pre-DWARF line info only, unless --cuda-noopt-device-debug is specified. Will this deal with the warnings about back-end being unable to handle particular debug options? On a side note, when DWARF is functional in NVPTX we need to seriously consider per-GPU control for it. Enabling debug info blows up cubin size (ptxas apparently packs compressed PTX inside *cubin*) and we run into ELF reloc overflows in some tensorflow builds if all GPU variants carry it. Repository: rC Clang https://reviews.llvm.org/D51554 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D51554: [CUDA][OPENMP][NVPTX]Improve logic of the debug info support.
ABataev created this revision. ABataev added reviewers: tra, echristo. Herald added subscribers: JDevlieghere, guansong, aprantl. Added support for the -gline-directives-only option + fixed logic of the debug info for CUDA devices. If optimization level is O0, then options --[no-]cuda-noopt-device-debug do not affect the debug info level. If the optimization level is >O0, debug info options are used + --no-cuda-noopt-device-debug is used or no --cuda-noopt-device-debug is used, the optimization level for the device code is kept and the emission of the debug directives is used. If the opt level is > O0, debug info is requested + --cuda-noopt-device-debug option is used, the optimization is disabled for the device code + required debug info is emitted. Repository: rC Clang https://reviews.llvm.org/D51554 Files: include/clang/Driver/ToolChain.h lib/Driver/ToolChains/Clang.cpp lib/Driver/ToolChains/Cuda.cpp lib/Driver/ToolChains/Cuda.h test/Driver/cuda-dwarf-2.cu test/Driver/openmp-offload-gpu.c Index: test/Driver/openmp-offload-gpu.c === --- test/Driver/openmp-offload-gpu.c +++ test/Driver/openmp-offload-gpu.c @@ -167,29 +167,39 @@ // CHK-BCLIB-WARN: No library 'libomptarget-nvptx-sm_20.bc' found in the default clang lib directory or in LIBRARY_PATH. Expect degraded performance due to no inlining of runtime functions on target devices. /// Check that debug info is emitted in dwarf-2 -// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --no-cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O1 --no-cuda-noopt-device-debug 2>&1 \ +// RUN: | FileCheck -check-prefix=LINE_TABLE %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O3 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: | FileCheck -check-prefix=LINE_TABLE %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O3 --no-cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG %s +// RUN: | FileCheck -check-prefix=LINE_TABLE %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g0 2>&1 \ // RUN: | FileCheck -check-prefix=NO_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb0 -O3 --cuda-noopt-device-debug 2>&1 \ // RUN: | FileCheck -check-prefix=NO_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -gline-tables-only 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG -check-prefix=LINE_TABLE %s +// RUN: | FileCheck -check-prefix=LINE_TABLE %s +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -gline-directives-only 2>&1 \ +// RUN: | FileCheck -check-prefix=LINE_TABLE %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -ggdb1 -O2 --cuda-noopt-device-debug 2>&1 \ -// RUN: | FileCheck -check-prefix=NO_DEBUG -check-prefix=LINE_TABLE %s +// RUN: | FileCheck -check-prefix=LINE_TABLE %s // LINE_TABLE-NOT: warning: debug // NO_DEBUG-NOT: warning: debug +// NO_DEBUG: "-fopenmp-is-device" +// NO_DEBUG-NOT: "-debug-info-kind= // NO_DEBUG: ptxas +// LINE_TABLE: "-triple" "nvptx64-nvidia-cuda" +// LINE_TABLE-SAME: "-debug-info-kind=line-directives-only" +// LINE_TABLE-SAME: "-fopenmp-is-device" +// LINE_TABLE: ptxas // LINE_TABLE: "-lineinfo" // NO_DEBUG-NOT: "-g" // NO_DEBUG: nvlink // NO_DEBUG-NOT: "-g" +// RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --no-cuda-noopt-device-debug 2>&1 \ +// RUN: | FileCheck -check-prefix=HAS_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g 2>&1 \ // RUN: | FileCheck -check-prefix=HAS_DEBUG %s // RUN: %clang -### -no-canonical-prefixes -fopenmp=libomp -fopenmp-targets=nvptx64-nvidia-cuda -Xopenmp-target -march=sm_60 %s -g -O0 --cuda-noopt-device-debug 2>&1 \ @@ -207,6 +217,7 @@ // HAS_DEBUG-NOT: warning: debug // HAS_DEBUG: "-triple" "nvptx64-nvidia-cuda" +// HAS_DEBUG-SAME: "-debug-info-kind=limited" // HAS_DEBUG-SAME: "-dwarf-version=2" // HAS_DEBUG-SAME: "-fopenmp-is-device" // HAS_DEBUG: ptxas Index: test/Driver/cuda-dwarf-2.cu
