Skip to content
This repository has been archived by the owner on Jan 26, 2024. It is now read-only.

[rocm] Fix segfault with -O0-compiled kernels #45

Closed
wants to merge 1 commit into from
Closed

[rocm] Fix segfault with -O0-compiled kernels #45

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions device/devhcprintf.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,6 +26,7 @@
#include <cstdio>
#include <cstring>
#include <string>
#include <cstdint>

static void checkPrintf(FILE* stream, int* outCount, const char* fmt, ...) {
va_list args;
Expand Down
6 changes: 3 additions & 3 deletions device/rocm/rocblit.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -290,13 +290,13 @@ class KernelBlitManager : public DmaBlitManager {
StreamOpsWait,
Scheduler,
GwsInit,
InitHeap,
BlitLinearTotal,
FillImage = BlitLinearTotal,
BlitCopyImage,
BlitCopyImage1DA,
BlitCopyImageToBuffer,
BlitCopyBufferToImage,
InitHeap,
BlitTotal
};

Expand Down Expand Up @@ -595,9 +595,9 @@ static const char* BlitName[KernelBlitManager::BlitTotal] = {
"__amd_rocclr_fillBufferAligned", "__amd_rocclr_fillBufferAligned2D", "__amd_rocclr_copyBuffer",
"__amd_rocclr_copyBufferAligned", "__amd_rocclr_copyBufferRect",
"__amd_rocclr_copyBufferRectAligned", "__amd_rocclr_streamOpsWrite", "__amd_rocclr_streamOpsWait",
"__amd_rocclr_scheduler", "__amd_rocclr_gwsInit", "__amd_rocclr_fillImage",
"__amd_rocclr_scheduler", "__amd_rocclr_gwsInit", "__amd_rocclr_initHeap", "__amd_rocclr_fillImage",
"__amd_rocclr_copyImage", "__amd_rocclr_copyImage1DA", "__amd_rocclr_copyImageToBuffer",
"__amd_rocclr_copyBufferToImage", "__amd_rocclr_initHeap"
"__amd_rocclr_copyBufferToImage"
};

inline void KernelBlitManager::setArgument(amd::Kernel* kernel, size_t index,
Expand Down