[DeviceSanitizer] Support check invalid kernel argument #14512
Conversation
yingcong-wu
left a comment
yingcong-wu
left a comment
There was a problem hiding this comment.
out-of-bounds and released-pointer checks seem a bit redundant. I mean those could also be caught when those errors happened within kernel execution, right?
Yeah, that's a problem. I'll reconsider this. |
Yes, this a bit redundant. But it costs very little to check these two errors while checking other errors. |
AllanZyne
had a problem deploying
to
WindowsCILock
GitHub Actions
Failure
|
Hi @intel/unified-runtime-reviewers, please review, thanks |
| int main() { | ||
| sycl::queue Q; | ||
| constexpr std::size_t N = 123456; | ||
| constexpr std::size_t N = 512; |
There was a problem hiding this comment.
NOTE: We'll add a new test in “private” for large workgroup case
|
@intel/dpcpp-sanitizers-review could you have another round of review on this as I merged the UR part of it, so we would prioritize merging that. Thanks! |
omarahmed1111
temporarily deployed
to
WindowsCILock
GitHub Actions
Inactive
omarahmed1111
temporarily deployed
to
WindowsCILock
GitHub Actions
Inactive
Sure. |
|
@intel/llvm-gatekeepers please merge, thanks. |
UR: oneapi-src/unified-runtime#1843 also fix parallel_for_double.cpp --------- Co-authored-by: omarahmed1111 <omar.ahmed@codeplay.com>
UR: oneapi-src/unified-runtime#1843
also fix parallel_for_double.cpp