[Driver][SYCL][FPGA] Improve expected triple and link behaviors for A… #1549
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
mdtoguchi
force-pushed
the
private/mdtoguchi/fpga-fsycl-link-lin-win
branch
from
eca5edb to
363e007
Compare
…OCX files When performing -fsycl-link=image behaviors on Windows, there was a mismatch of triples being used when checking for existence of the AOCX binary. This caused a problem when using clang-cl as the triples do not properly match. We also do not want to do any additional unbundling of any AOCX archive, which was causing a conflict of expected FPGA archives. Signed-off-by: Michael D Toguchi <michael.d.toguchi@intel.com>
mdtoguchi
force-pushed
the
private/mdtoguchi/fpga-fsycl-link-lin-win
branch
from
363e007 to
c23a03e
Compare
AGindinson
reviewed
Apr 20, 2020
|
Hi Mike,
What about -fsycl-link (without the =image part). Does this flow also get affected by the mismatched triple? |
Signed-off-by: Michael D Toguchi <michael.d.toguchi@intel.com>
Signed-off-by: Michael D Toguchi <michael.d.toguchi@intel.com>
…riple When unbunding for aocx, we were not checking for -coff Fix tests associated. Signed-off-by: Michael D Toguchi <michael.d.toguchi@intel.com>
Signed-off-by: Michael D Toguchi <michael.d.toguchi@intel.com>
@meiyacha, yes, aocr is also impacted here. |
AGindinson
reviewed
Apr 20, 2020
…t archive Signed-off-by: Michael D Toguchi <michael.d.toguchi@intel.com>
AGindinson
previously approved these changes
Apr 20, 2020
Co-Authored-By: Sergey Semenov <43845535+sergey-semenov@users.noreply.github.com>
AGindinson
approved these changes
Apr 21, 2020
sergey-semenov
approved these changes
Apr 21, 2020
dwoodwor-intel
pushed a commit
to dwoodwor-intel/llvm
that referenced
this pull request
Jul 22, 2022
This includes a rejigging of the BuiltinFuncMangleInfo to actually act as the source of information to pass this information into. Only some of the call sites of mangleBuiltin have been properly updated to pass in the pointer type element information. The various mutateCallInst* callers have yet to be updated, so those methods are temporarily relying on querying getPointerElementType() to generate this information. Original commit: KhronosGroup/SPIRV-LLVM-Translator@0eb9a7d
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
…OCX files
When performing -fsycl-link=image behaviors on Windows, there was a mismatch
of triples being used when checking for existence of the AOCX binary. This
caused a problem when using clang-cl as the triples do not properly match.
We also do not want to do any additional unbundling of any AOCX archive, which
was causing a conflict of expected FPGA archives.
Signed-off-by: Michael D Toguchi michael.d.toguchi@intel.com