User Details
- User Since
- Apr 12 2020, 10:42 PM (115 w, 3 d)
Aug 10 2021
@bondhugula Thanks for your review! I don't have commit access. So, could you land this patch?
Aug 9 2021
Aug 7 2021
@bondhugula Sorry again. Could you be a reviewer of this patch?
Aug 3 2021
@bondhugula I would like to support normalizing memrefs with MemRef_ReinterpretCastOp. As usual, this patch adds MemRefNormalizable trait. Could you review this patch?
Jul 8 2021
@MaskRay Yes. New GCC (>= 10.3) works.
Updated the comment again to reflect @MaskRay`s comment.
Updated the comment. We confirmed GCC version >= 10.3 works, not >= 11.
@MaskRay Thanks for comments.
Removed checking for MLIR build. Now the option is disabled for all configuration of SystemZ.
Jul 7 2021
Updated the condition to disable the option. It is disabled when GCC version is less than 10.3.
Condition to disable the option was updated since we found GCC version >= 11 is no problem.
Jul 6 2021
After D102453, MLIR installation test (cmake --build . --target check-mlir) issues 32 errors on SystemZ with GCC. This errors can be avoided by disabling fno-semantic-interposition. I used g++ (Ubuntu 9.4.0-1ubuntu1~18.04) 9.4.0.
Jul 5 2021
Jun 15 2021
@rriddle Thanks for quick review! I don't have commit access. So, could you land this patch?
Jun 14 2021
Fixed #ifdef.
May 20 2021
Rebased.
Could you land this patch?
May 14 2021
@bondhugula Much appreciate your reviews!
May 12 2021
@bondhugula Thanks for the review! I updated the code. Do you have any other comments?
Reflected comments.
May 11 2021
Hi, @rriddle , I got many testing errors after this patch on Systems Z (Big-endian machine)(Triple = "s390x-ibm-linux", CPU = "z14").
It seems I can avoid the errors by changing the bit order (TrailingOperandStorage() : numOperands(0), capacity(0), reserved(0) {}) , but I'm not sure why.
Do you have any suggestion?
Rebased and removed unnecessary include file.
May 10 2021
Used output parameters to reflect comments.
May 4 2021
Hi, @bondhugula , I updated this patch to reflect your suggestions. How do you think of this?
Rebased.
Apr 26 2021
Apr 25 2021
Hi, @ezhulenev
I'm testing MLIR on System z (Triple = "s390x-ibm-linux", CPU = "z14"), but I realized installation test (cmake --build . --target check-mlir) failed after this patch.
Endianness of System z is big-endian. Do you think this patch depends on the endianness? Do you have any suggestions to fix this?
Apr 22 2021
@bondhugula I updated the patch.
Fixed errors in clang-tidy.
Updated check routine for a tiled layout. It checks dimensions being floordiv'ed and then those same dimensions appearing with a mod and no other expressions involving those dimensions. Also, the routine provides those tile sizes back to the call site.
Apr 19 2021
Reflected comments for isTiledLayoutMap().
Apr 15 2021
Rebased.
Apr 14 2021
Updated to solve warning message in clang-tidy and rebased.
Apr 13 2021
Updated condition of tiled-layout map used in isTiledLayoutMap()
Apr 7 2021
@mehdi_amini Thanks for review. I don't have commit access. Could you commit this patch?
Apr 5 2021
@bondhugula Thanks for your review! I reflected your comments. Could you review again?
Fixed a comment.
Fixed code that accidentally deleted a previous fix, and updated comments.
Rebased and reflected comments.
Mar 29 2021
@bondhugula @avarmapml Sorry again, but I would like to use this patch in my use case. Could you merge this if you can approve?
Mar 22 2021
@bondhugula @avarmapml Thanks for your review! I reflected your comments. Do you have additional comments?
Mar 18 2021
Removed canonicalize.
Mar 17 2021
Reflected comments.
Mar 9 2021
@bondhugula @avarmapml Thanks for your review! I reflected your comments. Do you have any other comments on this?
Mar 4 2021
Rebased.
Mar 2 2021
@bondhugula @avarmapml Thanks for your reviews! I reflected your comments.
Rebased and Reflected comments.
Feb 28 2021
Clean up.
Nov 17 2020
@rriddle Thanks for reviewing many times! I don't have commit access. Could you commit this patch if you don't have any other comments?
Nov 15 2020
Hi, @rriddle, do you have any comments on this patch?
Nov 10 2020
Rebased and reflected comments.
Nov 1 2020
Oct 31 2020
Updated comments and inserted assert.
Oct 30 2020
Use ulittle for eidian conversion.
Oct 28 2020
Reflected the comment.
Oct 27 2020
@rriddle Thanks for your review. I don't have commit access. If you don't have other comments, could you commit this patch?
Avoid warning message.
Avoid warning in assert().
Rebased.
Reflect comments.
Oct 26 2020
@rriddle Thanks for your comments. I answered some of the comments.
Reflected comments and rebased.
Oct 20 2020
Removed unnecessary headers.
Moved the convEndianBE into DenseIntOrFPElementsAttr class and removed EndianUtilities.
@mehdi_amini Thanks for your review! I reflected all of your comments.
Reflected @mehdi_amini's comments.
Oct 19 2020
Implemented the conversion from LE to BE in AttributeParser and from BE to LE in AsmPrinter
, assuming that HEX in dense.attr is always LE.
Oct 5 2020
@bondhugula @avarmapml Thanks for your comments! I updated the code to reflect your comments. I moved part of code in the lambda to Util.cpp. Could you review again?
Rebased and fixed a comment.
Updated test case to include two operation results.
Reflected comments and rebased.
Oct 2 2020
@bondhugula @avarmapml I found the results(returend memref) of normalizable operation were not normalized by --normalize-memrefs.
In test case, %1 was not normalized. This patch fixes it.
%1 = "test.op_norm_ret"(%arg0, %arg1) : (memref<1x32768xf32>, memref<1x16x14x14xf32, #map1>) -> (memref<1x16x14x14xf32, #map1>)