[llvm-bugs] [Bug 47206] New: Request merge of another AArch64 SVE CodeGen bug fix in narrowExtractedVectorLoad

[via llvm-bugs]

https://bugs.llvm.org/show_bug.cgi?id=47206

            Bug ID: 47206
           Summary: Request merge of another AArch64 SVE CodeGen bug fix
                    in narrowExtractedVectorLoad
           Product: libraries
           Version: 11.0
          Hardware: All
                OS: All
            Status: NEW
          Severity: release blocker
          Priority: P
         Component: Common Code Generator Code
          Assignee: unassignedbugs at nondot.org
          Reporter: david.sherwood at arm.com
                CC: llvm-bugs at lists.llvm.org

Hi, can I ask if it's possible to get the following patch cherry-picked onto
the LLVM 11 release branch?

3ec3fcb97a6b5a42d89032d44d81bbe711d188a4 David Sherwood [CodeGen] In
narrowExtractedVectorLoad bail out for scalable vectors

This bug fix is related to CodeGen for AArch64 SVE - any behavioural
differences can only be observed by compiling with +sve. Primarily this fix is
trying to ensure that optimisations in the DAG combiner calculate vector
offsets correctly for SVE. The bug was observed when compiling user written SVE
ACLE code. This patch depends upon the following 3 other patches:

455ed56d48e365f7d095254109abed876dc10c65 Kerry McLaughlin [SVE][CodeGen]
Legalisation of INSERT_VECTOR_ELT for scalable vectors
85c7e89f3b334a66df63d760d7951ee9fa2a5df5 Kerry McLaughlin [CodeGen] Refactor
getMemBasePlusOffset & getObjectPtrOffset to accept a TypeSize
88bbd30736561190a6733d0ad60aec21446b914c David Sherwood [SVE][CodeGen] Fix
issues with EXTRACT_SUBVECTOR when using scalable FP vectors

The commnand to cherry-pick these patches is:

git cherry-pick 88bbd30736561190a6733d0ad60aec21446b914c
85c7e89f3b334a66df63d760d7951ee9fa2a5df5
455ed56d48e365f7d095254109abed876dc10c65
3ec3fcb97a6b5a42d89032d44d81bbe711d188a4

However, I will understand if you feel it is too risky to cherry-pick these
patches due to the refactoring taking place in one of the dependent patches.

Thanks!
David.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20200817/d6383ce2/attachment.html>