Page MenuHomePhabricator

Farhana (Farhana Aleen)
User

Projects

User does not belong to any projects.

User Details

User Since
Jul 21 2015, 7:48 AM (208 w, 1 d)

Recent Activity

Aug 11 2017

Farhana added inline comments to D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Aug 11 2017, 11:47 AM
Farhana added inline comments to D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Aug 11 2017, 7:24 AM

Jul 31 2017

Farhana updated the summary of D36036: Supported interleaved byte load-pattern of stride:4 VF(8, 16, 32)..
Jul 31 2017, 12:32 PM
Farhana added inline comments to D35829: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF16 stride 4)..
Jul 31 2017, 11:18 AM
Farhana added inline comments to D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Jul 31 2017, 11:10 AM

Jul 28 2017

Farhana created D36036: Supported interleaved byte load-pattern of stride:4 VF(8, 16, 32)..
Jul 28 2017, 4:43 PM

Jul 20 2017

Farhana added inline comments to D35638: A fix for bug33826.
Jul 20 2017, 3:15 PM
Farhana updated the diff for D35638: A fix for bug33826.
Jul 20 2017, 3:15 PM
Farhana added inline comments to D35638: A fix for bug33826.
Jul 20 2017, 10:59 AM
Farhana updated the diff for D35638: A fix for bug33826.
Jul 20 2017, 10:55 AM
Farhana added inline comments to D35638: A fix for bug33826.
Jul 20 2017, 7:03 AM
Farhana updated the diff for D35638: A fix for bug33826.
Jul 20 2017, 7:03 AM
Farhana added inline comments to D35638: A fix for bug33826.
Jul 20 2017, 6:52 AM

Jul 19 2017

Farhana added inline comments to D35638: A fix for bug33826.
Jul 19 2017, 2:07 PM
Farhana updated the diff for D35638: A fix for bug33826.
Jul 19 2017, 2:06 PM
Farhana added a reviewer for D35638: A fix for bug33826: DavidKreitzer.
Jul 19 2017, 11:49 AM
Farhana created D35638: A fix for bug33826.
Jul 19 2017, 11:47 AM

Jul 10 2017

Farhana updated subscribers of D34478: [BasicAliasAnalysis] Allow idAddofNonZero() for values coming from the same loop iteration..
Jul 10 2017, 1:07 PM

Jul 5 2017

Farhana added inline comments to D34601: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess..
Jul 5 2017, 5:07 PM

Jun 30 2017

Farhana added a comment to D34478: [BasicAliasAnalysis] Allow idAddofNonZero() for values coming from the same loop iteration..

Hi Daniel,

Jun 30 2017, 2:52 PM

Jun 23 2017

Farhana added a comment to D34478: [BasicAliasAnalysis] Allow idAddofNonZero() for values coming from the same loop iteration..

"It makes the assumption that the values(GEP indices) are from the same iteration if the analysis has not visited any phi's yet."

Can you prove that this is actually true for all codepaths through here?

I agree that my comment about loop iteration is misleading.

In order to call isAddOfNonZero() on index1 and index2 and check whether index2 equals to index1 + x( and, vice versa), we need to prove that we are working on the same copy of index1. Which means the instance that used in index1 is the same as the one used in index2.
If we haven’t visited any phis yet, we can assume that part of the code is flat and we have the same copy, right?

This is precisely my question: Are you guaranteed that all codepaths end up here only in that situation?

Jun 23 2017, 8:26 AM

Jun 22 2017

Farhana added a comment to D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..

Hi Simon,

Jun 22 2017, 8:29 AM

Jun 21 2017

Farhana added a comment to D34478: [BasicAliasAnalysis] Allow idAddofNonZero() for values coming from the same loop iteration..

"It makes the assumption that the values(GEP indices) are from the same iteration if the analysis has not visited any phi's yet."

Can you prove that this is actually true for all codepaths through here?

Jun 21 2017, 4:19 PM
Farhana updated the diff for D34478: [BasicAliasAnalysis] Allow idAddofNonZero() for values coming from the same loop iteration..
Jun 21 2017, 3:04 PM
Farhana created D34478: [BasicAliasAnalysis] Allow idAddofNonZero() for values coming from the same loop iteration..
Jun 21 2017, 2:43 PM
Farhana updated the diff for D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..

This change-set reflects the fixed typos.

Jun 21 2017, 8:22 AM

Jun 19 2017

Farhana added a comment to D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..

ping.

Jun 19 2017, 8:40 AM

Jun 13 2017

Farhana updated the diff for D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..

Updated diff with the latest changes in the trunk.

Jun 13 2017, 12:00 PM

Jun 5 2017

Farhana added a comment to D33136: [ValueTracking] Don't let isAddOfNonZero look at adds with a PHI node as input.

What is the performance implication of this change-set? Have you done any performance testing? This seems to make the alias analysis very conservative. It is now going to find two non-equal numbers equal very easily.

Jun 5 2017, 9:03 PM

May 30 2017

Farhana added a comment to D33262: [JumpThreading] Teach jump threading how to analyze (and (cmp A, C1), (cmp A, C2)) after InstCombine has turned it into (cmp (add A, C3), C4) .

It seems to me that the improvement is done in the wrong place which is JumpThreading and made the change-set very specific. In my opinion, it does not have to be specific to JumpThreading. It can be a very general improvement of LVI using your new utilities.

May 30 2017, 6:38 PM
Farhana added a reviewer for D33262: [JumpThreading] Teach jump threading how to analyze (and (cmp A, C1), (cmp A, C2)) after InstCombine has turned it into (cmp (add A, C3), C4) : Farhana.
May 30 2017, 5:05 PM
Farhana updated subscribers of D33684: Added tests for lowerInterleavedStore..
May 30 2017, 12:53 PM
Farhana updated the diff for D33684: Added tests for lowerInterleavedStore..
May 30 2017, 12:24 PM
Farhana added a comment to D33684: Added tests for lowerInterleavedStore..

What happened to test/CodeGen/X86/x86-interleaved-access.ll ?

May 30 2017, 10:31 AM
Farhana updated the diff for D33684: Added tests for lowerInterleavedStore..

updated x86-interleaved-access.ll

May 30 2017, 10:20 AM
Farhana added a comment to D33684: Added tests for lowerInterleavedStore..

Hi Simon,

May 30 2017, 9:51 AM
Farhana added reviewers for D33684: Added tests for lowerInterleavedStore.: RKSimon, DavidKreitzer.
May 30 2017, 9:44 AM
Farhana created D33684: Added tests for lowerInterleavedStore..
May 30 2017, 9:43 AM

May 24 2017

Farhana added inline comments to D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..
May 24 2017, 3:48 PM
Farhana updated the diff for D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..

Updated the changes with the feedback from Dave.

May 24 2017, 3:48 PM

May 4 2017

Farhana added a comment to D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..

Ping.

May 4 2017, 11:33 AM

May 1 2017

Farhana added a comment to D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..

Performance numbers?

Also, please add context to the diff.

May 1 2017, 5:01 PM
Farhana updated the diff for D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..
May 1 2017, 4:48 PM

Apr 28 2017

Farhana created D32658: Supports lowerInterleavedStore() in X86InterleavedAccess..
Apr 28 2017, 1:17 PM

Nov 22 2016

Farhana updated the diff for D25986: Re-factorization of X86InterleaveAccess into a class.

ArrayRefs are passed by value.

Nov 22 2016, 11:53 AM

Nov 21 2016

Farhana added a comment to D25986: Re-factorization of X86InterleaveAccess into a class.

Hi Guys,

Nov 21 2016, 9:44 AM

Nov 11 2016

Farhana updated the diff for D25986: Re-factorization of X86InterleaveAccess into a class.
Nov 11 2016, 12:17 PM
Farhana added inline comments to D25986: Re-factorization of X86InterleaveAccess into a class.
Nov 11 2016, 7:15 AM

Nov 10 2016

Farhana added inline comments to D25986: Re-factorization of X86InterleaveAccess into a class.
Nov 10 2016, 4:11 PM
Farhana updated the diff for D25986: Re-factorization of X86InterleaveAccess into a class.

This change-set addresses Elena's comments.

Nov 10 2016, 4:11 PM

Nov 9 2016

Farhana updated the diff for D25986: Re-factorization of X86InterleaveAccess into a class.

Replaces uint32_t with unsigned.

Nov 9 2016, 11:04 AM
Farhana added a comment to D25986: Re-factorization of X86InterleaveAccess into a class.

Please can you replace the uses of uint32_t with unsigned?

Nov 9 2016, 10:53 AM

Oct 31 2016

Farhana added inline comments to D25986: Re-factorization of X86InterleaveAccess into a class.
Oct 31 2016, 8:11 PM
Farhana updated the diff for D25986: Re-factorization of X86InterleaveAccess into a class.

Thanks Dave, the update addresses your comment.

Oct 31 2016, 8:11 PM
Farhana updated the diff for D25986: Re-factorization of X86InterleaveAccess into a class.

The update addresses Zvi's comment about declaring Grp as a functional local variable.

Oct 31 2016, 11:26 AM

Oct 27 2016

Farhana updated the diff for D25986: Re-factorization of X86InterleaveAccess into a class.

Thanks Simon and Zvi.

Oct 27 2016, 3:13 PM

Oct 26 2016

Farhana retitled D25986: Re-factorization of X86InterleaveAccess into a class from to Re-factorization of X86InterleaveAccess into a class.
Oct 26 2016, 7:20 AM

Oct 13 2016

Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 13 2016, 7:32 AM

Oct 11 2016

Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..

Hi Michael,

Oct 11 2016, 11:27 AM

Oct 10 2016

Farhana added a comment to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..

I'm really sorry for coming into this review so late, but could you please also add stand-alone IR-level tests for the pass itself (and not only llc-level tests)?
Not only we'll get better testing granularity, but it'll be easier to see what the produced IR looks like, and whether it's really in the form we want.

Oct 10 2016, 9:07 PM

Oct 7 2016

Farhana added inline comments to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 7 2016, 5:10 PM
Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 7 2016, 5:10 PM
Farhana added inline comments to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 7 2016, 10:20 AM
Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 7 2016, 10:20 AM
Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 7 2016, 9:43 AM

Oct 6 2016

Farhana added inline comments to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 6 2016, 5:49 PM
Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 6 2016, 5:47 PM

Oct 5 2016

Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Oct 5 2016, 11:56 AM

Sep 30 2016

Farhana added inline comments to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Sep 30 2016, 2:03 PM
Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..

Includes changes that reflect Dave and Simon's comment.

Sep 30 2016, 1:46 PM
Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..

Uploaded the whole patch after running clang-format.

Sep 30 2016, 10:47 AM
Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..

This version is created after running clang-format.

Sep 30 2016, 10:22 AM

Sep 28 2016

Farhana added inline comments to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Sep 28 2016, 10:39 AM
Farhana added a comment to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Sep 28 2016, 10:36 AM
Farhana added inline comments to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Sep 28 2016, 10:35 AM
Farhana added a comment to D24681: Optimize patterns of vectorized interleaved memory accesses for X86..

Hi Zvi,

Sep 28 2016, 10:34 AM
Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Sep 28 2016, 10:22 AM

Sep 26 2016

Farhana updated the diff for D24681: Optimize patterns of vectorized interleaved memory accesses for X86..
Sep 26 2016, 12:26 PM

Sep 16 2016

Farhana retitled D24681: Optimize patterns of vectorized interleaved memory accesses for X86. from to Optimize patterns of vectorized interleaved memory accesses for X86..
Sep 16 2016, 1:01 PM