Page MenuHomePhabricator

Please use GitHub pull requests for new patches. Phabricator shutdown timeline

m_zuckerman (michael zuckerman)
User

Projects

User does not belong to any projects.

User Details

User Since
Jun 17 2015, 1:09 AM (432 w, 5 d)

Recent Activity

Jun 2 2020

m_zuckerman added a reviewer for D57504: RFC: Prototype & Roadmap for vector predication in LLVM: m_zuckerman.
Jun 2 2020, 6:36 AM · Restricted Project, Restricted Project

Dec 18 2017

m_zuckerman updated the diff for D40465: [Clang][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Dec 18 2017, 12:03 AM

Dec 17 2017

m_zuckerman updated the diff for D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.

SORRY

Dec 17 2017, 11:50 PM
m_zuckerman updated the diff for D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Dec 17 2017, 7:31 AM
m_zuckerman updated the diff for D40465: [Clang][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Dec 17 2017, 6:39 AM
m_zuckerman added inline comments to D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Dec 17 2017, 6:27 AM
m_zuckerman updated the diff for D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Dec 17 2017, 6:25 AM
m_zuckerman updated the diff for D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Dec 17 2017, 6:24 AM
m_zuckerman added inline comments to D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Dec 17 2017, 6:23 AM
m_zuckerman updated the diff for D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Dec 17 2017, 5:56 AM

Dec 14 2017

m_zuckerman accepted D40287: [X86][AVX][AVX2]: Adding full coverage of MC encoding for the AVX, AVX2 isa set.<NFC>.
Dec 14 2017, 4:35 AM
m_zuckerman added a comment to D40287: [X86][AVX][AVX2]: Adding full coverage of MC encoding for the AVX, AVX2 isa set.<NFC>.

LGTM, if no one has something to add

Dec 14 2017, 4:35 AM
m_zuckerman abandoned D39569: [AVX512] Adding support for extract element(mask i8) from a vector .
Dec 14 2017, 4:25 AM

Dec 13 2017

m_zuckerman updated the diff for D39547: [AVX512] Adding support for load truncate store of I1.
Dec 13 2017, 3:19 AM
m_zuckerman updated the diff for D39547: [AVX512] Adding support for load truncate store of I1.
Dec 13 2017, 1:35 AM

Dec 12 2017

m_zuckerman accepted D34141: [X86] Recognize constant arrays with special values and replace loads from it with subtract and shift instructions, which then will be replaced by X86 BZHI machine instruction..
Dec 12 2017, 3:37 AM
m_zuckerman added a comment to D34141: [X86] Recognize constant arrays with special values and replace loads from it with subtract and shift instructions, which then will be replaced by X86 BZHI machine instruction..

LGTM with no one has nothing to add !!!

Dec 12 2017, 3:37 AM

Nov 26 2017

m_zuckerman updated the summary of D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Nov 26 2017, 3:41 AM
m_zuckerman created D40465: [Clang][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Nov 26 2017, 3:39 AM
m_zuckerman created D40462: [LLVM][X86] Converting X86 intrinisics movmsk{ps|pd}{|256} and pmovmskb{128|256}.
Nov 26 2017, 12:39 AM

Nov 2 2017

m_zuckerman updated the diff for D39547: [AVX512] Adding support for load truncate store of I1.
Nov 2 2017, 2:26 PM
m_zuckerman created D39569: [AVX512] Adding support for extract element(mask i8) from a vector .
Nov 2 2017, 2:22 PM
m_zuckerman updated the diff for D39547: [AVX512] Adding support for load truncate store of I1.
Nov 2 2017, 8:56 AM
m_zuckerman updated the diff for D39547: [AVX512] Adding support for load truncate store of I1.
Nov 2 2017, 7:26 AM
m_zuckerman created D39547: [AVX512] Adding support for load truncate store of I1.
Nov 2 2017, 6:24 AM

Oct 30 2017

m_zuckerman accepted D38312: adding pattern for broadcastm.

LGTM

Oct 30 2017, 2:21 AM
m_zuckerman added inline comments to D38312: adding pattern for broadcastm.
Oct 30 2017, 1:42 AM
m_zuckerman added inline comments to D38312: adding pattern for broadcastm.
Oct 30 2017, 1:16 AM

Oct 29 2017

m_zuckerman added inline comments to D34141: [X86] Recognize constant arrays with special values and replace loads from it with subtract and shift instructions, which then will be replaced by X86 BZHI machine instruction..
Oct 29 2017, 10:13 AM
m_zuckerman added a reviewer for D34141: [X86] Recognize constant arrays with special values and replace loads from it with subtract and shift instructions, which then will be replaced by X86 BZHI machine instruction.: m_zuckerman.
Oct 29 2017, 6:45 AM
m_zuckerman added a comment to D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 .

We shouldn’t rely on pattern order in the td file if we can avoid it. So a pattern that is for when DQI is disabled should have a no DQI predicate.

But in this case I see no good reason to have two different patterns. Why can’t we use KSHIFTW even when DQI is enabled? Is there a downside?

Oct 29 2017, 6:18 AM
m_zuckerman updated the diff for D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 .
Oct 29 2017, 6:17 AM

Oct 27 2017

m_zuckerman added inline comments to D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 .
Oct 27 2017, 8:16 AM
m_zuckerman updated the diff for D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 .
Oct 27 2017, 8:10 AM
m_zuckerman updated the diff for D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 .
Oct 27 2017, 7:59 AM

Oct 26 2017

m_zuckerman updated the diff for D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 .
Oct 26 2017, 8:03 AM

Oct 25 2017

m_zuckerman updated the summary of D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 .
Oct 25 2017, 8:57 AM
m_zuckerman added reviewers for D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 : guyblank, igorb, zvi, ayman, craig.topper.
Oct 25 2017, 8:56 AM
m_zuckerman created D39292: [X86][AVX512] Adding new patterns for extract_subvector of vXi1 .
Oct 25 2017, 8:54 AM

Oct 24 2017

m_zuckerman added inline comments to D38312: adding pattern for broadcastm.
Oct 24 2017, 9:57 AM

Oct 17 2017

m_zuckerman updated the diff for D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 17 2017, 3:09 AM

Oct 16 2017

m_zuckerman added inline comments to D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 16 2017, 8:53 AM
m_zuckerman updated the diff for D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 16 2017, 8:53 AM
m_zuckerman updated the diff for D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 16 2017, 4:14 AM

Oct 15 2017

m_zuckerman updated the diff for D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 15 2017, 1:58 PM

Oct 14 2017

m_zuckerman updated the diff for D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 14 2017, 2:18 PM
m_zuckerman added inline comments to D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 14 2017, 2:11 PM
m_zuckerman updated the diff for D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 14 2017, 2:10 PM

Oct 10 2017

m_zuckerman created D38762: Cost calculation for interleave load/store patterns {v8i8,v16i8,v32i8,v64i8}.
Oct 10 2017, 1:56 PM
m_zuckerman added inline comments to D38037: [InstCombine] Compacting or instructions whose operands are shift instructions.
Oct 10 2017, 1:25 PM

Oct 9 2017

m_zuckerman added a comment to D38689: [X86] Fix a bug with i386 subtarget in LowerCONCAT_VECTORSvXi1 func.

Test case?

Oct 9 2017, 11:33 PM
m_zuckerman accepted D38689: [X86] Fix a bug with i386 subtarget in LowerCONCAT_VECTORSvXi1 func.

LGTM

Oct 9 2017, 12:10 PM
m_zuckerman accepted D38672: [X86][AVX512] lowering shuffle f/i intrinsic - clang part.

after LLVM-SIDE

Oct 9 2017, 1:14 AM
m_zuckerman added a comment to D38672: [X86][AVX512] lowering shuffle f/i intrinsic - clang part.

Please add dependent parent.
LGTM After the parent commit (LLVM side).

Oct 9 2017, 1:12 AM

Oct 7 2017

m_zuckerman abandoned D31396: [X86][LLVM][Canonical Compare Intrinsics] Creating a canonical representation for X86 CMP intrinsics.
Oct 7 2017, 11:41 PM

Sep 29 2017

m_zuckerman added inline comments to D37687: [X86][LLVM]Expanding Supports lowerInterleaved{store|load}() in X86InterleavedAccess (VF64 stride 3) .
Sep 29 2017, 6:59 AM
m_zuckerman updated the diff for D37687: [X86][LLVM]Expanding Supports lowerInterleaved{store|load}() in X86InterleavedAccess (VF64 stride 3) .
Sep 29 2017, 6:57 AM

Sep 28 2017

m_zuckerman updated the diff for D37687: [X86][LLVM]Expanding Supports lowerInterleaved{store|load}() in X86InterleavedAccess (VF64 stride 3) .
Sep 28 2017, 6:56 AM

Sep 26 2017

m_zuckerman closed D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 26 2017, 3:30 PM

Sep 25 2017

m_zuckerman accepted D38220: [X86] Finishing broadcast[f|i]32x2 intrinsics lowering to IR..

LGTM THANKS ,

Sep 25 2017, 5:03 AM
m_zuckerman updated the diff for D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 25 2017, 5:03 AM
m_zuckerman added inline comments to D38220: [X86] Finishing broadcast[f|i]32x2 intrinsics lowering to IR..
Sep 25 2017, 5:03 AM
m_zuckerman added inline comments to D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 25 2017, 5:03 AM
m_zuckerman updated the diff for D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 25 2017, 5:03 AM
m_zuckerman updated the diff for D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Sep 25 2017, 5:03 AM
m_zuckerman updated the diff for D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 25 2017, 5:02 AM

Sep 21 2017

m_zuckerman updated the diff for D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 21 2017, 3:09 AM
m_zuckerman updated the diff for D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Sep 21 2017, 2:17 AM

Sep 14 2017

m_zuckerman updated the diff for D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Sep 14 2017, 1:58 AM

Sep 11 2017

m_zuckerman updated the diff for D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Sep 11 2017, 2:55 PM
m_zuckerman retitled D37687: [X86][LLVM]Expanding Supports lowerInterleaved{store|load}() in X86InterleavedAccess (VF64 stride 3) from [X86][LLVM]Expanding Supports lowerInterleaved{store|load}}() in X86InterleavedAccess (VF64 stride 3) to [X86][LLVM]Expanding Supports lowerInterleaved{store|load}() in X86InterleavedAccess (VF64 stride 3) .
Sep 11 2017, 7:13 AM
m_zuckerman retitled D37687: [X86][LLVM]Expanding Supports lowerInterleaved{store|load}() in X86InterleavedAccess (VF64 stride 3) from [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF64 stride 3) to [X86][LLVM]Expanding Supports lowerInterleaved{store|load}}() in X86InterleavedAccess (VF64 stride 3) .
Sep 11 2017, 7:04 AM
m_zuckerman created D37687: [X86][LLVM]Expanding Supports lowerInterleaved{store|load}() in X86InterleavedAccess (VF64 stride 3) .
Sep 11 2017, 7:02 AM

Sep 10 2017

m_zuckerman updated the diff for D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 10 2017, 6:56 AM
m_zuckerman updated the diff for D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 10 2017, 4:21 AM
m_zuckerman updated the diff for D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Sep 10 2017, 4:19 AM

Sep 7 2017

m_zuckerman closed D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..

commit to https://reviews.llvm.org/rL312722

Sep 7 2017, 7:35 AM

Sep 6 2017

m_zuckerman updated the diff for D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Sep 6 2017, 3:25 AM

Sep 5 2017

m_zuckerman updated the diff for D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Sep 5 2017, 4:24 AM

Sep 4 2017

m_zuckerman updated the diff for D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..

Update according to Zvi comments

Sep 4 2017, 2:05 PM

Sep 3 2017

m_zuckerman updated the diff for D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Sep 3 2017, 1:17 AM

Aug 30 2017

m_zuckerman added inline comments to D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Aug 30 2017, 5:12 AM
m_zuckerman updated the diff for D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Aug 30 2017, 4:00 AM

Aug 29 2017

m_zuckerman updated the diff for D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Aug 29 2017, 6:31 AM
m_zuckerman added inline comments to D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Aug 29 2017, 2:16 AM
m_zuckerman updated the diff for D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Aug 29 2017, 2:15 AM

Aug 28 2017

m_zuckerman added a comment to D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..

Ping

Aug 28 2017, 6:33 AM
m_zuckerman updated the diff for D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Aug 28 2017, 6:33 AM
m_zuckerman added inline comments to D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Aug 28 2017, 6:29 AM

Aug 24 2017

m_zuckerman updated the diff for D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Aug 24 2017, 12:52 PM
m_zuckerman created D37117: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF{8|16|32} stride 3)...
Aug 24 2017, 12:45 PM

Aug 22 2017

m_zuckerman added inline comments to D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Aug 22 2017, 12:58 AM
m_zuckerman updated the diff for D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Aug 22 2017, 12:56 AM

Aug 21 2017

m_zuckerman created D36960: [X86][LLVM]Expanding Supports lowerInterleavedLoad() in X86InterleavedAccess (VF{8|16|32} stride 3)..
Aug 21 2017, 7:34 AM
m_zuckerman abandoned D20055: [Clang][Intrinsics] Adding avx512f missing intrinsics {cast} .
Aug 21 2017, 4:45 AM

Aug 10 2017

m_zuckerman accepted D36570: [X86][AsmParser][AVX512] Error appropriately when K0 is tried as a write-mask.

It sounds reasonable
LGTM

Aug 10 2017, 4:45 AM
m_zuckerman added inline comments to D36570: [X86][AsmParser][AVX512] Error appropriately when K0 is tried as a write-mask.
Aug 10 2017, 3:18 AM

Aug 8 2017

m_zuckerman accepted D35785: [AsmParser][AVX512]Enhance OpMask/Zero/Merge syntax check rubostness.
Aug 8 2017, 5:44 AM

Aug 7 2017

m_zuckerman updated the diff for D36058: [X86][LLVM]Expanding Supports lowerInterleavedStore() in X86InterleavedAccess (VF8 stride 4)..
Aug 7 2017, 1:34 PM

Aug 1 2017

m_zuckerman abandoned D11215: Adding altmacro support in integrated assembler. continue of D10591 .
Aug 1 2017, 11:01 PM