Page MenuHomePhabricator

tpr (Tim Renouf)
User

Projects

User does not belong to any projects.

User Details

User Since
Jan 26 2017, 7:07 AM (191 w, 3 d)

Recent Activity

Jul 8 2020

tpr added a reviewer for D83372: Fix for memory leak reported by Valgrind: beanz.

+Chris Bieneman

Jul 8 2020, 10:03 AM · Restricted Project

May 27 2020

tpr added a reviewer for D80194: [AMDGPU] Allow PAL to put RO global in .text or .rodata: rampitec.
May 27 2020, 2:07 AM · Restricted Project
tpr added a reviewer for D80194: [AMDGPU] Allow PAL to put RO global in .text or .rodata: s-perron.
May 27 2020, 2:07 AM · Restricted Project
tpr abandoned D80338: [AMDGPU] GFX10 PAL: Only pad with s_code_end in pipeline compilation.

I am abandoning this, because it turns out that it is difficult to reliably detect whether we have a PAL pipeline or a PAL shader compilation. Instead, the LLPC linker will remove the padding, and then adds it back on after concatenating .text sections.

May 27 2020, 2:07 AM · Restricted Project

May 26 2020

tpr added a comment to D80194: [AMDGPU] Allow PAL to put RO global in .text or .rodata.

Ping.

May 26 2020, 3:44 AM · Restricted Project

May 25 2020

tpr updated the diff for D80338: [AMDGPU] GFX10 PAL: Only pad with s_code_end in pipeline compilation.

V2: Review comments: Separate predicate function; early exit on CS.

May 25 2020, 3:10 AM · Restricted Project

May 21 2020

tpr committed rGd13a50882006: [AMDGPU] Fixed incorrect PAL metadata register naming (authored by tpr).
[AMDGPU] Fixed incorrect PAL metadata register naming
May 21 2020, 2:39 PM
tpr committed rGdb16eb33ce43: [MsgPack] Added convenience assignment to MsgPackDocument (authored by tpr).
[MsgPack] Added convenience assignment to MsgPackDocument
May 21 2020, 2:39 PM
tpr closed D80121: [MsgPack] Added convenience assignment to MsgPackDocument.
May 21 2020, 2:39 PM · Restricted Project
tpr closed D78860: [AMDGPU] Fixed incorrect PAL metadata register naming.
May 21 2020, 2:39 PM · Restricted Project
tpr committed rGe79d002309cb: [MsgPack] MsgPackDocument::readFromBlob now merges (authored by tpr).
[MsgPack] MsgPackDocument::readFromBlob now merges
May 21 2020, 1:33 PM
tpr closed D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.
May 21 2020, 1:33 PM · Restricted Project
tpr added inline comments to D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.
May 21 2020, 12:59 PM · Restricted Project
tpr updated the diff for D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.

V5: API change so client can choose whether array merging does merge or append.

May 21 2020, 5:54 AM · Restricted Project
tpr added inline comments to D80338: [AMDGPU] GFX10 PAL: Only pad with s_code_end in pipeline compilation.
May 21 2020, 1:04 AM · Restricted Project

May 20 2020

tpr created D80338: [AMDGPU] GFX10 PAL: Only pad with s_code_end in pipeline compilation.
May 20 2020, 4:33 PM · Restricted Project
tpr added reviewers for D80338: [AMDGPU] GFX10 PAL: Only pad with s_code_end in pipeline compilation: nhaehnle, rampitec.
May 20 2020, 4:33 PM · Restricted Project
tpr updated the diff for D80121: [MsgPack] Added convenience assignment to MsgPackDocument.

V4: Added MapDocNode::erase methods.

May 20 2020, 3:27 PM · Restricted Project
tpr updated the diff for D80194: [AMDGPU] Allow PAL to put RO global in .text or .rodata.

V2: Fixed tests.

May 20 2020, 2:53 PM · Restricted Project

May 19 2020

tpr created D80194: [AMDGPU] Allow PAL to put RO global in .text or .rodata.
May 19 2020, 1:36 AM · Restricted Project
tpr added a reviewer for D80194: [AMDGPU] Allow PAL to put RO global in .text or .rodata: kzhuravl.
May 19 2020, 1:36 AM · Restricted Project
tpr added a reviewer for D80194: [AMDGPU] Allow PAL to put RO global in .text or .rodata: nhaehnle.
May 19 2020, 1:36 AM · Restricted Project

May 18 2020

tpr updated the diff for D80121: [MsgPack] Added convenience assignment to MsgPackDocument.

V3: Added assignment for const char *, otherwise it defaults to using

the bool one rather than the StringRef one.
May 18 2020, 9:39 AM · Restricted Project
tpr added reviewers for D80121: [MsgPack] Added convenience assignment to MsgPackDocument: scott.linder, dineshkb-amd.
May 18 2020, 6:57 AM · Restricted Project
tpr updated the diff for D80121: [MsgPack] Added convenience assignment to MsgPackDocument.

V2: Added tests

May 18 2020, 6:57 AM · Restricted Project
tpr created D80121: [MsgPack] Added convenience assignment to MsgPackDocument.
May 18 2020, 6:57 AM · Restricted Project
tpr added a comment to D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.

With the nil/empty change, there is a change in behavior in that doing a map lookup "map[key]" with a key not previously in the map creates a new map entry that is empty rather than nil. That is more correct, but I have just discovered some code in LLPC relying on the old behavior. I can fix the LLPC code, but is the change in behavior going to cause any HSA problems?

May 18 2020, 4:15 AM · Restricted Project
tpr added inline comments to D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.
May 18 2020, 3:42 AM · Restricted Project
tpr updated the diff for D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.
V4: Addressed review comments:
    * Separated the idea of an empty node from a nil node
    * Made default merge behavior to fail on any conflict
    * Array merging now appends
May 18 2020, 3:42 AM · Restricted Project
tpr added a comment to D78860: [AMDGPU] Fixed incorrect PAL metadata register naming.

Ping

May 18 2020, 1:01 AM · Restricted Project

May 13 2020

tpr added inline comments to D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.
May 13 2020, 1:02 AM · Restricted Project

May 12 2020

tpr added inline comments to D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.
May 12 2020, 11:49 AM · Restricted Project
tpr added a comment to D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.

Dinesh is working on merging metadata in ROCm CompilerSupport using MsgPackDocument, and it fell out that merging YAML metadata was also handled transparently. What is the advantage to doing this via a function pointer passed to readFromBlob? It doesn't seem to save any copies as you still need to be able to compare the nodes; is it just a matter of readability? Would it be reasonable to also add a callback for fromYAML?

May 12 2020, 11:49 AM · Restricted Project

May 11 2020

tpr updated the diff for D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.

V3: Removed unneeded nullptr arg.

May 11 2020, 6:56 AM · Restricted Project

May 10 2020

tpr added a comment to D78860: [AMDGPU] Fixed incorrect PAL metadata register naming.

Ping.

May 10 2020, 2:39 AM · Restricted Project

May 9 2020

tpr updated the diff for D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.

V2: Pass map key to merger function.

May 9 2020, 10:05 AM · Restricted Project
tpr created D79671: [MsgPack] MsgPackDocument::readFromBlob now merges.
May 9 2020, 7:25 AM · Restricted Project
tpr added reviewers for D79671: [MsgPack] MsgPackDocument::readFromBlob now merges: scott.linder, kzhuravl.
May 9 2020, 7:25 AM · Restricted Project

May 1 2020

tpr updated the diff for D78860: [AMDGPU] Fixed incorrect PAL metadata register naming.

V2: Also fixed GS regs. Added test.

May 1 2020, 2:57 AM · Restricted Project

Apr 25 2020

tpr created D78860: [AMDGPU] Fixed incorrect PAL metadata register naming.
Apr 25 2020, 4:44 AM · Restricted Project
tpr added reviewers for D78860: [AMDGPU] Fixed incorrect PAL metadata register naming: dstuttard, rampitec.
Apr 25 2020, 4:44 AM · Restricted Project

Mar 10 2020

tpr added a comment to D75789: [AMDGPU] Allow struct.buffer.*.format intrinsics to accept i32.

I don't know of any reason why it has to be fp. Maybe related to the prehistoric thing where isel used to need something to be fp to put it in a vgpr?

Mar 10 2020, 3:02 AM · Restricted Project

Feb 17 2020

tpr added a comment to D74594: [AMDGPU] Fix some tests that did not specify -mcpu.

This got landed in a slightly different form because of conflicts with other sdiv64 etc tests.

Feb 17 2020, 6:16 AM · Restricted Project
tpr committed rG1e926a9f9c51: [AMDGPU] Fix some tests that did not specify -mcpu (authored by tpr).
[AMDGPU] Fix some tests that did not specify -mcpu
Feb 17 2020, 6:08 AM
tpr closed D74594: [AMDGPU] Fix some tests that did not specify -mcpu.
Feb 17 2020, 6:07 AM · Restricted Project

Feb 14 2020

tpr updated the diff for D74594: [AMDGPU] Fix some tests that did not specify -mcpu.

V2: Also added gfx1010 to memtime test.

Feb 14 2020, 2:30 AM · Restricted Project
tpr added reviewers for D74594: [AMDGPU] Fix some tests that did not specify -mcpu: kzhuravl, rampitec.
Feb 14 2020, 1:47 AM · Restricted Project
tpr created D74594: [AMDGPU] Fix some tests that did not specify -mcpu.
Feb 14 2020, 1:45 AM · Restricted Project

Dec 13 2019

tpr committed rGfce1a6f5848d: Revert "AMDGPU: Try to commute sub of boolean ext" (authored by tpr).
Revert "AMDGPU: Try to commute sub of boolean ext"
Dec 13 2019, 4:54 AM
tpr added a reverting change for rG69fcfb7d3597: AMDGPU: Try to commute sub of boolean ext: rGfce1a6f5848d: Revert "AMDGPU: Try to commute sub of boolean ext".
Dec 13 2019, 4:54 AM
tpr closed D70978: Revert "AMDGPU: Try to commute sub of boolean ext".
Dec 13 2019, 4:54 AM · Restricted Project
tpr updated the diff for D70978: Revert "AMDGPU: Try to commute sub of boolean ext".
V3: Reinstated and fixed the removed test.
Dec 13 2019, 2:35 AM · Restricted Project

Dec 12 2019

tpr added a comment to D70978: Revert "AMDGPU: Try to commute sub of boolean ext".

Yes, sorry, I will do the test changes that Matt suggested so this can be approved and landed.

Dec 12 2019, 2:16 AM · Restricted Project

Dec 4 2019

tpr updated the diff for D70978: Revert "AMDGPU: Try to commute sub of boolean ext".

V2: Removed an existing test for the bogus fold.

Dec 4 2019, 12:44 AM · Restricted Project

Dec 3 2019

tpr added reviewers for D70978: Revert "AMDGPU: Try to commute sub of boolean ext": arsenm, rampitec.
Dec 3 2019, 1:11 PM · Restricted Project
tpr added a reverting change for rG69fcfb7d3597: AMDGPU: Try to commute sub of boolean ext: D70978: Revert "AMDGPU: Try to commute sub of boolean ext".
Dec 3 2019, 1:02 PM
tpr created D70978: Revert "AMDGPU: Try to commute sub of boolean ext".
Dec 3 2019, 1:02 PM · Restricted Project

Dec 2 2019

tpr committed rG3d5ba7c60f39: AMDGPU: Fixed indeterminate map iteration in SIPeepholeSDWA (authored by tpr).
AMDGPU: Fixed indeterminate map iteration in SIPeepholeSDWA
Dec 2 2019, 4:12 AM
tpr closed D70783: AMDGPU: Fixed indeterminate map iteration in SIPeepholeSDWA.
Dec 2 2019, 4:12 AM · Restricted Project

Nov 27 2019

tpr updated the diff for D70783: AMDGPU: Fixed indeterminate map iteration in SIPeepholeSDWA.

V2: Sorted includes alphabetically.

Nov 27 2019, 11:51 AM · Restricted Project
tpr added a comment to D70783: AMDGPU: Fixed indeterminate map iteration in SIPeepholeSDWA.

Did the problem manifest in any way?

Nov 27 2019, 11:23 AM · Restricted Project
tpr added reviewers for D70783: AMDGPU: Fixed indeterminate map iteration in SIPeepholeSDWA: SamWot, rampitec, nhaehnle.

No test because by definition it was indeterminate.

Nov 27 2019, 7:31 AM · Restricted Project
tpr created D70783: AMDGPU: Fixed indeterminate map iteration in SIPeepholeSDWA.
Nov 27 2019, 7:22 AM · Restricted Project

Nov 12 2019

tpr committed rG07ebd741546e: MCP: Fixed bug with dest overlapping copy source (authored by tpr).
MCP: Fixed bug with dest overlapping copy source
Nov 12 2019, 12:22 AM
tpr closed D69953: MCP: Fixed bug with dest overlapping copy source.
Nov 12 2019, 12:22 AM · Restricted Project

Nov 11 2019

tpr updated the diff for D69953: MCP: Fixed bug with dest overlapping copy source.

V3: Only bail if it is a copy and a partial def, to avoid spurious test changes.

Nov 11 2019, 6:41 AM · Restricted Project

Nov 8 2019

tpr added inline comments to D69953: MCP: Fixed bug with dest overlapping copy source.
Nov 8 2019, 8:17 AM · Restricted Project
tpr updated the diff for D69953: MCP: Fixed bug with dest overlapping copy source.

V2: Used modifiesRegister as suggested by Matt.

Nov 8 2019, 8:17 AM · Restricted Project
tpr committed rG0703db398929: [CostModel] Fixed isExtractSubvectorMask for undef index off end (authored by tpr).
[CostModel] Fixed isExtractSubvectorMask for undef index off end
Nov 8 2019, 7:49 AM
tpr closed D70005: [CostModel] Fixed isExtractSubvectorMask for undef index off end.
Nov 8 2019, 7:49 AM · Restricted Project
tpr updated subscribers of D70005: [CostModel] Fixed isExtractSubvectorMask for undef index off end.
Nov 8 2019, 5:56 AM · Restricted Project
tpr added a reviewer for D70005: [CostModel] Fixed isExtractSubvectorMask for undef index off end: RKSimon.
Nov 8 2019, 5:56 AM · Restricted Project
tpr added a comment to rL346510: [CostModel] Add SK_ExtractSubvector handling to getInstructionThroughput….

This is slightly broken for us. See D70005 for a fix.

Nov 8 2019, 5:49 AM
tpr created D70005: [CostModel] Fixed isExtractSubvectorMask for undef index off end.
Nov 8 2019, 5:49 AM · Restricted Project

Nov 7 2019

tpr added reviewers for D69953: MCP: Fixed bug with dest overlapping copy source: bogner, efriedma.
Nov 7 2019, 9:42 AM · Restricted Project
tpr created D69953: MCP: Fixed bug with dest overlapping copy source.
Nov 7 2019, 9:42 AM · Restricted Project

Oct 29 2019

tpr added a comment to D69557: AsmParser: Allow FMF on varargs call.

Does D69161 cover this change? If so, we could add the test there and fix the problem in 1 step.

I think this bug fix is orthogonal to D69161, which adds new functionality.

Oct 29 2019, 10:07 AM · Restricted Project
tpr added reviewers for D69557: AsmParser: Allow FMF on varargs call: spatel, hfinkel.
Oct 29 2019, 5:10 AM · Restricted Project
tpr created D69557: AsmParser: Allow FMF on varargs call.
Oct 29 2019, 5:01 AM · Restricted Project

Sep 30 2019

tpr updated the diff for D68231: [SLC] Allow llvm.pow(x,2.0) -> x*x etc even if no pow() lib func.

V2: Better fix that does not accidentally allow pow() transforms.

Sep 30 2019, 11:43 AM · Restricted Project
tpr updated subscribers of D68231: [SLC] Allow llvm.pow(x,2.0) -> x*x etc even if no pow() lib func.

@foad pointed out that this fix is wrong. TLI saying pow() is not supported means if we find a call to a function called pow() then we don't know its semantics. So I will push a revised fix.

Sep 30 2019, 11:06 AM · Restricted Project
tpr added reviewers for D68231: [SLC] Allow llvm.pow(x,2.0) -> x*x etc even if no pow() lib func: evandro, xbolva00.
Sep 30 2019, 9:35 AM · Restricted Project
tpr created D68231: [SLC] Allow llvm.pow(x,2.0) -> x*x etc even if no pow() lib func.
Sep 30 2019, 9:33 AM · Restricted Project

Sep 23 2019

tpr committed rL372563: Request commit access for tpr.
Request commit access for tpr
Sep 23 2019, 2:21 AM

Sep 18 2019

tpr committed rG178611711122: [AMDGPU] Allow FP inline constant in v_madak_f16 and v_fmaak_f16 (authored by tpr).
[AMDGPU] Allow FP inline constant in v_madak_f16 and v_fmaak_f16
Sep 18 2019, 2:31 AM
tpr committed rL372208: [AMDGPU] Allow FP inline constant in v_madak_f16 and v_fmaak_f16.
[AMDGPU] Allow FP inline constant in v_madak_f16 and v_fmaak_f16
Sep 18 2019, 2:30 AM
tpr closed D67680: [AMDGPU] Allow FP inline constant in v_madak_f16 and v_fmaak_f16.
Sep 18 2019, 2:30 AM · Restricted Project

Sep 17 2019

tpr added a comment to D67680: [AMDGPU] Allow FP inline constant in v_madak_f16 and v_fmaak_f16.

LGTM. Can you also add a test for the mad case

Sep 17 2019, 3:06 PM · Restricted Project
tpr added reviewers for D67680: [AMDGPU] Allow FP inline constant in v_madak_f16 and v_fmaak_f16: arsenm, kzhuravl, rampitec, vpykhtin.
Sep 17 2019, 2:34 PM · Restricted Project
tpr created D67680: [AMDGPU] Allow FP inline constant in v_madak_f16 and v_fmaak_f16.
Sep 17 2019, 2:27 PM · Restricted Project

Sep 11 2019

tpr committed rGc26b3940c320: [TLI][AMDGPU] AMDPAL does not have library functions (authored by tpr).
[TLI][AMDGPU] AMDPAL does not have library functions
Sep 11 2019, 12:30 AM
tpr committed rL371592: [TLI][AMDGPU] AMDPAL does not have library functions.
[TLI][AMDGPU] AMDPAL does not have library functions
Sep 11 2019, 12:26 AM
tpr closed D67406: [TLI][AMDGPU] AMDPAL does not have tan function.
Sep 11 2019, 12:26 AM · Restricted Project

Sep 10 2019

tpr updated the diff for D67406: [TLI][AMDGPU] AMDPAL does not have tan function.

V2: Disable all library functions, not just tan.

Sep 10 2019, 12:09 PM · Restricted Project
tpr added inline comments to D67406: [TLI][AMDGPU] AMDPAL does not have tan function.
Sep 10 2019, 9:34 AM · Restricted Project
tpr added reviewers for D67406: [TLI][AMDGPU] AMDPAL does not have tan function: dlj, nhaehnle.
Sep 10 2019, 9:19 AM · Restricted Project
tpr created D67406: [TLI][AMDGPU] AMDPAL does not have tan function.
Sep 10 2019, 9:13 AM · Restricted Project

Sep 2 2019

tpr added a comment to D67003: AMDGPU: Don't put constants in .text for Mesa.

I just noticed that this already came up in D65813 and it does the right thing, it's just waiting review.

Sep 2 2019, 4:19 AM · Restricted Project

Aug 13 2019

tpr committed rG10db641aabf0: [AMDGPU] Fix to 'Fold readlane from copy of SGPR or imm' (authored by tpr).
[AMDGPU] Fix to 'Fold readlane from copy of SGPR or imm'
Aug 13 2019, 12:02 PM
tpr committed rL368736: [AMDGPU] Fix to 'Fold readlane from copy of SGPR or imm'.
[AMDGPU] Fix to 'Fold readlane from copy of SGPR or imm'
Aug 13 2019, 11:57 AM