Page MenuHomePhabricator

yrouban (Yevgeny Rouban)
User

Projects

User does not belong to any projects.

User Details

User Since
Jul 4 2017, 1:39 AM (103 w, 1 d)

Recent Activity

Today

yrouban updated the diff for D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.

Addressed most comments. The main changes are the following:

  • added more cases to trivial-unswitch-profmd.ll
  • brushed up basictest-profmd.ll
  • added NewSI default case weight if the default successor was not an exit
Wed, Jun 26, 1:10 AM · Restricted Project
yrouban added inline comments to D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.
Wed, Jun 26, 1:02 AM · Restricted Project

Mon, Jun 24

yrouban added a comment to D63706: [InlineCost] Fix bug 42084: remember negative result when computing full inline cost.

Please, see my comment to PR42084: https://bugs.llvm.org/show_bug.cgi?id=42084#c3.
A core of the problem is in two difference Cost-vs-Threshold checks used in analyzeBlock (Cost >= Threshold) and analyzeCall (Cost < max(1, Threshold)).
I believe a proper fix for this bug would be to use a unified Cost-vs-Threshold check everywhere.

I agree that those two checks seem to be inconsistent with each other, but I insist on my fix. It makes the method analyzeBlock() return same result (negative or positive) regardless of the flag ComputeFullInlineCost.
The root cause is that the analyzeBlock() returns different results (negative or positive) for different ComputeFullInlineCost. And the checks you mentioned if fixed could just hide this difference.

Mon, Jun 24, 11:09 PM · Restricted Project
yrouban set the repository for D63706: [InlineCost] Fix bug 42084: remember negative result when computing full inline cost to rG LLVM Github Monorepo.
Mon, Jun 24, 2:54 AM · Restricted Project
yrouban created D63706: [InlineCost] Fix bug 42084: remember negative result when computing full inline cost.
Mon, Jun 24, 2:54 AM · Restricted Project

Fri, Jun 21

yrouban added a comment to D63058: [InlineCost] Fix bug 42084: return the first negative result.

One issue bothers me: results of the InlineCost analyzer will depend on the flag ComputeFullInlineCost. Negative or positive decision will persist but the message and highest computed cost will be changing.

We already have the cost difference depending on this flag.
It seems to be rather obvious consequence of "full inline cost" notion and the difference should not be confusing...

I still believe that we should separate a bug fix from introducing a new feature. As I read the comments this patch looks to be complex to be just a bugfix but not full enough to introduce a new and complete feature.
I could provide a one line fix with a test if I believe it is a complete fix for the bug. In other words I'm a bit stuck. Any suggestions are welcome.

Fri, Jun 21, 1:41 AM · Restricted Project

Thu, Jun 20

yrouban committed rGd5e1ce3f44b0: [LICM & MSSA] Fixed test to run only with assertions enabled as it uses -debug… (authored by yrouban).
[LICM & MSSA] Fixed test to run only with assertions enabled as it uses -debug…
Thu, Jun 20, 9:49 PM
yrouban added inline comments to D63582: [LICM & MSSA] Limit unsafe sinking and hoisting..
Thu, Jun 20, 9:47 PM · Restricted Project
yrouban committed rL364005: [LICM & MSSA] Fixed test to run only with assertions enabled as it uses -debug….
[LICM & MSSA] Fixed test to run only with assertions enabled as it uses -debug…
Thu, Jun 20, 9:46 PM

Tue, Jun 18

yrouban updated the diff for D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.

Rebased to the latest HEAD. Please review. Its sibling patches have been landed.

Tue, Jun 18, 10:47 PM · Restricted Project

Mon, Jun 17

yrouban committed rG69daf4a72daf: [SimplifyCFG] NFC, prof branch_weighs handling is simplified (authored by yrouban).
[SimplifyCFG] NFC, prof branch_weighs handling is simplified
Mon, Jun 17, 11:50 PM
yrouban committed rL363652: [SimplifyCFG] NFC, prof branch_weighs handling is simplified.
[SimplifyCFG] NFC, prof branch_weighs handling is simplified
Mon, Jun 17, 11:50 PM
yrouban closed D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified.
Mon, Jun 17, 11:50 PM · Restricted Project

Sun, Jun 16

yrouban committed rGee62c40eae93: [SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases (authored by yrouban).
[SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases
Sun, Jun 16, 10:53 PM
yrouban committed rL363527: [SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases.
[SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases
Sun, Jun 16, 10:53 PM
yrouban closed D62186: [SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases.
Sun, Jun 16, 10:53 PM · Restricted Project

Fri, Jun 14

yrouban added inline comments to D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified.
Fri, Jun 14, 5:14 AM · Restricted Project
yrouban updated the diff for D63058: [InlineCost] Fix bug 42084: return the first negative result.
  • extracted a separate function GetFirstNegative()
  • changed the test to check that the bug is fixed
Fri, Jun 14, 4:48 AM · Restricted Project
yrouban added a comment to D63058: [InlineCost] Fix bug 42084: return the first negative result.

I think this is somewhat the wrong approach.
I think we need to widen the interface a bit to return two pieces of information instead of one:

  1. The highest cost computed
  2. Any inline-blocking construct encountered And we should phrase #2 as a bitmask so that we can return multiple things in it. One of them can be that we exceeded the threshold, another can be any terminal condition we additionally reached like recursion. I think this will also make the code more clear as we won't be throwing away any information at any step.

In general I agree with this approach. But the bigger change it needs the stronger intention I have to fix the bug with a minimal change and to develop the feature as another separate patch. One issue bothers me: results of the InlineCost analyzer will depend on the flag ComputeFullInlineCost. Negative or positive decision will persist but the message and highest computed cost will be changing.

Fri, Jun 14, 12:43 AM · Restricted Project

Tue, Jun 11

yrouban planned changes to D63058: [InlineCost] Fix bug 42084: return the first negative result.
Tue, Jun 11, 3:20 AM · Restricted Project

Sun, Jun 9

yrouban added a comment to D63058: [InlineCost] Fix bug 42084: return the first negative result.

2 questions, since i've read through the diff:

  1. will this pick the smallest cost, or the first negative cost?
  2. is this missing some abstraction? maybe InlineResult Result should be some wrapper that should be assigned-to, that will internally only accept the new value if it's better than what it currently has,
  1. Not the smallest cost, but the first negative result with its message.
  2. I think such an abstraction would not be much shorter: Result = InlineResult::FirstNegative(Result, expression)
Sun, Jun 9, 8:45 PM · Restricted Project
yrouban added a comment to D63058: [InlineCost] Fix bug 42084: return the first negative result.

tests?

Makes sense. I will try.

Sun, Jun 9, 8:39 PM · Restricted Project
yrouban created D63058: [InlineCost] Fix bug 42084: return the first negative result.
Sun, Jun 9, 7:22 AM · Restricted Project

Wed, Jun 5

yrouban reopened D50435: [Inliner] Attribute callsites with inline remarks.

Hi @yrouban,
Since you did some work in this area, can you look at this bug report please?
https://bugs.llvm.org/show_bug.cgi?id=42084

Interesting. I will try but next week.

Wed, Jun 5, 4:52 AM

Tue, Jun 4

yrouban committed rGa3e16719c46a: Resubmit "[CorrelatedValuePropagation] Fix prof branch_weights metadata… (authored by yrouban).
Resubmit "[CorrelatedValuePropagation] Fix prof branch_weights metadata…
Tue, Jun 4, 10:48 PM
yrouban added a reverting change for rG5b32f60ec31c: Revert "[CorrelatedValuePropagation] Fix prof branch_weights metadata handling…: rGa3e16719c46a: Resubmit "[CorrelatedValuePropagation] Fix prof branch_weights metadata….
Tue, Jun 4, 10:48 PM
yrouban committed rL362583: Resubmit "[CorrelatedValuePropagation] Fix prof branch_weights metadata….
Resubmit "[CorrelatedValuePropagation] Fix prof branch_weights metadata…
Tue, Jun 4, 10:47 PM
yrouban closed D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.
Tue, Jun 4, 10:47 PM · Restricted Project
yrouban committed rG4f9e68148bd0: Make SwitchInstProfUpdateWrapper safer (authored by yrouban).
Make SwitchInstProfUpdateWrapper safer
Tue, Jun 4, 2:04 AM
yrouban committed rL362473: Make SwitchInstProfUpdateWrapper safer.
Make SwitchInstProfUpdateWrapper safer
Tue, Jun 4, 2:04 AM
yrouban closed D62656: Make SwitchInstProfUpdateWrapper safer.
Tue, Jun 4, 2:04 AM · Restricted Project

Sun, Jun 2

yrouban updated the diff for D62656: Make SwitchInstProfUpdateWrapper safer.

fixed typos

Sun, Jun 2, 10:03 PM · Restricted Project

Fri, May 31

yrouban updated the diff for D62656: Make SwitchInstProfUpdateWrapper safer.

added option switch-inst-prof-update-wrapper-strict to assert if prof data is inconsistent. It is off by default.
Probably it makes sense to enable prof branch_weights validation (patch D61179) in the same way.

Fri, May 31, 2:09 AM · Restricted Project
yrouban planned changes to D62656: Make SwitchInstProfUpdateWrapper safer.

I suggest the following path:

  1. add an option to enforce internal error on inconsistency; by default the inconsistency is ignored.

Ok. This option will probably simplify catching such bugs.

Fri, May 31, 12:06 AM · Restricted Project

Thu, May 30

yrouban updated the diff for D62656: Make SwitchInstProfUpdateWrapper safer.

Formatted and fixed one typo.

Thu, May 30, 9:06 PM · Restricted Project
yrouban added inline comments to D62656: Make SwitchInstProfUpdateWrapper safer.
Thu, May 30, 9:06 PM · Restricted Project
yrouban added a comment to D62656: Make SwitchInstProfUpdateWrapper safer.

What I meant is that we should never see inconsistent state and instead of setting invalid state, we assert there.

Yes. But in LLVM there are many places to fix. So we have to fix all of them at once and set the assertion here. The other option is to fix them one by one and tolerate such inconsistencies.
For example, SimplifyCFG is safe: it checks that the prof data is ok and only then makes its prof changes.

Thu, May 30, 8:26 PM · Restricted Project
yrouban added inline comments to D62656: Make SwitchInstProfUpdateWrapper safer.
Thu, May 30, 11:08 AM · Restricted Project
yrouban requested review of D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.

Providing an obviously incorrect branch_weights for one of the added test cases is enough to cause an assertion failure:

I have just submitted a fix. Please see D62656. Once it is landed this patch can be landed unchanged.
I decided to not include the crash test case as it would be caught by verification proposed in D61179.

Thu, May 30, 6:08 AM · Restricted Project
yrouban added a parent revision for D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst: D62656: Make SwitchInstProfUpdateWrapper safer.
Thu, May 30, 6:02 AM · Restricted Project
yrouban added a child revision for D62656: Make SwitchInstProfUpdateWrapper safer: D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.
Thu, May 30, 6:02 AM · Restricted Project
yrouban created D62656: Make SwitchInstProfUpdateWrapper safer.
Thu, May 30, 6:02 AM · Restricted Project

Wed, May 29

yrouban planned changes to D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.

Thank you very much for the revert, the analysis and the test.
I'm going make the wrapper class safe by introducing invalid sticky state.

Wed, May 29, 4:37 AM · Restricted Project

Tue, May 28

yrouban committed rG53f2f3286572: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for… (authored by yrouban).
[CorrelatedValuePropagation] Fix prof branch_weights metadata handling for…
Tue, May 28, 4:33 AM
yrouban committed rL361808: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for….
[CorrelatedValuePropagation] Fix prof branch_weights metadata handling for…
Tue, May 28, 4:32 AM
yrouban closed D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.
Tue, May 28, 4:32 AM · Restricted Project

May 27 2019

yrouban added a comment to D62186: [SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases.

@davidxl, @eraman, @reames please review.

May 27 2019, 9:02 PM · Restricted Project
yrouban added a comment to D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified.

@davidxl, @eraman, @reames please review.

May 27 2019, 9:02 PM · Restricted Project
yrouban abandoned D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.

@asbirlea, @fedor.sergeev, @mkazantsev, @chandlerc could you please review the patch or suggest somebody who can take a look?

May 27 2019, 9:02 PM · Restricted Project
yrouban reclaimed D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.
May 27 2019, 9:02 PM · Restricted Project
yrouban added a comment to D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.

I considered this way and found it error prone. The wrapper stays alive after ConstantFoldTerminator() and developer may accidentally reuse it but the stored weights might be unsync with the underlying SwitchInst's weights changed by ConstantFoldTerminator().

May 27 2019, 8:55 PM · Restricted Project
yrouban updated the diff for D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.
  • renamed SwitchInstProfBranchWeightsWrapper to SwitchInstProfUpdateWrapper
  • explicitly narrowed scope of the SwitchInstProfUpdateWrapper object so it does not overlap with ConstantFoldTerminator()
  • added a test case where switch turns into conditional branch
May 27 2019, 7:36 AM · Restricted Project
yrouban added inline comments to D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.
May 27 2019, 7:16 AM · Restricted Project

May 23 2019

yrouban committed rGc652b3455ec5: [NFC] SwitchInst: Introduce wrapper for prof branch_weights handling (authored by yrouban).
[NFC] SwitchInst: Introduce wrapper for prof branch_weights handling
May 23 2019, 9:33 PM
yrouban committed rL361596: [NFC] SwitchInst: Introduce wrapper for prof branch_weights handling.
[NFC] SwitchInst: Introduce wrapper for prof branch_weights handling
May 23 2019, 9:33 PM
yrouban closed D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling.
May 23 2019, 9:33 PM · Restricted Project

May 22 2019

yrouban updated the diff for D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling.

Addressed comments:

  • renamed to SwitchInstProfUpdateWrapper
  • added method descriptions
  • added one assertion
May 22 2019, 8:11 AM · Restricted Project
yrouban added inline comments to D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling.
May 22 2019, 8:10 AM · Restricted Project

May 21 2019

yrouban added a child revision for D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling: D62186: [SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases.
May 21 2019, 4:01 AM · Restricted Project
yrouban added a parent revision for D62186: [SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases: D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling.
May 21 2019, 4:01 AM · Restricted Project
yrouban created D62186: [SimplifyCFG] Fix prof branch_weights MD while removing unreachable switch cases.
May 21 2019, 4:01 AM · Restricted Project
yrouban added inline comments to D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified.
May 21 2019, 12:03 AM · Restricted Project
yrouban updated the diff for D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified.

Done early return in eliminateDeadSwitchCases as suggested.

May 21 2019, 12:03 AM · Restricted Project

May 20 2019

yrouban updated the summary of D61179: Verifier: check prof branch_weights.
May 20 2019, 8:24 PM · Restricted Project
yrouban abandoned D60604: [IR] Improve SwitchInst API to support prof branch_weights.

See 62122 and its children.

May 20 2019, 8:23 PM · Restricted Project
yrouban abandoned D60554: Fix number of args of prof branch_weights MD for SwitchInst.

See 62122 and its children.

May 20 2019, 8:22 PM · Restricted Project
yrouban retitled D61179: Verifier: check prof branch_weights from [WIP] Verifier: check prof branch_weights to Verifier: check prof branch_weights.
May 20 2019, 8:19 PM · Restricted Project
yrouban added a child revision for D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst: D61179: Verifier: check prof branch_weights.
May 20 2019, 5:23 AM · Restricted Project
yrouban added a parent revision for D61179: Verifier: check prof branch_weights: D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.
May 20 2019, 5:23 AM · Restricted Project
yrouban edited parent revisions for D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst, added: 1; removed: 1.
May 20 2019, 5:23 AM · Restricted Project
yrouban added a child revision for D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling: D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.
May 20 2019, 5:23 AM · Restricted Project
yrouban removed a child revision for D60604: [IR] Improve SwitchInst API to support prof branch_weights: D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.
May 20 2019, 5:23 AM · Restricted Project
yrouban updated the diff for D60606: [SimpleLoopUnswitch] Implement handling of prof branch_weights metadata for SwitchInst.
  • made use of SwitchInstProfBranchWeightsWrapper introduced in D62122.
  • created new profmd tests instead of changing the existing tests
May 20 2019, 5:20 AM · Restricted Project
yrouban added a parent revision for D61179: Verifier: check prof branch_weights: D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.
May 20 2019, 1:38 AM · Restricted Project
yrouban added a child revision for D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst: D61179: Verifier: check prof branch_weights.
May 20 2019, 1:38 AM · Restricted Project
yrouban added a child revision for D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling: D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.
May 20 2019, 1:38 AM · Restricted Project
yrouban added a parent revision for D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst: D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling.
May 20 2019, 1:38 AM · Restricted Project
yrouban created D62126: [CorrelatedValuePropagation] Fix prof branch_weights metadata handling for SwitchInst.
May 20 2019, 1:38 AM · Restricted Project
yrouban added a reviewer for D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified: reames.
May 20 2019, 12:47 AM · Restricted Project
yrouban added a child revision for D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling: D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified.
May 20 2019, 12:47 AM · Restricted Project
yrouban added a parent revision for D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified: D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling.
May 20 2019, 12:47 AM · Restricted Project
yrouban created D62123: [NFC] SimplifyCFG prof branch_weighs handling is simplified.
May 20 2019, 12:47 AM · Restricted Project

May 19 2019

yrouban created D62122: [NFC] Introduce SwitchInst wrapper for prof branch_weights handling.
May 19 2019, 11:54 PM · Restricted Project

May 15 2019

yrouban committed rGbf6df042a5d8: Fix prof branch_weights in entry_counts_missing_dbginfo.ll test (authored by yrouban).
Fix prof branch_weights in entry_counts_missing_dbginfo.ll test
May 15 2019, 8:37 PM
yrouban committed rL360843: Fix prof branch_weights in entry_counts_missing_dbginfo.ll test.
Fix prof branch_weights in entry_counts_missing_dbginfo.ll test
May 15 2019, 8:37 PM
yrouban closed D61932: Fix prof branch_weights in entry_counts_missing_dbginfo.ll test.
May 15 2019, 8:37 PM · Restricted Project
yrouban added a parent revision for D61179: Verifier: check prof branch_weights: D61932: Fix prof branch_weights in entry_counts_missing_dbginfo.ll test.
May 15 2019, 12:49 AM · Restricted Project
yrouban added a child revision for D61932: Fix prof branch_weights in entry_counts_missing_dbginfo.ll test: D61179: Verifier: check prof branch_weights.
May 15 2019, 12:49 AM · Restricted Project
yrouban created D61932: Fix prof branch_weights in entry_counts_missing_dbginfo.ll test.
May 15 2019, 12:42 AM · Restricted Project

Apr 29 2019

yrouban committed rG0822bfc6de4b: [LoopSimplifyCFG] Suppress expensive DomTree verification (authored by yrouban).
[LoopSimplifyCFG] Suppress expensive DomTree verification
Apr 29 2019, 6:29 AM
yrouban committed rL359446: [LoopSimplifyCFG] Suppress expensive DomTree verification.
[LoopSimplifyCFG] Suppress expensive DomTree verification
Apr 29 2019, 6:29 AM
yrouban closed D61055: [LoopSimplifyCFG] Suppress expensive DomTree verification in LoopSimplifyCFG.
Apr 29 2019, 6:29 AM · Restricted Project
yrouban updated the diff for D61055: [LoopSimplifyCFG] Suppress expensive DomTree verification in LoopSimplifyCFG.

changed verification level from Basic to Fast.

Apr 29 2019, 2:23 AM · Restricted Project

Apr 28 2019

yrouban added a comment to D61055: [LoopSimplifyCFG] Suppress expensive DomTree verification in LoopSimplifyCFG.

It of course depends on exactly location, but we found that Fast is generally not quick enough to be run during a typical loop pass (on large files, or loopy code from non-clang frontends). I claim that it's better to make it obvious that these checks are slow, and make developers either guard them with EXEPNSIVE_CHECKS or opt-in with Fast.

Does it mean that you lgtm if I will remove the else part of the change (дштуы 604-606).

Apr 28 2019, 8:41 PM · Restricted Project
yrouban added a comment to D61055: [LoopSimplifyCFG] Suppress expensive DomTree verification in LoopSimplifyCFG.

Hello. I think it's fine to always use DominatorTree::VerificationLevel::Fast, the others are more about checking that the DomTree construction code was correct. Fast will compare against a new tree, so for updates will check the the tree is correct.

Apr 28 2019, 7:08 AM · Restricted Project

Apr 26 2019

yrouban created D61179: Verifier: check prof branch_weights.
Apr 26 2019, 3:05 AM · Restricted Project

Apr 23 2019

yrouban added inline comments to D43938: [RewriteStatepoints] Do not instert Store before Phi.
Apr 23 2019, 10:38 PM
yrouban created D61055: [LoopSimplifyCFG] Suppress expensive DomTree verification in LoopSimplifyCFG.
Apr 23 2019, 9:39 PM · Restricted Project

Apr 18 2019

yrouban added inline comments to D60751: [NFC][InlineCost] cleanup - comments, overflow handling..
Apr 18 2019, 12:42 AM · Restricted Project

Apr 16 2019

yrouban created D60769: [SimpleLoopUnswitch] Discard stale prof branch_weights for partial unswitched branches.
Apr 16 2019, 4:32 AM · Restricted Project