osa1 (Ömer Sinan Ağacan)
User

Projects

User does not belong to any projects.

User Details

User Since
Jun 6 2014, 2:22 AM (214 w, 2 d)

Recent Activity

Fri, Jul 13

osa1 committed rGHC2625f1310ede: Fix processHeapClosureForDead CONSTR_NOCAF case (authored by osa1).
Fix processHeapClosureForDead CONSTR_NOCAF case
Fri, Jul 13, 7:22 AM
osa1 closed D4928: Fix processHeapClosureForDead CONSTR_NOCAF case:.
Fri, Jul 13, 7:21 AM
osa1 added inline comments to D4962: Run StgCse after unarise, fixes #15300.
Fri, Jul 13, 7:19 AM
osa1 updated the diff for D4962: Run StgCse after unarise, fixes #15300.
  • Fix existing test
Fri, Jul 13, 7:19 AM
osa1 added a comment to D4928: Fix processHeapClosureForDead CONSTR_NOCAF case:.

@simonmar could you take a look at the patch and my explanation? Am I missing anything?

Fri, Jul 13, 4:50 AM
osa1 created D4962: Run StgCse after unarise, fixes #15300.
Fri, Jul 13, 4:27 AM

Thu, Jul 12

osa1 committed rGHC3efd7cd9bacc: Minor refactoring in CmmUtils.mkLiveness (authored by osa1).
Minor refactoring in CmmUtils.mkLiveness
Thu, Jul 12, 9:12 AM
osa1 closed D4957: Minor refactoring in CmmUtils.mkLiveness.
Thu, Jul 12, 9:12 AM

Wed, Jul 11

osa1 created D4957: Minor refactoring in CmmUtils.mkLiveness.
Wed, Jul 11, 11:43 PM
osa1 abandoned D4954: Document why we can't apply MVar ownership semantics to BlockedOnSTM.

D4956 the ticket by removing all locking so no need for this comment now.

Wed, Jul 11, 1:25 PM
osa1 abandoned D4945: SAT: Fix hacky name shadowing.

We discussed this with @simonpj -- it turns out we can shadow binders using the same unique. Closing this. (see also the ticket)

Wed, Jul 11, 1:24 PM
osa1 accepted D4956: Fix deadlock between STM and throwTo.

Thanks! I was looking at this code today and was terribly confused ...

Wed, Jul 11, 1:21 PM
osa1 created D4954: Document why we can't apply MVar ownership semantics to BlockedOnSTM.
Wed, Jul 11, 3:24 AM

Tue, Jul 10

osa1 added inline comments to D4945: SAT: Fix hacky name shadowing.
Tue, Jul 10, 5:12 AM
osa1 updated the diff for D4945: SAT: Fix hacky name shadowing.
  • Implement RHS renaming
Tue, Jul 10, 5:11 AM
osa1 abandoned D4950: Minor refactoring in messageBlackHole.

Just realized that this is used by BLACKHOLE entry code.

Tue, Jul 10, 4:09 AM
osa1 created D4950: Minor refactoring in messageBlackHole.
Tue, Jul 10, 4:05 AM

Mon, Jul 9

osa1 updated the test plan for D4945: SAT: Fix hacky name shadowing.
Mon, Jul 9, 8:49 AM
osa1 updated the test plan for D4945: SAT: Fix hacky name shadowing.
Mon, Jul 9, 8:22 AM
osa1 created D4945: SAT: Fix hacky name shadowing.
Mon, Jul 9, 3:50 AM

Fri, Jul 6

osa1 added a comment to D4942: docs: remove leftovers of static flags.

Could we also remove :type: fields in ghc-flags?

Fri, Jul 6, 11:28 AM

Wed, Jul 4

osa1 committed rGHC8f449955e041: Revert "Don't lock the MVar closure on tryReadMVar" (authored by osa1).
Revert "Don't lock the MVar closure on tryReadMVar"
Wed, Jul 4, 9:26 AM
osa1 added a reverting change for rGHC6bb0c5db818c: Don't lock the MVar closure on tryReadMVar: rGHC8f449955e041: Revert "Don't lock the MVar closure on tryReadMVar".
Wed, Jul 4, 9:26 AM
osa1 added a comment to D4905: Don't lock the MVar closure on tryReadMVar.

Reverted with 8f449955e0417ca7b2d3b324262aa8d1a87ad822 in coordination with @dfeuer

Wed, Jul 4, 9:12 AM
osa1 committed rGHCf6ac0833b733: Add regression test for #15007 (authored by osa1).
Add regression test for #15007
Wed, Jul 4, 9:04 AM
osa1 committed rGHCe835fdb18cca: Add regression test for #15321 (authored by osa1).
Add regression test for #15321
Wed, Jul 4, 9:04 AM
osa1 committed rGHC39de4e3d33dd: Fix errors caused by invalid candidates leaking from hole fits (authored by Tritlo).
Fix errors caused by invalid candidates leaking from hole fits
Wed, Jul 4, 7:48 AM
osa1 closed D4909: Fix errors caused by invalid candidates leaking from hole fits.
Wed, Jul 4, 7:48 AM
osa1 committed rGHC15bb4e0b6c08: Fix nptr field alignment in RtClosureInspect (authored by osa1).
Fix nptr field alignment in RtClosureInspect
Wed, Jul 4, 1:14 AM
osa1 closed D4906: Fix nptr field alignment in RtClosureInspect.
Wed, Jul 4, 1:14 AM
osa1 committed rGHC6cb189d1d465: RtClosureInspect: add some docs, remove unused stuff (authored by osa1).
RtClosureInspect: add some docs, remove unused stuff
Wed, Jul 4, 1:14 AM
osa1 closed D4911: RtClosureInspect: add some docs, remove unused stuff.
Wed, Jul 4, 1:14 AM
osa1 updated the diff for D4906: Fix nptr field alignment in RtClosureInspect.
  • Say more about offset calculation
Wed, Jul 4, 1:11 AM

Tue, Jul 3

osa1 added a comment to D4905: Don't lock the MVar closure on tryReadMVar.

Ping @dfeuer

Tue, Jul 3, 2:40 PM
osa1 updated the test plan for D4928: Fix processHeapClosureForDead CONSTR_NOCAF case:.
Tue, Jul 3, 5:16 AM
osa1 added a comment to D4567: Fix processHeapClosureForDead CONSTR_NOCAF case:.

Reopened as D4928.

Tue, Jul 3, 4:36 AM
osa1 added a comment to D4928: Fix processHeapClosureForDead CONSTR_NOCAF case:.

This patch was previously submitted as D4567, but it got merged accidentally and then revereted and now I can't reopen it, so I'm submitting a new one.

Tue, Jul 3, 4:33 AM
osa1 created D4928: Fix processHeapClosureForDead CONSTR_NOCAF case:.
Tue, Jul 3, 4:23 AM
osa1 abandoned D4900: GC: remove unused field, fix gen spinlock alignment.
Tue, Jul 3, 1:00 AM

Sun, Jul 1

osa1 committed rGHCb4e64839a95d: testsuite: remove unused scc001 target (authored by osa1).
testsuite: remove unused scc001 target
Sun, Jul 1, 6:39 AM

Sat, Jun 30

osa1 added a comment to D4906: Fix nptr field alignment in RtClosureInspect.

@simonpj, I started documenting this module in D4911.

Sat, Jun 30, 1:30 AM
osa1 created D4911: RtClosureInspect: add some docs, remove unused stuff.
Sat, Jun 30, 1:01 AM
osa1 added a comment to D4905: Don't lock the MVar closure on tryReadMVar.

Perhaps we should revert this before we forget about this.

Sat, Jun 30, 12:50 AM

Fri, Jun 29

osa1 accepted D4905: Don't lock the MVar closure on tryReadMVar.

This makes sense to me. There's only one read operation here and no need to lock the closure for that.

Fri, Jun 29, 1:12 PM
osa1 updated the summary of D4906: Fix nptr field alignment in RtClosureInspect.
Fri, Jun 29, 4:39 AM
osa1 added inline comments to D4906: Fix nptr field alignment in RtClosureInspect.
Fri, Jun 29, 4:39 AM
osa1 updated the summary of D4906: Fix nptr field alignment in RtClosureInspect.
Fri, Jun 29, 4:32 AM
osa1 committed rGHC4760a8c1d5f5: Add -ddump-rtti to user's guide and man page (authored by osa1).
Add -ddump-rtti to user's guide and man page
Fri, Jun 29, 1:48 AM
osa1 closed D4903: Add -ddump-rtti to user's guide and man page.
Fri, Jun 29, 1:48 AM
osa1 updated the Trac tickets for D4906: Fix nptr field alignment in RtClosureInspect.
Fri, Jun 29, 1:44 AM
osa1 created D4906: Fix nptr field alignment in RtClosureInspect.
Fri, Jun 29, 1:44 AM

Thu, Jun 28

osa1 created D4903: Add -ddump-rtti to user's guide and man page.
Thu, Jun 28, 4:37 AM
osa1 committed rGHC904abd4e6922: Document SRT scavenging behavior of scavenge_block() and scavenge_one() (authored by osa1).
Document SRT scavenging behavior of scavenge_block() and scavenge_one()
Thu, Jun 28, 12:44 AM
osa1 closed D4893: Document SRT scavenging behavior of scavenge_block() and scavenge_one().
Thu, Jun 28, 12:44 AM

Wed, Jun 27

osa1 committed rGHC437ff69e2f4d: Add ghc-prim as dependency to ghc-bin (authored by osa1).
Add ghc-prim as dependency to ghc-bin
Wed, Jun 27, 5:58 AM
osa1 closed D4901: Add ghc-prim as dependency to ghc-bin.
Wed, Jun 27, 5:58 AM
osa1 updated the summary of D4901: Add ghc-prim as dependency to ghc-bin.
Wed, Jun 27, 5:06 AM
osa1 added inline comments to D4892: Show addresses of live objects in GHCi leak check.
Wed, Jun 27, 4:27 AM
osa1 created D4901: Add ghc-prim as dependency to ghc-bin.
Wed, Jun 27, 4:27 AM
osa1 added a comment to D4900: GC: remove unused field, fix gen spinlock alignment.

Ooh wait, I think I get it. The next field should also be aligned, correct?

Wed, Jun 27, 4:21 AM
osa1 added a comment to D4900: GC: remove unused field, fix gen spinlock alignment.

OK. What changes do you want exactly? The alignment was previously broken and I fixed it in this patch.

Wed, Jun 27, 4:21 AM
osa1 committed rGHCa54c94f08b93: Show addresses of live objects in GHCi leak check (authored by osa1).
Show addresses of live objects in GHCi leak check
Wed, Jun 27, 2:38 AM
osa1 closed D4892: Show addresses of live objects in GHCi leak check.
Wed, Jun 27, 2:38 AM
osa1 committed rGHC4168ee3a503f: rts: Update some comments, minor refactoring (authored by osa1).
rts: Update some comments, minor refactoring
Wed, Jun 27, 2:38 AM
osa1 added inline comments to D4892: Show addresses of live objects in GHCi leak check.
Wed, Jun 27, 2:32 AM
osa1 added a comment to D4900: GC: remove unused field, fix gen spinlock alignment.

I don't know how useful this alignment is. It seems like we don't align the generation itself so it seems to me that this alignment is useless. Am I missing anything?

Wed, Jun 27, 2:29 AM
osa1 created D4900: GC: remove unused field, fix gen spinlock alignment.
Wed, Jun 27, 2:28 AM

Tue, Jun 26

osa1 committed rGHC5865e9a75c26: Typo fix in rts [skip ci] (authored by osa1).
Typo fix in rts [skip ci]
Tue, Jun 26, 9:25 AM
osa1 added a comment to D4893: Document SRT scavenging behavior of scavenge_block() and scavenge_one().

@simonmar how does it look now?

Tue, Jun 26, 9:24 AM
osa1 updated the diff for D4893: Document SRT scavenging behavior of scavenge_block() and scavenge_one().
  • Update
Tue, Jun 26, 9:24 AM

Mon, Jun 25

osa1 added inline comments to D4892: Show addresses of live objects in GHCi leak check.
Mon, Jun 25, 2:09 PM
osa1 created D4893: Document SRT scavenging behavior of scavenge_block() and scavenge_one().
Mon, Jun 25, 8:37 AM
osa1 updated the diff for D4892: Show addresses of live objects in GHCi leak check.
  • Mask tag bits
Mon, Jun 25, 8:22 AM
osa1 created D4892: Show addresses of live objects in GHCi leak check.
Mon, Jun 25, 8:02 AM

Tue, Jun 19

osa1 committed rGHCa9b01c08aab0: Mark some TH tests as req_interp (authored by osa1).
Mark some TH tests as req_interp
Tue, Jun 19, 2:24 AM

Sun, Jun 17

osa1 added a comment to D4862: Use __FILE__ for Cmm assertion locations, fix #8619.

It does!

Sun, Jun 17, 10:45 AM
osa1 created D4862: Use __FILE__ for Cmm assertion locations, fix #8619.
Sun, Jun 17, 6:01 AM

Jun 14 2018

osa1 committed rGHCf7b9456cbec6: Minor refactoring and docs in selector optimisation (authored by osa1).
Minor refactoring and docs in selector optimisation
Jun 14 2018, 6:52 AM
osa1 closed D4835: Minor refactoring and docs in selector optimisation.
Jun 14 2018, 6:52 AM
osa1 updated the diff for D4835: Minor refactoring and docs in selector optimisation.
  • Minor refactoring
Jun 14 2018, 6:26 AM
osa1 committed rGHCa3c0b42ee6ab: testsuite: Print summary even if interrupted (authored by bgamari).
testsuite: Print summary even if interrupted
Jun 14 2018, 4:34 AM
osa1 closed D4841: testsuite: Print summary even if interrupted.
Jun 14 2018, 4:34 AM
osa1 added a comment to D4840: Update user manual sections for -rtsopts and -with-rtsopts.

@AndreasK any other concerns?

Jun 14 2018, 1:25 AM
osa1 retitled D4840: Update user manual sections for -rtsopts and -with-rtsopts from Update user manual sections for -rtsopts, -no-rtsopts-suggestions to Update user manual sections for -rtsopts and -with-rtsopts.
Jun 14 2018, 1:25 AM
osa1 committed rGHCb5ccee49f96e: Do not skip conc004 in GHCi way (authored by osa1).
Do not skip conc004 in GHCi way
Jun 14 2018, 1:09 AM
osa1 closed D4831: Do not skip conc004 in GHCi way.
Jun 14 2018, 1:09 AM

Jun 13 2018

osa1 added a comment to D4835: Minor refactoring and docs in selector optimisation.

@simonmar Does my example makes sense? The notation may be a bit off but hopefully I get the idea right.

Jun 13 2018, 2:51 PM
osa1 updated the diff for D4835: Minor refactoring and docs in selector optimisation.
  • Apply Simon's suggestion, an example selector chain
Jun 13 2018, 2:51 PM
osa1 added a comment to D4828: Make Control.Exception.throw levity polymorphic..

No way to delete it. In the "add action" menu select "abandon revision" and then submit.

Jun 13 2018, 2:19 PM
osa1 updated the diff for D4840: Update user manual sections for -rtsopts and -with-rtsopts.
  • Put -no-rtsopts-suggestions back
Jun 13 2018, 9:12 AM
osa1 updated the diff for D4840: Update user manual sections for -rtsopts and -with-rtsopts.
  • Put -no-rtsopts-suggestions back
Jun 13 2018, 9:12 AM
osa1 committed rGHCa610c215580c: Fix some of the failures in sanity way (authored by osa1).
Fix some of the failures in sanity way
Jun 13 2018, 7:31 AM
osa1 closed D4839: Fix some of the failures in sanity way.
Jun 13 2018, 7:31 AM
osa1 accepted D4841: testsuite: Print summary even if interrupted.
Jun 13 2018, 7:27 AM
osa1 added a comment to D4841: testsuite: Print summary even if interrupted.

Awesome, thanks! Just tested this, this really improves my workflow.

Jun 13 2018, 7:27 AM
osa1 added inline comments to D4840: Update user manual sections for -rtsopts and -with-rtsopts.
Jun 13 2018, 5:42 AM
osa1 updated subscribers of D4840: Update user manual sections for -rtsopts and -with-rtsopts.
Jun 13 2018, 4:49 AM
osa1 updated the Trac tickets for D4840: Update user manual sections for -rtsopts and -with-rtsopts.
Jun 13 2018, 4:46 AM
osa1 updated the diff for D4840: Update user manual sections for -rtsopts and -with-rtsopts.
  • Add a few clarifications about -rtsopts and -with-rtsopts interaction
Jun 13 2018, 4:45 AM
osa1 created D4840: Update user manual sections for -rtsopts and -with-rtsopts.
Jun 13 2018, 4:17 AM