- User Since
- Apr 3 2019, 5:55 PM (115 w, 22 h)
Google keeps zero counts in the profile. It treats missing lines conservatively and tries to infer their hotness. I vaguely remember there is a flag added to control it.
Tue, Jun 15
Be consistent with dwarf line number based profile.
This is blocking our internal adoption of llvm-12, going to commit on behalf of Igor now.
Makes sense to me to still allow overriding. Thanks for the change.
Be consistent with non-probe profile.
Thanks for working on this. Among all deleted probes, if the majority of them are actually dangling (i.e removed due to hoisting, merging like optimization as opposed to dce), it makes sense to default to dangling and explicitly marking non-dangling. Previously we assumed that majority of deleted probes are not dangled, hence we choose to mark dangled ones explicitly.
Mon, Jun 14
Sat, Jun 12
Fri, Jun 11
lgtm, thanks for working on this Sergey!
Thanks for the patch. I think this can be useful for tuning. However as we discussed, it'd be nice if the level of merging (number of frames to keep after merging) can be adaptive based on hotness. But to do that, the preinliner in llvm-profgen is the way to go.
Thu, Jun 10
Tue, Jun 8
lgtm, thanks for the fix.
Sun, Jun 6
We also saw ipa-cp-clone being a very noticeable difference between gcc and llvm when we tried to move workloads from gcc to llvm. Thanks for working on this for llvm optimization parity with gcc.
Thu, Jun 3
Thu, May 27
Wed, May 26
Thanks for the change. I tried a few limits on the repro case I mentioned in D91722. It does look like 16 is a sweet spot at least for that case. :) Without forcing a limit, the case runs for 10+ hours..
Tue, May 25
We're seeing a pathological case with 36ec97f76ac0d8be76fb16ac521f55126766267d (3rd Reapply "[DebugInfo] Use variadic debug values to salvage BinOps and GEP instrs with non-const operands).
Mon, May 24
lgtm with a nit, thanks for the fix.
Wed, May 19
Tue, May 18
May 18 2021
lgtm too, thanks. looking forward to complete fs-adfo support and we'd be happy to give it a try.
May 17 2021
May 15 2021
Thanks for bringing this up - it makes sure we give full context profile precedence even when we had to use context-less profile in prelink.
lgtm, with some nit on comments.
May 14 2021
May 7 2021
May 3 2021
good catch, thanks for the fix!
Apr 27 2021
lgtm, thanks for the fix.
So without this fix, it would AV as well if a block only had dbg instruction?
Apr 26 2021
Apr 23 2021
lgtm, thanks for addressing the ICP issues.
Apr 22 2021
Apr 21 2021
Ignore trim/merge flag for non-CS profile.
The dependency on Analysis should be removed in the latest version.
Move threshold knobs into ProfileSummaryBuilder, avoid pulling Analysis lib into tools.
Apr 19 2021
Apr 14 2021
Apr 13 2021
Apr 12 2021
Thanks for the fix. How about using DAG check for things under other "Getting base profile for function" too?