The dwarf discriminator field for pseudo probes is not supposed to be used until the first FS discriminator pass. Unfortunately there are always corner cases that accidientally set this field. For example, the inliner could set this field for an inlined instruction if the instruction does not come with any debug information. While fixing all such spots is possible, but for future-proff I'd like to enforce a general cleanup before assigning probes any FS discriminator.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
llvm/test/CodeGen/X86/fsafdo_probe.ll | ||
---|---|---|
68–69 | why is this test change? If anything, I was expecting non-zero -> zero for probe discriminators.. |
llvm/test/CodeGen/X86/fsafdo_probe.ll | ||
---|---|---|
68–69 | This is to test the non-zero discriminator can be cleaned up by the first FS discriminator pass. |
llvm/test/CodeGen/X86/fsafdo_probe.ll | ||
---|---|---|
68–69 | Oh, I see. Can you maybe add a comment? Is the "from 0 -> xxx" covering the check on clean up non-zero discriminator? |
llvm/test/CodeGen/X86/fsafdo_probe.ll | ||
---|---|---|
68–69 | Done. |
Comment Actions
Updating D150741: [FS-AFDO] Clean up non-zero discriminator for pseudo probes at the first FS discriminator pass.
why is this test change? If anything, I was expecting non-zero -> zero for probe discriminators..