Change Details

This is somewhat partial. Overview: * Latencies are reasonably-good {F7345689} * There are still some inconsistencies. * *Many* of the inconsistencies are noise * fp measurements are flaky * Non-fp measurements are somewhat flaky too * NumMicroOps are somewhat good {F7345689} * Most of the remaining inconsistencies are from `Ld` / `Ld_ReadAfterLd` classes * Actual unit occupation (pipes, `ResourceCycles`) are undiscovered lands, i did not really look there. They are basically verbatum copy from `btver2` * Many `InstRW`. And there are still inconsistencies left... What's not here: * llvm-mca test coverage. I understand how to not just add the coverage here, but actually show a diff, but i did not look there yet. * benchmarks. I'd say it's too soon for any measurements. To be noted: I think this is the first new schedule profile produced with the new next-gen tools like llvm-exegesis! llvm-exegesis details: {F7345685} {F7345686} {F7345687} {F7345689} {F7345690}

This is somewhat partial. Overview: * Latencies are good {F7354935} * All of these remaining inconsistencies //appear// to be noise/noisy/flaky. * NumMicroOps are somewhat good {F7354949} * Most of the remaining inconsistencies are from `Ld` / `Ld_ReadAfterLd` classes * Actual unit occupation (pipes, `ResourceCycles`) are undiscovered lands, i did not really look there. They are basically verbatum copy from `btver2` * Many `InstRW`. And there are still inconsistencies left... What's not here: * llvm-mca test coverage. I understand how to not just add the coverage here, but actually show a diff, but i did not look there yet. * benchmarks. I'd say it's too soon for any measurements. To be noted: I think this is the first new schedule profile produced with the new next-gen tools like llvm-exegesis! llvm-exegesis details: {F7354931} {F7354935} {F7354950} {F7354952} {F7354949}

This is somewhat partial. Overview: * Latencies are reasonably-good {F7345689}54935} * There are still some* All of these remaining inconsistenciesies //appear// to be noise/noisy/flaky. * *Many* of the inconsistencies are noise * fp measurements are flaky * Non-fp measurements are somewhat flaky too * NumMicroOps are somewhat good {F7345689}54949} * Most of the remaining inconsistencies are from `Ld` / `Ld_ReadAfterLd` classes * Actual unit occupation (pipes, `ResourceCycles`) are undiscovered lands, i did not really look there. They are basically verbatum copy from `btver2` * Many `InstRW`. And there are still inconsistencies left... What's not here: * llvm-mca test coverage. I understand how to not just add the coverage here, but actually show a diff, but i did not look there yet. * benchmarks. I'd say it's too soon for any measurements. To be noted: I think this is the first new schedule profile produced with the new next-gen tools like llvm-exegesis! llvm-exegesis details: {F7345685} 54931} {F7345686} 54935} {F7345687} 54950} {F7345689} 54952} {F7345690}54949}