Craig Topper f4ddf70574 [X86] Improve the gather scheduler models for SkylakeClient and SkylakeServer ...

The load ports need a cycle for each potentially loaded element just like Haswell and Skylake. Unlike Haswell and Broadwell, the number of uops does not scale with the number of elements. Instead the load uops run for multiple cycles.

I've taken the latency number from the uops.info. The port binding for the non-load uops is taken from the original IACA data I have.

Differential Revision: https://reviews.llvm.org/D74000

2020-02-05 13:26:47 -08:00

..

AArch64

[MCA] Fix test cases (NFC)

2019-11-22 16:19:58 -06:00

ARM

[ARM][Thumb2] Fix ADD/SUB invalid writes to SP

2020-01-14 11:47:19 +00:00

SystemZ

[MCA] Show aggregate over Average Wait times for the whole snippet (PR43219)

2019-10-10 14:46:21 +00:00

X86

[X86] Improve the gather scheduler models for SkylakeClient and SkylakeServer

2020-02-05 13:26:47 -08:00

invalid_input_file_name.test

Replace unused output filenames with /dev/null in tests

2018-07-02 18:16:44 +00:00

lit.local.cfg

[lit] Delete empty lines at the end of lit.local.cfg NFC

2019-06-17 09:51:07 +00:00