Commit Graph

9 Commits

Author SHA1 Message Date
Ryan Houdek
2152d1b2e9
Scripts: Fix issue in aarch64_fit_native
Apparently I messed this up in testing, is now fixed.
2024-08-09 21:04:26 -07:00
Ryan Houdek
4beb29141b
Scripts: Workaround deprecated parse_version
Different approach from #3579

Instead of completely ddropping the deprecated path, support the new
path and the old path using python try-except import exceptions.

This allows us to continue using old packages in CI, while supporting
the future API once pkg_resources gets deprecated and removed. Best of
both worlds.
2024-07-31 01:57:31 -07:00
Ryan Houdek
7f1464b135 Scripts: Update fit_native script for X1C/A78C
Cortex-X1C and A78C are relatively minor changes to their non-C
counterparts. Support classifying them in case clang understands them.

Fixes a minor perf regression noticed on the Lenovo X13s while testing.
2023-02-18 23:18:48 -08:00
Ryan Houdek
fba698cb74 Scripts: Updates AArch64 fit for Clang 14
Clang now supports these latest ARMv9 CPUs
2022-04-01 18:08:02 -07:00
Ryan Houdek
9911fe68d4 Scripts: Stop using deprecated Distutils
According to PEP 386: https://www.python.org/dev/peps/pep-0386/

distutils is deprecated and will be removed in an upcoming python
version.

Switch over to pkg_resources for version parsing and comparison
2022-03-01 07:10:41 -08:00
Ryan Houdek
bb7fa84fb8 Scripts: Updates CPU fitting script for latest CPUs
Clang-13 doesn't yet understand the latest ARM CPUs so just document them
and set to the closest thing.
2022-02-26 02:14:52 -08:00
Ryan Houdek
0bfc1bbe70 Arm64: Don't fall back to native
In the case of Arm64, make sure not to fallback to native if we hit an
unsupported CPU.
Can cause issues depending on system configuration.
2021-10-21 20:39:49 -07:00
Ryan Houdek
b06ebf486f Work around M1 Parallels hypervisor not showing CPU type
Parallels claims that the CPU is of implementor 0x41 and part number 0.
Easy enough to work around before real numbers get exposed
2021-04-27 22:26:35 -07:00
Ryan Houdek
921867de7e Works around Clang failing to identify new Kryo CPUs
Some of the newer CPU cores in LLVM's source claim to be a Cortex-A73,
which means they become limited to an ARMv8.0 feature set.

This is what you get if you compile FEX with -mcpu=native

To work around this issue, manually parse /proc/cpuinfo ourselves and
pull out the CPU type to pass to clang directly.
This also fixes the issue that we were using -march on AArch64, which no
longer works on newer clang versions. We instead need to use mcpu or
mtune.

Should improve all atomic op performance outside of the JITs, where they
were turning in to loadstore exclusive pairs.
2021-01-19 03:21:10 -08:00