Commit graph

6 commits

Author SHA1 Message Date
Sylvia Harrington e4e628ab84 sb-simd: Fix a bunch of bugs
64-bit integer min/max don't have actual instructions, fake instead
Fix signed lane extraction not sign-extending
Add float abs/neg
Fix fsqrt encoding
Fix u32.4-from-u64.2/s32.4-from-s64.2 family, wrong size
Remove 64-bit integer vector mul, instruction doesn't exist
Add missing exports for integer min/max
Align spilled neon registers properly
Validate more vector instruction forms properly
Fix movi immediates
2026-08-28 22:38:41 +03:00
Stas Boukarev 8ee5b24141 sb-simd: separate x86-64 and arm64 things 2026-08-02 17:02:43 +03:00
Sylvia Harrington cb716e24d6 sb-simd: arm64 support 2026-07-01 01:15:08 +03:00
Stas Boukarev 5dd9e7af5c Spell things. 2023-11-30 01:31:40 +03:00
Sergio Durigan Junior 7790dffb61 Check if AVX2 is supported when running sb-simd.impure.lisp 2023-03-17 01:17:35 +03:00
Douglas Katzman fbd46c52ea Run contrib tests with all other tests and not in make-target-contrib
Most build systems distinguish a failed build from a failed test run,
but make.sh has a hard time doing that because of the conflation of the two.

The whole regression suite should be run if you want to ensure a good build,
so that would be a good time to test contribs. Not only that, with ASDF
we obscured a ton of style-warnings, we lost sandboxing of input files,
the ability to use --evaluator-mode, automatic generation and cleanup of
scratch pathnames, and automatic sb-sprof profiling.

So convert contrib tests to use WITH-TEST except some that gave me trouble.
This makes the output a ton more readable, and makes bisection on seldom-used
configurations quicker, not to mention that SB-RT is very lame anyway.
And there is quite literally less code to maintain now. Go figure.
2022-09-01 21:17:47 -04:00