gingerBill
620bf162a0
Cache const string16 in LLVM
2025-08-02 12:32:18 +01:00
gingerBill
bb4bc316a4
for in string16; Support string16 across core
2025-08-02 12:20:35 +01:00
gingerBill
ae02d3d02d
Begin supporting string16 across the core library
2025-08-02 11:55:16 +01:00
gingerBill
2561427dd3
Add string16 and cstring16 (UTF-16 based strings)
2025-08-02 11:00:15 +01:00
Laytan
710203eadb
Merge pull request #5505 from RoboMage/sdl2_audio_bitset
...
SDL2 - AudioAllowChangeFlags bit_set
2025-08-01 23:46:26 +02:00
Laytan
a7a6216e08
Merge pull request #5510 from taylormck/add-cgltf-enums
...
add cgltf filter type and wrap mode enums
2025-08-01 23:46:13 +02:00
Laytan
d343b93ef2
fix check
2025-08-01 20:31:53 +02:00
Laytan
20318e7a3b
concrete types to make llvm 14 happy
...
Fixes #5463
Fixes #5244
Fixes #5435
2025-08-01 19:27:28 +02:00
Laytan
d0890c2135
fix dropping part of particular vecs
2025-08-01 19:18:43 +02:00
gingerBill
3eb9c5fa65
Merge pull request #5525 from Barinzaya/xxh3-simd
...
`core:hash/xxhash`: Static SIMD Support for XXH3
2025-08-01 16:22:29 +01:00
gingerBill
ab866653a8
Merge pull request #5526 from laytan/fixes-for-vector-abi
...
amd64 abi fixes regarding vectors
2025-08-01 16:21:44 +01:00
Barinzaya
4ef7ed1cbd
Skip bounds checking on the inner accumulate loop.
...
This helps performance with SSE (somewhat) and AVX-512 (quite a bit),
but not AVX2 for some reason.
2025-07-31 16:51:42 -04:00
Laytan
5527982c20
fix vec offset calculation
2025-07-31 22:01:17 +02:00
Laytan
f32ee28e6d
amd64 support half in the abi too
2025-07-31 20:50:46 +02:00
Laytan
c3bae964d0
amd64 abi fixes regarding vectors
...
- Fixes the code so SSEUp is grouped/skipped over properly (Fixes #5429 )
- Fixes f16 vectors using garbage widths, because it would call
LLVMGetIntTypeWidth and an f16 is not an int so doesn't have that
function
2025-07-31 20:26:22 +02:00
Barinzaya
f61dc7d071
Remove favor_size attributes inhibiting SIMD optimizations.
...
This makes a tremendous (2x with SSE2, 3x with AVX2) difference on big
datasets on my system, but this may be hardware-dependent (e.g.
instruction cache sizes).
Naturally, this also results in somewhat larger code for the large-data
case (~75% larger).
2025-07-31 13:05:10 -04:00
Barinzaya
2f8b390c19
Various minor changes in XXH3.
...
This includes various minor things that didn't seem right or could be
improved, including:
- XXH3_state is documented to have a strict alignment requirement of 64
bytes, and thus came with a disclaimer not to use `new` because it
wouldn't be aligned correctly. It now has an `#align(64)` so that it
will.
- An _internal proc being marked #force_no_inline (every other one is
#force_inline)
- Unnecessarily casting the product of two u32s through u128 (and
ultimately truncating to u64 anyway)
2025-07-31 13:05:10 -04:00
Barinzaya
9d40f371be
Add static SIMD support to XXH3 in core:hash/xxhash.
...
This uses compile-time features to decide how large of a SIMD vector to
use. It currently has checks for amd64/i386 to size its vectors for
SSE2/AVX2/AVX512 as necessary.
The generalized SIMD functions could also be useful for multiversioning
of the hash procs, to allow for run-time dispatch based on available CPU
features.
2025-07-31 13:05:08 -04:00
Jeroen van Rijn
393e00bec3
Placate linker
2025-07-31 16:40:40 +02:00
Jeroen van Rijn
a41500de28
Merge pull request #5520 from Mecso2/master
...
fix amd64 no-crt entry assembly
2025-07-31 16:38:21 +02:00
gingerBill
077ba0b6cb
Merge pull request #5513 from colrdavidson/spall_v3
...
update to spall format v3
2025-07-31 12:37:09 +01:00
gingerBill
9b8223dd69
Remove use of .no_copy
2025-07-30 23:19:14 +01:00
gingerBill
6a10cfdc30
Fix typo
2025-07-30 23:15:14 +01:00
gingerBill
a2e7a62a26
Add virtual.new_clone
2025-07-30 23:14:53 +01:00
gingerBill
7057fc8dfc
Remove the semantics of #no_copy, keep the grammar
2025-07-30 23:14:29 +01:00
gingerBill
aa6a749804
Improve atomic logic for sync.Wait_Group
2025-07-30 23:11:18 +01:00
Colin Davidson
3e10684630
adjust scale with new format, fix segfault for auto-trace
2025-07-30 14:09:50 -07:00
Colin Davidson
7720a32120
Merge remote-tracking branch 'live/master' into spall_v3
2025-07-30 12:51:56 -07:00
Mecso2
e54bf24746
fix amd64 no-crt entry assembly
2025-07-30 12:44:53 +02:00
Laytan
939fed592d
Merge pull request #5517 from colrdavidson/freq_hotfix
...
fix x86 osx build
2025-07-30 06:22:25 +02:00
Colin Davidson
b5a8a3cf94
require to reduce files for a one line function
2025-07-29 13:58:30 -07:00
Colin Davidson
42f2891f35
attempt to fix x86 osx build
2025-07-29 13:54:56 -07:00
Colin Davidson
e869b9351b
Merge remote-tracking branch 'live/master' into spall_v3
2025-07-29 13:48:32 -07:00
Laytan
861fa4ab68
Merge pull request #4140 from colrdavidson/macharena
...
Mach Process control
2025-07-29 22:37:42 +02:00
Colin Davidson
35a95fc3e4
workaround to avoid excessive file-splitting
2025-07-29 13:28:04 -07:00
Colin Davidson
517e392abe
if -> when
2025-07-29 13:18:40 -07:00
Colin Davidson
4d9fdf5bd3
move to checker error, rather than panic
2025-07-29 13:09:05 -07:00
Colin Davidson
9a2e6fff48
switch to read_cycle_counter_frequency from arm64_read_cycle_counter_freq, trap on invalid arch
2025-07-29 12:21:46 -07:00
Colin Davidson
c1b3d035e4
remove spawn from os, comment sys/posix/spawn
2025-07-29 12:12:15 -07:00
Colin Davidson
0a102bd757
shuffle to Kern_Return
2025-07-29 11:52:58 -07:00
Colin Davidson
fe2b48dec4
enum pass
2025-07-29 11:48:58 -07:00
Laytan
435f7f6503
Merge pull request #5515 from laytan/ignored-arg-multi-return
...
fix ignored args in multi return handling
2025-07-29 19:16:32 +02:00
Laytan
e8a591386b
Merge pull request #5514 from laytan/saner-default-stack-linker-flags
...
saner stack linker flags for WASM
2025-07-29 19:16:19 +02:00
Laytan
af0b90bd3e
fix flag
2025-07-29 19:07:28 +02:00
Laytan
c966e63520
fix ignored args in multi return handling
...
Fixes #5483
2025-07-29 19:04:49 +02:00
Laytan
e7670e58dd
saner stack linker flags for WASM
...
Supersedes #5425
2025-07-29 18:50:27 +02:00
Taylor
6c2b72f701
fix: align the enum values
2025-07-30 00:17:40 +09:00
Colin Davidson
b5629aeed9
copy-paste typo fix
2025-07-28 16:34:53 -07:00
Colin Davidson
1a9b128bfc
update linux-arm to use the asm intrin for freq
2025-07-28 16:32:10 -07:00
Colin Davidson
e1fd69f573
oops, one more manual_header
2025-07-28 16:27:54 -07:00
