danielkrause/colibri-strix
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

VSX integer-dot kernels for POWER8 (12.8x int8 / 7.6x int4 over scalar, #ifdef-gated, x86 path untouched) (#98)

Browse Source 
* Makefile: support Linux PowerPC (ppc64le) builds

PowerPC GCC uses -mcpu instead of -march, so the Linux branch failed
with unrecognized option -march=native on ppc64le. Detect ppc64le and
ppc64 via uname -m and use -mcpu=$(ARCH) there. The x86-64 path is
unchanged.

Validated on an IBM POWER8 S824 (Ubuntu 20.04, gcc 9.4): make test-c
passes, teacher forcing 32/32 positions and greedy 20/20 tokens against
the transformers oracle, engine reports the scalar idot fallback.

Signed-off-by: Scott <scottbphone12@gmail.com>

* VSX integer-dot kernels for POWER8 (12.8x int8, 7.6x int4 over scalar)

Adds a VSX path to dot_i8i8 and dot_i4i8 using vec_msum, which sums
byte products directly into s32 lanes, so the 16-bit saturation bound
of the AVX2 maddubs trick does not apply. abs(w) is built with a
modulo-subtract select instead of vec_abs so w=-128 wraps to 128
unsigned instead of saturating to 127. Nibble unpack uses
vec_mergeh/vec_mergel, which interleave like x86 unpacklo/unpackhi on
ppc64le (verified on hardware). g_i4s=1 on VSX since the f32 fallback
is plain scalar there: measured 5.5x for int4 IDOT at S=1.

Measured on an IBM POWER8 S824 (gcc 9.4, Ubuntu 20.04 ppc64le),
single thread, 1536x6144:
  dot_i8i8  1.48 -> 18.99 Gops/s (12.8x)
  dot_i4i8  2.33 -> 17.72 Gops/s (7.6x)
  S=1 int4 matmul path: 3.925 -> 0.505 ms/call (7.8x vs scalar build)

Adds tests/test_idot.c: exactness test of the compiled idot kernels
(any arch) against a plain-C reference, covering odd tails and the
w=-128 edge. Passes on avx512-vnni (x86) and vsx (POWER8). The tiny
oracle stays token-exact on the VSX build: TF 32/32, greedy 20/20.

Signed-off-by: Scott <scottbphone12@gmail.com>

---------

Signed-off-by: Scott <scottbphone12@gmail.com>
Co-authored-by: Scott <scottbphone12@gmail.com>
This commit is contained in:
AutoJanitor
2026-07-12 15:44:56 -05:00
committed by GitHub
parent 9bba681ae2
commit bc6bc9c250
3 changed files with 104 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
c/Makefile
+4 -1
View File
@@ -66,7 +66,7 @@ CUDA_ARCH ?= native
NVCCFLAGS ?= -O3 -std=c++17 -arch=$(CUDA_ARCH) -Xcompiler=-Wall,-Wextra
PYTHON ?= python3
CUDA_OBJ =
TEST_BINS = tests/test_json$(EXE) tests/test_st$(EXE) tests/test_tier$(EXE) tests/test_grammar$(EXE)
TEST_BINS = tests/test_json$(EXE) tests/test_st$(EXE) tests/test_tier$(EXE) tests/test_grammar$(EXE) tests/test_idot$(EXE)
ifeq ($(CUDA),1)
ifeq ($(UNAME_S),Darwin)
$(error CUDA=1 is supported only on Linux)
@@ -119,6 +119,9 @@ tests/test_tier$(EXE): tests/test_tier.c tier.h
tests/test_grammar$(EXE): tests/test_grammar.c grammar.h
$(CC) $(CFLAGS) $< -o $@ $(LDFLAGS)
tests/test_idot$(EXE): tests/test_idot.c glm.c st.h json.h tok.h tok_unicode.h compat.h grammar.h tier.h
$(CC) $(CFLAGS) $< -o $@ $(LDFLAGS)
test-c: $(TEST_BINS)
@for test in $(TEST_BINS); do ./$$test || exit 1; done