happyz
/
gemma.cpp
mirror of https://github.com/google/gemma.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
468 Commits 18 Branches 5 Tags 35 MiB
Commit Graph

190 Commits

Author SHA1 Message Date
Jan Wassenberg 96d2ab7d31 Minor fix to profiler zone and add comment 
PiperOrigin-RevId: 681350546
 2024-10-02 01:37:50 -07:00
Jan Wassenberg 7d9fcda0d8 -467ms startup: parallel Reshape 
Also split Softmax into Argmax helper, add comments;
add profiler zones + fix IDE warning

PiperOrigin-RevId: 680954573
 2024-10-01 04:11:35 -07:00
Jan Wassenberg 2d14d796e3 1.09x decode speedup for topk=1/temp0: fuse softmax and sample 
PiperOrigin-RevId: 680589099
 2024-09-30 08:37:41 -07:00
Jan Wassenberg 897f902d28 Fix include order, required to build with profiler enabled 
PiperOrigin-RevId: 680574177
 2024-09-30 07:52:50 -07:00
Daniel Keysers 606427022c Fix compiler errors when trying to generate (unused) code for the ConfigNoVit struct. 
PiperOrigin-RevId: 679049377
 2024-09-26 01:55:26 -07:00
RangerUFO d1010337c3 Fix prefix-LM mode assertion 2024-09-25 22:22:28 +08:00
Daniel Keysers f8835fe4a4 Add support for PaliGemma Vision-LM (224x224) to gemma.cpp 
See https://arxiv.org/abs/2407.07726 for a description of the model.
Because PaliGemma operates as a prefix-LM on the image+prompt, add support for that.

PiperOrigin-RevId: 677841119
 2024-09-23 10:09:38 -07:00
Jan Wassenberg 8c0a8834c1 Major compression update, arbitrary-len unpack + new Dot 
Compression:
* Implement {any packed} x {bf16, f32} 'Load2' and DecompressAndZeroPad
* New compression test for all packed formats, add to GEMMA_TEST_FILES, remove from sfp/nuq_test
* Decompress->DecompressAndZeroPad, use PackedSpan for args with bounds checking
* NUQ: support arbitrary-length enc/dec
* New compression/shared, remove sfp.h and nuq.h
* Move Store2 into Traits and provide Compress2 wrapper
* Remove unused Decompress()-with-pool overload
* Simplify CompressedArrayLen, rename to CompressedArrayElements
* Remove unused DistortionStats b_l1_

Misc:
* Add compensated and Kahan dot, support any length
* Use same Dot function everywhere
* Move exact arithmetic functions into fp_arith
* use FloatPtr and MatPtr typedefs in tests; less stack usage
* Rename args to packed/raw
* Remove Traits::Name, instead TypeName<T>()
* Move kMaxSFP and kClusters/kGroupSize into Sfp/NuqStream
PiperOrigin-RevId: 672868468
 2024-09-10 02:22:19 -07:00
Jan Wassenberg c29e9752c7 Refactor/cleanup, remove even_odd 
* New compression/shared.h, remove sfp.h
* Remove unused DistortionStats b_l1_
* Move exact arithmetic functions into fp_arith
* Remove even_odd optimization for MatVec (mostly unused)
* use BF16 typedef more widely
* Add kMaxSFP constant

PiperOrigin-RevId: 670996386
 2024-09-04 09:25:13 -07:00
Daniel Keysers a8e08778d4 Add an additional QueryModel() overload to GemmaEnv. 
Use args only in GemmaEnv constructor, store everything else in RuntimeConfig.
Add runtime option to turn off thread spinning.

PiperOrigin-RevId: 670467320
 2024-09-03 02:25:19 -07:00
Zoltan Szabadka f6abbab3a4 Fix asan failure in local attention computation. 
PiperOrigin-RevId: 670207380
 2024-09-02 07:06:10 -07:00
Jan Wassenberg 4033ed9e78 Avoid duplication of RMSNorm, support all activation/weight types 
Add test for RMSNorm
Rename VectorizedRopeAndMulBy -> RopeAndMulBy

Move test_util to util/

PiperOrigin-RevId: 668332927
 2024-08-28 01:26:55 -07:00
Daniel Keysers 18e6012872 Fix prefill for batched queries. 
This lets gemma_test/GeographyBatched pass now also for gemma2-27B.

PiperOrigin-RevId: 664827485
 2024-08-19 08:50:42 -07:00
Apoorv Reddy c6eb3b6f0d VectorizedRopeAndMulBy. 
~8x reduction (tested on few prompts) in Rope.
~3.8% prefill latency improvement.
~2.6% decode latency improvement.

PiperOrigin-RevId: 664650108
 2024-08-18 23:17:01 -07:00
Paul Chang 773333e5be Expose underlying model configuration: number of layers, heads, etc. 
PiperOrigin-RevId: 663747853
 2024-08-16 09:03:24 -07:00
Jan Wassenberg 301dc8067a Major MatMul update, 1.9-2.3x speedup on Zen4 via bf16 mul 
Supports converting all weight/activation formats to native MulT (bf16/f32)

Also:
- ConstMat/MutableMat for const correctness
- Move RowVectorBatch to allocator.h so it can be used from Matmul
- Add matmul.h so MatMulEnv can be used from Activations
- Remove kMaxThreads, detect from PerClusterPools
- Build fix: -inl.h files must be textual_hdrs, and highway.h should precede -inl.h

```
zen4 new
64, 24576, 3072, add=0, MatTA=bf16, MatTB=sfp:   616.6 GFLOPS.
64, 3072, 24576, add=0, MatTA=bf16, MatTB=sfp:   460.7 GFLOPS.
64, 24576, 3072, add=0, MatTA=f32, MatTB=sfp:    598.6 GFLOPS.
64, 3072, 24576, add=0, MatTA=f32, MatTB=sfp:    435.6 GFLOPS.

zen4 old
64, 24576, 3072, add=0, MatTA=f32, MatTB=sfp:    257.5 GFLOPS.
64, 3072, 24576, add=0, MatTA=f32, MatTB=sfp:    231.9 GFLOPS.
```

PiperOrigin-RevId: 663729812
 2024-08-16 07:52:20 -07:00
Paul Chang b9ed12a325 Support directly observing activations, partially replacing LayersOutputFunc 
LayersOutputFunc is no longer invoked for "blocks" and "final_norm" outputs.
Instead, we directly expose the Activations structure.

PiperOrigin-RevId: 663409316
 2024-08-15 12:39:07 -07:00
Jan Wassenberg 22995c699d Simplify pos handling, auto-increment output arg 
- no longer multiply by num_queries
- remove unused interleaved prompts
- Rename to Queries*
- Rename batch_start/interleaved_pos/pos to queries_pos

PiperOrigin-RevId: 663331823
 2024-08-15 09:25:26 -07:00
Copybara-Service 6763afcd1c Merge pull request #348 from ufownl:feature/start_pos_per_query_reopen 
PiperOrigin-RevId: 662533529
 2024-08-13 08:51:06 -07:00
RangerUFO 8c634f6486 Fix the position calculation issue in the generation phase 2024-08-12 18:50:23 +02:00
RangerUFO 730b6bfc94 Implement `start_pos` per query for batch interface 2024-08-12 18:50:23 +02:00
Jan Wassenberg b831fa8482 1.3x prefill, 0.95x decode: matmul replacing last matvec 
Before 38.28, 9.17 (with profiler enabled, prompt = 330 tok)
```
Gen.FFW                                 :      15414 x         4692352 = 24.166318
Gen.Attention.SumHeads                  :      15414 x         1394804 =  7.183451 !!
Gen.Embedding                           :        361 x        49961894 =  6.026297
Gen.Attention.QKV                       :      15414 x         1005125 =  5.176546
Gen.Attention.DotSoftmax                :      15414 x          885480 =  4.560357
RopeAndMulBy                            :     696528 x           11867 =  2.761818
```

After 49.80, 8.68
```
Gen.FFW                                 :      14448 x         5312783 = 25.646868
Gen.Embedding                           :        338 x        63044815 =  7.119845
Gen.Attention.QKV                       :      14448 x         1115003 =  5.382557
Gen.Attention.DotSoftmax                :      14448 x          897577 =  4.332957
RopeAndMulBy                            :     673344 x           11886 =  2.674156
Gen.Attention.SumHeads                  :      14448 x          518291 =  2.501993 !!
```
PiperOrigin-RevId: 662024085
 2024-08-12 03:36:01 -07:00
Jan Wassenberg 282f73ec2f Add pin flag to disable pinning. Refs #338 
PiperOrigin-RevId: 661389171
 2024-08-09 13:47:12 -07:00
Apoorv Reddy fd1b0743a7 Rename Gemma9B and Gemma27B to Gemma2_9B and Gemma2_27B. 
This is to make it clear that these models are part of the Gemma2 family of models.

PiperOrigin-RevId: 661181682
 2024-08-09 02:09:06 -07:00
Jan Wassenberg 2ebbe4076f 1.03-1.08x decode speedup: precompute Rope theta, fuse 
Split attention into functions, move into class.
Fuse Rope and MulBy, allow non-in-place version to avoid copy from q to KV.
Sink if() into MaybeLogitsSoftCap.

PiperOrigin-RevId: 661168418
 2024-08-09 01:23:24 -07:00
The gemma.cpp Authors 27258b03e6 Improve performance logging 
PiperOrigin-RevId: 660534330
 2024-08-07 14:15:43 -07:00
Jan Wassenberg 5e433e774a 1.1x prefill speedup, revamp threading in preparation for hierarchical parallelism. 
Limit thread counts to detected. Add max_clusters arg.
Update detection logic to check for smt0 - previously we pinned to some siblings.

PiperOrigin-RevId: 659755311
 2024-08-05 18:50:09 -07:00
Phil Culliton 1982a6ba00 Internal change 
PiperOrigin-RevId: 657831926
 2024-07-30 20:24:54 -07:00
Jan Wassenberg a24eda8d02 Split matmul into matvec; add large matrix benchmark 
Rename var names to row/col for more clarity.
Better estimate error tolerance via max abs col sum.

PiperOrigin-RevId: 657601791
 2024-07-30 08:29:11 -07:00
Paul Chang d37c088e44 Extend LayersOutputFunc to take query index and auxillary int 
PiperOrigin-RevId: 657574814
 2024-07-30 06:53:56 -07:00
Jan Wassenberg 8b4915f321 Fix Windows build - macro conflict with param name 
PiperOrigin-RevId: 657518587
 2024-07-30 03:22:32 -07:00
Jan Wassenberg 6ea4232b2e MatMul cleanup: Mat struct, simplify args. 
Add large benchmark to test, use 4 threads, skip some targets.
Also use Traits::Name instead of typeid.

PiperOrigin-RevId: 657496185
 2024-07-30 01:55:50 -07:00
Jan Wassenberg f27683152c 1.05x prefill speedup: matvec -> matmul for !MHA 
Also add C_stride and make shape normal non-template arguments.

PiperOrigin-RevId: 657285945
 2024-07-29 12:18:06 -07:00
Jan Wassenberg 2721f54446 Add offset arg to MatMul, rename, Matmul for logits = ~1.1x decode speedup 
PiperOrigin-RevId: 657167257
 2024-07-29 05:34:26 -07:00
Jan Wassenberg aaf51898b6 Major revamp #2 of Prefill: fix token order, parallel for multi-query 
- Allocate only the required KV caches and activation batch size
- Add flags for batch sizes
- Const-correct interface: Span of const int.
- Also clean up the KVCache arg to a span.
- Move kPrefillBatchSize into RuntimeConfig and remove related global constants.

PiperOrigin-RevId: 655893197
 2024-07-25 03:28:55 -07:00
Daniel Keysers 2346b5a434 Minor polishing: adding comments, renaming variables. 
PiperOrigin-RevId: 655235006
 2024-07-23 11:17:44 -07:00
Daniel Keysers 33334ad454 Fix msan uninitialized scale in optimize_test 
PiperOrigin-RevId: 654817460
 2024-07-22 10:50:25 -07:00
Jan Wassenberg 85cac13fb1 Split up ops.h into ops/ops-inl and matmul-inl 
PiperOrigin-RevId: 654068303
 2024-07-19 11:21:48 -07:00
Jan Wassenberg 5844e6a1e5 Cleanup: add wrapper functions and rename vars to interleaved 
Simplifies the TransformerLayer function.
Use interleaved* instead of _and_queries.

PiperOrigin-RevId: 653929449
 2024-07-19 02:04:11 -07:00
Jan Wassenberg 12016d31c3 Major Prefill/Generate cleanup, 1.3x Prefill speedup 
This fixes TTFT, which was not including prefill.

PiperOrigin-RevId: 653690626
 2024-07-18 11:16:46 -07:00
Jan Wassenberg 3fe79b3876 Fix msan uninitialized scale 
PiperOrigin-RevId: 653655471
 2024-07-18 09:42:31 -07:00
Daniel Keysers e87e65ca45 Add scale parameter to MatMul. 
Add accessor to CompressedArray that asserts the scale is 1 and use it.

PiperOrigin-RevId: 653604840
 2024-07-18 06:58:56 -07:00
Daniel Keysers 5a751a9a44 Update gemma-27b to the correct query scaling. 
PiperOrigin-RevId: 653201646
 2024-07-17 05:43:52 -07:00
Jan Wassenberg 992a2cbbc0 De-templatize Activations, add RowVectorBatch class 
Also remove most kBatchSize args.

PiperOrigin-RevId: 653185525
 2024-07-17 04:38:15 -07:00
Daniel Keysers ff34370aac Simplify FFW by using MatMul_4x4_Batch_Add. 
Affects only the griffin model, where prefill TPS improves by about 70%.

PiperOrigin-RevId: 652878176
 2024-07-16 09:41:23 -07:00
Jan Wassenberg cd530374b3 Further 1.02x prefill speedup from batch 64->512 
Measured on SKX. Larger speedup expected for Zen4/SPR.

PiperOrigin-RevId: 652472928
 2024-07-15 07:26:10 -07:00
The gemma.cpp Authors c879133a5a Increase the prefill batch size to 64. 
PiperOrigin-RevId: 651754772
 2024-07-12 06:28:37 -07:00
The gemma.cpp Authors df3fb70802 Improve readability with RepeatedAttentionWindowSizes 
PiperOrigin-RevId: 651431738
 2024-07-11 09:11:46 -07:00
Jan Wassenberg edaf61b983 SVE build fix: avoid capturing vectors directly. 
Also use more V typedef instead of auto.

PiperOrigin-RevId: 651423685
 2024-07-11 08:43:56 -07:00
Jan Wassenberg be765afce2 Simplify matmul: only 2 overloads 
Also add StoreHorizontalSumsMaybeAdd wrapper function,
move MatMulSlowBatch into test.

1.02-1.06x speedup.

PiperOrigin-RevId: 651394791
 2024-07-11 06:58:42 -07:00