-
Notifications
You must be signed in to change notification settings - Fork 6.8k
[MXNET-1446] Quantization: intgemm matrix multiply wrappers #17559
Commits on Nov 28, 2019
-
Kenneth Heafield committed
Nov 28, 2019 Configuration menu - View commit details
-
Copy full SHA for 1320de3 - Browse repository at this point
Copy the full SHA 1320de3View commit details -
Update to remove DEFAULT macro
Kenneth Heafield committedNov 28, 2019 Configuration menu - View commit details
-
Copy full SHA for 0c68e33 - Browse repository at this point
Copy the full SHA 0c68e33View commit details -
Kenneth Heafield committed
Nov 28, 2019 Configuration menu - View commit details
-
Copy full SHA for 3bf28e5 - Browse repository at this point
Copy the full SHA 3bf28e5View commit details -
Kenneth Heafield committed
Nov 28, 2019 Configuration menu - View commit details
-
Copy full SHA for 5b01d0b - Browse repository at this point
Copy the full SHA 5b01d0bView commit details
Commits on Nov 29, 2019
-
Consolidate CPU inline into cc since there's only one dispatch
Kenneth Heafield committedNov 29, 2019 Configuration menu - View commit details
-
Copy full SHA for 0d7b54a - Browse repository at this point
Copy the full SHA 0d7b54aView commit details -
Kenneth Heafield committed
Nov 29, 2019 Configuration menu - View commit details
-
Copy full SHA for 88fb3a5 - Browse repository at this point
Copy the full SHA 88fb3a5View commit details
Commits on Dec 2, 2019
-
intgemm fully_connected operator
import mxnet as mx a = mx.nd.random_uniform(low=-1.0, high=1.0, shape=[5, 64]) b = mx.nd.random_uniform(low=-1.0, high=1.0, shape=[8, 64]) b_scale = 127.0 / mx.nd.contrib.intgemm_maxabsolute(b).asscalar() b_prepared = mx.nd.contrib.intgemm_prepareb(b, multiplier = b_scale) mx.nd.FullyConnected(a, b, num_hidden=8, no_bias=True, flatten=False) mx.nd.contrib.intgemm_fully_connected(a, b_prepared, out_float_multiplier=1.0/b_scale, num_hidden=8, no_bias=True, flatten=False)
Kenneth Heafield committedDec 2, 2019 Configuration menu - View commit details
-
Copy full SHA for 9e5d7d5 - Browse repository at this point
Copy the full SHA 9e5d7d5View commit details -
Kenneth Heafield committed
Dec 2, 2019 Configuration menu - View commit details
-
Copy full SHA for 897bf6e - Browse repository at this point
Copy the full SHA 897bf6eView commit details -
Kenneth Heafield committed
Dec 2, 2019 Configuration menu - View commit details
-
Copy full SHA for b65e33f - Browse repository at this point
Copy the full SHA b65e33fView commit details -
Extract common code from Prepare* operations
Kenneth Heafield committedDec 2, 2019 Configuration menu - View commit details
-
Copy full SHA for b615ee8 - Browse repository at this point
Copy the full SHA b615ee8View commit details
Commits on Dec 3, 2019
-
Disable in-place, zero gradients following existing quantization code
Kenneth Heafield committedDec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for ed6be7e - Browse repository at this point
Copy the full SHA ed6be7eView commit details -
Remove commented out parameter
Kenneth Heafield committedDec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for 153a628 - Browse repository at this point
Copy the full SHA 153a628View commit details -
Better documentation/parameter naming for intgemm fully connected
Kenneth Heafield committedDec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for f1cd4ab - Browse repository at this point
Copy the full SHA f1cd4abView commit details -
Rename preparea to prepare_data, prepareb to prepare_weight
Kenneth Heafield committedDec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for 8b5d107 - Browse repository at this point
Copy the full SHA 8b5d107View commit details -
Allow all request types for max_absolute
Kenneth Heafield committedDec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for 6e801f4 - Browse repository at this point
Copy the full SHA 6e801f4View commit details -
Kenneth Heafield committed
Dec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for f492f26 - Browse repository at this point
Copy the full SHA f492f26View commit details -
Add operator to slice a B matrix
Kenneth Heafield committedDec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for 7a02d05 - Browse repository at this point
Copy the full SHA 7a02d05View commit details -
Kenneth Heafield committed
Dec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for 947f911 - Browse repository at this point
Copy the full SHA 947f911View commit details -
Revert "Update intgemm with VNNI". It's not ready for compilers that …
…lack VNNI support yet. This reverts commit 947f911.
Kenneth Heafield committedDec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for b28c699 - Browse repository at this point
Copy the full SHA b28c699View commit details -
Remove op suffix on intgemm_take_weight
Kenneth Heafield committedDec 3, 2019 Configuration menu - View commit details
-
Copy full SHA for 8f7deb6 - Browse repository at this point
Copy the full SHA 8f7deb6View commit details
Commits on Dec 20, 2019
-
Merge branch 'master' of https://github.com/apache/incubator-mxnet in…
…to intgemm
Kenneth Heafield committedDec 20, 2019 Configuration menu - View commit details
-
Copy full SHA for 502bcf5 - Browse repository at this point
Copy the full SHA 502bcf5View commit details -
Kenneth Heafield committed
Dec 20, 2019 Configuration menu - View commit details
-
Copy full SHA for 63c1a3b - Browse repository at this point
Copy the full SHA 63c1a3bView commit details -
Merge remote-tracking branch 'origin/master' into intgemm
Kenneth Heafield committedDec 20, 2019 Configuration menu - View commit details
-
Copy full SHA for 409fe0e - Browse repository at this point
Copy the full SHA 409fe0eView commit details
Commits on Jan 16, 2020
-
Kenneth Heafield committed
Jan 16, 2020 Configuration menu - View commit details
-
Copy full SHA for d777fed - Browse repository at this point
Copy the full SHA d777fedView commit details -
Merge branch 'master' into intgemm
Kenneth Heafield committedJan 16, 2020 Configuration menu - View commit details
-
Copy full SHA for c55076c - Browse repository at this point
Copy the full SHA c55076cView commit details
Commits on Feb 3, 2020
-
Refactor prepare operations to take scaling as tensors. PrepareBQuant…
…izedTransposed. This will make it easier to store a consistent file on disk.
Kenneth Heafield committedFeb 3, 2020 Configuration menu - View commit details
-
Copy full SHA for c6b47a1 - Browse repository at this point
Copy the full SHA c6b47a1View commit details -
Kenneth Heafield committed
Feb 3, 2020 Configuration menu - View commit details
-
Copy full SHA for 07cf577 - Browse repository at this point
Copy the full SHA 07cf577View commit details -
Merge remote-tracking branch 'origin/master' into intgemm
Kenneth Heafield committedFeb 3, 2020 Configuration menu - View commit details
-
Copy full SHA for c7dab72 - Browse repository at this point
Copy the full SHA c7dab72View commit details
Commits on Feb 10, 2020
-
Merge remote-tracking branch 'origin' into intgemm
Kenneth Heafield committedFeb 10, 2020 Configuration menu - View commit details
-
Copy full SHA for 389d7e3 - Browse repository at this point
Copy the full SHA 389d7e3View commit details -
Makefile compilation for intgemm
Kenneth Heafield committedFeb 10, 2020 Configuration menu - View commit details
-
Copy full SHA for 6c1a388 - Browse repository at this point
Copy the full SHA 6c1a388View commit details -
Fix order of arguments to filter-out in Makefile
Kenneth Heafield committedFeb 10, 2020 Configuration menu - View commit details
-
Copy full SHA for 85c5afd - Browse repository at this point
Copy the full SHA 85c5afdView commit details -
Kenneth Heafield committed
Feb 10, 2020 Configuration menu - View commit details
-
Copy full SHA for a17ba65 - Browse repository at this point
Copy the full SHA a17ba65View commit details
Commits on Mar 3, 2020
-
Quantizer with arbitrarily many arguments and OpenMP support
Benchmark program: import mxnet as mx import time def time_procedure(shape, count, proc): data = mx.nd.random_uniform(shape=shape, low=-1.0, high = 1.0) mx.nd.waitall() begin = time.time() for i in range(0, count): proc(data) mx.nd.waitall() return (time.time() - begin) / count shapes = [(1,1), (128, 128), (256,256), (512, 512), (1024, 1024), (2048, 2048), (20971520,), (8, 4096), (4096, 8)] count = 1000 one = mx.nd.ones(shape=(1)) minusone = -one procedures = { "quantize" : (lambda data : mx.nd.contrib.quantize(data, minusone, one)), "quantize_v2" : (lambda data : mx.nd.contrib.quantize_v2(data, min_calib_range = -1.0, max_calib_range = 1.0)), "intgemm" : (lambda data : mx.nd.contrib.intgemm_prepare_data(data, one)), "quantize_v2_fit" : (lambda data : mx.nd.contrib.quantize_v2(data)), "intgemm_fit" : (lambda data : mx.nd.contrib.intgemm_prepare_data(data, mx.nd.contrib.intgemm_maxabsolute(data))), } for s in shapes: print("Shape " + str(s)) stats = {} for name, l in procedures.items(): stats[name] = time_procedure(s, count, l) print("{:.7f} seconds for {}".format(stats[name], name)) best_baseline = min(stats["quantize"], stats["quantize_v2"]) ratio = best_baseline / stats["intgemm"] print("intgemm is {:.1f}x faster with calibration".format(ratio)) fit_ratio = stats["quantize_v2_fit"] / stats["intgemm_fit"] print("intgemm is {:.1f}x faster without calibration".format(fit_ratio)) OMP_NUM_THREADS=24 ./quant_bench.py Shape (1, 1) 0.0001304 seconds for quantize 0.0001076 seconds for quantize_v2 0.0000310 seconds for intgemm 0.0001114 seconds for quantize_v2_fit 0.0000479 seconds for intgemm_fit intgemm is 3.5x faster with calibration intgemm is 2.3x faster without calibration Shape (128, 128) 0.0001649 seconds for quantize 0.0001399 seconds for quantize_v2 0.0000329 seconds for intgemm 0.0001533 seconds for quantize_v2_fit 0.0000502 seconds for intgemm_fit intgemm is 4.2x faster with calibration intgemm is 3.1x faster without calibration Shape (256, 256) 0.0001660 seconds for quantize 0.0001404 seconds for quantize_v2 0.0000335 seconds for intgemm 0.0001599 seconds for quantize_v2_fit 0.0000505 seconds for intgemm_fit intgemm is 4.2x faster with calibration intgemm is 3.2x faster without calibration Shape (512, 512) 0.0001691 seconds for quantize 0.0001434 seconds for quantize_v2 0.0000342 seconds for intgemm 0.0001813 seconds for quantize_v2_fit 0.0000540 seconds for intgemm_fit intgemm is 4.2x faster with calibration intgemm is 3.4x faster without calibration Shape (1024, 1024) 0.0001920 seconds for quantize 0.0001538 seconds for quantize_v2 0.0000511 seconds for intgemm 0.0002390 seconds for quantize_v2_fit 0.0000827 seconds for intgemm_fit intgemm is 3.0x faster with calibration intgemm is 2.9x faster without calibration Shape (2048, 2048) 0.0002364 seconds for quantize 0.0001989 seconds for quantize_v2 0.0000875 seconds for intgemm 0.0004747 seconds for quantize_v2_fit 0.0001531 seconds for intgemm_fit intgemm is 2.3x faster with calibration intgemm is 3.1x faster without calibration Shape (20971520,) 0.0011446 seconds for quantize 0.0010902 seconds for quantize_v2 0.0008950 seconds for intgemm 0.0023337 seconds for quantize_v2_fit 0.0015005 seconds for intgemm_fit intgemm is 1.2x faster with calibration intgemm is 1.6x faster without calibration Shape (8, 4096) 0.0001636 seconds for quantize 0.0001392 seconds for quantize_v2 0.0000364 seconds for intgemm 0.0001508 seconds for quantize_v2_fit 0.0000651 seconds for intgemm_fit intgemm is 3.8x faster with calibration intgemm is 2.3x faster without calibration Shape (4096, 8) 0.0001642 seconds for quantize 0.0001392 seconds for quantize_v2 0.0000370 seconds for intgemm 0.0001515 seconds for quantize_v2_fit 0.0000654 seconds for intgemm_fit intgemm is 3.8x faster with calibration intgemm is 2.3x faster without calibration OMP_NUM_THREADS=1 ./quant_bench.py Shape (1, 1) 0.0000630 seconds for quantize 0.0000706 seconds for quantize_v2 0.0000294 seconds for intgemm 0.0000632 seconds for quantize_v2_fit 0.0000475 seconds for intgemm_fit intgemm is 2.1x faster with calibration intgemm is 1.3x faster without calibration Shape (128, 128) 0.0000860 seconds for quantize 0.0000898 seconds for quantize_v2 0.0000324 seconds for intgemm 0.0000996 seconds for quantize_v2_fit 0.0000464 seconds for intgemm_fit intgemm is 2.6x faster with calibration intgemm is 2.1x faster without calibration Shape (256, 256) 0.0000976 seconds for quantize 0.0001028 seconds for quantize_v2 0.0000339 seconds for intgemm 0.0001513 seconds for quantize_v2_fit 0.0000521 seconds for intgemm_fit intgemm is 2.9x faster with calibration intgemm is 2.9x faster without calibration Shape (512, 512) 0.0001724 seconds for quantize 0.0001693 seconds for quantize_v2 0.0000839 seconds for intgemm 0.0004351 seconds for quantize_v2_fit 0.0001420 seconds for intgemm_fit intgemm is 2.0x faster with calibration intgemm is 3.1x faster without calibration Shape (1024, 1024) 0.0003559 seconds for quantize 0.0003481 seconds for quantize_v2 0.0002384 seconds for intgemm 0.0013803 seconds for quantize_v2_fit 0.0004667 seconds for intgemm_fit intgemm is 1.5x faster with calibration intgemm is 3.0x faster without calibration Shape (2048, 2048) 0.0011425 seconds for quantize 0.0010880 seconds for quantize_v2 0.0008497 seconds for intgemm 0.0051828 seconds for quantize_v2_fit 0.0018427 seconds for intgemm_fit intgemm is 1.3x faster with calibration intgemm is 2.8x faster without calibration Shape (20971520,) 0.0101917 seconds for quantize 0.0096956 seconds for quantize_v2 0.0071391 seconds for intgemm 0.0305159 seconds for quantize_v2_fit 0.0140535 seconds for intgemm_fit intgemm is 1.4x faster with calibration intgemm is 2.2x faster without calibration Shape (8, 4096) 0.0000880 seconds for quantize 0.0000950 seconds for quantize_v2 0.0000334 seconds for intgemm 0.0001183 seconds for quantize_v2_fit 0.0000423 seconds for intgemm_fit intgemm is 2.6x faster with calibration intgemm is 2.8x faster without calibration Shape (4096, 8) 0.0000900 seconds for quantize 0.0000949 seconds for quantize_v2 0.0000332 seconds for intgemm 0.0001215 seconds for quantize_v2_fit 0.0000433 seconds for intgemm_fit intgemm is 2.7x faster with calibration intgemm is 2.8x faster without calibration
Kenneth Heafield committedMar 3, 2020 Configuration menu - View commit details
-
Copy full SHA for 2e6bf75 - Browse repository at this point
Copy the full SHA 2e6bf75View commit details -
Update intgemm with less warnings
Kenneth Heafield committedMar 3, 2020 Configuration menu - View commit details
-
Copy full SHA for 804d78c - Browse repository at this point
Copy the full SHA 804d78cView commit details
Commits on Mar 17, 2020
-
Updated intgemm, should fix compiler issues.
Kenneth Heafield committedMar 17, 2020 Configuration menu - View commit details
-
Copy full SHA for 792bf72 - Browse repository at this point
Copy the full SHA 792bf72View commit details -
Kenneth Heafield committed
Mar 17, 2020 Configuration menu - View commit details
-
Copy full SHA for edc00f6 - Browse repository at this point
Copy the full SHA edc00f6View commit details -
gcc < 5 is a lost cause for intrinsics.
Kenneth Heafield committedMar 17, 2020 Configuration menu - View commit details
-
Copy full SHA for 5f3dc65 - Browse repository at this point
Copy the full SHA 5f3dc65View commit details -
Exclude intgemm operators when compiling with -DUSE_INTGEMM=OFF
Kenneth Heafield committedMar 17, 2020 Configuration menu - View commit details
-
Copy full SHA for a9c26db - Browse repository at this point
Copy the full SHA a9c26dbView commit details
Commits on Apr 6, 2020
-
intgemm with OMP support for multiply
Kenneth Heafield committedApr 6, 2020 Configuration menu - View commit details
-
Copy full SHA for a438d3d - Browse repository at this point
Copy the full SHA a438d3dView commit details
Commits on Apr 13, 2020
-
Kenneth Heafield committed
Apr 13, 2020 Configuration menu - View commit details
-
Copy full SHA for f04e70a - Browse repository at this point
Copy the full SHA f04e70aView commit details
Commits on Apr 20, 2020
-
Properly allocate temporary space for quantized A
Kenneth Heafield committedApr 20, 2020 Configuration menu - View commit details
-
Copy full SHA for b02dbc3 - Browse repository at this point
Copy the full SHA b02dbc3View commit details
Commits on May 25, 2020
-
Fix compile test path for avx512bw
Kenneth Heafield committedMay 25, 2020 Configuration menu - View commit details
-
Copy full SHA for d7cda47 - Browse repository at this point
Copy the full SHA d7cda47View commit details -
Kenneth Heafield committed
May 25, 2020 Configuration menu - View commit details
-
Copy full SHA for 7a20b91 - Browse repository at this point
Copy the full SHA 7a20b91View commit details
Commits on Jun 8, 2020
-
Merge branch 'master' into intgemm
Kenneth Heafield committedJun 8, 2020 Configuration menu - View commit details
-
Copy full SHA for 3faebb5 - Browse repository at this point
Copy the full SHA 3faebb5View commit details -
Kenneth Heafield committed
Jun 8, 2020 Configuration menu - View commit details
-
Copy full SHA for acf325d - Browse repository at this point
Copy the full SHA acf325dView commit details
Commits on Jul 13, 2020
-
Merge branch 'master' into intgemm
Kenneth Heafield committedJul 13, 2020 Configuration menu - View commit details
-
Copy full SHA for 952965f - Browse repository at this point
Copy the full SHA 952965fView commit details
Commits on Jul 20, 2020
-
Merge branch 'master' of https://github.com/apache/incubator-mxnet in…
…to intgemm
Kenneth Heafield committedJul 20, 2020 Configuration menu - View commit details
-
Copy full SHA for 2006d03 - Browse repository at this point
Copy the full SHA 2006d03View commit details
Commits on Jul 22, 2020
-
Update intgemm including -mno-avx fix
Kenneth Heafield committedJul 22, 2020 Configuration menu - View commit details
-
Copy full SHA for 91c729d - Browse repository at this point
Copy the full SHA 91c729dView commit details -
Merge https://github.com/apache/incubator-mxnet into intgemm
Kenneth Heafield committedJul 22, 2020 Configuration menu - View commit details
-
Copy full SHA for 528b77d - Browse repository at this point
Copy the full SHA 528b77dView commit details
Commits on Jul 24, 2020
-
Merge branch 'master' of https://github.com/apache/incubator-mxnet in…
…to intgemm
Kenneth Heafield committedJul 24, 2020 Configuration menu - View commit details
-
Copy full SHA for 3a1dfb6 - Browse repository at this point
Copy the full SHA 3a1dfb6View commit details
Commits on Aug 3, 2020
-
Merge branch 'master' into intgemm
Kenneth Heafield committedAug 3, 2020 Configuration menu - View commit details
-
Copy full SHA for 83db9fd - Browse repository at this point
Copy the full SHA 83db9fdView commit details -
Align to 64 bytes for intgemm too
Kenneth Heafield committedAug 3, 2020 Configuration menu - View commit details
-
Copy full SHA for 7ab838e - Browse repository at this point
Copy the full SHA 7ab838eView commit details -
Use intgemm MaxAbsolute's OMP support
Kenneth Heafield committedAug 3, 2020 Configuration menu - View commit details
-
Copy full SHA for b1a9725 - Browse repository at this point
Copy the full SHA b1a9725View commit details
Commits on Aug 10, 2020
-
Update intgemm including MSVC support
Kenneth Heafield committedAug 10, 2020 Configuration menu - View commit details
-
Copy full SHA for 784889e - Browse repository at this point
Copy the full SHA 784889eView commit details -
Merge branch 'master' into intgemm
Kenneth Heafield committedAug 10, 2020 Configuration menu - View commit details
-
Copy full SHA for 42e6bdf - Browse repository at this point
Copy the full SHA 42e6bdfView commit details -
More checks for 64-byte alignment
Kenneth Heafield committedAug 10, 2020 Configuration menu - View commit details
-
Copy full SHA for ea3bb9d - Browse repository at this point
Copy the full SHA ea3bb9dView commit details -
Don't take a scaling factor for int32 output
Kenneth Heafield committedAug 10, 2020 Configuration menu - View commit details
-
Copy full SHA for 7618fb1 - Browse repository at this point
Copy the full SHA 7618fb1View commit details -
Note these currently fail due to 64-byte alignment being broken in master apache#18854
Kenneth Heafield committedAug 10, 2020 Configuration menu - View commit details
-
Copy full SHA for 9fa5fff - Browse repository at this point
Copy the full SHA 9fa5fffView commit details -
Kenneth Heafield committed
Aug 10, 2020 Configuration menu - View commit details
-
Copy full SHA for 9209105 - Browse repository at this point
Copy the full SHA 9209105View commit details
Commits on Aug 17, 2020
-
Merge branch 'master' into intgemm
Kenneth Heafield committedAug 17, 2020 Configuration menu - View commit details
-
Copy full SHA for 26bd2dd - Browse repository at this point
Copy the full SHA 26bd2ddView commit details -
Update intgemm to remove MSVC warnings
Kenneth Heafield committedAug 17, 2020 Configuration menu - View commit details
-
Copy full SHA for a3fa6a0 - Browse repository at this point
Copy the full SHA a3fa6a0View commit details -
Also allow intgemm without MKLDNN to have 64-byte alignment
Kenneth Heafield committedAug 17, 2020 Configuration menu - View commit details
-
Copy full SHA for c0d93db - Browse repository at this point
Copy the full SHA c0d93dbView commit details -
Kenneth Heafield committed
Aug 17, 2020 Configuration menu - View commit details
-
Copy full SHA for 5d6279a - Browse repository at this point
Copy the full SHA 5d6279aView commit details -
Kenneth Heafield committed
Aug 17, 2020 Configuration menu - View commit details
-
Copy full SHA for 3842401 - Browse repository at this point
Copy the full SHA 3842401View commit details -
Slight fix for compilers without AVX512BW support
Kenneth Heafield committedAug 17, 2020 Configuration menu - View commit details
-
Copy full SHA for de3c19d - Browse repository at this point
Copy the full SHA de3c19dView commit details -
Fix flaky test whereby 0.5 could round either way
Kenneth Heafield committedAug 17, 2020 Configuration menu - View commit details
-
Copy full SHA for 98588da - Browse repository at this point
Copy the full SHA 98588daView commit details
Commits on Aug 24, 2020
-
Merge https://github.com/apache/incubator-mxnet into intgemm
Kenneth Heafield committedAug 24, 2020 Configuration menu - View commit details
-
Copy full SHA for 1952b9f - Browse repository at this point
Copy the full SHA 1952b9fView commit details -
Kenneth Heafield committed
Aug 24, 2020 Configuration menu - View commit details
-
Copy full SHA for a436cbd - Browse repository at this point
Copy the full SHA a436cbdView commit details -
Update tests to support numpy, refactor to pytest.mark.parametrize
Kenneth Heafield committedAug 24, 2020 Configuration menu - View commit details
-
Copy full SHA for 8e6739b - Browse repository at this point
Copy the full SHA 8e6739bView commit details -
Kenneth Heafield committed
Aug 24, 2020 Configuration menu - View commit details
-
Copy full SHA for 29cc970 - Browse repository at this point
Copy the full SHA 29cc970View commit details
Commits on Aug 28, 2020
-
Merge branch 'master' into intgemm
Kenneth Heafield committedAug 28, 2020 Configuration menu - View commit details
-
Copy full SHA for 524b79a - Browse repository at this point
Copy the full SHA 524b79aView commit details -
gcc7 is already required. You don't need any special handling here.
Kenneth Heafield committedAug 28, 2020 Configuration menu - View commit details
-
Copy full SHA for d03342a - Browse repository at this point
Copy the full SHA d03342aView commit details -
Kenneth Heafield committed
Aug 28, 2020 Configuration menu - View commit details
-
Copy full SHA for d7a8ef4 - Browse repository at this point
Copy the full SHA d7a8ef4View commit details -
Kenneth Heafield committed
Aug 28, 2020 Configuration menu - View commit details
-
Copy full SHA for a5a441e - Browse repository at this point
Copy the full SHA a5a441eView commit details -
Change intgemm.cc to linked library
Kenneth Heafield committedAug 28, 2020 Configuration menu - View commit details
-
Copy full SHA for 33ad782 - Browse repository at this point
Copy the full SHA 33ad782View commit details -
Use target_link_libraries to pick up intgemm compilation test header
Kenneth Heafield committedAug 28, 2020 Configuration menu - View commit details
-
Copy full SHA for 03732c7 - Browse repository at this point
Copy the full SHA 03732c7View commit details -
Change to a cmake_dependent_option
Kenneth Heafield committedAug 28, 2020 Configuration menu - View commit details
-
Copy full SHA for 8aaa23c - Browse repository at this point
Copy the full SHA 8aaa23cView commit details -
Revert "Change to downloaded intgemm" and remove header reference fro…
…m CMakeLists.txt This reverts commit a5a441e.
Kenneth Heafield committedAug 28, 2020 Configuration menu - View commit details
-
Copy full SHA for 8ac7fe6 - Browse repository at this point
Copy the full SHA 8ac7fe6View commit details
Commits on Aug 31, 2020
-
Change to #include <intgemm/intgemm.h>
Kenneth Heafield committedAug 31, 2020 Configuration menu - View commit details
-
Copy full SHA for e6ddba8 - Browse repository at this point
Copy the full SHA e6ddba8View commit details -
Merge branch 'master' into intgemm
Kenneth Heafield committedAug 31, 2020 Configuration menu - View commit details
-
Copy full SHA for 4578c8d - Browse repository at this point
Copy the full SHA 4578c8dView commit details -
Kenneth Heafield committed
Aug 31, 2020 Configuration menu - View commit details
-
Copy full SHA for 7e7b0c2 - Browse repository at this point
Copy the full SHA 7e7b0c2View commit details