eigen/bench/perf_monitoring/changesets.txt

#3.0.1
#3.1.1
#3.2.0
3.2.4
#5745:37f59e65eb6c
5891:d8652709345d  # introduce AVX
#5893:24b4dc92c6d3  # merge
5895:997c2ef9fc8b  # introduce FMA
#5904:e1eafd14eaa1  # complex and AVX
5908:f8ee3c721251  # improve packing with ptranspose
#5921:ca808bb456b0  # merge
#5927:8b1001f9e3ac
5937:5a4ca1ad8c53  # New gebp kernel handling up to 3 packets x 4 register-level blocks
#5949:f3488f4e45b2  # merge
#5969:e09031dccfd9  # Disable 3pX4 kernel on Altivec
#5992:4a429f5e0483  # merge
before-evaluators
#6334:f6a45e5b8b7c  # Implement evaluator for sparse outer products
#6639:c9121c60b5c7
#6655:06f163b5221f  # Properly detect FMA support on ARM
#6677:700e023044e7   # FMA has been wrongly disabled
#6681:11d31dafb0e3
#6699:5e6e8e10aad1   # merge default to tensors
#6726:ff2d2388e7b9   # merge default to tensors
#6742:0cbd6195e829   # merge default to tensors
#6747:853d2bafeb8f   # Generalized the gebp apis
6765:71584fd55762   # Made the blocking computation aware of the l3 cache; Also optimized the blocking parameters to take into account the number of threads used for a computation
6781:9cc5a931b2c6   # generalized gemv
6792:f6e1daab600a   # ensured that contractions that can be reduced to a matrix vector product
#6844:039efd86b75c   # merge tensor
6845:7333ed40c6ef   # change prefetching in gebp
#6856:b5be5e10eb7f   # merge index conversion
6893:c3a64aba7c70   # clean blocking size computation
6899:877facace746   # rotating kernel for ARM only
#6904:c250623ae9fa   # result_of
6921:915f1b1fc158   # fix prefetching change for ARM
6923:9ff25f6dacc6   # prefetching
6933:52572e60b5d3   # blocking size strategy
6937:c8c042f286b2   # avoid redundant pack_rhs
6981:7e5d6f78da59   # dynamic loop swapping
6984:45f26866c091   # rm dynamic loop swapping, adjust lhs's micro panel height to fully exploit L1 cache
6986:a675d05b6f8f   # blocking heuristic: block on the rhs in L1 if the lhs fit in L1.
7013:f875e75f07e5   # organize a little our default cache sizes, and use a saner default L1 outside of x86 (10% faster on Nexus 5)
7015:8aad8f35c955   # Refactor computeProductBlockingSizes to make room for the possibility of using lookup tables
7016:a58d253e8c91   # Polish lookup tables generation
7018:9b27294a8186   # actual_panel_rows computation should always be resilient to parameters not consistent with the known L1 cache size, see comment
7019:c758b1e2c073   # Provide a empirical lookup table for blocking sizes measured on a Nexus 5. Only for float, only for Android on ARM 32bit for now.
7085:627e039fba68   # Bug 986: add support for coefficient-based product with 0 depth.
7098:b6f1db9cf9ec   # Bug 992: don't select a 3p GEMM path with non-vectorizable scalar types, this hits unsupported paths in symm/triangular products code
7591:09a8e2186610   # 3.3-alpha1
7650:b0f3c8f43025   # help clang inlining
7708:dfc6ab9d9458   # Improve numerical accuracy in LLT and triangular solve by using true scalar divisions (instead of x * (1/y))
#8744:74b789ada92a   # Improved the matrix multiplication blocking in the case where mr is not a power of 2 (e.g on Haswell CPUs)
8789:efcb912e4356   # Made the index type a template parameter to evaluateProductBlockingSizes. Use numext::mini and numext::maxi instead of std::min/std::max to compute blocking sizes
8972:81d53c711775   # Don't optimize the processing of the last rows of a matrix matrix product in cases that violate the assumptions made by the optimized code path
8985:d935df21a082   # Remove the rotating kernel.
8988:6c2dc56e73b3   # Bug 256: enable vectorization with unaligned loads/stores.
9148:b8b8c421e36c   # Relax mixing-type constraints for binary coefficient-wise operators
9174:d228bc282ac9   # merge
9175:abc7a3600098   # Include the cost of stores in unrolling
9212:c90098affa7b   # Fix performance regression introduced in changeset 8aad8f35c955
9213:9f1c14e4694b   # Fix performance regression in dgemm introduced by changeset 81d53c711775
9361:69d418c06999   # 3.3-beta2
9445:f27ff0ad77a3   # Optimize expression matching "d?=a-b*c" as "d?=a; d?=b*c;"
9583:bef509908b9d   # 3.3-rc1
9593:2f24280cf59a   # Bug 1311: fix alignment logic in some cases of (scalar*small).lazyProduct(small)
9722:040d861b88b5   # Disabled part of the matrix matrix peeling code that's incompatible with 512 bit registers
9792:26667be4f70b   # 3.3.0
9891:41260bdfc23b   # Fix a performance regression in (mat*mat)*vec for which mat*mat was evaluated multiple times.
9942:b1d3eba60130   # Operators += and -= do not resize!
9943:79bb9887afd4   # Ease compiler job to generate clean and efficient code in mat*vec
9946:2213991340ea   # Complete rewrite of column-major-matrix * vector product to deliver higher performance of modern CPU.
9955:630471c3298c   # Improve performance of row-major-dense-matrix * vector products for recent CPUs. (this is the next changeset fixing a typo)
9975:2eeed9de710c   # Revert vec/y to vec*(1/y) in row-major TRSM
10442:e3f17da72a40  # Bug 1435: fix aliasing issue in exressions like: A = C - B*A;
10735:6913f0cf7d06  # Adds missing EIGEN_STRONG_INLINE to support MSVC properly inlining small vector calculations
10943:4db388d946bd  # Bug 1562: optimize evaluation of small products of the form s*A*B by rewriting them as: s*(A.lazyProduct(B)) to save a costly temporary. Measured speedup from 2x to 5x.
10961:5007ff66c9f6  # Introduce the macro ei_declare_local_nested_eval to help allocating on the stack local temporaries via alloca, and let outer-products makes a good use of it.
11083:30a528a984bb  # Bug 1578: Improve prefetching in matrix multiplication on MIPS.
11533:71609c41e9f8  # PR 526: Speed up multiplication of small, dynamically sized matrices
11535:6d348dc9b092  # Vectorize row-by-row gebp loop iterations on 16 packets as well
11568:efda481cbd7a  # Bug 1624: improve matrix-matrix product on ARM 64, 20% speedup
11596:b8d3f548a9d9  # do not read buffers out of bounds
11628:22f9cc0079bd  # Implement AVX512 vectorization of std::complex<float/double>
11638:81172653b67b  # Bug 1515: disable gebp's 3pX4 micro kernel for MSVC<=19.14 because of register spilling.
11659:b500fef42ced  # Artificially increase l1-blocking size for AVX512. +10% speedup with current kernels.
Update gemm performance monitoring tool: - permit to recompute a subset of changesets - update changeset list - add a few more cases 2015-03-06 18:47:13 +08:00			`#3.0.1`
			`#3.1.1`
			`#3.2.0`
Initial version of a small script to help tracking performance regressions 2015-02-21 02:20:34 +08:00			`3.2.4`
Update gemm performance monitoring tool: - permit to recompute a subset of changesets - update changeset list - add a few more cases 2015-03-06 18:47:13 +08:00			`#5745:37f59e65eb6c`
			`5891:d8652709345d # introduce AVX`
			`#5893:24b4dc92c6d3 # merge`
			`5895:997c2ef9fc8b # introduce FMA`
			`#5904:e1eafd14eaa1 # complex and AVX`
			`5908:f8ee3c721251 # improve packing with ptranspose`
			`#5921:ca808bb456b0 # merge`
			`#5927:8b1001f9e3ac`
			`5937:5a4ca1ad8c53 # New gebp kernel handling up to 3 packets x 4 register-level blocks`
			`#5949:f3488f4e45b2 # merge`
			`#5969:e09031dccfd9 # Disable 3pX4 kernel on Altivec`
			`#5992:4a429f5e0483 # merge`
Initial version of a small script to help tracking performance regressions 2015-02-21 02:20:34 +08:00			`before-evaluators`
Update gemm performance monitoring tool: - permit to recompute a subset of changesets - update changeset list - add a few more cases 2015-03-06 18:47:13 +08:00			`#6334:f6a45e5b8b7c # Implement evaluator for sparse outer products`
			`#6639:c9121c60b5c7`
			`#6655:06f163b5221f # Properly detect FMA support on ARM`
			`#6677:700e023044e7 # FMA has been wrongly disabled`
			`#6681:11d31dafb0e3`
			`#6699:5e6e8e10aad1 # merge default to tensors`
			`#6726:ff2d2388e7b9 # merge default to tensors`
			`#6742:0cbd6195e829 # merge default to tensors`
			`#6747:853d2bafeb8f # Generalized the gebp apis`
Update changeset list to be checked by perf_monitoring/gemm. 2015-02-26 23:13:33 +08:00			`6765:71584fd55762 # Made the blocking computation aware of the l3 cache; Also optimized the blocking parameters to take into account the number of threads used for a computation`
extend monitoring benchmarks with transpose matrix-vector and triangular matrix-vectors. 2016-12-05 20:36:26 +08:00			`6781:9cc5a931b2c6 # generalized gemv`
			`6792:f6e1daab600a # ensured that contractions that can be reduced to a matrix vector product`
Update gemm performance monitoring tool: - permit to recompute a subset of changesets - update changeset list - add a few more cases 2015-03-06 18:47:13 +08:00			`#6844:039efd86b75c # merge tensor`
Update changeset list to be checked by perf_monitoring/gemm. 2015-02-26 23:13:33 +08:00			`6845:7333ed40c6ef # change prefetching in gebp`
Update gemm performance monitoring tool: - permit to recompute a subset of changesets - update changeset list - add a few more cases 2015-03-06 18:47:13 +08:00			`#6856:b5be5e10eb7f # merge index conversion`
extend monitoring benchmarks with transpose matrix-vector and triangular matrix-vectors. 2016-12-05 20:36:26 +08:00			`6893:c3a64aba7c70 # clean blocking size computation`
Update changeset list to be checked by perf_monitoring/gemm. 2015-02-26 23:13:33 +08:00			`6899:877facace746 # rotating kernel for ARM only`
Update gemm performance monitoring tool: - permit to recompute a subset of changesets - update changeset list - add a few more cases 2015-03-06 18:47:13 +08:00			`#6904:c250623ae9fa # result_of`
Update changeset list to be checked by perf_monitoring/gemm. 2015-02-26 23:13:33 +08:00			`6921:915f1b1fc158 # fix prefetching change for ARM`
			`6923:9ff25f6dacc6 # prefetching`
Update gemm performance monitoring tool: - permit to recompute a subset of changesets - update changeset list - add a few more cases 2015-03-06 18:47:13 +08:00			`6933:52572e60b5d3 # blocking size strategy`
			`6937:c8c042f286b2 # avoid redundant pack_rhs`
			`6981:7e5d6f78da59 # dynamic loop swapping`
			`6984:45f26866c091 # rm dynamic loop swapping, adjust lhs's micro panel height to fully exploit L1 cache`
update gemm changeset list 2015-03-06 22:08:20 +08:00			`6986:a675d05b6f8f # blocking heuristic: block on the rhs in L1 if the lhs fit in L1.`
update perf_monitoring/gemm/changesets.txt 2015-03-14 05:57:05 +08:00			`7013:f875e75f07e5 # organize a little our default cache sizes, and use a saner default L1 outside of x86 (10% faster on Nexus 5)`
Update relevent list of changesets. 2016-07-04 20:32:34 +08:00			`7015:8aad8f35c955 # Refactor computeProductBlockingSizes to make room for the possibility of using lookup tables`
			`7016:a58d253e8c91 # Polish lookup tables generation`
			`7018:9b27294a8186 # actual_panel_rows computation should always be resilient to parameters not consistent with the known L1 cache size, see comment`
			`7019:c758b1e2c073 # Provide a empirical lookup table for blocking sizes measured on a Nexus 5. Only for float, only for Android on ARM 32bit for now.`
			`7085:627e039fba68 # Bug 986: add support for coefficient-based product with 0 depth.`
			`7098:b6f1db9cf9ec # Bug 992: don't select a 3p GEMM path with non-vectorizable scalar types, this hits unsupported paths in symm/triangular products code`
Unify gemm and lazy_gemm benchmarks 2015-10-07 22:06:48 +08:00			`7591:09a8e2186610 # 3.3-alpha1`
			`7650:b0f3c8f43025 # help clang inlining`
Update monitored changeset lists 2016-12-06 22:07:46 +08:00			`7708:dfc6ab9d9458 # Improve numerical accuracy in LLT and triangular solve by using true scalar divisions (instead of x * (1/y))`
Add performance monitoring for LLT 2016-12-06 06:01:52 +08:00			`#8744:74b789ada92a # Improved the matrix multiplication blocking in the case where mr is not a power of 2 (e.g on Haswell CPUs)`
Include recent changesets that played with product's kernel 2016-06-09 23:13:33 +08:00			`8789:efcb912e4356 # Made the index type a template parameter to evaluateProductBlockingSizes. Use numext::mini and numext::maxi instead of std::min/std::max to compute blocking sizes`
			`8972:81d53c711775 # Don't optimize the processing of the last rows of a matrix matrix product in cases that violate the assumptions made by the optimized code path`
			`8985:d935df21a082 # Remove the rotating kernel.`
Update relevent list of changesets. 2016-07-04 20:32:34 +08:00			`8988:6c2dc56e73b3 # Bug 256: enable vectorization with unaligned loads/stores.`
			`9148:b8b8c421e36c # Relax mixing-type constraints for binary coefficient-wise operators`
			`9174:d228bc282ac9 # merge`
add changesets related to matrix product perf. 2018-12-13 17:33:29 +08:00			`9175:abc7a3600098 # Include the cost of stores in unrolling`
Update relevent list of changesets. 2016-07-04 20:32:34 +08:00			`9212:c90098affa7b # Fix performance regression introduced in changeset 8aad8f35c955`
			`9213:9f1c14e4694b # Fix performance regression in dgemm introduced by changeset 81d53c711775`
extend monitoring benchmarks with transpose matrix-vector and triangular matrix-vectors. 2016-12-05 20:36:26 +08:00			`9361:69d418c06999 # 3.3-beta2`
add changesets related to matrix product perf. 2018-12-13 17:33:29 +08:00			`9445:f27ff0ad77a3 # Optimize expression matching "d?=a-bc" as "d?=a; d?=bc;"`
extend monitoring benchmarks with transpose matrix-vector and triangular matrix-vectors. 2016-12-05 20:36:26 +08:00			`9583:bef509908b9d # 3.3-rc1`
add changesets related to matrix product perf. 2018-12-13 17:33:29 +08:00			`9593:2f24280cf59a # Bug 1311: fix alignment logic in some cases of (scalar*small).lazyProduct(small)`
			`9722:040d861b88b5 # Disabled part of the matrix matrix peeling code that's incompatible with 512 bit registers`
extend monitoring benchmarks with transpose matrix-vector and triangular matrix-vectors. 2016-12-05 20:36:26 +08:00			`9792:26667be4f70b # 3.3.0`
add changesets related to matrix product perf. 2018-12-13 17:33:29 +08:00			`9891:41260bdfc23b # Fix a performance regression in (matmat)vec for which mat*mat was evaluated multiple times.`
extend monitoring benchmarks with transpose matrix-vector and triangular matrix-vectors. 2016-12-05 20:36:26 +08:00			`9942:b1d3eba60130 # Operators += and -= do not resize!`
			`9943:79bb9887afd4 # Ease compiler job to generate clean and efficient code in mat*vec`
			`9946:2213991340ea # Complete rewrite of column-major-matrix * vector product to deliver higher performance of modern CPU.`
Add performance monitoring for LLT 2016-12-06 06:01:52 +08:00			`9955:630471c3298c # Improve performance of row-major-dense-matrix * vector products for recent CPUs. (this is the next changeset fixing a typo)`
Update monitored changeset lists 2016-12-06 22:07:46 +08:00			`9975:2eeed9de710c # Revert vec/y to vec*(1/y) in row-major TRSM`
add changesets related to matrix product perf. 2018-12-13 17:33:29 +08:00			`10442:e3f17da72a40 # Bug 1435: fix aliasing issue in exressions like: A = C - B*A;`
			`10735:6913f0cf7d06 # Adds missing EIGEN_STRONG_INLINE to support MSVC properly inlining small vector calculations`
			`10943:4db388d946bd # Bug 1562: optimize evaluation of small products of the form sAB by rewriting them as: s*(A.lazyProduct(B)) to save a costly temporary. Measured speedup from 2x to 5x.`
			`10961:5007ff66c9f6 # Introduce the macro ei_declare_local_nested_eval to help allocating on the stack local temporaries via alloca, and let outer-products makes a good use of it.`
			`11083:30a528a984bb # Bug 1578: Improve prefetching in matrix multiplication on MIPS.`
			`11533:71609c41e9f8 # PR 526: Speed up multiplication of small, dynamically sized matrices`
			`11535:6d348dc9b092 # Vectorize row-by-row gebp loop iterations on 16 packets as well`
			`11568:efda481cbd7a # Bug 1624: improve matrix-matrix product on ARM 64, 20% speedup`
			`11596:b8d3f548a9d9 # do not read buffers out of bounds`
			`11628:22f9cc0079bd # Implement AVX512 vectorization of std::complex<float/double>`
			`11638:81172653b67b # Bug 1515: disable gebp's 3pX4 micro kernel for MSVC<=19.14 because of register spilling.`
			`11659:b500fef42ced # Artificially increase l1-blocking size for AVX512. +10% speedup with current kernels.`
extend monitoring benchmarks with transpose matrix-vector and triangular matrix-vectors. 2016-12-05 20:36:26 +08:00