Compare commits
525 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| daae8fd197 | |||
| a1fb7670f7 | |||
| 6c99c97489 | |||
| 6a0930560e | |||
| 24f8d5b624 | |||
| 77b4dbd53b | |||
| bc4c3bca01 | |||
| 6b0a9d135c | |||
| 137ccd9dd9 | |||
| 84923dedb7 | |||
| 8ec28ff461 | |||
| ca8ca796d3 | |||
| 8f811a9312 | |||
| 36a17536ca | |||
| bb9876db33 | |||
| d636b418af | |||
| a460c92577 | |||
| 33f838393c | |||
| a41d241a0e | |||
| 8da6b6ae52 | |||
| 01c4b82f04 | |||
| 93db123f7e | |||
| 752fdb5dd8 | |||
| 07ed01e97f | |||
| 35c5a32309 | |||
| c7b55b6082 | |||
| 840e01061f | |||
| 28ca97015d | |||
| 73c5ca74fa | |||
| e453555d97 | |||
| 6a6ffaff1e | |||
| 28ac9ea5a6 | |||
| a55694dd5b | |||
| 85a41e9cdb | |||
| 2c7392f07b | |||
| 81215711a2 | |||
| 809fd0d451 | |||
| 72e65157df | |||
| 69a8aa6de2 | |||
| 0ab5bf1746 | |||
| 22167170b3 | |||
| 69d9f36ff4 | |||
| f81815e48a | |||
| 5f855d965d | |||
| fa9ca65c0e | |||
| 719b68f077 | |||
| fe9f15f2d8 | |||
| 497f0c3d8a | |||
| ea37db828e | |||
| e6a0a3de73 | |||
| 6e70287776 | |||
| 58f236ad73 | |||
| e207107150 | |||
| c9d408064a | |||
| 288d1a3f6e | |||
| 7c1925acec | |||
| 2359c7c1a9 | |||
| 7646974227 | |||
| e3a80e6aa8 | |||
| 8866e393a2 | |||
| 9e87e6f3d8 | |||
| 3119b2ab4c | |||
| e7366a4161 | |||
| 6c0f79b787 | |||
| 7cd7acf71e | |||
| 72f14a0363 | |||
| 53026dc63a | |||
| 798f1595d5 | |||
| 2c0a008281 | |||
| c5425daa6b | |||
| eaab622f03 | |||
| 3cda1ce50a | |||
| 0391c07b17 | |||
| f4b095b1bb | |||
| 6940c59a88 | |||
| 42514fa24c | |||
| 650077074a | |||
| fe16a94fc2 | |||
| 632b8e0f05 | |||
| a1bc0fcf07 | |||
| 0464aa6784 | |||
| 55840f0bc9 | |||
| a7485f3222 | |||
| d31f62cf02 | |||
| 150c7294a6 | |||
| 8f5f614615 | |||
| f5fc109fbd | |||
| ce2028b425 | |||
| b47e6822aa | |||
| 0ae5e14923 | |||
| 95f2cea45b | |||
| e3c50643bb | |||
| efa84afd00 | |||
| d1b512a01a | |||
| e0b02789ff | |||
| 60596a1abc | |||
| afd514c25d | |||
| 25c7e3992f | |||
| f45776ec1f | |||
| e388459a27 | |||
| 0ac824f6a5 | |||
| f653e7a18d | |||
| 09e397e4f1 | |||
| 913e9546c0 | |||
| f946a89432 | |||
| 485df77612 | |||
| 038bfbb86c | |||
| 114fc0bae3 | |||
| 4a4f6658de | |||
| 0cf8046324 | |||
| e4c71a799a | |||
| e5752ff9b3 | |||
| 2619ad7ea5 | |||
| e5cc3d72c0 | |||
| 4938faa822 | |||
| 9fa986337d | |||
| 8a0b086b28 | |||
| 3eed97f6b9 | |||
| 13e137fbc9 | |||
| b3f24c9664 | |||
| 3d23f45107 | |||
| 87669d1c0a | |||
| 42285d8e70 | |||
| a76d1bb9e6 | |||
| 7c7e2d9dc6 | |||
| d602b99386 | |||
| 8aafa0473c | |||
| 11a627c54e | |||
| 4d0b005e5b | |||
| b81656936f | |||
| b973990df2 | |||
| 8fef2414b5 | |||
| 1e31124eb0 | |||
| cc9500db41 | |||
| b0652184ae | |||
| 723f396a20 | |||
| 03e5ff0687 | |||
| 47deec2c1a | |||
| bfc2a88594 | |||
| d741fc13d8 | |||
| 374260027d | |||
| 599de9e598 | |||
| 893bd14e92 | |||
| 43c0622e7b | |||
| 6aba7b66ce | |||
| 0623636c98 | |||
| 177b78c8b4 | |||
| 281a2b952f | |||
| c49c6b237d | |||
| e2469a9ebc | |||
| 5b71f3a8e4 | |||
| 9381ac2748 | |||
| 28ae3ca76f | |||
| b414283f48 | |||
| 38763ec4f3 | |||
| 452fbef0bf | |||
| 8c8313983b | |||
| 881a50c093 | |||
| 8213385ab8 | |||
| bede1c4fb4 | |||
| 1d2da67841 | |||
| 0dc291d3fa | |||
| e0ddd7d124 | |||
| adf4316f0e | |||
| 7fce11a5b8 | |||
| c40f01ccea | |||
| c567e34e6b | |||
| c917278d23 | |||
| 0639ed1258 | |||
| f017e169dc | |||
| 7e860acd38 | |||
| db00a51e6b | |||
| 7a4b3cfbf8 | |||
| 6c77b5f267 | |||
| d8b3c3c7db | |||
| beb18492fd | |||
| 441a9c8385 | |||
| 1236dbe5a6 | |||
| ef95cd471f | |||
| c92cd6d162 | |||
| cae5d9a20b | |||
| 3d891c3106 | |||
| 4fbdcfa823 | |||
| 1bb6a96ebc | |||
| 6bd163f37a | |||
| f0333333d1 | |||
| e14d50d86e | |||
| e89b979b2c | |||
| 7e9b29b9b8 | |||
| 5056a044b2 | |||
| 4054d32def | |||
| 2d52f0f4c3 | |||
| 3be5c3d343 | |||
| b18730f9e1 | |||
| be9e63c021 | |||
| 21844238ca | |||
| 07e7c36dac | |||
| 9251a2efde | |||
| 06adbf2e41 | |||
| b46e2b57cc | |||
| 3ce401f51b | |||
| 1c9f4422b5 | |||
| 27575d200a | |||
| 6157d0902a | |||
| 3fea849bbf | |||
| 8f177621bc | |||
| 5f402b7759 | |||
| 65bf0a343c | |||
| acf3d34bc5 | |||
| 8e75f7dcb4 | |||
| bd3546704c | |||
| 2df1e3372d | |||
| 4271b2b158 | |||
| 148493df89 | |||
| 415555a9c1 | |||
| 2a6fef9a55 | |||
| 307305aeb5 | |||
| cc26cdce0c | |||
| d8576826c4 | |||
| c6968edec4 | |||
| 9268314290 | |||
| 3ace0fda3f | |||
| 3cfc64404a | |||
| 72956e8950 | |||
| 505dc08635 | |||
| 61587b0670 | |||
| 632fc75d77 | |||
| 2c222f1faa | |||
| 66ac898f64 | |||
| ab87ee6b48 | |||
| a07807caac | |||
| b71f4fe681 | |||
| a0128aa489 | |||
| 627133f9ad | |||
| 0e2cf102e1 | |||
| 5e3e91d0fc | |||
| 28c3fa8950 | |||
| 8ac87c1cb6 | |||
| b7cee00455 | |||
| 962b20a9bb | |||
| fbf83f4833 | |||
| 78cec6209c | |||
| c460027dbe | |||
| bfa9b9f6b2 | |||
| c7a8512d12 | |||
| db72ad8f6a | |||
| 97ecd4996a | |||
| 1eb43cccad | |||
| 9d92f526dd | |||
| 514d237257 | |||
| 30486a356c | |||
| e1b2502840 | |||
| 9de52b489a | |||
| 860dcfc703 | |||
| f96afd94b0 | |||
| ebe84215e4 | |||
| 845e6d750f | |||
| a89d6711c6 | |||
| 9c017a2218 | |||
| 0e6b11b708 | |||
| 6aaa107865 | |||
| 00c42dc815 | |||
| 79e754e548 | |||
| 2ccd7f6e0c | |||
| e3d844b062 | |||
| def146efed | |||
| 7543e578a4 | |||
| 601c71fe54 | |||
| 3810a6fd99 | |||
| 742f54c235 | |||
| d674fbb4c7 | |||
| 2922c15f36 | |||
| 3a245a376f | |||
| 46c9357c72 | |||
| 1c3e2d3dd5 | |||
| f66d908282 | |||
| ba1f91f17b | |||
| f460776f0f | |||
| e882f3d6f3 | |||
| dddedbab5d | |||
| 1cfdb2295d | |||
| 00740c0e34 | |||
| 254db9bd7c | |||
| f2074f9ac1 | |||
| aece65ea29 | |||
| ef64991506 | |||
| 7a867082d8 | |||
| a1b87eac6b | |||
| 47ebce4d1a | |||
| 69b560751c | |||
| 0a7a527a92 | |||
| 50715e8945 | |||
| 11911fd941 | |||
| 408b4fe83f | |||
| 4474465438 | |||
| b9ec72546c | |||
| 37858d1146 | |||
| 1763e01567 | |||
| 6aac06587d | |||
| 7c1acc07f0 | |||
| 38d273ea03 | |||
| 7242cdc4ec | |||
| 90a4dab501 | |||
| a8a342ccc4 | |||
| 9e9a9553db | |||
| be7c1b6324 | |||
| be8e8228ae | |||
| 77b197bc93 | |||
| 9cc6599056 | |||
| da611dd81a | |||
| e6b9b660c3 | |||
| 62635b1fb5 | |||
| 75ddfbd687 | |||
| 2f12ea017b | |||
| d3b677fe87 | |||
| 6f83004252 | |||
| 505b218829 | |||
| 719fcc56b0 | |||
| 1296c89688 | |||
| dfb91f2a31 | |||
| 2ec76011f9 | |||
| 7abbe40980 | |||
| ca32b66a1c | |||
| d9346930dd | |||
| 0bfe5c6822 | |||
| 4899d67f7d | |||
| 1d1854032b | |||
| ebb04e3265 | |||
| 2c51a990ac | |||
| 7892434572 | |||
| d798487213 | |||
| 0ba64cee60 | |||
| 53aee36076 | |||
| c4e5ba1bfe | |||
| 486a485bb7 | |||
| 08c7d1ddf8 | |||
| e0bd5b5c0e | |||
| ae65f75572 | |||
| 1e9247c276 | |||
| 251715d9ef | |||
| a6f533b248 | |||
| 50deeb49b7 | |||
| 4260215adf | |||
| 89ec2be110 | |||
| 8381591f56 | |||
| d245caa49a | |||
| c9c13a8a0c | |||
| e70a6b92bf | |||
| 7a96499b29 | |||
| 63cfa32691 | |||
| d537e0de8c | |||
| f00bbb9dbf | |||
| 1a02a087a1 | |||
| 02c1f86055 | |||
| 5e0f67c666 | |||
| 7345795e64 | |||
| a36e976491 | |||
| 1169f489a4 | |||
| 7a4ebf825b | |||
| ff17e3eb9f | |||
| c56d4881f9 | |||
| b03d50b794 | |||
| ea1095135e | |||
| 4c5df489db | |||
| eb98fdddfc | |||
| ca17b4b75c | |||
| dc24914415 | |||
| 585c0010a5 | |||
| 857f61bc5d | |||
| 9332042d5f | |||
| ae93532fd3 | |||
| c4af196a2d | |||
| 480e697681 | |||
| 3c4c472584 | |||
| a797666fbe | |||
| 5dde4e65d3 | |||
| 27a9df6477 | |||
| 7224022473 | |||
| 468ac3df9e | |||
| 376048156b | |||
| d1c5b8f913 | |||
| 91bde7d315 | |||
| 80373ea039 | |||
| d12b75a6c4 | |||
| 7294fb1d9d | |||
| 31e086d6a6 | |||
| cbb47736af | |||
| 2a7c6930ac | |||
| 00774b1105 | |||
| 6497aae57c | |||
| c4ec882020 | |||
| d33fc32cf3 | |||
| 529bfc36ec | |||
| 88249ca5f7 | |||
| 731c518cff | |||
| 29fc429d9a | |||
| e2d3b1561a | |||
| 4a012c3d20 | |||
| d5ef0dee9a | |||
| a590e6135c | |||
| 4239dd65ce | |||
| 3db2adf872 | |||
| ad2462811a | |||
| c1cf62d2c0 | |||
| bfe1656b8b | |||
| f02d535fde | |||
| 49e62c0e77 | |||
| 3381f23709 | |||
| fa6a920caa | |||
| a6515bb858 | |||
| c66b842d66 | |||
| df2dfe65d6 | |||
| 2c8d634619 | |||
| 37efb5bc1d | |||
| 97d671eb61 | |||
| 305cd2e8b4 | |||
| 09bc6ebe5b | |||
| 872a11a2bf | |||
| eda9e8632a | |||
| 8f83d3f961 | |||
| e5e47cfdb5 | |||
| ebf9e9dabe | |||
| 83bd547517 | |||
| e25f4c01d6 | |||
| 54915ce343 | |||
| 0150fabdb6 | |||
| 4f0773f07d | |||
| aa5edebc80 | |||
| 89924b3d5b | |||
| da7f0ff425 | |||
| 0d5c8e5386 | |||
| 912410f214 | |||
| b122413fb0 | |||
| 9b7b5f7fdc | |||
| 34513be726 | |||
| 482015f8d6 | |||
| 639000e34f | |||
| 5de7727cc7 | |||
| 96df4b9b17 | |||
| 29dc8e0c61 | |||
| 65e56cb29d | |||
| bd831a03a8 | |||
| edc97918f8 | |||
| e0034de22d | |||
| 32c7fe6bff | |||
| 19bdf9d52b | |||
| 4f09030fdc | |||
| 6f4eca5ea4 | |||
| be55f96cbd | |||
| 96dd0ef4f7 | |||
| 8f0d6c06a9 | |||
| 410a07cbec | |||
| 72f95a0acc | |||
| e545b81e76 | |||
| d7afdf9137 | |||
| 4f4daaa42a | |||
| 42bbe74791 | |||
| c8322c65e4 | |||
| 87dde1fde6 | |||
| 42466e54fa | |||
| 3b0624d50f | |||
| fd4e68128e | |||
| 6464d1723a | |||
| 59c97cfee4 | |||
| de7875ca5d | |||
| 67836c2ab4 | |||
| 5fecfe0f42 | |||
| bba6676803 | |||
| 5649b2c53a | |||
| 6e972994b2 | |||
| 5b04cf7ab4 | |||
| d5ea8fd823 | |||
| 4beffaaa4b | |||
| fb28e4adc9 | |||
| 26faa3ca47 | |||
| 4f75989634 | |||
| 1e06b49854 | |||
| 7f546f54fa | |||
| a809431e34 | |||
| 5ee1cf0223 | |||
| 9aea7a0d9a | |||
| da0987507c | |||
| 81fed55782 | |||
| 35387edb8d | |||
| 9c884986ad | |||
| f2f0e98bb5 | |||
| 166d64eb7c | |||
| e078339e8d | |||
| 832a272784 | |||
| 356606314c | |||
| ed79a29d87 | |||
| 77d16ffc69 | |||
| 56762d5e4c | |||
| 90dd190a6d | |||
| ab9ec4ab4e | |||
| 0cbd2d34e4 | |||
| 62979fd104 | |||
| 20a413e154 | |||
| dc40bc7368 | |||
| 1acfc78c8f | |||
| b4071d0d16 | |||
| 7908efafc8 | |||
| 66dc10b019 | |||
| c9ff735da6 | |||
| 99880f7906 | |||
| cd135e2b59 | |||
| ad124a5e8b | |||
| 211d2eceb5 | |||
| 5813ed095b | |||
| e44b028fe5 | |||
| a6efabf155 | |||
| ea26b00c06 | |||
| 08786c4b95 | |||
| 12e476f7a2 | |||
| 8de40955ad | |||
| 9b24688eed | |||
| 43224f7273 | |||
| 9254a701f3 | |||
| 26a614fdd1 | |||
| 7ae64f4f9c | |||
| 82e80fa82b | |||
| bcfc298c38 | |||
| ce7c6c6b2d | |||
| 85636ff1a0 | |||
| 12ab1804b6 |
@@ -5,6 +5,7 @@
|
||||
*.def
|
||||
*.o
|
||||
*.out
|
||||
*.tmp
|
||||
lapack-3.1.1
|
||||
lapack-3.1.1.tgz
|
||||
lapack-3.4.1
|
||||
|
||||
+153
-29
@@ -1,5 +1,158 @@
|
||||
# XXX: Precise is already deprecated, new default is Trusty.
|
||||
# https://blog.travis-ci.com/2017-07-11-trusty-as-default-linux-is-coming
|
||||
dist: precise
|
||||
sudo: true
|
||||
language: c
|
||||
|
||||
jobs:
|
||||
include:
|
||||
- &test-ubuntu
|
||||
stage: test
|
||||
compiler: gcc
|
||||
addons:
|
||||
apt:
|
||||
packages:
|
||||
- gfortran
|
||||
before_script: &common-before
|
||||
- COMMON_FLAGS="DYNAMIC_ARCH=1 TARGET=NEHALEM NUM_THREADS=32"
|
||||
script:
|
||||
- set -e
|
||||
- travis_wait 45 make QUIET_MAKE=1 $COMMON_FLAGS $BTYPE
|
||||
- make -C test $COMMON_FLAGS $BTYPE
|
||||
- make -C ctest $COMMON_FLAGS $BTYPE
|
||||
- make -C utest $COMMON_FLAGS $BTYPE
|
||||
env:
|
||||
- TARGET_BOX=LINUX64
|
||||
- BTYPE="BINARY=64"
|
||||
|
||||
- <<: *test-ubuntu
|
||||
env:
|
||||
- TARGET_BOX=LINUX64
|
||||
- BTYPE="BINARY=64 USE_OPENMP=1"
|
||||
|
||||
- <<: *test-ubuntu
|
||||
env:
|
||||
- TARGET_BOX=LINUX64
|
||||
- BTYPE="BINARY=64 INTERFACE64=1"
|
||||
|
||||
- <<: *test-ubuntu
|
||||
compiler: clang
|
||||
env:
|
||||
- TARGET_BOX=LINUX64
|
||||
- BTYPE="BINARY=64 CC=clang"
|
||||
|
||||
- <<: *test-ubuntu
|
||||
compiler: clang
|
||||
env:
|
||||
- TARGET_BOX=LINUX64
|
||||
- BTYPE="BINARY=64 INTERFACE64=1 CC=clang"
|
||||
|
||||
- <<: *test-ubuntu
|
||||
addons:
|
||||
apt:
|
||||
packages:
|
||||
- gcc-multilib
|
||||
- gfortran-multilib
|
||||
env:
|
||||
- TARGET_BOX=LINUX32
|
||||
- BTYPE="BINARY=32"
|
||||
|
||||
- stage: test
|
||||
compiler: gcc
|
||||
addons:
|
||||
apt:
|
||||
packages:
|
||||
- binutils-mingw-w64-x86-64
|
||||
- gcc-mingw-w64-x86-64
|
||||
- gfortran-mingw-w64-x86-64
|
||||
before_script: *common-before
|
||||
script:
|
||||
- travis_wait 45 make QUIET_MAKE=1 $COMMON_FLAGS $BTYPE
|
||||
env:
|
||||
- TARGET_BOX=WIN64
|
||||
- BTYPE="BINARY=64 HOSTCC=gcc CC=x86_64-w64-mingw32-gcc FC=x86_64-w64-mingw32-gfortran"
|
||||
|
||||
# Build & test on Alpine Linux inside chroot, i.e. on system with musl libc.
|
||||
# These jobs needs sudo, so Travis runs them on VM-based infrastructure
|
||||
# which is slower than container-based infrastructure used for jobs
|
||||
# that don't require sudo.
|
||||
- &test-alpine
|
||||
stage: test
|
||||
dist: trusty
|
||||
sudo: true
|
||||
language: minimal
|
||||
before_install:
|
||||
- "wget 'https://raw.githubusercontent.com/alpinelinux/alpine-chroot-install/v0.6.0/alpine-chroot-install' \
|
||||
&& echo 'a827a4ba3d0817e7c88bae17fe34e50204983d1e alpine-chroot-install' | sha1sum -c || exit 1"
|
||||
- alpine() { /alpine/enter-chroot -u "$USER" "$@"; }
|
||||
install:
|
||||
- sudo sh alpine-chroot-install -p 'build-base gfortran perl linux-headers'
|
||||
before_script: *common-before
|
||||
script:
|
||||
- set -e
|
||||
# XXX: Disable some warnings for now to avoid exceeding Travis limit for log size.
|
||||
- alpine make QUIET_MAKE=1 $COMMON_FLAGS $BTYPE
|
||||
CFLAGS="-Wno-misleading-indentation -Wno-sign-conversion -Wno-incompatible-pointer-types"
|
||||
- alpine make -C test $COMMON_FLAGS $BTYPE
|
||||
- alpine make -C ctest $COMMON_FLAGS $BTYPE
|
||||
- alpine make -C utest $COMMON_FLAGS $BTYPE
|
||||
env:
|
||||
- TARGET_BOX=LINUX64_MUSL
|
||||
- BTYPE="BINARY=64"
|
||||
|
||||
# XXX: This job segfaults in TESTS OF THE COMPLEX LEVEL 3 BLAS,
|
||||
# but only on Travis CI, cannot reproduce it elsewhere.
|
||||
#- &test-alpine-openmp
|
||||
# <<: *test-alpine
|
||||
# env:
|
||||
# - TARGET_BOX=LINUX64_MUSL
|
||||
# - BTYPE="BINARY=64 USE_OPENMP=1"
|
||||
|
||||
- <<: *test-alpine
|
||||
env:
|
||||
- TARGET_BOX=LINUX64_MUSL
|
||||
- BTYPE="BINARY=64 INTERFACE64=1"
|
||||
|
||||
# Build with the same flags as Alpine do in OpenBLAS package.
|
||||
- <<: *test-alpine
|
||||
env:
|
||||
- TARGET_BOX=LINUX64_MUSL
|
||||
- BTYPE="BINARY=64 NO_AFFINITY=1 USE_OPENMP=0 NO_LAPACK=0 TARGET=core2"
|
||||
|
||||
- &test-cmake
|
||||
stage: test
|
||||
compiler: clang
|
||||
addons:
|
||||
apt:
|
||||
packages:
|
||||
- gfortran
|
||||
- cmake
|
||||
dist: trusty
|
||||
sudo: true
|
||||
before_script:
|
||||
- COMMON_ARGS="-DTARGET=NEHALEM -DNUM_THREADS=32"
|
||||
script:
|
||||
- set -e
|
||||
- mkdir build
|
||||
- CONFIG=Release
|
||||
- cmake -Bbuild -H. $CMAKE_ARGS $COMMON_ARGS -DCMAKE_BUILD_TYPE=$CONFIG
|
||||
- cmake --build build --config $CONFIG -- -j2
|
||||
env:
|
||||
- CMAKE=1
|
||||
- <<: *test-cmake
|
||||
env:
|
||||
- CMAKE=1 CMAKE_ARGS="-DNOFORTRAN=1"
|
||||
- <<: *test-cmake
|
||||
compiler: gcc
|
||||
env:
|
||||
- CMAKE=1
|
||||
|
||||
# whitelist
|
||||
branches:
|
||||
only:
|
||||
- master
|
||||
- develop
|
||||
|
||||
notifications:
|
||||
webhooks:
|
||||
urls:
|
||||
@@ -7,32 +160,3 @@ notifications:
|
||||
on_success: change # options: [always|never|change] default: always
|
||||
on_failure: always # options: [always|never|change] default: always
|
||||
on_start: never # options: [always|never|change] default: always
|
||||
|
||||
compiler:
|
||||
- gcc
|
||||
|
||||
env:
|
||||
- TARGET_BOX=LINUX64 BTYPE="BINARY=64"
|
||||
- TARGET_BOX=LINUX64 BTYPE="BINARY=64 USE_OPENMP=1"
|
||||
- TARGET_BOX=LINUX64 BTYPE="BINARY=64 INTERFACE64=1"
|
||||
- TARGET_BOX=LINUX32 BTYPE="BINARY=32"
|
||||
- TARGET_BOX=WIN64 BTYPE="BINARY=64 HOSTCC=gcc CC=x86_64-w64-mingw32-gcc FC=x86_64-w64-mingw32-gfortran"
|
||||
|
||||
before_install:
|
||||
- sudo apt-get update -qq
|
||||
- sudo apt-get install -qq gfortran
|
||||
- if [[ "$TARGET_BOX" == "WIN64" ]]; then sudo apt-get install -qq binutils-mingw-w64-x86-64 gcc-mingw-w64-x86-64 gfortran-mingw-w64-x86-64; fi
|
||||
- if [[ "$TARGET_BOX" == "LINUX32" ]]; then sudo apt-get install -qq gcc-multilib gfortran-multilib; fi
|
||||
|
||||
script:
|
||||
- set -e
|
||||
- make QUIET_MAKE=1 DYNAMIC_ARCH=1 TARGET=NEHALEM NUM_THREADS=32 $BTYPE
|
||||
- if [ "$TARGET_BOX" == "LINUX32" ] || [ "$TARGET_BOX" == "LINUX64" ]; then make -C test DYNAMIC_ARCH=1 TARGET=NEHALEM NUM_THREADS=32 $BTYPE; fi
|
||||
- if [ "$TARGET_BOX" == "LINUX32" ] || [ "$TARGET_BOX" == "LINUX64" ]; then make -C ctest DYNAMIC_ARCH=1 TARGET=NEHALEM NUM_THREADS=32 $BTYPE; fi
|
||||
- if [ "$TARGET_BOX" == "LINUX32" ] || [ "$TARGET_BOX" == "LINUX64" ]; then make -C utest DYNAMIC_ARCH=1 TARGET=NEHALEM NUM_THREADS=32 $BTYPE; fi
|
||||
|
||||
# whitelist
|
||||
branches:
|
||||
only:
|
||||
- master
|
||||
- develop
|
||||
+136
-105
@@ -2,50 +2,31 @@
|
||||
## Author: Hank Anderson <hank@statease.com>
|
||||
##
|
||||
|
||||
cmake_minimum_required(VERSION 2.8.4)
|
||||
project(OpenBLAS)
|
||||
cmake_minimum_required(VERSION 2.8.5)
|
||||
project(OpenBLAS C ASM)
|
||||
set(OpenBLAS_MAJOR_VERSION 0)
|
||||
set(OpenBLAS_MINOR_VERSION 2)
|
||||
set(OpenBLAS_PATCH_VERSION 20.dev)
|
||||
set(OpenBLAS_MINOR_VERSION 3)
|
||||
set(OpenBLAS_PATCH_VERSION 0.dev)
|
||||
set(OpenBLAS_VERSION "${OpenBLAS_MAJOR_VERSION}.${OpenBLAS_MINOR_VERSION}.${OpenBLAS_PATCH_VERSION}")
|
||||
|
||||
enable_language(ASM)
|
||||
enable_language(C)
|
||||
# Adhere to GNU filesystem layout conventions
|
||||
include(GNUInstallDirs)
|
||||
|
||||
if(MSVC)
|
||||
set(OpenBLAS_LIBNAME libopenblas)
|
||||
else()
|
||||
set(OpenBLAS_LIBNAME openblas)
|
||||
endif()
|
||||
|
||||
#######
|
||||
if(MSVC)
|
||||
option(BUILD_WITHOUT_LAPACK "Without LAPACK and LAPACKE (Only BLAS or CBLAS)" ON)
|
||||
endif()
|
||||
option(BUILD_WITHOUT_CBLAS "Without CBLAS" OFF)
|
||||
option(BUILD_DEBUG "Build Debug Version" OFF)
|
||||
option(DYNAMIC_ARCH "Build with DYNAMIC_ARCH" OFF)
|
||||
option(BUILD_RELAPACK "Build with ReLAPACK (recursive LAPACK" OFF)
|
||||
#######
|
||||
if(BUILD_WITHOUT_LAPACK)
|
||||
set(NO_LAPACK 1)
|
||||
set(NO_LAPACKE 1)
|
||||
endif()
|
||||
|
||||
if(CMAKE_CONFIGURATION_TYPES) # multiconfig generator?
|
||||
set(CMAKE_CONFIGURATION_TYPES "Debug;Release" CACHE STRING "" FORCE)
|
||||
set(CMAKE_BUILD_TYPE
|
||||
Debug Debug
|
||||
Release Release
|
||||
)
|
||||
else()
|
||||
if( NOT CMAKE_BUILD_TYPE )
|
||||
if(BUILD_DEBUG)
|
||||
set(CMAKE_BUILD_TYPE Debug)
|
||||
else()
|
||||
set(CMAKE_BUILD_TYPE Release)
|
||||
endif()
|
||||
endif()
|
||||
endif()
|
||||
|
||||
if(BUILD_WITHOUT_CBLAS)
|
||||
set(NO_CBLAS 1)
|
||||
endif()
|
||||
@@ -71,6 +52,9 @@ endif ()
|
||||
set(SUBDIRS ${BLASDIRS})
|
||||
if (NOT NO_LAPACK)
|
||||
list(APPEND SUBDIRS lapack)
|
||||
if(BUILD_RELAPACK)
|
||||
list(APPEND SUBDIRS relapack/src)
|
||||
endif()
|
||||
endif ()
|
||||
|
||||
# set which float types we want to build for
|
||||
@@ -82,6 +66,10 @@ if (NOT DEFINED BUILD_SINGLE AND NOT DEFINED BUILD_DOUBLE AND NOT DEFINED BUILD_
|
||||
set(BUILD_COMPLEX16 true)
|
||||
endif ()
|
||||
|
||||
if (NOT DEFINED BUILD_MATGEN)
|
||||
set(BUILD_MATGEN true)
|
||||
endif()
|
||||
|
||||
set(FLOAT_TYPES "")
|
||||
if (BUILD_SINGLE)
|
||||
message(STATUS "Building Single Precision")
|
||||
@@ -103,23 +91,17 @@ if (BUILD_COMPLEX16)
|
||||
list(APPEND FLOAT_TYPES "ZCOMPLEX") # defines COMPLEX and DOUBLE
|
||||
endif ()
|
||||
|
||||
set(SUBDIRS_ALL ${SUBDIRS} test ctest utest exports benchmark ../laswp ../bench)
|
||||
|
||||
# all :: libs netlib tests shared
|
||||
|
||||
# libs :
|
||||
if (NOT DEFINED CORE OR "${CORE}" STREQUAL "UNKNOWN")
|
||||
message(FATAL_ERROR "Detecting CPU failed. Please set TARGET explicitly, e.g. make TARGET=your_cpu_target. Please read README for details.")
|
||||
endif ()
|
||||
|
||||
if (${NO_STATIC} AND ${NO_SHARED})
|
||||
message(FATAL_ERROR "Neither static nor shared are enabled.")
|
||||
endif ()
|
||||
|
||||
#Set default output directory
|
||||
set( CMAKE_LIBRARY_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR}/lib )
|
||||
set( CMAKE_ARCHIVE_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR}/lib )
|
||||
|
||||
set( CMAKE_LIBRARY_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR}/lib)
|
||||
set( CMAKE_ARCHIVE_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR}/lib)
|
||||
if(MSVC)
|
||||
set( CMAKE_LIBRARY_OUTPUT_DIRECTORY_DEBUG ${PROJECT_BINARY_DIR}/lib/Debug)
|
||||
set( CMAKE_ARCHIVE_OUTPUT_DIRECTORY_RELEASE ${PROJECT_BINARY_DIR}/lib/Release)
|
||||
endif ()
|
||||
# get obj vars into format that add_library likes: $<TARGET_OBJS:objlib> (see http://www.cmake.org/cmake/help/v3.0/command/add_library.html)
|
||||
set(TARGET_OBJS "")
|
||||
foreach (SUBDIR ${SUBDIRS})
|
||||
@@ -134,20 +116,51 @@ endforeach ()
|
||||
# Not using add_subdirectory here because lapack-netlib already has its own CMakeLists.txt. Instead include a cmake script with the sources we want.
|
||||
if (NOT NOFORTRAN AND NOT NO_LAPACK)
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/lapack.cmake")
|
||||
if (NOT NO_LAPACKE)
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/lapacke.cmake")
|
||||
endif ()
|
||||
if (NOT NO_LAPACKE)
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/lapacke.cmake")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
#Only generate .def for dll on MSVC
|
||||
# Only generate .def for dll on MSVC and always produce pdb files for debug and release
|
||||
if(MSVC)
|
||||
set(OpenBLAS_DEF_FILE "${PROJECT_BINARY_DIR}/openblas.def")
|
||||
if (${CMAKE_MAJOR_VERSION}.${CMAKE_MINOR_VERSION} LESS 3.4)
|
||||
set(OpenBLAS_DEF_FILE "${PROJECT_BINARY_DIR}/openblas.def")
|
||||
endif()
|
||||
set(CMAKE_C_FLAGS_RELEASE "${CMAKE_C_FLAGS_RELEASE} /Zi")
|
||||
set(CMAKE_SHARED_LINKER_FLAGS_RELEASE "${CMAKE_SHARED_LINKER_FLAGS_RELEASE} /DEBUG /OPT:REF /OPT:ICF")
|
||||
endif()
|
||||
|
||||
# add objects to the openblas lib
|
||||
add_library(${OpenBLAS_LIBNAME} SHARED ${LA_SOURCES} ${LAPACKE_SOURCES} ${TARGET_OBJS} ${OpenBLAS_DEF_FILE})
|
||||
if (${DYNAMIC_ARCH})
|
||||
add_subdirectory(kernel)
|
||||
foreach(TARGET_CORE ${DYNAMIC_CORE})
|
||||
message("${TARGET_CORE}")
|
||||
list(APPEND TARGET_OBJS "$<TARGET_OBJECTS:kernel_${TARGET_CORE}>")
|
||||
endforeach()
|
||||
endif ()
|
||||
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/export.cmake")
|
||||
# Only build shared libs for MSVC
|
||||
if (MSVC)
|
||||
set(BUILD_SHARED_LIBS ON)
|
||||
endif()
|
||||
|
||||
|
||||
# add objects to the openblas lib
|
||||
add_library(${OpenBLAS_LIBNAME} ${LA_SOURCES} ${LAPACKE_SOURCES} ${RELA_SOURCES} ${TARGET_OBJS} ${OpenBLAS_DEF_FILE})
|
||||
|
||||
# Android needs to explicitly link against libm
|
||||
if(ANDROID)
|
||||
target_link_libraries(${OpenBLAS_LIBNAME} m)
|
||||
endif()
|
||||
|
||||
# Handle MSVC exports
|
||||
if(MSVC AND BUILD_SHARED_LIBS)
|
||||
if (${CMAKE_MAJOR_VERSION}.${CMAKE_MINOR_VERSION} LESS 3.4)
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/export.cmake")
|
||||
else()
|
||||
# Creates verbose .def file (51KB vs 18KB)
|
||||
set_target_properties(${OpenBLAS_LIBNAME} PROPERTIES WINDOWS_EXPORT_ALL_SYMBOLS true)
|
||||
endif()
|
||||
endif()
|
||||
|
||||
# Set output for libopenblas
|
||||
set_target_properties( ${OpenBLAS_LIBNAME} PROPERTIES RUNTIME_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR}/lib)
|
||||
@@ -155,84 +168,102 @@ set_target_properties( ${OpenBLAS_LIBNAME} PROPERTIES LIBRARY_OUTPUT_NAME_DEBUG
|
||||
|
||||
foreach (OUTPUTCONFIG ${CMAKE_CONFIGURATION_TYPES})
|
||||
string( TOUPPER ${OUTPUTCONFIG} OUTPUTCONFIG )
|
||||
|
||||
set_target_properties( ${OpenBLAS_LIBNAME} PROPERTIES RUNTIME_OUTPUT_DIRECTORY_${OUTPUTCONFIG} ${PROJECT_BINARY_DIR}/lib)
|
||||
set_target_properties( ${OpenBLAS_LIBNAME} PROPERTIES LIBRARY_OUTPUT_DIRECTORY_${OUTPUTCONFIG} ${PROJECT_BINARY_DIR}/lib)
|
||||
set_target_properties( ${OpenBLAS_LIBNAME} PROPERTIES ARCHIVE_OUTPUT_DIRECTORY_${OUTPUTCONFIG} ${PROJECT_BINARY_DIR}/lib)
|
||||
|
||||
set_target_properties( ${OpenBLAS_LIBNAME} PROPERTIES RUNTIME_OUTPUT_DIRECTORY_${OUTPUTCONFIG} ${PROJECT_BINARY_DIR}/lib/${OUTPUTCONFIG} )
|
||||
set_target_properties( ${OpenBLAS_LIBNAME} PROPERTIES LIBRARY_OUTPUT_DIRECTORY_${OUTPUTCONFIG} ${PROJECT_BINARY_DIR}/lib/${OUTPUTCONFIG} )
|
||||
set_target_properties( ${OpenBLAS_LIBNAME} PROPERTIES ARCHIVE_OUTPUT_DIRECTORY_${OUTPUTCONFIG} ${PROJECT_BINARY_DIR}/lib/${OUTPUTCONFIG} )
|
||||
endforeach()
|
||||
|
||||
enable_testing()
|
||||
add_subdirectory(utest)
|
||||
|
||||
if(NOT MSVC)
|
||||
#only build shared library for MSVC
|
||||
|
||||
add_library(${OpenBLAS_LIBNAME}_static STATIC ${LA_SOURCES} ${LAPACKE_SOURCES} ${TARGET_OBJS})
|
||||
set_target_properties(${OpenBLAS_LIBNAME}_static PROPERTIES OUTPUT_NAME ${OpenBLAS_LIBNAME})
|
||||
set_target_properties(${OpenBLAS_LIBNAME}_static PROPERTIES CLEAN_DIRECT_OUTPUT 1)
|
||||
|
||||
if(SMP)
|
||||
target_link_libraries(${OpenBLAS_LIBNAME} pthread)
|
||||
target_link_libraries(${OpenBLAS_LIBNAME}_static pthread)
|
||||
if (USE_THREAD)
|
||||
# Add threading library to linker
|
||||
find_package(Threads)
|
||||
if (THREADS_HAVE_PTHREAD_ARG)
|
||||
set_property(TARGET ${OpenBLAS_LIBNAME} PROPERTY COMPILE_OPTIONS "-pthread")
|
||||
set_property(TARGET ${OpenBLAS_LIBNAME} PROPERTY INTERFACE_COMPILE_OPTIONS "-pthread")
|
||||
endif()
|
||||
target_link_libraries(${OpenBLAS_LIBNAME} ${CMAKE_THREAD_LIBS_INIT})
|
||||
endif()
|
||||
|
||||
#build test and ctest
|
||||
add_subdirectory(test)
|
||||
if(NOT NO_CBLAS)
|
||||
add_subdirectory(ctest)
|
||||
endif()
|
||||
if (MSVC OR NOT NOFORTRAN)
|
||||
# Broken without fortran on unix
|
||||
add_subdirectory(utest)
|
||||
endif()
|
||||
|
||||
set_target_properties(${OpenBLAS_LIBNAME} PROPERTIES
|
||||
if (NOT MSVC AND NOT NOFORTRAN)
|
||||
# Build test and ctest
|
||||
add_subdirectory(test)
|
||||
if(NOT NO_CBLAS)
|
||||
add_subdirectory(ctest)
|
||||
endif()
|
||||
endif()
|
||||
|
||||
set_target_properties(${OpenBLAS_LIBNAME} PROPERTIES
|
||||
VERSION ${OpenBLAS_MAJOR_VERSION}.${OpenBLAS_MINOR_VERSION}
|
||||
SOVERSION ${OpenBLAS_MAJOR_VERSION}
|
||||
)
|
||||
|
||||
|
||||
# TODO: Why is the config saved here? Is this necessary with CMake?
|
||||
#Save the config files for installation
|
||||
# @cp Makefile.conf Makefile.conf_last
|
||||
# @cp config.h config_last.h
|
||||
#ifdef QUAD_PRECISION
|
||||
# @echo "#define QUAD_PRECISION">> config_last.h
|
||||
#endif
|
||||
#ifeq ($(EXPRECISION), 1)
|
||||
# @echo "#define EXPRECISION">> config_last.h
|
||||
#endif
|
||||
###
|
||||
#ifeq ($(DYNAMIC_ARCH), 1)
|
||||
# @$(MAKE) -C kernel commonlibs || exit 1
|
||||
# @for d in $(DYNAMIC_CORE) ; \
|
||||
# do $(MAKE) GOTOBLAS_MAKEFILE= -C kernel TARGET_CORE=$$d kernel || exit 1 ;\
|
||||
# done
|
||||
# @echo DYNAMIC_ARCH=1 >> Makefile.conf_last
|
||||
#endif
|
||||
#ifdef USE_THREAD
|
||||
# @echo USE_THREAD=$(USE_THREAD) >> Makefile.conf_last
|
||||
#endif
|
||||
# @touch lib.grd
|
||||
|
||||
# Install project
|
||||
|
||||
# Install libraries
|
||||
install(TARGETS ${OpenBLAS_LIBNAME}
|
||||
RUNTIME DESTINATION bin
|
||||
ARCHIVE DESTINATION lib
|
||||
LIBRARY DESTINATION lib )
|
||||
RUNTIME DESTINATION ${CMAKE_INSTALL_BINDIR}
|
||||
ARCHIVE DESTINATION ${CMAKE_INSTALL_LIBDIR}
|
||||
LIBRARY DESTINATION ${CMAKE_INSTALL_LIBDIR} )
|
||||
|
||||
# Install include files
|
||||
FILE(GLOB_RECURSE INCLUDE_FILES "${CMAKE_CURRENT_SOURCE_DIR}/*.h")
|
||||
install (FILES ${INCLUDE_FILES} DESTINATION include)
|
||||
|
||||
if(NOT MSVC)
|
||||
install (TARGETS ${OpenBLAS_LIBNAME}_static DESTINATION lib)
|
||||
message(STATUS "Generating openblas_config.h in ${CMAKE_INSTALL_INCLUDEDIR}")
|
||||
|
||||
set(OPENBLAS_CONFIG_H ${CMAKE_BINARY_DIR}/openblas_config.h)
|
||||
file(WRITE ${OPENBLAS_CONFIG_H} "#ifndef OPENBLAS_CONFIG_H\n")
|
||||
file(APPEND ${OPENBLAS_CONFIG_H} "#define OPENBLAS_CONFIG_H\n")
|
||||
file(STRINGS ${PROJECT_BINARY_DIR}/config.h __lines)
|
||||
foreach(line ${__lines})
|
||||
string(REPLACE "#define " "" line ${line})
|
||||
file(APPEND ${OPENBLAS_CONFIG_H} "#define OPENBLAS_${line}\n")
|
||||
endforeach()
|
||||
file(APPEND ${OPENBLAS_CONFIG_H} "#define OPENBLAS_VERSION \"OpenBLAS ${OpenBLAS_VERSION}\"\n")
|
||||
file(READ ${CMAKE_CURRENT_SOURCE_DIR}/openblas_config_template.h OPENBLAS_CONFIG_TEMPLATE_H_CONTENTS)
|
||||
file(APPEND ${OPENBLAS_CONFIG_H} "${OPENBLAS_CONFIG_TEMPLATE_H_CONTENTS}\n")
|
||||
file(APPEND ${OPENBLAS_CONFIG_H} "#endif /* OPENBLAS_CONFIG_H */\n")
|
||||
install (FILES ${OPENBLAS_CONFIG_H} DESTINATION ${CMAKE_INSTALL_INCLUDEDIR})
|
||||
|
||||
if(NOT NOFORTRAN)
|
||||
message(STATUS "Generating f77blas.h in ${CMAKE_INSTALL_INCLUDEDIR}")
|
||||
|
||||
set(F77BLAS_H ${CMAKE_BINARY_DIR}/f77blas.h)
|
||||
file(WRITE ${F77BLAS_H} "#ifndef OPENBLAS_F77BLAS_H\n")
|
||||
file(APPEND ${F77BLAS_H} "#define OPENBLAS_F77BLAS_H\n")
|
||||
file(APPEND ${F77BLAS_H} "#include \"openblas_config.h\"\n")
|
||||
file(READ ${CMAKE_CURRENT_SOURCE_DIR}/common_interface.h COMMON_INTERFACE_H_CONTENTS)
|
||||
file(APPEND ${F77BLAS_H} "${COMMON_INTERFACE_H_CONTENTS}\n")
|
||||
file(APPEND ${F77BLAS_H} "#endif")
|
||||
install (FILES ${F77BLAS_H} DESTINATION ${CMAKE_INSTALL_INCLUDEDIR})
|
||||
endif()
|
||||
|
||||
if(NOT NO_CBLAS)
|
||||
message (STATUS "Generating cblas.h in ${CMAKE_INSTALL_INCLUDEDIR}")
|
||||
|
||||
file(READ ${CMAKE_CURRENT_SOURCE_DIR}/cblas.h CBLAS_H_CONTENTS)
|
||||
string(REPLACE "common" "openblas_config" CBLAS_H_CONTENTS_NEW "${CBLAS_H_CONTENTS}")
|
||||
file(WRITE ${CMAKE_BINARY_DIR}/cblas.tmp "${CBLAS_H_CONTENTS_NEW}")
|
||||
install (FILES ${CMAKE_BINARY_DIR}/cblas.tmp DESTINATION ${CMAKE_INSTALL_INCLUDEDIR} RENAME cblas.h)
|
||||
endif()
|
||||
|
||||
if(NOT NO_LAPACKE)
|
||||
message (STATUS "Copying LAPACKE header files to ${CMAKE_INSTALL_INCLUDEDIR}")
|
||||
add_dependencies( ${OpenBLAS_LIBNAME} genlapacke)
|
||||
FILE(GLOB_RECURSE INCLUDE_FILES "${CMAKE_CURRENT_SOURCE_DIR}/lapack-netlib/LAPACKE/*.h")
|
||||
install (FILES ${INCLUDE_FILES} DESTINATION ${CMAKE_INSTALL_INCLUDEDIR})
|
||||
|
||||
ADD_CUSTOM_TARGET(genlapacke
|
||||
COMMAND ${CMAKE_COMMAND} -E copy ${CMAKE_CURRENT_SOURCE_DIR}/lapack-netlib/LAPACKE/include/lapacke_mangling_with_flags.h.in "${CMAKE_BINARY_DIR}/lapacke_mangling.h"
|
||||
)
|
||||
install (FILES ${CMAKE_BINARY_DIR}/lapacke_mangling.h DESTINATION ${CMAKE_INSTALL_INCLUDEDIR})
|
||||
endif()
|
||||
|
||||
include(FindPkgConfig QUIET)
|
||||
if(PKG_CONFIG_FOUND)
|
||||
set(prefix ${CMAKE_INSTALL_PREFIX})
|
||||
set(libdir ${CMAKE_INSTALL_PREFIX}/lib)
|
||||
set(includedir ${CMAKE_INSTALL_PREFIX}/include)
|
||||
configure_file(${PROJECT_SOURCE_DIR}/cmake/openblas.pc.in ${PROJECT_BINARY_DIR}/openblas.pc @ONLY)
|
||||
install (FILES ${PROJECT_BINARY_DIR}/openblas.pc DESTINATION lib/pkgconfig/)
|
||||
install (FILES ${PROJECT_BINARY_DIR}/openblas.pc DESTINATION ${CMAKE_INSTALL_LIBDIR}/pkgconfig/)
|
||||
endif()
|
||||
|
||||
+2
-1
@@ -165,5 +165,6 @@ In chronological order:
|
||||
* Abdelrauf <https://github.com/quickwritereader>
|
||||
* [2017-01-01] dgemm and dtrmm kernels for IBM z13
|
||||
* [2017-02-26] ztrmm kernel for IBM z13
|
||||
|
||||
* [2017-03-13] strmm and ctrmm kernel for IBM z13
|
||||
* [2017-09-01] initial Blas Level-1,2 (double precision) for IBM z13
|
||||
|
||||
|
||||
@@ -1,4 +1,45 @@
|
||||
OpenBLAS ChangeLog
|
||||
====================================================================
|
||||
Version 0.2.20
|
||||
24-Jul-2017
|
||||
|
||||
common:
|
||||
* Improved CMake support
|
||||
* Fixed several thread race and locking bugs
|
||||
* Fixed default LAPACK optimization level
|
||||
* Updated LAPACK to 3.7.0
|
||||
* Added ReLAPACK (https://github.com/HPAC/ReLAPACK, make BUILD_RELAPACK=1)
|
||||
|
||||
POWER:
|
||||
* Optimizations for Power9
|
||||
* Fixed several Power8 assembly bugs
|
||||
|
||||
ARM:
|
||||
* New optimized Vulcan and ThunderX2T99 targets
|
||||
* Support for ARMV7 SOFT_FP ABI (make ARM_SOFTFP_ABI=1)
|
||||
* Detect all cpu cores including offline ones
|
||||
* Fix compilation with CLANG
|
||||
* Support building a shared library for Android
|
||||
|
||||
MIPS:
|
||||
* Fixed several threading issues
|
||||
* Fix compilation with CLANG
|
||||
|
||||
x86_64:
|
||||
* Detect Intel Bay Trail and Apollo Lake
|
||||
* Detect Intel Sky Lake and Kaby Lake
|
||||
* Detect Intel Knights Landing
|
||||
* Detect AMD A8, A10, A12 and Ryzen
|
||||
* Support 64bit builds with Visual Studio
|
||||
* Fix building with Intel and PGI compilers
|
||||
* Fix building with MINGW and TDM-GCC
|
||||
* Fix cmake builds for Haswell and related cpus
|
||||
* Fix building for Sandybridge with CLANG 3.9
|
||||
* Add support for the FLANG compiler
|
||||
|
||||
IBM Z:
|
||||
* New target z13 with BLAS3 optimizations
|
||||
|
||||
====================================================================
|
||||
Version 0.2.19
|
||||
1-Sep-2016
|
||||
|
||||
@@ -16,14 +16,19 @@ ifneq ($(NO_LAPACK), 1)
|
||||
SUBDIRS += lapack
|
||||
endif
|
||||
|
||||
RELA =
|
||||
ifeq ($(BUILD_RELAPACK), 1)
|
||||
RELA = re_lapack
|
||||
endif
|
||||
|
||||
LAPACK_NOOPT := $(filter-out -O0 -O1 -O2 -O3 -Ofast,$(LAPACK_FFLAGS))
|
||||
|
||||
SUBDIRS_ALL = $(SUBDIRS) test ctest utest exports benchmark ../laswp ../bench
|
||||
|
||||
.PHONY : all libs netlib test ctest shared install
|
||||
.NOTPARALLEL : all libs prof lapack-test install blas-test
|
||||
.PHONY : all libs netlib $(RELA) test ctest shared install
|
||||
.NOTPARALLEL : all libs $(RELA) prof lapack-test install blas-test
|
||||
|
||||
all :: libs netlib tests shared
|
||||
all :: libs netlib $(RELA) tests shared
|
||||
@echo
|
||||
@echo " OpenBLAS build complete. ($(LIB_COMPONENTS))"
|
||||
@echo
|
||||
@@ -81,16 +86,12 @@ endif
|
||||
|
||||
shared :
|
||||
ifndef NO_SHARED
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),Linux SunOS))
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),Linux SunOS Android))
|
||||
@$(MAKE) -C exports so
|
||||
@ln -fs $(LIBSONAME) $(LIBPREFIX).so
|
||||
@ln -fs $(LIBSONAME) $(LIBPREFIX).so.$(MAJOR_VERSION)
|
||||
endif
|
||||
ifeq ($(OSNAME), FreeBSD)
|
||||
@$(MAKE) -C exports so
|
||||
@ln -fs $(LIBSONAME) $(LIBPREFIX).so
|
||||
endif
|
||||
ifeq ($(OSNAME), NetBSD)
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),FreeBSD OpenBSD NetBSD DragonFly))
|
||||
@$(MAKE) -C exports so
|
||||
@ln -fs $(LIBSONAME) $(LIBPREFIX).so
|
||||
endif
|
||||
@@ -215,6 +216,14 @@ ifndef NO_LAPACKE
|
||||
endif
|
||||
endif
|
||||
|
||||
ifeq ($(NO_LAPACK), 1)
|
||||
re_lapack :
|
||||
|
||||
else
|
||||
re_lapack :
|
||||
@$(MAKE) -C relapack
|
||||
endif
|
||||
|
||||
prof_lapack : lapack_prebuild
|
||||
@$(MAKE) -C $(NETLIB_LAPACK_DIR) lapack_prof
|
||||
|
||||
@@ -229,7 +238,7 @@ ifndef NOFORTRAN
|
||||
-@echo "CC = $(CC)" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
-@echo "override CFLAGS = $(LAPACK_CFLAGS)" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
-@echo "ARCH = $(AR)" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
-@echo "ARCHFLAGS = -ru" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
-@echo "ARCHFLAGS = $(ARFLAGS) -ru" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
-@echo "RANLIB = $(RANLIB)" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
-@echo "LAPACKLIB = ../$(LIBNAME)" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
-@echo "TMGLIB = ../$(LIBNAME)" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
@@ -257,6 +266,7 @@ endif
|
||||
ifeq ($(BUILD_LAPACK_DEPRECATED), 1)
|
||||
-@echo "BUILD_DEPRECATED = 1" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
endif
|
||||
-@echo "LAPACKE_WITH_TMG = 1" >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
-@cat make.inc >> $(NETLIB_LAPACK_DIR)/make.inc
|
||||
endif
|
||||
|
||||
@@ -326,6 +336,7 @@ endif
|
||||
@touch $(NETLIB_LAPACK_DIR)/make.inc
|
||||
@$(MAKE) -C $(NETLIB_LAPACK_DIR) clean
|
||||
@rm -f $(NETLIB_LAPACK_DIR)/make.inc $(NETLIB_LAPACK_DIR)/lapacke/include/lapacke_mangling.h
|
||||
@$(MAKE) -C relapack clean
|
||||
@rm -f *.grd Makefile.conf_last config_last.h
|
||||
@(cd $(NETLIB_LAPACK_DIR)/TESTING && rm -f x* *.out testing_results.txt)
|
||||
@echo Done.
|
||||
|
||||
+3
-20
@@ -1,5 +1,4 @@
|
||||
#ifeq logical or
|
||||
ifeq ($(CORE), $(filter $(CORE),CORTEXA9 CORTEXA15))
|
||||
ifeq ($(CORE), $(filter $(CORE),ARMV7 CORTEXA9 CORTEXA15))
|
||||
ifeq ($(OSNAME), Android)
|
||||
CCOMMON_OPT += -mfpu=neon -march=armv7-a
|
||||
FCOMMON_OPT += -mfpu=neon -march=armv7-a
|
||||
@@ -9,28 +8,12 @@ FCOMMON_OPT += -mfpu=vfpv3 -march=armv7-a
|
||||
endif
|
||||
endif
|
||||
|
||||
ifeq ($(CORE), ARMV7)
|
||||
ifeq ($(OSNAME), Android)
|
||||
ifeq ($(ARM_SOFTFP_ABI), 1)
|
||||
CCOMMON_OPT += -mfpu=neon -march=armv7-a
|
||||
FCOMMON_OPT += -mfpu=neon -march=armv7-a
|
||||
else
|
||||
CCOMMON_OPT += -mfpu=neon -march=armv7-a -Wl,--no-warn-mismatch
|
||||
FCOMMON_OPT += -mfpu=neon -march=armv7-a -Wl,--no-warn-mismatch
|
||||
endif
|
||||
else
|
||||
CCOMMON_OPT += -mfpu=vfpv3 -march=armv7-a
|
||||
FCOMMON_OPT += -mfpu=vfpv3 -march=armv7-a
|
||||
endif
|
||||
endif
|
||||
|
||||
ifeq ($(CORE), ARMV6)
|
||||
CCOMMON_OPT += -mfpu=vfp -march=armv6
|
||||
FCOMMON_OPT += -mfpu=vfp -march=armv6
|
||||
endif
|
||||
|
||||
|
||||
ifeq ($(CORE), ARMV5)
|
||||
CCOMMON_OPT += -marm -march=armv5
|
||||
FCOMMON_OPT += -marm -march=armv5
|
||||
CCOMMON_OPT += -march=armv5
|
||||
FCOMMON_OPT += -march=armv5
|
||||
endif
|
||||
|
||||
+2
-2
@@ -20,6 +20,6 @@ FCOMMON_OPT += -mtune=thunderx -mcpu=thunderx
|
||||
endif
|
||||
|
||||
ifeq ($(CORE), THUNDERX2T99)
|
||||
CCOMMON_OPT += -mtune=vulcan -mcpu=vulcan
|
||||
FCOMMON_OPT += -mtune=vulcan -mcpu=vulcan
|
||||
CCOMMON_OPT += -mtune=thunderx2t99 -mcpu=thunderx2t99
|
||||
FCOMMON_OPT += -mtune=thunderx2t99 -mcpu=thunderx2t99
|
||||
endif
|
||||
|
||||
+11
-15
@@ -66,18 +66,13 @@ endif
|
||||
#for install shared library
|
||||
ifndef NO_SHARED
|
||||
@echo Copying the shared library to $(DESTDIR)$(OPENBLAS_LIBRARY_DIR)
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),Linux SunOS))
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),Linux SunOS Android))
|
||||
@install -pm755 $(LIBSONAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
|
||||
@cd "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)" ; \
|
||||
ln -fs $(LIBSONAME) $(LIBPREFIX).so ; \
|
||||
ln -fs $(LIBSONAME) $(LIBPREFIX).so.$(MAJOR_VERSION)
|
||||
endif
|
||||
ifeq ($(OSNAME), FreeBSD)
|
||||
@cp $(LIBSONAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
|
||||
@cd "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)" ; \
|
||||
ln -fs $(LIBSONAME) $(LIBPREFIX).so
|
||||
endif
|
||||
ifeq ($(OSNAME), NetBSD)
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),FreeBSD OpenBSD NetBSD DragonFly))
|
||||
@cp $(LIBSONAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
|
||||
@cd "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)" ; \
|
||||
ln -fs $(LIBSONAME) $(LIBPREFIX).so
|
||||
@@ -90,21 +85,22 @@ ifeq ($(OSNAME), Darwin)
|
||||
endif
|
||||
ifeq ($(OSNAME), WINNT)
|
||||
@-cp $(LIBDLLNAME) "$(DESTDIR)$(OPENBLAS_BINARY_DIR)"
|
||||
@-cp $(LIBDLLNAME).a "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
|
||||
@-cp $(IMPLIBNAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
|
||||
endif
|
||||
ifeq ($(OSNAME), CYGWIN_NT)
|
||||
@-cp $(LIBDLLNAME) "$(DESTDIR)$(OPENBLAS_BINARY_DIR)"
|
||||
@-cp $(IMPLIBNAME) "$(DESTDIR)$(OPENBLAS_LIBRARY_DIR)"
|
||||
endif
|
||||
endif
|
||||
|
||||
|
||||
#Generating openblas.pc
|
||||
@echo Generating openblas.pc in $(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)
|
||||
@echo 'libdir='$(OPENBLAS_LIBRARY_DIR) >> $(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc
|
||||
@echo 'includedir='$(OPENBLAS_INCLUDE_DIR) >> $(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc
|
||||
@echo 'version='$(VERSION) >> $(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc
|
||||
@echo 'extralib='$(EXTRALIB) >> $(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc
|
||||
@cat openblas.pc.in >> $(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc
|
||||
@echo Generating openblas.pc in "$(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)"
|
||||
@echo 'libdir='$(OPENBLAS_LIBRARY_DIR) >> "$(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc"
|
||||
@echo 'includedir='$(OPENBLAS_INCLUDE_DIR) >> "$(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc"
|
||||
@echo 'version='$(VERSION) >> "$(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc"
|
||||
@echo 'extralib='$(EXTRALIB) >> "$(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc"
|
||||
@cat openblas.pc.in >> "$(DESTDIR)$(OPENBLAS_PKGCONFIG_DIR)/openblas.pc"
|
||||
|
||||
|
||||
#Generating OpenBLASConfig.cmake
|
||||
@@ -114,7 +110,7 @@ endif
|
||||
|
||||
ifndef NO_SHARED
|
||||
#ifeq logical or
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),Linux FreeBSD NetBSD))
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),Linux FreeBSD NetBSD OpenBSD DragonFly))
|
||||
@echo "SET(OpenBLAS_LIBRARIES ${OPENBLAS_LIBRARY_DIR}/$(LIBPREFIX).so)" >> "$(DESTDIR)$(OPENBLAS_CMAKE_DIR)/$(OPENBLAS_CMAKE_CONFIG)"
|
||||
endif
|
||||
ifeq ($(OSNAME), $(filter $(OSNAME),WINNT CYGWIN_NT))
|
||||
|
||||
+1
-1
@@ -43,7 +43,7 @@ endif
|
||||
|
||||
ifeq ($(USE_MASS), 1)
|
||||
# Path to MASS libs, change it if the libs are installed at any other location
|
||||
MASSPATH = /opt/ibm/xlmass/8.1.3/lib
|
||||
MASSPATH = /opt/ibm/xlmass/8.1.5/lib
|
||||
COMMON_OPT += -mveclibabi=mass -ftree-vectorize -funsafe-math-optimizations -DUSE_MASS
|
||||
EXTRALIB += -L$(MASSPATH) -lmass -lmassvp8 -lmass_simdp8
|
||||
endif
|
||||
|
||||
@@ -29,6 +29,10 @@ ifeq ($(TARGET), P6600)
|
||||
TARGET_FLAGS = -mips64r6
|
||||
endif
|
||||
|
||||
ifeq ($(TARGET), I6500)
|
||||
TARGET_FLAGS = -mips64r6
|
||||
endif
|
||||
|
||||
all: getarch_2nd
|
||||
./getarch_2nd 0 >> $(TARGET_MAKE)
|
||||
./getarch_2nd 1 >> $(TARGET_CONF)
|
||||
|
||||
+4
-1
@@ -3,7 +3,7 @@
|
||||
#
|
||||
|
||||
# This library's version
|
||||
VERSION = 0.2.20.dev
|
||||
VERSION = 0.3.0.dev
|
||||
|
||||
# If you set the suffix, the library name will be libopenblas_$(LIBNAMESUFFIX).a
|
||||
# and libopenblas_$(LIBNAMESUFFIX).so. Meanwhile, the soname in shared library
|
||||
@@ -83,6 +83,9 @@ VERSION = 0.2.20.dev
|
||||
# Build LAPACK Deprecated functions since LAPACK 3.6.0
|
||||
BUILD_LAPACK_DEPRECATED = 1
|
||||
|
||||
# Build RecursiveLAPACK on top of LAPACK
|
||||
# BUILD_RELAPACK = 1
|
||||
|
||||
# If you want to use legacy threaded Level 3 implementation.
|
||||
# USE_SIMPLE_THREADED_LEVEL3 = 1
|
||||
|
||||
|
||||
+86
-17
@@ -17,15 +17,24 @@ NETLIB_LAPACK_DIR = $(TOPDIR)/lapack-netlib
|
||||
# http://stackoverflow.com/questions/4029274/mingw-and-make-variables
|
||||
# - Default value is 'cc' which is not always a valid command (e.g. MinGW).
|
||||
ifeq ($(origin CC),default)
|
||||
|
||||
# Check if $(CC) refers to a valid command and set the value to gcc if not
|
||||
ifneq ($(findstring cmd.exe,$(SHELL)),)
|
||||
ifeq ($(shell where $(CC) 2>NUL),)
|
||||
CC = gcc
|
||||
# Change the default compile to clang on Mac OSX.
|
||||
# http://stackoverflow.com/questions/714100/os-detecting-makefile
|
||||
UNAME_S := $(shell uname -s)
|
||||
ifeq ($(UNAME_S),Darwin)
|
||||
CC = clang
|
||||
# EXTRALIB += -Wl,-no_compact_unwind
|
||||
endif
|
||||
endif
|
||||
else # POSIX-ish
|
||||
ifeq ($(shell command -v $(CC) 2>/dev/null),)
|
||||
ifeq ($(shell uname -s),Darwin)
|
||||
CC = clang
|
||||
# EXTRALIB += -Wl,-no_compact_unwind
|
||||
else
|
||||
CC = gcc
|
||||
endif # Darwin
|
||||
endif # CC exists
|
||||
endif # Shell is sane
|
||||
|
||||
endif # CC is set to default
|
||||
|
||||
# Default Fortran compiler (FC) is selected by f_check.
|
||||
|
||||
@@ -68,6 +77,9 @@ endif
|
||||
ifeq ($(TARGET), EXCAVATOR)
|
||||
GETARCH_FLAGS := -DFORCE_BARCELONA
|
||||
endif
|
||||
ifeq ($(TARGET), ZEN)
|
||||
GETARCH_FLAGS := -DFORCE_BARCELONA
|
||||
endif
|
||||
endif
|
||||
|
||||
|
||||
@@ -98,6 +110,9 @@ endif
|
||||
ifeq ($(TARGET_CORE), EXCAVATOR)
|
||||
GETARCH_FLAGS := -DFORCE_BARCELONA
|
||||
endif
|
||||
ifeq ($(TARGET_CORE), ZEN)
|
||||
GETARCH_FLAGS := -DFORCE_BARCELONA
|
||||
endif
|
||||
endif
|
||||
|
||||
|
||||
@@ -175,6 +190,7 @@ endif
|
||||
|
||||
ifeq ($(NUM_THREADS), 1)
|
||||
override USE_THREAD = 0
|
||||
override USE_OPENMP = 0
|
||||
endif
|
||||
|
||||
ifdef USE_THREAD
|
||||
@@ -223,7 +239,7 @@ endif
|
||||
MD5SUM = md5 -r
|
||||
endif
|
||||
|
||||
ifeq ($(OSNAME), FreeBSD)
|
||||
ifneq (,$(findstring $(OSNAME), FreeBSD OpenBSD DragonFly))
|
||||
MD5SUM = md5 -r
|
||||
endif
|
||||
|
||||
@@ -236,6 +252,10 @@ EXTRALIB += -lm
|
||||
NO_EXPRECISION = 1
|
||||
endif
|
||||
|
||||
ifeq ($(OSNAME), Android)
|
||||
EXTRALIB += -lm
|
||||
endif
|
||||
|
||||
ifeq ($(OSNAME), AIX)
|
||||
EXTRALIB += -lm
|
||||
endif
|
||||
@@ -293,6 +313,7 @@ endif
|
||||
ifeq ($(OSNAME), CYGWIN_NT)
|
||||
NEED_PIC = 0
|
||||
NO_EXPRECISION = 1
|
||||
OS_CYGWIN_NT = 1
|
||||
endif
|
||||
|
||||
ifneq ($(OSNAME), WINNT)
|
||||
@@ -408,7 +429,6 @@ CCOMMON_OPT += -fopenmp
|
||||
endif
|
||||
|
||||
ifeq ($(C_COMPILER), CLANG)
|
||||
$(error OpenBLAS: Clang didn't support OpenMP yet.)
|
||||
CCOMMON_OPT += -fopenmp
|
||||
endif
|
||||
|
||||
@@ -443,12 +463,13 @@ ifneq ($(NO_AVX), 1)
|
||||
DYNAMIC_CORE += SANDYBRIDGE BULLDOZER PILEDRIVER STEAMROLLER EXCAVATOR
|
||||
endif
|
||||
ifneq ($(NO_AVX2), 1)
|
||||
DYNAMIC_CORE += HASWELL
|
||||
DYNAMIC_CORE += HASWELL ZEN
|
||||
endif
|
||||
endif
|
||||
|
||||
# If DYNAMIC_CORE is not set, DYNAMIC_ARCH cannot do anything, so force it to empty
|
||||
ifndef DYNAMIC_CORE
|
||||
DYNAMIC_ARCH =
|
||||
override DYNAMIC_ARCH=
|
||||
endif
|
||||
endif
|
||||
|
||||
@@ -480,12 +501,18 @@ BINARY_DEFINED = 1
|
||||
CCOMMON_OPT += -marm
|
||||
FCOMMON_OPT += -marm
|
||||
|
||||
# If softfp abi is mentioned on the command line, force it.
|
||||
ifeq ($(ARM_SOFTFP_ABI), 1)
|
||||
CCOMMON_OPT += -mfloat-abi=softfp -DARM_SOFTFP_ABI
|
||||
FCOMMON_OPT += -mfloat-abi=softfp -DARM_SOFTFP_ABI
|
||||
CCOMMON_OPT += -mfloat-abi=softfp
|
||||
FCOMMON_OPT += -mfloat-abi=softfp
|
||||
endif
|
||||
|
||||
ifeq ($(OSNAME), Android)
|
||||
ifeq ($(ARM_SOFTFP_ABI), 1)
|
||||
EXTRALIB += -lm
|
||||
else
|
||||
CCOMMON_OPT += -mfloat-abi=hard
|
||||
FCOMMON_OPT += -mfloat-abi=hard
|
||||
EXTRALIB += -Wl,-lm_hard
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
|
||||
@@ -552,6 +579,11 @@ CCOMMON_OPT += -mips64r6 -mnan=2008 -mtune=p6600 $(MSA_FLAGS)
|
||||
FCOMMON_OPT += -mips64r6 -mnan=2008 -mtune=p6600 $(MSA_FLAGS)
|
||||
endif
|
||||
|
||||
ifeq ($(CORE), I6500)
|
||||
CCOMMON_OPT += -mips64r6 -mnan=2008 -mtune=i6400 $(MSA_FLAGS)
|
||||
FCOMMON_OPT += -mips64r6 -mnan=2008 -mtune=i6400 $(MSA_FLAGS)
|
||||
endif
|
||||
|
||||
ifeq ($(OSNAME), AIX)
|
||||
BINARY_DEFINED = 1
|
||||
endif
|
||||
@@ -559,12 +591,14 @@ endif
|
||||
endif
|
||||
|
||||
ifndef BINARY_DEFINED
|
||||
ifneq ($(OSNAME), AIX)
|
||||
ifdef BINARY64
|
||||
CCOMMON_OPT += -m64
|
||||
else
|
||||
CCOMMON_OPT += -m32
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
|
||||
endif
|
||||
|
||||
@@ -588,10 +622,28 @@ endif
|
||||
# Fortran Compiler dependent settings
|
||||
#
|
||||
|
||||
ifeq ($(F_COMPILER), FLANG)
|
||||
CCOMMON_OPT += -DF_INTERFACE_FLANG
|
||||
ifdef BINARY64
|
||||
ifdef INTERFACE64
|
||||
ifneq ($(INTERFACE64), 0)
|
||||
FCOMMON_OPT += -i8
|
||||
endif
|
||||
endif
|
||||
FCOMMON_OPT += -Wall
|
||||
else
|
||||
FCOMMON_OPT += -Wall
|
||||
endif
|
||||
ifeq ($(USE_OPENMP), 1)
|
||||
FCOMMON_OPT += -fopenmp
|
||||
endif
|
||||
endif
|
||||
|
||||
ifeq ($(F_COMPILER), G77)
|
||||
CCOMMON_OPT += -DF_INTERFACE_G77
|
||||
FCOMMON_OPT += -Wall
|
||||
ifndef NO_BINARY_MODE
|
||||
ifneq ($(OSNAME), AIX)
|
||||
ifdef BINARY64
|
||||
FCOMMON_OPT += -m64
|
||||
else
|
||||
@@ -599,10 +651,12 @@ FCOMMON_OPT += -m32
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
|
||||
ifeq ($(F_COMPILER), G95)
|
||||
CCOMMON_OPT += -DF_INTERFACE_G95
|
||||
FCOMMON_OPT += -Wall
|
||||
ifneq ($(OSNAME), AIX)
|
||||
ifndef NO_BINARY_MODE
|
||||
ifdef BINARY64
|
||||
FCOMMON_OPT += -m64
|
||||
@@ -611,6 +665,7 @@ FCOMMON_OPT += -m32
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
|
||||
ifeq ($(F_COMPILER), GFORTRAN)
|
||||
CCOMMON_OPT += -DF_INTERFACE_GFORT
|
||||
@@ -631,16 +686,20 @@ FCOMMON_OPT += -mabi=32
|
||||
endif
|
||||
else
|
||||
ifdef BINARY64
|
||||
ifneq ($(OSNAME), AIX)
|
||||
FCOMMON_OPT += -m64
|
||||
endif
|
||||
ifdef INTERFACE64
|
||||
ifneq ($(INTERFACE64), 0)
|
||||
FCOMMON_OPT += -fdefault-integer-8
|
||||
endif
|
||||
endif
|
||||
else
|
||||
ifneq ($(OSNAME), AIX)
|
||||
FCOMMON_OPT += -m32
|
||||
endif
|
||||
endif
|
||||
endif
|
||||
ifeq ($(USE_OPENMP), 1)
|
||||
FCOMMON_OPT += -fopenmp
|
||||
endif
|
||||
@@ -915,9 +974,15 @@ SYMBOLSUFFIX =
|
||||
endif
|
||||
|
||||
ifndef LIBNAMESUFFIX
|
||||
LIBPREFIX = lib$(SYMBOLPREFIX)openblas$(SYMBOLSUFFIX)
|
||||
LIBNAMEBASE = $(SYMBOLPREFIX)openblas$(SYMBOLSUFFIX)
|
||||
else
|
||||
LIBPREFIX = lib$(SYMBOLPREFIX)openblas$(SYMBOLSUFFIX)_$(LIBNAMESUFFIX)
|
||||
LIBNAMEBASE = $(SYMBOLPREFIX)openblas$(SYMBOLSUFFIX)_$(LIBNAMESUFFIX)
|
||||
endif
|
||||
|
||||
ifeq ($(OSNAME), CYGWIN_NT)
|
||||
LIBPREFIX = cyg$(LIBNAMEBASE)
|
||||
else
|
||||
LIBPREFIX = lib$(LIBNAMEBASE)
|
||||
endif
|
||||
|
||||
KERNELDIR = $(TOPDIR)/kernel/$(ARCH)
|
||||
@@ -1076,6 +1141,7 @@ endif
|
||||
|
||||
|
||||
LIBDLLNAME = $(LIBPREFIX).dll
|
||||
IMPLIBNAME = lib$(LIBNAMEBASE).dll.a
|
||||
LIBSONAME = $(LIBNAME:.$(LIBSUFFIX)=.so)
|
||||
LIBDYNNAME = $(LIBNAME:.$(LIBSUFFIX)=.dylib)
|
||||
LIBDEFNAME = $(LIBNAME:.$(LIBSUFFIX)=.def)
|
||||
@@ -1096,6 +1162,9 @@ LIB_COMPONENTS += LAPACK
|
||||
ifneq ($(NO_LAPACKE), 1)
|
||||
LIB_COMPONENTS += LAPACKE
|
||||
endif
|
||||
ifeq ($(BUILD_RELAPACK), 1)
|
||||
LIB_COMPONENTS += ReLAPACK
|
||||
endif
|
||||
endif
|
||||
|
||||
ifeq ($(ONLY_CBLAS), 1)
|
||||
|
||||
@@ -2,175 +2,222 @@
|
||||
|
||||
[](https://gitter.im/xianyi/OpenBLAS?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
|
||||
|
||||
Travis CI: [](https://travis-ci.org/xianyi/OpenBLAS)
|
||||
Travis CI: [](https://travis-ci.org/xianyi/OpenBLAS)
|
||||
|
||||
AppVeyor: [](https://ci.appveyor.com/project/xianyi/openblas/branch/develop)
|
||||
|
||||
## Introduction
|
||||
|
||||
OpenBLAS is an optimized BLAS library based on GotoBLAS2 1.13 BSD version.
|
||||
|
||||
Please read the documents on OpenBLAS wiki pages <http://github.com/xianyi/OpenBLAS/wiki>.
|
||||
Please read the documentation on the OpenBLAS wiki pages: <http://github.com/xianyi/OpenBLAS/wiki>.
|
||||
|
||||
## Binary Packages
|
||||
We provide binary packages for the following platform.
|
||||
|
||||
We provide official binary packages for the following platform:
|
||||
|
||||
* Windows x86/x86_64
|
||||
|
||||
You can download them from [file hosting on sourceforge.net](https://sourceforge.net/projects/openblas/files/).
|
||||
|
||||
## Installation from Source
|
||||
Download from project homepage. http://xianyi.github.com/OpenBLAS/
|
||||
|
||||
Or, check out codes from git://github.com/xianyi/OpenBLAS.git
|
||||
Download from project homepage, http://xianyi.github.com/OpenBLAS/, or check out the code
|
||||
using Git from https://github.com/xianyi/OpenBLAS.git.
|
||||
|
||||
### Dependencies
|
||||
|
||||
Building OpenBLAS requires the following to be installed:
|
||||
|
||||
* GNU Make
|
||||
* A C compiler, e.g. GCC or Clang
|
||||
* A Fortran compiler (optional, for LAPACK)
|
||||
* IBM MASS (optional, see below)
|
||||
|
||||
### Normal compile
|
||||
* type "make" to detect the CPU automatically.
|
||||
or
|
||||
* type "make TARGET=xxx" to set target CPU, e.g. "make TARGET=NEHALEM". The full target list is in file TargetList.txt.
|
||||
|
||||
Simply invoking `make` (or `gmake` on BSD) will detect the CPU automatically.
|
||||
To set a specific target CPU, use `make TARGET=xxx`, e.g. `make TARGET=NEHALEM`.
|
||||
The full target list is in the file `TargetList.txt`.
|
||||
|
||||
### Cross compile
|
||||
Please set CC and FC with the cross toolchains. Then, set HOSTCC with your host C compiler. At last, set TARGET explicitly.
|
||||
|
||||
Set `CC` and `FC` to point to the cross toolchains, and set `HOSTCC` to your host C compiler.
|
||||
The target must be specified explicitly when cross compiling.
|
||||
|
||||
Examples:
|
||||
|
||||
On X86 box, compile this library for loongson3a CPU.
|
||||
* On an x86 box, compile this library for a loongson3a CPU:
|
||||
```sh
|
||||
make BINARY=64 CC=mips64el-unknown-linux-gnu-gcc FC=mips64el-unknown-linux-gnu-gfortran HOSTCC=gcc TARGET=LOONGSON3A
|
||||
```
|
||||
|
||||
make BINARY=64 CC=mips64el-unknown-linux-gnu-gcc FC=mips64el-unknown-linux-gnu-gfortran HOSTCC=gcc TARGET=LOONGSON3A
|
||||
|
||||
On X86 box, compile this library for loongson3a CPU with loongcc (based on Open64) compiler.
|
||||
|
||||
make CC=loongcc FC=loongf95 HOSTCC=gcc TARGET=LOONGSON3A CROSS=1 CROSS_SUFFIX=mips64el-st-linux-gnu- NO_LAPACKE=1 NO_SHARED=1 BINARY=32
|
||||
* On an x86 box, compile this library for a loongson3a CPU with loongcc (based on Open64) compiler:
|
||||
```sh
|
||||
make CC=loongcc FC=loongf95 HOSTCC=gcc TARGET=LOONGSON3A CROSS=1 CROSS_SUFFIX=mips64el-st-linux-gnu- NO_LAPACKE=1 NO_SHARED=1 BINARY=32
|
||||
```
|
||||
|
||||
### Debug version
|
||||
|
||||
make DEBUG=1
|
||||
A debug version can be built using `make DEBUG=1`.
|
||||
|
||||
### Compile with MASS Support on Power CPU (Optional dependency)
|
||||
### Compile with MASS support on Power CPU (optional)
|
||||
|
||||
[IBM MASS](http://www-01.ibm.com/software/awdtools/mass/linux/mass-linux.html) library consists of a set of mathematical functions for C, C++, and
|
||||
Fortran-language applications that are tuned for optimum performance on POWER architectures. OpenBLAS with MASS requires 64-bit, little-endian OS on POWER.
|
||||
The library can be installed as below -
|
||||
The [IBM MASS](http://www-01.ibm.com/software/awdtools/mass/linux/mass-linux.html) library
|
||||
consists of a set of mathematical functions for C, C++, and Fortran applications that are
|
||||
are tuned for optimum performance on POWER architectures.
|
||||
OpenBLAS with MASS requires a 64-bit, little-endian OS on POWER.
|
||||
The library can be installed as shown:
|
||||
|
||||
* On Ubuntu:
|
||||
* On Ubuntu:
|
||||
```sh
|
||||
wget -q http://public.dhe.ibm.com/software/server/POWER/Linux/xl-compiler/eval/ppc64le/ubuntu/public.gpg -O- | sudo apt-key add -
|
||||
echo "deb http://public.dhe.ibm.com/software/server/POWER/Linux/xl-compiler/eval/ppc64le/ubuntu/ trusty main" | sudo tee /etc/apt/sources.list.d/ibm-xl-compiler-eval.list
|
||||
sudo apt-get update
|
||||
sudo apt-get install libxlmass-devel.8.1.5
|
||||
```
|
||||
|
||||
wget -q http://public.dhe.ibm.com/software/server/POWER/Linux/xl-compiler/eval/ppc64le/ubuntu/public.gpg -O- | sudo apt-key add -
|
||||
echo "deb http://public.dhe.ibm.com/software/server/POWER/Linux/xl-compiler/eval/ppc64le/ubuntu/ trusty main" | sudo tee /etc/apt/sources.list.d/ibm-xl-compiler-eval.list
|
||||
sudo apt-get update
|
||||
sudo apt-get install libxlmass-devel.8.1.3
|
||||
* On RHEL/CentOS:
|
||||
```sh
|
||||
wget http://public.dhe.ibm.com/software/server/POWER/Linux/xl-compiler/eval/ppc64le/rhel7/repodata/repomd.xml.key
|
||||
sudo rpm --import repomd.xml.key
|
||||
wget http://public.dhe.ibm.com/software/server/POWER/Linux/xl-compiler/eval/ppc64le/rhel7/ibm-xl-compiler-eval.repo
|
||||
sudo cp ibm-xl-compiler-eval.repo /etc/yum.repos.d/
|
||||
sudo yum install libxlmass-devel.8.1.5
|
||||
```
|
||||
|
||||
* On RHEL/CentOS:
|
||||
After installing the MASS library, compile OpenBLAS with `USE_MASS=1`.
|
||||
For example, to compile on Power8 with MASS support: `make USE_MASS=1 TARGET=POWER8`.
|
||||
|
||||
wget http://public.dhe.ibm.com/software/server/POWER/Linux/xl-compiler/eval/ppc64le/rhel7/repodata/repomd.xml.key
|
||||
sudo rpm --import repomd.xml.key
|
||||
wget http://public.dhe.ibm.com/software/server/POWER/Linux/xl-compiler/eval/ppc64le/rhel7/ibm-xl-compiler-eval.repo
|
||||
sudo cp ibm-xl-compiler-eval.repo /etc/yum.repos.d/
|
||||
sudo yum install libxlmass-devel.8.1.3
|
||||
### Install to a specific directory (optional)
|
||||
|
||||
After installing MASS library, compile openblas with USE_MASS=1.
|
||||
Use `PREFIX=` when invoking `make`, for example
|
||||
|
||||
Example:
|
||||
```sh
|
||||
make install PREFIX=your_installation_directory
|
||||
```
|
||||
|
||||
Compiling on Power8 with MASS support -
|
||||
The default installation directory is `/opt/OpenBLAS`.
|
||||
|
||||
make USE_MASS=1 TARGET=POWER8
|
||||
## Supported CPUs and Operating Systems
|
||||
|
||||
### Install to the directory (optional)
|
||||
Please read `GotoBLAS_01Readme.txt`.
|
||||
|
||||
Example:
|
||||
### Additional supported CPUs
|
||||
|
||||
make install PREFIX=your_installation_directory
|
||||
#### x86/x86-64
|
||||
|
||||
The default directory is /opt/OpenBLAS
|
||||
|
||||
## Support CPU & OS
|
||||
Please read GotoBLAS_01Readme.txt
|
||||
|
||||
### Additional support CPU:
|
||||
|
||||
#### x86/x86-64:
|
||||
- **Intel Xeon 56xx (Westmere)**: Used GotoBLAS2 Nehalem codes.
|
||||
- **Intel Sandy Bridge**: Optimized Level-3 and Level-2 BLAS with AVX on x86-64.
|
||||
- **Intel Haswell**: Optimized Level-3 and Level-2 BLAS with AVX2 and FMA on x86-64.
|
||||
- **AMD Bobcat**: Used GotoBLAS2 Barcelona codes.
|
||||
- **AMD Bulldozer**: x86-64 ?GEMM FMA4 kernels. (Thank Werner Saar)
|
||||
- **AMD Bulldozer**: x86-64 ?GEMM FMA4 kernels. (Thanks to Werner Saar)
|
||||
- **AMD PILEDRIVER**: Uses Bulldozer codes with some optimizations.
|
||||
- **AMD STEAMROLLER**: Uses Bulldozer codes with some optimizations.
|
||||
|
||||
#### MIPS64:
|
||||
#### MIPS64
|
||||
|
||||
- **ICT Loongson 3A**: Optimized Level-3 BLAS and the part of Level-1,2.
|
||||
- **ICT Loongson 3B**: Experimental
|
||||
|
||||
#### ARM:
|
||||
- **ARMV6**: Optimized BLAS for vfpv2 and vfpv3-d16 ( e.g. BCM2835, Cortex M0+ )
|
||||
- **ARMV7**: Optimized BLAS for vfpv3-d32 ( e.g. Cortex A8, A9 and A15 )
|
||||
#### ARM
|
||||
|
||||
#### ARM64:
|
||||
- **ARMV8**: Experimental
|
||||
- **ARMv6**: Optimized BLAS for vfpv2 and vfpv3-d16 (e.g. BCM2835, Cortex M0+)
|
||||
- **ARMv7**: Optimized BLAS for vfpv3-d32 (e.g. Cortex A8, A9 and A15)
|
||||
|
||||
#### ARM64
|
||||
|
||||
- **ARMv8**: Experimental
|
||||
- **ARM Cortex-A57**: Experimental
|
||||
|
||||
#### IBM zEnterprise System:
|
||||
- **Z13**: blas3 for double
|
||||
|
||||
#### PPC/PPC64
|
||||
|
||||
- **POWER8**: Optmized Level-3 BLAS and some Level-1, only with `USE_OPENMP=1`
|
||||
|
||||
#### IBM zEnterprise System
|
||||
|
||||
- **Z13**: Optimized Level-3 BLAS and Level-1,2 (double precision)
|
||||
|
||||
### Supported OS
|
||||
|
||||
### Support OS:
|
||||
- **GNU/Linux**
|
||||
- **MingWin or Visual Studio(CMake)/Windows**: Please read <https://github.com/xianyi/OpenBLAS/wiki/How-to-use-OpenBLAS-in-Microsoft-Visual-Studio>.
|
||||
- **Darwin/Mac OS X**: Experimental. Although GotoBLAS2 supports Darwin, we are the beginner on Mac OS X.
|
||||
- **FreeBSD**: Supported by community. We didn't test the library on this OS.
|
||||
- **Android**: Supported by community. Please read <https://github.com/xianyi/OpenBLAS/wiki/How-to-build-OpenBLAS-for-Android>.
|
||||
- **MinGW or Visual Studio (CMake)/Windows**: Please read <https://github.com/xianyi/OpenBLAS/wiki/How-to-use-OpenBLAS-in-Microsoft-Visual-Studio>.
|
||||
- **Darwin/macOS**: Experimental. Although GotoBLAS2 supports Darwin, we are not macOS experts.
|
||||
- **FreeBSD**: Supported by the community. We don't actively test the library on this OS.
|
||||
- **OpenBSD**: Supported by the community. We don't actively test the library on this OS.
|
||||
- **DragonFly BSD**: Supported by the community. We don't actively test the library on this OS.
|
||||
- **Android**: Supported by the community. Please read <https://github.com/xianyi/OpenBLAS/wiki/How-to-build-OpenBLAS-for-Android>.
|
||||
|
||||
## Usages
|
||||
Link with libopenblas.a or -lopenblas for shared library.
|
||||
## Usage
|
||||
|
||||
### Set the number of threads with environment variables.
|
||||
Statically link with `libopenblas.a` or dynamically link with `-lopenblas` if OpenBLAS was
|
||||
compiled as a shared library.
|
||||
|
||||
Examples:
|
||||
### Setting the number of threads using environment variables
|
||||
|
||||
export OPENBLAS_NUM_THREADS=4
|
||||
Environment variables are used to specify a maximum number of threads.
|
||||
For example,
|
||||
|
||||
or
|
||||
```sh
|
||||
export OPENBLAS_NUM_THREADS=4
|
||||
export GOTO_NUM_THREADS=4
|
||||
export OMP_NUM_THREADS=4
|
||||
```
|
||||
|
||||
export GOTO_NUM_THREADS=4
|
||||
The priorities are `OPENBLAS_NUM_THREADS` > `GOTO_NUM_THREADS` > `OMP_NUM_THREADS`.
|
||||
|
||||
or
|
||||
If you compile this library with `USE_OPENMP=1`, you should set the `OMP_NUM_THREADS`
|
||||
environment variable; OpenBLAS ignores `OPENBLAS_NUM_THREADS` and `GOTO_NUM_THREADS` when
|
||||
compiled with `USE_OPENMP=1`.
|
||||
|
||||
export OMP_NUM_THREADS=4
|
||||
### Setting the number of threads at runtime
|
||||
|
||||
The priorities are OPENBLAS_NUM_THREADS > GOTO_NUM_THREADS > OMP_NUM_THREADS.
|
||||
We provide the following functions to control the number of threads at runtime:
|
||||
|
||||
If you compile this lib with USE_OPENMP=1, you should set OMP_NUM_THREADS environment variable. OpenBLAS ignores OPENBLAS_NUM_THREADS and GOTO_NUM_THREADS with USE_OPENMP=1.
|
||||
```c
|
||||
void goto_set_num_threads(int num_threads);
|
||||
void openblas_set_num_threads(int num_threads);
|
||||
```
|
||||
|
||||
### Set the number of threads on runtime.
|
||||
If you compile this library with `USE_OPENMP=1`, you should use the above functions too.
|
||||
|
||||
We provided the below functions to control the number of threads on runtime.
|
||||
## Reporting bugs
|
||||
|
||||
void goto_set_num_threads(int num_threads);
|
||||
|
||||
void openblas_set_num_threads(int num_threads);
|
||||
|
||||
If you compile this lib with USE_OPENMP=1, you should use the above functions, too.
|
||||
|
||||
## Report Bugs
|
||||
Please add a issue in https://github.com/xianyi/OpenBLAS/issues
|
||||
Please submit an issue in https://github.com/xianyi/OpenBLAS/issues.
|
||||
|
||||
## Contact
|
||||
|
||||
* OpenBLAS users mailing list: https://groups.google.com/forum/#!forum/openblas-users
|
||||
* OpenBLAS developers mailing list: https://groups.google.com/forum/#!forum/openblas-dev
|
||||
|
||||
## ChangeLog
|
||||
Please see Changelog.txt to obtain the differences between GotoBLAS2 1.13 BSD version.
|
||||
## Change log
|
||||
|
||||
Please see Changelog.txt to view the differences between OpenBLAS and GotoBLAS2 1.13 BSD version.
|
||||
|
||||
## Troubleshooting
|
||||
* Please read [Faq](https://github.com/xianyi/OpenBLAS/wiki/Faq) at first.
|
||||
* Please use gcc version 4.6 and above to compile Sandy Bridge AVX kernels on Linux/MingW/BSD.
|
||||
* Please use Clang version 3.1 and above to compile the library on Sandy Bridge microarchitecture. The Clang 3.0 will generate the wrong AVX binary code.
|
||||
* The number of CPUs/Cores should less than or equal to 256. On Linux x86_64(amd64), there is experimental support for up to 1024 CPUs/Cores and 128 numa nodes if you build the library with BIGNUMA=1.
|
||||
* OpenBLAS does not set processor affinity by default. On Linux, you can enable processor affinity by commenting the line NO_AFFINITY=1 in Makefile.rule. But this may cause [the conflict with R parallel](https://stat.ethz.ch/pipermail/r-sig-hpc/2012-April/001348.html).
|
||||
* On Loongson 3A. make test would be failed because of pthread_create error. The error code is EAGAIN. However, it will be OK when you run the same testcase on shell.
|
||||
|
||||
* Please read the [FAQ](https://github.com/xianyi/OpenBLAS/wiki/Faq) first.
|
||||
* Please use GCC version 4.6 and above to compile Sandy Bridge AVX kernels on Linux/MinGW/BSD.
|
||||
* Please use Clang version 3.1 and above to compile the library on Sandy Bridge microarchitecture.
|
||||
Clang 3.0 will generate the wrong AVX binary code.
|
||||
* The number of CPUs/cores should less than or equal to 256. On Linux `x86_64` (`amd64`),
|
||||
there is experimental support for up to 1024 CPUs/cores and 128 numa nodes if you build
|
||||
the library with `BIGNUMA=1`.
|
||||
* OpenBLAS does not set processor affinity by default.
|
||||
On Linux, you can enable processor affinity by commenting out the line `NO_AFFINITY=1` in
|
||||
Makefile.rule. However, note that this may cause
|
||||
[a conflict with R parallel](https://stat.ethz.ch/pipermail/r-sig-hpc/2012-April/001348.html).
|
||||
* On Loongson 3A, `make test` may fail with a `pthread_create` error (`EAGAIN`).
|
||||
However, it will be okay when you run the same test case on the shell.
|
||||
|
||||
## Contributing
|
||||
1. [Check for open issues](https://github.com/xianyi/OpenBLAS/issues) or open a fresh issue to start a discussion around a feature idea or a bug.
|
||||
1. Fork the [OpenBLAS](https://github.com/xianyi/OpenBLAS) repository to start making your changes.
|
||||
1. Write a test which shows that the bug was fixed or that the feature works as expected.
|
||||
1. Send a pull request. Make sure to add yourself to `CONTRIBUTORS.md`.
|
||||
|
||||
1. [Check for open issues](https://github.com/xianyi/OpenBLAS/issues) or open a fresh issue
|
||||
to start a discussion around a feature idea or a bug.
|
||||
2. Fork the [OpenBLAS](https://github.com/xianyi/OpenBLAS) repository to start making your changes.
|
||||
3. Write a test which shows that the bug was fixed or that the feature works as expected.
|
||||
4. Send a pull request. Make sure to add yourself to `CONTRIBUTORS.md`.
|
||||
|
||||
## Donation
|
||||
|
||||
Please read [this wiki page](https://github.com/xianyi/OpenBLAS/wiki/Donation).
|
||||
|
||||
@@ -34,6 +34,7 @@ BULLDOZER
|
||||
PILEDRIVER
|
||||
STEAMROLLER
|
||||
EXCAVATOR
|
||||
ZEN
|
||||
|
||||
c)VIA CPU:
|
||||
SSE_GENERIC
|
||||
@@ -62,6 +63,7 @@ LOONGSON3A
|
||||
LOONGSON3B
|
||||
I6400
|
||||
P6600
|
||||
I6500
|
||||
|
||||
5.IA64 CPU:
|
||||
ITANIUM2
|
||||
@@ -84,3 +86,6 @@ VULCAN
|
||||
THUNDERX
|
||||
THUNDERX2T99
|
||||
|
||||
9.System Z:
|
||||
ZARCH_GENERIC
|
||||
Z13
|
||||
|
||||
@@ -14,6 +14,20 @@ Please build OpenBLAS with larger `NUM_THREADS`. For example, `make
|
||||
NUM_THREADS=32` or `make NUM_THREADS=64`. In `Makefile.system`, we will set
|
||||
`MAX_CPU_NUMBER=NUM_THREADS`.
|
||||
|
||||
Despite its name, and due to the use of memory buffers in functions like SGEMM,
|
||||
the setting of NUM_THREADS can be relevant even for a single-threaded build
|
||||
of OpenBLAS, if such functions get called by multiple threads of a program
|
||||
that uses OpenBLAS. In some cases, the affected code may simply crash or throw
|
||||
a segmentation fault without displaying the above warning first.
|
||||
|
||||
Note that the number of threads used at runtime can be altered to differ from the
|
||||
value NUM_THREADS was set to at build time. At runtime, the actual number of
|
||||
threads can be set anywhere from 1 to the build's NUM_THREADS (note however,
|
||||
that this does not change the number of memory buffers that will be allocated,
|
||||
which is set at build time). The number of threads for a process can be set by
|
||||
using the mechanisms described below.
|
||||
|
||||
|
||||
#### How can I use OpenBLAS in multi-threaded applications?
|
||||
|
||||
If your application is already multi-threaded, it will conflict with OpenBLAS
|
||||
|
||||
+37
-15
@@ -5,6 +5,8 @@ version: 0.2.19.{build}
|
||||
platform:
|
||||
- x64
|
||||
|
||||
os: Visual Studio 2017
|
||||
|
||||
configuration: Release
|
||||
|
||||
clone_folder: c:\projects\OpenBLAS
|
||||
@@ -12,33 +14,53 @@ clone_folder: c:\projects\OpenBLAS
|
||||
init:
|
||||
- git config --global core.autocrlf input
|
||||
|
||||
build:
|
||||
project: OpenBLAS.sln
|
||||
|
||||
clone_depth: 5
|
||||
|
||||
#branches to build
|
||||
branches:
|
||||
only:
|
||||
- master
|
||||
- develop
|
||||
- cmake
|
||||
|
||||
skip_tags: true
|
||||
|
||||
matrix:
|
||||
fast_finish: true
|
||||
fast_finish: false
|
||||
|
||||
skip_commits:
|
||||
# Add [av skip] to commit messages
|
||||
message: /\[av skip\]/
|
||||
|
||||
environment:
|
||||
global:
|
||||
CONDA_INSTALL_LOCN: C:\\Miniconda36-x64
|
||||
matrix:
|
||||
- COMPILER: clang-cl
|
||||
WITH_FORTRAN: yes
|
||||
- COMPILER: clang-cl
|
||||
DYNAMIC_ARCH: ON
|
||||
WITH_FORTRAN: no
|
||||
- COMPILER: cl
|
||||
|
||||
install:
|
||||
- if [%COMPILER%]==[clang-cl] call %CONDA_INSTALL_LOCN%\Scripts\activate.bat
|
||||
- if [%COMPILER%]==[clang-cl] conda config --add channels conda-forge --force
|
||||
- if [%COMPILER%]==[clang-cl] conda install --yes --quiet clangdev cmake
|
||||
|
||||
- if [%WITH_FORTRAN%]==[no] conda install --yes --quiet ninja
|
||||
- if [%WITH_FORTRAN%]==[yes] conda install --yes --quiet -c isuruf kitware-ninja
|
||||
- if [%WITH_FORTRAN%]==[yes] conda install --yes --quiet flang
|
||||
|
||||
- if [%COMPILER%]==[clang-cl] call "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvarsall.bat" x64
|
||||
- if [%COMPILER%]==[clang-cl] set "LIB=%CONDA_INSTALL_LOCN%\Library\lib;%LIB%"
|
||||
- if [%COMPILER%]==[clang-cl] set "CPATH=%CONDA_INSTALL_LOCN%\Library\include;%CPATH%"
|
||||
|
||||
before_build:
|
||||
- echo Running cmake...
|
||||
- cd c:\projects\OpenBLAS
|
||||
- cmake -G "Visual Studio 12 Win64" .
|
||||
- ps: if (-Not (Test-Path .\build)) { mkdir build }
|
||||
- cd build
|
||||
- if [%COMPILER%]==[cl] cmake -G "Visual Studio 15 2017 Win64" ..
|
||||
- if [%WITH_FORTRAN%]==[no] cmake -G "Ninja" -DCMAKE_CXX_COMPILER=clang-cl -DCMAKE_C_COMPILER=clang-cl ..
|
||||
- if [%WITH_FORTRAN%]==[yes] cmake -G "Ninja" -DCMAKE_CXX_COMPILER=clang-cl -DCMAKE_C_COMPILER=clang-cl -DCMAKE_Fortran_COMPILER=flang -DBUILD_WITHOUT_LAPACK=no -DNOFORTRAN=0 ..
|
||||
- if [%DYNAMIC_ARCH%]==[ON] cmake -DDYNAMIC_ARCH=ON ..
|
||||
|
||||
build_script:
|
||||
- cmake --build .
|
||||
|
||||
test_script:
|
||||
- echo Running Test
|
||||
- cd c:\projects\OpenBLAS\utest
|
||||
- cd utest
|
||||
- openblas_utest
|
||||
|
||||
+80
-56
@@ -121,13 +121,15 @@ static void *huge_malloc(BLASLONG size){
|
||||
int main(int argc, char *argv[]){
|
||||
|
||||
FLOAT *a, *b, *c;
|
||||
FLOAT alpha[] = {1.0, 1.0};
|
||||
FLOAT alpha[] = {1.0, 0.0};
|
||||
FLOAT beta [] = {0.0, 0.0};
|
||||
char trans='N';
|
||||
blasint m, n, i, j;
|
||||
char transa = 'N';
|
||||
char transb = 'N';
|
||||
blasint m, n, k, i, j, lda, ldb, ldc;
|
||||
int loops = 1;
|
||||
int has_param_n=0;
|
||||
int l;
|
||||
int has_param_m = 0;
|
||||
int has_param_n = 0;
|
||||
int has_param_k = 0;
|
||||
char *p;
|
||||
|
||||
int from = 1;
|
||||
@@ -135,86 +137,108 @@ int main(int argc, char *argv[]){
|
||||
int step = 1;
|
||||
|
||||
struct timeval start, stop;
|
||||
double time1,timeg;
|
||||
double time1, timeg;
|
||||
|
||||
argc--;argv++;
|
||||
|
||||
if (argc > 0) { from = atol(*argv); argc--; argv++;}
|
||||
if (argc > 0) { to = MAX(atol(*argv), from); argc--; argv++;}
|
||||
if (argc > 0) { step = atol(*argv); argc--; argv++;}
|
||||
if (argc > 0) { from = atol(*argv); argc--; argv++; }
|
||||
if (argc > 0) { to = MAX(atol(*argv), from); argc--; argv++; }
|
||||
if (argc > 0) { step = atol(*argv); argc--; argv++; }
|
||||
|
||||
if ((p = getenv("OPENBLAS_TRANS"))) trans=*p;
|
||||
|
||||
fprintf(stderr, "From : %3d To : %3d Step=%d : Trans=%c\n", from, to, step, trans);
|
||||
|
||||
if (( a = (FLOAT *)malloc(sizeof(FLOAT) * to * to * COMPSIZE)) == NULL){
|
||||
fprintf(stderr,"Out of Memory!!\n");exit(1);
|
||||
if ((p = getenv("OPENBLAS_TRANS"))) {
|
||||
transa=*p;
|
||||
transb=*p;
|
||||
}
|
||||
|
||||
if (( b = (FLOAT *)malloc(sizeof(FLOAT) * to * to * COMPSIZE)) == NULL){
|
||||
fprintf(stderr,"Out of Memory!!\n");exit(1);
|
||||
if ((p = getenv("OPENBLAS_TRANSA"))) {
|
||||
transa=*p;
|
||||
}
|
||||
|
||||
if (( c = (FLOAT *)malloc(sizeof(FLOAT) * to * to * COMPSIZE)) == NULL){
|
||||
fprintf(stderr,"Out of Memory!!\n");exit(1);
|
||||
if ((p = getenv("OPENBLAS_TRANSB"))) {
|
||||
transb=*p;
|
||||
}
|
||||
TOUPPER(transa);
|
||||
TOUPPER(transb);
|
||||
|
||||
fprintf(stderr, "From : %3d To : %3d Step=%d : Transa=%c : Transb=%c\n", from, to, step, transa, transb);
|
||||
|
||||
p = getenv("OPENBLAS_LOOPS");
|
||||
if ( p != NULL )
|
||||
loops = atoi(p);
|
||||
if ( p != NULL ) {
|
||||
loops = atoi(p);
|
||||
}
|
||||
|
||||
if ((p = getenv("OPENBLAS_PARAM_M"))) {
|
||||
m = atoi(p);
|
||||
has_param_m=1;
|
||||
} else {
|
||||
m = to;
|
||||
}
|
||||
if ((p = getenv("OPENBLAS_PARAM_N"))) {
|
||||
n = atoi(p);
|
||||
has_param_n=1;
|
||||
n = atoi(p);
|
||||
has_param_n=1;
|
||||
} else {
|
||||
n = to;
|
||||
}
|
||||
if ((p = getenv("OPENBLAS_PARAM_K"))) {
|
||||
k = atoi(p);
|
||||
has_param_k=1;
|
||||
} else {
|
||||
k = to;
|
||||
}
|
||||
|
||||
if (( a = (FLOAT *)malloc(sizeof(FLOAT) * m * k * COMPSIZE)) == NULL) {
|
||||
fprintf(stderr,"Out of Memory!!\n");exit(1);
|
||||
}
|
||||
if (( b = (FLOAT *)malloc(sizeof(FLOAT) * k * n * COMPSIZE)) == NULL) {
|
||||
fprintf(stderr,"Out of Memory!!\n");exit(1);
|
||||
}
|
||||
if (( c = (FLOAT *)malloc(sizeof(FLOAT) * m * n * COMPSIZE)) == NULL) {
|
||||
fprintf(stderr,"Out of Memory!!\n");exit(1);
|
||||
}
|
||||
|
||||
#ifdef linux
|
||||
srandom(getpid());
|
||||
#endif
|
||||
|
||||
for (i = 0; i < m * k * COMPSIZE; i++) {
|
||||
a[i] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
|
||||
}
|
||||
for (i = 0; i < k * n * COMPSIZE; i++) {
|
||||
b[i] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
|
||||
}
|
||||
for (i = 0; i < m * n * COMPSIZE; i++) {
|
||||
c[i] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
|
||||
}
|
||||
|
||||
for(j = 0; j < to; j++){
|
||||
for(i = 0; i < to * COMPSIZE; i++){
|
||||
a[i + j * to * COMPSIZE] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
|
||||
b[i + j * to * COMPSIZE] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
|
||||
c[i + j * to * COMPSIZE] = ((FLOAT) rand() / (FLOAT) RAND_MAX) - 0.5;
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
|
||||
fprintf(stderr, " SIZE Flops Time\n");
|
||||
|
||||
for(m = from; m <= to; m += step)
|
||||
{
|
||||
fprintf(stderr, " SIZE Flops Time\n");
|
||||
|
||||
for (i = from; i <= to; i += step) {
|
||||
|
||||
timeg=0;
|
||||
|
||||
if ( has_param_n == 1 && n <= m )
|
||||
n=n;
|
||||
else
|
||||
n=m;
|
||||
if (!has_param_m) { m = i; }
|
||||
if (!has_param_n) { n = i; }
|
||||
if (!has_param_k) { k = i; }
|
||||
|
||||
if (transa == 'N') { lda = m; }
|
||||
else { lda = k; }
|
||||
if (transb == 'N') { ldb = k; }
|
||||
else { ldb = n; }
|
||||
ldc = m;
|
||||
|
||||
|
||||
fprintf(stderr, " %6dx%d : ", (int)m, (int)n);
|
||||
fprintf(stderr, " M=%4d, N=%4d, K=%4d : ", (int)m, (int)n, (int)k);
|
||||
gettimeofday( &start, (struct timezone *)0);
|
||||
|
||||
for (l=0; l<loops; l++)
|
||||
{
|
||||
|
||||
GEMM (&trans, &trans, &m, &n, &m, alpha, a, &m, b, &m, beta, c, &m );
|
||||
|
||||
|
||||
|
||||
for (j=0; j<loops; j++) {
|
||||
GEMM (&transa, &transb, &m, &n, &k, alpha, a, &lda, b, &ldb, beta, c, &ldc);
|
||||
}
|
||||
gettimeofday( &stop, (struct timezone *)0);
|
||||
time1 = (double)(stop.tv_sec - start.tv_sec) + (double)((stop.tv_usec - start.tv_usec)) * 1.e-6;
|
||||
|
||||
gettimeofday( &stop, (struct timezone *)0);
|
||||
time1 = (double)(stop.tv_sec - start.tv_sec) + (double)((stop.tv_usec - start.tv_usec)) * 1.e-6;
|
||||
|
||||
timeg = time1/loops;
|
||||
fprintf(stderr,
|
||||
" %10.2f MFlops %10.6f sec\n",
|
||||
COMPSIZE * COMPSIZE * 2. * (double)m * (double)m * (double)n / timeg * 1.e-6, time1);
|
||||
|
||||
COMPSIZE * COMPSIZE * 2. * (double)k * (double)m * (double)n / timeg * 1.e-6, time1);
|
||||
|
||||
}
|
||||
|
||||
return 0;
|
||||
|
||||
@@ -54,6 +54,8 @@ $compiler = GCC if ($compiler eq "");
|
||||
$os = Linux if ($data =~ /OS_LINUX/);
|
||||
$os = FreeBSD if ($data =~ /OS_FREEBSD/);
|
||||
$os = NetBSD if ($data =~ /OS_NETBSD/);
|
||||
$os = OpenBSD if ($data =~ /OS_OPENBSD/);
|
||||
$os = DragonFly if ($data =~ /OS_DRAGONFLY/);
|
||||
$os = Darwin if ($data =~ /OS_DARWIN/);
|
||||
$os = SunOS if ($data =~ /OS_SUNOS/);
|
||||
$os = AIX if ($data =~ /OS_AIX/);
|
||||
|
||||
@@ -57,45 +57,45 @@ double cblas_dsdot (OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_
|
||||
float cblas_sdot(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST float *y, OPENBLAS_CONST blasint incy);
|
||||
double cblas_ddot(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST double *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
openblas_complex_float cblas_cdotu(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST float *y, OPENBLAS_CONST blasint incy);
|
||||
openblas_complex_float cblas_cdotc(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST float *y, OPENBLAS_CONST blasint incy);
|
||||
openblas_complex_double cblas_zdotu(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST double *y, OPENBLAS_CONST blasint incy);
|
||||
openblas_complex_double cblas_zdotc(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST double *y, OPENBLAS_CONST blasint incy);
|
||||
openblas_complex_float cblas_cdotu(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *y, OPENBLAS_CONST blasint incy);
|
||||
openblas_complex_float cblas_cdotc(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *y, OPENBLAS_CONST blasint incy);
|
||||
openblas_complex_double cblas_zdotu(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *y, OPENBLAS_CONST blasint incy);
|
||||
openblas_complex_double cblas_zdotc(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
void cblas_cdotu_sub(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST float *y, OPENBLAS_CONST blasint incy, openblas_complex_float *ret);
|
||||
void cblas_cdotc_sub(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST float *y, OPENBLAS_CONST blasint incy, openblas_complex_float *ret);
|
||||
void cblas_zdotu_sub(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST double *y, OPENBLAS_CONST blasint incy, openblas_complex_double *ret);
|
||||
void cblas_zdotc_sub(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST double *y, OPENBLAS_CONST blasint incy, openblas_complex_double *ret);
|
||||
void cblas_cdotu_sub(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *y, OPENBLAS_CONST blasint incy, void *ret);
|
||||
void cblas_cdotc_sub(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *y, OPENBLAS_CONST blasint incy, void *ret);
|
||||
void cblas_zdotu_sub(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *y, OPENBLAS_CONST blasint incy, void *ret);
|
||||
void cblas_zdotc_sub(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *y, OPENBLAS_CONST blasint incy, void *ret);
|
||||
|
||||
float cblas_sasum (OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx);
|
||||
double cblas_dasum (OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx);
|
||||
float cblas_scasum(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx);
|
||||
double cblas_dzasum(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx);
|
||||
float cblas_scasum(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx);
|
||||
double cblas_dzasum(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx);
|
||||
|
||||
float cblas_snrm2 (OPENBLAS_CONST blasint N, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX);
|
||||
double cblas_dnrm2 (OPENBLAS_CONST blasint N, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX);
|
||||
float cblas_scnrm2(OPENBLAS_CONST blasint N, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX);
|
||||
double cblas_dznrm2(OPENBLAS_CONST blasint N, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX);
|
||||
float cblas_scnrm2(OPENBLAS_CONST blasint N, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX);
|
||||
double cblas_dznrm2(OPENBLAS_CONST blasint N, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX);
|
||||
|
||||
CBLAS_INDEX cblas_isamax(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx);
|
||||
CBLAS_INDEX cblas_idamax(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx);
|
||||
CBLAS_INDEX cblas_icamax(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx);
|
||||
CBLAS_INDEX cblas_izamax(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx);
|
||||
CBLAS_INDEX cblas_icamax(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx);
|
||||
CBLAS_INDEX cblas_izamax(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx);
|
||||
|
||||
void cblas_saxpy(OPENBLAS_CONST blasint n, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, float *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_daxpy(OPENBLAS_CONST blasint n, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, double *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_caxpy(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, float *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_zaxpy(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, double *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_caxpy(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, void *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_zaxpy(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, void *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
void cblas_scopy(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, float *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_dcopy(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, double *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_ccopy(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, float *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_zcopy(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, double *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_ccopy(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, void *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_zcopy(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, void *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
void cblas_sswap(OPENBLAS_CONST blasint n, float *x, OPENBLAS_CONST blasint incx, float *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_dswap(OPENBLAS_CONST blasint n, double *x, OPENBLAS_CONST blasint incx, double *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_cswap(OPENBLAS_CONST blasint n, float *x, OPENBLAS_CONST blasint incx, float *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_zswap(OPENBLAS_CONST blasint n, double *x, OPENBLAS_CONST blasint incx, double *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_cswap(OPENBLAS_CONST blasint n, void *x, OPENBLAS_CONST blasint incx, void *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_zswap(OPENBLAS_CONST blasint n, void *x, OPENBLAS_CONST blasint incx, void *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
void cblas_srot(OPENBLAS_CONST blasint N, float *X, OPENBLAS_CONST blasint incX, float *Y, OPENBLAS_CONST blasint incY, OPENBLAS_CONST float c, OPENBLAS_CONST float s);
|
||||
void cblas_drot(OPENBLAS_CONST blasint N, double *X, OPENBLAS_CONST blasint incX, double *Y, OPENBLAS_CONST blasint incY, OPENBLAS_CONST double c, OPENBLAS_CONST double s);
|
||||
@@ -111,59 +111,59 @@ void cblas_drotmg(double *d1, double *d2, double *b1, OPENBLAS_CONST double b2,
|
||||
|
||||
void cblas_sscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_dscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_cscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST float *alpha, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_zscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST double *alpha, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_csscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_zdscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_cscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_zscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_csscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_zdscal(OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, void *X, OPENBLAS_CONST blasint incX);
|
||||
|
||||
void cblas_sgemv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_TRANSPOSE trans, OPENBLAS_CONST blasint m, OPENBLAS_CONST blasint n,
|
||||
OPENBLAS_CONST float alpha, OPENBLAS_CONST float *a, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST float beta, float *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_dgemv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_TRANSPOSE trans, OPENBLAS_CONST blasint m, OPENBLAS_CONST blasint n,
|
||||
OPENBLAS_CONST double alpha, OPENBLAS_CONST double *a, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST double beta, double *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_cgemv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_TRANSPOSE trans, OPENBLAS_CONST blasint m, OPENBLAS_CONST blasint n,
|
||||
OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *a, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST float *beta, float *y, OPENBLAS_CONST blasint incy);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *a, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *beta, void *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_zgemv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_TRANSPOSE trans, OPENBLAS_CONST blasint m, OPENBLAS_CONST blasint n,
|
||||
OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *a, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST double *beta, double *y, OPENBLAS_CONST blasint incy);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *a, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx, OPENBLAS_CONST void *beta, void *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
void cblas_sger (OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *Y, OPENBLAS_CONST blasint incY, float *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_dger (OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *Y, OPENBLAS_CONST blasint incY, double *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_cgeru(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *Y, OPENBLAS_CONST blasint incY, float *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_cgerc(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *Y, OPENBLAS_CONST blasint incY, float *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_zgeru(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *Y, OPENBLAS_CONST blasint incY, double *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_zgerc(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *Y, OPENBLAS_CONST blasint incY, double *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_cgeru(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *Y, OPENBLAS_CONST blasint incY, void *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_cgerc(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *Y, OPENBLAS_CONST blasint incY, void *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_zgeru(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *Y, OPENBLAS_CONST blasint incY, void *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_zgerc(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *Y, OPENBLAS_CONST blasint incY, void *A, OPENBLAS_CONST blasint lda);
|
||||
|
||||
void cblas_strsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_dtrsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ctrsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ztrsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ctrsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ztrsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *X, OPENBLAS_CONST blasint incX);
|
||||
|
||||
void cblas_strmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_dtrmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ctrmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ztrmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ctrmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ztrmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *X, OPENBLAS_CONST blasint incX);
|
||||
|
||||
void cblas_ssyr(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, float *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_dsyr(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, double *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_cher(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, float *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_zher(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, double *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_cher(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, void *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_zher(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, void *A, OPENBLAS_CONST blasint lda);
|
||||
|
||||
void cblas_ssyr2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo,OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *X,
|
||||
OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *Y, OPENBLAS_CONST blasint incY, float *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_dsyr2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *X,
|
||||
OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *Y, OPENBLAS_CONST blasint incY, double *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_cher2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX,
|
||||
OPENBLAS_CONST float *Y, OPENBLAS_CONST blasint incY, float *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_zher2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX,
|
||||
OPENBLAS_CONST double *Y, OPENBLAS_CONST blasint incY, double *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_cher2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX,
|
||||
OPENBLAS_CONST void *Y, OPENBLAS_CONST blasint incY, void *A, OPENBLAS_CONST blasint lda);
|
||||
void cblas_zher2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX,
|
||||
OPENBLAS_CONST void *Y, OPENBLAS_CONST blasint incY, void *A, OPENBLAS_CONST blasint lda);
|
||||
|
||||
void cblas_sgbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST blasint KL, OPENBLAS_CONST blasint KU, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float beta, float *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_dgbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST blasint KL, OPENBLAS_CONST blasint KU, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double beta, double *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_cgbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST blasint KL, OPENBLAS_CONST blasint KU, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *beta, float *Y, OPENBLAS_CONST blasint incY);
|
||||
OPENBLAS_CONST blasint KL, OPENBLAS_CONST blasint KU, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *beta, void *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_zgbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST blasint KL, OPENBLAS_CONST blasint KU, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *beta, double *Y, OPENBLAS_CONST blasint incY);
|
||||
OPENBLAS_CONST blasint KL, OPENBLAS_CONST blasint KU, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *beta, void *Y, OPENBLAS_CONST blasint incY);
|
||||
|
||||
void cblas_ssbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A,
|
||||
OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float beta, float *Y, OPENBLAS_CONST blasint incY);
|
||||
@@ -176,45 +176,45 @@ void cblas_stbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLA
|
||||
void cblas_dtbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ctbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *X, OPENBLAS_CONST blasint incX);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ztbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *X, OPENBLAS_CONST blasint incX);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *X, OPENBLAS_CONST blasint incX);
|
||||
|
||||
void cblas_stbsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_dtbsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ctbsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *X, OPENBLAS_CONST blasint incX);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ztbsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *X, OPENBLAS_CONST blasint incX);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *X, OPENBLAS_CONST blasint incX);
|
||||
|
||||
void cblas_stpmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST float *Ap, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_dtpmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST double *Ap, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ctpmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST float *Ap, float *X, OPENBLAS_CONST blasint incX);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST void *Ap, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ztpmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST double *Ap, double *X, OPENBLAS_CONST blasint incX);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST void *Ap, void *X, OPENBLAS_CONST blasint incX);
|
||||
|
||||
void cblas_stpsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST float *Ap, float *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_dtpsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST double *Ap, double *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ctpsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST float *Ap, float *X, OPENBLAS_CONST blasint incX);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST void *Ap, void *X, OPENBLAS_CONST blasint incX);
|
||||
void cblas_ztpsv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_DIAG Diag,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST double *Ap, double *X, OPENBLAS_CONST blasint incX);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST void *Ap, void *X, OPENBLAS_CONST blasint incX);
|
||||
|
||||
void cblas_ssymv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A,
|
||||
OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float beta, float *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_dsymv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A,
|
||||
OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double beta, double *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_chemv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A,
|
||||
OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *beta, float *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_zhemv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A,
|
||||
OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *beta, double *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_chemv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A,
|
||||
OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *beta, void *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_zhemv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A,
|
||||
OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *beta, void *Y, OPENBLAS_CONST blasint incY);
|
||||
|
||||
|
||||
void cblas_sspmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *Ap,
|
||||
@@ -225,36 +225,36 @@ void cblas_dspmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLA
|
||||
void cblas_sspr(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, float *Ap);
|
||||
void cblas_dspr(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, double *Ap);
|
||||
|
||||
void cblas_chpr(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, float *A);
|
||||
void cblas_zhpr(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *X,OPENBLAS_CONST blasint incX, double *A);
|
||||
void cblas_chpr(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, void *A);
|
||||
void cblas_zhpr(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST void *X,OPENBLAS_CONST blasint incX, void *A);
|
||||
|
||||
void cblas_sspr2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *Y, OPENBLAS_CONST blasint incY, float *A);
|
||||
void cblas_dspr2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *Y, OPENBLAS_CONST blasint incY, double *A);
|
||||
void cblas_chpr2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *Y, OPENBLAS_CONST blasint incY, float *Ap);
|
||||
void cblas_zhpr2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *Y, OPENBLAS_CONST blasint incY, double *Ap);
|
||||
void cblas_chpr2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *Y, OPENBLAS_CONST blasint incY, void *Ap);
|
||||
void cblas_zhpr2(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *Y, OPENBLAS_CONST blasint incY, void *Ap);
|
||||
|
||||
void cblas_chbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *beta, float *Y, OPENBLAS_CONST blasint incY);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *beta, void *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_zhbmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *beta, double *Y, OPENBLAS_CONST blasint incY);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *beta, void *Y, OPENBLAS_CONST blasint incY);
|
||||
|
||||
void cblas_chpmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *Ap, OPENBLAS_CONST float *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST float *beta, float *Y, OPENBLAS_CONST blasint incY);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *Ap, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *beta, void *Y, OPENBLAS_CONST blasint incY);
|
||||
void cblas_zhpmv(OPENBLAS_CONST enum CBLAS_ORDER order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *Ap, OPENBLAS_CONST double *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST double *beta, double *Y, OPENBLAS_CONST blasint incY);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *Ap, OPENBLAS_CONST void *X, OPENBLAS_CONST blasint incX, OPENBLAS_CONST void *beta, void *Y, OPENBLAS_CONST blasint incY);
|
||||
|
||||
void cblas_sgemm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransB, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_dgemm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransB, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_cgemm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransB, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float *beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_cgemm3m(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransB, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float *beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_zgemm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransB, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double *beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_zgemm3m(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransB, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double *beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
|
||||
|
||||
void cblas_ssymm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
@@ -262,60 +262,60 @@ void cblas_ssymm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLA
|
||||
void cblas_dsymm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_csymm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float *beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_zsymm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double *beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
|
||||
void cblas_ssyrk(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_dsyrk(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_csyrk(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_zsyrk(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
|
||||
void cblas_ssyr2k(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_dsyr2k(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_csyr2k(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float *beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_zsyr2k(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans,
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double *beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
|
||||
void cblas_strmm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA,
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *B, OPENBLAS_CONST blasint ldb);
|
||||
void cblas_dtrmm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA,
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *B, OPENBLAS_CONST blasint ldb);
|
||||
void cblas_ctrmm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA,
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *B, OPENBLAS_CONST blasint ldb);
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *B, OPENBLAS_CONST blasint ldb);
|
||||
void cblas_ztrmm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA,
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *B, OPENBLAS_CONST blasint ldb);
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *B, OPENBLAS_CONST blasint ldb);
|
||||
|
||||
void cblas_strsm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA,
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *B, OPENBLAS_CONST blasint ldb);
|
||||
void cblas_dtrsm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA,
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *B, OPENBLAS_CONST blasint ldb);
|
||||
void cblas_ctrsm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA,
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, float *B, OPENBLAS_CONST blasint ldb);
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *B, OPENBLAS_CONST blasint ldb);
|
||||
void cblas_ztrsm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE TransA,
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, double *B, OPENBLAS_CONST blasint ldb);
|
||||
OPENBLAS_CONST enum CBLAS_DIAG Diag, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, void *B, OPENBLAS_CONST blasint ldb);
|
||||
|
||||
void cblas_chemm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float *beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_zhemm(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_SIDE Side, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST blasint M, OPENBLAS_CONST blasint N,
|
||||
OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double *beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST void *beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
|
||||
void cblas_cherk(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST float alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST float alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_zherk(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST double alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST double alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
|
||||
void cblas_cher2k(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST float *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float beta, float *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST float beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
void cblas_zher2k(OPENBLAS_CONST enum CBLAS_ORDER Order, OPENBLAS_CONST enum CBLAS_UPLO Uplo, OPENBLAS_CONST enum CBLAS_TRANSPOSE Trans, OPENBLAS_CONST blasint N, OPENBLAS_CONST blasint K,
|
||||
OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST double *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double beta, double *C, OPENBLAS_CONST blasint ldc);
|
||||
OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *A, OPENBLAS_CONST blasint lda, OPENBLAS_CONST void *B, OPENBLAS_CONST blasint ldb, OPENBLAS_CONST double beta, void *C, OPENBLAS_CONST blasint ldc);
|
||||
|
||||
void cblas_xerbla(blasint p, char *rout, char *form, ...);
|
||||
|
||||
@@ -325,9 +325,9 @@ void cblas_saxpby(OPENBLAS_CONST blasint n, OPENBLAS_CONST float alpha, OPENBLAS
|
||||
|
||||
void cblas_daxpby(OPENBLAS_CONST blasint n, OPENBLAS_CONST double alpha, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx,OPENBLAS_CONST double beta, double *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
void cblas_caxpby(OPENBLAS_CONST blasint n, OPENBLAS_CONST float *alpha, OPENBLAS_CONST float *x, OPENBLAS_CONST blasint incx,OPENBLAS_CONST float *beta, float *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_caxpby(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx,OPENBLAS_CONST void *beta, void *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
void cblas_zaxpby(OPENBLAS_CONST blasint n, OPENBLAS_CONST double *alpha, OPENBLAS_CONST double *x, OPENBLAS_CONST blasint incx,OPENBLAS_CONST double *beta, double *y, OPENBLAS_CONST blasint incy);
|
||||
void cblas_zaxpby(OPENBLAS_CONST blasint n, OPENBLAS_CONST void *alpha, OPENBLAS_CONST void *x, OPENBLAS_CONST blasint incx,OPENBLAS_CONST void *beta, void *y, OPENBLAS_CONST blasint incy);
|
||||
|
||||
void cblas_somatcopy(OPENBLAS_CONST enum CBLAS_ORDER CORDER, OPENBLAS_CONST enum CBLAS_TRANSPOSE CTRANS, OPENBLAS_CONST blasint crows, OPENBLAS_CONST blasint ccols, OPENBLAS_CONST float calpha, OPENBLAS_CONST float *a,
|
||||
OPENBLAS_CONST blasint clda, float *b, OPENBLAS_CONST blasint cldb);
|
||||
|
||||
+18
-37
@@ -3,9 +3,9 @@
|
||||
## Description: Ported from portion of OpenBLAS/Makefile.system
|
||||
## Sets various variables based on architecture.
|
||||
|
||||
if (${ARCH} STREQUAL "x86" OR ${ARCH} STREQUAL "x86_64")
|
||||
if (X86 OR X86_64)
|
||||
|
||||
if (${ARCH} STREQUAL "x86")
|
||||
if (X86)
|
||||
if (NOT BINARY)
|
||||
set(NO_BINARY_MODE 1)
|
||||
endif ()
|
||||
@@ -33,47 +33,28 @@ if (${CMAKE_C_COMPILER_ID} STREQUAL "Intel")
|
||||
endif ()
|
||||
|
||||
if (USE_OPENMP)
|
||||
|
||||
if (${CMAKE_C_COMPILER_ID} STREQUAL "GNU" OR ${CMAKE_C_COMPILER_ID} STREQUAL "LSB")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -fopenmp")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_C_COMPILER_ID} STREQUAL "Clang")
|
||||
message(WARNING "Clang doesn't support OpenMP yet.")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -fopenmp")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_C_COMPILER_ID} STREQUAL "Intel")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -openmp")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_C_COMPILER_ID} STREQUAL "PGI")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -mp")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_C_COMPILER_ID} STREQUAL "OPEN64")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -mp")
|
||||
set(CEXTRALIB "${CEXTRALIB} -lstdc++")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_C_COMPILER_ID} STREQUAL "PATHSCALE")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -mp")
|
||||
endif ()
|
||||
# USE_SIMPLE_THREADED_LEVEL3 = 1
|
||||
# NO_AFFINITY = 1
|
||||
find_package(OpenMP REQUIRED)
|
||||
if (OpenMP_FOUND)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} ${OpenMP_C_FLAGS} -DUSE_OPENMP")
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} ${OpenMP_Fortran_FLAGS}")
|
||||
endif()
|
||||
endif ()
|
||||
|
||||
|
||||
if (DYNAMIC_ARCH)
|
||||
if (${ARCH} STREQUAL "x86")
|
||||
set(DYNAMIC_CORE "KATMAI COPPERMINE NORTHWOOD PRESCOTT BANIAS CORE2 PENRYN DUNNINGTON NEHALEM ATHLON OPTERON OPTERON_SSE3 BARCELONA BOBCAT ATOM NANO")
|
||||
if (X86)
|
||||
set(DYNAMIC_CORE KATMAI COPPERMINE NORTHWOOD PRESCOTT BANIAS CORE2 PENRYN DUNNINGTON NEHALEM ATHLON OPTERON OPTERON_SSE3 BARCELONA BOBCAT ATOM NANO)
|
||||
endif ()
|
||||
|
||||
if (${ARCH} STREQUAL "x86_64")
|
||||
set(DYNAMIC_CORE "PRESCOTT CORE2 PENRYN DUNNINGTON NEHALEM OPTERON OPTERON_SSE3 BARCELONA BOBCAT ATOM NANO")
|
||||
if (X86_64)
|
||||
set(DYNAMIC_CORE PRESCOTT CORE2 PENRYN DUNNINGTON NEHALEM OPTERON OPTERON_SSE3 BARCELONA BOBCAT ATOM NANO)
|
||||
if (NOT NO_AVX)
|
||||
set(DYNAMIC_CORE "${DYNAMIC_CORE} SANDYBRIDGE BULLDOZER PILEDRIVER STEAMROLLER")
|
||||
set(DYNAMIC_CORE ${DYNAMIC_CORE} SANDYBRIDGE BULLDOZER PILEDRIVER STEAMROLLER EXCAVATOR)
|
||||
endif ()
|
||||
if (NOT NO_AVX2)
|
||||
set(DYNAMIC_CORE "${DYNAMIC_CORE} HASWELL")
|
||||
set(DYNAMIC_CORE ${DYNAMIC_CORE} HASWELL ZEN)
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
@@ -94,7 +75,7 @@ if (${ARCH} STREQUAL "ia64")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
if (${ARCH} STREQUAL "mips64")
|
||||
if (MIPS64)
|
||||
set(NO_BINARY_MODE 1)
|
||||
endif ()
|
||||
|
||||
@@ -103,12 +84,12 @@ if (${ARCH} STREQUAL "alpha")
|
||||
set(BINARY_DEFINED 1)
|
||||
endif ()
|
||||
|
||||
if (${ARCH} STREQUAL "arm")
|
||||
if (ARM)
|
||||
set(NO_BINARY_MODE 1)
|
||||
set(BINARY_DEFINED 1)
|
||||
endif ()
|
||||
|
||||
if (${ARCH} STREQUAL "arm64")
|
||||
if (ARM64)
|
||||
set(NO_BINARY_MODE 1)
|
||||
set(BINARY_DEFINED 1)
|
||||
endif ()
|
||||
|
||||
@@ -1,89 +0,0 @@
|
||||
##
|
||||
## Author: Hank Anderson <hank@statease.com>
|
||||
## Description: Ported from the OpenBLAS/c_check perl script.
|
||||
## This is triggered by prebuild.cmake and runs before any of the code is built.
|
||||
## Creates config.h and Makefile.conf.
|
||||
|
||||
# CMake vars set by this file:
|
||||
# OSNAME (use CMAKE_SYSTEM_NAME)
|
||||
# ARCH
|
||||
# C_COMPILER (use CMAKE_C_COMPILER)
|
||||
# BINARY32
|
||||
# BINARY64
|
||||
# FU
|
||||
# CROSS_SUFFIX
|
||||
# CROSS
|
||||
# CEXTRALIB
|
||||
|
||||
# Defines set by this file:
|
||||
# OS_
|
||||
# ARCH_
|
||||
# C_
|
||||
# __32BIT__
|
||||
# __64BIT__
|
||||
# FUNDERSCORE
|
||||
# PTHREAD_CREATE_FUNC
|
||||
|
||||
# N.B. c_check (and ctest.c) is not cross-platform, so instead try to use CMake variables.
|
||||
set(FU "")
|
||||
if(APPLE)
|
||||
set(FU "_")
|
||||
elseif(MSVC)
|
||||
set(FU "_")
|
||||
elseif(UNIX)
|
||||
set(FU "")
|
||||
endif()
|
||||
|
||||
# Convert CMake vars into the format that OpenBLAS expects
|
||||
string(TOUPPER ${CMAKE_SYSTEM_NAME} HOST_OS)
|
||||
if (${HOST_OS} STREQUAL "WINDOWS")
|
||||
set(HOST_OS WINNT)
|
||||
endif ()
|
||||
|
||||
# added by hpa - check size of void ptr to detect 64-bit compile
|
||||
if (NOT DEFINED BINARY)
|
||||
set(BINARY 32)
|
||||
if (CMAKE_SIZEOF_VOID_P EQUAL 8)
|
||||
set(BINARY 64)
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
if (BINARY EQUAL 64)
|
||||
set(BINARY64 1)
|
||||
else ()
|
||||
set(BINARY32 1)
|
||||
endif ()
|
||||
|
||||
# CMake docs define these:
|
||||
# CMAKE_SYSTEM_PROCESSOR - The name of the CPU CMake is building for.
|
||||
# CMAKE_HOST_SYSTEM_PROCESSOR - The name of the CPU CMake is running on.
|
||||
#
|
||||
# TODO: CMAKE_SYSTEM_PROCESSOR doesn't seem to be correct - instead get it from the compiler a la c_check
|
||||
set(ARCH ${CMAKE_SYSTEM_PROCESSOR})
|
||||
if (${ARCH} STREQUAL "AMD64")
|
||||
set(ARCH "x86_64")
|
||||
endif ()
|
||||
|
||||
# If you are using a 32-bit compiler on a 64-bit system CMAKE_SYSTEM_PROCESSOR will be wrong
|
||||
if (${ARCH} STREQUAL "x86_64" AND BINARY EQUAL 32)
|
||||
set(ARCH x86)
|
||||
endif ()
|
||||
|
||||
if (${ARCH} STREQUAL "X86")
|
||||
set(ARCH x86)
|
||||
endif ()
|
||||
|
||||
set(COMPILER_ID ${CMAKE_CXX_COMPILER_ID})
|
||||
if (${COMPILER_ID} STREQUAL "GNU")
|
||||
set(COMPILER_ID "GCC")
|
||||
endif ()
|
||||
|
||||
string(TOUPPER ${ARCH} UC_ARCH)
|
||||
|
||||
file(WRITE ${TARGET_CONF}
|
||||
"#define OS_${HOST_OS}\t1\n"
|
||||
"#define ARCH_${UC_ARCH}\t1\n"
|
||||
"#define C_${COMPILER_ID}\t1\n"
|
||||
"#define __${BINARY}BIT__\t1\n"
|
||||
"#define FUNDERSCORE\t${FU}\n")
|
||||
|
||||
+6
-11
@@ -15,7 +15,7 @@ if (${CMAKE_C_COMPILER} STREQUAL "GNU" OR ${CMAKE_C_COMPILER} STREQUAL "LSB" OR
|
||||
|
||||
if (NO_BINARY_MODE)
|
||||
|
||||
if (${ARCH} STREQUAL "mips64")
|
||||
if (MIPS64)
|
||||
if (BINARY64)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -mabi=64")
|
||||
else ()
|
||||
@@ -24,17 +24,12 @@ if (${CMAKE_C_COMPILER} STREQUAL "GNU" OR ${CMAKE_C_COMPILER} STREQUAL "LSB" OR
|
||||
set(BINARY_DEFINED 1)
|
||||
endif ()
|
||||
|
||||
if (${CORE} STREQUAL "LOONGSON3A")
|
||||
if (${CORE} STREQUAL "LOONGSON3A" OR ${CORE} STREQUAL "LOONGSON3B")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -march=mips64")
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -march=mips64")
|
||||
endif ()
|
||||
|
||||
if (${CORE} STREQUAL "LOONGSON3B")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -march=mips64")
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -march=mips64")
|
||||
endif ()
|
||||
|
||||
if (${OSNAME} STREQUAL "AIX")
|
||||
if (CMAKE_SYSTEM_NAME STREQUAL "AIX")
|
||||
set(BINARY_DEFINED 1)
|
||||
endif ()
|
||||
endif ()
|
||||
@@ -66,7 +61,7 @@ endif ()
|
||||
|
||||
if (${CMAKE_C_COMPILER} STREQUAL "OPEN64")
|
||||
|
||||
if (${ARCH} STREQUAL "mips64")
|
||||
if (MIPS64)
|
||||
|
||||
if (NOT BINARY64)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -n32")
|
||||
@@ -94,10 +89,10 @@ endif ()
|
||||
|
||||
if (${CMAKE_C_COMPILER} STREQUAL "SUN")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -w")
|
||||
if (${ARCH} STREQUAL "x86")
|
||||
if (X86)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -m32")
|
||||
else ()
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -m64")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -m64")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
|
||||
+2
-1
@@ -51,7 +51,8 @@ else()
|
||||
endif()
|
||||
|
||||
add_custom_command(
|
||||
TARGET ${OpenBLAS_LIBNAME} PRE_LINK
|
||||
OUTPUT ${PROJECT_BINARY_DIR}/openblas.def
|
||||
#TARGET ${OpenBLAS_LIBNAME} PRE_LINK
|
||||
COMMAND perl
|
||||
ARGS "${PROJECT_SOURCE_DIR}/exports/gensymbol" "win2k" "${ARCH_IN}" "dummy" "${EXPRECISION_IN}" "${NO_CBLAS_IN}" "${NO_LAPACK_IN}" "${NO_LAPACKE_IN}" "${NEED2UNDERSCORES_IN}" "${ONLY_CBLAS_IN}" "${SYMBOLPREFIX}" "${SYMBOLSUFFIX}" > "${PROJECT_BINARY_DIR}/openblas.def"
|
||||
COMMENT "Create openblas.def file"
|
||||
|
||||
+2
-12
@@ -20,12 +20,6 @@
|
||||
# NEEDBUNDERSCORE
|
||||
# NEED2UNDERSCORES
|
||||
|
||||
if (MSVC)
|
||||
# had to do this for MSVC, else CMake automatically assumes I have ifort... -hpa
|
||||
include(CMakeForceCompiler)
|
||||
CMAKE_FORCE_Fortran_COMPILER(gfortran GNU)
|
||||
endif ()
|
||||
|
||||
if (NOT NO_LAPACK)
|
||||
enable_language(Fortran)
|
||||
else()
|
||||
@@ -34,17 +28,13 @@ else()
|
||||
endif()
|
||||
|
||||
if (NOT ONLY_CBLAS)
|
||||
# N.B. f_check is not cross-platform, so instead try to use CMake variables
|
||||
# run f_check (appends to TARGET files)
|
||||
# message(STATUS "Running f_check...")
|
||||
# execute_process(COMMAND perl f_check ${TARGET_MAKE} ${TARGET_CONF} ${CMAKE_Fortran_COMPILER}
|
||||
# WORKING_DIRECTORY ${PROJECT_SOURCE_DIR})
|
||||
|
||||
# TODO: detect whether underscore needed, set #defines and BU appropriately - use try_compile
|
||||
# TODO: set FEXTRALIB flags a la f_check?
|
||||
|
||||
set(BU "_")
|
||||
file(APPEND ${TARGET_CONF}
|
||||
file(APPEND ${TARGET_CONF_TEMP}
|
||||
"#define BUNDERSCORE _\n"
|
||||
"#define NEEDBUNDERSCORE 1\n"
|
||||
"#define NEED2UNDERSCORES 0\n")
|
||||
@@ -56,7 +46,7 @@ else ()
|
||||
set(NO_FBLAS 1)
|
||||
#set(F_COMPILER GFORTRAN) # CMake handles the fortran compiler
|
||||
set(BU "_")
|
||||
file(APPEND ${TARGET_CONF}
|
||||
file(APPEND ${TARGET_CONF_TEMP}
|
||||
"#define BUNDERSCORE _\n"
|
||||
"#define NEEDBUNDERSCORE 1\n")
|
||||
endif()
|
||||
|
||||
+14
-4
@@ -3,6 +3,16 @@
|
||||
## Description: Ported from portion of OpenBLAS/Makefile.system
|
||||
## Sets Fortran related variables.
|
||||
|
||||
if (${F_COMPILER} STREQUAL "FLANG")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DF_INTERFACE_FLANG")
|
||||
if (BINARY64 AND INTERFACE64)
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -i8")
|
||||
endif ()
|
||||
if (USE_OPENMP)
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -fopenmp")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
if (${F_COMPILER} STREQUAL "G77")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DF_INTERFACE_G77")
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -Wall")
|
||||
@@ -35,7 +45,7 @@ if (${F_COMPILER} STREQUAL "GFORTRAN")
|
||||
set(EXTRALIB "{EXTRALIB} -lgfortran")
|
||||
endif ()
|
||||
if (NO_BINARY_MODE)
|
||||
if (${ARCH} STREQUAL "mips64")
|
||||
if (MIPS64)
|
||||
if (BINARY64)
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -mabi=64")
|
||||
else ()
|
||||
@@ -115,7 +125,7 @@ if (${F_COMPILER} STREQUAL "PATHSCALE")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
if (NOT ${ARCH} STREQUAL "mips64")
|
||||
if (NOT MIPS64)
|
||||
if (NOT BINARY64)
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -m32")
|
||||
else ()
|
||||
@@ -143,7 +153,7 @@ if (${F_COMPILER} STREQUAL "OPEN64")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
if (${ARCH} STREQUAL "mips64")
|
||||
if (MIPS64)
|
||||
|
||||
if (NOT BINARY64)
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -n32")
|
||||
@@ -174,7 +184,7 @@ endif ()
|
||||
|
||||
if (${F_COMPILER} STREQUAL "SUN")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DF_INTERFACE_SUN")
|
||||
if (${ARCH} STREQUAL "x86")
|
||||
if (X86)
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -m32")
|
||||
else ()
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -m64")
|
||||
|
||||
+459
-361
@@ -1,387 +1,485 @@
|
||||
# Sources for compiling lapack-netlib. Can't use CMakeLists.txt because lapack-netlib already has its own cmake files.
|
||||
|
||||
set(ALLAUX
|
||||
ilaenv.f ieeeck.f lsamen.f xerbla_array.f iparmq.f
|
||||
ilaprec.f ilatrans.f ilauplo.f iladiag.f iparam2stage.F chla_transtype.f
|
||||
../INSTALL/ilaver.f ../INSTALL/slamch.f
|
||||
)
|
||||
set(ALLAUX ilaenv.f ilaenv2stage.f ieeeck.f lsamen.f iparmq.f iparam2stage.F
|
||||
ilaprec.f ilatrans.f ilauplo.f iladiag.f chla_transtype.f
|
||||
../INSTALL/ilaver.f xerbla_array.f
|
||||
../INSTALL/slamch.f)
|
||||
|
||||
set(SCLAUX
|
||||
sbdsdc.f
|
||||
sbdsqr.f sdisna.f slabad.f slacpy.f sladiv.f slae2.f slaebz.f
|
||||
slaed0.f slaed1.f slaed2.f slaed3.f slaed4.f slaed5.f slaed6.f
|
||||
slaed7.f slaed8.f slaed9.f slaeda.f slaev2.f slagtf.f
|
||||
slagts.f slamrg.f slanst.f
|
||||
slapy2.f slapy3.f slarnv.f
|
||||
slarra.f slarrb.f slarrc.f slarrd.f slarre.f slarrf.f slarrj.f
|
||||
slarrk.f slarrr.f slaneg.f
|
||||
slartg.f slaruv.f slas2.f slascl.f
|
||||
slasd0.f slasd1.f slasd2.f slasd3.f slasd4.f slasd5.f slasd6.f
|
||||
slasd7.f slasd8.f slasda.f slasdq.f slasdt.f
|
||||
slaset.f slasq1.f slasq2.f slasq3.f slasq4.f slasq5.f slasq6.f
|
||||
slasr.f slasrt.f slassq.f slasv2.f spttrf.f sstebz.f sstedc.f
|
||||
ssteqr.f ssterf.f slaisnan.f sisnan.f
|
||||
slartgp.f slartgs.f
|
||||
../INSTALL/second_${TIMER}.f
|
||||
)
|
||||
sbdsdc.f
|
||||
sbdsqr.f sdisna.f slabad.f slacpy.f sladiv.f slae2.f slaebz.f
|
||||
slaed0.f slaed1.f slaed2.f slaed3.f slaed4.f slaed5.f slaed6.f
|
||||
slaed7.f slaed8.f slaed9.f slaeda.f slaev2.f slagtf.f
|
||||
slagts.f slamrg.f slanst.f
|
||||
slapy2.f slapy3.f slarnv.f
|
||||
slarra.f slarrb.f slarrc.f slarrd.f slarre.f slarrf.f slarrj.f
|
||||
slarrk.f slarrr.f slaneg.f
|
||||
slartg.f slaruv.f slas2.f slascl.f
|
||||
slasd0.f slasd1.f slasd2.f slasd3.f slasd4.f slasd5.f slasd6.f
|
||||
slasd7.f slasd8.f slasda.f slasdq.f slasdt.f
|
||||
slaset.f slasq1.f slasq2.f slasq3.f slasq4.f slasq5.f slasq6.f
|
||||
slasr.f slasrt.f slassq.f slasv2.f spttrf.f sstebz.f sstedc.f
|
||||
ssteqr.f ssterf.f slaisnan.f sisnan.f
|
||||
slartgp.f slartgs.f
|
||||
../INSTALL/second_${TIMER}.f)
|
||||
|
||||
set(DZLAUX
|
||||
dbdsdc.f dbdsvdx.f
|
||||
dbdsqr.f ddisna.f dlabad.f dlacpy.f dladiv.f dlae2.f dlaebz.f
|
||||
dlaed0.f dlaed1.f dlaed2.f dlaed3.f dlaed4.f dlaed5.f dlaed6.f
|
||||
dlaed7.f dlaed8.f dlaed9.f dlaeda.f dlaev2.f dlagtf.f
|
||||
dlagts.f dlamrg.f dlanst.f
|
||||
dlapy2.f dlapy3.f dlarnv.f
|
||||
dlarra.f dlarrb.f dlarrc.f dlarrd.f dlarre.f dlarrf.f dlarrj.f
|
||||
dlarrk.f dlarrr.f dlaneg.f
|
||||
dlartg.f dlaruv.f dlas2.f dlascl.f
|
||||
dlasd0.f dlasd1.f dlasd2.f dlasd3.f dlasd4.f dlasd5.f dlasd6.f
|
||||
dlasd7.f dlasd8.f dlasda.f dlasdq.f dlasdt.f
|
||||
dlaset.f dlasq1.f dlasq2.f dlasq3.f dlasq4.f dlasq5.f dlasq6.f
|
||||
dlasr.f dlasrt.f dlassq.f dlasv2.f dpttrf.f dstebz.f dstedc.f
|
||||
dsteqr.f dsterf.f dlaisnan.f disnan.f
|
||||
dlartgp.f dlartgs.f
|
||||
../INSTALL/dlamch.f ../INSTALL/dsecnd_${TIMER}.f
|
||||
dgelq.f dgelqt.f dgelqt3.f dgemlq.f dgemlqt.f dgemqr.f dgeqr.f
|
||||
dgetsls.f dlamswlq.f dlamtsqr.f dlaswlq.f dlatsqr.f dtplqt.f
|
||||
dtplqt2.f dtpmlqt.f dsysv_aa.f dsytrf_aa.f dsytrs_aa.f dlasyf_aa.f
|
||||
dsytf2_rk.f dlasyf_rk.f dsytrf_rk.f dsytrs_3.f dsycon_3.f dsytri_3.f
|
||||
dsytri_3x.f dsysv_rk.f dsb2st_kernels.f dsbev_2stage.f dsbevd_2stage.f
|
||||
dsbevx_2stage.f dsyev_2stage.f dsyevd_2stage.f dsyevr_2stage.f
|
||||
dsyevx_2stage.f dsygv_2stage.f dsytrd_2stage.f dsytrd_sb2st.F
|
||||
dsytrd_sy2sb.f dlarfy.f
|
||||
)
|
||||
dbdsdc.f
|
||||
dbdsqr.f ddisna.f dlabad.f dlacpy.f dladiv.f dlae2.f dlaebz.f
|
||||
dlaed0.f dlaed1.f dlaed2.f dlaed3.f dlaed4.f dlaed5.f dlaed6.f
|
||||
dlaed7.f dlaed8.f dlaed9.f dlaeda.f dlaev2.f dlagtf.f
|
||||
dlagts.f dlamrg.f dlanst.f
|
||||
dlapy2.f dlapy3.f dlarnv.f
|
||||
dlarra.f dlarrb.f dlarrc.f dlarrd.f dlarre.f dlarrf.f dlarrj.f
|
||||
dlarrk.f dlarrr.f dlaneg.f
|
||||
dlartg.f dlaruv.f dlas2.f dlascl.f
|
||||
dlasd0.f dlasd1.f dlasd2.f dlasd3.f dlasd4.f dlasd5.f dlasd6.f
|
||||
dlasd7.f dlasd8.f dlasda.f dlasdq.f dlasdt.f
|
||||
dlaset.f dlasq1.f dlasq2.f dlasq3.f dlasq4.f dlasq5.f dlasq6.f
|
||||
dlasr.f dlasrt.f dlassq.f dlasv2.f dpttrf.f dstebz.f dstedc.f
|
||||
dsteqr.f dsterf.f dlaisnan.f disnan.f
|
||||
dlartgp.f dlartgs.f
|
||||
../INSTALL/dlamch.f ../INSTALL/dsecnd_${TIMER}.f)
|
||||
|
||||
set(SLASRC
|
||||
sbdsvdx.f sgbbrd.f sgbcon.f sgbequ.f sgbrfs.f sgbsv.f
|
||||
sgbsvx.f sgbtf2.f sgbtrf.f sgbtrs.f sgebak.f sgebal.f sgebd2.f
|
||||
sgebrd.f sgecon.f sgeequ.f sgees.f sgeesx.f sgeev.f sgeevx.f
|
||||
DEPRECATED/sgegs.f DEPRECATED/sgegv.f sgehd2.f sgehrd.f sgelq2.f sgelqf.f
|
||||
sgels.f sgelsd.f sgelss.f DEPRECATED/sgelsx.f sgelsy.f sgeql2.f sgeqlf.f
|
||||
sgeqp3.f DEPRECATED/sgeqpf.f sgeqr2.f sgeqr2p.f sgeqrf.f sgeqrfp.f sgerfs.f
|
||||
sgerq2.f sgerqf.f sgesc2.f sgesdd.f sgesvd.f sgesvdx.f sgesvx.f
|
||||
sgetc2.f sgetri.f sgetrf2.f
|
||||
sggbak.f sggbal.f sgghd3.f sgges.f sgges3.f sggesx.f sggev.f sggev3.f sggevx.f
|
||||
sggglm.f sgghrd.f sgglse.f sggqrf.f
|
||||
sggrqf.f DEPRECATED/sggsvd.f sggsvd3.f DEPRECATED/sggsvp.f sggsvp3.f sgtcon.f sgtrfs.f sgtsv.f
|
||||
sgtsvx.f sgttrf.f sgttrs.f sgtts2.f shgeqz.f
|
||||
shsein.f shseqr.f slabrd.f slacon.f slacn2.f
|
||||
slaein.f slaexc.f slag2.f slags2.f slagtm.f slagv2.f slahqr.f
|
||||
DEPRECATED/slahrd.f slahr2.f slaic1.f slaln2.f slals0.f slalsa.f slalsd.f
|
||||
slangb.f slange.f slangt.f slanhs.f slansb.f slansp.f
|
||||
slansy.f slantb.f slantp.f slantr.f slanv2.f
|
||||
slapll.f slapmt.f
|
||||
slaqgb.f slaqge.f slaqp2.f slaqps.f slaqsb.f slaqsp.f slaqsy.f
|
||||
slaqr0.f slaqr1.f slaqr2.f slaqr3.f slaqr4.f slaqr5.f
|
||||
slaqtr.f slar1v.f slar2v.f ilaslr.f ilaslc.f
|
||||
slarf.f slarfb.f slarfg.f slarfgp.f slarft.f slarfx.f slargv.f
|
||||
slarrv.f slartv.f
|
||||
slarz.f slarzb.f slarzt.f slasy2.f slasyf.f slasyf_rook.f
|
||||
slatbs.f slatdf.f slatps.f slatrd.f slatrs.f slatrz.f DEPRECATED/slatzm.f
|
||||
sopgtr.f sopmtr.f sorg2l.f sorg2r.f
|
||||
sorgbr.f sorghr.f sorgl2.f sorglq.f sorgql.f sorgqr.f sorgr2.f
|
||||
sorgrq.f sorgtr.f sorm2l.f sorm2r.f sorm22.f
|
||||
sormbr.f sormhr.f sorml2.f sormlq.f sormql.f sormqr.f sormr2.f
|
||||
sormr3.f sormrq.f sormrz.f sormtr.f spbcon.f spbequ.f spbrfs.f
|
||||
spbstf.f spbsv.f spbsvx.f
|
||||
spbtf2.f spbtrf.f spbtrs.f spocon.f spoequ.f sporfs.f sposv.f
|
||||
sposvx.f spstrf.f spstf2.f
|
||||
sppcon.f sppequ.f
|
||||
spprfs.f sppsv.f sppsvx.f spptrf.f spptri.f spptrs.f sptcon.f
|
||||
spteqr.f sptrfs.f sptsv.f sptsvx.f spttrs.f sptts2.f srscl.f
|
||||
ssbev.f ssbevd.f ssbevx.f ssbgst.f ssbgv.f ssbgvd.f ssbgvx.f
|
||||
ssbtrd.f sspcon.f sspev.f sspevd.f sspevx.f sspgst.f
|
||||
sspgv.f sspgvd.f sspgvx.f ssprfs.f sspsv.f sspsvx.f ssptrd.f
|
||||
ssptrf.f ssptri.f ssptrs.f sstegr.f sstein.f sstev.f sstevd.f sstevr.f
|
||||
sstevx.f
|
||||
ssycon.f ssyev.f ssyevd.f ssyevr.f ssyevx.f ssygs2.f
|
||||
ssygst.f ssygv.f ssygvd.f ssygvx.f ssyrfs.f ssysv.f ssysvx.f
|
||||
ssytd2.f ssytf2.f ssytrd.f ssytrf.f ssytri.f ssytri2.f ssytri2x.f
|
||||
ssyswapr.f ssytrs.f ssytrs2.f ssyconv.f
|
||||
ssytf2_rook.f ssytrf_rook.f ssytrs_rook.f
|
||||
ssytri_rook.f ssycon_rook.f ssysv_rook.f
|
||||
stbcon.f
|
||||
stbrfs.f stbtrs.f stgevc.f stgex2.f stgexc.f stgsen.f
|
||||
stgsja.f stgsna.f stgsy2.f stgsyl.f stpcon.f stprfs.f stptri.f
|
||||
stptrs.f
|
||||
strcon.f strevc.f strevc3.f strexc.f strrfs.f strsen.f strsna.f strsyl.f
|
||||
strtrs.f DEPRECATED/stzrqf.f stzrzf.f sstemr.f
|
||||
slansf.f spftrf.f spftri.f spftrs.f ssfrk.f stfsm.f stftri.f stfttp.f
|
||||
stfttr.f stpttf.f stpttr.f strttf.f strttp.f
|
||||
sgejsv.f sgesvj.f sgsvj0.f sgsvj1.f
|
||||
sgeequb.f ssyequb.f spoequb.f sgbequb.f
|
||||
sbbcsd.f slapmr.f sorbdb.f sorbdb1.f sorbdb2.f sorbdb3.f sorbdb4.f
|
||||
sorbdb5.f sorbdb6.f sorcsd.f sorcsd2by1.f
|
||||
sgeqrt.f sgeqrt2.f sgeqrt3.f sgemqrt.f
|
||||
stpqrt.f stpqrt2.f stpmqrt.f stprfb.f spotri.f
|
||||
sgelq.f sgelqt.f sgelqt3.f sgemlq.f sgemlqt.f sgemqr.f sgeqr.f sgetsls.f
|
||||
slamswlq.f slamtsqr.f slaswlq.f slatsqr.f stplqt.f stplqt2.f stpmlqt.f
|
||||
ssysv_aa.f ssytrf_aa.f ssytrs_aa.f slasyf_aa.f ssytf2_rk.f slasyf_rk.f
|
||||
ssytrf_rk.f ssytrs_3.f ssycon_3.f ssytri_3.f ssytri_3x.f ssysv_rk.f
|
||||
ssb2st_kernels.f ssbev_2stage.f ssbevd_2stage.f ssbevx_2stage.f
|
||||
ssyev_2stage.f ssyevd_2stage.f ssyevr_2stage.f ssyevx_2stage.f
|
||||
ssygv_2stage.f ssytrd_2stage.f ssytrd_sb2st.F ssytrd_sy2sb.f slarfy.f
|
||||
)
|
||||
sbdsvdx.f sgbbrd.f sgbcon.f sgbequ.f sgbrfs.f sgbsv.f
|
||||
sgbsvx.f sgbtf2.f sgbtrf.f sgbtrs.f sgebak.f sgebal.f sgebd2.f
|
||||
sgebrd.f sgecon.f sgeequ.f sgees.f sgeesx.f sgeev.f sgeevx.f
|
||||
sgehd2.f sgehrd.f sgelq2.f sgelqf.f
|
||||
sgels.f sgelsd.f sgelss.f sgelsy.f sgeql2.f sgeqlf.f
|
||||
sgeqp3.f sgeqr2.f sgeqr2p.f sgeqrf.f sgeqrfp.f sgerfs.f sgerq2.f sgerqf.f
|
||||
sgesc2.f sgesdd.f sgesvd.f sgesvdx.f sgesvx.f sgetc2.f
|
||||
sgetrf2.f sgetri.f
|
||||
sggbak.f sggbal.f
|
||||
sgges.f sgges3.f sggesx.f sggev.f sggev3.f sggevx.f
|
||||
sggglm.f sgghrd.f sgghd3.f sgglse.f sggqrf.f
|
||||
sggrqf.f sggsvd3.f sggsvp3.f sgtcon.f sgtrfs.f sgtsv.f
|
||||
sgtsvx.f sgttrf.f sgttrs.f sgtts2.f shgeqz.f
|
||||
shsein.f shseqr.f slabrd.f slacon.f slacn2.f
|
||||
slaein.f slaexc.f slag2.f slags2.f slagtm.f slagv2.f slahqr.f
|
||||
slahr2.f slaic1.f slaln2.f slals0.f slalsa.f slalsd.f
|
||||
slangb.f slange.f slangt.f slanhs.f slansb.f slansp.f
|
||||
slansy.f slantb.f slantp.f slantr.f slanv2.f
|
||||
slapll.f slapmt.f
|
||||
slaqgb.f slaqge.f slaqp2.f slaqps.f slaqsb.f slaqsp.f slaqsy.f
|
||||
slaqr0.f slaqr1.f slaqr2.f slaqr3.f slaqr4.f slaqr5.f
|
||||
slaqtr.f slar1v.f slar2v.f ilaslr.f ilaslc.f
|
||||
slarf.f slarfb.f slarfg.f slarfgp.f slarft.f slarfx.f slarfy.f slargv.f
|
||||
slarrv.f slartv.f
|
||||
slarz.f slarzb.f slarzt.f slasy2.f
|
||||
slasyf.f slasyf_rook.f slasyf_rk.f slasyf_aa.f
|
||||
slatbs.f slatdf.f slatps.f slatrd.f slatrs.f slatrz.f
|
||||
sopgtr.f sopmtr.f sorg2l.f sorg2r.f
|
||||
sorgbr.f sorghr.f sorgl2.f sorglq.f sorgql.f sorgqr.f sorgr2.f
|
||||
sorgrq.f sorgtr.f sorm2l.f sorm2r.f sorm22.f
|
||||
sormbr.f sormhr.f sorml2.f sormlq.f sormql.f sormqr.f sormr2.f
|
||||
sormr3.f sormrq.f sormrz.f sormtr.f spbcon.f spbequ.f spbrfs.f
|
||||
spbstf.f spbsv.f spbsvx.f
|
||||
spbtf2.f spbtrf.f spbtrs.f spocon.f spoequ.f sporfs.f sposv.f
|
||||
sposvx.f spotrf2.f spotri.f spstrf.f spstf2.f
|
||||
sppcon.f sppequ.f
|
||||
spprfs.f sppsv.f sppsvx.f spptrf.f spptri.f spptrs.f sptcon.f
|
||||
spteqr.f sptrfs.f sptsv.f sptsvx.f spttrs.f sptts2.f srscl.f
|
||||
ssbev.f ssbevd.f ssbevx.f ssbgst.f ssbgv.f ssbgvd.f ssbgvx.f
|
||||
ssbtrd.f sspcon.f sspev.f sspevd.f sspevx.f sspgst.f
|
||||
sspgv.f sspgvd.f sspgvx.f ssprfs.f sspsv.f sspsvx.f ssptrd.f
|
||||
ssptrf.f ssptri.f ssptrs.f sstegr.f sstein.f sstev.f sstevd.f sstevr.f
|
||||
sstevx.f ssycon.f ssyev.f ssyevd.f ssyevr.f ssyevx.f ssygs2.f
|
||||
ssygst.f ssygv.f ssygvd.f ssygvx.f ssyrfs.f ssysv.f ssysvx.f
|
||||
ssytd2.f ssytf2.f ssytrd.f ssytrf.f ssytri.f ssytri2.f ssytri2x.f
|
||||
ssyswapr.f ssytrs.f ssytrs2.f
|
||||
ssyconv.f ssyconvf.f ssyconvf_rook.f
|
||||
ssysv_aa.f ssysv_aa_2stage.f ssytrf_aa.f ssytrf_aa_2stage.f ssytrs_aa.f ssytrs_aa_2stage.f
|
||||
ssytf2_rook.f ssytrf_rook.f ssytrs_rook.f
|
||||
ssytri_rook.f ssycon_rook.f ssysv_rook.f
|
||||
ssytf2_rk.f ssytrf_rk.f ssytrs_3.f
|
||||
ssytri_3.f ssytri_3x.f ssycon_3.f ssysv_rk.f
|
||||
ssysv_aa.f ssytrf_aa.f ssytrs_aa.f
|
||||
stbcon.f
|
||||
stbrfs.f stbtrs.f stgevc.f stgex2.f stgexc.f stgsen.f
|
||||
stgsja.f stgsna.f stgsy2.f stgsyl.f stpcon.f stprfs.f stptri.f
|
||||
stptrs.f
|
||||
strcon.f strevc.f strevc3.f strexc.f strrfs.f strsen.f strsna.f strsyl.f
|
||||
strtrs.f stzrzf.f sstemr.f
|
||||
slansf.f spftrf.f spftri.f spftrs.f ssfrk.f stfsm.f stftri.f stfttp.f
|
||||
stfttr.f stpttf.f stpttr.f strttf.f strttp.f
|
||||
sgejsv.f sgesvj.f sgsvj0.f sgsvj1.f
|
||||
sgeequb.f ssyequb.f spoequb.f sgbequb.f
|
||||
sbbcsd.f slapmr.f sorbdb.f sorbdb1.f sorbdb2.f sorbdb3.f sorbdb4.f
|
||||
sorbdb5.f sorbdb6.f sorcsd.f sorcsd2by1.f
|
||||
sgeqrt.f sgeqrt2.f sgeqrt3.f sgemqrt.f
|
||||
stpqrt.f stpqrt2.f stpmqrt.f stprfb.f
|
||||
sgelqt.f sgelqt3.f sgemlqt.f
|
||||
sgetsls.f sgeqr.f slatsqr.f slamtsqr.f sgemqr.f
|
||||
sgelq.f slaswlq.f slamswlq.f sgemlq.f
|
||||
stplqt.f stplqt2.f stpmlqt.f
|
||||
ssytrd_2stage.f ssytrd_sy2sb.f ssytrd_sb2st.F ssb2st_kernels.f
|
||||
ssyevd_2stage.f ssyev_2stage.f ssyevx_2stage.f ssyevr_2stage.f
|
||||
ssbev_2stage.f ssbevx_2stage.f ssbevd_2stage.f ssygv_2stage.f)
|
||||
|
||||
set(DSLASRC spotrs.f spotrf2.f)
|
||||
set(SXLASRC sgesvxx.f sgerfsx.f sla_gerfsx_extended.f sla_geamv.f
|
||||
sla_gercond.f sla_gerpvgrw.f ssysvxx.f ssyrfsx.f
|
||||
sla_syrfsx_extended.f sla_syamv.f sla_syrcond.f sla_syrpvgrw.f
|
||||
sposvxx.f sporfsx.f sla_porfsx_extended.f sla_porcond.f
|
||||
sla_porpvgrw.f sgbsvxx.f sgbrfsx.f sla_gbrfsx_extended.f
|
||||
sla_gbamv.f sla_gbrcond.f sla_gbrpvgrw.f sla_lin_berr.f slarscl2.f
|
||||
slascl2.f sla_wwaddw.f)
|
||||
|
||||
set(CLASRC
|
||||
cbdsqr.f cgbbrd.f cgbcon.f cgbequ.f cgbrfs.f cgbsv.f cgbsvx.f
|
||||
cgbtf2.f cgbtrf.f cgbtrs.f cgebak.f cgebal.f cgebd2.f cgebrd.f
|
||||
cgecon.f cgeequ.f cgees.f cgeesx.f cgeev.f cgeevx.f
|
||||
DEPRECATED/cgegs.f DEPRECATED/cgegv.f cgehd2.f cgehrd.f cgelq2.f cgelqf.f
|
||||
cgels.f cgelsd.f cgelss.f DEPRECATED/cgelsx.f cgelsy.f cgeql2.f cgeqlf.f cgeqp3.f
|
||||
DEPRECATED/cgeqpf.f cgeqr2.f cgeqr2p.f cgeqrf.f cgeqrfp.f cgerfs.f
|
||||
cgerq2.f cgerqf.f cgesc2.f cgesdd.f cgesvd.f
|
||||
cgesvx.f cgetc2.f cgetri.f
|
||||
cggbak.f cggbal.f cgges.f cggesx.f cggev.f cggevx.f cggglm.f
|
||||
cgghrd.f cgglse.f cggqrf.f cggrqf.f
|
||||
DEPRECATED/cggsvd.f DEPRECATED/cggsvp.f
|
||||
cgtcon.f cgtrfs.f cgtsv.f cgtsvx.f cgttrf.f cgttrs.f cgtts2.f chbev.f
|
||||
chbevd.f chbevx.f chbgst.f chbgv.f chbgvd.f chbgvx.f chbtrd.f
|
||||
checon.f cheev.f cheevd.f cheevr.f cheevx.f chegs2.f chegst.f
|
||||
chegv.f chegvd.f chegvx.f cherfs.f chesv.f chesvx.f chetd2.f
|
||||
chetf2.f chetrd.f
|
||||
chetrf.f chetri.f chetri2.f chetri2x.f cheswapr.f
|
||||
chetrs.f chetrs2.f
|
||||
chetf2_rook.f chetrf_rook.f chetri_rook.f chetrs_rook.f checon_rook.f chesv_rook.f
|
||||
chgeqz.f chpcon.f chpev.f chpevd.f
|
||||
chpevx.f chpgst.f chpgv.f chpgvd.f chpgvx.f chprfs.f chpsv.f
|
||||
chpsvx.f
|
||||
chptrd.f chptrf.f chptri.f chptrs.f chsein.f chseqr.f clabrd.f
|
||||
clacgv.f clacon.f clacn2.f clacp2.f clacpy.f clacrm.f clacrt.f cladiv.f
|
||||
claed0.f claed7.f claed8.f
|
||||
claein.f claesy.f claev2.f clags2.f clagtm.f
|
||||
clahef.f clahef_rook.f clahqr.f
|
||||
DEPRECATED/clahrd.f clahr2.f claic1.f clals0.f clalsa.f clalsd.f clangb.f clange.f clangt.f
|
||||
clanhb.f clanhe.f
|
||||
clanhp.f clanhs.f clanht.f clansb.f clansp.f clansy.f clantb.f
|
||||
clantp.f clantr.f clapll.f clapmt.f clarcm.f claqgb.f claqge.f
|
||||
claqhb.f claqhe.f claqhp.f claqp2.f claqps.f claqsb.f
|
||||
claqr0.f claqr1.f claqr2.f claqr3.f claqr4.f claqr5.f
|
||||
claqsp.f claqsy.f clar1v.f clar2v.f ilaclr.f ilaclc.f
|
||||
clarf.f clarfb.f clarfg.f clarft.f clarfgp.f
|
||||
clarfx.f clargv.f clarnv.f clarrv.f clartg.f clartv.f
|
||||
clarz.f clarzb.f clarzt.f clascl.f claset.f clasr.f classq.f
|
||||
clasyf.f clasyf_rook.f clatbs.f clatdf.f clatps.f clatrd.f clatrs.f clatrz.f
|
||||
DEPRECATED/clatzm.f cpbcon.f cpbequ.f cpbrfs.f cpbstf.f cpbsv.f
|
||||
cpbsvx.f cpbtf2.f cpbtrf.f cpbtrs.f cpocon.f cpoequ.f cporfs.f
|
||||
cposv.f cposvx.f cpstrf.f cpstf2.f
|
||||
cppcon.f cppequ.f cpprfs.f cppsv.f cppsvx.f cpptrf.f cpptri.f cpptrs.f
|
||||
cptcon.f cpteqr.f cptrfs.f cptsv.f cptsvx.f cpttrf.f cpttrs.f cptts2.f
|
||||
crot.f cspcon.f csprfs.f cspsv.f
|
||||
cspsvx.f csptrf.f csptri.f csptrs.f csrscl.f cstedc.f
|
||||
cstegr.f cstein.f csteqr.f
|
||||
csycon.f
|
||||
csyrfs.f csysv.f csysvx.f csytf2.f csytrf.f csytri.f csytri2.f csytri2x.f
|
||||
csyswapr.f csytrs.f csytrs2.f csyconv.f
|
||||
csytf2_rook.f csytrf_rook.f csytrs_rook.f
|
||||
csytri_rook.f csycon_rook.f csysv_rook.f
|
||||
ctbcon.f ctbrfs.f ctbtrs.f ctgevc.f ctgex2.f
|
||||
ctgexc.f ctgsen.f ctgsja.f ctgsna.f ctgsy2.f ctgsyl.f ctpcon.f
|
||||
ctprfs.f ctptri.f
|
||||
ctptrs.f ctrcon.f ctrevc.f ctrevc3.f ctrexc.f ctrrfs.f ctrsen.f ctrsna.f
|
||||
ctrsyl.f ctrtrs.f DEPRECATED/ctzrqf.f ctzrzf.f cung2l.f cung2r.f
|
||||
cungbr.f cunghr.f cungl2.f cunglq.f cungql.f cungqr.f cungr2.f
|
||||
cungrq.f cungtr.f cunm2l.f cunm2r.f cunmbr.f cunmhr.f cunml2.f
|
||||
cunmlq.f cunmql.f cunmqr.f cunmr2.f cunmr3.f cunmrq.f cunmrz.f
|
||||
cunmtr.f cupgtr.f cupmtr.f icmax1.f scsum1.f cstemr.f
|
||||
chfrk.f ctfttp.f clanhf.f cpftrf.f cpftri.f cpftrs.f ctfsm.f ctftri.f
|
||||
ctfttr.f ctpttf.f ctpttr.f ctrttf.f ctrttp.f
|
||||
cgeequb.f cgbequb.f csyequb.f cpoequb.f cheequb.f
|
||||
cbbcsd.f clapmr.f cunbdb.f cunbdb1.f cunbdb2.f cunbdb3.f cunbdb4.f
|
||||
cunbdb5.f cunbdb6.f cuncsd.f cuncsd2by1.f
|
||||
cgeqrt.f cgeqrt2.f cgeqrt3.f cgemqrt.f
|
||||
ctpqrt.f ctpqrt2.f ctpmqrt.f ctprfb.f cpotri.f
|
||||
cgelq.f cgelqt.f cgelqt3.f cgemlq.f cgemlqt.f cgemqr.f cgeqr.f cgetsls.f
|
||||
clamswlq.f clamtsqr.f claswlq.f clatsqr.f ctplqt.f ctplqt2.f ctpmlqt.f
|
||||
chesv_aa.f chetrf_aa.f chetrs_aa.f clahef_aa.f csytf2_rk.f clasyf_rk.f
|
||||
csytrf_rk.f csytrs_3.f csycon_3.f csytri_3.f csytri_3x.f csysv_rk.f
|
||||
chetf2_rk.f clahef_rk.f chetrf_rk.f chetrs_3.f checon_3.f chetri_3.f
|
||||
chetri_3x.f chesv_rk.f chb2st_kernels.f chbev_2stage.f chbevd_2stage.f
|
||||
chbevx_2stage.f cheev_2stage.f cheevd_2stage.f cheevr_2stage.f cheevx_2stage.f
|
||||
chegv_2stage.f chetrd_2stage.f chetrd_hb2st.F chetrd_he2hb.f clarfy.f
|
||||
)
|
||||
cbdsqr.f cgbbrd.f cgbcon.f cgbequ.f cgbrfs.f cgbsv.f cgbsvx.f
|
||||
cgbtf2.f cgbtrf.f cgbtrs.f cgebak.f cgebal.f cgebd2.f cgebrd.f
|
||||
cgecon.f cgeequ.f cgees.f cgeesx.f cgeev.f cgeevx.f
|
||||
cgehd2.f cgehrd.f cgelq2.f cgelqf.f
|
||||
cgels.f cgelsd.f cgelss.f cgelsy.f cgeql2.f cgeqlf.f cgeqp3.f
|
||||
cgeqr2.f cgeqr2p.f cgeqrf.f cgeqrfp.f cgerfs.f cgerq2.f cgerqf.f
|
||||
cgesc2.f cgesdd.f cgesvd.f cgesvdx.f
|
||||
cgesvj.f cgejsv.f cgsvj0.f cgsvj1.f
|
||||
cgesvx.f cgetc2.f cgetrf2.f
|
||||
cgetri.f
|
||||
cggbak.f cggbal.f
|
||||
cgges.f cgges3.f cggesx.f cggev.f cggev3.f cggevx.f
|
||||
cggglm.f cgghrd.f cgghd3.f cgglse.f cggqrf.f cggrqf.f
|
||||
cggsvd3.f cggsvp3.f
|
||||
cgtcon.f cgtrfs.f cgtsv.f cgtsvx.f cgttrf.f cgttrs.f cgtts2.f chbev.f
|
||||
chbevd.f chbevx.f chbgst.f chbgv.f chbgvd.f chbgvx.f chbtrd.f
|
||||
checon.f cheev.f cheevd.f cheevr.f cheevx.f chegs2.f chegst.f
|
||||
chegv.f chegvd.f chegvx.f cherfs.f chesv.f chesvx.f chetd2.f
|
||||
chetf2.f chetrd.f
|
||||
chetrf.f chetri.f chetri2.f chetri2x.f cheswapr.f
|
||||
chetrs.f chetrs2.f
|
||||
chetf2_rook.f chetrf_rook.f chetri_rook.f
|
||||
chetrs_rook.f checon_rook.f chesv_rook.f
|
||||
chetf2_rk.f chetrf_rk.f chetri_3.f chetri_3x.f
|
||||
chetrs_3.f checon_3.f chesv_rk.f
|
||||
chesv_aa.f chesv_aa_2stage.f chetrf_aa.f chetrf_aa_2stage.f chetrs_aa.f chetrs_aa_2stage.f
|
||||
chgeqz.f chpcon.f chpev.f chpevd.f
|
||||
chpevx.f chpgst.f chpgv.f chpgvd.f chpgvx.f chprfs.f chpsv.f
|
||||
chpsvx.f
|
||||
chptrd.f chptrf.f chptri.f chptrs.f chsein.f chseqr.f clabrd.f
|
||||
clacgv.f clacon.f clacn2.f clacp2.f clacpy.f clacrm.f clacrt.f cladiv.f
|
||||
claed0.f claed7.f claed8.f
|
||||
claein.f claesy.f claev2.f clags2.f clagtm.f
|
||||
clahef.f clahef_rook.f clahef_rk.f clahef_aa.f clahqr.f
|
||||
clahr2.f claic1.f clals0.f clalsa.f clalsd.f clangb.f clange.f clangt.f
|
||||
clanhb.f clanhe.f
|
||||
clanhp.f clanhs.f clanht.f clansb.f clansp.f clansy.f clantb.f
|
||||
clantp.f clantr.f clapll.f clapmt.f clarcm.f claqgb.f claqge.f
|
||||
claqhb.f claqhe.f claqhp.f claqp2.f claqps.f claqsb.f
|
||||
claqr0.f claqr1.f claqr2.f claqr3.f claqr4.f claqr5.f
|
||||
claqsp.f claqsy.f clar1v.f clar2v.f ilaclr.f ilaclc.f
|
||||
clarf.f clarfb.f clarfg.f clarfgp.f clarft.f
|
||||
clarfx.f clarfy.f clargv.f clarnv.f clarrv.f clartg.f clartv.f
|
||||
clarz.f clarzb.f clarzt.f clascl.f claset.f clasr.f classq.f
|
||||
clasyf.f clasyf_rook.f clasyf_rk.f clasyf_aa.f
|
||||
clatbs.f clatdf.f clatps.f clatrd.f clatrs.f clatrz.f
|
||||
cpbcon.f cpbequ.f cpbrfs.f cpbstf.f cpbsv.f
|
||||
cpbsvx.f cpbtf2.f cpbtrf.f cpbtrs.f cpocon.f cpoequ.f cporfs.f
|
||||
cposv.f cposvx.f cpotrf2.f cpotri.f cpstrf.f cpstf2.f
|
||||
cppcon.f cppequ.f cpprfs.f cppsv.f cppsvx.f cpptrf.f cpptri.f cpptrs.f
|
||||
cptcon.f cpteqr.f cptrfs.f cptsv.f cptsvx.f cpttrf.f cpttrs.f cptts2.f
|
||||
crot.f cspcon.f csprfs.f cspsv.f
|
||||
cspsvx.f csptrf.f csptri.f csptrs.f csrscl.f cstedc.f
|
||||
cstegr.f cstein.f csteqr.f csycon.f
|
||||
csyrfs.f csysv.f csysvx.f csytf2.f csytrf.f csytri.f
|
||||
csytri2.f csytri2x.f csyswapr.f
|
||||
csytrs.f csytrs2.f
|
||||
csyconv.f csyconvf.f csyconvf_rook.f
|
||||
csytf2_rook.f csytrf_rook.f csytrs_rook.f
|
||||
csytri_rook.f csycon_rook.f csysv_rook.f
|
||||
csytf2_rk.f csytrf_rk.f csytrf_aa.f csytrf_aa_2stage.f csytrs_3.f csytrs_aa.f csytrs_aa_2stage.f
|
||||
csytri_3.f csytri_3x.f csycon_3.f csysv_rk.f csysv_aa.f csysv_aa_2stage.f
|
||||
ctbcon.f ctbrfs.f ctbtrs.f ctgevc.f ctgex2.f
|
||||
ctgexc.f ctgsen.f ctgsja.f ctgsna.f ctgsy2.f ctgsyl.f ctpcon.f
|
||||
ctprfs.f ctptri.f
|
||||
ctptrs.f ctrcon.f ctrevc.f ctrevc3.f ctrexc.f ctrrfs.f ctrsen.f ctrsna.f
|
||||
ctrsyl.f ctrtrs.f ctzrzf.f cung2l.f cung2r.f
|
||||
cungbr.f cunghr.f cungl2.f cunglq.f cungql.f cungqr.f cungr2.f
|
||||
cungrq.f cungtr.f cunm2l.f cunm2r.f cunmbr.f cunmhr.f cunml2.f cunm22.f
|
||||
cunmlq.f cunmql.f cunmqr.f cunmr2.f cunmr3.f cunmrq.f cunmrz.f
|
||||
cunmtr.f cupgtr.f cupmtr.f icmax1.f scsum1.f cstemr.f
|
||||
chfrk.f ctfttp.f clanhf.f cpftrf.f cpftri.f cpftrs.f ctfsm.f ctftri.f
|
||||
ctfttr.f ctpttf.f ctpttr.f ctrttf.f ctrttp.f
|
||||
cgeequb.f cgbequb.f csyequb.f cpoequb.f cheequb.f
|
||||
cbbcsd.f clapmr.f cunbdb.f cunbdb1.f cunbdb2.f cunbdb3.f cunbdb4.f
|
||||
cunbdb5.f cunbdb6.f cuncsd.f cuncsd2by1.f
|
||||
cgeqrt.f cgeqrt2.f cgeqrt3.f cgemqrt.f
|
||||
ctpqrt.f ctpqrt2.f ctpmqrt.f ctprfb.f
|
||||
cgelqt.f cgelqt3.f cgemlqt.f
|
||||
cgetsls.f cgeqr.f clatsqr.f clamtsqr.f cgemqr.f
|
||||
cgelq.f claswlq.f clamswlq.f cgemlq.f
|
||||
ctplqt.f ctplqt2.f ctpmlqt.f
|
||||
chetrd_2stage.f chetrd_he2hb.f chetrd_hb2st.F chb2st_kernels.f
|
||||
cheevd_2stage.f cheev_2stage.f cheevx_2stage.f cheevr_2stage.f
|
||||
chbev_2stage.f chbevx_2stage.f chbevd_2stage.f chegv_2stage.f)
|
||||
|
||||
set(CXLASRC cgesvxx.f cgerfsx.f cla_gerfsx_extended.f cla_geamv.f
|
||||
cla_gercond_c.f cla_gercond_x.f cla_gerpvgrw.f
|
||||
csysvxx.f csyrfsx.f cla_syrfsx_extended.f cla_syamv.f
|
||||
cla_syrcond_c.f cla_syrcond_x.f cla_syrpvgrw.f
|
||||
cposvxx.f cporfsx.f cla_porfsx_extended.f
|
||||
cla_porcond_c.f cla_porcond_x.f cla_porpvgrw.f
|
||||
cgbsvxx.f cgbrfsx.f cla_gbrfsx_extended.f cla_gbamv.f
|
||||
cla_gbrcond_c.f cla_gbrcond_x.f cla_gbrpvgrw.f
|
||||
chesvxx.f cherfsx.f cla_herfsx_extended.f cla_heamv.f
|
||||
cla_hercond_c.f cla_hercond_x.f cla_herpvgrw.f
|
||||
cla_lin_berr.f clarscl2.f clascl2.f cla_wwaddw.f)
|
||||
|
||||
set(DLASRC
|
||||
dbdsvdx.f dgbbrd.f dgbcon.f dgbequ.f dgbrfs.f dgbsv.f
|
||||
dgbsvx.f dgbtf2.f dgbtrf.f dgbtrs.f dgebak.f dgebal.f dgebd2.f
|
||||
dgebrd.f dgecon.f dgeequ.f dgees.f dgeesx.f dgeev.f dgeevx.f
|
||||
dgehd2.f dgehrd.f dgelq2.f dgelqf.f
|
||||
dgels.f dgelsd.f dgelss.f dgelsy.f dgeql2.f dgeqlf.f
|
||||
dgeqp3.f dgeqr2.f dgeqr2p.f dgeqrf.f dgeqrfp.f dgerfs.f dgerq2.f dgerqf.f
|
||||
dgesc2.f dgesdd.f dgesvd.f dgesvdx.f dgesvx.f dgetc2.f
|
||||
dgetrf2.f dgetri.f
|
||||
dggbak.f dggbal.f
|
||||
dgges.f dgges3.f dggesx.f dggev.f dggev3.f dggevx.f
|
||||
dggglm.f dgghrd.f dgghd3.f dgglse.f dggqrf.f
|
||||
dggrqf.f dggsvd3.f dggsvp3.f dgtcon.f dgtrfs.f dgtsv.f
|
||||
dgtsvx.f dgttrf.f dgttrs.f dgtts2.f dhgeqz.f
|
||||
dhsein.f dhseqr.f dlabrd.f dlacon.f dlacn2.f
|
||||
dlaein.f dlaexc.f dlag2.f dlags2.f dlagtm.f dlagv2.f dlahqr.f
|
||||
dlahr2.f dlaic1.f dlaln2.f dlals0.f dlalsa.f dlalsd.f
|
||||
dlangb.f dlange.f dlangt.f dlanhs.f dlansb.f dlansp.f
|
||||
dlansy.f dlantb.f dlantp.f dlantr.f dlanv2.f
|
||||
dlapll.f dlapmt.f
|
||||
dlaqgb.f dlaqge.f dlaqp2.f dlaqps.f dlaqsb.f dlaqsp.f dlaqsy.f
|
||||
dlaqr0.f dlaqr1.f dlaqr2.f dlaqr3.f dlaqr4.f dlaqr5.f
|
||||
dlaqtr.f dlar1v.f dlar2v.f iladlr.f iladlc.f
|
||||
dlarf.f dlarfb.f dlarfg.f dlarfgp.f dlarft.f dlarfx.f dlarfy.f
|
||||
dlargv.f dlarrv.f dlartv.f
|
||||
dlarz.f dlarzb.f dlarzt.f dlasy2.f
|
||||
dlasyf.f dlasyf_rook.f dlasyf_rk.f dlasyf_aa.f
|
||||
dlatbs.f dlatdf.f dlatps.f dlatrd.f dlatrs.f dlatrz.f
|
||||
dopgtr.f dopmtr.f dorg2l.f dorg2r.f
|
||||
dorgbr.f dorghr.f dorgl2.f dorglq.f dorgql.f dorgqr.f dorgr2.f
|
||||
dorgrq.f dorgtr.f dorm2l.f dorm2r.f dorm22.f
|
||||
dormbr.f dormhr.f dorml2.f dormlq.f dormql.f dormqr.f dormr2.f
|
||||
dormr3.f dormrq.f dormrz.f dormtr.f dpbcon.f dpbequ.f dpbrfs.f
|
||||
dpbstf.f dpbsv.f dpbsvx.f
|
||||
dpbtf2.f dpbtrf.f dpbtrs.f dpocon.f dpoequ.f dporfs.f dposv.f
|
||||
dposvx.f dpotrf2.f dpotri.f dpotrs.f dpstrf.f dpstf2.f
|
||||
dppcon.f dppequ.f
|
||||
dpprfs.f dppsv.f dppsvx.f dpptrf.f dpptri.f dpptrs.f dptcon.f
|
||||
dpteqr.f dptrfs.f dptsv.f dptsvx.f dpttrs.f dptts2.f drscl.f
|
||||
dsbev.f dsbevd.f dsbevx.f dsbgst.f dsbgv.f dsbgvd.f dsbgvx.f
|
||||
dsbtrd.f dspcon.f dspev.f dspevd.f dspevx.f dspgst.f
|
||||
dspgv.f dspgvd.f dspgvx.f dsprfs.f dspsv.f dspsvx.f dsptrd.f
|
||||
dsptrf.f dsptri.f dsptrs.f dstegr.f dstein.f dstev.f dstevd.f dstevr.f
|
||||
dstevx.f dsycon.f dsyev.f dsyevd.f dsyevr.f
|
||||
dsyevx.f dsygs2.f dsygst.f dsygv.f dsygvd.f dsygvx.f dsyrfs.f
|
||||
dsysv.f dsysvx.f
|
||||
dsytd2.f dsytf2.f dsytrd.f dsytrf.f dsytri.f dsytrs.f dsytrs2.f
|
||||
dsytri2.f dsytri2x.f dsyswapr.f
|
||||
dsyconv.f dsyconvf.f dsyconvf_rook.f
|
||||
dsytf2_rook.f dsytrf_rook.f dsytrs_rook.f
|
||||
dsytri_rook.f dsycon_rook.f dsysv_rook.f
|
||||
dsytf2_rk.f dsytrf_rk.f dsytrs_3.f
|
||||
dsytri_3.f dsytri_3x.f dsycon_3.f dsysv_rk.f
|
||||
dsysv_aa.f dsysv_aa_2stage.f dsytrf_aa.f dsytrf_aa_2stage.f dsytrs_aa.f dsytrs_aa_2stage.f
|
||||
dtbcon.f
|
||||
dtbrfs.f dtbtrs.f dtgevc.f dtgex2.f dtgexc.f dtgsen.f
|
||||
dtgsja.f dtgsna.f dtgsy2.f dtgsyl.f dtpcon.f dtprfs.f dtptri.f
|
||||
dtptrs.f
|
||||
dtrcon.f dtrevc.f dtrevc3.f dtrexc.f dtrrfs.f dtrsen.f dtrsna.f dtrsyl.f
|
||||
dtrtrs.f dtzrzf.f dstemr.f
|
||||
dsgesv.f dsposv.f dlag2s.f slag2d.f dlat2s.f
|
||||
dlansf.f dpftrf.f dpftri.f dpftrs.f dsfrk.f dtfsm.f dtftri.f dtfttp.f
|
||||
dtfttr.f dtpttf.f dtpttr.f dtrttf.f dtrttp.f
|
||||
dgejsv.f dgesvj.f dgsvj0.f dgsvj1.f
|
||||
dgeequb.f dsyequb.f dpoequb.f dgbequb.f
|
||||
dbbcsd.f dlapmr.f dorbdb.f dorbdb1.f dorbdb2.f dorbdb3.f dorbdb4.f
|
||||
dorbdb5.f dorbdb6.f dorcsd.f dorcsd2by1.f
|
||||
dgeqrt.f dgeqrt2.f dgeqrt3.f dgemqrt.f
|
||||
dtpqrt.f dtpqrt2.f dtpmqrt.f dtprfb.f
|
||||
dgelqt.f dgelqt3.f dgemlqt.f
|
||||
dgetsls.f dgeqr.f dlatsqr.f dlamtsqr.f dgemqr.f
|
||||
dgelq.f dlaswlq.f dlamswlq.f dgemlq.f
|
||||
dtplqt.f dtplqt2.f dtpmlqt.f
|
||||
dsytrd_2stage.f dsytrd_sy2sb.f dsytrd_sb2st.F dsb2st_kernels.f
|
||||
dsyevd_2stage.f dsyev_2stage.f dsyevx_2stage.f dsyevr_2stage.f
|
||||
dsbev_2stage.f dsbevx_2stage.f dsbevd_2stage.f dsygv_2stage.f)
|
||||
|
||||
set(DXLASRC dgesvxx.f dgerfsx.f dla_gerfsx_extended.f dla_geamv.f
|
||||
dla_gercond.f dla_gerpvgrw.f dsysvxx.f dsyrfsx.f
|
||||
dla_syrfsx_extended.f dla_syamv.f dla_syrcond.f dla_syrpvgrw.f
|
||||
dposvxx.f dporfsx.f dla_porfsx_extended.f dla_porcond.f
|
||||
dla_porpvgrw.f dgbsvxx.f dgbrfsx.f dla_gbrfsx_extended.f
|
||||
dla_gbamv.f dla_gbrcond.f dla_gbrpvgrw.f dla_lin_berr.f dlarscl2.f
|
||||
dlascl2.f dla_wwaddw.f)
|
||||
|
||||
set(ZLASRC
|
||||
zbdsqr.f zgbbrd.f zgbcon.f zgbequ.f zgbrfs.f zgbsv.f zgbsvx.f
|
||||
zgbtf2.f zgbtrf.f zgbtrs.f zgebak.f zgebal.f zgebd2.f zgebrd.f
|
||||
zgecon.f zgeequ.f zgees.f zgeesx.f zgeev.f zgeevx.f
|
||||
zgehd2.f zgehrd.f zgelq2.f zgelqf.f
|
||||
zgels.f zgelsd.f zgelss.f zgelsy.f zgeql2.f zgeqlf.f zgeqp3.f
|
||||
zgeqr2.f zgeqr2p.f zgeqrf.f zgeqrfp.f zgerfs.f zgerq2.f zgerqf.f
|
||||
zgesc2.f zgesdd.f zgesvd.f zgesvdx.f zgesvx.f
|
||||
zgesvj.f zgejsv.f zgsvj0.f zgsvj1.f
|
||||
zgetc2.f zgetrf2.f
|
||||
zgetri.f
|
||||
zggbak.f zggbal.f
|
||||
zgges.f zgges3.f zggesx.f zggev.f zggev3.f zggevx.f
|
||||
zggglm.f zgghrd.f zgghd3.f zgglse.f zggqrf.f zggrqf.f
|
||||
zggsvd3.f zggsvp3.f
|
||||
zgtcon.f zgtrfs.f zgtsv.f zgtsvx.f zgttrf.f zgttrs.f zgtts2.f zhbev.f
|
||||
zhbevd.f zhbevx.f zhbgst.f zhbgv.f zhbgvd.f zhbgvx.f zhbtrd.f
|
||||
zhecon.f zheev.f zheevd.f zheevr.f zheevx.f zhegs2.f zhegst.f
|
||||
zhegv.f zhegvd.f zhegvx.f zherfs.f zhesv.f zhesvx.f zhetd2.f
|
||||
zhetf2.f zhetrd.f
|
||||
zhetrf.f zhetri.f zhetri2.f zhetri2x.f zheswapr.f
|
||||
zhetrs.f zhetrs2.f
|
||||
zhetf2_rook.f zhetrf_rook.f zhetri_rook.f
|
||||
zhetrs_rook.f zhecon_rook.f zhesv_rook.f
|
||||
zhetf2_rk.f zhetrf_rk.f zhetri_3.f zhetri_3x.f
|
||||
zhetrs_3.f zhecon_3.f zhesv_rk.f
|
||||
zhesv_aa.f zhesv_aa_2stage.f zhetrf_aa.f zhetrf_aa_2stage.f zhetrs_aa.f zhetrs_aa_2stage.f
|
||||
zhgeqz.f zhpcon.f zhpev.f zhpevd.f
|
||||
zhpevx.f zhpgst.f zhpgv.f zhpgvd.f zhpgvx.f zhprfs.f zhpsv.f
|
||||
zhpsvx.f
|
||||
zhptrd.f zhptrf.f zhptri.f zhptrs.f zhsein.f zhseqr.f zlabrd.f
|
||||
zlacgv.f zlacon.f zlacn2.f zlacp2.f zlacpy.f zlacrm.f zlacrt.f zladiv.f
|
||||
zlaed0.f zlaed7.f zlaed8.f
|
||||
zlaein.f zlaesy.f zlaev2.f zlags2.f zlagtm.f
|
||||
zlahef.f zlahef_rook.f zlahef_rk.f zlahef_aa.f zlahqr.f
|
||||
zlahr2.f zlaic1.f zlals0.f zlalsa.f zlalsd.f zlangb.f zlange.f
|
||||
zlangt.f zlanhb.f
|
||||
zlanhe.f
|
||||
zlanhp.f zlanhs.f zlanht.f zlansb.f zlansp.f zlansy.f zlantb.f
|
||||
zlantp.f zlantr.f zlapll.f zlapmt.f zlaqgb.f zlaqge.f
|
||||
zlaqhb.f zlaqhe.f zlaqhp.f zlaqp2.f zlaqps.f zlaqsb.f
|
||||
zlaqr0.f zlaqr1.f zlaqr2.f zlaqr3.f zlaqr4.f zlaqr5.f
|
||||
zlaqsp.f zlaqsy.f zlar1v.f zlar2v.f ilazlr.f ilazlc.f
|
||||
zlarcm.f zlarf.f zlarfb.f
|
||||
zlarfg.f zlarfgp.f zlarft.f
|
||||
zlarfx.f zlarfy.f zlargv.f zlarnv.f zlarrv.f zlartg.f zlartv.f
|
||||
zlarz.f zlarzb.f zlarzt.f zlascl.f zlaset.f zlasr.f
|
||||
zlassq.f zlasyf.f zlasyf_rook.f zlasyf_rk.f zlasyf_aa.f
|
||||
zlatbs.f zlatdf.f zlatps.f zlatrd.f zlatrs.f zlatrz.f
|
||||
zpbcon.f zpbequ.f zpbrfs.f zpbstf.f zpbsv.f
|
||||
zpbsvx.f zpbtf2.f zpbtrf.f zpbtrs.f zpocon.f zpoequ.f zporfs.f
|
||||
zposv.f zposvx.f zpotrf2.f zpotri.f zpotrs.f zpstrf.f zpstf2.f
|
||||
zppcon.f zppequ.f zpprfs.f zppsv.f zppsvx.f zpptrf.f zpptri.f zpptrs.f
|
||||
zptcon.f zpteqr.f zptrfs.f zptsv.f zptsvx.f zpttrf.f zpttrs.f zptts2.f
|
||||
zrot.f zspcon.f zsprfs.f zspsv.f
|
||||
zspsvx.f zsptrf.f zsptri.f zsptrs.f zdrscl.f zstedc.f
|
||||
zstegr.f zstein.f zsteqr.f zsycon.f
|
||||
zsyrfs.f zsysv.f zsysvx.f zsytf2.f zsytrf.f zsytri.f
|
||||
zsytri2.f zsytri2x.f zsyswapr.f
|
||||
zsytrs.f zsytrs2.f
|
||||
zsyconv.f zsyconvf.f zsyconvf_rook.f
|
||||
zsytf2_rook.f zsytrf_rook.f zsytrs_rook.f zsytrs_aa.f zsytrs_aa_2stage.f
|
||||
zsytri_rook.f zsycon_rook.f zsysv_rook.f
|
||||
zsytf2_rk.f zsytrf_rk.f zsytrf_aa.f zsytrf_aa_2stage.f zsytrs_3.f
|
||||
zsytri_3.f zsytri_3x.f zsycon_3.f zsysv_rk.f zsysv_aa.f zsysv_aa_2stage.f
|
||||
ztbcon.f ztbrfs.f ztbtrs.f ztgevc.f ztgex2.f
|
||||
ztgexc.f ztgsen.f ztgsja.f ztgsna.f ztgsy2.f ztgsyl.f ztpcon.f
|
||||
ztprfs.f ztptri.f
|
||||
ztptrs.f ztrcon.f ztrevc.f ztrevc3.f ztrexc.f ztrrfs.f ztrsen.f ztrsna.f
|
||||
ztrsyl.f ztrtrs.f ztzrzf.f zung2l.f
|
||||
zung2r.f zungbr.f zunghr.f zungl2.f zunglq.f zungql.f zungqr.f zungr2.f
|
||||
zungrq.f zungtr.f zunm2l.f zunm2r.f zunmbr.f zunmhr.f zunml2.f zunm22.f
|
||||
zunmlq.f zunmql.f zunmqr.f zunmr2.f zunmr3.f zunmrq.f zunmrz.f
|
||||
zunmtr.f zupgtr.f
|
||||
zupmtr.f izmax1.f dzsum1.f zstemr.f
|
||||
zcgesv.f zcposv.f zlag2c.f clag2z.f zlat2c.f
|
||||
zhfrk.f ztfttp.f zlanhf.f zpftrf.f zpftri.f zpftrs.f ztfsm.f ztftri.f
|
||||
ztfttr.f ztpttf.f ztpttr.f ztrttf.f ztrttp.f
|
||||
zgeequb.f zgbequb.f zsyequb.f zpoequb.f zheequb.f
|
||||
zbbcsd.f zlapmr.f zunbdb.f zunbdb1.f zunbdb2.f zunbdb3.f zunbdb4.f
|
||||
zunbdb5.f zunbdb6.f zuncsd.f zuncsd2by1.f
|
||||
zgeqrt.f zgeqrt2.f zgeqrt3.f zgemqrt.f
|
||||
ztpqrt.f ztpqrt2.f ztpmqrt.f ztprfb.f
|
||||
ztplqt.f ztplqt2.f ztpmlqt.f
|
||||
zgelqt.f zgelqt3.f zgemlqt.f
|
||||
zgetsls.f zgeqr.f zlatsqr.f zlamtsqr.f zgemqr.f
|
||||
zgelq.f zlaswlq.f zlamswlq.f zgemlq.f
|
||||
zhetrd_2stage.f zhetrd_he2hb.f zhetrd_hb2st.F zhb2st_kernels.f
|
||||
zheevd_2stage.f zheev_2stage.f zheevx_2stage.f zheevr_2stage.f
|
||||
zhbev_2stage.f zhbevx_2stage.f zhbevd_2stage.f zhegv_2stage.f)
|
||||
|
||||
set(ZXLASRC zgesvxx.f zgerfsx.f zla_gerfsx_extended.f zla_geamv.f
|
||||
zla_gercond_c.f zla_gercond_x.f zla_gerpvgrw.f zsysvxx.f zsyrfsx.f
|
||||
zla_syrfsx_extended.f zla_syamv.f zla_syrcond_c.f zla_syrcond_x.f
|
||||
zla_syrpvgrw.f zposvxx.f zporfsx.f zla_porfsx_extended.f
|
||||
zla_porcond_c.f zla_porcond_x.f zla_porpvgrw.f zgbsvxx.f zgbrfsx.f
|
||||
zla_gbrfsx_extended.f zla_gbamv.f zla_gbrcond_c.f zla_gbrcond_x.f
|
||||
zla_gbrpvgrw.f zhesvxx.f zherfsx.f zla_herfsx_extended.f
|
||||
zla_heamv.f zla_hercond_c.f zla_hercond_x.f zla_herpvgrw.f
|
||||
zla_lin_berr.f zlarscl2.f zlascl2.f zla_wwaddw.f)
|
||||
|
||||
|
||||
if(USE_XBLAS)
|
||||
set(ALLXOBJ ${SXLASRC} ${DXLASRC} ${CXLASRC} ${ZXLASRC})
|
||||
endif()
|
||||
|
||||
list(APPEND SLASRC DEPRECATED/sgegs.f DEPRECATED/sgegv.f
|
||||
DEPRECATED/sgeqpf.f DEPRECATED/sgelsx.f DEPRECATED/sggsvd.f
|
||||
DEPRECATED/sggsvp.f DEPRECATED/slahrd.f DEPRECATED/slatzm.f DEPRECATED/stzrqf.f)
|
||||
list(APPEND DLASRC DEPRECATED/dgegs.f DEPRECATED/dgegv.f
|
||||
DEPRECATED/dgeqpf.f DEPRECATED/dgelsx.f DEPRECATED/dggsvd.f
|
||||
DEPRECATED/dggsvp.f DEPRECATED/dlahrd.f DEPRECATED/dlatzm.f DEPRECATED/dtzrqf.f)
|
||||
list(APPEND CLASRC DEPRECATED/cgegs.f DEPRECATED/cgegv.f
|
||||
DEPRECATED/cgeqpf.f DEPRECATED/cgelsx.f DEPRECATED/cggsvd.f
|
||||
DEPRECATED/cggsvp.f DEPRECATED/clahrd.f DEPRECATED/clatzm.f DEPRECATED/ctzrqf.f)
|
||||
list(APPEND ZLASRC DEPRECATED/zgegs.f DEPRECATED/zgegv.f
|
||||
DEPRECATED/zgeqpf.f DEPRECATED/zgelsx.f DEPRECATED/zggsvd.f
|
||||
DEPRECATED/zggsvp.f DEPRECATED/zlahrd.f DEPRECATED/zlatzm.f DEPRECATED/ztzrqf.f)
|
||||
message(STATUS "Building deprecated routines")
|
||||
|
||||
set(DSLASRC spotrs.f)
|
||||
|
||||
set(ZCLASRC cpotrs.f)
|
||||
|
||||
set(DLASRC
|
||||
dgbbrd.f dgbcon.f dgbequ.f dgbrfs.f dgbsv.f
|
||||
dgbsvx.f dgbtf2.f dgbtrf.f dgbtrs.f dgebak.f dgebal.f dgebd2.f
|
||||
dgebrd.f dgecon.f dgeequ.f dgees.f dgeesx.f dgeev.f dgeevx.f
|
||||
DEPRECATED/dgegs.f DEPRECATED/dgegv.f dgehd2.f dgehrd.f dgelq2.f dgelqf.f
|
||||
dgels.f dgelsd.f dgelss.f DEPRECATED/dgelsx.f dgelsy.f dgeql2.f dgeqlf.f
|
||||
dgeqp3.f DEPRECATED/dgeqpf.f dgeqr2.f dgeqr2p.f dgeqrf.f dgeqrfp.f dgerfs.f
|
||||
dgerq2.f dgerqf.f dgesc2.f dgesdd.f dgesvd.f dgesvdx.f dgesvx.f
|
||||
dgetc2.f dgetri.f dgetrf2.f
|
||||
dggbak.f dggbal.f dgges.f dgges3.f dggesx.f dggev.f dggev3.f dggevx.f
|
||||
dggglm.f dgghd3.f dgghrd.f dgglse.f dggqrf.f
|
||||
dggrqf.f dggsvd3.f dggsvp3.f DEPRECATED/dggsvd.f DEPRECATED/dggsvp.f dgtcon.f dgtrfs.f dgtsv.f
|
||||
dgtsvx.f dgttrf.f dgttrs.f dgtts2.f dhgeqz.f
|
||||
dhsein.f dhseqr.f dlabrd.f dlacon.f dlacn2.f
|
||||
dlaein.f dlaexc.f dlag2.f dlags2.f dlagtm.f dlagv2.f dlahqr.f
|
||||
DEPRECATED/dlahrd.f dlahr2.f dlaic1.f dlaln2.f dlals0.f dlalsa.f dlalsd.f
|
||||
dlangb.f dlange.f dlangt.f dlanhs.f dlansb.f dlansp.f
|
||||
dlansy.f dlantb.f dlantp.f dlantr.f dlanv2.f
|
||||
dlapll.f dlapmt.f
|
||||
dlaqgb.f dlaqge.f dlaqp2.f dlaqps.f dlaqsb.f dlaqsp.f dlaqsy.f
|
||||
dlaqr0.f dlaqr1.f dlaqr2.f dlaqr3.f dlaqr4.f dlaqr5.f
|
||||
dlaqtr.f dlar1v.f dlar2v.f iladlr.f iladlc.f
|
||||
dlarf.f dlarfb.f dlarfg.f dlarfgp.f dlarft.f dlarfx.f
|
||||
dlargv.f dlarrv.f dlartv.f
|
||||
dlarz.f dlarzb.f dlarzt.f dlasy2.f dlasyf.f dlasyf_rook.f
|
||||
dlatbs.f dlatdf.f dlatps.f dlatrd.f dlatrs.f dlatrz.f DEPRECATED/dlatzm.f
|
||||
dopgtr.f dopmtr.f dorg2l.f dorg2r.f
|
||||
dorgbr.f dorghr.f dorgl2.f dorglq.f dorgql.f dorgqr.f dorgr2.f
|
||||
dorgrq.f dorgtr.f dorm2l.f dorm2r.f dorm22.f
|
||||
dormbr.f dormhr.f dorml2.f dormlq.f dormql.f dormqr.f dormr2.f
|
||||
dormr3.f dormrq.f dormrz.f dormtr.f dpbcon.f dpbequ.f dpbrfs.f
|
||||
dpbstf.f dpbsv.f dpbsvx.f
|
||||
dpbtf2.f dpbtrf.f dpbtrs.f dpocon.f dpoequ.f dporfs.f dposv.f
|
||||
dposvx.f dpotrf2.f dpotrs.f dpstrf.f dpstf2.f
|
||||
dppcon.f dppequ.f
|
||||
dpprfs.f dppsv.f dppsvx.f dpptrf.f dpptri.f dpptrs.f dptcon.f
|
||||
dpteqr.f dptrfs.f dptsv.f dptsvx.f dpttrs.f dptts2.f drscl.f
|
||||
dsbev.f dsbevd.f dsbevx.f dsbgst.f dsbgv.f dsbgvd.f dsbgvx.f
|
||||
dsbtrd.f dspcon.f dspev.f dspevd.f dspevx.f dspgst.f
|
||||
dspgv.f dspgvd.f dspgvx.f dsprfs.f dspsv.f dspsvx.f dsptrd.f
|
||||
dsptrf.f dsptri.f dsptrs.f dstegr.f dstein.f dstev.f dstevd.f dstevr.f
|
||||
dstevx.f
|
||||
dsycon.f dsyev.f dsyevd.f dsyevr.f
|
||||
dsyevx.f dsygs2.f dsygst.f dsygv.f dsygvd.f dsygvx.f dsyrfs.f
|
||||
dsysv.f dsysvx.f
|
||||
dsytd2.f dsytf2.f dsytrd.f dsytrf.f dsytri.f dsytri2.f dsytri2x.f
|
||||
dsyswapr.f dsytrs.f dsytrs2.f dsyconv.f
|
||||
dsytf2_rook.f dsytrf_rook.f dsytrs_rook.f
|
||||
dsytri_rook.f dsycon_rook.f dsysv_rook.f
|
||||
dtbcon.f dtbrfs.f dtbtrs.f dtgevc.f dtgex2.f dtgexc.f dtgsen.f
|
||||
dtgsja.f dtgsna.f dtgsy2.f dtgsyl.f dtpcon.f dtprfs.f dtptri.f
|
||||
dtptrs.f
|
||||
dtrcon.f dtrevc.f dtrevc3.f dtrexc.f dtrrfs.f dtrsen.f dtrsna.f dtrsyl.f
|
||||
dtrtrs.f DEPRECATED/dtzrqf.f dtzrzf.f dstemr.f
|
||||
dsgesv.f dsposv.f dlag2s.f slag2d.f dlat2s.f
|
||||
dlansf.f dpftrf.f dpftri.f dpftrs.f dsfrk.f dtfsm.f dtftri.f dtfttp.f
|
||||
dtfttr.f dtpttf.f dtpttr.f dtrttf.f dtrttp.f
|
||||
dgejsv.f dgesvj.f dgsvj0.f dgsvj1.f
|
||||
dgeequb.f dsyequb.f dpoequb.f dgbequb.f
|
||||
dbbcsd.f dlapmr.f dorbdb.f dorbdb1.f dorbdb2.f dorbdb3.f dorbdb4.f
|
||||
dorbdb5.f dorbdb6.f dorcsd.f dorcsd2by1.f
|
||||
dgeqrt.f dgeqrt2.f dgeqrt3.f dgemqrt.f
|
||||
dtpqrt.f dtpqrt2.f dtpmqrt.f dtprfb.f dpotri.f
|
||||
dgelq.f dgelqt.f dgelqt3.f dgemlq.f dgemlqt.f dgemqr.f dgeqr.f dgetsls.f
|
||||
dlamswlq.f dlamtsqr.f dlaswlq.f dlatsqr.f dtplqt.f dtplqt2.f dtpmlqt.f
|
||||
dsysv_aa.f dsytrf_aa.f dsytrs_aa.f dlasyf_aa.f dsytf2_rk.f dlasyf_rk.f
|
||||
dsytrf_rk.f dsytrs_3.f dsycon_3.f dsytri_3.f dsytri_3x.f dsysv_rk.f
|
||||
dsb2st_kernels.f dsbev_2stage.f dsbevd_2stage.f dsbevx_2stage.f
|
||||
dsyev_2stage.f dsyevd_2stage.f dsyevr_2stage.f dsyevx_2stage.f
|
||||
dsygv_2stage.f dsytrd_2stage.f dsytrd_sb2st.F dsytrd_sy2sb.f dlarfy.f
|
||||
)
|
||||
set(SCATGEN slatm1.f slaran.f slarnd.f)
|
||||
|
||||
set(ZLASRC
|
||||
zbdsqr.f zgbbrd.f zgbcon.f zgbequ.f zgbrfs.f zgbsv.f zgbsvx.f
|
||||
zgbtf2.f zgbtrf.f zgbtrs.f zgebak.f zgebal.f zgebd2.f zgebrd.f
|
||||
zgecon.f zgeequ.f zgees.f zgeesx.f zgeev.f zgeevx.f
|
||||
DEPRECATED/zgegs.f DEPRECATED/zgegv.f zgehd2.f zgehrd.f zgejsv.f zgelq2.f zgelqf.f
|
||||
zgels.f zgelsd.f zgelss.f DEPRECATED/zgelsx.f zgelsy.f zgeql2.f zgeqlf.f zgeqp3.f
|
||||
DEPRECATED/zgeqpf.f zgeqr2.f zgeqr2p.f zgeqrf.f zgeqrfp.f zgerfs.f zgerq2.f zgerqf.f
|
||||
zgesc2.f zgesdd.f zgesvd.f zgesvdx.f zgesvj.f zgesvx.f zgetc2.f
|
||||
zgetri.f zgetrf2.f
|
||||
zggbak.f zggbal.f zgges.f zgges3.f zggesx.f zggev.f zggev3.f zggevx.f zggglm.f
|
||||
zgghd3.f zgghrd.f zgglse.f zggqrf.f zggrqf.f
|
||||
DEPRECATED/zggsvd.f zggsvd3.f DEPRECATED/zggsvp.f zggsvp3.f
|
||||
zgsvj0.f zgsvj1.f
|
||||
zgtcon.f zgtrfs.f zgtsv.f zgtsvx.f zgttrf.f zgttrs.f zgtts2.f zhbev.f
|
||||
zhbevd.f zhbevx.f zhbgst.f zhbgv.f zhbgvd.f zhbgvx.f zhbtrd.f
|
||||
zhecon.f zheev.f zheevd.f zheevr.f zheevx.f zhegs2.f zhegst.f
|
||||
zhegv.f zhegvd.f zhegvx.f zherfs.f zhesv.f zhesvx.f zhetd2.f
|
||||
zhetf2.f zhetrd.f
|
||||
zhetrf.f zhetri.f zhetri2.f zhetri2x.f zheswapr.f
|
||||
zhetrs.f zhetrs2.f
|
||||
zhetf2_rook.f zhetrf_rook.f zhetri_rook.f zhetrs_rook.f zhecon_rook.f zhesv_rook.f
|
||||
zhgeqz.f zhpcon.f zhpev.f zhpevd.f
|
||||
zhpevx.f zhpgst.f zhpgv.f zhpgvd.f zhpgvx.f zhprfs.f zhpsv.f
|
||||
zhpsvx.f
|
||||
zhptrd.f zhptrf.f zhptri.f zhptrs.f zhsein.f zhseqr.f zlabrd.f
|
||||
zlacgv.f zlacon.f zlacn2.f zlacp2.f zlacpy.f zlacrm.f zlacrt.f zladiv.f
|
||||
zlaed0.f zlaed7.f zlaed8.f
|
||||
zlaein.f zlaesy.f zlaev2.f zlags2.f zlagtm.f
|
||||
zlahef.f zlahef_rook.f zlahqr.f
|
||||
DEPRECATED/zlahrd.f zlahr2.f zlaic1.f zlals0.f zlalsa.f zlalsd.f zlangb.f zlange.f
|
||||
zlangt.f zlanhb.f
|
||||
zlanhe.f
|
||||
zlanhp.f zlanhs.f zlanht.f zlansb.f zlansp.f zlansy.f zlantb.f
|
||||
zlantp.f zlantr.f zlapll.f zlapmt.f zlaqgb.f zlaqge.f
|
||||
zlaqhb.f zlaqhe.f zlaqhp.f zlaqp2.f zlaqps.f zlaqsb.f
|
||||
zlaqr0.f zlaqr1.f zlaqr2.f zlaqr3.f zlaqr4.f zlaqr5.f
|
||||
zlaqsp.f zlaqsy.f zlar1v.f zlar2v.f ilazlr.f ilazlc.f
|
||||
zlarcm.f zlarf.f zlarfb.f
|
||||
zlarfg.f zlarft.f zlarfgp.f
|
||||
zlarfx.f zlargv.f zlarnv.f zlarrv.f zlartg.f zlartv.f
|
||||
zlarz.f zlarzb.f zlarzt.f zlascl.f zlaset.f zlasr.f
|
||||
zlassq.f zlasyf.f zlasyf_rook.f zlasy_aa.f
|
||||
zlatbs.f zlatdf.f zlatps.f zlatrd.f zlatrs.f zlatrz.f DEPRECATED/zlatzm.f
|
||||
zpbcon.f zpbequ.f zpbrfs.f zpbstf.f zpbsv.f
|
||||
zpbsvx.f zpbtf2.f zpbtrf.f zpbtrs.f zpocon.f zpoequ.f zporfs.f
|
||||
zposv.f zposvx.f zpotrf2.f zpotrs.f zpstrf.f zpstf2.f
|
||||
zppcon.f zppequ.f zpprfs.f zppsv.f zppsvx.f zpptrf.f zpptri.f zpptrs.f
|
||||
zptcon.f zpteqr.f zptrfs.f zptsv.f zptsvx.f zpttrf.f zpttrs.f zptts2.f
|
||||
zrot.f zspcon.f zsprfs.f zspsv.f
|
||||
zspsvx.f zsptrf.f zsptri.f zsptrs.f zdrscl.f zstedc.f
|
||||
zstegr.f zstein.f zsteqr.f
|
||||
zsycon.f zsysv_aa.f
|
||||
zsyrfs.f zsysv.f zsysvx.f zsytf2.f zsytrf.f zsytri.f zsytri2.f zsytri2x.f
|
||||
zsyswapr.f zsytrs.f zsytrs_aa.f zsytrs2.f zsyconv.f
|
||||
zsytf2_rook.f zsytrf_rook.f zsytrs_rook.f
|
||||
zsytri_rook.f zsycon_rook.f zsysv_rook.f
|
||||
ztbcon.f ztbrfs.f ztbtrs.f ztgevc.f ztgex2.f
|
||||
ztgexc.f ztgsen.f ztgsja.f ztgsna.f ztgsy2.f ztgsyl.f ztpcon.f
|
||||
ztprfs.f ztptri.f
|
||||
ztptrs.f ztrcon.f ztrevc.f ztrevc3.f ztrexc.f ztrrfs.f ztrsen.f ztrsna.f
|
||||
ztrsyl.f ztrtrs.f DEPRECATED/ztzrqf.f ztzrzf.f zung2l.f
|
||||
zung2r.f zungbr.f zunghr.f zungl2.f zunglq.f zungql.f zungqr.f zungr2.f
|
||||
zungrq.f zungtr.f zunm2l.f zunm2r.f zunmbr.f zunmhr.f zunm22.f zunml2.f
|
||||
zunmlq.f zunmql.f zunmqr.f zunmr2.f zunmr3.f zunmrq.f zunmrz.f
|
||||
zunmtr.f zupgtr.f
|
||||
zupmtr.f izmax1.f dzsum1.f zstemr.f
|
||||
zcgesv.f zcposv.f zlag2c.f clag2z.f zlat2c.f
|
||||
zhfrk.f ztfttp.f zlanhf.f zpftrf.f zpftri.f zpftrs.f ztfsm.f ztftri.f
|
||||
ztfttr.f ztpttf.f ztpttr.f ztrttf.f ztrttp.f
|
||||
zgeequb.f zgbequb.f zsyequb.f zpoequb.f zheequb.f
|
||||
zbbcsd.f zlapmr.f zunbdb.f zunbdb1.f zunbdb2.f zunbdb3.f zunbdb4.f
|
||||
zunbdb5.f zunbdb6.f zuncsd.f zuncsd2by1.f
|
||||
zgeqrt.f zgeqrt2.f zgeqrt3.f zgemqrt.f
|
||||
ztpqrt.f ztpqrt2.f ztpmqrt.f ztprfb.f zpotri.f
|
||||
zgelq.f zgelqt.f zgelqt3.f zgemlq.f zgemlqt.f zgemqr.f zgeqr.f zgetsls.f
|
||||
zlamswlq.f zlamtsqr.f zlaswlq.f zlatsqr.f ztplqt.f ztplqt2.f ztpmlqt.f
|
||||
zhesv_aa.f zhetrf_aa.f zhetrs_aa.f zlahef_aa.f zsytf2_rk.f zlasyf_rk.f
|
||||
zsytrf_aa.f zsytrf_rk.f zsytrs_3.f zsycon_3.f zsytri_3.f zsytri_3x.f zsysv_rk.f
|
||||
zhetf2_rk.f zlahef_rk.f zhetrf_rk.f zhetrs_3.f zhecon_3.f zhetri_3.f
|
||||
zhetri_3x.f zhesv_rk.f zhb2st_kernels.f zhbev_2stage.f zhbevd_2stage.f
|
||||
zhbevx_2stage.f zheev_2stage.f zheevd_2stage.f zheevr_2stage.f
|
||||
zheevx_2stage.f zhegv_2stage.f zhetrd_2stage.f zhetrd_hb2st.F zhetrd_he2hb.f
|
||||
zlarfy.f
|
||||
)
|
||||
set(SMATGEN slatms.f slatme.f slatmr.f slatmt.f
|
||||
slagge.f slagsy.f slakf2.f slarge.f slaror.f slarot.f slatm2.f
|
||||
slatm3.f slatm5.f slatm6.f slatm7.f slahilb.f)
|
||||
|
||||
set(LA_REL_SRC ${ALLAUX})
|
||||
if (BUILD_SINGLE)
|
||||
list(APPEND LA_REL_SRC ${SLASRC} ${DSLASRC} ${SCLAUX})
|
||||
endif ()
|
||||
set(CMATGEN clatms.f clatme.f clatmr.f clatmt.f
|
||||
clagge.f claghe.f clagsy.f clakf2.f clarge.f claror.f clarot.f
|
||||
clatm1.f clarnd.f clatm2.f clatm3.f clatm5.f clatm6.f clahilb.f slatm7.f)
|
||||
|
||||
if (BUILD_DOUBLE)
|
||||
list(APPEND LA_REL_SRC ${DLASRC} ${DSLASRC} ${DZLAUX})
|
||||
endif ()
|
||||
set(DZATGEN dlatm1.f dlaran.f dlarnd.f)
|
||||
|
||||
if (BUILD_COMPLEX)
|
||||
list(APPEND LA_REL_SRC ${CLASRC} ${ZCLASRC} ${SCLAUX})
|
||||
endif ()
|
||||
set(DMATGEN dlatms.f dlatme.f dlatmr.f dlatmt.f
|
||||
dlagge.f dlagsy.f dlakf2.f dlarge.f dlaror.f dlarot.f dlatm2.f
|
||||
dlatm3.f dlatm5.f dlatm6.f dlatm7.f dlahilb.f)
|
||||
|
||||
if (BUILD_COMPLEX16)
|
||||
list(APPEND LA_REL_SRC ${ZLASRC} ${ZCLASRC} ${DZLAUX})
|
||||
endif ()
|
||||
set(ZMATGEN zlatms.f zlatme.f zlatmr.f zlatmt.f
|
||||
zlagge.f zlaghe.f zlagsy.f zlakf2.f zlarge.f zlaror.f zlarot.f
|
||||
zlatm1.f zlarnd.f zlatm2.f zlatm3.f zlatm5.f zlatm6.f zlahilb.f dlatm7.f)
|
||||
|
||||
if(BUILD_SINGLE)
|
||||
set(LA_REL_SRC ${SLASRC} ${DSLASRC} ${ALLAUX} ${SCLAUX})
|
||||
set(LA_GEN_SRC ${SMATGEN} ${SCATGEN})
|
||||
message(STATUS "Building Single Precision")
|
||||
endif()
|
||||
if(BUILD_DOUBLE)
|
||||
set(LA_REL_SRC ${LA_REL_SRC} ${DLASRC} ${DSLASRC} ${ALLAUX} ${DZLAUX})
|
||||
set(LA_GEN_SRC ${LA_GEN_SRC} ${DMATGEN} ${DZATGEN})
|
||||
message(STATUS "Building Double Precision")
|
||||
endif()
|
||||
if(BUILD_COMPLEX)
|
||||
set(LA_REL_SRC ${LA_REL_SRC} ${CLASRC} ${ZCLASRC} ${ALLAUX} ${SCLAUX})
|
||||
SET(LA_GEN_SRC ${LA_GEN_SRC} ${CMATGEN} ${SCATGEN})
|
||||
message(STATUS "Building Complex Precision")
|
||||
endif()
|
||||
if(BUILD_COMPLEX16)
|
||||
set(LA_REL_SRC ${LA_REL_SRC} ${ZLASRC} ${ZCLASRC} ${ALLAUX} ${DZLAUX})
|
||||
SET(LA_GEN_SRC ${LA_GEN_SRC} ${ZMATGEN} ${DZATGEN})
|
||||
message(STATUS "Building Double Complex Precision")
|
||||
endif()
|
||||
|
||||
# add lapack-netlib folder to the sources
|
||||
set(LA_SOURCES "")
|
||||
foreach (LA_FILE ${LA_REL_SRC})
|
||||
list(APPEND LA_SOURCES "${NETLIB_LAPACK_DIR}/SRC/${LA_FILE}")
|
||||
endforeach ()
|
||||
foreach (LA_FILE ${LA_GEN_SRC})
|
||||
list(APPEND LA_SOURCES "${NETLIB_LAPACK_DIR}/TESTING/MATGEN/${LA_FILE}")
|
||||
endforeach ()
|
||||
set_source_files_properties(${LA_SOURCES} PROPERTIES COMPILE_FLAGS "${LAPACK_FFLAGS}")
|
||||
|
||||
+193
-91
@@ -1,5 +1,5 @@
|
||||
|
||||
set(C_SRC
|
||||
set(CSRC
|
||||
lapacke_cbbcsd.c
|
||||
lapacke_cbbcsd_work.c
|
||||
lapacke_cbdsqr.c
|
||||
@@ -46,6 +46,8 @@ set(C_SRC
|
||||
lapacke_cgehrd_work.c
|
||||
lapacke_cgejsv.c
|
||||
lapacke_cgejsv_work.c
|
||||
lapacke_cgelq.c
|
||||
lapacke_cgelq_work.c
|
||||
lapacke_cgelq2.c
|
||||
lapacke_cgelq2_work.c
|
||||
lapacke_cgelqf.c
|
||||
@@ -60,12 +62,18 @@ set(C_SRC
|
||||
lapacke_cgelsy_work.c
|
||||
lapacke_cgemqr.c
|
||||
lapacke_cgemqr_work.c
|
||||
lapacke_cgemlq.c
|
||||
lapacke_cgemlq_work.c
|
||||
lapacke_cgemqrt.c
|
||||
lapacke_cgemqrt_work.c
|
||||
lapacke_cgeqlf.c
|
||||
lapacke_cgeqlf_work.c
|
||||
lapacke_cgeqp3.c
|
||||
lapacke_cgeqp3_work.c
|
||||
lapacke_cgeqpf.c
|
||||
lapacke_cgeqpf_work.c
|
||||
lapacke_cgeqr.c
|
||||
lapacke_cgeqr_work.c
|
||||
lapacke_cgeqr2.c
|
||||
lapacke_cgeqr2_work.c
|
||||
lapacke_cgeqrf.c
|
||||
@@ -134,8 +142,12 @@ set(C_SRC
|
||||
lapacke_cggqrf_work.c
|
||||
lapacke_cggrqf.c
|
||||
lapacke_cggrqf_work.c
|
||||
lapacke_cggsvd.c
|
||||
lapacke_cggsvd_work.c
|
||||
lapacke_cggsvd3.c
|
||||
lapacke_cggsvd3_work.c
|
||||
lapacke_cggsvp.c
|
||||
lapacke_cggsvp_work.c
|
||||
lapacke_cggsvp3.c
|
||||
lapacke_cggsvp3_work.c
|
||||
lapacke_cgtcon.c
|
||||
@@ -210,6 +222,8 @@ set(C_SRC
|
||||
lapacke_chesv_work.c
|
||||
lapacke_chesv_aa.c
|
||||
lapacke_chesv_aa_work.c
|
||||
lapacke_chesv_aa_2stage.c
|
||||
lapacke_chesv_aa_2stage_work.c
|
||||
lapacke_chesv_rk.c
|
||||
lapacke_chesv_rk_work.c
|
||||
lapacke_chesvx.c
|
||||
@@ -224,6 +238,8 @@ set(C_SRC
|
||||
lapacke_chetrf_rook_work.c
|
||||
lapacke_chetrf_aa.c
|
||||
lapacke_chetrf_aa_work.c
|
||||
lapacke_chetrf_aa_2stage.c
|
||||
lapacke_chetrf_aa_2stage_work.c
|
||||
lapacke_chetrf_rk.c
|
||||
lapacke_chetrf_rk_work.c
|
||||
lapacke_chetri.c
|
||||
@@ -242,6 +258,9 @@ set(C_SRC
|
||||
lapacke_chetrs_rook_work.c
|
||||
lapacke_chetrs_aa.c
|
||||
lapacke_chetrs_aa_work.c
|
||||
lapacke_chetrs_aa_2stage.c
|
||||
lapacke_chetrs_aa_2stage_work.c
|
||||
lapacke_chetrf_rk.c
|
||||
lapacke_chetrs_3.c
|
||||
lapacke_chetrs_3_work.c
|
||||
lapacke_chfrk.c
|
||||
@@ -290,6 +309,11 @@ set(C_SRC
|
||||
lapacke_clacp2_work.c
|
||||
lapacke_clacpy.c
|
||||
lapacke_clacpy_work.c
|
||||
lapacke_clacrm.c
|
||||
lapacke_clacrm_work.c
|
||||
lapacke_clarcm.c
|
||||
lapacke_clarcm_work.c
|
||||
lapacke_clacn2.c
|
||||
lapacke_clag2z.c
|
||||
lapacke_clag2z_work.c
|
||||
lapacke_clange.c
|
||||
@@ -318,6 +342,8 @@ set(C_SRC
|
||||
lapacke_clascl_work.c
|
||||
lapacke_claset.c
|
||||
lapacke_claset_work.c
|
||||
lapacke_classq.c
|
||||
lapacke_classq_work.c
|
||||
lapacke_claswp.c
|
||||
lapacke_claswp_work.c
|
||||
lapacke_clauum.c
|
||||
@@ -436,6 +462,8 @@ set(C_SRC
|
||||
lapacke_csysv_work.c
|
||||
lapacke_csysv_aa.c
|
||||
lapacke_csysv_aa_work.c
|
||||
lapacke_csysv_aa_2stage.c
|
||||
lapacke_csysv_aa_2stage_work.c
|
||||
lapacke_csysv_rk.c
|
||||
lapacke_csysv_rk_work.c
|
||||
lapacke_csysvx.c
|
||||
@@ -448,6 +476,8 @@ set(C_SRC
|
||||
lapacke_csytrf_rook_work.c
|
||||
lapacke_csytrf_aa.c
|
||||
lapacke_csytrf_aa_work.c
|
||||
lapacke_csytrf_aa_2stage.c
|
||||
lapacke_csytrf_aa_2stage_work.c
|
||||
lapacke_csytrf_rk.c
|
||||
lapacke_csytrf_rk_work.c
|
||||
lapacke_csytri.c
|
||||
@@ -466,6 +496,8 @@ set(C_SRC
|
||||
lapacke_csytrs_rook_work.c
|
||||
lapacke_csytrs_aa.c
|
||||
lapacke_csytrs_aa_work.c
|
||||
lapacke_csytrs_aa_2stage.c
|
||||
lapacke_csytrs_aa_2stage_work.c
|
||||
lapacke_csytrs_3.c
|
||||
lapacke_csytrs_3_work.c
|
||||
lapacke_ctbcon.c
|
||||
@@ -633,6 +665,8 @@ set(DSRC
|
||||
lapacke_dgehrd_work.c
|
||||
lapacke_dgejsv.c
|
||||
lapacke_dgejsv_work.c
|
||||
lapacke_dgelq.c
|
||||
lapacke_dgelq_work.c
|
||||
lapacke_dgelq2.c
|
||||
lapacke_dgelq2_work.c
|
||||
lapacke_dgelqf.c
|
||||
@@ -645,6 +679,8 @@ set(DSRC
|
||||
lapacke_dgelss_work.c
|
||||
lapacke_dgelsy.c
|
||||
lapacke_dgelsy_work.c
|
||||
lapacke_dgemlq.c
|
||||
lapacke_dgemlq_work.c
|
||||
lapacke_dgemqr.c
|
||||
lapacke_dgemqr_work.c
|
||||
lapacke_dgemqrt.c
|
||||
@@ -653,6 +689,10 @@ set(DSRC
|
||||
lapacke_dgeqlf_work.c
|
||||
lapacke_dgeqp3.c
|
||||
lapacke_dgeqp3_work.c
|
||||
lapacke_dgeqpf.c
|
||||
lapacke_dgeqpf_work.c
|
||||
lapacke_dgeqr.c
|
||||
lapacke_dgeqr_work.c
|
||||
lapacke_dgeqr2.c
|
||||
lapacke_dgeqr2_work.c
|
||||
lapacke_dgeqrf.c
|
||||
@@ -721,8 +761,12 @@ set(DSRC
|
||||
lapacke_dggqrf_work.c
|
||||
lapacke_dggrqf.c
|
||||
lapacke_dggrqf_work.c
|
||||
lapacke_dggsvd.c
|
||||
lapacke_dggsvd_work.c
|
||||
lapacke_dggsvd3.c
|
||||
lapacke_dggsvd3_work.c
|
||||
lapacke_dggsvp.c
|
||||
lapacke_dggsvp_work.c
|
||||
lapacke_dggsvp3.c
|
||||
lapacke_dggsvp3_work.c
|
||||
lapacke_dgtcon.c
|
||||
@@ -785,6 +829,9 @@ set(DSRC
|
||||
lapacke_dlaset_work.c
|
||||
lapacke_dlasrt.c
|
||||
lapacke_dlasrt_work.c
|
||||
lapacke_dlassq.c
|
||||
lapacke_dlassq_work.c
|
||||
lapacke_dlaswp.c
|
||||
lapacke_dlaswp.c
|
||||
lapacke_dlaswp_work.c
|
||||
lapacke_dlauum.c
|
||||
@@ -1025,6 +1072,8 @@ set(DSRC
|
||||
lapacke_dsysv_work.c
|
||||
lapacke_dsysv_aa.c
|
||||
lapacke_dsysv_aa_work.c
|
||||
lapacke_dsysv_aa_2stage.c
|
||||
lapacke_dsysv_aa_2stage_work.c
|
||||
lapacke_dsysv_rk.c
|
||||
lapacke_dsysv_rk_work.c
|
||||
lapacke_dsysvx.c
|
||||
@@ -1039,6 +1088,8 @@ set(DSRC
|
||||
lapacke_dsytrf_rook_work.c
|
||||
lapacke_dsytrf_aa.c
|
||||
lapacke_dsytrf_aa_work.c
|
||||
lapacke_dsytrf_aa_2stage.c
|
||||
lapacke_dsytrf_aa_2stage_work.c
|
||||
lapacke_dsytrf_rk.c
|
||||
lapacke_dsytrf_rk_work.c
|
||||
lapacke_dsytri.c
|
||||
@@ -1055,6 +1106,8 @@ set(DSRC
|
||||
lapacke_dsytrs2_work.c
|
||||
lapacke_dsytrs_aa.c
|
||||
lapacke_dsytrs_aa_work.c
|
||||
lapacke_dsytrs_aa_2stage.c
|
||||
lapacke_dsytrs_aa_2stage_work.c
|
||||
lapacke_dsytrs_3.c
|
||||
lapacke_dsytrs_3_work.c
|
||||
lapacke_dsytrs_work.c
|
||||
@@ -1184,6 +1237,8 @@ set(SSRC
|
||||
lapacke_sgehrd_work.c
|
||||
lapacke_sgejsv.c
|
||||
lapacke_sgejsv_work.c
|
||||
lapacke_sgelq.c
|
||||
lapacke_sgelq_work.c
|
||||
lapacke_sgelq2.c
|
||||
lapacke_sgelq2_work.c
|
||||
lapacke_sgelqf.c
|
||||
@@ -1196,6 +1251,8 @@ set(SSRC
|
||||
lapacke_sgelss_work.c
|
||||
lapacke_sgelsy.c
|
||||
lapacke_sgelsy_work.c
|
||||
lapacke_sgemlq.c
|
||||
lapacke_sgemlq_work.c
|
||||
lapacke_sgemqr.c
|
||||
lapacke_sgemqr_work.c
|
||||
lapacke_sgemqrt.c
|
||||
@@ -1204,6 +1261,10 @@ set(SSRC
|
||||
lapacke_sgeqlf_work.c
|
||||
lapacke_sgeqp3.c
|
||||
lapacke_sgeqp3_work.c
|
||||
lapacke_sgeqpf.c
|
||||
lapacke_sgeqpf_work.c
|
||||
lapacke_sgeqr.c
|
||||
lapacke_sgeqr_work.c
|
||||
lapacke_sgeqr2.c
|
||||
lapacke_sgeqr2_work.c
|
||||
lapacke_sgeqrf.c
|
||||
@@ -1272,8 +1333,12 @@ set(SSRC
|
||||
lapacke_sggqrf_work.c
|
||||
lapacke_sggrqf.c
|
||||
lapacke_sggrqf_work.c
|
||||
lapacke_sggsvd.c
|
||||
lapacke_sggsvd_work.c
|
||||
lapacke_sggsvd3.c
|
||||
lapacke_sggsvd3_work.c
|
||||
lapacke_sggsvp.c
|
||||
lapacke_sggsvp_work.c
|
||||
lapacke_sggsvp3.c
|
||||
lapacke_sggsvp3_work.c
|
||||
lapacke_sgtcon.c
|
||||
@@ -1336,6 +1401,8 @@ set(SSRC
|
||||
lapacke_slaset_work.c
|
||||
lapacke_slasrt.c
|
||||
lapacke_slasrt_work.c
|
||||
lapacke_slassq.c
|
||||
lapacke_slassq_work.c
|
||||
lapacke_slaswp.c
|
||||
lapacke_slaswp_work.c
|
||||
lapacke_slauum.c
|
||||
@@ -1572,6 +1639,8 @@ set(SSRC
|
||||
lapacke_ssysv_work.c
|
||||
lapacke_ssysv_aa.c
|
||||
lapacke_ssysv_aa_work.c
|
||||
lapacke_ssysv_aa_2stage.c
|
||||
lapacke_ssysv_aa_2stage_work.c
|
||||
lapacke_ssysv_rk.c
|
||||
lapacke_ssysv_rk_work.c
|
||||
lapacke_ssysvx.c
|
||||
@@ -1586,6 +1655,9 @@ set(SSRC
|
||||
lapacke_ssytrf_rook_work.c
|
||||
lapacke_ssytrf_aa.c
|
||||
lapacke_ssytrf_aa_work.c
|
||||
lapacke_ssytrf_aa_2stage.c
|
||||
lapacke_ssytrf_aa_2stage_work.c
|
||||
lapacke_ssytrf_rook.c
|
||||
lapacke_ssytrf_rk.c
|
||||
lapacke_ssytrf_rk_work.c
|
||||
lapacke_ssytri.c
|
||||
@@ -1602,6 +1674,8 @@ set(SSRC
|
||||
lapacke_ssytrs2_work.c
|
||||
lapacke_ssytrs_aa.c
|
||||
lapacke_ssytrs_aa_work.c
|
||||
lapacke_ssytrs_aa_2stage.c
|
||||
lapacke_ssytrs_aa_2stage_work.c
|
||||
lapacke_ssytrs_3.c
|
||||
lapacke_ssytrs_3_work.c
|
||||
lapacke_ssytrs_work.c
|
||||
@@ -1729,6 +1803,8 @@ set(ZSRC
|
||||
lapacke_zgehrd_work.c
|
||||
lapacke_zgejsv.c
|
||||
lapacke_zgejsv_work.c
|
||||
lapacke_zgelq.c
|
||||
lapacke_zgelq_work.c
|
||||
lapacke_zgelq2.c
|
||||
lapacke_zgelq2_work.c
|
||||
lapacke_zgelqf.c
|
||||
@@ -1741,6 +1817,8 @@ set(ZSRC
|
||||
lapacke_zgelss_work.c
|
||||
lapacke_zgelsy.c
|
||||
lapacke_zgelsy_work.c
|
||||
lapacke_zgemlq.c
|
||||
lapacke_zgemlq_work.c
|
||||
lapacke_zgemqr.c
|
||||
lapacke_zgemqr_work.c
|
||||
lapacke_zgemqrt.c
|
||||
@@ -1749,6 +1827,10 @@ set(ZSRC
|
||||
lapacke_zgeqlf_work.c
|
||||
lapacke_zgeqp3.c
|
||||
lapacke_zgeqp3_work.c
|
||||
lapacke_zgeqpf.c
|
||||
lapacke_zgeqpf_work.c
|
||||
lapacke_zgeqr.c
|
||||
lapacke_zgeqr_work.c
|
||||
lapacke_zgeqr2.c
|
||||
lapacke_zgeqr2_work.c
|
||||
lapacke_zgeqrf.c
|
||||
@@ -1817,8 +1899,12 @@ set(ZSRC
|
||||
lapacke_zggqrf_work.c
|
||||
lapacke_zggrqf.c
|
||||
lapacke_zggrqf_work.c
|
||||
lapacke_zggsvd.c
|
||||
lapacke_zggsvd_work.c
|
||||
lapacke_zggsvd3.c
|
||||
lapacke_zggsvd3_work.c
|
||||
lapacke_zggsvp.c
|
||||
lapacke_zggsvp_work.c
|
||||
lapacke_zggsvp3.c
|
||||
lapacke_zggsvp3_work.c
|
||||
lapacke_zgtcon.c
|
||||
@@ -1839,6 +1925,12 @@ set(ZSRC
|
||||
lapacke_zhbevd_work.c
|
||||
lapacke_zhbevx.c
|
||||
lapacke_zhbevx_work.c
|
||||
lapacke_zhbev_2stage.c
|
||||
lapacke_zhbev_2stage_work.c
|
||||
lapacke_zhbevd_2stage.c
|
||||
lapacke_zhbevd_2stage_work.c
|
||||
lapacke_zhbevx_2stage.c
|
||||
lapacke_zhbevx_2stage_work.c
|
||||
lapacke_zhbgst.c
|
||||
lapacke_zhbgst_work.c
|
||||
lapacke_zhbgv.c
|
||||
@@ -1887,6 +1979,8 @@ set(ZSRC
|
||||
lapacke_zhesv_work.c
|
||||
lapacke_zhesv_aa.c
|
||||
lapacke_zhesv_aa_work.c
|
||||
lapacke_zhesv_aa_2stage.c
|
||||
lapacke_zhesv_aa_2stage_work.c
|
||||
lapacke_zhesv_rk.c
|
||||
lapacke_zhesv_rk_work.c
|
||||
lapacke_zhesvx.c
|
||||
@@ -1901,6 +1995,8 @@ set(ZSRC
|
||||
lapacke_zhetrf_rook_work.c
|
||||
lapacke_zhetrf_aa.c
|
||||
lapacke_zhetrf_aa_work.c
|
||||
lapacke_zhetrf_aa_2stage.c
|
||||
lapacke_zhetrf_aa_2stage_work.c
|
||||
lapacke_zhetrf_rk.c
|
||||
lapacke_zhetrf_rk_work.c
|
||||
lapacke_zhetri.c
|
||||
@@ -1918,6 +2014,8 @@ set(ZSRC
|
||||
lapacke_zhetrs_work.c
|
||||
lapacke_zhetrs_aa.c
|
||||
lapacke_zhetrs_aa_work.c
|
||||
lapacke_zhetrs_aa_2stage.c
|
||||
lapacke_zhetrs_aa_2stage_work.c
|
||||
lapacke_zhetrs_3.c
|
||||
lapacke_zhetrs_3_work.c
|
||||
lapacke_zhetrs_rook_work.c
|
||||
@@ -1967,6 +2065,8 @@ set(ZSRC
|
||||
lapacke_zlacp2_work.c
|
||||
lapacke_zlacpy.c
|
||||
lapacke_zlacpy_work.c
|
||||
lapacke_zlacrm.c
|
||||
lapacke_zlacrm_work.c
|
||||
lapacke_zlag2c.c
|
||||
lapacke_zlag2c_work.c
|
||||
lapacke_zlange.c
|
||||
@@ -1981,6 +2081,8 @@ set(ZSRC
|
||||
lapacke_zlapmr_work.c
|
||||
lapacke_zlapmt.c
|
||||
lapacke_zlapmt_work.c
|
||||
lapacke_zlarcm.c
|
||||
lapacke_zlarcm_work.c
|
||||
lapacke_zlarfb.c
|
||||
lapacke_zlarfb_work.c
|
||||
lapacke_zlarfg.c
|
||||
@@ -1995,6 +2097,8 @@ set(ZSRC
|
||||
lapacke_zlascl_work.c
|
||||
lapacke_zlaset.c
|
||||
lapacke_zlaset_work.c
|
||||
lapacke_zlassq.c
|
||||
lapacke_zlassq_work.c
|
||||
lapacke_zlaswp.c
|
||||
lapacke_zlaswp_work.c
|
||||
lapacke_zlauum.c
|
||||
@@ -2113,6 +2217,8 @@ set(ZSRC
|
||||
lapacke_zsysv_work.c
|
||||
lapacke_zsysv_aa.c
|
||||
lapacke_zsysv_aa_work.c
|
||||
lapacke_zsysv_aa_2stage.c
|
||||
lapacke_zsysv_aa_2stage_work.c
|
||||
lapacke_zsysv_rk.c
|
||||
lapacke_zsysv_rk_work.c
|
||||
lapacke_zsysvx.c
|
||||
@@ -2125,6 +2231,8 @@ set(ZSRC
|
||||
lapacke_zsytrf_rook_work.c
|
||||
lapacke_zsytrf_aa.c
|
||||
lapacke_zsytrf_aa_work.c
|
||||
lapacke_zsytrf_aa_2stage.c
|
||||
lapacke_zsytrf_aa_2stage_work.c
|
||||
lapacke_zsytrf_rk.c
|
||||
lapacke_zsytrf_rk_work.c
|
||||
lapacke_zsytri.c
|
||||
@@ -2143,6 +2251,8 @@ set(ZSRC
|
||||
lapacke_zsytrs_rook_work.c
|
||||
lapacke_zsytrs_aa.c
|
||||
lapacke_zsytrs_aa_work.c
|
||||
lapacke_zsytrs_aa_2stage.c
|
||||
lapacke_zsytrs_aa_2stage_work.c
|
||||
lapacke_zsytrs_3.c
|
||||
lapacke_zsytrs_3_work.c
|
||||
lapacke_ztbcon.c
|
||||
@@ -2263,104 +2373,92 @@ set(ZSRC
|
||||
)
|
||||
|
||||
set(SRCX
|
||||
lapacke_cgbrfsx.c lapacke_cporfsx.c lapacke_dgerfsx.c lapacke_sgbrfsx.c lapacke_ssyrfsx.c lapacke_zherfsx.c
|
||||
lapacke_cgbrfsx_work.c lapacke_cporfsx_work.c lapacke_dgerfsx_work.c lapacke_sgbrfsx_work.c lapacke_ssyrfsx_work.c lapacke_zherfsx_work.c
|
||||
lapacke_cgerfsx.c lapacke_csyrfsx.c lapacke_dporfsx.c lapacke_sgerfsx.c lapacke_zgbrfsx.c lapacke_zporfsx.c
|
||||
lapacke_cgerfsx_work.c lapacke_csyrfsx_work.c lapacke_dporfsx_work.c lapacke_sgerfsx_work.c lapacke_zgbrfsx_work.c lapacke_zporfsx_work.c
|
||||
lapacke_cherfsx.c lapacke_dgbrfsx.c lapacke_dsyrfsx.c lapacke_sporfsx.c lapacke_zgerfsx.c lapacke_zsyrfsx.c
|
||||
lapacke_cherfsx_work.c lapacke_dgbrfsx_work.c lapacke_dsyrfsx_work.c lapacke_sporfsx_work.c lapacke_zgerfsx_work.c lapacke_zsyrfsx_work.c
|
||||
lapacke_cgbsvxx.c lapacke_cposvxx.c lapacke_dgesvxx.c lapacke_sgbsvxx.c lapacke_ssysvxx.c lapacke_zhesvxx.c
|
||||
lapacke_cgbsvxx_work.c lapacke_cposvxx_work.c lapacke_dgesvxx_work.c lapacke_sgbsvxx_work.c lapacke_ssysvxx_work.c lapacke_zhesvxx_work.c
|
||||
lapacke_cgesvxx.c lapacke_csysvxx.c lapacke_dposvxx.c lapacke_sgesvxx.c lapacke_zgbsvxx.c lapacke_zposvxx.c
|
||||
lapacke_cgesvxx_work.c lapacke_csysvxx_work.c lapacke_dposvxx_work.c lapacke_sgesvxx_work.c lapacke_zgbsvxx_work.c lapacke_zposvxx_work.c
|
||||
lapacke_chesvxx.c lapacke_dgbsvxx.c lapacke_dsysvxx.c lapacke_sposvxx.c lapacke_zgesvxx.c lapacke_zsysvxx.c
|
||||
lapacke_cgbrfsx.c lapacke_cporfsx.c lapacke_dgerfsx.c lapacke_sgbrfsx.c lapacke_ssyrfsx.c lapacke_zherfsx.c
|
||||
lapacke_cgbrfsx_work.c lapacke_cporfsx_work.c lapacke_dgerfsx_work.c lapacke_sgbrfsx_work.c lapacke_ssyrfsx_work.c lapacke_zherfsx_work.c
|
||||
lapacke_cgerfsx.c lapacke_csyrfsx.c lapacke_dporfsx.c lapacke_sgerfsx.c lapacke_zgbrfsx.c lapacke_zporfsx.c
|
||||
lapacke_cgerfsx_work.c lapacke_csyrfsx_work.c lapacke_dporfsx_work.c lapacke_sgerfsx_work.c lapacke_zgbrfsx_work.c lapacke_zporfsx_work.c
|
||||
lapacke_cherfsx.c lapacke_dgbrfsx.c lapacke_dsyrfsx.c lapacke_sporfsx.c lapacke_zgerfsx.c lapacke_zsyrfsx.c
|
||||
lapacke_cherfsx_work.c lapacke_dgbrfsx_work.c lapacke_dsyrfsx_work.c lapacke_sporfsx_work.c lapacke_zgerfsx_work.c lapacke_zsyrfsx_work.c
|
||||
lapacke_cgbsvxx.c lapacke_cposvxx.c lapacke_dgesvxx.c lapacke_sgbsvxx.c lapacke_ssysvxx.c lapacke_zhesvxx.c
|
||||
lapacke_cgbsvxx_work.c lapacke_cposvxx_work.c lapacke_dgesvxx_work.c lapacke_sgbsvxx_work.c lapacke_ssysvxx_work.c lapacke_zhesvxx_work.c
|
||||
lapacke_cgesvxx.c lapacke_csysvxx.c lapacke_dposvxx.c lapacke_sgesvxx.c lapacke_zgbsvxx.c lapacke_zposvxx.c
|
||||
lapacke_cgesvxx_work.c lapacke_csysvxx_work.c lapacke_dposvxx_work.c lapacke_sgesvxx_work.c lapacke_zgbsvxx_work.c lapacke_zposvxx_work.c
|
||||
lapacke_chesvxx.c lapacke_dgbsvxx.c lapacke_dsysvxx.c lapacke_sposvxx.c lapacke_zgesvxx.c lapacke_zsysvxx.c
|
||||
lapacke_chesvxx_work.c lapacke_dgbsvxx_work.c lapacke_dsysvxx_work.c lapacke_sposvxx_work.c lapacke_zgesvxx_work.c lapacke_zsysvxx_work.c
|
||||
)
|
||||
|
||||
|
||||
# FILE PARTS OF TMGLIB
|
||||
# FILE PARTS OF TMGLIB
|
||||
set(MATGEN
|
||||
lapacke_clatms.c
|
||||
lapacke_clatms_work.c
|
||||
lapacke_dlatms.c
|
||||
lapacke_dlatms_work.c
|
||||
lapacke_slatms.c
|
||||
lapacke_slatms_work.c
|
||||
lapacke_zlatms.c
|
||||
lapacke_zlatms_work.c
|
||||
lapacke_clagge.c
|
||||
lapacke_clagge_work.c
|
||||
lapacke_dlagge.c
|
||||
lapacke_dlagge_work.c
|
||||
lapacke_slagge.c
|
||||
lapacke_slagge_work.c
|
||||
lapacke_zlagge.c
|
||||
lapacke_zlagge_work.c
|
||||
lapacke_claghe.c
|
||||
lapacke_claghe_work.c
|
||||
lapacke_zlaghe.c
|
||||
lapacke_zlaghe_work.c
|
||||
lapacke_clagsy.c
|
||||
lapacke_clagsy_work.c
|
||||
lapacke_dlagsy.c
|
||||
lapacke_dlagsy_work.c
|
||||
lapacke_slagsy.c
|
||||
lapacke_slagsy_work.c
|
||||
lapacke_zlagsy.c
|
||||
lapacke_clatms.c
|
||||
lapacke_clatms_work.c
|
||||
lapacke_dlatms.c
|
||||
lapacke_dlatms_work.c
|
||||
lapacke_slatms.c
|
||||
lapacke_slatms_work.c
|
||||
lapacke_zlatms.c
|
||||
lapacke_zlatms_work.c
|
||||
lapacke_clagge.c
|
||||
lapacke_clagge_work.c
|
||||
lapacke_dlagge.c
|
||||
lapacke_dlagge_work.c
|
||||
lapacke_slagge.c
|
||||
lapacke_slagge_work.c
|
||||
lapacke_zlagge.c
|
||||
lapacke_zlagge_work.c
|
||||
lapacke_claghe.c
|
||||
lapacke_claghe_work.c
|
||||
lapacke_zlaghe.c
|
||||
lapacke_zlaghe_work.c
|
||||
lapacke_clagsy.c
|
||||
lapacke_clagsy_work.c
|
||||
lapacke_dlagsy.c
|
||||
lapacke_dlagsy_work.c
|
||||
lapacke_slagsy.c
|
||||
lapacke_slagsy_work.c
|
||||
lapacke_zlagsy.c
|
||||
lapacke_zlagsy_work.c
|
||||
lapacke_nancheck.c
|
||||
)
|
||||
|
||||
set(Utils_SRC
|
||||
lapacke_cgb_nancheck.c lapacke_dpf_nancheck.c lapacke_ssy_trans.c
|
||||
lapacke_cgb_trans.c lapacke_dpf_trans.c lapacke_stb_nancheck.c
|
||||
lapacke_cge_nancheck.c lapacke_dpo_nancheck.c lapacke_stb_trans.c
|
||||
lapacke_cge_trans.c lapacke_dpo_trans.c lapacke_stf_nancheck.c
|
||||
lapacke_cgg_nancheck.c lapacke_dpp_nancheck.c lapacke_stf_trans.c
|
||||
lapacke_cgg_trans.c lapacke_dpp_trans.c lapacke_stp_nancheck.c
|
||||
lapacke_cgt_nancheck.c lapacke_dpt_nancheck.c lapacke_stp_trans.c
|
||||
lapacke_chb_nancheck.c lapacke_dsb_nancheck.c lapacke_str_nancheck.c
|
||||
lapacke_chb_trans.c lapacke_dsb_trans.c lapacke_str_trans.c
|
||||
lapacke_che_nancheck.c lapacke_dsp_nancheck.c lapacke_xerbla.c
|
||||
lapacke_che_trans.c lapacke_dsp_trans.c lapacke_zgb_nancheck.c
|
||||
lapacke_chp_nancheck.c lapacke_dst_nancheck.c lapacke_zgb_trans.c
|
||||
lapacke_chp_trans.c lapacke_dsy_nancheck.c lapacke_zge_nancheck.c
|
||||
lapacke_chs_nancheck.c lapacke_dsy_trans.c lapacke_zge_trans.c
|
||||
lapacke_chs_trans.c lapacke_dtb_nancheck.c lapacke_zgg_nancheck.c
|
||||
lapacke_c_nancheck.c lapacke_dtb_trans.c lapacke_zgg_trans.c
|
||||
lapacke_cpb_nancheck.c lapacke_dtf_nancheck.c lapacke_zgt_nancheck.c
|
||||
lapacke_cpb_trans.c lapacke_dtf_trans.c lapacke_zhb_nancheck.c
|
||||
lapacke_cpf_nancheck.c lapacke_dtp_nancheck.c lapacke_zhb_trans.c
|
||||
lapacke_cpf_trans.c lapacke_dtp_trans.c lapacke_zhe_nancheck.c
|
||||
lapacke_cpo_nancheck.c lapacke_dtr_nancheck.c lapacke_zhe_trans.c
|
||||
lapacke_cpo_trans.c lapacke_dtr_trans.c lapacke_zhp_nancheck.c
|
||||
lapacke_cpp_nancheck.c lapacke_lsame.c lapacke_zhp_trans.c
|
||||
lapacke_cpp_trans.c lapacke_make_complex_double.c lapacke_zhs_nancheck.c
|
||||
lapacke_cpt_nancheck.c lapacke_make_complex_float.c lapacke_zhs_trans.c
|
||||
lapacke_csp_nancheck.c lapacke_sgb_nancheck.c lapacke_z_nancheck.c
|
||||
lapacke_csp_trans.c lapacke_sgb_trans.c lapacke_zpb_nancheck.c
|
||||
lapacke_cst_nancheck.c lapacke_sge_nancheck.c lapacke_zpb_trans.c
|
||||
lapacke_csy_nancheck.c lapacke_sge_trans.c lapacke_zpf_nancheck.c
|
||||
lapacke_csy_trans.c lapacke_sgg_nancheck.c lapacke_zpf_trans.c
|
||||
lapacke_ctb_nancheck.c lapacke_sgg_trans.c lapacke_zpo_nancheck.c
|
||||
lapacke_ctb_trans.c lapacke_sgt_nancheck.c lapacke_zpo_trans.c
|
||||
lapacke_ctf_nancheck.c lapacke_shs_nancheck.c lapacke_zpp_nancheck.c
|
||||
lapacke_ctf_trans.c lapacke_shs_trans.c lapacke_zpp_trans.c
|
||||
lapacke_ctp_nancheck.c lapacke_s_nancheck.c lapacke_zpt_nancheck.c
|
||||
lapacke_ctp_trans.c lapacke_spb_nancheck.c lapacke_zsp_nancheck.c
|
||||
lapacke_ctr_nancheck.c lapacke_spb_trans.c lapacke_zsp_trans.c
|
||||
lapacke_ctr_trans.c lapacke_spf_nancheck.c lapacke_zst_nancheck.c
|
||||
lapacke_dgb_nancheck.c lapacke_spf_trans.c lapacke_zsy_nancheck.c
|
||||
lapacke_dgb_trans.c lapacke_spo_nancheck.c lapacke_zsy_trans.c
|
||||
lapacke_dge_nancheck.c lapacke_spo_trans.c lapacke_ztb_nancheck.c
|
||||
lapacke_dge_trans.c lapacke_spp_nancheck.c lapacke_ztb_trans.c
|
||||
lapacke_dgg_nancheck.c lapacke_spp_trans.c lapacke_ztf_nancheck.c
|
||||
lapacke_dgg_trans.c lapacke_spt_nancheck.c lapacke_ztf_trans.c
|
||||
lapacke_dgt_nancheck.c lapacke_ssb_nancheck.c lapacke_ztp_nancheck.c
|
||||
lapacke_dhs_nancheck.c lapacke_ssb_trans.c lapacke_ztp_trans.c
|
||||
lapacke_dhs_trans.c lapacke_ssp_nancheck.c lapacke_ztr_nancheck.c
|
||||
lapacke_d_nancheck.c lapacke_ssp_trans.c lapacke_ztr_trans.c
|
||||
lapacke_dpb_nancheck.c lapacke_sst_nancheck.c
|
||||
lapacke_dpb_trans.c lapacke_ssy_nancheck.c
|
||||
lapacke_c_nancheck.c lapacke_ctr_trans.c lapacke_make_complex_float.c lapacke_zgb_nancheck.c
|
||||
lapacke_cgb_nancheck.c lapacke_d_nancheck.c lapacke_s_nancheck.c lapacke_zgb_trans.c
|
||||
lapacke_cgb_trans.c lapacke_dgb_nancheck.c lapacke_sgb_nancheck.c lapacke_zge_nancheck.c
|
||||
lapacke_cge_nancheck.c lapacke_dgb_trans.c lapacke_sgb_trans.c lapacke_zge_trans.c
|
||||
lapacke_cge_trans.c lapacke_dge_nancheck.c lapacke_sge_nancheck.c lapacke_zgg_nancheck.c
|
||||
lapacke_cgg_nancheck.c lapacke_dge_trans.c lapacke_sge_trans.c lapacke_zgg_trans.c
|
||||
lapacke_cgg_trans.c lapacke_dgg_nancheck.c lapacke_sgg_nancheck.c lapacke_zgt_nancheck.c
|
||||
lapacke_cgt_nancheck.c lapacke_dgg_trans.c lapacke_sgg_trans.c lapacke_zhb_nancheck.c
|
||||
lapacke_chb_nancheck.c lapacke_dgt_nancheck.c lapacke_sgt_nancheck.c lapacke_zhb_trans.c
|
||||
lapacke_chb_trans.c lapacke_dhs_nancheck.c lapacke_shs_nancheck.c lapacke_zhe_nancheck.c
|
||||
lapacke_che_nancheck.c lapacke_dhs_trans.c lapacke_shs_trans.c lapacke_zhe_trans.c
|
||||
lapacke_che_trans.c lapacke_dpb_nancheck.c lapacke_spb_nancheck.c lapacke_zhp_nancheck.c
|
||||
lapacke_chp_nancheck.c lapacke_dpb_trans.c lapacke_spb_trans.c lapacke_zhp_trans.c
|
||||
lapacke_chp_trans.c lapacke_dpf_nancheck.c lapacke_spf_nancheck.c lapacke_zhs_nancheck.c
|
||||
lapacke_chs_nancheck.c lapacke_dpf_trans.c lapacke_spf_trans.c lapacke_zhs_trans.c
|
||||
lapacke_chs_trans.c lapacke_dpo_nancheck.c lapacke_spo_nancheck.c lapacke_zpb_nancheck.c
|
||||
lapacke_cpb_nancheck.c lapacke_dpo_trans.c lapacke_spo_trans.c lapacke_zpb_trans.c
|
||||
lapacke_cpb_trans.c lapacke_dpp_nancheck.c lapacke_spp_nancheck.c lapacke_zpf_nancheck.c
|
||||
lapacke_cpf_nancheck.c lapacke_dpp_trans.c lapacke_spp_trans.c lapacke_zpf_trans.c
|
||||
lapacke_cpf_trans.c lapacke_dpt_nancheck.c lapacke_spt_nancheck.c lapacke_zpo_nancheck.c
|
||||
lapacke_cpo_nancheck.c lapacke_dsb_nancheck.c lapacke_ssb_nancheck.c lapacke_zpo_trans.c
|
||||
lapacke_cpo_trans.c lapacke_dsb_trans.c lapacke_ssb_trans.c lapacke_zpp_nancheck.c
|
||||
lapacke_cpp_nancheck.c lapacke_dsp_nancheck.c lapacke_ssp_nancheck.c lapacke_zpp_trans.c
|
||||
lapacke_cpp_trans.c lapacke_dsp_trans.c lapacke_ssp_trans.c lapacke_zpt_nancheck.c
|
||||
lapacke_cpt_nancheck.c lapacke_dst_nancheck.c lapacke_sst_nancheck.c lapacke_zsp_nancheck.c
|
||||
lapacke_csp_nancheck.c lapacke_dsy_nancheck.c lapacke_ssy_nancheck.c lapacke_zsp_trans.c
|
||||
lapacke_csp_trans.c lapacke_dsy_trans.c lapacke_ssy_trans.c lapacke_zst_nancheck.c
|
||||
lapacke_cst_nancheck.c lapacke_dtb_nancheck.c lapacke_stb_nancheck.c lapacke_zsy_nancheck.c
|
||||
lapacke_csy_nancheck.c lapacke_dtb_trans.c lapacke_stb_trans.c lapacke_zsy_trans.c
|
||||
lapacke_csy_trans.c lapacke_dtf_nancheck.c lapacke_stf_nancheck.c lapacke_ztb_nancheck.c
|
||||
lapacke_ctb_nancheck.c lapacke_dtf_trans.c lapacke_stf_trans.c lapacke_ztb_trans.c
|
||||
lapacke_ctb_trans.c lapacke_dtp_nancheck.c lapacke_stp_nancheck.c lapacke_ztf_nancheck.c
|
||||
lapacke_ctf_nancheck.c lapacke_dtp_trans.c lapacke_stp_trans.c lapacke_ztf_trans.c
|
||||
lapacke_ctf_trans.c lapacke_dtr_nancheck.c lapacke_str_nancheck.c lapacke_ztp_nancheck.c
|
||||
lapacke_ctp_nancheck.c lapacke_dtr_trans.c lapacke_str_trans.c lapacke_ztp_trans.c
|
||||
lapacke_ctp_trans.c lapacke_lsame.c lapacke_xerbla.c lapacke_ztr_nancheck.c
|
||||
lapacke_ctr_nancheck.c lapacke_make_complex_double.c lapacke_z_nancheck.c lapacke_ztr_trans.c
|
||||
)
|
||||
|
||||
set(LAPACKE_REL_SRC "")
|
||||
@@ -2380,6 +2478,10 @@ if (BUILD_COMPLEX16)
|
||||
list(APPEND LAPACKE_REL_SRC ${ZSRC})
|
||||
endif ()
|
||||
|
||||
if (BUILD_MATGEN)
|
||||
list(APPEND LAPACKE_REL_SRC ${MATGEN})
|
||||
endif ()
|
||||
|
||||
# add lapack-netlib folder to the sources
|
||||
set(LAPACKE_SOURCES "")
|
||||
foreach (LAE_FILE ${LAPACKE_REL_SRC})
|
||||
@@ -2391,6 +2493,6 @@ foreach (Utils_FILE ${Utils_SRC})
|
||||
endforeach ()
|
||||
|
||||
set(lapacke_include_dir "${NETLIB_LAPACK_DIR}/LAPACKE/include")
|
||||
execute_process(COMMAND ${CMAKE_COMMAND} -E copy "${lapacke_include_dir}/lapacke_mangling_with_flags.h.in" "${lapacke_include_dir}/lapacke_mangling.h")
|
||||
configure_file("${lapacke_include_dir}/lapacke_mangling_with_flags.h.in" "${lapacke_include_dir}/lapacke_mangling.h" COPYONLY)
|
||||
include_directories(${lapacke_include_dir})
|
||||
set_source_files_properties(${LAPACKE_SOURCES} PROPERTIES COMPILE_FLAGS "${LAPACK_CFLAGS}")
|
||||
|
||||
@@ -1,6 +1,5 @@
|
||||
prefix=@prefix@
|
||||
libdir=@libdir@
|
||||
includedir=@includedir@
|
||||
libdir=@CMAKE_INSTALL_FULL_LIBDIR@
|
||||
includedir=@CMAKE_INSTALL_FULL_INCLUDEDIR@
|
||||
|
||||
Name: OpenBLAS
|
||||
Description: OpenBLAS is an optimized BLAS library based on GotoBLAS2 1.13 BSD version
|
||||
|
||||
+4
-17
@@ -3,19 +3,6 @@
|
||||
## Description: Ported from portion of OpenBLAS/Makefile.system
|
||||
## Detects the OS and sets appropriate variables.
|
||||
|
||||
if (${CMAKE_SYSTEM_NAME} STREQUAL "Darwin")
|
||||
set(ENV{MACOSX_DEPLOYMENT_TARGET} "10.2") # TODO: should be exported as an env var
|
||||
set(MD5SUM "md5 -r")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_SYSTEM_NAME} STREQUAL "FreeBSD")
|
||||
set(MD5SUM "md5 -r")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_SYSTEM_NAME} STREQUAL "NetBSD")
|
||||
set(MD5SUM "md5 -n")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_SYSTEM_NAME} STREQUAL "Linux")
|
||||
set(EXTRALIB "${EXTRALIB} -lm")
|
||||
set(NO_EXPRECISION 1)
|
||||
@@ -56,7 +43,7 @@ if (${CMAKE_SYSTEM_NAME} STREQUAL "Windows")
|
||||
|
||||
# Ensure the correct stack alignment on Win32
|
||||
# http://permalink.gmane.org/gmane.comp.lib.openblas.general/97
|
||||
if (${ARCH} STREQUAL "x86")
|
||||
if (X86)
|
||||
if (NOT MSVC AND NOT ${CMAKE_C_COMPILER_ID} STREQUAL "Clang")
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -mincoming-stack-boundary=2")
|
||||
endif ()
|
||||
@@ -77,8 +64,8 @@ if (CYGWIN)
|
||||
set(NO_EXPRECISION 1)
|
||||
endif ()
|
||||
|
||||
if (NOT ${CMAKE_SYSTEM_NAME} STREQUAL "Windows" AND NOT ${CMAKE_SYSTEM_NAME} STREQUAL "Interix")
|
||||
if (SMP)
|
||||
if (NOT ${CMAKE_SYSTEM_NAME} STREQUAL "Windows" AND NOT ${CMAKE_SYSTEM_NAME} STREQUAL "Interix" AND NOT ${CMAKE_SYSTEM_NAME} STREQUAL "Android")
|
||||
if (USE_THREAD)
|
||||
set(EXTRALIB "${EXTRALIB} -lpthread")
|
||||
endif ()
|
||||
endif ()
|
||||
@@ -88,7 +75,7 @@ if (QUAD_PRECISION)
|
||||
set(NO_EXPRECISION 1)
|
||||
endif ()
|
||||
|
||||
if (${ARCH} STREQUAL "x86")
|
||||
if (X86)
|
||||
set(NO_EXPRECISION 1)
|
||||
endif ()
|
||||
|
||||
|
||||
+188
-57
@@ -4,7 +4,8 @@
|
||||
## This is triggered by system.cmake and runs before any of the code is built.
|
||||
## Creates config.h and Makefile.conf by first running the c_check perl script (which creates those files).
|
||||
## Next it runs f_check and appends some fortran information to the files.
|
||||
## Finally it runs getarch and getarch_2nd for even more environment information.
|
||||
## Then it runs getarch and getarch_2nd for even more environment information.
|
||||
## Finally it builds gen_config_h for use at build time to generate config.h.
|
||||
|
||||
# CMake vars set by this file:
|
||||
# CORE
|
||||
@@ -36,78 +37,208 @@
|
||||
|
||||
# CPUIDEMU = ../../cpuid/table.o
|
||||
|
||||
|
||||
if (DEFINED CPUIDEMU)
|
||||
set(EXFLAGS "-DCPUIDEMU -DVENDOR=99")
|
||||
endif ()
|
||||
|
||||
if (DEFINED TARGET_CORE)
|
||||
if (BUILD_KERNEL)
|
||||
# set the C flags for just this file
|
||||
set(GETARCH2_FLAGS "-DBUILD_KERNEL")
|
||||
set(TARGET_MAKE "Makefile_kernel.conf")
|
||||
set(TARGET_CONF "config_kernel.h")
|
||||
set(TARGET_CONF_DIR ${PROJECT_BINARY_DIR}/kernel_config/${TARGET_CORE})
|
||||
else()
|
||||
set(TARGET_MAKE "Makefile.conf")
|
||||
set(TARGET_CONF "config.h")
|
||||
set(TARGET_CONF_DIR ${PROJECT_BINARY_DIR})
|
||||
endif ()
|
||||
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/c_check.cmake")
|
||||
set(TARGET_CONF_TEMP "${PROJECT_BINARY_DIR}/${TARGET_CONF}.tmp")
|
||||
|
||||
# c_check
|
||||
set(FU "")
|
||||
if (APPLE OR (MSVC AND NOT ${CMAKE_C_COMPILER_ID} MATCHES "Clang"))
|
||||
set(FU "_")
|
||||
endif()
|
||||
|
||||
set(COMPILER_ID ${CMAKE_C_COMPILER_ID})
|
||||
if (${COMPILER_ID} STREQUAL "GNU")
|
||||
set(COMPILER_ID "GCC")
|
||||
endif ()
|
||||
|
||||
string(TOUPPER ${ARCH} UC_ARCH)
|
||||
|
||||
file(WRITE ${TARGET_CONF_TEMP}
|
||||
"#define OS_${HOST_OS}\t1\n"
|
||||
"#define ARCH_${UC_ARCH}\t1\n"
|
||||
"#define C_${COMPILER_ID}\t1\n"
|
||||
"#define __${BINARY}BIT__\t1\n"
|
||||
"#define FUNDERSCORE\t${FU}\n")
|
||||
|
||||
if (${HOST_OS} STREQUAL "WINDOWSSTORE")
|
||||
file(APPEND ${TARGET_CONF_TEMP}
|
||||
"#define OS_WINNT\t1\n")
|
||||
endif ()
|
||||
|
||||
# f_check
|
||||
if (NOT NOFORTRAN)
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/f_check.cmake")
|
||||
endif ()
|
||||
|
||||
# compile getarch
|
||||
set(GETARCH_SRC
|
||||
${PROJECT_SOURCE_DIR}/getarch.c
|
||||
${CPUIDEMO}
|
||||
)
|
||||
# Cannot run getarch on target if we are cross-compiling
|
||||
if (DEFINED CORE AND CMAKE_CROSSCOMPILING)
|
||||
# Write to config as getarch would
|
||||
|
||||
if (NOT MSVC)
|
||||
list(APPEND GETARCH_SRC ${PROJECT_SOURCE_DIR}/cpuid.S)
|
||||
endif ()
|
||||
# TODO: Set up defines that getarch sets up based on every other target
|
||||
# Perhaps this should be inside a different file as it grows larger
|
||||
file(APPEND ${TARGET_CONF_TEMP}
|
||||
"#define ${CORE}\n"
|
||||
"#define CHAR_CORENAME \"${CORE}\"\n")
|
||||
if ("${CORE}" STREQUAL "ARMV7")
|
||||
file(APPEND ${TARGET_CONF_TEMP}
|
||||
"#define L1_DATA_SIZE\t65536\n"
|
||||
"#define L1_DATA_LINESIZE\t32\n"
|
||||
"#define L2_SIZE\t512488\n"
|
||||
"#define L2_LINESIZE\t32\n"
|
||||
"#define DTB_DEFAULT_ENTRIES\t64\n"
|
||||
"#define DTB_SIZE\t4096\n"
|
||||
"#define L2_ASSOCIATIVE\t4\n"
|
||||
"#define HAVE_VFPV3\n"
|
||||
"#define HAVE_VFP\n")
|
||||
set(SGEMM_UNROLL_M 4)
|
||||
set(SGEMM_UNROLL_N 4)
|
||||
set(DGEMM_UNROLL_M 4)
|
||||
set(DGEMM_UNROLL_N 4)
|
||||
elseif ("${CORE}" STREQUAL "ARMV8")
|
||||
file(APPEND ${TARGET_CONF_TEMP}
|
||||
"#define L1_DATA_SIZE\t32768\n"
|
||||
"#define L1_DATA_LINESIZE\t64\n"
|
||||
"#define L2_SIZE\t262144\n"
|
||||
"#define L2_LINESIZE\t64\n"
|
||||
"#define DTB_DEFAULT_ENTRIES\t64\n"
|
||||
"#define DTB_SIZE\t4096\n"
|
||||
"#define L2_ASSOCIATIVE\t32\n")
|
||||
set(SGEMM_UNROLL_M 4)
|
||||
set(SGEMM_UNROLL_N 4)
|
||||
elseif ("${CORE}" STREQUAL "CORTEXA57")
|
||||
file(APPEND ${TARGET_CONF_TEMP}
|
||||
"#define L1_CODE_SIZE\t49152\n"
|
||||
"#define L1_CODE_LINESIZE\t64\n"
|
||||
"#define L1_CODE_ASSOCIATIVE\t3\n"
|
||||
"#define L1_DATA_SIZE\t32768\n"
|
||||
"#define L1_DATA_LINESIZE\t64\n"
|
||||
"#define L1_DATA_ASSOCIATIVE\t2\n"
|
||||
"#define L2_SIZE\t2097152\n"
|
||||
"#define L2_LINESIZE\t64\n"
|
||||
"#define L2_ASSOCIATIVE\t16\n"
|
||||
"#define DTB_DEFAULT_ENTRIES\t64\n"
|
||||
"#define DTB_SIZE\t4096\n"
|
||||
"#define HAVE_VFPV4\n"
|
||||
"#define HAVE_VFPV3\n"
|
||||
"#define HAVE_VFP\n"
|
||||
"#define HAVE_NEON\n")
|
||||
set(SGEMM_UNROLL_M 16)
|
||||
set(SGEMM_UNROLL_N 4)
|
||||
set(DGEMM_UNROLL_M 8)
|
||||
set(DGEMM_UNROLL_N 4)
|
||||
set(CGEMM_UNROLL_M 8)
|
||||
set(CGEMM_UNROLL_N 4)
|
||||
set(ZGEMM_UNROLL_M 8)
|
||||
set(ZGEMM_UNROLL_N 4)
|
||||
endif()
|
||||
|
||||
# Or should this actually be NUM_CORES?
|
||||
if (${NUM_THREADS} GREATER 0)
|
||||
file(APPEND ${TARGET_CONF_TEMP} "#define NUM_CORES\t${NUM_THREADS}\n")
|
||||
endif()
|
||||
|
||||
# GetArch_2nd
|
||||
foreach(float_char S;D;Q;C;Z;X)
|
||||
if (NOT DEFINED ${float_char}GEMM_UNROLL_M)
|
||||
set(${float_char}GEMM_UNROLL_M 2)
|
||||
endif()
|
||||
if (NOT DEFINED ${float_char}GEMM_UNROLL_N)
|
||||
set(${float_char}GEMM_UNROLL_N 2)
|
||||
endif()
|
||||
endforeach()
|
||||
file(APPEND ${TARGET_CONF_TEMP}
|
||||
"#define GEMM_MULTITHREAD_THRESHOLD\t${GEMM_MULTITHREAD_THRESHOLD}\n")
|
||||
# Move to where gen_config_h would place it
|
||||
file(RENAME ${TARGET_CONF_TEMP} "${TARGET_CONF_DIR}/${TARGET_CONF}")
|
||||
|
||||
else(NOT CMAKE_CROSSCOMPILING)
|
||||
# compile getarch
|
||||
set(GETARCH_SRC
|
||||
${PROJECT_SOURCE_DIR}/getarch.c
|
||||
${CPUIDEMU}
|
||||
)
|
||||
|
||||
if ("${CMAKE_C_COMPILER_ID}" STREQUAL "MSVC")
|
||||
#Use generic for MSVC now
|
||||
message("MSVC")
|
||||
set(GETARCH_FLAGS ${GETARCH_FLAGS} -DFORCE_GENERIC)
|
||||
else()
|
||||
list(APPEND GETARCH_SRC ${PROJECT_SOURCE_DIR}/cpuid.S)
|
||||
endif ()
|
||||
|
||||
if ("${CMAKE_SYSTEM_NAME}" STREQUAL "WindowsStore")
|
||||
# disable WindowsStore strict CRT checks
|
||||
set(GETARCH_FLAGS ${GETARCH_FLAGS} -D_CRT_SECURE_NO_WARNINGS)
|
||||
endif ()
|
||||
|
||||
set(GETARCH_DIR "${PROJECT_BINARY_DIR}/getarch_build")
|
||||
set(GETARCH_BIN "getarch${CMAKE_EXECUTABLE_SUFFIX}")
|
||||
file(MAKE_DIRECTORY ${GETARCH_DIR})
|
||||
configure_file(${TARGET_CONF_TEMP} ${GETARCH_DIR}/${TARGET_CONF} COPYONLY)
|
||||
if (NOT "${CMAKE_SYSTEM_NAME}" STREQUAL "WindowsStore")
|
||||
try_compile(GETARCH_RESULT ${GETARCH_DIR}
|
||||
SOURCES ${GETARCH_SRC}
|
||||
COMPILE_DEFINITIONS ${EXFLAGS} ${GETARCH_FLAGS} -I${GETARCH_DIR} -I"${PROJECT_SOURCE_DIR}" -I"${PROJECT_BINARY_DIR}"
|
||||
OUTPUT_VARIABLE GETARCH_LOG
|
||||
COPY_FILE ${PROJECT_BINARY_DIR}/${GETARCH_BIN}
|
||||
)
|
||||
|
||||
if (NOT ${GETARCH_RESULT})
|
||||
MESSAGE(FATAL_ERROR "Compiling getarch failed ${GETARCH_LOG}")
|
||||
endif ()
|
||||
endif ()
|
||||
message(STATUS "Running getarch")
|
||||
|
||||
# use the cmake binary w/ the -E param to run a shell command in a cross-platform way
|
||||
execute_process(COMMAND "${PROJECT_BINARY_DIR}/${GETARCH_BIN}" 0 OUTPUT_VARIABLE GETARCH_MAKE_OUT)
|
||||
execute_process(COMMAND "${PROJECT_BINARY_DIR}/${GETARCH_BIN}" 1 OUTPUT_VARIABLE GETARCH_CONF_OUT)
|
||||
|
||||
message(STATUS "GETARCH results:\n${GETARCH_MAKE_OUT}")
|
||||
|
||||
# append config data from getarch to the TARGET file and read in CMake vars
|
||||
file(APPEND ${TARGET_CONF_TEMP} ${GETARCH_CONF_OUT})
|
||||
ParseGetArchVars(${GETARCH_MAKE_OUT})
|
||||
|
||||
set(GETARCH2_DIR "${PROJECT_BINARY_DIR}/getarch2_build")
|
||||
set(GETARCH2_BIN "getarch_2nd${CMAKE_EXECUTABLE_SUFFIX}")
|
||||
file(MAKE_DIRECTORY ${GETARCH2_DIR})
|
||||
configure_file(${TARGET_CONF_TEMP} ${GETARCH2_DIR}/${TARGET_CONF} COPYONLY)
|
||||
if (NOT "${CMAKE_SYSTEM_NAME}" STREQUAL "WindowsStore")
|
||||
try_compile(GETARCH2_RESULT ${GETARCH2_DIR}
|
||||
SOURCES ${PROJECT_SOURCE_DIR}/getarch_2nd.c
|
||||
COMPILE_DEFINITIONS ${EXFLAGS} ${GETARCH_FLAGS} ${GETARCH2_FLAGS} -I${GETARCH2_DIR} -I"${PROJECT_SOURCE_DIR}" -I"${PROJECT_BINARY_DIR}"
|
||||
OUTPUT_VARIABLE GETARCH2_LOG
|
||||
COPY_FILE ${PROJECT_BINARY_DIR}/${GETARCH2_BIN}
|
||||
)
|
||||
|
||||
if (NOT ${GETARCH2_RESULT})
|
||||
MESSAGE(FATAL_ERROR "Compiling getarch_2nd failed ${GETARCH2_LOG}")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
# use the cmake binary w/ the -E param to run a shell command in a cross-platform way
|
||||
execute_process(COMMAND "${PROJECT_BINARY_DIR}/${GETARCH2_BIN}" 0 OUTPUT_VARIABLE GETARCH2_MAKE_OUT)
|
||||
execute_process(COMMAND "${PROJECT_BINARY_DIR}/${GETARCH2_BIN}" 1 OUTPUT_VARIABLE GETARCH2_CONF_OUT)
|
||||
|
||||
# append config data from getarch_2nd to the TARGET file and read in CMake vars
|
||||
file(APPEND ${TARGET_CONF_TEMP} ${GETARCH2_CONF_OUT})
|
||||
|
||||
configure_file(${TARGET_CONF_TEMP} ${TARGET_CONF_DIR}/${TARGET_CONF} COPYONLY)
|
||||
|
||||
ParseGetArchVars(${GETARCH2_MAKE_OUT})
|
||||
|
||||
if (MSVC)
|
||||
#Use generic for MSVC now
|
||||
set(GETARCH_FLAGS ${GETARCH_FLAGS} -DFORCE_GENERIC)
|
||||
endif()
|
||||
|
||||
set(GETARCH_DIR "${PROJECT_BINARY_DIR}/getarch_build")
|
||||
set(GETARCH_BIN "getarch${CMAKE_EXECUTABLE_SUFFIX}")
|
||||
file(MAKE_DIRECTORY ${GETARCH_DIR})
|
||||
try_compile(GETARCH_RESULT ${GETARCH_DIR}
|
||||
SOURCES ${GETARCH_SRC}
|
||||
COMPILE_DEFINITIONS ${EXFLAGS} ${GETARCH_FLAGS} -I${PROJECT_SOURCE_DIR}
|
||||
OUTPUT_VARIABLE GETARCH_LOG
|
||||
COPY_FILE ${PROJECT_BINARY_DIR}/${GETARCH_BIN}
|
||||
)
|
||||
|
||||
message(STATUS "Running getarch")
|
||||
|
||||
# use the cmake binary w/ the -E param to run a shell command in a cross-platform way
|
||||
execute_process(COMMAND ${PROJECT_BINARY_DIR}/${GETARCH_BIN} 0 OUTPUT_VARIABLE GETARCH_MAKE_OUT)
|
||||
execute_process(COMMAND ${PROJECT_BINARY_DIR}/${GETARCH_BIN} 1 OUTPUT_VARIABLE GETARCH_CONF_OUT)
|
||||
|
||||
message(STATUS "GETARCH results:\n${GETARCH_MAKE_OUT}")
|
||||
|
||||
# append config data from getarch to the TARGET file and read in CMake vars
|
||||
file(APPEND ${TARGET_CONF} ${GETARCH_CONF_OUT})
|
||||
ParseGetArchVars(${GETARCH_MAKE_OUT})
|
||||
|
||||
set(GETARCH2_DIR "${PROJECT_BINARY_DIR}/getarch2_build")
|
||||
set(GETARCH2_BIN "getarch_2nd${CMAKE_EXECUTABLE_SUFFIX}")
|
||||
file(MAKE_DIRECTORY ${GETARCH2_DIR})
|
||||
try_compile(GETARCH2_RESULT ${GETARCH2_DIR}
|
||||
SOURCES ${PROJECT_SOURCE_DIR}/getarch_2nd.c
|
||||
COMPILE_DEFINITIONS ${EXFLAGS} ${GETARCH_FLAGS} ${GETARCH2_FLAGS} -I${PROJECT_SOURCE_DIR}
|
||||
OUTPUT_VARIABLE GETARCH2_LOG
|
||||
COPY_FILE ${PROJECT_BINARY_DIR}/${GETARCH2_BIN}
|
||||
)
|
||||
|
||||
# use the cmake binary w/ the -E param to run a shell command in a cross-platform way
|
||||
execute_process(COMMAND ${PROJECT_BINARY_DIR}/${GETARCH2_BIN} 0 OUTPUT_VARIABLE GETARCH2_MAKE_OUT)
|
||||
execute_process(COMMAND ${PROJECT_BINARY_DIR}/${GETARCH2_BIN} 1 OUTPUT_VARIABLE GETARCH2_CONF_OUT)
|
||||
|
||||
# append config data from getarch_2nd to the TARGET file and read in CMake vars
|
||||
file(APPEND ${TARGET_CONF} ${GETARCH2_CONF_OUT})
|
||||
ParseGetArchVars(${GETARCH2_MAKE_OUT})
|
||||
|
||||
|
||||
+69
-134
@@ -2,13 +2,27 @@
|
||||
## Author: Hank Anderson <hank@statease.com>
|
||||
## Description: Ported from OpenBLAS/Makefile.system
|
||||
##
|
||||
|
||||
set(NETLIB_LAPACK_DIR "${PROJECT_SOURCE_DIR}/lapack-netlib")
|
||||
|
||||
# TODO: Makefile.system detects Darwin (mac) and switches to clang here -hpa
|
||||
# http://stackoverflow.com/questions/714100/os-detecting-makefile
|
||||
# System detection, via CMake.
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/system_check.cmake")
|
||||
|
||||
# TODO: Makefile.system sets HOSTCC = $(CC) here if not already set -hpa
|
||||
if(CMAKE_CROSSCOMPILING AND NOT DEFINED TARGET)
|
||||
# Detect target without running getarch
|
||||
if (ARM64)
|
||||
set(TARGET "ARMV8")
|
||||
elseif(ARM)
|
||||
set(TARGET "ARMV7") # TODO: Ask compiler which arch this is
|
||||
else()
|
||||
message(FATAL_ERROR "When cross compiling, a TARGET is required.")
|
||||
endif()
|
||||
endif()
|
||||
|
||||
# Other files expect CORE, which is actually TARGET and will become TARGET_CORE for kernel build. Confused yet?
|
||||
# It seems we are meant to use TARGET as input and CORE internally as kernel.
|
||||
if(NOT DEFINED CORE AND DEFINED TARGET)
|
||||
set(CORE ${TARGET})
|
||||
endif()
|
||||
|
||||
# TARGET_CORE will override TARGET which is used in DYNAMIC_ARCH=1.
|
||||
if (DEFINED TARGET_CORE)
|
||||
@@ -22,13 +36,13 @@ if (DEFINED BINARY AND DEFINED TARGET AND BINARY EQUAL 32)
|
||||
if (${TARGET} STREQUAL "HASWELL" OR ${TARGET} STREQUAL "SANDYBRIDGE")
|
||||
set(TARGET "NEHALEM")
|
||||
endif ()
|
||||
if (${TARGET} STREQUAL "BULLDOZER" OR ${TARGET} STREQUAL "PILEDRIVER")
|
||||
if (${TARGET} STREQUAL "BULLDOZER" OR ${TARGET} STREQUAL "PILEDRIVER" OR ${TARGET} STREQUAL "ZEN")
|
||||
set(TARGET "BARCELONA")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
if (DEFINED TARGET)
|
||||
message(STATUS "Targetting the ${TARGET} architecture.")
|
||||
message(STATUS "Targeting the ${TARGET} architecture.")
|
||||
set(GETARCH_FLAGS "-DFORCE_${TARGET}")
|
||||
endif ()
|
||||
|
||||
@@ -53,21 +67,16 @@ if (NO_AVX2)
|
||||
set(GETARCH_FLAGS "${GETARCH_FLAGS} -DNO_AVX2")
|
||||
endif ()
|
||||
|
||||
if (CMAKE_BUILD_TYPE STREQUAL Debug)
|
||||
set(GETARCH_FLAGS "${GETARCH_FLAGS} -g")
|
||||
if (CMAKE_BUILD_TYPE STREQUAL "Debug")
|
||||
set(GETARCH_FLAGS "${GETARCH_FLAGS} ${CMAKE_C_FLAGS_DEBUG}")
|
||||
endif ()
|
||||
|
||||
# TODO: let CMake handle this? -hpa
|
||||
#if (${QUIET_MAKE})
|
||||
# set(MAKE "${MAKE} -s")
|
||||
#endif()
|
||||
|
||||
if (NOT DEFINED NO_PARALLEL_MAKE)
|
||||
set(NO_PARALLEL_MAKE 0)
|
||||
endif ()
|
||||
set(GETARCH_FLAGS "${GETARCH_FLAGS} -DNO_PARALLEL_MAKE=${NO_PARALLEL_MAKE}")
|
||||
|
||||
if (CMAKE_CXX_COMPILER STREQUAL loongcc)
|
||||
if (CMAKE_C_COMPILER STREQUAL loongcc)
|
||||
set(GETARCH_FLAGS "${GETARCH_FLAGS} -static")
|
||||
endif ()
|
||||
|
||||
@@ -78,51 +87,40 @@ else ()
|
||||
set(ONLY_CBLAS 0)
|
||||
endif ()
|
||||
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/prebuild.cmake")
|
||||
# N.B. this is NUM_THREAD in Makefile.system which is probably a bug -hpa
|
||||
if (NOT CMAKE_CROSSCOMPILING)
|
||||
if (NOT DEFINED NUM_CORES)
|
||||
include(ProcessorCount)
|
||||
ProcessorCount(NUM_CORES)
|
||||
endif()
|
||||
|
||||
endif()
|
||||
|
||||
if (NOT DEFINED NUM_THREADS)
|
||||
set(NUM_THREADS ${NUM_CORES})
|
||||
endif ()
|
||||
if (DEFINED NUM_CORES AND NOT NUM_CORES EQUAL 0)
|
||||
# HT?
|
||||
set(NUM_THREADS ${NUM_CORES})
|
||||
else ()
|
||||
set(NUM_THREADS 0)
|
||||
endif ()
|
||||
endif()
|
||||
|
||||
if (${NUM_THREADS} EQUAL 1)
|
||||
if (${NUM_THREADS} LESS 2)
|
||||
set(USE_THREAD 0)
|
||||
elseif(NOT DEFINED USE_THREAD)
|
||||
set(USE_THREAD 1)
|
||||
endif ()
|
||||
|
||||
if (DEFINED USE_THREAD)
|
||||
if (NOT ${USE_THREAD})
|
||||
unset(SMP)
|
||||
else ()
|
||||
set(SMP 1)
|
||||
endif ()
|
||||
else ()
|
||||
# N.B. this is NUM_THREAD in Makefile.system which is probably a bug -hpa
|
||||
if (${NUM_THREADS} EQUAL 1)
|
||||
unset(SMP)
|
||||
else ()
|
||||
set(SMP 1)
|
||||
endif ()
|
||||
if (USE_THREAD)
|
||||
message(STATUS "Multi-threading enabled with ${NUM_THREADS} threads.")
|
||||
endif ()
|
||||
|
||||
if (${SMP})
|
||||
message(STATUS "SMP enabled.")
|
||||
endif ()
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/prebuild.cmake")
|
||||
|
||||
if (NOT DEFINED NEED_PIC)
|
||||
set(NEED_PIC 1)
|
||||
endif ()
|
||||
|
||||
# TODO: I think CMake should be handling all this stuff -hpa
|
||||
unset(ARFLAGS)
|
||||
set(CPP "${COMPILER} -E")
|
||||
set(AR "${CROSS_SUFFIX}ar")
|
||||
set(AS "${CROSS_SUFFIX}as")
|
||||
set(LD "${CROSS_SUFFIX}ld")
|
||||
set(RANLIB "${CROSS_SUFFIX}ranlib")
|
||||
set(NM "${CROSS_SUFFIX}nm")
|
||||
set(DLLWRAP "${CROSS_SUFFIX}dllwrap")
|
||||
set(OBJCOPY "${CROSS_SUFFIX}objcopy")
|
||||
set(OBJCONV "${CROSS_SUFFIX}objconv")
|
||||
|
||||
# OS dependent settings
|
||||
include("${PROJECT_SOURCE_DIR}/cmake/os.cmake")
|
||||
|
||||
@@ -150,11 +148,13 @@ if (NEED_PIC)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -fPIC")
|
||||
endif ()
|
||||
|
||||
if (${F_COMPILER} STREQUAL "SUN")
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -pic")
|
||||
else ()
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -fPIC")
|
||||
endif ()
|
||||
if (NOT NOFORTRAN)
|
||||
if (${F_COMPILER} STREQUAL "SUN")
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -pic")
|
||||
else ()
|
||||
set(FCOMMON_OPT "${FCOMMON_OPT} -fPIC")
|
||||
endif ()
|
||||
endif()
|
||||
endif ()
|
||||
|
||||
if (DYNAMIC_ARCH)
|
||||
@@ -175,7 +175,7 @@ if (NO_AVX)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DNO_AVX")
|
||||
endif ()
|
||||
|
||||
if (${ARCH} STREQUAL "x86")
|
||||
if (X86)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DNO_AVX")
|
||||
endif ()
|
||||
|
||||
@@ -183,25 +183,20 @@ if (NO_AVX2)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DNO_AVX2")
|
||||
endif ()
|
||||
|
||||
if (SMP)
|
||||
if (USE_THREAD)
|
||||
# USE_SIMPLE_THREADED_LEVEL3 = 1
|
||||
# NO_AFFINITY = 1
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DSMP_SERVER")
|
||||
|
||||
if (${ARCH} STREQUAL "mips64")
|
||||
if (MIPS64)
|
||||
if (NOT ${CORE} STREQUAL "LOONGSON3B")
|
||||
set(USE_SIMPLE_THREADED_LEVEL3 1)
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
if (USE_OPENMP)
|
||||
# USE_SIMPLE_THREADED_LEVEL3 = 1
|
||||
# NO_AFFINITY = 1
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DUSE_OPENMP")
|
||||
endif ()
|
||||
|
||||
if (BIGNUMA)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DBIGNUMA")
|
||||
endif ()
|
||||
|
||||
endif ()
|
||||
|
||||
if (NO_WARMUP)
|
||||
@@ -264,7 +259,7 @@ if (NOT ${CMAKE_SYSTEM_NAME} STREQUAL "Linux")
|
||||
set(NO_AFFINITY 1)
|
||||
endif ()
|
||||
|
||||
if (NOT ${ARCH} STREQUAL "x86_64" AND NOT ${ARCH} STREQUAL "x86" AND NOT ${CORE} STREQUAL "LOONGSON3B")
|
||||
if (NOT X86_64 AND NOT X86 AND NOT ${CORE} STREQUAL "LOONGSON3B")
|
||||
set(NO_AFFINITY 1)
|
||||
endif ()
|
||||
|
||||
@@ -296,50 +291,19 @@ if (MIXED_MEMORY_ALLOCATION)
|
||||
set(CCOMMON_OPT "${CCOMMON_OPT} -DMIXED_MEMORY_ALLOCATION")
|
||||
endif ()
|
||||
|
||||
if (${CMAKE_SYSTEM_NAME} STREQUAL "SunOS")
|
||||
set(TAR gtar)
|
||||
set(PATCH gpatch)
|
||||
set(GREP ggrep)
|
||||
else ()
|
||||
set(TAR tar)
|
||||
set(PATCH patch)
|
||||
set(GREP grep)
|
||||
endif ()
|
||||
|
||||
if (NOT DEFINED MD5SUM)
|
||||
set(MD5SUM md5sum)
|
||||
endif ()
|
||||
|
||||
set(AWK awk)
|
||||
|
||||
set(REVISION "-r${OpenBLAS_VERSION}")
|
||||
set(MAJOR_VERSION ${OpenBLAS_MAJOR_VERSION})
|
||||
|
||||
if (DEBUG)
|
||||
set(COMMON_OPT "${COMMON_OPT} -g")
|
||||
endif ()
|
||||
|
||||
if (NOT DEFINED COMMON_OPT)
|
||||
set(COMMON_OPT "-O2")
|
||||
endif ()
|
||||
|
||||
#For x86 32-bit
|
||||
if (DEFINED BINARY AND BINARY EQUAL 32)
|
||||
if (NOT MSVC)
|
||||
set(COMMON_OPT "${COMMON_OPT} -m32")
|
||||
endif()
|
||||
endif()
|
||||
|
||||
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${COMMON_OPT} ${CCOMMON_OPT}")
|
||||
set(CMAKE_C_FLAGS "${CMAKE_C_FLAGS} ${CCOMMON_OPT}")
|
||||
if(NOT MSVC)
|
||||
set(CMAKE_ASM_FLAGS "${CMAKE_ASM_FLAGS} ${COMMON_OPT} ${CCOMMON_OPT}")
|
||||
set(CMAKE_ASM_FLAGS "${CMAKE_ASM_FLAGS} ${CCOMMON_OPT}")
|
||||
endif()
|
||||
# TODO: not sure what PFLAGS is -hpa
|
||||
set(PFLAGS "${PFLAGS} ${COMMON_OPT} ${CCOMMON_OPT} -I${TOPDIR} -DPROFILE ${COMMON_PROF}")
|
||||
set(PFLAGS "${PFLAGS} ${CCOMMON_OPT} -I${TOPDIR} -DPROFILE ${COMMON_PROF}")
|
||||
|
||||
set(CMAKE_Fortran_FLAGS "${CMAKE_Fortran_FLAGS} ${COMMON_OPT} ${FCOMMON_OPT}")
|
||||
set(CMAKE_Fortran_FLAGS "${CMAKE_Fortran_FLAGS} ${FCOMMON_OPT}")
|
||||
# TODO: not sure what FPFLAGS is -hpa
|
||||
set(FPFLAGS "${FPFLAGS} ${COMMON_OPT} ${FCOMMON_OPT} ${COMMON_PROF}")
|
||||
set(FPFLAGS "${FPFLAGS} ${FCOMMON_OPT} ${COMMON_PROF}")
|
||||
|
||||
#For LAPACK Fortran codes.
|
||||
set(LAPACK_FFLAGS "${LAPACK_FFLAGS} ${CMAKE_Fortran_FLAGS}")
|
||||
@@ -347,7 +311,7 @@ set(LAPACK_FPFLAGS "${LAPACK_FPFLAGS} ${FPFLAGS}")
|
||||
|
||||
#Disable -fopenmp for LAPACK Fortran codes on Windows.
|
||||
if (${CMAKE_SYSTEM_NAME} STREQUAL "Windows")
|
||||
set(FILTER_FLAGS "-fopenmp;-mp;-openmp;-xopenmp=parralel")
|
||||
set(FILTER_FLAGS "-fopenmp;-mp;-openmp;-xopenmp=parallel")
|
||||
foreach (FILTER_FLAG ${FILTER_FLAGS})
|
||||
string(REPLACE ${FILTER_FLAG} "" LAPACK_FFLAGS ${LAPACK_FFLAGS})
|
||||
string(REPLACE ${FILTER_FLAG} "" LAPACK_FPFLAGS ${LAPACK_FPFLAGS})
|
||||
@@ -385,7 +349,7 @@ if (NOT DEFINED LIBSUFFIX)
|
||||
endif ()
|
||||
|
||||
if (DYNAMIC_ARCH)
|
||||
if (DEFINED SMP)
|
||||
if (USE_THREAD)
|
||||
set(LIBNAME "${LIBPREFIX}p${REVISION}.${LIBSUFFIX}")
|
||||
set(LIBNAME_P "${LIBPREFIX}p${REVISION}_p.${LIBSUFFIX}")
|
||||
else ()
|
||||
@@ -393,7 +357,7 @@ if (DYNAMIC_ARCH)
|
||||
set(LIBNAME_P "${LIBPREFIX}${REVISION}_p.${LIBSUFFIX}")
|
||||
endif ()
|
||||
else ()
|
||||
if (DEFINED SMP)
|
||||
if (USE_THREAD)
|
||||
set(LIBNAME "${LIBPREFIX}_${LIBCORE}p${REVISION}.${LIBSUFFIX}")
|
||||
set(LIBNAME_P "${LIBPREFIX}_${LIBCORE}p${REVISION}_p.${LIBSUFFIX}")
|
||||
else ()
|
||||
@@ -424,6 +388,9 @@ if (NOT NO_LAPACK)
|
||||
if (NOT NO_LAPACKE)
|
||||
set(LIB_COMPONENTS "${LIB_COMPONENTS} LAPACKE")
|
||||
endif ()
|
||||
if (BUILD_RELAPACK)
|
||||
set(LIB_COMPONENTS "${LIB_COMPONENTS} ReLAPACK")
|
||||
endif ()
|
||||
endif ()
|
||||
|
||||
if (ONLY_CBLAS)
|
||||
@@ -435,7 +402,7 @@ endif ()
|
||||
set(USE_GEMM3M 0)
|
||||
|
||||
if (DEFINED ARCH)
|
||||
if (${ARCH} STREQUAL "x86" OR ${ARCH} STREQUAL "x86_64" OR ${ARCH} STREQUAL "ia64" OR ${ARCH} STREQUAL "MIPS")
|
||||
if (X86 OR X86_64 OR ${ARCH} STREQUAL "ia64" OR MIPS64)
|
||||
set(USE_GEMM3M 1)
|
||||
endif ()
|
||||
|
||||
@@ -518,35 +485,3 @@ endif ()
|
||||
# export CUFLAGS
|
||||
# export CULIB
|
||||
#endif
|
||||
|
||||
#.SUFFIXES: .$(PSUFFIX) .$(SUFFIX) .f
|
||||
#
|
||||
#.f.$(SUFFIX):
|
||||
# $(FC) $(FFLAGS) -c $< -o $(@F)
|
||||
#
|
||||
#.f.$(PSUFFIX):
|
||||
# $(FC) $(FPFLAGS) -pg -c $< -o $(@F)
|
||||
|
||||
# these are not cross-platform
|
||||
#ifdef BINARY64
|
||||
#PATHSCALEPATH = /opt/pathscale/lib/3.1
|
||||
#PGIPATH = /opt/pgi/linux86-64/7.1-5/lib
|
||||
#else
|
||||
#PATHSCALEPATH = /opt/pathscale/lib/3.1/32
|
||||
#PGIPATH = /opt/pgi/linux86/7.1-5/lib
|
||||
#endif
|
||||
|
||||
#ACMLPATH = /opt/acml/4.3.0
|
||||
#ifneq ($(OSNAME), Darwin)
|
||||
#MKLPATH = /opt/intel/mkl/10.2.2.025/lib
|
||||
#else
|
||||
#MKLPATH = /Library/Frameworks/Intel_MKL.framework/Versions/10.0.1.014/lib
|
||||
#endif
|
||||
#ATLASPATH = /opt/atlas/3.9.17/opteron
|
||||
#FLAMEPATH = $(HOME)/flame/lib
|
||||
#ifneq ($(OSNAME), SunOS)
|
||||
#SUNPATH = /opt/sunstudio12.1
|
||||
#else
|
||||
#SUNPATH = /opt/SUNWspro
|
||||
#endif
|
||||
|
||||
|
||||
@@ -0,0 +1,68 @@
|
||||
##
|
||||
## Author: Hank Anderson <hank@statease.com>
|
||||
## Description: Ported from the OpenBLAS/c_check perl script.
|
||||
## This is triggered by prebuild.cmake and runs before any of the code is built.
|
||||
## Creates config.h and Makefile.conf.
|
||||
|
||||
# Convert CMake vars into the format that OpenBLAS expects
|
||||
string(TOUPPER ${CMAKE_SYSTEM_NAME} HOST_OS)
|
||||
if (${HOST_OS} STREQUAL "WINDOWS")
|
||||
set(HOST_OS WINNT)
|
||||
endif ()
|
||||
|
||||
if(CMAKE_COMPILER_IS_GNUCC AND WIN32)
|
||||
execute_process(COMMAND ${CMAKE_C_COMPILER} -dumpmachine
|
||||
OUTPUT_VARIABLE OPENBLAS_GCC_TARGET_MACHINE
|
||||
OUTPUT_STRIP_TRAILING_WHITESPACE)
|
||||
if(OPENBLAS_GCC_TARGET_MACHINE MATCHES "amd64|x86_64|AMD64")
|
||||
set(MINGW64 1)
|
||||
endif()
|
||||
endif()
|
||||
|
||||
# Pretty thorough determination of arch. Add more if needed
|
||||
if(CMAKE_CL_64 OR MINGW64)
|
||||
set(X86_64 1)
|
||||
elseif(MINGW OR (MSVC AND NOT CMAKE_CROSSCOMPILING))
|
||||
set(X86 1)
|
||||
elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "ppc.*|power.*|Power.*")
|
||||
set(PPC 1)
|
||||
elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "mips64.*")
|
||||
set(MIPS64 1)
|
||||
elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "amd64.*|x86_64.*|AMD64.*")
|
||||
set(X86_64 1)
|
||||
elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "i686.*|i386.*|x86.*|amd64.*|AMD64.*")
|
||||
set(X86 1)
|
||||
elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "^(arm.*|ARM.*)")
|
||||
set(ARM 1)
|
||||
elseif(CMAKE_SYSTEM_PROCESSOR MATCHES "^(aarch64.*|AARCH64.*)")
|
||||
set(ARM64 1)
|
||||
endif()
|
||||
|
||||
if (X86_64)
|
||||
set(ARCH "x86_64")
|
||||
elseif(X86)
|
||||
set(ARCH "x86")
|
||||
elseif(PPC)
|
||||
set(ARCH "power")
|
||||
elseif(ARM)
|
||||
set(ARCH "arm")
|
||||
elseif(ARM64)
|
||||
set(ARCH "arm64")
|
||||
else()
|
||||
set(ARCH ${CMAKE_SYSTEM_PROCESSOR} CACHE STRING "Target Architecture")
|
||||
endif ()
|
||||
|
||||
if (NOT BINARY)
|
||||
if (X86_64 OR ARM64 OR PPC OR MIPS64)
|
||||
set(BINARY 64)
|
||||
else ()
|
||||
set(BINARY 32)
|
||||
endif ()
|
||||
endif()
|
||||
|
||||
if(BINARY EQUAL 64)
|
||||
set(BINARY64 1)
|
||||
else()
|
||||
set(BINARY32 1)
|
||||
endif()
|
||||
|
||||
+5
-1
@@ -202,6 +202,8 @@ function(GenerateNamedObjects sources_in)
|
||||
if (use_cblas)
|
||||
set(obj_name "cblas_${obj_name}")
|
||||
list(APPEND obj_defines "CBLAS")
|
||||
elseif (NOT "${obj_name}" MATCHES "${ARCH_SUFFIX}")
|
||||
set(obj_name "${obj_name}${ARCH_SUFFIX}")
|
||||
endif ()
|
||||
|
||||
list(APPEND obj_defines "ASMNAME=${FU}${obj_name};ASMFNAME=${FU}${obj_name}${BU};NAME=${obj_name}${BU};CNAME=${obj_name};CHAR_NAME=\"${obj_name}${BU}\";CHAR_CNAME=\"${obj_name}\"")
|
||||
@@ -234,7 +236,9 @@ function(GenerateNamedObjects sources_in)
|
||||
|
||||
string(REPLACE ";" "\n#define " define_source "${obj_defines}")
|
||||
string(REPLACE "=" " " define_source "${define_source}")
|
||||
file(WRITE ${new_source_file} "#define ${define_source}\n#include \"${old_source_file}\"")
|
||||
file(WRITE ${new_source_file}.tmp "#define ${define_source}\n#include \"${old_source_file}\"")
|
||||
configure_file(${new_source_file}.tmp ${new_source_file} COPYONLY)
|
||||
file(REMOVE ${new_source_file}.tmp)
|
||||
list(APPEND SRC_LIST_OUT ${new_source_file})
|
||||
|
||||
endforeach ()
|
||||
|
||||
@@ -93,7 +93,7 @@ extern "C" {
|
||||
#include <sched.h>
|
||||
#endif
|
||||
|
||||
#if defined(OS_DARWIN) || defined(OS_FREEBSD) || defined(OS_NETBSD) || defined(OS_ANDROID)
|
||||
#if defined(OS_DARWIN) || defined(OS_FREEBSD) || defined(OS_NETBSD) || defined(OS_OPENBSD) || defined(OS_DRAGONFLY) || defined(OS_ANDROID)
|
||||
#include <sched.h>
|
||||
#endif
|
||||
|
||||
@@ -425,6 +425,10 @@ please https://github.com/xianyi/OpenBLAS/issues/246
|
||||
#endif
|
||||
|
||||
#ifndef ASSEMBLER
|
||||
#ifdef OS_WINDOWSSTORE
|
||||
typedef char env_var_t[MAX_PATH];
|
||||
#define readenv(p, n) 0
|
||||
#else
|
||||
#ifdef OS_WINDOWS
|
||||
typedef char env_var_t[MAX_PATH];
|
||||
#define readenv(p, n) GetEnvironmentVariable((LPCTSTR)(n), (LPTSTR)(p), sizeof(p))
|
||||
@@ -432,6 +436,7 @@ typedef char env_var_t[MAX_PATH];
|
||||
typedef char* env_var_t;
|
||||
#define readenv(p, n) ((p)=getenv(n))
|
||||
#endif
|
||||
#endif
|
||||
|
||||
#if !defined(RPCC_DEFINED) && !defined(OS_WINDOWS)
|
||||
#ifdef _POSIX_MONOTONIC_CLOCK
|
||||
@@ -490,6 +495,33 @@ static void __inline blas_lock(volatile BLASULONG *address){
|
||||
#define MMAP_POLICY (MAP_PRIVATE | MAP_ANONYMOUS)
|
||||
#endif
|
||||
|
||||
#ifndef ASSEMBLER
|
||||
/* C99 supports complex floating numbers natively, which GCC also offers as an
|
||||
extension since version 3.0. If neither are available, use a compatible
|
||||
structure as fallback (see Clause 6.2.5.13 of the C99 standard). */
|
||||
#if ((defined(__STDC_IEC_559_COMPLEX__) || __STDC_VERSION__ >= 199901L || \
|
||||
(__GNUC__ >= 3 && !defined(__cplusplus))) && !(defined(FORCE_OPENBLAS_COMPLEX_STRUCT))) && !defined(_MSC_VER)
|
||||
#define OPENBLAS_COMPLEX_C99
|
||||
#ifndef __cplusplus
|
||||
#include <complex.h>
|
||||
#endif
|
||||
typedef float _Complex openblas_complex_float;
|
||||
typedef double _Complex openblas_complex_double;
|
||||
typedef xdouble _Complex openblas_complex_xdouble;
|
||||
#define openblas_make_complex_float(real, imag) ((real) + ((imag) * _Complex_I))
|
||||
#define openblas_make_complex_double(real, imag) ((real) + ((imag) * _Complex_I))
|
||||
#define openblas_make_complex_xdouble(real, imag) ((real) + ((imag) * _Complex_I))
|
||||
#else
|
||||
#define OPENBLAS_COMPLEX_STRUCT
|
||||
typedef struct { float real, imag; } openblas_complex_float;
|
||||
typedef struct { double real, imag; } openblas_complex_double;
|
||||
typedef struct { xdouble real, imag; } openblas_complex_xdouble;
|
||||
#define openblas_make_complex_float(real, imag) {(real), (imag)}
|
||||
#define openblas_make_complex_double(real, imag) {(real), (imag)}
|
||||
#define openblas_make_complex_xdouble(real, imag) {(real), (imag)}
|
||||
#endif
|
||||
#endif
|
||||
|
||||
#include "param.h"
|
||||
#include "common_param.h"
|
||||
|
||||
@@ -519,31 +551,6 @@ static void __inline blas_lock(volatile BLASULONG *address){
|
||||
#include <stdio.h>
|
||||
#endif // NOINCLUDE
|
||||
|
||||
/* C99 supports complex floating numbers natively, which GCC also offers as an
|
||||
extension since version 3.0. If neither are available, use a compatible
|
||||
structure as fallback (see Clause 6.2.5.13 of the C99 standard). */
|
||||
#if ((defined(__STDC_IEC_559_COMPLEX__) || __STDC_VERSION__ >= 199901L || \
|
||||
(__GNUC__ >= 3 && !defined(__cplusplus))) && !(defined(FORCE_OPENBLAS_COMPLEX_STRUCT)))
|
||||
#define OPENBLAS_COMPLEX_C99
|
||||
#ifndef __cplusplus
|
||||
#include <complex.h>
|
||||
#endif
|
||||
typedef float _Complex openblas_complex_float;
|
||||
typedef double _Complex openblas_complex_double;
|
||||
typedef xdouble _Complex openblas_complex_xdouble;
|
||||
#define openblas_make_complex_float(real, imag) ((real) + ((imag) * _Complex_I))
|
||||
#define openblas_make_complex_double(real, imag) ((real) + ((imag) * _Complex_I))
|
||||
#define openblas_make_complex_xdouble(real, imag) ((real) + ((imag) * _Complex_I))
|
||||
#else
|
||||
#define OPENBLAS_COMPLEX_STRUCT
|
||||
typedef struct { float real, imag; } openblas_complex_float;
|
||||
typedef struct { double real, imag; } openblas_complex_double;
|
||||
typedef struct { xdouble real, imag; } openblas_complex_xdouble;
|
||||
#define openblas_make_complex_float(real, imag) {(real), (imag)}
|
||||
#define openblas_make_complex_double(real, imag) {(real), (imag)}
|
||||
#define openblas_make_complex_xdouble(real, imag) {(real), (imag)}
|
||||
#endif
|
||||
|
||||
#ifdef XDOUBLE
|
||||
#define OPENBLAS_COMPLEX_FLOAT openblas_complex_xdouble
|
||||
#define OPENBLAS_MAKE_COMPLEX_FLOAT(r,i) openblas_make_complex_xdouble(r,i)
|
||||
@@ -556,8 +563,13 @@ static void __inline blas_lock(volatile BLASULONG *address){
|
||||
#endif
|
||||
|
||||
#if defined(C_PGI) || defined(C_SUN)
|
||||
#define CREAL(X) (*((FLOAT *)&X + 0))
|
||||
#define CIMAG(X) (*((FLOAT *)&X + 1))
|
||||
#if defined(__STDC_IEC_559_COMPLEX__)
|
||||
#define CREAL(X) creal(X)
|
||||
#define CIMAG(X) cimag(X)
|
||||
#else
|
||||
#define CREAL(X) (*((FLOAT *)&X + 0))
|
||||
#define CIMAG(X) (*((FLOAT *)&X + 1))
|
||||
#endif
|
||||
#else
|
||||
#ifdef OPENBLAS_COMPLEX_STRUCT
|
||||
#define CREAL(Z) ((Z).real)
|
||||
@@ -649,7 +661,11 @@ static __inline void blas_unlock(volatile BLASULONG *address){
|
||||
*address = 0;
|
||||
}
|
||||
|
||||
|
||||
#ifdef OS_WINDOWSSTORE
|
||||
static __inline int readenv_atoi(char *env) {
|
||||
return 0;
|
||||
}
|
||||
#else
|
||||
#ifdef OS_WINDOWS
|
||||
static __inline int readenv_atoi(char *env) {
|
||||
env_var_t p;
|
||||
@@ -664,7 +680,7 @@ static __inline int readenv_atoi(char *env) {
|
||||
return(0);
|
||||
}
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
||||
#if !defined(XDOUBLE) || !defined(QUAD_PRECISION)
|
||||
|
||||
|
||||
@@ -111,11 +111,6 @@ REALNAME:
|
||||
|
||||
#define PROFCODE
|
||||
|
||||
#ifdef __ARM_PCS
|
||||
//-mfloat-abi=softfp
|
||||
#define SOFT_FLOAT_ABI
|
||||
#endif
|
||||
|
||||
#endif
|
||||
|
||||
|
||||
|
||||
@@ -39,7 +39,11 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
|
||||
#define INLINE inline
|
||||
|
||||
#ifdef F_INTERFACE_FLANG
|
||||
#define RETURN_BY_STACK
|
||||
#else
|
||||
#define RETURN_BY_COMPLEX
|
||||
#endif
|
||||
|
||||
#ifndef ASSEMBLER
|
||||
|
||||
|
||||
+2
-7
@@ -33,8 +33,8 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#ifndef COMMON_MIPS
|
||||
#define COMMON_MIPS
|
||||
|
||||
#define MB
|
||||
#define WMB
|
||||
#define MB __sync_synchronize()
|
||||
#define WMB __sync_synchronize()
|
||||
|
||||
#define INLINE inline
|
||||
|
||||
@@ -42,11 +42,6 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
|
||||
#ifndef ASSEMBLER
|
||||
|
||||
static void INLINE blas_lock(volatile unsigned long *address){
|
||||
|
||||
}
|
||||
#define BLAS_LOCK_DEFINED
|
||||
|
||||
static inline unsigned int rpcc(void){
|
||||
unsigned long ret;
|
||||
|
||||
|
||||
+2
-24
@@ -71,35 +71,13 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#ifndef COMMON_MIPS64
|
||||
#define COMMON_MIPS64
|
||||
|
||||
#define MB
|
||||
#define WMB
|
||||
#define MB __sync_synchronize()
|
||||
#define WMB __sync_synchronize()
|
||||
|
||||
#define INLINE inline
|
||||
|
||||
#ifndef ASSEMBLER
|
||||
|
||||
static void INLINE blas_lock(volatile unsigned long *address){
|
||||
|
||||
long int ret, val = 1;
|
||||
|
||||
do {
|
||||
while (*address) {YIELDING;};
|
||||
|
||||
__asm__ __volatile__(
|
||||
"1: ll %0, %3\n"
|
||||
" ori %2, %0, 1\n"
|
||||
" sc %2, %1\n"
|
||||
" beqz %2, 1b\n"
|
||||
" andi %2, %0, 1\n"
|
||||
" sync\n"
|
||||
: "=&r" (val), "=m" (address), "=&r" (ret)
|
||||
: "m" (address)
|
||||
: "memory");
|
||||
|
||||
} while (ret);
|
||||
}
|
||||
#define BLAS_LOCK_DEFINED
|
||||
|
||||
static inline unsigned int rpcc(void){
|
||||
unsigned long ret;
|
||||
|
||||
|
||||
+7
-7
@@ -333,8 +333,8 @@ BLASLONG (*icamin_k)(BLASLONG, float *, BLASLONG);
|
||||
float (*cnrm2_k) (BLASLONG, float *, BLASLONG);
|
||||
float (*casum_k) (BLASLONG, float *, BLASLONG);
|
||||
int (*ccopy_k) (BLASLONG, float *, BLASLONG, float *, BLASLONG);
|
||||
float _Complex (*cdotu_k) (BLASLONG, float *, BLASLONG, float *, BLASLONG);
|
||||
float _Complex (*cdotc_k) (BLASLONG, float *, BLASLONG, float *, BLASLONG);
|
||||
openblas_complex_float (*cdotu_k) (BLASLONG, float *, BLASLONG, float *, BLASLONG);
|
||||
openblas_complex_float (*cdotc_k) (BLASLONG, float *, BLASLONG, float *, BLASLONG);
|
||||
int (*csrot_k) (BLASLONG, float *, BLASLONG, float *, BLASLONG, float, float);
|
||||
|
||||
int (*caxpy_k) (BLASLONG, BLASLONG, BLASLONG, float, float, float *, BLASLONG, float *, BLASLONG, float *, BLASLONG);
|
||||
@@ -496,8 +496,8 @@ BLASLONG (*izamin_k)(BLASLONG, double *, BLASLONG);
|
||||
double (*znrm2_k) (BLASLONG, double *, BLASLONG);
|
||||
double (*zasum_k) (BLASLONG, double *, BLASLONG);
|
||||
int (*zcopy_k) (BLASLONG, double *, BLASLONG, double *, BLASLONG);
|
||||
double _Complex (*zdotu_k) (BLASLONG, double *, BLASLONG, double *, BLASLONG);
|
||||
double _Complex (*zdotc_k) (BLASLONG, double *, BLASLONG, double *, BLASLONG);
|
||||
openblas_complex_double (*zdotu_k) (BLASLONG, double *, BLASLONG, double *, BLASLONG);
|
||||
openblas_complex_double (*zdotc_k) (BLASLONG, double *, BLASLONG, double *, BLASLONG);
|
||||
int (*zdrot_k) (BLASLONG, double *, BLASLONG, double *, BLASLONG, double, double);
|
||||
|
||||
int (*zaxpy_k) (BLASLONG, BLASLONG, BLASLONG, double, double, double *, BLASLONG, double *, BLASLONG, double *, BLASLONG);
|
||||
@@ -661,8 +661,8 @@ BLASLONG (*ixamin_k)(BLASLONG, xdouble *, BLASLONG);
|
||||
xdouble (*xnrm2_k) (BLASLONG, xdouble *, BLASLONG);
|
||||
xdouble (*xasum_k) (BLASLONG, xdouble *, BLASLONG);
|
||||
int (*xcopy_k) (BLASLONG, xdouble *, BLASLONG, xdouble *, BLASLONG);
|
||||
xdouble _Complex (*xdotu_k) (BLASLONG, xdouble *, BLASLONG, xdouble *, BLASLONG);
|
||||
xdouble _Complex (*xdotc_k) (BLASLONG, xdouble *, BLASLONG, xdouble *, BLASLONG);
|
||||
openblas_complex_xdouble (*xdotu_k) (BLASLONG, xdouble *, BLASLONG, xdouble *, BLASLONG);
|
||||
openblas_complex_xdouble (*xdotc_k) (BLASLONG, xdouble *, BLASLONG, xdouble *, BLASLONG);
|
||||
int (*xqrot_k) (BLASLONG, xdouble *, BLASLONG, xdouble *, BLASLONG, xdouble, xdouble);
|
||||
|
||||
int (*xaxpy_k) (BLASLONG, BLASLONG, BLASLONG, xdouble, xdouble, xdouble *, BLASLONG, xdouble *, BLASLONG, xdouble *, BLASLONG);
|
||||
@@ -888,7 +888,7 @@ BLASLONG (*ixamin_k)(BLASLONG, xdouble *, BLASLONG);
|
||||
int (*sgeadd_k) (BLASLONG, BLASLONG, float, float *, BLASLONG, float, float *, BLASLONG);
|
||||
int (*dgeadd_k) (BLASLONG, BLASLONG, double, double *, BLASLONG, double, double *, BLASLONG);
|
||||
int (*cgeadd_k) (BLASLONG, BLASLONG, float, float, float *, BLASLONG, float, float, float *, BLASLONG);
|
||||
int (*zgeadd_k) (BLASLONG, BLASLONG, float, double, double *, BLASLONG, double, double, double *, BLASLONG);
|
||||
int (*zgeadd_k) (BLASLONG, BLASLONG, double, double, double *, BLASLONG, double, double, double *, BLASLONG);
|
||||
|
||||
} gotoblas_t;
|
||||
|
||||
|
||||
+1
-1
@@ -327,7 +327,7 @@ REALNAME:
|
||||
#endif
|
||||
#endif
|
||||
|
||||
#if defined(OS_LINUX) || defined(OS_FREEBSD) || defined(OS_NETBSD) || defined(__ELF__)
|
||||
#if defined(OS_LINUX) || defined(OS_FREEBSD) || defined(OS_NETBSD) || defined(OS_OPENBSD) || defined(__ELF__)
|
||||
#define PROLOGUE \
|
||||
.text; \
|
||||
.align 16; \
|
||||
|
||||
+5
-1
@@ -245,6 +245,10 @@ static __inline int blas_quickdivide(unsigned int x, unsigned int y){
|
||||
#define RETURN_BY_STACK
|
||||
#endif
|
||||
|
||||
#ifdef F_INTERFACE_FLANG
|
||||
#define RETURN_BY_STACK
|
||||
#endif
|
||||
|
||||
#ifdef F_INTERFACE_PGI
|
||||
#define RETURN_BY_STACK
|
||||
#endif
|
||||
@@ -399,7 +403,7 @@ REALNAME:
|
||||
#define EPILOGUE .end
|
||||
#endif
|
||||
|
||||
#if defined(OS_LINUX) || defined(OS_FREEBSD) || defined(OS_NETBSD) || defined(__ELF__) || defined(C_PGI)
|
||||
#if defined(OS_LINUX) || defined(OS_FREEBSD) || defined(OS_NETBSD) || defined(OS_OPENBSD) || defined(OS_DRAGONFLY) || defined(__ELF__) || defined(C_PGI)
|
||||
#define PROLOGUE \
|
||||
.text; \
|
||||
.align 512; \
|
||||
|
||||
@@ -114,6 +114,7 @@
|
||||
#define CORE_HASWELL 24
|
||||
#define CORE_STEAMROLLER 25
|
||||
#define CORE_EXCAVATOR 26
|
||||
#define CORE_ZEN 27
|
||||
|
||||
#define HAVE_SSE (1 << 0)
|
||||
#define HAVE_SSE2 (1 << 1)
|
||||
@@ -209,5 +210,6 @@ typedef struct {
|
||||
#define CPUTYPE_HASWELL 48
|
||||
#define CPUTYPE_STEAMROLLER 49
|
||||
#define CPUTYPE_EXCAVATOR 50
|
||||
#define CPUTYPE_ZEN 51
|
||||
|
||||
#endif
|
||||
|
||||
+4
-3
@@ -114,8 +114,9 @@ int detect(void)
|
||||
|
||||
fclose(infile);
|
||||
if(cpu_part != NULL && cpu_implementer != NULL) {
|
||||
if (strstr(cpu_part, "0xd07") && strstr(cpu_implementer, "0x41"))
|
||||
return CPU_CORTEXA57;
|
||||
if (strstr(cpu_implementer, "0x41") &&
|
||||
(strstr(cpu_part, "0xd07") || strstr(cpu_part,"0xd08") || strstr(cpu_part,"0xd03") ))
|
||||
return CPU_CORTEXA57; //or compatible A53, A72
|
||||
else if (strstr(cpu_part, "0x516") && strstr(cpu_implementer, "0x42"))
|
||||
return CPU_VULCAN;
|
||||
else if (strstr(cpu_part, "0x0a1") && strstr(cpu_implementer, "0x43"))
|
||||
@@ -142,7 +143,7 @@ int detect(void)
|
||||
if(p != NULL)
|
||||
{
|
||||
|
||||
if (strstr(p, "AArch64"))
|
||||
if ((strstr(p, "AArch64")) || (strstr(p, "8")))
|
||||
{
|
||||
return CPU_ARMV8;
|
||||
|
||||
|
||||
+16
-1
@@ -76,6 +76,7 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
||||
#define CPU_LOONGSON3B 3
|
||||
#define CPU_I6400 4
|
||||
#define CPU_P6600 5
|
||||
#define CPU_I6500 6
|
||||
|
||||
static char *cpuname[] = {
|
||||
"UNKOWN",
|
||||
@@ -83,7 +84,8 @@ static char *cpuname[] = {
|
||||
"LOONGSON3A",
|
||||
"LOONGSON3B",
|
||||
"I6400",
|
||||
"P6600"
|
||||
"P6600",
|
||||
"I6500"
|
||||
};
|
||||
|
||||
int detect(void){
|
||||
@@ -165,6 +167,8 @@ void get_subarchitecture(void){
|
||||
printf("I6400");
|
||||
}else if(detect()==CPU_P6600){
|
||||
printf("P6600");
|
||||
}else if(detect()==CPU_I6500){
|
||||
printf("I6500");
|
||||
}else{
|
||||
printf("SICORTEX");
|
||||
}
|
||||
@@ -211,6 +215,15 @@ void get_cpuconfig(void){
|
||||
printf("#define DTB_DEFAULT_ENTRIES 64\n");
|
||||
printf("#define DTB_SIZE 4096\n");
|
||||
printf("#define L2_ASSOCIATIVE 8\n");
|
||||
}else if(detect()==CPU_I6500){
|
||||
printf("#define I6500\n");
|
||||
printf("#define L1_DATA_SIZE 65536\n");
|
||||
printf("#define L1_DATA_LINESIZE 32\n");
|
||||
printf("#define L2_SIZE 1048576\n");
|
||||
printf("#define L2_LINESIZE 32\n");
|
||||
printf("#define DTB_DEFAULT_ENTRIES 64\n");
|
||||
printf("#define DTB_SIZE 4096\n");
|
||||
printf("#define L2_ASSOCIATIVE 8\n");
|
||||
}else{
|
||||
printf("#define SICORTEX\n");
|
||||
printf("#define L1_DATA_SIZE 32768\n");
|
||||
@@ -232,6 +245,8 @@ void get_libname(void){
|
||||
printf("i6400\n");
|
||||
}else if(detect()==CPU_P6600) {
|
||||
printf("p6600\n");
|
||||
}else if(detect()==CPU_I6500) {
|
||||
printf("i6500\n");
|
||||
}else{
|
||||
printf("mips64\n");
|
||||
}
|
||||
|
||||
@@ -49,6 +49,7 @@ void get_subdirname(void){
|
||||
}
|
||||
|
||||
void get_cpuconfig(void){
|
||||
printf("#define SPARC\n");
|
||||
printf("#define V9\n");
|
||||
printf("#define DTB_DEFAULT_ENTRIES 32\n");
|
||||
}
|
||||
@@ -56,3 +57,8 @@ void get_cpuconfig(void){
|
||||
void get_libname(void){
|
||||
printf("v9\n");
|
||||
}
|
||||
|
||||
char *get_corename(void){
|
||||
return "sparc";
|
||||
}
|
||||
|
||||
|
||||
+151
-25
@@ -71,12 +71,23 @@ void cpuid(int op, int *eax, int *ebx, int *ecx, int *edx)
|
||||
*edx = cpuInfo[3];
|
||||
}
|
||||
|
||||
void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx, int *edx)
|
||||
{
|
||||
int cpuInfo[4] = {-1};
|
||||
__cpuidex(cpuInfo, op, count);
|
||||
*eax = cpuInfo[0];
|
||||
*ebx = cpuInfo[1];
|
||||
*ecx = cpuInfo[2];
|
||||
*edx = cpuInfo[3];
|
||||
}
|
||||
|
||||
#else
|
||||
|
||||
#ifndef CPUIDEMU
|
||||
|
||||
#if defined(__APPLE__) && defined(__i386__)
|
||||
void cpuid(int op, int *eax, int *ebx, int *ecx, int *edx);
|
||||
void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx, int *edx);
|
||||
#else
|
||||
static C_INLINE void cpuid(int op, int *eax, int *ebx, int *ecx, int *edx){
|
||||
#if defined(__i386__) && defined(__PIC__)
|
||||
@@ -90,6 +101,19 @@ static C_INLINE void cpuid(int op, int *eax, int *ebx, int *ecx, int *edx){
|
||||
("cpuid": "=a" (*eax), "=b" (*ebx), "=c" (*ecx), "=d" (*edx) : "a" (op) : "cc");
|
||||
#endif
|
||||
}
|
||||
|
||||
static C_INLINE void cpuid_count(int op, int count ,int *eax, int *ebx, int *ecx, int *edx){
|
||||
#if defined(__i386__) && defined(__PIC__)
|
||||
__asm__ __volatile__
|
||||
("mov %%ebx, %%edi;"
|
||||
"cpuid;"
|
||||
"xchgl %%ebx, %%edi;"
|
||||
: "=a" (*eax), "=D" (*ebx), "=c" (*ecx), "=d" (*edx) : "0" (op), "2" (count) : "cc");
|
||||
#else
|
||||
__asm__ __volatile__
|
||||
("cpuid": "=a" (*eax), "=b" (*ebx), "=c" (*ecx), "=d" (*edx) : "0" (op), "2" (count) : "cc");
|
||||
#endif
|
||||
}
|
||||
#endif
|
||||
|
||||
#else
|
||||
@@ -133,6 +157,10 @@ void cpuid(unsigned int op, unsigned int *eax, unsigned int *ebx, unsigned int *
|
||||
*edx = idlist[current].d;
|
||||
}
|
||||
|
||||
void cpuid_count (unsigned int op, unsigned int count, unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx) {
|
||||
return cpuid (op, eax, ebx, ecx, edx);
|
||||
}
|
||||
|
||||
#endif
|
||||
|
||||
#endif // _MSC_VER
|
||||
@@ -312,9 +340,9 @@ int get_cacheinfo(int type, cache_info_t *cacheinfo){
|
||||
cpuid(0, &cpuid_level, &ebx, &ecx, &edx);
|
||||
|
||||
if (cpuid_level > 1) {
|
||||
|
||||
int numcalls =0 ;
|
||||
cpuid(2, &eax, &ebx, &ecx, &edx);
|
||||
|
||||
numcalls = BITMASK(eax, 0, 0xff); //FIXME some systems may require repeated calls to read all entries
|
||||
info[ 0] = BITMASK(eax, 8, 0xff);
|
||||
info[ 1] = BITMASK(eax, 16, 0xff);
|
||||
info[ 2] = BITMASK(eax, 24, 0xff);
|
||||
@@ -335,7 +363,6 @@ int get_cacheinfo(int type, cache_info_t *cacheinfo){
|
||||
info[14] = BITMASK(edx, 24, 0xff);
|
||||
|
||||
for (i = 0; i < 15; i++){
|
||||
|
||||
switch (info[i]){
|
||||
|
||||
/* This table is from http://www.sandpile.org/ia32/cpuid.htm */
|
||||
@@ -637,12 +664,13 @@ int get_cacheinfo(int type, cache_info_t *cacheinfo){
|
||||
LD1.linesize = 64;
|
||||
break;
|
||||
case 0x63 :
|
||||
DTB.size = 2048;
|
||||
DTB.associative = 4;
|
||||
DTB.linesize = 32;
|
||||
LDTB.size = 4096;
|
||||
LDTB.associative= 4;
|
||||
LDTB.linesize = 32;
|
||||
DTB.size = 2048;
|
||||
DTB.associative = 4;
|
||||
DTB.linesize = 32;
|
||||
LDTB.size = 4096;
|
||||
LDTB.associative= 4;
|
||||
LDTB.linesize = 32;
|
||||
break;
|
||||
case 0x66 :
|
||||
LD1.size = 8;
|
||||
LD1.associative = 4;
|
||||
@@ -675,12 +703,13 @@ int get_cacheinfo(int type, cache_info_t *cacheinfo){
|
||||
LC1.associative = 8;
|
||||
break;
|
||||
case 0x76 :
|
||||
ITB.size = 2048;
|
||||
ITB.associative = 0;
|
||||
ITB.linesize = 8;
|
||||
LITB.size = 4096;
|
||||
LITB.associative= 0;
|
||||
LITB.linesize = 8;
|
||||
ITB.size = 2048;
|
||||
ITB.associative = 0;
|
||||
ITB.linesize = 8;
|
||||
LITB.size = 4096;
|
||||
LITB.associative= 0;
|
||||
LITB.linesize = 8;
|
||||
break;
|
||||
case 0x77 :
|
||||
LC1.size = 16;
|
||||
LC1.associative = 4;
|
||||
@@ -891,6 +920,67 @@ int get_cacheinfo(int type, cache_info_t *cacheinfo){
|
||||
}
|
||||
|
||||
if (get_vendor() == VENDOR_INTEL) {
|
||||
if(LD1.size<=0 || LC1.size<=0){
|
||||
//If we didn't detect L1 correctly before,
|
||||
int count;
|
||||
for (count=0;count <4;count++) {
|
||||
cpuid_count(4, count, &eax, &ebx, &ecx, &edx);
|
||||
switch (eax &0x1f) {
|
||||
case 0:
|
||||
continue;
|
||||
case 1:
|
||||
case 3:
|
||||
{
|
||||
switch ((eax >>5) &0x07)
|
||||
{
|
||||
case 1:
|
||||
{
|
||||
// fprintf(stderr,"L1 data cache...\n");
|
||||
int sets = ecx+1;
|
||||
int lines = (ebx & 0x0fff) +1;
|
||||
ebx>>=12;
|
||||
int part = (ebx&0x03ff)+1;
|
||||
ebx >>=10;
|
||||
int assoc = (ebx&0x03ff)+1;
|
||||
LD1.size = (assoc*part*lines*sets)/1024;
|
||||
LD1.associative = assoc;
|
||||
LD1.linesize= lines;
|
||||
break;
|
||||
}
|
||||
default:
|
||||
break;
|
||||
}
|
||||
break;
|
||||
}
|
||||
case 2:
|
||||
{
|
||||
switch ((eax >>5) &0x07)
|
||||
{
|
||||
case 1:
|
||||
{
|
||||
// fprintf(stderr,"L1 instruction cache...\n");
|
||||
int sets = ecx+1;
|
||||
int lines = (ebx & 0x0fff) +1;
|
||||
ebx>>=12;
|
||||
int part = (ebx&0x03ff)+1;
|
||||
ebx >>=10;
|
||||
int assoc = (ebx&0x03ff)+1;
|
||||
LC1.size = (assoc*part*lines*sets)/1024;
|
||||
LC1.associative = assoc;
|
||||
LC1.linesize= lines;
|
||||
break;
|
||||
}
|
||||
default:
|
||||
break;
|
||||
}
|
||||
break;
|
||||
|
||||
}
|
||||
default:
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
cpuid(0x80000000, &cpuid_level, &ebx, &ecx, &edx);
|
||||
if (cpuid_level >= 0x80000006) {
|
||||
if(L2.size<=0){
|
||||
@@ -1279,8 +1369,11 @@ int get_cpuname(void){
|
||||
return CPUTYPE_OPTERON;
|
||||
case 1:
|
||||
case 3:
|
||||
case 7:
|
||||
case 10:
|
||||
return CPUTYPE_BARCELONA;
|
||||
case 5:
|
||||
return CPUTYPE_BOBCAT;
|
||||
case 6:
|
||||
switch (model) {
|
||||
case 1:
|
||||
@@ -1295,12 +1388,13 @@ int get_cpuname(void){
|
||||
return CPUTYPE_PILEDRIVER;
|
||||
else
|
||||
return CPUTYPE_BARCELONA; //OS don't support AVX.
|
||||
case 5: // New EXCAVATOR CPUS
|
||||
if(support_avx())
|
||||
case 5: // New EXCAVATOR CPUS
|
||||
if(support_avx())
|
||||
return CPUTYPE_EXCAVATOR;
|
||||
else
|
||||
return CPUTYPE_BARCELONA; //OS don't support AVX.
|
||||
case 0:
|
||||
case 8:
|
||||
switch(exmodel){
|
||||
case 1: //AMD Trinity
|
||||
if(support_avx())
|
||||
@@ -1322,8 +1416,19 @@ int get_cpuname(void){
|
||||
break;
|
||||
}
|
||||
break;
|
||||
case 5:
|
||||
return CPUTYPE_BOBCAT;
|
||||
case 8:
|
||||
switch (model) {
|
||||
case 1:
|
||||
// AMD Ryzen
|
||||
if(support_avx())
|
||||
#ifndef NO_AVX2
|
||||
return CPUTYPE_ZEN;
|
||||
#else
|
||||
return CPUTYPE_SANDYBRIDGE; // Zen is closer in architecture to Sandy Bridge than to Excavator
|
||||
#endif
|
||||
else
|
||||
return CPUTYPE_BARCELONA;
|
||||
}
|
||||
}
|
||||
break;
|
||||
}
|
||||
@@ -1450,6 +1555,7 @@ static char *cpuname[] = {
|
||||
"HASWELL",
|
||||
"STEAMROLLER",
|
||||
"EXCAVATOR",
|
||||
"ZEN",
|
||||
};
|
||||
|
||||
static char *lowercpuname[] = {
|
||||
@@ -1503,6 +1609,7 @@ static char *lowercpuname[] = {
|
||||
"haswell",
|
||||
"steamroller",
|
||||
"excavator",
|
||||
"zen",
|
||||
};
|
||||
|
||||
static char *corename[] = {
|
||||
@@ -1533,6 +1640,7 @@ static char *corename[] = {
|
||||
"HASWELL",
|
||||
"STEAMROLLER",
|
||||
"EXCAVATOR",
|
||||
"ZEN",
|
||||
};
|
||||
|
||||
static char *corename_lower[] = {
|
||||
@@ -1563,6 +1671,7 @@ static char *corename_lower[] = {
|
||||
"haswell",
|
||||
"steamroller",
|
||||
"excavator",
|
||||
"zen",
|
||||
};
|
||||
|
||||
|
||||
@@ -1668,6 +1777,8 @@ int get_coretype(void){
|
||||
break;
|
||||
case 3:
|
||||
switch (model) {
|
||||
case 7:
|
||||
return CORE_ATOM;
|
||||
case 10:
|
||||
case 14:
|
||||
if(support_avx())
|
||||
@@ -1776,15 +1887,16 @@ int get_coretype(void){
|
||||
break;
|
||||
case 9:
|
||||
case 8:
|
||||
if (model == 14) // Kaby Lake
|
||||
if (model == 14) { // Kaby Lake
|
||||
if(support_avx())
|
||||
#ifndef NO_AVX2
|
||||
return CORE_HASWELL;
|
||||
return CORE_HASWELL;
|
||||
#else
|
||||
return CORE_SANDYBRIDGE;
|
||||
return CORE_SANDYBRIDGE;
|
||||
#endif
|
||||
else
|
||||
return CORE_NEHALEM;
|
||||
}
|
||||
}
|
||||
break;
|
||||
|
||||
@@ -1820,6 +1932,7 @@ int get_coretype(void){
|
||||
else
|
||||
return CORE_BARCELONA; //OS don't support AVX.
|
||||
case 0:
|
||||
case 8:
|
||||
switch(exmodel){
|
||||
case 1: //AMD Trinity
|
||||
if(support_avx())
|
||||
@@ -1841,9 +1954,22 @@ int get_coretype(void){
|
||||
}
|
||||
break;
|
||||
}
|
||||
|
||||
|
||||
}else return CORE_BARCELONA;
|
||||
} else if (exfamily == 8) {
|
||||
switch (model) {
|
||||
case 1:
|
||||
// AMD Ryzen
|
||||
if(support_avx())
|
||||
#ifndef NO_AVX2
|
||||
return CORE_ZEN;
|
||||
#else
|
||||
return CORE_SANDYBRIDGE; // Zen is closer in architecture to Sandy Bridge than to Excavator
|
||||
#endif
|
||||
else
|
||||
return CORE_BARCELONA;
|
||||
}
|
||||
} else {
|
||||
return CORE_BARCELONA;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
+21
-3
@@ -42,9 +42,27 @@ static char *cpuname_lower[] = {
|
||||
|
||||
int detect(void)
|
||||
{
|
||||
// return CPU_GENERIC;
|
||||
return CPU_Z13;
|
||||
|
||||
FILE *infile;
|
||||
char buffer[512], *p;
|
||||
|
||||
p = (char *)NULL;
|
||||
infile = fopen("/proc/sysinfo", "r");
|
||||
while (fgets(buffer, sizeof(buffer), infile)){
|
||||
if (!strncmp("Type", buffer, 4)){
|
||||
p = strchr(buffer, ':') + 2;
|
||||
#if 0
|
||||
fprintf(stderr, "%s\n", p);
|
||||
#endif
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
fclose(infile);
|
||||
|
||||
if (strstr(p, "2964")) return CPU_Z13;
|
||||
if (strstr(p, "2965")) return CPU_Z13;
|
||||
|
||||
return CPU_GENERIC;
|
||||
}
|
||||
|
||||
void get_libname(void)
|
||||
|
||||
@@ -60,6 +60,14 @@ OS_FREEBSD
|
||||
OS_NETBSD
|
||||
#endif
|
||||
|
||||
#if defined(__OpenBSD__)
|
||||
OS_OPENBSD
|
||||
#endif
|
||||
|
||||
#if defined(__DragonFly__)
|
||||
OS_DRAGONFLY
|
||||
#endif
|
||||
|
||||
#if defined(__sun)
|
||||
OS_SUNOS
|
||||
#endif
|
||||
|
||||
@@ -1,4 +1,5 @@
|
||||
include_directories(${PROJECT_SOURCE_DIR})
|
||||
include_directories(${PROJECT_BINARY_DIR})
|
||||
|
||||
enable_language(Fortran)
|
||||
|
||||
@@ -15,7 +16,7 @@ foreach(float_type ${FLOAT_TYPES})
|
||||
add_executable(x${float_char}cblat1
|
||||
c_${float_char}blat1.f
|
||||
c_${float_char}blas1.c)
|
||||
target_link_libraries(x${float_char}cblat1 ${OpenBLAS_LIBNAME}_static)
|
||||
target_link_libraries(x${float_char}cblat1 ${OpenBLAS_LIBNAME})
|
||||
add_test(NAME "x${float_char}cblat1"
|
||||
COMMAND "${CMAKE_CURRENT_BINARY_DIR}/x${float_char}cblat1")
|
||||
|
||||
@@ -27,7 +28,7 @@ foreach(float_type ${FLOAT_TYPES})
|
||||
auxiliary.c
|
||||
c_xerbla.c
|
||||
constant.c)
|
||||
target_link_libraries(x${float_char}cblat2 ${OpenBLAS_LIBNAME}_static)
|
||||
target_link_libraries(x${float_char}cblat2 ${OpenBLAS_LIBNAME})
|
||||
add_test(NAME "x${float_char}cblat2"
|
||||
COMMAND sh "${CMAKE_CURRENT_BINARY_DIR}/test_cblas_helper.sh" "${CMAKE_CURRENT_BINARY_DIR}/x${float_char}cblat2" "${PROJECT_SOURCE_DIR}/ctest/${float_char}in2")
|
||||
|
||||
@@ -39,7 +40,7 @@ foreach(float_type ${FLOAT_TYPES})
|
||||
auxiliary.c
|
||||
c_xerbla.c
|
||||
constant.c)
|
||||
target_link_libraries(x${float_char}cblat3 ${OpenBLAS_LIBNAME}_static)
|
||||
target_link_libraries(x${float_char}cblat3 ${OpenBLAS_LIBNAME})
|
||||
add_test(NAME "x${float_char}cblat3"
|
||||
COMMAND sh "${CMAKE_CURRENT_BINARY_DIR}/test_cblas_helper.sh" "${CMAKE_CURRENT_BINARY_DIR}/x${float_char}cblat3" "${PROJECT_SOURCE_DIR}/ctest/${float_char}in3")
|
||||
|
||||
|
||||
+3
-3
@@ -9,7 +9,7 @@
|
||||
#include "common.h"
|
||||
#include "cblas_test.h"
|
||||
|
||||
void F77_caxpy(const int *N, const void *alpha, void *X,
|
||||
void F77_caxpy(const int *N, OPENBLAS_CONST void *alpha, void *X,
|
||||
const int *incX, void *Y, const int *incY)
|
||||
{
|
||||
cblas_caxpy(*N, alpha, X, *incX, Y, *incY);
|
||||
@@ -58,13 +58,13 @@ void F77_cswap( const int *N, void *X, const int *incX,
|
||||
return;
|
||||
}
|
||||
|
||||
int F77_icamax(const int *N, const void *X, const int *incX)
|
||||
int F77_icamax(const int *N, OPENBLAS_CONST void *X, const int *incX)
|
||||
{
|
||||
if (*N < 1 || *incX < 1) return(0);
|
||||
return (cblas_icamax(*N, X, *incX)+1);
|
||||
}
|
||||
|
||||
float F77_scnrm2(const int *N, const void *X, const int *incX)
|
||||
float F77_scnrm2(const int *N, OPENBLAS_CONST void *X, const int *incX)
|
||||
{
|
||||
return cblas_scnrm2(*N, X, *incX);
|
||||
}
|
||||
|
||||
+3
-3
@@ -9,9 +9,9 @@
|
||||
#include "cblas_test.h"
|
||||
|
||||
void F77_cgemv(int *order, char *transp, int *m, int *n,
|
||||
const void *alpha,
|
||||
CBLAS_TEST_COMPLEX *a, int *lda, const void *x, int *incx,
|
||||
const void *beta, void *y, int *incy) {
|
||||
OPENBLAS_CONST void *alpha,
|
||||
CBLAS_TEST_COMPLEX *a, int *lda, OPENBLAS_CONST void *x, int *incx,
|
||||
OPENBLAS_CONST void *beta, void *y, int *incy) {
|
||||
|
||||
CBLAS_TEST_COMPLEX *A;
|
||||
int i,j,LDA;
|
||||
|
||||
+3
-3
@@ -349,13 +349,13 @@
|
||||
CALL CCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NKB, KB, NINC, INC,
|
||||
$ NMAX, INCMAX, A, AA, AS, Y, YY, YS, YT, G, Z,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL CCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NKB, KB, NINC, INC,
|
||||
$ NMAX, INCMAX, A, AA, AS, Y, YY, YS, YT, G, Z,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 200
|
||||
* Test CGERC, 12, CGERU, 13.
|
||||
@@ -2660,7 +2660,7 @@
|
||||
50 CONTINUE
|
||||
END IF
|
||||
*
|
||||
60 CONTINUE
|
||||
C 60 CONTINUE
|
||||
LCERES = .TRUE.
|
||||
GO TO 80
|
||||
70 CONTINUE
|
||||
|
||||
+26
-26
@@ -329,13 +329,13 @@
|
||||
CALL CCHK3(SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NMAX, AB,
|
||||
$ AA, AS, AB( 1, NMAX + 1 ), BB, BS, CT, G, C,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL CCHK3(SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NMAX, AB,
|
||||
$ AA, AS, AB( 1, NMAX + 1 ), BB, BS, CT, G, C,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 190
|
||||
* Test CHERK, 06, CSYRK, 07.
|
||||
@@ -357,13 +357,13 @@
|
||||
CALL CCHK5(SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NBET, BET,
|
||||
$ NMAX, AB, AA, AS, BB, BS, C, CC, CS, CT, G, W,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL CCHK5(SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NBET, BET,
|
||||
$ NMAX, AB, AA, AS, BB, BS, C, CC, CS, CT, G, W,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 190
|
||||
*
|
||||
@@ -707,9 +707,9 @@
|
||||
9998 FORMAT(' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ', A12,'(''', A1, ''',''', A1, ''',',
|
||||
$ 3( I3, ',' ), '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3,
|
||||
$ ',(', F4.1, ',', F4.1, '), C,', I3, ').' )
|
||||
C 9995 FORMAT( 1X, I6, ': ', A12,'(''', A1, ''',''', A1, ''',',
|
||||
C $ 3( I3, ',' ), '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3,
|
||||
C $ ',(', F4.1, ',', F4.1, '), C,', I3, ').' )
|
||||
9994 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1033,9 +1033,9 @@
|
||||
9998 FORMAT(' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',(', F4.1,
|
||||
$ ',', F4.1, '), C,', I3, ') .' )
|
||||
C 9995 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',(', F4.1,
|
||||
C $ ',', F4.1, '), C,', I3, ') .' )
|
||||
9994 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1385,9 +1385,9 @@
|
||||
9998 FORMAT(' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT(' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT(1X, I6, ': ', A12,'(', 4( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ') ',
|
||||
$ ' .' )
|
||||
C 9995 FORMAT(1X, I6, ': ', A12,'(', 4( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ') ',
|
||||
C $ ' .' )
|
||||
9994 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1768,12 +1768,12 @@
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
$ ' .' )
|
||||
9993 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, ') , A,', I3, ',(', F4.1, ',', F4.1,
|
||||
$ '), C,', I3, ') .' )
|
||||
C 9994 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
C $ ' .' )
|
||||
C 9993 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, ') , A,', I3, ',(', F4.1, ',', F4.1,
|
||||
C $ '), C,', I3, ') .' )
|
||||
9992 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -2221,12 +2221,12 @@
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',', F4.1,
|
||||
$ ', C,', I3, ') .' )
|
||||
9993 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',(', F4.1,
|
||||
$ ',', F4.1, '), C,', I3, ') .' )
|
||||
C 9994 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',', F4.1,
|
||||
C $ ', C,', I3, ') .' )
|
||||
C 9993 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',(', F4.1,
|
||||
C $ ',', F4.1, '), C,', I3, ') .' )
|
||||
9992 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -2702,7 +2702,7 @@
|
||||
50 CONTINUE
|
||||
END IF
|
||||
*
|
||||
60 CONTINUE
|
||||
C 60 CONTINUE
|
||||
LCERES = .TRUE.
|
||||
GO TO 80
|
||||
70 CONTINUE
|
||||
|
||||
+6
-6
@@ -14,7 +14,7 @@ double F77_dasum(const int *N, double *X, const int *incX)
|
||||
return cblas_dasum(*N, X, *incX);
|
||||
}
|
||||
|
||||
void F77_daxpy(const int *N, const double *alpha, const double *X,
|
||||
void F77_daxpy(const int *N, const double *alpha, OPENBLAS_CONST double *X,
|
||||
const int *incX, double *Y, const int *incY)
|
||||
{
|
||||
cblas_daxpy(*N, *alpha, X, *incX, Y, *incY);
|
||||
@@ -28,13 +28,13 @@ void F77_dcopy(const int *N, double *X, const int *incX,
|
||||
return;
|
||||
}
|
||||
|
||||
double F77_ddot(const int *N, const double *X, const int *incX,
|
||||
const double *Y, const int *incY)
|
||||
double F77_ddot(const int *N, OPENBLAS_CONST double *X, const int *incX,
|
||||
OPENBLAS_CONST double *Y, const int *incY)
|
||||
{
|
||||
return cblas_ddot(*N, X, *incX, Y, *incY);
|
||||
}
|
||||
|
||||
double F77_dnrm2(const int *N, const double *X, const int *incX)
|
||||
double F77_dnrm2(const int *N, OPENBLAS_CONST double *X, const int *incX)
|
||||
{
|
||||
return cblas_dnrm2(*N, X, *incX);
|
||||
}
|
||||
@@ -72,12 +72,12 @@ double F77_dzasum(const int *N, void *X, const int *incX)
|
||||
return cblas_dzasum(*N, X, *incX);
|
||||
}
|
||||
|
||||
double F77_dznrm2(const int *N, const void *X, const int *incX)
|
||||
double F77_dznrm2(const int *N, OPENBLAS_CONST void *X, const int *incX)
|
||||
{
|
||||
return cblas_dznrm2(*N, X, *incX);
|
||||
}
|
||||
|
||||
int F77_idamax(const int *N, const double *X, const int *incX)
|
||||
int F77_idamax(const int *N, OPENBLAS_CONST double *X, const int *incX)
|
||||
{
|
||||
if (*N < 1 || *incX < 1) return(0);
|
||||
return (cblas_idamax(*N, X, *incX)+1);
|
||||
|
||||
+2
-2
@@ -211,11 +211,11 @@
|
||||
IF (ICASE.EQ.7) THEN
|
||||
* .. DNRM2TEST ..
|
||||
STEMP(1) = DTRUE1(NP1)
|
||||
CALL STEST1(DNRM2TEST(N,SX,INCX),STEMP,STEMP,SFAC)
|
||||
CALL STEST1(DNRM2TEST(N,SX,INCX),STEMP(1),STEMP,SFAC)
|
||||
ELSE IF (ICASE.EQ.8) THEN
|
||||
* .. DASUMTEST ..
|
||||
STEMP(1) = DTRUE3(NP1)
|
||||
CALL STEST1(DASUMTEST(N,SX,INCX),STEMP,STEMP,SFAC)
|
||||
CALL STEST1(DASUMTEST(N,SX,INCX),STEMP(1),STEMP,SFAC)
|
||||
ELSE IF (ICASE.EQ.9) THEN
|
||||
* .. DSCALTEST ..
|
||||
CALL DSCALTEST(N,SA((INCX-1)*5+NP1),SX,INCX)
|
||||
|
||||
+28
-28
@@ -345,13 +345,13 @@
|
||||
CALL DCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NKB, KB, NINC, INC,
|
||||
$ NMAX, INCMAX, A, AA, AS, Y, YY, YS, YT, G, Z,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL DCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NKB, KB, NINC, INC,
|
||||
$ NMAX, INCMAX, A, AA, AS, Y, YY, YS, YT, G, Z,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 200
|
||||
* Test DGER, 12.
|
||||
@@ -797,9 +797,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ',A12, '(', A14, ',', 4( I3, ',' ), F4.1,
|
||||
$ ', A,', I3, ',',/ 10x,'X,', I2, ',', F4.1, ', Y,',
|
||||
@@ -1004,7 +1004,7 @@
|
||||
$ REWIND NTRA
|
||||
CALL CDSBMV( IORDER, UPLO, N, K, ALPHA,
|
||||
$ AA, LDA, XX, INCX, BETA, YY,
|
||||
$ INCY )
|
||||
$ INCY )
|
||||
ELSE IF( PACKED )THEN
|
||||
IF( TRACE )
|
||||
$ WRITE( NTRA, FMT = 9995 )NC, SNAME,
|
||||
@@ -1156,9 +1156,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ',A12, '(', A14, ',', I3, ',', F4.1, ', AP',
|
||||
$ ', X,', I2, ',', F4.1, ', Y,', I2, ') .' )
|
||||
@@ -1191,7 +1191,7 @@
|
||||
* .. Scalar Arguments ..
|
||||
DOUBLE PRECISION EPS, THRESH
|
||||
INTEGER INCMAX, NIDIM, NINC, NKB, NMAX, NOUT, NTRA,
|
||||
$ IORDER
|
||||
$ IORDER
|
||||
LOGICAL FATAL, REWI, TRACE
|
||||
CHARACTER*12 SNAME
|
||||
* .. Array Arguments ..
|
||||
@@ -1216,7 +1216,7 @@
|
||||
EXTERNAL LDE, LDERES
|
||||
* .. External Subroutines ..
|
||||
EXTERNAL DMAKE, DMVCH, CDTBMV, CDTBSV, CDTPMV,
|
||||
$ CDTPSV, CDTRMV, CDTRSV
|
||||
$ CDTPSV, CDTRMV, CDTRSV
|
||||
* .. Intrinsic Functions ..
|
||||
INTRINSIC ABS, MAX
|
||||
* .. Scalars in Common ..
|
||||
@@ -1544,9 +1544,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ',A12, '(', 3( A14,',' ),/ 10x, I3, ', AP, ',
|
||||
$ 'X,', I2, ') .' )
|
||||
@@ -1579,7 +1579,7 @@
|
||||
* .. Scalar Arguments ..
|
||||
DOUBLE PRECISION EPS, THRESH
|
||||
INTEGER INCMAX, NALF, NIDIM, NINC, NMAX, NOUT, NTRA,
|
||||
$ IORDER
|
||||
$ IORDER
|
||||
LOGICAL FATAL, REWI, TRACE
|
||||
CHARACTER*12 SNAME
|
||||
* .. Array Arguments ..
|
||||
@@ -1819,9 +1819,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ',A12, '(', 2( I3, ',' ), F4.1, ', X,', I2,
|
||||
@@ -1851,7 +1851,7 @@
|
||||
* .. Scalar Arguments ..
|
||||
DOUBLE PRECISION EPS, THRESH
|
||||
INTEGER INCMAX, NALF, NIDIM, NINC, NMAX, NOUT, NTRA,
|
||||
$ IORDER
|
||||
$ IORDER
|
||||
LOGICAL FATAL, REWI, TRACE
|
||||
CHARACTER*12 SNAME
|
||||
* .. Array Arguments ..
|
||||
@@ -1973,7 +1973,7 @@
|
||||
IF( REWI )
|
||||
$ REWIND NTRA
|
||||
CALL CDSYR( IORDER, UPLO, N, ALPHA, XX, INCX,
|
||||
$ AA, LDA )
|
||||
$ AA, LDA )
|
||||
ELSE IF( PACKED )THEN
|
||||
IF( TRACE )
|
||||
$ WRITE( NTRA, FMT = 9994 )NC, SNAME, CUPLO, N,
|
||||
@@ -2113,9 +2113,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ',A12, '(', A14, ',', I3, ',', F4.1, ', X,',
|
||||
@@ -2147,7 +2147,7 @@
|
||||
* .. Scalar Arguments ..
|
||||
DOUBLE PRECISION EPS, THRESH
|
||||
INTEGER INCMAX, NALF, NIDIM, NINC, NMAX, NOUT, NTRA,
|
||||
$ IORDER
|
||||
$ IORDER
|
||||
LOGICAL FATAL, REWI, TRACE
|
||||
CHARACTER*12 SNAME
|
||||
* .. Array Arguments ..
|
||||
@@ -2445,9 +2445,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ',A12, '(', A14, ',', I3, ',', F4.1, ', X,',
|
||||
@@ -2833,7 +2833,7 @@
|
||||
50 CONTINUE
|
||||
END IF
|
||||
*
|
||||
60 CONTINUE
|
||||
C 60 CONTINUE
|
||||
LDERES = .TRUE.
|
||||
GO TO 80
|
||||
70 CONTINUE
|
||||
|
||||
+25
-25
@@ -56,7 +56,7 @@
|
||||
* .. Local Scalars ..
|
||||
DOUBLE PRECISION EPS, ERR, THRESH
|
||||
INTEGER I, ISNUM, J, N, NALF, NBET, NIDIM, NTRA,
|
||||
$ LAYOUT
|
||||
$ LAYOUT
|
||||
LOGICAL FATAL, LTESTT, REWI, SAME, SFATAL, TRACE,
|
||||
$ TSTERR, CORDER, RORDER
|
||||
CHARACTER*1 TRANSA, TRANSB
|
||||
@@ -78,7 +78,7 @@
|
||||
EXTERNAL DDIFF, LDE
|
||||
* .. External Subroutines ..
|
||||
EXTERNAL DCHK1, DCHK2, DCHK3, DCHK4, DCHK5, CD3CHKE,
|
||||
$ DMMCH
|
||||
$ DMMCH
|
||||
* .. Intrinsic Functions ..
|
||||
INTRINSIC MAX, MIN
|
||||
* .. Scalars in Common ..
|
||||
@@ -323,13 +323,13 @@
|
||||
CALL DCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NMAX, AB,
|
||||
$ AA, AS, AB( 1, NMAX + 1 ), BB, BS, CT, G, C,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL DCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NMAX, AB,
|
||||
$ AA, AS, AB( 1, NMAX + 1 ), BB, BS, CT, G, C,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 190
|
||||
* Test DSYRK, 05.
|
||||
@@ -351,13 +351,13 @@
|
||||
CALL DCHK5( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NBET, BET,
|
||||
$ NMAX, AB, AA, AS, BB, BS, C, CC, CS, CT, G, W,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL DCHK5( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NBET, BET,
|
||||
$ NMAX, AB, AA, AS, BB, BS, C, CC, CS, CT, G, W,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 190
|
||||
*
|
||||
@@ -588,7 +588,7 @@
|
||||
$ REWIND NTRA
|
||||
CALL CDGEMM( IORDER, TRANSA, TRANSB, M, N,
|
||||
$ K, ALPHA, AA, LDA, BB, LDB,
|
||||
$ BETA, CC, LDC )
|
||||
$ BETA, CC, LDC )
|
||||
*
|
||||
* Check if error-exit was taken incorrectly.
|
||||
*
|
||||
@@ -694,9 +694,9 @@
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ', A12,'(''', A1, ''',''', A1, ''',',
|
||||
$ 3( I3, ',' ), F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', ',
|
||||
$ 'C,', I3, ').' )
|
||||
C 9995 FORMAT( 1X, I6, ': ', A12,'(''', A1, ''',''', A1, ''',',
|
||||
C $ 3( I3, ',' ), F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', ',
|
||||
C $ 'C,', I3, ').' )
|
||||
9994 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1007,9 +1007,9 @@
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
$ ' .' )
|
||||
C 9995 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
C $ ' .' )
|
||||
9994 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1201,7 +1201,7 @@
|
||||
$ REWIND NTRA
|
||||
CALL CDTRMM( IORDER, SIDE, UPLO, TRANSA,
|
||||
$ DIAG, M, N, ALPHA, AA, LDA,
|
||||
$ BB, LDB )
|
||||
$ BB, LDB )
|
||||
ELSE IF( SNAME( 10: 11 ).EQ.'sm' )THEN
|
||||
IF( TRACE )
|
||||
$ CALL DPRCN3( NTRA, NC, SNAME, IORDER,
|
||||
@@ -1211,7 +1211,7 @@
|
||||
$ REWIND NTRA
|
||||
CALL CDTRSM( IORDER, SIDE, UPLO, TRANSA,
|
||||
$ DIAG, M, N, ALPHA, AA, LDA,
|
||||
$ BB, LDB )
|
||||
$ BB, LDB )
|
||||
END IF
|
||||
*
|
||||
* Check if error-exit was taken incorrectly.
|
||||
@@ -1355,8 +1355,8 @@
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ', A12,'(', 4( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ', B,', I3, ') .' )
|
||||
C 9995 FORMAT( 1X, I6, ': ', A12,'(', 4( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ', B,', I3, ') .' )
|
||||
9994 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1681,8 +1681,8 @@
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ',', F4.1, ', C,', I3, ') .' )
|
||||
C 9994 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ',', F4.1, ', C,', I3, ') .' )
|
||||
9993 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1726,7 +1726,7 @@
|
||||
SUBROUTINE DCHK5( SNAME, EPS, THRESH, NOUT, NTRA, TRACE, REWI,
|
||||
$ FATAL, NIDIM, IDIM, NALF, ALF, NBET, BET, NMAX,
|
||||
$ AB, AA, AS, BB, BS, C, CC, CS, CT, G, W,
|
||||
$ IORDER )
|
||||
$ IORDER )
|
||||
*
|
||||
* Tests DSYR2K.
|
||||
*
|
||||
@@ -1888,7 +1888,7 @@
|
||||
$ REWIND NTRA
|
||||
CALL CDSYR2K( IORDER, UPLO, TRANS, N, K,
|
||||
$ ALPHA, AA, LDA, BB, LDB, BETA,
|
||||
$ CC, LDC )
|
||||
$ CC, LDC )
|
||||
*
|
||||
* Check if error-exit was taken incorrectly.
|
||||
*
|
||||
@@ -2037,9 +2037,9 @@
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
$ ' .' )
|
||||
C 9994 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
C $ ' .' )
|
||||
9993 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -2399,7 +2399,7 @@
|
||||
50 CONTINUE
|
||||
END IF
|
||||
*
|
||||
60 CONTINUE
|
||||
C 60 CONTINUE
|
||||
LDERES = .TRUE.
|
||||
GO TO 80
|
||||
70 CONTINUE
|
||||
|
||||
+7
-7
@@ -14,7 +14,7 @@ float F77_sasum(blasint *N, float *X, blasint *incX)
|
||||
return cblas_sasum(*N, X, *incX);
|
||||
}
|
||||
|
||||
void F77_saxpy(blasint *N, const float *alpha, const float *X,
|
||||
void F77_saxpy(blasint *N, const float *alpha, OPENBLAS_CONST float *X,
|
||||
blasint *incX, float *Y, blasint *incY)
|
||||
{
|
||||
cblas_saxpy(*N, *alpha, X, *incX, Y, *incY);
|
||||
@@ -26,25 +26,25 @@ float F77_scasum(blasint *N, float *X, blasint *incX)
|
||||
return cblas_scasum(*N, X, *incX);
|
||||
}
|
||||
|
||||
float F77_scnrm2(blasint *N, const float *X, blasint *incX)
|
||||
float F77_scnrm2(blasint *N, OPENBLAS_CONST float *X, blasint *incX)
|
||||
{
|
||||
return cblas_scnrm2(*N, X, *incX);
|
||||
}
|
||||
|
||||
void F77_scopy(blasint *N, const float *X, blasint *incX,
|
||||
void F77_scopy(blasint *N, OPENBLAS_CONST float *X, blasint *incX,
|
||||
float *Y, blasint *incY)
|
||||
{
|
||||
cblas_scopy(*N, X, *incX, Y, *incY);
|
||||
return;
|
||||
}
|
||||
|
||||
float F77_sdot(blasint *N, const float *X, blasint *incX,
|
||||
const float *Y, blasint *incY)
|
||||
float F77_sdot(blasint *N, OPENBLAS_CONST float *X, blasint *incX,
|
||||
OPENBLAS_CONST float *Y, blasint *incY)
|
||||
{
|
||||
return cblas_sdot(*N, X, *incX, Y, *incY);
|
||||
}
|
||||
|
||||
float F77_snrm2(blasint *N, const float *X, blasint *incX)
|
||||
float F77_snrm2(blasint *N, OPENBLAS_CONST float *X, blasint *incX)
|
||||
{
|
||||
return cblas_snrm2(*N, X, *incX);
|
||||
}
|
||||
@@ -76,7 +76,7 @@ void F77_sswap( blasint *N, float *X, blasint *incX,
|
||||
return;
|
||||
}
|
||||
|
||||
int F77_isamax(blasint *N, const float *X, blasint *incX)
|
||||
int F77_isamax(blasint *N, OPENBLAS_CONST float *X, blasint *incX)
|
||||
{
|
||||
if (*N < 1 || *incX < 1) return(0);
|
||||
return (cblas_isamax(*N, X, *incX)+1);
|
||||
|
||||
+2
-2
@@ -211,11 +211,11 @@
|
||||
IF (ICASE.EQ.7) THEN
|
||||
* .. SNRM2TEST ..
|
||||
STEMP(1) = DTRUE1(NP1)
|
||||
CALL STEST1(SNRM2TEST(N,SX,INCX),STEMP,STEMP,SFAC)
|
||||
CALL STEST1(SNRM2TEST(N,SX,INCX),STEMP(1),STEMP,SFAC)
|
||||
ELSE IF (ICASE.EQ.8) THEN
|
||||
* .. SASUMTEST ..
|
||||
STEMP(1) = DTRUE3(NP1)
|
||||
CALL STEST1(SASUMTEST(N,SX,INCX),STEMP,STEMP,SFAC)
|
||||
CALL STEST1(SASUMTEST(N,SX,INCX),STEMP(1),STEMP,SFAC)
|
||||
ELSE IF (ICASE.EQ.9) THEN
|
||||
* .. SSCALTEST ..
|
||||
CALL SSCALTEST(N,SA((INCX-1)*5+NP1),SX,INCX)
|
||||
|
||||
+28
-28
@@ -345,13 +345,13 @@
|
||||
CALL SCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NKB, KB, NINC, INC,
|
||||
$ NMAX, INCMAX, A, AA, AS, Y, YY, YS, YT, G, Z,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL SCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NKB, KB, NINC, INC,
|
||||
$ NMAX, INCMAX, A, AA, AS, Y, YY, YS, YT, G, Z,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 200
|
||||
* Test SGER, 12.
|
||||
@@ -797,9 +797,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ',A12, '(', A14, ',', 4( I3, ',' ), F4.1,
|
||||
$ ', A,', I3, ',',/ 10x, 'X,', I2, ',', F4.1, ', Y,',
|
||||
@@ -1004,7 +1004,7 @@
|
||||
$ REWIND NTRA
|
||||
CALL CSSBMV( IORDER, UPLO, N, K, ALPHA,
|
||||
$ AA, LDA, XX, INCX, BETA, YY,
|
||||
$ INCY )
|
||||
$ INCY )
|
||||
ELSE IF( PACKED )THEN
|
||||
IF( TRACE )
|
||||
$ WRITE( NTRA, FMT = 9995 )NC, SNAME,
|
||||
@@ -1156,9 +1156,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ',A12, '(', A14, ',', I3, ',', F4.1, ', AP',
|
||||
$ ', X,', I2, ',', F4.1, ', Y,', I2, ') .' )
|
||||
@@ -1191,7 +1191,7 @@
|
||||
* .. Scalar Arguments ..
|
||||
REAL EPS, THRESH
|
||||
INTEGER INCMAX, NIDIM, NINC, NKB, NMAX, NOUT, NTRA,
|
||||
$ IORDER
|
||||
$ IORDER
|
||||
LOGICAL FATAL, REWI, TRACE
|
||||
CHARACTER*12 SNAME
|
||||
* .. Array Arguments ..
|
||||
@@ -1216,7 +1216,7 @@
|
||||
EXTERNAL LSE, LSERES
|
||||
* .. External Subroutines ..
|
||||
EXTERNAL SMAKE, SMVCH, CSTBMV, CSTBSV, CSTPMV,
|
||||
$ CSTPSV, CSTRMV, CSTRSV
|
||||
$ CSTPSV, CSTRMV, CSTRSV
|
||||
* .. Intrinsic Functions ..
|
||||
INTRINSIC ABS, MAX
|
||||
* .. Scalars in Common ..
|
||||
@@ -1544,9 +1544,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ',A12, '(', 3( A14,',' ),/ 10x, I3, ', AP, ',
|
||||
$ 'X,', I2, ') .' )
|
||||
@@ -1579,7 +1579,7 @@
|
||||
* .. Scalar Arguments ..
|
||||
REAL EPS, THRESH
|
||||
INTEGER INCMAX, NALF, NIDIM, NINC, NMAX, NOUT, NTRA,
|
||||
$ IORDER
|
||||
$ IORDER
|
||||
LOGICAL FATAL, REWI, TRACE
|
||||
CHARACTER*12 SNAME
|
||||
* .. Array Arguments ..
|
||||
@@ -1819,9 +1819,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ',A12, '(', 2( I3, ',' ), F4.1, ', X,', I2,
|
||||
@@ -1851,7 +1851,7 @@
|
||||
* .. Scalar Arguments ..
|
||||
REAL EPS, THRESH
|
||||
INTEGER INCMAX, NALF, NIDIM, NINC, NMAX, NOUT, NTRA,
|
||||
$ IORDER
|
||||
$ IORDER
|
||||
LOGICAL FATAL, REWI, TRACE
|
||||
CHARACTER*12 SNAME
|
||||
* .. Array Arguments ..
|
||||
@@ -1973,7 +1973,7 @@
|
||||
IF( REWI )
|
||||
$ REWIND NTRA
|
||||
CALL CSSYR( IORDER, UPLO, N, ALPHA, XX, INCX,
|
||||
$ AA, LDA )
|
||||
$ AA, LDA )
|
||||
ELSE IF( PACKED )THEN
|
||||
IF( TRACE )
|
||||
$ WRITE( NTRA, FMT = 9994 )NC, SNAME, CUPLO, N,
|
||||
@@ -2113,9 +2113,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ',A12, '(', A14, ',', I3, ',', F4.1, ', X,',
|
||||
@@ -2147,7 +2147,7 @@
|
||||
* .. Scalar Arguments ..
|
||||
REAL EPS, THRESH
|
||||
INTEGER INCMAX, NALF, NIDIM, NINC, NMAX, NOUT, NTRA,
|
||||
$ IORDER
|
||||
$ IORDER
|
||||
LOGICAL FATAL, REWI, TRACE
|
||||
CHARACTER*12 SNAME
|
||||
* .. Array Arguments ..
|
||||
@@ -2445,9 +2445,9 @@
|
||||
$ ' (', I6, ' CALL', 'S)' )
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
$ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
$ ' - SUSPECT *******' )
|
||||
C 9997 FORMAT( ' ',A12, ' COMPLETED THE COMPUTATIONAL TESTS (', I6, ' C',
|
||||
C $ 'ALLS)', /' ******* BUT WITH MAXIMUM TEST RATIO', F8.2,
|
||||
C $ ' - SUSPECT *******' )
|
||||
9996 FORMAT( ' ******* ',A12, ' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ',A12, '(', A14, ',', I3, ',', F4.1, ', X,',
|
||||
@@ -2833,7 +2833,7 @@
|
||||
50 CONTINUE
|
||||
END IF
|
||||
*
|
||||
60 CONTINUE
|
||||
C 60 CONTINUE
|
||||
LSERES = .TRUE.
|
||||
GO TO 80
|
||||
70 CONTINUE
|
||||
|
||||
+14
-14
@@ -694,9 +694,9 @@
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ', A12,'(''', A1, ''',''', A1, ''',',
|
||||
$ 3( I3, ',' ), F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', ',
|
||||
$ 'C,', I3, ').' )
|
||||
C 9995 FORMAT( 1X, I6, ': ', A12,'(''', A1, ''',''', A1, ''',',
|
||||
C $ 3( I3, ',' ), F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', ',
|
||||
C $ 'C,', I3, ').' )
|
||||
9994 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1011,9 +1011,9 @@
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
$ ' .' )
|
||||
C 9995 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
C $ ' .' )
|
||||
9994 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1359,8 +1359,8 @@
|
||||
9998 FORMAT( ' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ', A12,'(', 4( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ', B,', I3, ') .' )
|
||||
C 9995 FORMAT( 1X, I6, ': ', A12,'(', 4( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ', B,', I3, ') .' )
|
||||
9994 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1686,8 +1686,8 @@
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ',', F4.1, ', C,', I3, ') .' )
|
||||
C 9994 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ',', F4.1, ', C,', I3, ') .' )
|
||||
9993 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -2041,9 +2041,9 @@
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
$ ' .' )
|
||||
C 9994 FORMAT( 1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ', B,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
C $ ' .' )
|
||||
9993 FORMAT( ' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -2403,7 +2403,7 @@
|
||||
50 CONTINUE
|
||||
END IF
|
||||
*
|
||||
60 CONTINUE
|
||||
C 60 CONTINUE
|
||||
LSERES = .TRUE.
|
||||
GO TO 80
|
||||
70 CONTINUE
|
||||
|
||||
+1
-1
@@ -131,7 +131,7 @@ void F77_xerbla(char *srname, void *vinfo)
|
||||
int BLASFUNC(xerbla)(char *name, blasint *info, blasint length) {
|
||||
|
||||
F77_xerbla(name, info);
|
||||
|
||||
return 0;
|
||||
};
|
||||
|
||||
|
||||
|
||||
+5
-5
@@ -9,7 +9,7 @@
|
||||
#include "common.h"
|
||||
#include "cblas_test.h"
|
||||
|
||||
void F77_zaxpy(const int *N, const void *alpha, void *X,
|
||||
void F77_zaxpy(const int *N, OPENBLAS_CONST void *alpha, void *X,
|
||||
const int *incX, void *Y, const int *incY)
|
||||
{
|
||||
cblas_zaxpy(*N, alpha, X, *incX, Y, *incY);
|
||||
@@ -23,8 +23,8 @@ void F77_zcopy(const int *N, void *X, const int *incX,
|
||||
return;
|
||||
}
|
||||
|
||||
void F77_zdotc(const int *N, const void *X, const int *incX,
|
||||
const void *Y, const int *incY,void *dotc)
|
||||
void F77_zdotc(const int *N, OPENBLAS_CONST void *X, const int *incX,
|
||||
OPENBLAS_CONST void *Y, const int *incY,void *dotc)
|
||||
{
|
||||
cblas_zdotc_sub(*N, X, *incX, Y, *incY, dotc);
|
||||
return;
|
||||
@@ -58,13 +58,13 @@ void F77_zswap( const int *N, void *X, const int *incX,
|
||||
return;
|
||||
}
|
||||
|
||||
int F77_izamax(const int *N, const void *X, const int *incX)
|
||||
int F77_izamax(const int *N, OPENBLAS_CONST void *X, const int *incX)
|
||||
{
|
||||
if (*N < 1 || *incX < 1) return(0);
|
||||
return(cblas_izamax(*N, X, *incX)+1);
|
||||
}
|
||||
|
||||
double F77_dznrm2(const int *N, const void *X, const int *incX)
|
||||
double F77_dznrm2(const int *N, OPENBLAS_CONST void *X, const int *incX)
|
||||
{
|
||||
return cblas_dznrm2(*N, X, *incX);
|
||||
}
|
||||
|
||||
+3
-3
@@ -9,9 +9,9 @@
|
||||
#include "cblas_test.h"
|
||||
|
||||
void F77_zgemv(int *order, char *transp, int *m, int *n,
|
||||
const void *alpha,
|
||||
CBLAS_TEST_ZOMPLEX *a, int *lda, const void *x, int *incx,
|
||||
const void *beta, void *y, int *incy) {
|
||||
OPENBLAS_CONST void *alpha,
|
||||
CBLAS_TEST_ZOMPLEX *a, int *lda, OPENBLAS_CONST void *x, int *incx,
|
||||
OPENBLAS_CONST void *beta, void *y, int *incy) {
|
||||
|
||||
CBLAS_TEST_ZOMPLEX *A;
|
||||
int i,j,LDA;
|
||||
|
||||
+2
-2
@@ -349,13 +349,13 @@
|
||||
CALL ZCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NKB, KB, NINC, INC,
|
||||
$ NMAX, INCMAX, A, AA, AS, Y, YY, YS, YT, G, Z,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL ZCHK3( SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NKB, KB, NINC, INC,
|
||||
$ NMAX, INCMAX, A, AA, AS, Y, YY, YS, YT, G, Z,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 200
|
||||
* Test ZGERC, 12, ZGERU, 13.
|
||||
|
||||
+26
-26
@@ -330,13 +330,13 @@
|
||||
CALL ZCHK3(SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NMAX, AB,
|
||||
$ AA, AS, AB( 1, NMAX + 1 ), BB, BS, CT, G, C,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL ZCHK3(SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NMAX, AB,
|
||||
$ AA, AS, AB( 1, NMAX + 1 ), BB, BS, CT, G, C,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 190
|
||||
* Test ZHERK, 06, ZSYRK, 07.
|
||||
@@ -358,13 +358,13 @@
|
||||
CALL ZCHK5(SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NBET, BET,
|
||||
$ NMAX, AB, AA, AS, BB, BS, C, CC, CS, CT, G, W,
|
||||
$ 0 )
|
||||
$ 0 )
|
||||
END IF
|
||||
IF (RORDER) THEN
|
||||
CALL ZCHK5(SNAMES( ISNUM ), EPS, THRESH, NOUT, NTRA, TRACE,
|
||||
$ REWI, FATAL, NIDIM, IDIM, NALF, ALF, NBET, BET,
|
||||
$ NMAX, AB, AA, AS, BB, BS, C, CC, CS, CT, G, W,
|
||||
$ 1 )
|
||||
$ 1 )
|
||||
END IF
|
||||
GO TO 190
|
||||
*
|
||||
@@ -708,9 +708,9 @@
|
||||
9998 FORMAT(' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( 1X, I6, ': ', A12,'(''', A1, ''',''', A1, ''',',
|
||||
$ 3( I3, ',' ), '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3,
|
||||
$ ',(', F4.1, ',', F4.1, '), C,', I3, ').' )
|
||||
C 9995 FORMAT( 1X, I6, ': ', A12,'(''', A1, ''',''', A1, ''',',
|
||||
C $ 3( I3, ',' ), '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3,
|
||||
C $ ',(', F4.1, ',', F4.1, '), C,', I3, ').' )
|
||||
9994 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1034,9 +1034,9 @@
|
||||
9998 FORMAT(' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',(', F4.1,
|
||||
$ ',', F4.1, '), C,', I3, ') .' )
|
||||
C 9995 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',(', F4.1,
|
||||
C $ ',', F4.1, '), C,', I3, ') .' )
|
||||
9994 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1386,9 +1386,9 @@
|
||||
9998 FORMAT(' ******* FATAL ERROR - PARAMETER NUMBER ', I2, ' WAS CH',
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT(' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT(1X, I6, ': ', A12,'(', 4( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ') ',
|
||||
$ ' .' )
|
||||
C 9995 FORMAT(1X, I6, ': ', A12,'(', 4( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ') ',
|
||||
C $ ' .' )
|
||||
9994 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -1769,12 +1769,12 @@
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ F4.1, ', A,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
$ ' .' )
|
||||
9993 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, ') , A,', I3, ',(', F4.1, ',', F4.1,
|
||||
$ '), C,', I3, ') .' )
|
||||
C 9994 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ F4.1, ', A,', I3, ',', F4.1, ', C,', I3, ') ',
|
||||
C $ ' .' )
|
||||
C 9993 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, ') , A,', I3, ',(', F4.1, ',', F4.1,
|
||||
C $ '), C,', I3, ') .' )
|
||||
9992 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -2222,12 +2222,12 @@
|
||||
$ 'ANGED INCORRECTLY *******' )
|
||||
9996 FORMAT( ' ******* ', A12,' FAILED ON CALL NUMBER:' )
|
||||
9995 FORMAT( ' THESE ARE THE RESULTS FOR COLUMN ', I3 )
|
||||
9994 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',', F4.1,
|
||||
$ ', C,', I3, ') .' )
|
||||
9993 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
$ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',(', F4.1,
|
||||
$ ',', F4.1, '), C,', I3, ') .' )
|
||||
C 9994 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',', F4.1,
|
||||
C $ ', C,', I3, ') .' )
|
||||
C 9993 FORMAT(1X, I6, ': ', A12,'(', 2( '''', A1, ''',' ), 2( I3, ',' ),
|
||||
C $ '(', F4.1, ',', F4.1, '), A,', I3, ', B,', I3, ',(', F4.1,
|
||||
C $ ',', F4.1, '), C,', I3, ') .' )
|
||||
9992 FORMAT(' ******* FATAL ERROR - ERROR-EXIT TAKEN ON VALID CALL *',
|
||||
$ '******' )
|
||||
*
|
||||
@@ -2706,7 +2706,7 @@
|
||||
50 CONTINUE
|
||||
END IF
|
||||
*
|
||||
60 CONTINUE
|
||||
C 60 CONTINUE
|
||||
LZERES = .TRUE.
|
||||
GO TO 80
|
||||
70 CONTINUE
|
||||
|
||||
@@ -1,5 +1,6 @@
|
||||
|
||||
include_directories(${PROJECT_SOURCE_DIR})
|
||||
include_directories(${PROJECT_BINARY_DIR})
|
||||
|
||||
# sources that need to be compiled twice, once with no flags and once with LOWER
|
||||
set(UL_SOURCES
|
||||
@@ -72,7 +73,7 @@ GenerateNamedObjects("zgbmv_k.c" "TRANS;CONJ;XCONJ" "gbmv_d" false "" "" "" 2)
|
||||
# special defines for complex
|
||||
foreach (float_type ${FLOAT_TYPES})
|
||||
|
||||
if (SMP)
|
||||
if (USE_THREAD)
|
||||
GenerateNamedObjects("gemv_thread.c" "" "gemv_thread_n" false "" "" false ${float_type})
|
||||
GenerateNamedObjects("gemv_thread.c" "TRANSA" "gemv_thread_t" false "" "" false ${float_type})
|
||||
|
||||
@@ -106,7 +107,7 @@ foreach (float_type ${FLOAT_TYPES})
|
||||
GenerateNamedObjects("z${ulvm_source}" "LOWER;HEMVREV" "${op_name}_M" false "" "" false ${float_type})
|
||||
endforeach()
|
||||
|
||||
if (SMP)
|
||||
if (USE_THREAD)
|
||||
|
||||
GenerateNamedObjects("gemv_thread.c" "CONJ" "gemv_thread_r" false "" "" false ${float_type})
|
||||
GenerateNamedObjects("gemv_thread.c" "CONJ;TRANSA" "gemv_thread_c" false "" "" false ${float_type})
|
||||
@@ -185,7 +186,7 @@ foreach (float_type ${FLOAT_TYPES})
|
||||
GenerateCombinationObjects("${l_source}" "UNIT" "N" "TRANSA" 0 "${op_name}_TU" false ${float_type})
|
||||
endforeach ()
|
||||
|
||||
if (SMP)
|
||||
if (USE_THREAD)
|
||||
GenerateNamedObjects("ger_thread.c" "" "" false "" "" false ${float_type})
|
||||
foreach(nu_smp_source ${NU_SMP_SOURCES})
|
||||
string(REGEX MATCH "[a-z]+_[a-z]+" op_name ${nu_smp_source})
|
||||
@@ -196,7 +197,7 @@ foreach (float_type ${FLOAT_TYPES})
|
||||
endif ()
|
||||
endforeach ()
|
||||
|
||||
if (SMP)
|
||||
if (USE_THREAD)
|
||||
GenerateCombinationObjects("${UL_SMP_SOURCES}" "LOWER" "U" "" 2)
|
||||
endif ()
|
||||
|
||||
|
||||
@@ -62,13 +62,13 @@ void CNAME(BLASLONG m, BLASLONG n, BLASLONG ku, BLASLONG kl, FLOAT alpha,
|
||||
if (incy != 1) {
|
||||
Y = bufferY;
|
||||
bufferX = (FLOAT *)(((BLASLONG)bufferY + M * sizeof(FLOAT) + 4095) & ~4095);
|
||||
gemvbuffer = bufferX;
|
||||
// gemvbuffer = bufferX;
|
||||
COPY_K(M, y, incy, Y, 1);
|
||||
}
|
||||
|
||||
if (incx != 1) {
|
||||
X = bufferX;
|
||||
gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + N * sizeof(FLOAT) + 4095) & ~4095);
|
||||
// gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + N * sizeof(FLOAT) + 4095) & ~4095);
|
||||
COPY_K(N, x, incx, X, 1);
|
||||
}
|
||||
|
||||
|
||||
@@ -96,7 +96,7 @@ static int gbmv_kernel(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, F
|
||||
COPY_K(args -> m, x, incx, buffer, 1);
|
||||
|
||||
x = buffer;
|
||||
buffer += ((COMPSIZE * args -> m + 1023) & ~1023);
|
||||
// buffer += ((COMPSIZE * args -> m + 1023) & ~1023);
|
||||
}
|
||||
#endif
|
||||
|
||||
@@ -177,7 +177,7 @@ int CNAME(BLASLONG m, BLASLONG n, BLASLONG ku, BLASLONG kl, FLOAT *alpha, FLOAT
|
||||
|
||||
blas_arg_t args;
|
||||
blas_queue_t queue[MAX_CPU_NUMBER];
|
||||
BLASLONG range_m[MAX_CPU_NUMBER];
|
||||
BLASLONG range_m[MAX_CPU_NUMBER + 1];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER + 1];
|
||||
|
||||
BLASLONG width, i, num_cpu;
|
||||
@@ -230,8 +230,10 @@ int CNAME(BLASLONG m, BLASLONG n, BLASLONG ku, BLASLONG kl, FLOAT *alpha, FLOAT
|
||||
|
||||
#ifndef TRANSA
|
||||
range_m[num_cpu] = num_cpu * ((m + 15) & ~15);
|
||||
if (range_m[num_cpu] > m * num_cpu) range_m[num_cpu] = m * num_cpu;
|
||||
#else
|
||||
range_m[num_cpu] = num_cpu * ((n + 15) & ~15);
|
||||
if (range_m[num_cpu] > n * num_cpu) range_m[num_cpu] = n * num_cpu;
|
||||
#endif
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
|
||||
@@ -55,13 +55,13 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT alpha,
|
||||
if (incy != 1) {
|
||||
Y = bufferY;
|
||||
bufferX = (FLOAT *)(((BLASLONG)bufferY + n * sizeof(FLOAT) + 4095) & ~4095);
|
||||
sbmvbuffer = bufferX;
|
||||
// sbmvbuffer = bufferX;
|
||||
COPY_K(n, y, incy, Y, 1);
|
||||
}
|
||||
|
||||
if (incx != 1) {
|
||||
X = bufferX;
|
||||
sbmvbuffer = (FLOAT *)(((BLASLONG)bufferX + n * sizeof(FLOAT) + 4095) & ~4095);
|
||||
// sbmvbuffer = (FLOAT *)(((BLASLONG)bufferX + n * sizeof(FLOAT) + 4095) & ~4095);
|
||||
COPY_K(n, x, incx, X, 1);
|
||||
}
|
||||
|
||||
|
||||
@@ -91,7 +91,7 @@ static int sbmv_kernel(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, F
|
||||
COPY_K(n, x, incx, buffer, 1);
|
||||
|
||||
x = buffer;
|
||||
buffer += ((COMPSIZE * n + 1023) & ~1023);
|
||||
// buffer += ((COMPSIZE * n + 1023) & ~1023);
|
||||
}
|
||||
|
||||
SCAL_K(n, 0, 0, ZERO,
|
||||
@@ -177,7 +177,7 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT *alpha, FLOAT *a, BLASLONG lda, FLOAT *x
|
||||
#endif
|
||||
|
||||
blas_arg_t args;
|
||||
blas_queue_t queue[MAX_CPU_NUMBER];
|
||||
blas_queue_t queue[MAX_CPU_NUMBER + 1];
|
||||
BLASLONG range_m[MAX_CPU_NUMBER + 1];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER];
|
||||
|
||||
@@ -246,6 +246,7 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT *alpha, FLOAT *a, BLASLONG lda, FLOAT *x
|
||||
|
||||
range_m[MAX_CPU_NUMBER - num_cpu - 1] = range_m[MAX_CPU_NUMBER - num_cpu] - width;
|
||||
range_n[num_cpu] = num_cpu * (((n + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > n * num_cpu) range_n[num_cpu] = n * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = sbmv_kernel;
|
||||
@@ -285,6 +286,7 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT *alpha, FLOAT *a, BLASLONG lda, FLOAT *x
|
||||
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
range_n[num_cpu] = num_cpu * (((n + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > n * num_cpu) range_n[num_cpu] = n * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = sbmv_kernel;
|
||||
@@ -316,6 +318,7 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT *alpha, FLOAT *a, BLASLONG lda, FLOAT *x
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
|
||||
range_n[num_cpu] = num_cpu * ((n + 15) & ~15);
|
||||
if (range_n[num_cpu] > n * num_cpu) range_n[num_cpu] = n * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = sbmv_kernel;
|
||||
|
||||
@@ -53,13 +53,13 @@ int CNAME(BLASLONG m, FLOAT alpha, FLOAT *a,
|
||||
if (incy != 1) {
|
||||
Y = bufferY;
|
||||
bufferX = (FLOAT *)(((BLASLONG)bufferY + m * sizeof(FLOAT) + 4095) & ~4095);
|
||||
gemvbuffer = bufferX;
|
||||
// gemvbuffer = bufferX;
|
||||
COPY_K(m, y, incy, Y, 1);
|
||||
}
|
||||
|
||||
if (incx != 1) {
|
||||
X = bufferX;
|
||||
gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + m * sizeof(FLOAT) + 4095) & ~4095);
|
||||
// gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + m * sizeof(FLOAT) + 4095) & ~4095);
|
||||
COPY_K(m, x, incx, X, 1);
|
||||
}
|
||||
|
||||
|
||||
@@ -182,7 +182,7 @@ int CNAME(BLASLONG m, FLOAT *alpha, FLOAT *a, FLOAT *x, BLASLONG incx, FLOAT *y,
|
||||
blas_arg_t args;
|
||||
blas_queue_t queue[MAX_CPU_NUMBER];
|
||||
BLASLONG range_m[MAX_CPU_NUMBER + 1];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER + 1];
|
||||
|
||||
BLASLONG width, i, num_cpu;
|
||||
|
||||
@@ -246,6 +246,7 @@ int CNAME(BLASLONG m, FLOAT *alpha, FLOAT *a, FLOAT *x, BLASLONG incx, FLOAT *y,
|
||||
|
||||
range_m[MAX_CPU_NUMBER - num_cpu - 1] = range_m[MAX_CPU_NUMBER - num_cpu] - width;
|
||||
range_n[num_cpu] = num_cpu * (((m + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > m * num_cpu) range_n[num_cpu] = m * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = spmv_kernel;
|
||||
@@ -285,6 +286,7 @@ int CNAME(BLASLONG m, FLOAT *alpha, FLOAT *a, FLOAT *x, BLASLONG incx, FLOAT *y,
|
||||
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
range_n[num_cpu] = num_cpu * (((m + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > m * num_cpu) range_n[num_cpu] = m * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = spmv_kernel;
|
||||
|
||||
@@ -177,7 +177,8 @@ int CNAME(BLASLONG m, FLOAT *alpha, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG i
|
||||
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
range_n[num_cpu] = num_cpu * (((m + 15) & ~15) + 16);
|
||||
|
||||
if (range_n[num_cpu] > m * num_cpu) range_n[num_cpu] = m * num_cpu;
|
||||
|
||||
queue[MAX_CPU_NUMBER - num_cpu - 1].mode = mode;
|
||||
queue[MAX_CPU_NUMBER - num_cpu - 1].routine = symv_kernel;
|
||||
queue[MAX_CPU_NUMBER - num_cpu - 1].args = &args;
|
||||
@@ -225,6 +226,7 @@ int CNAME(BLASLONG m, FLOAT *alpha, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG i
|
||||
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
range_n[num_cpu] = num_cpu * (((m + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > m * num_cpu) range_n[num_cpu] = m * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = symv_kernel;
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -107,7 +107,7 @@ static int trmv_kernel(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, F
|
||||
COPY_K(args -> n, x, incx, buffer, 1);
|
||||
|
||||
x = buffer;
|
||||
buffer += ((args -> n * COMPSIZE + 1023) & ~1023);
|
||||
// buffer += ((args -> n * COMPSIZE + 1023) & ~1023);
|
||||
}
|
||||
|
||||
if (range_n) y += *range_n * COMPSIZE;
|
||||
@@ -221,7 +221,7 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG inc
|
||||
blas_arg_t args;
|
||||
blas_queue_t queue[MAX_CPU_NUMBER];
|
||||
BLASLONG range_m[MAX_CPU_NUMBER + 1];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER + 1];
|
||||
|
||||
BLASLONG width, i, num_cpu;
|
||||
|
||||
@@ -288,6 +288,7 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG inc
|
||||
|
||||
range_m[MAX_CPU_NUMBER - num_cpu - 1] = range_m[MAX_CPU_NUMBER - num_cpu] - width;
|
||||
range_n[num_cpu] = num_cpu * (((n + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > n * num_cpu) range_n[num_cpu] = n * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = trmv_kernel;
|
||||
@@ -327,6 +328,7 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG inc
|
||||
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
range_n[num_cpu] = num_cpu * (((n + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > n * num_cpu) range_n[num_cpu] = n * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = trmv_kernel;
|
||||
@@ -356,6 +358,7 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG inc
|
||||
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
range_n[num_cpu] = num_cpu * (((n + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > n * num_cpu) range_n[num_cpu] = n * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = trmv_kernel;
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG m, FLOAT *a, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG m, FLOAT *a, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -112,7 +112,7 @@ static int tpmv_kernel(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, F
|
||||
#endif
|
||||
|
||||
x = buffer;
|
||||
buffer += ((COMPSIZE * args -> m + 1023) & ~1023);
|
||||
// buffer += ((COMPSIZE * args -> m + 1023) & ~1023);
|
||||
}
|
||||
|
||||
#ifndef TRANS
|
||||
@@ -234,16 +234,12 @@ static int tpmv_kernel(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, F
|
||||
return 0;
|
||||
}
|
||||
|
||||
#ifndef COMPLEX
|
||||
int CNAME(BLASLONG m, FLOAT *a, FLOAT *x, BLASLONG incx, FLOAT *buffer, int nthreads){
|
||||
#else
|
||||
int CNAME(BLASLONG m, FLOAT *a, FLOAT *x, BLASLONG incx, FLOAT *buffer, int nthreads){
|
||||
#endif
|
||||
|
||||
blas_arg_t args;
|
||||
blas_queue_t queue[MAX_CPU_NUMBER];
|
||||
BLASLONG range_m[MAX_CPU_NUMBER + 1];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER + 1];
|
||||
|
||||
BLASLONG width, i, num_cpu;
|
||||
|
||||
@@ -307,7 +303,8 @@ int CNAME(BLASLONG m, FLOAT *a, FLOAT *x, BLASLONG incx, FLOAT *buffer, int nthr
|
||||
|
||||
range_m[MAX_CPU_NUMBER - num_cpu - 1] = range_m[MAX_CPU_NUMBER - num_cpu] - width;
|
||||
range_n[num_cpu] = num_cpu * (((m + 15) & ~15) + 16);
|
||||
|
||||
if (range_n[num_cpu] > m * num_cpu) range_n[num_cpu] = m * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = tpmv_kernel;
|
||||
queue[num_cpu].args = &args;
|
||||
@@ -346,6 +343,7 @@ int CNAME(BLASLONG m, FLOAT *a, FLOAT *x, BLASLONG incx, FLOAT *buffer, int nthr
|
||||
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
range_n[num_cpu] = num_cpu * (((m + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > m * num_cpu) range_n[num_cpu] = m * num_cpu;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = tpmv_kernel;
|
||||
|
||||
@@ -54,12 +54,16 @@ int CNAME(BLASLONG m, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, FLOAT *bu
|
||||
COPY_K(m, b, incb, buffer, 1);
|
||||
}
|
||||
|
||||
for (is = 0; is < m; is += DTB_ENTRIES){
|
||||
/*FIXME the GEMV unrolling performed here was found to be broken, see issue 1332 */
|
||||
/* Multiplying DTB size by 100 is just a quick-and-dirty hack to disable it for now[B */
|
||||
|
||||
min_i = MIN(m - is, DTB_ENTRIES);
|
||||
for (is = 0; is < m; is += DTB_ENTRIES * 100){
|
||||
|
||||
min_i = MIN(m - is, DTB_ENTRIES * 100);
|
||||
|
||||
#ifndef TRANSA
|
||||
if (is > 0){
|
||||
fprintf(stderr,"WARNING unrolling of the trmv_U loop may give wrong results\n");
|
||||
GEMV_N(is, min_i, 0, dp1,
|
||||
a + is * lda, lda,
|
||||
B + is, 1,
|
||||
|
||||
@@ -281,7 +281,7 @@ int CNAME(BLASLONG m, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG incx, FLOAT *bu
|
||||
blas_arg_t args;
|
||||
blas_queue_t queue[MAX_CPU_NUMBER];
|
||||
BLASLONG range_m[MAX_CPU_NUMBER + 1];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER];
|
||||
BLASLONG range_n[MAX_CPU_NUMBER + 1];
|
||||
|
||||
BLASLONG width, i, num_cpu;
|
||||
|
||||
@@ -346,6 +346,7 @@ int CNAME(BLASLONG m, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG incx, FLOAT *bu
|
||||
|
||||
range_m[MAX_CPU_NUMBER - num_cpu - 1] = range_m[MAX_CPU_NUMBER - num_cpu] - width;
|
||||
range_n[num_cpu] = num_cpu * (((m + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > m) range_n[num_cpu] = m;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = trmv_kernel;
|
||||
@@ -385,6 +386,7 @@ int CNAME(BLASLONG m, FLOAT *a, BLASLONG lda, FLOAT *x, BLASLONG incx, FLOAT *bu
|
||||
|
||||
range_m[num_cpu + 1] = range_m[num_cpu] + width;
|
||||
range_n[num_cpu] = num_cpu * (((m + 15) & ~15) + 16);
|
||||
if (range_n[num_cpu] > m) range_n[num_cpu] = m;
|
||||
|
||||
queue[num_cpu].mode = mode;
|
||||
queue[num_cpu].routine = trmv_kernel;
|
||||
|
||||
@@ -83,13 +83,13 @@ void CNAME(BLASLONG m, BLASLONG n, BLASLONG ku, BLASLONG kl, FLOAT alpha_r, FLOA
|
||||
if (incy != 1) {
|
||||
Y = bufferY;
|
||||
bufferX = (FLOAT *)(((BLASLONG)bufferY + M * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
gemvbuffer = bufferX;
|
||||
// gemvbuffer = bufferX;
|
||||
COPY_K(M, y, incy, Y, 1);
|
||||
}
|
||||
|
||||
if (incx != 1) {
|
||||
X = bufferX;
|
||||
gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + N * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
// gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + N * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
COPY_K(N, x, incx, X, 1);
|
||||
}
|
||||
|
||||
|
||||
@@ -61,13 +61,13 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT alpha_r, FLOAT alpha_i,
|
||||
if (incy != 1) {
|
||||
Y = bufferY;
|
||||
bufferX = (FLOAT *)(((BLASLONG)bufferY + n * sizeof(FLOAT) * COMPSIZE + 4095) & ~4095);
|
||||
sbmvbuffer = bufferX;
|
||||
// sbmvbuffer = bufferX;
|
||||
COPY_K(n, y, incy, Y, 1);
|
||||
}
|
||||
|
||||
if (incx != 1) {
|
||||
X = bufferX;
|
||||
sbmvbuffer = (FLOAT *)(((BLASLONG)bufferX + n * sizeof(FLOAT) * COMPSIZE + 4095) & ~4095);
|
||||
// sbmvbuffer = (FLOAT *)(((BLASLONG)bufferX + n * sizeof(FLOAT) * COMPSIZE + 4095) & ~4095);
|
||||
COPY_K(n, x, incx, X, 1);
|
||||
}
|
||||
|
||||
|
||||
@@ -56,13 +56,13 @@ int CNAME(BLASLONG m, FLOAT alpha_r, FLOAT alpha_i,
|
||||
if (incy != 1) {
|
||||
Y = bufferY;
|
||||
bufferX = (FLOAT *)(((BLASLONG)bufferY + m * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
gemvbuffer = bufferX;
|
||||
// gemvbuffer = bufferX;
|
||||
COPY_K(m, y, incy, Y, 1);
|
||||
}
|
||||
|
||||
if (incx != 1) {
|
||||
X = bufferX;
|
||||
gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + m * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
// gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + m * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
COPY_K(m, x, incx, X, 1);
|
||||
}
|
||||
|
||||
|
||||
@@ -60,13 +60,13 @@ int CNAME(BLASLONG n, BLASLONG k, FLOAT alpha_r, FLOAT alpha_i,
|
||||
if (incy != 1) {
|
||||
Y = bufferY;
|
||||
bufferX = (FLOAT *)(((BLASLONG)bufferY + n * sizeof(FLOAT) * COMPSIZE + 4095) & ~4095);
|
||||
sbmvbuffer = bufferX;
|
||||
// sbmvbuffer = bufferX;
|
||||
COPY_K(n, y, incy, Y, 1);
|
||||
}
|
||||
|
||||
if (incx != 1) {
|
||||
X = bufferX;
|
||||
sbmvbuffer = (FLOAT *)(((BLASLONG)bufferX + n * sizeof(FLOAT) * COMPSIZE + 4095) & ~4095);
|
||||
// sbmvbuffer = (FLOAT *)(((BLASLONG)bufferX + n * sizeof(FLOAT) * COMPSIZE + 4095) & ~4095);
|
||||
COPY_K(n, x, incx, X, 1);
|
||||
}
|
||||
|
||||
|
||||
@@ -55,13 +55,13 @@ int CNAME(BLASLONG m, FLOAT alpha_r, FLOAT alpha_i,
|
||||
if (incy != 1) {
|
||||
Y = bufferY;
|
||||
bufferX = (FLOAT *)(((BLASLONG)bufferY + m * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
gemvbuffer = bufferX;
|
||||
// gemvbuffer = bufferX;
|
||||
COPY_K(m, y, incy, Y, 1);
|
||||
}
|
||||
|
||||
if (incx != 1) {
|
||||
X = bufferX;
|
||||
gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + m * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
// gemvbuffer = (FLOAT *)(((BLASLONG)bufferX + m * sizeof(FLOAT) * 2 + 4095) & ~4095);
|
||||
COPY_K(m, x, incx, X, 1);
|
||||
}
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dp1 = 1.;
|
||||
// const static FLOAT dp1 = 1.;
|
||||
|
||||
int CNAME(BLASLONG n, BLASLONG k, FLOAT *a, BLASLONG lda, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -40,7 +40,7 @@
|
||||
#include <ctype.h>
|
||||
#include "common.h"
|
||||
|
||||
const static FLOAT dm1 = -1.;
|
||||
// const static FLOAT dm1 = -1.;
|
||||
|
||||
int CNAME(BLASLONG m, FLOAT *a, FLOAT *b, BLASLONG incb, void *buffer){
|
||||
|
||||
|
||||
@@ -1,4 +1,5 @@
|
||||
include_directories(${PROJECT_SOURCE_DIR})
|
||||
include_directories(${PROJECT_BINARY_DIR})
|
||||
|
||||
# N.B. In the original makefile there was a BLOCKS define used in the compilation of these files but I don't see any evidence of it being set anywhere. -hpa
|
||||
|
||||
@@ -8,7 +9,7 @@ set(GEMM_COMPLEX_DEFINES RN CN RT CT NR TR RR CR NC TC RC CC)
|
||||
foreach (GEMM_DEFINE ${GEMM_DEFINES})
|
||||
string(TOLOWER ${GEMM_DEFINE} GEMM_DEFINE_LC)
|
||||
GenerateNamedObjects("gemm.c" "${GEMM_DEFINE}" "gemm_${GEMM_DEFINE_LC}" 0)
|
||||
if (SMP AND NOT USE_SIMPLE_THREADED_LEVEL3)
|
||||
if (USE_THREAD AND NOT USE_SIMPLE_THREADED_LEVEL3)
|
||||
GenerateNamedObjects("gemm.c" "${GEMM_DEFINE};THREADED_LEVEL3" "gemm_thread_${GEMM_DEFINE_LC}" 0)
|
||||
endif ()
|
||||
endforeach ()
|
||||
@@ -31,7 +32,7 @@ GenerateCombinationObjects("syrk_k.c" "LOWER;TRANS" "U;N" "" 1)
|
||||
GenerateCombinationObjects("syr2k_k.c" "LOWER;TRANS" "U;N" "" 1)
|
||||
GenerateCombinationObjects("syrk_kernel.c" "LOWER" "U" "" 2)
|
||||
GenerateCombinationObjects("syr2k_kernel.c" "LOWER" "U" "" 2)
|
||||
if (SMP)
|
||||
if (USE_THREAD)
|
||||
|
||||
# N.B. these do NOT have a float type (e.g. DOUBLE) defined!
|
||||
GenerateNamedObjects("gemm_thread_m.c;gemm_thread_n.c;gemm_thread_mn.c;gemm_thread_variable.c;syrk_thread.c" "" "" 0 "" "" 1)
|
||||
@@ -70,7 +71,7 @@ foreach (float_type ${FLOAT_TYPES})
|
||||
GenerateNamedObjects("zher2k_k.c" "HER2K;LOWER" "her2k_LN" false "" "" false ${float_type})
|
||||
GenerateNamedObjects("zher2k_k.c" "HER2K;LOWER;TRANS;CONJ" "her2k_LC" false "" "" false ${float_type})
|
||||
|
||||
if (SMP AND NOT USE_SIMPLE_THREADED_LEVEL3)
|
||||
if (USE_THREAD AND NOT USE_SIMPLE_THREADED_LEVEL3)
|
||||
#herk
|
||||
GenerateCombinationObjects("zherk_k.c" "LOWER" "U" "HERK;THREADED_LEVEL3" 3 "herk_thread_N" false ${float_type})
|
||||
GenerateCombinationObjects("zherk_k.c" "LOWER" "U" "HERK;THREADED_LEVEL3;TRANS;CONJ" 3 "herk_thread_C" false ${float_type})
|
||||
@@ -92,7 +93,7 @@ foreach (float_type ${FLOAT_TYPES})
|
||||
if(USE_GEMM3M)
|
||||
GenerateNamedObjects("gemm3m.c" "${gemm_define}" "gemm3m_${gemm_define_LC}" false "" "" false ${float_type})
|
||||
endif()
|
||||
if (SMP AND NOT USE_SIMPLE_THREADED_LEVEL3)
|
||||
if (USE_THREAD AND NOT USE_SIMPLE_THREADED_LEVEL3)
|
||||
GenerateNamedObjects("gemm.c" "${gemm_define};THREADED_LEVEL3" "gemm_thread_${gemm_define_LC}" false "" "" false ${float_type})
|
||||
if(USE_GEMM3M)
|
||||
GenerateNamedObjects("gemm3m.c" "${gemm_define};THREADED_LEVEL3" "gemm3m_thread_${gemm_define_LC}" false "" "" false ${float_type})
|
||||
@@ -105,7 +106,7 @@ foreach (float_type ${FLOAT_TYPES})
|
||||
foreach (GEMM_DEFINE ${GEMM_DEFINES})
|
||||
string(TOLOWER ${GEMM_DEFINE} GEMM_DEFINE_LC)
|
||||
GenerateNamedObjects("gemm3m.c" "${GEMM_DEFINE}" "gemm3m_${GEMM_DEFINE_LC}" false "" "" false ${float_type})
|
||||
if (SMP AND NOT USE_SIMPLE_THREADED_LEVEL3)
|
||||
if (USE_THREAD AND NOT USE_SIMPLE_THREADED_LEVEL3)
|
||||
GenerateNamedObjects("gemm3m.c" "${GEMM_DEFINE};THREADED_LEVEL3" "gemm3m_thread_${GEMM_DEFINE_LC}" false "" "" false ${float_type})
|
||||
endif ()
|
||||
endforeach ()
|
||||
|
||||
@@ -251,11 +251,11 @@ int CNAME(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n,
|
||||
if ((k == 0) || (alpha == NULL)) return 0;
|
||||
|
||||
#if !defined(XDOUBLE) || !defined(QUAD_PRECISION)
|
||||
if ((alpha[0] == ZERO)
|
||||
if ( alpha[0] == ZERO
|
||||
#ifdef COMPLEX
|
||||
&& (alpha[1] == ZERO)
|
||||
&& alpha[1] == ZERO
|
||||
#endif
|
||||
) return 0;
|
||||
) return 0;
|
||||
#else
|
||||
if (((alpha[0].x[0] | alpha[0].x[1]
|
||||
#ifdef COMPLEX
|
||||
@@ -293,7 +293,7 @@ int CNAME(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n,
|
||||
min_l = k - ls;
|
||||
|
||||
if (min_l >= GEMM_Q * 2) {
|
||||
gemm_p = GEMM_P;
|
||||
// gemm_p = GEMM_P;
|
||||
min_l = GEMM_Q;
|
||||
} else {
|
||||
if (min_l > GEMM_Q) {
|
||||
|
||||
@@ -974,7 +974,7 @@ static int gemm_driver(blas_arg_t *args, BLASLONG *range_m, BLASLONG
|
||||
int CNAME(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, FLOAT *sa, FLOAT *sb, BLASLONG mypos){
|
||||
|
||||
BLASLONG m = args -> m;
|
||||
BLASLONG n = args -> n;
|
||||
// BLASLONG n = args -> n;
|
||||
BLASLONG nthreads = args -> nthreads;
|
||||
BLASLONG divN, divT;
|
||||
int mode;
|
||||
@@ -985,13 +985,14 @@ int CNAME(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, FLOAT *sa, FLO
|
||||
|
||||
m = m_to - m_from;
|
||||
}
|
||||
|
||||
/*
|
||||
if (range_n) {
|
||||
BLASLONG n_from = *(((BLASLONG *)range_n) + 0);
|
||||
BLASLONG n_to = *(((BLASLONG *)range_n) + 1);
|
||||
|
||||
n = n_to - n_from;
|
||||
}
|
||||
*/
|
||||
|
||||
if ((args -> m < nthreads * SWITCH_RATIO) || (args -> n < nthreads * SWITCH_RATIO)) {
|
||||
GEMM3M_LOCAL(args, range_m, range_n, sa, sb, 0);
|
||||
|
||||
@@ -154,9 +154,9 @@ int CNAME(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, FLOAT *sa, FLO
|
||||
|
||||
if ((k == 0) || (alpha == NULL)) return 0;
|
||||
|
||||
if ((alpha[0] == ZERO)
|
||||
if (alpha[0] == ZERO
|
||||
#ifdef COMPLEX
|
||||
&& (alpha[1] == ZERO)
|
||||
&& alpha[1] == ZERO
|
||||
#endif
|
||||
) return 0;
|
||||
|
||||
|
||||
@@ -158,9 +158,9 @@ int CNAME(blas_arg_t *args, BLASLONG *range_m, BLASLONG *range_n, FLOAT *sa, FLO
|
||||
|
||||
if ((k == 0) || (alpha == NULL)) return 0;
|
||||
|
||||
if ((alpha[0] == ZERO)
|
||||
if (alpha[0] == ZERO
|
||||
#if defined(COMPLEX) && !defined(HERK)
|
||||
&& (alpha[1] == ZERO)
|
||||
&& alpha[1] == ZERO
|
||||
#endif
|
||||
) return 0;
|
||||
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user