You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

cgemm_kernel_power9.S 6.8 kB

rebase? (#1) * With the Intel compiler on Linux, prefer ifort for the final link step icc has known problems with mixed-language builds that ifort can handle just fine. Fixes #1956 * Rename operands to put lda on the input/output constraint list * Fix wrong constraints in inline assembly for #2009 * Fix inline assembly constraints rework indices to allow marking argument lda4 as input and output. For #2009 * Fix inline assembly constraints rework indices to allow marking argument lda as input and output. * Fix inline assembly constraints * Fix inline assembly constraints * Fix inline assembly constraints in Bulldozer TRSM kernels rework indices to allow marking i,as and bs as both input and output (marked operand n1 as well for simplicity). For #2009 * Correct range_n limiting same bug as seen in #1388, somehow missed in corresponding PR #1389 * Allow multithreading TRMV again revert workaround introduced for issue #1332 as the actual cause appears to be my incorrect fix from #1262 (see #1388) * Fix error introduced during cleanup * Reduce list of kernels in the dynamic arch build to make compilation complete reliably within the 1h limit again * init * move fix to right place * Fix missing -c option in AVX512 test * Fix AVX512 test always returning false due to missing compiler option * Make x86_32 imply NO_AVX2, NO_AVX512 in addition to NO_AVX fixes #2033 * Keep xcode8.3 for osx BINARY=32 build as xcode10 deprecated i386 * Make sure that AVX512 is disabled in 32bit builds for #2033 * Improve handling of NO_STATIC and NO_SHARED to avoid surprises from defining either as zero. Fixes #2035 by addressing some concerns from #1422 * init * address warning introed with #1814 et al * Restore locking optimizations for OpenMP case restore another accidentally dropped part of #1468 that was missed in #2004 to address performance regression reported in #1461 * HiSilicon tsv110 CPUs optimization branch add HiSilicon tsv110 CPUs optimization branch * add TARGET support for HiSilicon tsv110 CPUs * add TARGET support for HiSilicon tsv110 CPUs * add TARGET support for HiSilicon tsv110 CPUs * Fix module definition conflicts between LAPACK and ReLAPACK for #2043 * Do not compile in AVX512 check if AVX support is disabled xgetbv is function depends on NO_AVX being undefined - we could change that too, but that combo is unlikely to work anyway * ctest.c : add __POWERPC__ for PowerMac * Fix crash in sgemm SSE/nano kernel on x86_64 Fix bug #2047. Signed-off-by: Celelibi <celelibi@gmail.com> * param.h : enable defines for PPC970 on DarwinOS fixes: gemm.c: In function 'sgemm_': ../common_param.h:981:18: error: 'SGEMM_DEFAULT_P' undeclared (first use in this function) #define SGEMM_P SGEMM_DEFAULT_P ^ * common_power.h: force DCBT_ARG 0 on PPC970 Darwin without this, we see ../kernel/power/gemv_n.S:427:Parameter syntax error and many more similar entries that relates to this assembly command dcbt 8, r24, r18 this change makes the DCBT_ARG = 0 and openblas builds through to completion on PowerMac 970 Tests pass * Make TARGET=GENERIC compatible with DYNAMIC_ARCH=1 for issue #2048 * make DYNAMIC_ARCH=1 package work on TSV110. * make DYNAMIC_ARCH=1 package work on TSV110 * Add Intel Denverton for #2048 * Add Intel Denverton * Change 64-bit detection as explained in #2056 * Trivial typo fix as suggested in #2022 * Disable the AVX512 DGEMM kernel (again) Due to as yet unresolved errors seen in #1955 and #2029 * Use POSIX getenv on Cygwin The Windows-native GetEnvironmentVariable cannot be relied on, as Cygwin does not always copy environment variables set through Cygwin to the Windows environment block, particularly after fork(). * Fix for #2063: The DllMain used in Cygwin did not run the thread memory pool cleanup upon THREAD_DETACH which is needed when compiled with USE_TLS=1. * Also call CloseHandle on each thread, as well as on the event so as to not leak thread handles. * AIX asm syntax changes needed for shared object creation * power9 makefile. dgemm based on power8 kernel with following changes : 32x unrolled 16x4 kernel and 8x4 kernel using (lxv stxv butterfly rank1 update). improvement from 17 to 22-23gflops. dtrmm cases were added into dgemm itself * Expose CBLAS interfaces for I?MIN and I?MAX * Build CBLAS interfaces for I?MIN and I?MAX * Add declarations for ?sum and cblas_?sum * Add interface for ?sum (derived from ?asum) * Add ?sum * Add implementations of ssum/dsum and csum/zsum as trivial copies of asum/zsasum with the fabs calls replaced by fmov to preserve code structure * Add ARM implementations of ?sum (trivial copies of the respective ?asum with the fabs calls removed) * Add ARM64 implementations of ?sum as trivial copies of the respective ?asum kernels with the fabs calls removed * Add ia64 implementation of ?sum as trivial copy of asum with the fabs calls removed * Add MIPS implementation of ?sum as trivial copy of ?asum with the fabs calls removed * Add MIPS64 implementation of ?sum as trivial copy of ?asum with the fabs replaced by mov to preserve code structure * Add POWER implementation of ?sum as trivial copy of ?asum with the fabs replaced by fmr to preserve code structure * Add SPARC implementation of ?sum as trivial copy of ?asum with the fabs replaced by fmov to preserve code structure * Add x86 implementation of ?sum as trivial copy of ?asum with the fabs calls removed * Add x86_64 implementation of ?sum as trivial copy of ?asum with the fabs calls removed * Add ZARCH implementation of ?sum as trivial copies of the respective ?asum kernels with the ABS and vflpsb calls removed * Detect 32bit environment on 64bit ARM hardware for #2056, using same approach as #2058 * Add cmake defaults for ?sum kernels * Add ?sum * Add ?sum definitions for generic kernel * Add declarations for ?sum * Add -lm and disable EXPRECISION support on *BSD fixes #2075 * Add in runtime CPU detection for POWER. * snprintf define consolidated to common.h * Support INTERFACE64=1 * Add support for INTERFACE64 and fix XERBLA calls 1. Replaced all instances of "int" with "blasint" 2. Added string length as "hidden" third parameter in calls to fortran XERBLA * Correct length of name string in xerbla call * Avoid out-of-bounds accesses in LAPACK EIG tests see https://github.com/Reference-LAPACK/lapack/issues/333 * Correct INFO=4 condition * Disable reallocation of work array in xSYTRF as it appears to cause memory management problems (seen in the LAPACK tests) * Disable repeated recursion on Ab_BR in ReLAPACK xGBTRF due to crashes in LAPACK tests * sgemm/strmm * Update Changelog with changes from 0.3.6 * Increment version to 0.3.7.dev * Increment version to 0.3.7.dev * Misc. typo fixes Found via `codespell -q 3 -w -L ith,als,dum,nd,amin,nto,wis,ba -S ./relapack,./kernel,./lapack-netlib` * Correct argument of CPU_ISSET for glibc <2.5 fixes #2104 * conflict resolve * Revert reference/ fixes * Revert Changelog.txt typos * Disable the SkyLakeX DGEMMITCOPY kernel as well as a stopgap measure for https://github.com/numpy/numpy/issues/13401 as mentioned in #1955 * Disable DGEMMINCOPY as well for now #1955 * init * Fix errors in cpu enumeration with glibc 2.6 for #2114 * Change two http links to https Closes #2109 * remove redundant code #2113 * Set up CI with Azure Pipelines [skip ci] * TST: add native POWER8 to CI * add native POWER8 testing to Travis CI matrix with ppc64le os entry * Update link to IBM MASS library, update cpu support status * first try migrating one of the arm builds from travis * fix tabbing in azure commands * Update azure-pipelines.yml take out offending lines (although stolen from https://github.com/conda-forge/opencv-feedstock azure-pipelines fiie) * Update azure-pipelines.yml * Update azure-pipelines.yml * Update azure-pipelines.yml * Update azure-pipelines.yml * DOC: Add Azure CI status badge * Add ARMV6 build to azure CI setup (#2122) using aytekinar's Alpine image and docker script from the Travis setup [skip ci] * TST: Azure manylinux1 & clean-up * remove some of the steps & comments from the original Azure yml template * modify the trigger section to use develop since OpenBLAS primarily uses this branch; use the same batching behavior as downstream projects NumPy/ SciPy * remove Travis emulated ARMv6 gcc build because this now happens in Azure * use documented Ubuntu vmImage name for Azure and add in a manylinux1 test run to the matrix [skip appveyor] * Add NO_AFFINITY to available options on Linux, and set it to ON to match the gmake default. Fixes second part of #2114 * Replace ISMIN and ISAMIN kernels on all x86_64 platforms (#2125) * Mark iamax_sse.S as unsuitable for MIN due to issue #2116 * Use iamax.S rather than iamax_sse.S for ISMIN/ISAMIN on all x86_64 as workaround for #2116 * Move ARMv8 gcc build from Travis to Azure * Move ARMv8 gcc build from Travis to Azure * Update .travis.yml * Test drone CI * install make * remove sudo * Install gcc * Install perl * Install gfortran and add a clang job * gfortran->gcc-gfortran * Switch to ubuntu and parallel jobs * apt update * Fix typo * update yes * no need of gcc in clang build * Add a cmake build as well * Add cmake builds and print options * build without lapack on cmake * parallel build * See if ubuntu 19.04 fixes the ICE * Remove qemu armv8 builds * arm32 build * Fix typo * TST: add SkylakeX AVX512 CI test * adapt the C-level reproducer code for some recent SkylakeX AVX512 kernel issues, provided by Isuru Fernando and modified by Martin Kroeker, for usage in the utest suite * add an Intel SDE SkylakeX emulation utest run to the Azure CI matrix; a custom Docker build was required because Ubuntu image provided by Azure does not support AVX512VL instructions * Add option USE_LOCKING for single-threaded build with locking support for calling from concurrent threads * Add option USE_LOCKING for single-threaded build with locking support * Add option USE_LOCKING for SMP-like locking in USE_THREAD=0 builds * Add option USE_LOCKING but keep default settings intact * Remove unrelated change * Do not try ancient PGI hacks with recent versions of that compiler should fix #2139 * Build and run utests in any case, they do their own checks for fortran availability * Add softfp support in min/max kernels fix for #1912 * Revert "Add softfp support in min/max kernels" * Separate implementations of AMAX and IAMAX on arm As noted in #1912 and comment on #1942, the combined implementation happens to "do the right thing" on hardfp, but cannot return both value and index on softfp where they would have to share the return register * Ensure correct output for DAMAX with softfp * Use generic kernels for complex (I)AMAX to support softfp * improved zgemm power9 based on power8 * upload thread safety test folder * hook up c++ thread safety test (main Makefile) * add c++ thread test option to Makefile.rule * Document NO_AVX512 for #2151 * sgemm pipeline improved, zgemm rewritten without inner packs, ABI lxvx v20 fixed with vs52 * Fix detection of AVX512 capable compilers in getarch 21eda8b5 introduced a check in getarch.c to test if the compiler is capable of AVX512. This check currently fails, since the used __AVX2__ macro is only defined if getarch itself was compiled with AVX2/AVX512 support. Make sure this is the case by building getarch with -march=native on x86_64. It is only supposed to run on the build host anyway. * c_check: Unlink correct file * power9 zgemm ztrmm optimized * conflict resolve * Add gfortran workaround for ABI violations in LAPACKE for #2154 (see gcc bug 90329) * Add gfortran workaround for ABI violations for #2154 (see gcc bug 90329) * Add gfortran workaround for potential ABI violation for #2154 * Update fc.cmake * Remove any inadvertent use of -march=native from DYNAMIC_ARCH builds from #2143, -march=native precludes use of more specific options like -march=skylake-avx512 in individual kernels, and defeats the purpose of dynamic arch anyway. * Avoid unintentional activation of TLS code via USE_TLS=0 fixes #2149 * Do not force gcc options on non-gcc compilers fixes compile failure with pgi 18.10 as reported on OpenBLAS-users * Update Makefile.x86_64 * Zero ecx with a mov instruction PGI assembler does not like the initialization in the constraints. * Fix mov syntax * new sgemm 8x16 * Update dtrmm_kernel_16x4_power8.S * PGI compiler does not like -march=native * Fix build on FreeBSD/powerpc64. Signed-off-by: Piotr Kubaj <pkubaj@anongoth.pl> * Fix build for PPC970 on FreeBSD pt. 1 FreeBSD needs DCBT_ARG=0 as well. * Fix build for PPC970 on FreeBSD pt.2 FreeBSD needs those macros too. * cgemm/ctrmm power9 * Utest needs CBLAS but not necessarily FORTRAN * Add mingw builds to Appveyor config * Add getarch flags to disable AVX on x86 (and other small fixes to match Makefile behaviour) * Make disabling DYNAMIC_ARCH on unsupported systems work needs to be unset in the cache for the change to have any effect * Mingw32 needs leading underscore on object names (also copy BUNDERSCORE settings for FORTRAN from the corresponding Makefile)
7 years ago
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293
  1. /***************************************************************************
  2. Copyright (c) 2013-2019, The OpenBLAS Project
  3. All rights reserved.
  4. Redistribution and use in source and binary forms, with or without
  5. modification, are permitted provided that the following conditions are
  6. met:
  7. 1. Redistributions of source code must retain the above copyright
  8. notice, this list of conditions and the following disclaimer.
  9. 2. Redistributions in binary form must reproduce the above copyright
  10. notice, this list of conditions and the following disclaimer in
  11. the documentation and/or other materials provided with the
  12. distribution.
  13. 3. Neither the name of the OpenBLAS project nor the names of
  14. its contributors may be used to endorse or promote products
  15. derived from this software without specific prior written permission.
  16. THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
  17. AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
  18. IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
  19. ARE DISCLAIMED. IN NO EVENT SHALL THE OPENBLAS PROJECT OR CONTRIBUTORS BE
  20. LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
  21. DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
  22. SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
  23. CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
  24. OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE
  25. USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  26. *****************************************************************************/
  27. /**************************************************************************************
  28. * Abdelrauf(quickwritereader@gmail.com)
  29. * BLASTEST : OK
  30. * CTEST : OK
  31. * TEST : OK
  32. * LAPACK-TEST : OK
  33. **************************************************************************************/
  34. #define ASSEMBLER
  35. #include "common.h"
  36. #include "def_vsx.h"
  37. #define LOAD ld
  38. #define STACKSIZE (512 )
  39. #define FLINK_SAVE (STACKSIZE+16) /* 16($r12) */
  40. #define M r3
  41. #define N r4
  42. #define K r5
  43. #define A r8
  44. #define B r9
  45. #define C r10
  46. #define LDC r6
  47. #define OFFSET r7
  48. #define alpha_r vs19
  49. #define alpha_i vs20
  50. #define save_permute_1 vs21
  51. #define permute_mask vs22
  52. #define o0 0
  53. #define T1 r11
  54. #define T2 r12
  55. #define T3 r14
  56. #define T4 r15
  57. #define T5 r16
  58. #define T6 r17
  59. #define L r18
  60. #define T7 r19
  61. #define T8 r20
  62. #define TEMP_REG r21
  63. #define I r22
  64. #define J r23
  65. #define AO r24
  66. #define BO r25
  67. #define CO r26
  68. #define T9 r27
  69. #define T10 r28
  70. #define PRE r29
  71. #define T12 r30
  72. #define T13 r31
  73. #include "cgemm_macros_power9.S"
  74. .equ perm_const1, 0x0405060700010203
  75. .equ perm_const2, 0x0c0d0e0f08090a0b
  76. .equ save_permute_12, 0x0c0d0e0f1c1d1e1f
  77. .equ save_permute_11, 0x0405060714151617
  78. #ifndef NEEDPARAM
  79. PROLOGUE
  80. PROFCODE
  81. addi SP, SP, -STACKSIZE
  82. mflr r0
  83. stfd f14, 0(SP)
  84. stfd f15, 8(SP)
  85. stfd f16, 16(SP)
  86. stfd f17, 24(SP)
  87. stfd f18, 32(SP)
  88. stfd f19, 40(SP)
  89. stfd f20, 48(SP)
  90. stfd f21, 56(SP)
  91. stfd f22, 64(SP)
  92. stfd f23, 72(SP)
  93. stfd f24, 80(SP)
  94. stfd f25, 88(SP)
  95. stfd f26, 96(SP)
  96. stfd f27, 104(SP)
  97. stfd f28, 112(SP)
  98. stfd f29, 120(SP)
  99. stfd f30, 128(SP)
  100. stfd f31, 136(SP)
  101. std r31, 144(SP)
  102. std r30, 152(SP)
  103. std r29, 160(SP)
  104. std r28, 168(SP)
  105. std r27, 176(SP)
  106. std r26, 184(SP)
  107. std r25, 192(SP)
  108. std r24, 200(SP)
  109. std r23, 208(SP)
  110. std r22, 216(SP)
  111. std r21, 224(SP)
  112. std r20, 232(SP)
  113. std r19, 240(SP)
  114. std r18, 248(SP)
  115. std r17, 256(SP)
  116. std r16, 264(SP)
  117. std r15, 272(SP)
  118. std r14, 280(SP)
  119. stxv vs52, 288(SP)
  120. stxv vs53, 304(SP)
  121. stxv vs54, 320(SP)
  122. stxv vs55, 336(SP)
  123. stxv vs56, 352(SP)
  124. stxv vs57, 368(SP)
  125. stxv vs58, 384(SP)
  126. stxv vs59, 400(SP)
  127. stxv vs60, 416(SP)
  128. stxv vs61, 432(SP)
  129. stxv vs62, 448(SP)
  130. stxv vs63, 464(SP)
  131. std r0, FLINK_SAVE(SP)
  132. ld LDC, FRAMESLOT(0) + STACKSIZE(SP)
  133. #ifdef TRMMKERNEL
  134. ld OFFSET, FRAMESLOT(1) + STACKSIZE(SP)
  135. #endif
  136. slwi LDC, LDC, ZBASE_SHIFT
  137. /*alpha is stored in f1. convert to single and splat*/
  138. xscvdpspn alpha_r,vs1
  139. xscvdpspn alpha_i,vs2
  140. xxspltw alpha_r,alpha_r,0
  141. xxspltw alpha_i,alpha_i,0
  142. /*load reverse permute mask for big endian
  143. uint128 = 0xc0d0e0f08090a0b0405060700010203
  144. */
  145. lis T2, perm_const2@highest
  146. lis T1, perm_const1@highest
  147. lis T3, save_permute_12@highest
  148. lis T4, save_permute_11@highest
  149. ori T2, T2, perm_const2@higher
  150. ori T1, T1, perm_const1@higher
  151. ori T3, T3, save_permute_12@higher
  152. ori T4, T4, save_permute_11@higher
  153. rldicr T2, T2, 32, 31
  154. rldicr T1, T1, 32, 31
  155. rldicr T3, T3, 32, 31
  156. rldicr T4, T4, 32, 31
  157. oris T2, T2, perm_const2@h
  158. oris T1, T1, perm_const1@h
  159. oris T3, T3, save_permute_12@h
  160. oris T4, T4, save_permute_11@h
  161. ori T2, T2, perm_const2@l
  162. ori T1, T1, perm_const1@l
  163. ori T3, T3, save_permute_12@l
  164. ori T4, T4, save_permute_11@l
  165. li r0,0
  166. li PRE,512
  167. #if defined(CC) || defined(CR) || defined(RC) || defined(RR)
  168. /*negate for this case as we will use addition -1*(a+b) */
  169. xvnegsp alpha_r,alpha_r
  170. xvnegsp alpha_i,alpha_i
  171. #endif
  172. mtvsrdd permute_mask,T2,T1
  173. mtvsrdd save_permute_1,T3,T4
  174. /*mask is reverse permute so we have to make it inner permute */
  175. xxpermdi permute_mask, permute_mask, permute_mask,2
  176. #include "cgemm_logic_power9.S"
  177. .L999:
  178. lfd f14, 0(SP)
  179. lfd f15, 8(SP)
  180. lfd f16, 16(SP)
  181. lfd f17, 24(SP)
  182. lfd f18, 32(SP)
  183. lfd f19, 40(SP)
  184. lfd f20, 48(SP)
  185. lfd f21, 56(SP)
  186. lfd f22, 64(SP)
  187. lfd f23, 72(SP)
  188. lfd f24, 80(SP)
  189. lfd f25, 88(SP)
  190. lfd f26, 96(SP)
  191. lfd f27, 104(SP)
  192. lfd f28, 112(SP)
  193. lfd f29, 120(SP)
  194. lfd f30, 128(SP)
  195. lfd f31, 136(SP)
  196. ld r31, 144(SP)
  197. ld r30, 152(SP)
  198. ld r29, 160(SP)
  199. ld r28, 168(SP)
  200. ld r27, 176(SP)
  201. ld r26, 184(SP)
  202. ld r25, 192(SP)
  203. ld r24, 200(SP)
  204. ld r23, 208(SP)
  205. ld r22, 216(SP)
  206. ld r21, 224(SP)
  207. ld r20, 232(SP)
  208. ld r19, 240(SP)
  209. ld r18, 248(SP)
  210. ld r17, 256(SP)
  211. ld r16, 264(SP)
  212. ld r15, 272(SP)
  213. ld r14, 280(SP)
  214. ld r0, FLINK_SAVE(SP)
  215. lxv vs52, 288(SP)
  216. lxv vs53, 304(SP)
  217. lxv vs54, 320(SP)
  218. lxv vs55, 336(SP)
  219. lxv vs56, 352(SP)
  220. lxv vs57, 368(SP)
  221. lxv vs58, 384(SP)
  222. lxv vs59, 400(SP)
  223. mtlr r0
  224. lxv vs60, 416(SP)
  225. lxv vs61, 432(SP)
  226. lxv vs62, 448(SP)
  227. lxv vs63, 464(SP)
  228. addi SP, SP, STACKSIZE
  229. blr
  230. EPILOGUE
  231. #endif