Skip to content

Commit

Permalink
Upgrade gensymbol (#125)
Browse files Browse the repository at this point in the history
* Upgrade gensymbol

* Update CI scripts
  • Loading branch information
amontoison committed May 1, 2024
1 parent a46dbfb commit 3c451b9
Show file tree
Hide file tree
Showing 4 changed files with 351 additions and 49 deletions.
2 changes: 1 addition & 1 deletion .buildkite/pipeline.yml
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@ steps:
}
# We'll test on the following julia versions
for JULIA_VERSION in 1.8 nightly; do
for JULIA_VERSION in 1.10 nightly; do
# First, linux
OS="linux"
Expand Down
2 changes: 1 addition & 1 deletion .cirrus.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@ task:
matrix:
- name: FreeBSD
freebsd_instance:
image: freebsd-13-1-release-amd64
image: freebsd-13-2-release-amd64
env:
matrix:
- JULIA_VERSION: 1
Expand Down
204 changes: 184 additions & 20 deletions ext/gensymbol/gensymbol
100755 → 100644
Original file line number Diff line number Diff line change
Expand Up @@ -17,38 +17,38 @@
# removed blas_thread_shutdown_
#
@blasobjsc = (
caxpy,caxpby,ccopy,cdotc,cdotu,cgbmv,cgemm,cgemmt,cgemv,cgerc,cgeru,
caxpy,caxpby,ccopy,cdotc,cdotu,cgbmv,cgemm,cgemv,cgerc,cgeru,
chbmv,chemm,chemv,cher2,cher2k,cher,cherk,scabs1,scamax,
chpmv,chpr2,chpr,crotg,cscal,csrot,csscal,cswap,scamin,scasum,scnrm2,
csymm,csyr2k,csyrk,ctbmv,ctbsv,ctpmv,ctpsv,ctrmm,ctrmv,ctrsm,
ctrsv,icamax,icamin,cimatcopy,comatcopy,cgeadd,scsum);
ctrsv,icamax,icamin,cimatcopy,comatcopy,cgeadd,scsum,cgemmt);

@blasobjsd = (
damax,damin,dasum,daxpy,daxpby,dcabs1,dcopy,ddot,dgbmv,dgemm,dgemmt,
damax,damin,dasum,daxpy,daxpby,dcabs1,dcopy,ddot,dgbmv,dgemm,
dgemv,dger,dmax,dmin,dnrm2,drot,drotg,drotm,drotmg,dsbmv,
dscal,dsdot,dspmv,dspr2,dimatcopy,domatcopy,
dspr,dswap,dsymm,dsymv,dsyr2,dsyr2k,dsyr,dsyrk,dtbmv,dtbsv,
dtpmv,dtpsv,dtrmm,dtrmv,dtrsm,dtrsv,
idamax,idamin,idmax,idmin,dgeadd,dsum);
idamax,idamin,idmax,idmin,dgeadd,dsum,dgemmt);

@blasobjss = (
isamax,isamin,ismax,ismin,
samax,samin,sasum,saxpy, saxpby,
scopy,sdot,sdsdot,sgbmv,sgemm,sgemmt,sgemv,sger,
scopy,sdot,sdsdot,sgbmv,sgemm,sgemv,sger,
smax,smin,snrm2,simatcopy,somatcopy,
srot,srotg,srotm,srotmg,ssbmv,sscal,sspmv,sspr2,sspr,sswap,
ssymm,ssymv,ssyr2,ssyr2k,ssyr,ssyrk,stbmv,stbsv,stpmv,stpsv,
strmm,strmv,strsm,strsv, sgeadd,ssum);
strmm,strmv,strsm,strsv, sgeadd,ssum,sgemmt);

@blasobjsz = (
izamax,izamin,,
zaxpy,zaxpby,zcopy,zdotc,zdotu,zdrot,
zdscal,zgbmv,zgemm,zgemmt,zgemv,zgerc,zgeru,
zdscal,zgbmv,zgemm,zgemv,zgerc,zgeru,
zhbmv,zhemm,zhemv,zher2,zher2k,zher,zherk,zhpmv,zhpr2,
zhpr,zrotg,zscal,zswap,zsymm,zsyr2k,zsyrk,ztbmv,
ztbsv,ztpmv,ztpsv,ztrmm,ztrmv,ztrsm,ztrsv,
zomatcopy, zimatcopy,dzamax,dzamin,dzasum,dznrm2,
zgeadd, dzsum);
zomatcopy,zimatcopy,dzamax,dzamin,dzasum,dznrm2,
zgeadd,dzsum,zgemmt);

@blasobjs = (lsame, xerbla);
@bfblasobjs = (sbgemm, sbgemv, sbdot, sbstobf16, sbdtobf16, sbf16tos, dbf16tod);
Expand All @@ -59,17 +59,17 @@
cblas_csscal, cblas_cswap, cblas_csymm, cblas_csyr2k, cblas_csyrk, cblas_ctbmv, cblas_cgeadd,
cblas_ctbsv, cblas_ctpmv, cblas_ctpsv, cblas_ctrmm, cblas_ctrmv, cblas_ctrsm, cblas_ctrsv,
cblas_scnrm2, cblas_scasum,
cblas_icamax, cblas_icamin, cblas_icmin, cblas_icmax, cblas_scsum,cblas_cimatcopy,cblas_comatcopy
);
cblas_icamax, cblas_icamin, cblas_icmin, cblas_icmax, cblas_scsum,cblas_cimatcopy,cblas_comatcopy,
cblas_cgemmt, cblas_caxpyc, cblas_crotg, cblas_csrot, cblas_scamax, cblas_scamin);
@cblasobjsd = (
cblas_dasum, cblas_daxpy, cblas_dcopy, cblas_ddot,
cblas_dgbmv, cblas_dgemm, cblas_dgemv, cblas_dger, cblas_dnrm2,
cblas_drot, cblas_drotg, cblas_drotm, cblas_drotmg, cblas_dsbmv, cblas_dscal, cblas_dsdot,
cblas_dspmv, cblas_dspr2, cblas_dspr, cblas_dswap, cblas_dsymm, cblas_dsymv, cblas_dsyr2,
cblas_dsyr2k, cblas_dsyr, cblas_dsyrk, cblas_dtbmv, cblas_dtbsv, cblas_dtpmv, cblas_dtpsv,
cblas_dtrmm, cblas_dtrmv, cblas_dtrsm, cblas_dtrsv, cblas_daxpby, cblas_dgeadd,
cblas_idamax, cblas_idamin, cblas_idmin, cblas_idmax, cblas_dsum,cblas_dimatcopy,cblas_domatcopy
);
cblas_idamax, cblas_idamin, cblas_idmin, cblas_idmax, cblas_dsum,cblas_dimatcopy,cblas_domatcopy,
cblas_dgemmt, cblas_damax, cblas_damin);

@cblasobjss = (
cblas_sasum, cblas_saxpy, cblas_saxpby,
Expand All @@ -79,8 +79,8 @@
cblas_sswap, cblas_ssymm, cblas_ssymv, cblas_ssyr2, cblas_ssyr2k, cblas_ssyr, cblas_ssyrk,
cblas_stbmv, cblas_stbsv, cblas_stpmv, cblas_stpsv, cblas_strmm, cblas_strmv, cblas_strsm,
cblas_strsv, cblas_sgeadd,
cblas_isamax, cblas_isamin, cblas_ismin, cblas_ismax, cblas_ssum,cblas_simatcopy,cblas_somatcopy
);
cblas_isamax, cblas_isamin, cblas_ismin, cblas_ismax, cblas_ssum,cblas_simatcopy,cblas_somatcopy,
cblas_sgemmt, cblas_samax, cblas_samin);
@cblasobjsz = (
cblas_dzasum, cblas_dznrm2, cblas_zaxpy, cblas_zcopy, cblas_zdotc, cblas_zdotu, cblas_zdscal,
cblas_zgbmv, cblas_zgemm, cblas_zgemv, cblas_zgerc, cblas_zgeru, cblas_zhbmv, cblas_zhemm,
Expand All @@ -89,8 +89,8 @@
cblas_ztbmv, cblas_ztbsv, cblas_ztpmv, cblas_ztpsv, cblas_ztrmm, cblas_ztrmv, cblas_ztrsm,
cblas_ztrsv, cblas_cdotc_sub, cblas_cdotu_sub, cblas_zdotc_sub, cblas_zdotu_sub,
cblas_zaxpby, cblas_zgeadd,
cblas_izamax, cblas_izamin, cblas_izmin, cblas_izmax, cblas_dzsum,cblas_zimatcopy,cblas_zomatcopy
);
cblas_izamax, cblas_izamin, cblas_izmin, cblas_izmax, cblas_dzsum,cblas_zimatcopy,cblas_zomatcopy,
cblas_zgemmt, cblas_dzamax, cblas_dzamin);

@cblasobjs = ( cblas_xerbla );

Expand Down Expand Up @@ -709,6 +709,7 @@ zpotri,
# functions added for lapack-3.7.0
@lapackobjs2s = (@lapackobjs2s,
slarfy,
ssyconvf,
strevc3,
sgelqt,
sgelqt3,
Expand Down Expand Up @@ -813,7 +814,86 @@ zpotri,
@lapackobjs2z = (@lapackobjs2z,
zgesvdq,
zungtsqr
);
);
#functions added for lapack-3.10
@lapackobjs2c = (@lapackobjs2c,
cgetsqrhrt,
cungtsqr_row
);
@lapackobjs2d = (@lapackobjs2d,
dgetsqrhrt,
dorgtsqr_row
);
@lapackobjs2s = (@lapackobjs2s,
sgetsqrhrt,
sorgtsqr_row
);
@lapackobjs2z = (@lapackobjs2z,
zgetsqrhrt,
zungtsqr_row
);
#functions added for lapack-3.11
@lapackobjs2c = (@lapackobjs2c,
cgedmd,
cgedmdq
);
@lapackobjs2d = (@lapackobjs2d,
dgedmd,
dgedmdq
);
@lapackobjs2s = (@lapackobjs2s,
sgedmd,
sgedmdq
);
@lapackobjs2z = (@lapackobjs2z,
zgedmd,
zgedmdq
);
#functions added for lapack-3.12
@lapackobjs2c = (@lapackobjs2c,
cgelqs,
cgelst,
cgeqp3rk,
cgeqrs,
claqp2rk,
claqp3rk,
clatrs3,
ctrsyl3,
crscl,
);
@lapackobjs2d = (@lapackobjs2d,
dgelqs,
dgelst,
dgeqp3rk,
dgeqrs,
dlaqp2rk,
dlaqp3rk,
dlatrs3,
dtrsyl3,
dlarmm,
);
@lapackobjs2s = (@lapackobjs2s,
sgelqs,
sgelst,
sgeqp3rk,
sgeqrs,
slaqp2rk,
slaqp3rk,
slatrs3,
strsyl3,
slarmm,
);
@lapackobjs2z = (@lapackobjs2z,
zgelqs,
zgelst,
zgeqp3rk,
zgeqrs,
zlaqp2rk,
zlaqp3rk,
zlatrs3,
ztrsyl3,
zrscl,
);

@lapack_extendedprecision_objs = (
zposvxx, clagge, clatms, chesvxx, cposvxx, cgesvxx, ssyrfssx, csyrfsx,
Expand Down Expand Up @@ -841,7 +921,8 @@ zpotri,
dggsvp,
dlahrd,
dlatzm,
dtzrqf);
dtzrqf,
);

@lapack_deprecated_objss = (
sgelsx,
Expand Down Expand Up @@ -989,6 +1070,10 @@ zpotri,
LAPACKE_cgebrd_work,
LAPACKE_cgecon,
LAPACKE_cgecon_work,
LAPACKE_cgedmd,
LAPACKE_cgedmd_work,
LAPACKE_cgedmdq,
LAPACKE_cgedmdq_work,
LAPACKE_cgeequ,
LAPACKE_cgeequ_work,
LAPACKE_cgeequb,
Expand Down Expand Up @@ -1571,6 +1656,20 @@ zpotri,
LAPACKE_csytrf_aa_2stage_work,
LAPACKE_csytrs_aa_2stage,
LAPACKE_csytrs_aa_2stage_work,
LAPACKE_cgesvdq,
LAPACKE_cgesvdq_work,
LAPACKE_cgetsqrhrt,
LAPACKE_cgetsqrhrt_work,
LAPACKE_cungtsqr_row,
LAPACKE_cungtsqr_row_work,
LAPACKE_clangb,
LAPACKE_clangb_work,
LAPACKE_ctrsyl3,
LAPACKE_ctrsyl3_work,
LAPACKE_ctz_nancheck,
LAPACKE_ctz_trans,
LAPACKE_cunhr_col,
LAPACKE_cunhr_col_work,
);
@lapackeobjsd = (
LAPACKE_dgb_nancheck,
Expand Down Expand Up @@ -1641,6 +1740,10 @@ zpotri,
LAPACKE_dgebrd_work,
LAPACKE_dgecon,
LAPACKE_dgecon_work,
LAPACKE_dgedmd,
LAPACKE_dgedmd_work,
LAPACKE_dgedmdq,
LAPACKE_dgedmdq_work,
LAPACKE_dgeequ,
LAPACKE_dgeequ_work,
LAPACKE_dgeequb,
Expand Down Expand Up @@ -2179,6 +2282,18 @@ zpotri,
LAPACKE_dgesvdq_work,
LAPACKE_slag2d,
LAPACKE_slag2d_work,
LAPACKE_dgetsqrhrt,
LAPACKE_dgetsqrhrt_work,
LAPACKE_dorgtsqr_row,
LAPACKE_dorgtsqr_row_work,
LAPACKE_dlangb,
LAPACKE_dlangb_work,
LAPACKE_dorhr_col,
LAPACKE_dorhr_col_work,
LAPACKE_dtrsyl3,
LAPACKE_dtrsyl3_work,
LAPACKE_dtz_nancheck,
LAPACKE_dtz_trans,
);
@lapackeobjss = (
LAPACKE_sgb_nancheck,
Expand Down Expand Up @@ -2249,6 +2364,10 @@ zpotri,
LAPACKE_sgebrd_work,
LAPACKE_sgecon,
LAPACKE_sgecon_work,
LAPACKE_sgedmd,
LAPACKE_sgedmd_work,
LAPACKE_sgedmdq,
LAPACKE_sgedmdq_work,
LAPACKE_sgeequ,
LAPACKE_sgeequ_work,
LAPACKE_sgeequb,
Expand Down Expand Up @@ -2779,6 +2898,18 @@ zpotri,
LAPACKE_ssytrs_aa_2stage_work,
LAPACKE_sgesvdq,
LAPACKE_sgesvdq_work,
LAPACKE_sgetsqrhrt,
LAPACKE_sgetsqrhrt_work,
LAPACKE_sorgtsqr_row,
LAPACKE_sorgtsqr_row_work,
LAPACKE_slangb,
LAPACKE_slangb_work,
LAPACKE_sorhr_col,
LAPACKE_sorhr_col_work,
LAPACKE_strsyl3,
LAPACKE_strsyl3_work,
LAPACKE_stz_nancheck,
LAPACKE_stz_trans,
);
@lapackeobjsz = (
LAPACKE_zgb_nancheck,
Expand Down Expand Up @@ -2853,6 +2984,10 @@ zpotri,
LAPACKE_zgebrd_work,
LAPACKE_zgecon,
LAPACKE_zgecon_work,
LAPACKE_zgedmd,
LAPACKE_zgedmd_work,
LAPACKE_zgedmdq,
LAPACKE_zgedmdq_work,
LAPACKE_zgeequ,
LAPACKE_zgeequ_work,
LAPACKE_zgeequb,
Expand Down Expand Up @@ -3317,6 +3452,19 @@ zpotri,
LAPACKE_zupmtr_work,
LAPACKE_zsyr,
LAPACKE_zsyr_work,
LAPACKE_zgetsqrhrt,
LAPACKE_zgetsqrhrt_work,
LAPACKE_zungtsqr_row,
LAPACKE_zungtsqr_row_work,
LAPACKE_zlangb,
LAPACKE_zlangb_work,
LAPACKE_ztrsyl3,
LAPACKE_ztrsyl3_work,
LAPACKE_ztz_nancheck,
LAPACKE_ztz_trans,
LAPACKE_zunhr_col,
LAPACKE_zunhr_col_work,

## @(SRCX_OBJ) from `lapack-3.4.1/lapacke/src/Makefile`
## Not exported: requires LAPACKE_EXTENDED to be set and depends on the
## corresponding LAPACK extended precision routines.
Expand Down Expand Up @@ -3544,6 +3692,7 @@ zpotri,
ssysv_aa_2stage, ssytrf_aa_2stage,
ssytrs_aa_2stage,
slaorhr_col_getrfnp, slaorhr_col_getrfnp2, sorhr_col,
slarfb_gett,
);
@lapack_embeded_underscore_objs_c=(
chetf2_rook, chetrf_rook, chetri_rook,
Expand All @@ -3569,6 +3718,7 @@ zpotri,
csysv_aa_2stage, csytrf_aa_2stage,
csytrs_aa_2stage,
claunhr_col_getrfnp, claunhr_col_getrfnp2, cunhr_col,
clarfb_gett,
);
@lapack_embeded_underscore_objs_d=(
dlasyf_rook,
Expand All @@ -3586,6 +3736,7 @@ zpotri,
dsysv_aa_2stage,
dsytrf_aa_2stage, dsytrs_aa_2stage,
dlaorhr_col_getrfnp, dlaorhr_col_getrfnp2, dorhr_col,
dlarfb_gett,
);
@lapack_embeded_underscore_objs_z=(
zhetf2_rook, zhetrf_rook, zhetri_rook,
Expand All @@ -3609,7 +3760,8 @@ zpotri,
zhesv_aa_2stage, zhetrf_aa_2stage,
zhetrs_aa_2stage, zsysv_aa_2stage,
zsytrf_aa_2stage, zsytrs_aa_2stage,
zlaunhr_col_getrfnp, zlaunhr_col_getrfnp2, zunhr_col
zlaunhr_col_getrfnp, zlaunhr_col_getrfnp2, zunhr_col,
zlarfb_gett,
);


Expand Down Expand Up @@ -3930,6 +4082,18 @@ if ($ARGV[0] eq "linktest") {
@no_underscore_objs = (@no_underscore_objs, @misc_common_objs);

print "int main(void){\n";
foreach $objs (@underscore_objs) {
print "extern void ", $symbolprefix, $objs, $bu, $symbolsuffix, "();\n" if $objs ne "xerbla";
}

foreach $objs (@need_2underscore_objs) {
print "extern void ", $symbolprefix, $objs, $bu, $bu, $symbolsuffix, "();\n";
}

foreach $objs (@no_underscore_objs) {
print "extern void ", $symbolprefix, $objs, $symbolsuffix, "();\n";
}

foreach $objs (@underscore_objs) {
print $symbolprefix, $objs, $bu, $symbolsuffix, "();\n" if $objs ne "xerbla";
}
Expand Down
Loading

0 comments on commit 3c451b9

Please sign in to comment.