Skip to content

add riscv level3 C,Z kernel functions. #3893

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Mar 15, 2023
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
85 changes: 61 additions & 24 deletions kernel/riscv64/KERNEL.x280
Original file line number Diff line number Diff line change
Expand Up @@ -118,8 +118,8 @@ DGEMVTKERNEL = gemv_t_rvv.c
CGEMVTKERNEL = zgemv_t_rvv.c
ZGEMVTKERNEL = zgemv_t_rvv.c

CTRMMKERNEL = ztrmmkernel_2x2_rvv.c
ZTRMMKERNEL = ztrmmkernel_2x2_rvv.c
CTRMMKERNEL = ztrmmkernel_rvv_v1x4.c
ZTRMMKERNEL = ztrmmkernel_rvv_v1x4.c

# SGEMM_UNROLL_N set in params.h
ifeq ($(SGEMM_UNROLL_N), 8)
Expand Down Expand Up @@ -168,17 +168,28 @@ DSYMMUCOPY_M = symm_ucopy_rvv_v1.c
DSYMMLCOPY_M = symm_lcopy_rvv_v1.c
endif

CGEMMKERNEL = ../generic/zgemmkernel_2x2.c
CGEMMONCOPY = ../generic/zgemm_ncopy_2.c
CGEMMOTCOPY = ../generic/zgemm_tcopy_2.c
CGEMMONCOPYOBJ = cgemm_oncopy.o
CGEMMOTCOPYOBJ = cgemm_otcopy.o
CGEMMKERNEL = zgemmkernel_rvv_v1x4.c
CGEMMINCOPY = zgemm_ncopy_rvv_v1.c
CGEMMITCOPY = zgemm_tcopy_rvv_v1.c
CGEMMONCOPY = zgemm_ncopy_4_rvv.c
CGEMMOTCOPY = zgemm_tcopy_4_rvv.c

ZGEMMKERNEL = ../generic/zgemmkernel_2x2.c
ZGEMMONCOPY = ../generic/zgemm_ncopy_2.c
ZGEMMOTCOPY = ../generic/zgemm_tcopy_2.c
ZGEMMONCOPYOBJ = zgemm_oncopy.o
ZGEMMOTCOPYOBJ = zgemm_otcopy.o
CGEMMINCOPYOBJ = cgemm_incopy$(TSUFFIX).$(SUFFIX)
CGEMMITCOPYOBJ = cgemm_itcopy$(TSUFFIX).$(SUFFIX)
CGEMMONCOPYOBJ = cgemm_oncopy$(TSUFFIX).$(SUFFIX)
CGEMMOTCOPYOBJ = cgemm_otcopy$(TSUFFIX).$(SUFFIX)

ZGEMMKERNEL = zgemmkernel_rvv_v1x4.c

ZGEMMINCOPY = zgemm_ncopy_rvv_v1.c
ZGEMMITCOPY = zgemm_tcopy_rvv_v1.c
ZGEMMONCOPY = zgemm_ncopy_4_rvv.c
ZGEMMOTCOPY = zgemm_tcopy_4_rvv.c

ZGEMMINCOPYOBJ = zgemm_incopy$(TSUFFIX).$(SUFFIX)
ZGEMMITCOPYOBJ = zgemm_itcopy$(TSUFFIX).$(SUFFIX)
ZGEMMONCOPYOBJ = zgemm_oncopy$(TSUFFIX).$(SUFFIX)
ZGEMMOTCOPYOBJ = zgemm_otcopy$(TSUFFIX).$(SUFFIX)

STRSMKERNEL_LN = trsm_kernel_LN_rvv_v1.c
STRSMKERNEL_LT = trsm_kernel_LT_rvv_v1.c
Expand All @@ -190,20 +201,25 @@ DTRSMKERNEL_LT = trsm_kernel_LT_rvv_v1.c
DTRSMKERNEL_RN = trsm_kernel_RN_rvv_v1.c
DTRSMKERNEL_RT = trsm_kernel_RT_rvv_v1.c

CTRSMKERNEL_LN = ../generic/trsm_kernel_LN.c
CTRSMKERNEL_LT = ../generic/trsm_kernel_LT.c
CTRSMKERNEL_RN = ../generic/trsm_kernel_RN.c
CTRSMKERNEL_RT = ../generic/trsm_kernel_RT.c
CTRSMKERNEL_LN = trsm_kernel_LN_rvv_v1.c
CTRSMKERNEL_LT = trsm_kernel_LT_rvv_v1.c
CTRSMKERNEL_RN = trsm_kernel_RN_rvv_v1.c
CTRSMKERNEL_RT = trsm_kernel_RT_rvv_v1.c

ZTRSMKERNEL_LN = ../generic/trsm_kernel_LN.c
ZTRSMKERNEL_LT = ../generic/trsm_kernel_LT.c
ZTRSMKERNEL_RN = ../generic/trsm_kernel_RN.c
ZTRSMKERNEL_RT = ../generic/trsm_kernel_RT.c
ZTRSMKERNEL_LN = trsm_kernel_LN_rvv_v1.c
ZTRSMKERNEL_LT = trsm_kernel_LT_rvv_v1.c
ZTRSMKERNEL_RN = trsm_kernel_RN_rvv_v1.c
ZTRSMKERNEL_RT = trsm_kernel_RT_rvv_v1.c

TRSMCOPYLN_M = trsm_lncopy_rvv_v1.c
TRSMCOPYLT_M = trsm_ltcopy_rvv_v1.c
TRSMCOPYUN_M = trsm_uncopy_rvv_v1.c
TRSMCOPYUT_M = trsm_utcopy_rvv_v1.c
TRSMCOPYLN_M = trsm_lncopy_rvv_v1.c
TRSMCOPYLT_M = trsm_ltcopy_rvv_v1.c
TRSMCOPYUN_M = trsm_uncopy_rvv_v1.c
TRSMCOPYUT_M = trsm_utcopy_rvv_v1.c

ZTRSMCOPYLN_M = ztrsm_lncopy_rvv_v1.c
ZTRSMCOPYLT_M = ztrsm_ltcopy_rvv_v1.c
ZTRSMCOPYUN_M = ztrsm_uncopy_rvv_v1.c
ZTRSMCOPYUT_M = ztrsm_utcopy_rvv_v1.c

SSYMV_U_KERNEL = symv_U_rvv.c
SSYMV_L_KERNEL = symv_L_rvv.c
Expand All @@ -214,6 +230,27 @@ CSYMV_L_KERNEL = ../generic/zsymv_k.c
ZSYMV_U_KERNEL = ../generic/zsymv_k.c
ZSYMV_L_KERNEL = ../generic/zsymv_k.c

ZHEMMLTCOPY_M = zhemm_ltcopy_rvv_v1.c
ZHEMMUTCOPY_M = zhemm_utcopy_rvv_v1.c

CHEMMLTCOPY_M = zhemm_ltcopy_rvv_v1.c
CHEMMUTCOPY_M = zhemm_utcopy_rvv_v1.c

ZSYMMUCOPY_M = zsymm_ucopy_rvv_v1.c
ZSYMMLCOPY_M = zsymm_lcopy_rvv_v1.c

CSYMMUCOPY_M = zsymm_ucopy_rvv_v1.c
CSYMMLCOPY_M = zsymm_lcopy_rvv_v1.c

ZTRMMUNCOPY_M = ztrmm_uncopy_rvv_v1.c
ZTRMMLNCOPY_M = ztrmm_lncopy_rvv_v1.c
ZTRMMUTCOPY_M = ztrmm_utcopy_rvv_v1.c
ZTRMMLTCOPY_M = ztrmm_ltcopy_rvv_v1.c

CTRMMUNCOPY_M = ztrmm_uncopy_rvv_v1.c
CTRMMLNCOPY_M = ztrmm_lncopy_rvv_v1.c
CTRMMUTCOPY_M = ztrmm_utcopy_rvv_v1.c
CTRMMLTCOPY_M = ztrmm_ltcopy_rvv_v1.c

LSAME_KERNEL = ../generic/lsame.c

Expand Down
8 changes: 4 additions & 4 deletions kernel/riscv64/trmm_lncopy_rvv_v1.c
Original file line number Diff line number Diff line change
Expand Up @@ -36,10 +36,10 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#define VSEV_FLOAT vse32_v_f32m2
#define VLSEV_FLOAT vlse32_v_f32m2
#define VBOOL_T vbool16_t
#define UINT_V_T vint32m2_t
#define VID_V_UINT vid_v_i32m2
#define VMSGTU_VX_UINT vmsgt_vx_i32m2_b16
#define VMSEQ_VX_UINT vmseq_vx_i32m2_b16
#define UINT_V_T vuint32m2_t
#define VID_V_UINT vid_v_u32m2
#define VMSGTU_VX_UINT vmsgtu_vx_u32m2_b16
#define VMSEQ_VX_UINT vmseq_vx_u32m2_b16
#define VFMERGE_VFM_FLOAT vfmerge_vfm_f32m2
#else
#define VSETVL(n) vsetvl_e64m2(n)
Expand Down
Loading