mirrors/qemu
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

target/arm: Macroize helper_gvec_{s,u}dot_idx_{b,h}

Browse Source 
We're about to add more variations on this theme.
Accept the inner loop for the _h variants, rather
than keep it unrolled.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20210525010358.152808-66-richard.henderson@linaro.org
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
This commit is contained in:
Richard Henderson 2021-05-24 18:03:31 -07:00 committed by Peter Maydell
parent 5c57e3b954
commit 7020ffd656
1 changed files with 29 additions and 131 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

160
target/arm/vec_helper.c
View File

@ -569,139 +569,37 @@ DO_DOT(gvec_udot_b, uint32_t, uint8_t, uint8_t)
DO_DOT(gvec_sdot_h, int64_t, int16_t, int16_t) DO_DOT(gvec_sdot_h, int64_t, int16_t, int16_t)
DO_DOT(gvec_udot_h, uint64_t, uint16_t, uint16_t) DO_DOT(gvec_udot_h, uint64_t, uint16_t, uint16_t)
void HELPER(gvec_sdot_idx_b)(void *vd, void *vn, void *vm, #define DO_DOT_IDX(NAME, TYPED, TYPEN, TYPEM, HD) \
void *va, uint32_t desc) void HELPER(NAME)(void *vd, void *vn, void *vm, void *va, uint32_t desc) \
{ { \
intptr_t i, segend, opr_sz = simd_oprsz(desc), opr_sz_4 = opr_sz / 4; intptr_t i = 0, opr_sz = simd_oprsz(desc); \
intptr_t index = simd_data(desc); intptr_t opr_sz_n = opr_sz / sizeof(TYPED); \
int32_t *d = vd, *a = va; intptr_t segend = MIN(16 / sizeof(TYPED), opr_sz_n); \
int8_t *n = vn; intptr_t index = simd_data(desc); \
int8_t *m_indexed = (int8_t *)vm + H4(index) * 4; TYPED *d = vd, *a = va; \
TYPEN *n = vn; \
/* Notice the special case of opr_sz == 8, from aa64/aa32 advsimd. TYPEM *m_indexed = (TYPEM *)vm + HD(index) * 4; \
* Otherwise opr_sz is a multiple of 16. do { \
*/ TYPED m0 = m_indexed[i * 4 + 0]; \
segend = MIN(4, opr_sz_4); TYPED m1 = m_indexed[i * 4 + 1]; \
i = 0; TYPED m2 = m_indexed[i * 4 + 2]; \
do { TYPED m3 = m_indexed[i * 4 + 3]; \
int8_t m0 = m_indexed[i * 4 + 0]; do { \
int8_t m1 = m_indexed[i * 4 + 1]; d[i] = (a[i] + \
int8_t m2 = m_indexed[i * 4 + 2]; n[i * 4 + 0] * m0 + \
int8_t m3 = m_indexed[i * 4 + 3]; n[i * 4 + 1] * m1 + \
n[i * 4 + 2] * m2 + \
do { n[i * 4 + 3] * m3); \
d[i] = (a[i] + } while (++i < segend); \
n[i * 4 + 0] * m0 + segend = i + 4; \
n[i * 4 + 1] * m1 + } while (i < opr_sz_n); \
n[i * 4 + 2] * m2 + clear_tail(d, opr_sz, simd_maxsz(desc)); \
n[i * 4 + 3] * m3);
} while (++i < segend);
segend = i + 4;
} while (i < opr_sz_4);
clear_tail(d, opr_sz, simd_maxsz(desc));
} }
void HELPER(gvec_udot_idx_b)(void *vd, void *vn, void *vm, DO_DOT_IDX(gvec_sdot_idx_b, int32_t, int8_t, int8_t, H4)
void *va, uint32_t desc) DO_DOT_IDX(gvec_udot_idx_b, uint32_t, uint8_t, uint8_t, H4)
{ DO_DOT_IDX(gvec_sdot_idx_h, int64_t, int16_t, int16_t, )
intptr_t i, segend, opr_sz = simd_oprsz(desc), opr_sz_4 = opr_sz / 4; DO_DOT_IDX(gvec_udot_idx_h, uint64_t, uint16_t, uint16_t, )
intptr_t index = simd_data(desc);
uint32_t *d = vd, *a = va;
uint8_t *n = vn;
uint8_t *m_indexed = (uint8_t *)vm + H4(index) * 4;
/* Notice the special case of opr_sz == 8, from aa64/aa32 advsimd.
* Otherwise opr_sz is a multiple of 16.
*/
segend = MIN(4, opr_sz_4);
i = 0;
do {
uint8_t m0 = m_indexed[i * 4 + 0];
uint8_t m1 = m_indexed[i * 4 + 1];
uint8_t m2 = m_indexed[i * 4 + 2];
uint8_t m3 = m_indexed[i * 4 + 3];
do {
d[i] = (a[i] +
n[i * 4 + 0] * m0 +
n[i * 4 + 1] * m1 +
n[i * 4 + 2] * m2 +
n[i * 4 + 3] * m3);
} while (++i < segend);
segend = i + 4;
} while (i < opr_sz_4);
clear_tail(d, opr_sz, simd_maxsz(desc));
}
void HELPER(gvec_sdot_idx_h)(void *vd, void *vn, void *vm,
void *va, uint32_t desc)
{
intptr_t i, opr_sz = simd_oprsz(desc), opr_sz_8 = opr_sz / 8;
intptr_t index = simd_data(desc);
int64_t *d = vd, *a = va;
int16_t *n = vn;
int16_t *m_indexed = (int16_t *)vm + index * 4;
/* This is supported by SVE only, so opr_sz is always a multiple of 16.
* Process the entire segment all at once, writing back the results
* only after we've consumed all of the inputs.
*/
for (i = 0; i < opr_sz_8; i += 2) {
int64_t d0, d1;
d0 = a[i + 0];
d0 += n[i * 4 + 0] * (int64_t)m_indexed[i * 4 + 0];
d0 += n[i * 4 + 1] * (int64_t)m_indexed[i * 4 + 1];
d0 += n[i * 4 + 2] * (int64_t)m_indexed[i * 4 + 2];
d0 += n[i * 4 + 3] * (int64_t)m_indexed[i * 4 + 3];
d1 = a[i + 1];
d1 += n[i * 4 + 4] * (int64_t)m_indexed[i * 4 + 0];
d1 += n[i * 4 + 5] * (int64_t)m_indexed[i * 4 + 1];
d1 += n[i * 4 + 6] * (int64_t)m_indexed[i * 4 + 2];
d1 += n[i * 4 + 7] * (int64_t)m_indexed[i * 4 + 3];
d[i + 0] = d0;
d[i + 1] = d1;
}
clear_tail(d, opr_sz, simd_maxsz(desc));
}
void HELPER(gvec_udot_idx_h)(void *vd, void *vn, void *vm,
void *va, uint32_t desc)
{
intptr_t i, opr_sz = simd_oprsz(desc), opr_sz_8 = opr_sz / 8;
intptr_t index = simd_data(desc);
uint64_t *d = vd, *a = va;
uint16_t *n = vn;
uint16_t *m_indexed = (uint16_t *)vm + index * 4;
/* This is supported by SVE only, so opr_sz is always a multiple of 16.
* Process the entire segment all at once, writing back the results
* only after we've consumed all of the inputs.
*/
for (i = 0; i < opr_sz_8; i += 2) {
uint64_t d0, d1;
d0 = a[i + 0];
d0 += n[i * 4 + 0] * (uint64_t)m_indexed[i * 4 + 0];
d0 += n[i * 4 + 1] * (uint64_t)m_indexed[i * 4 + 1];
d0 += n[i * 4 + 2] * (uint64_t)m_indexed[i * 4 + 2];
d0 += n[i * 4 + 3] * (uint64_t)m_indexed[i * 4 + 3];
d1 = a[i + 1];
d1 += n[i * 4 + 4] * (uint64_t)m_indexed[i * 4 + 0];
d1 += n[i * 4 + 5] * (uint64_t)m_indexed[i * 4 + 1];
d1 += n[i * 4 + 6] * (uint64_t)m_indexed[i * 4 + 2];
d1 += n[i * 4 + 7] * (uint64_t)m_indexed[i * 4 + 3];
d[i + 0] = d0;
d[i + 1] = d1;
}
clear_tail(d, opr_sz, simd_maxsz(desc));
}
void HELPER(gvec_fcaddh)(void *vd, void *vn, void *vm, void HELPER(gvec_fcaddh)(void *vd, void *vn, void *vm,
void *vfpst, uint32_t desc) void *vfpst, uint32_t desc)