motorhead/qemu
1
0
Fork 0
mirror of https://github.com/Motorhead1991/qemu.git synced 2025-08-05 16:53:55 -06:00
Code Issues Projects Releases Packages Wiki Activity Actions

target/loongarch: Implement vaddw/vsubw

Browse source 
This patch includes:
- VADDW{EV/OD}.{H.B/W.H/D.W/Q.D}[U];
- VSUBW{EV/OD}.{H.B/W.H/D.W/Q.D}[U];
- VADDW{EV/OD}.{H.BU.B/W.HU.H/D.WU.W/Q.DU.D}.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Song Gao <gaosong@loongson.cn>
Message-Id: <20230504122810.4094787-10-gaosong@loongson.cn>
This commit is contained in:
Song Gao 2023-05-04 20:27:35 +08:00
parent c037fbc97d
commit 2d5f950c05
No known key found for this signature in database
GPG key ID: 40A2FFF239263EDF
5 changed files with 1116 additions and 0 deletions
Download patch file Download diff file Expand all files Collapse all files

190
target/loongarch/lsx_helper.c
View file

@ -85,3 +85,193 @@ void HELPER(vhsubw_qu_du)(CPULoongArchState *env,
Vd->Q(0) = int128_sub(int128_make64((uint64_t)Vj->D(1)),
int128_make64((uint64_t)Vk->D(0)));
}
#define DO_EVEN(NAME, BIT, E1, E2, DO_OP) \
void HELPER(NAME)(void *vd, void *vj, void *vk, uint32_t v) \
{ \
int i; \
VReg *Vd = (VReg *)vd; \
VReg *Vj = (VReg *)vj; \
VReg *Vk = (VReg *)vk; \
typedef __typeof(Vd->E1(0)) TD; \
for (i = 0; i < LSX_LEN/BIT; i++) { \
Vd->E1(i) = DO_OP((TD)Vj->E2(2 * i) ,(TD)Vk->E2(2 * i)); \
} \
}
#define DO_ODD(NAME, BIT, E1, E2, DO_OP) \
void HELPER(NAME)(void *vd, void *vj, void *vk, uint32_t v) \
{ \
int i; \
VReg *Vd = (VReg *)vd; \
VReg *Vj = (VReg *)vj; \
VReg *Vk = (VReg *)vk; \
typedef __typeof(Vd->E1(0)) TD; \
for (i = 0; i < LSX_LEN/BIT; i++) { \
Vd->E1(i) = DO_OP((TD)Vj->E2(2 * i + 1), (TD)Vk->E2(2 * i + 1)); \
} \
}
void HELPER(vaddwev_q_d)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_add(int128_makes64(Vj->D(0)), int128_makes64(Vk->D(0)));
}
DO_EVEN(vaddwev_h_b, 16, H, B, DO_ADD)
DO_EVEN(vaddwev_w_h, 32, W, H, DO_ADD)
DO_EVEN(vaddwev_d_w, 64, D, W, DO_ADD)
void HELPER(vaddwod_q_d)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_add(int128_makes64(Vj->D(1)), int128_makes64(Vk->D(1)));
}
DO_ODD(vaddwod_h_b, 16, H, B, DO_ADD)
DO_ODD(vaddwod_w_h, 32, W, H, DO_ADD)
DO_ODD(vaddwod_d_w, 64, D, W, DO_ADD)
void HELPER(vsubwev_q_d)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_sub(int128_makes64(Vj->D(0)), int128_makes64(Vk->D(0)));
}
DO_EVEN(vsubwev_h_b, 16, H, B, DO_SUB)
DO_EVEN(vsubwev_w_h, 32, W, H, DO_SUB)
DO_EVEN(vsubwev_d_w, 64, D, W, DO_SUB)
void HELPER(vsubwod_q_d)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_sub(int128_makes64(Vj->D(1)), int128_makes64(Vk->D(1)));
}
DO_ODD(vsubwod_h_b, 16, H, B, DO_SUB)
DO_ODD(vsubwod_w_h, 32, W, H, DO_SUB)
DO_ODD(vsubwod_d_w, 64, D, W, DO_SUB)
void HELPER(vaddwev_q_du)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_add(int128_make64((uint64_t)Vj->D(0)),
int128_make64((uint64_t)Vk->D(0)));
}
DO_EVEN(vaddwev_h_bu, 16, UH, UB, DO_ADD)
DO_EVEN(vaddwev_w_hu, 32, UW, UH, DO_ADD)
DO_EVEN(vaddwev_d_wu, 64, UD, UW, DO_ADD)
void HELPER(vaddwod_q_du)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_add(int128_make64((uint64_t)Vj->D(1)),
int128_make64((uint64_t)Vk->D(1)));
}
DO_ODD(vaddwod_h_bu, 16, UH, UB, DO_ADD)
DO_ODD(vaddwod_w_hu, 32, UW, UH, DO_ADD)
DO_ODD(vaddwod_d_wu, 64, UD, UW, DO_ADD)
void HELPER(vsubwev_q_du)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_sub(int128_make64((uint64_t)Vj->D(0)),
int128_make64((uint64_t)Vk->D(0)));
}
DO_EVEN(vsubwev_h_bu, 16, UH, UB, DO_SUB)
DO_EVEN(vsubwev_w_hu, 32, UW, UH, DO_SUB)
DO_EVEN(vsubwev_d_wu, 64, UD, UW, DO_SUB)
void HELPER(vsubwod_q_du)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_sub(int128_make64((uint64_t)Vj->D(1)),
int128_make64((uint64_t)Vk->D(1)));
}
DO_ODD(vsubwod_h_bu, 16, UH, UB, DO_SUB)
DO_ODD(vsubwod_w_hu, 32, UW, UH, DO_SUB)
DO_ODD(vsubwod_d_wu, 64, UD, UW, DO_SUB)
#define DO_EVEN_U_S(NAME, BIT, ES1, EU1, ES2, EU2, DO_OP) \
void HELPER(NAME)(void *vd, void *vj, void *vk, uint32_t v) \
{ \
int i; \
VReg *Vd = (VReg *)vd; \
VReg *Vj = (VReg *)vj; \
VReg *Vk = (VReg *)vk; \
typedef __typeof(Vd->ES1(0)) TDS; \
typedef __typeof(Vd->EU1(0)) TDU; \
for (i = 0; i < LSX_LEN/BIT; i++) { \
Vd->ES1(i) = DO_OP((TDU)Vj->EU2(2 * i) ,(TDS)Vk->ES2(2 * i)); \
} \
}
#define DO_ODD_U_S(NAME, BIT, ES1, EU1, ES2, EU2, DO_OP) \
void HELPER(NAME)(void *vd, void *vj, void *vk, uint32_t v) \
{ \
int i; \
VReg *Vd = (VReg *)vd; \
VReg *Vj = (VReg *)vj; \
VReg *Vk = (VReg *)vk; \
typedef __typeof(Vd->ES1(0)) TDS; \
typedef __typeof(Vd->EU1(0)) TDU; \
for (i = 0; i < LSX_LEN/BIT; i++) { \
Vd->ES1(i) = DO_OP((TDU)Vj->EU2(2 * i + 1), (TDS)Vk->ES2(2 * i + 1)); \
} \
}
void HELPER(vaddwev_q_du_d)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_add(int128_make64((uint64_t)Vj->D(0)),
int128_makes64(Vk->D(0)));
}
DO_EVEN_U_S(vaddwev_h_bu_b, 16, H, UH, B, UB, DO_ADD)
DO_EVEN_U_S(vaddwev_w_hu_h, 32, W, UW, H, UH, DO_ADD)
DO_EVEN_U_S(vaddwev_d_wu_w, 64, D, UD, W, UW, DO_ADD)
void HELPER(vaddwod_q_du_d)(void *vd, void *vj, void *vk, uint32_t v)
{
VReg *Vd = (VReg *)vd;
VReg *Vj = (VReg *)vj;
VReg *Vk = (VReg *)vk;
Vd->Q(0) = int128_add(int128_make64((uint64_t)Vj->D(1)),
int128_makes64(Vk->D(1)));
}
DO_ODD_U_S(vaddwod_h_bu_b, 16, H, UH, B, UB, DO_ADD)
DO_ODD_U_S(vaddwod_w_hu_h, 32, W, UW, H, UH, DO_ADD)
DO_ODD_U_S(vaddwod_d_wu_w, 64, D, UD, W, UW, DO_ADD)