diff options
| author | Tom Musta <tommusta@gmail.com> | 2014-02-12 15:23:00 -0600 |
|---|---|---|
| committer | Alexander Graf <agraf@suse.de> | 2014-03-05 03:06:54 +0100 |
| commit | aa9e930c8870d06a20b356785d3ec7d9a942a29f (patch) | |
| tree | 2bb553f66befa7e3d5809386847b53baf994d967 /target-ppc | |
| parent | 56eabc750862b985a6ddfc3905b534576eeee33e (diff) | |
target-ppc: Altivec 2.07: Change VMUL_DO to Support 64-bit Integers
This VMUL_DO macro provides support for the various vmule* and vmulo*
instructions. These instructions multiply vector elements, producing
products that are one size larger; e.g. vmuleub multiplies unsigned 8-bit
elements and produces a 16 bit unsigned element.
The existing macro works correctly for the existing instructions (8-bit,
and 16-bit source elements) but does not work correctly for 32-bit
source elements.
This patch adds an explicit cast to the multiplicands, forcing them to be
of the target element type. This is required for the forthcoming patches
that add the vmul[eo][us]w instructions.
Signed-off-by: Tom Musta <tommusta@gmail.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Diffstat (limited to 'target-ppc')
| -rw-r--r-- | target-ppc/int_helper.c | 26 |
1 files changed, 14 insertions, 12 deletions
diff --git a/target-ppc/int_helper.c b/target-ppc/int_helper.c index 3e36c0aa3c..20d34e6231 100644 --- a/target-ppc/int_helper.c +++ b/target-ppc/int_helper.c @@ -983,28 +983,30 @@ void helper_vmsumuhs(CPUPPCState *env, ppc_avr_t *r, ppc_avr_t *a, } } -#define VMUL_DO(name, mul_element, prod_element, evenp) \ +#define VMUL_DO(name, mul_element, prod_element, cast, evenp) \ void helper_v##name(ppc_avr_t *r, ppc_avr_t *a, ppc_avr_t *b) \ { \ int i; \ \ VECTOR_FOR_INORDER_I(i, prod_element) { \ if (evenp) { \ - r->prod_element[i] = a->mul_element[i * 2 + HI_IDX] * \ - b->mul_element[i * 2 + HI_IDX]; \ + r->prod_element[i] = \ + (cast)a->mul_element[i * 2 + HI_IDX] * \ + (cast)b->mul_element[i * 2 + HI_IDX]; \ } else { \ - r->prod_element[i] = a->mul_element[i * 2 + LO_IDX] * \ - b->mul_element[i * 2 + LO_IDX]; \ + r->prod_element[i] = \ + (cast)a->mul_element[i * 2 + LO_IDX] * \ + (cast)b->mul_element[i * 2 + LO_IDX]; \ } \ } \ } -#define VMUL(suffix, mul_element, prod_element) \ - VMUL_DO(mule##suffix, mul_element, prod_element, 1) \ - VMUL_DO(mulo##suffix, mul_element, prod_element, 0) -VMUL(sb, s8, s16) -VMUL(sh, s16, s32) -VMUL(ub, u8, u16) -VMUL(uh, u16, u32) +#define VMUL(suffix, mul_element, prod_element, cast) \ + VMUL_DO(mule##suffix, mul_element, prod_element, cast, 1) \ + VMUL_DO(mulo##suffix, mul_element, prod_element, cast, 0) +VMUL(sb, s8, s16, int16_t) +VMUL(sh, s16, s32, int32_t) +VMUL(ub, u8, u16, uint16_t) +VMUL(uh, u16, u32, uint32_t) #undef VMUL_DO #undef VMUL |