[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v6 18/61] target/riscv: vector single-width integer multiply
From: |
LIU Zhiwei |
Subject: |
Re: [PATCH v6 18/61] target/riscv: vector single-width integer multiply instructions |
Date: |
Sun, 29 Mar 2020 12:00:37 +0800 |
User-agent: |
Mozilla/5.0 (Windows NT 10.0; WOW64; rv:68.0) Gecko/20100101 Thunderbird/68.6.0 |
On 2020/3/29 0:13, LIU Zhiwei wrote:
On 2020/3/28 23:47, Richard Henderson wrote:
On 3/28/20 8:17 AM, LIU Zhiwei wrote:
Missed the improvement here. See
tcg_gen_mulsu2_i64.
Though I have not gotten the principle, the code in
tcg_gen_mulsu2_i64 is much
tidier.
Let A = signed operand,
B = unsigned operand
P = unsigned product
If the sign bit A is set, then P is too large.
In that case we subtract 2**64 * B to fix that:
HI_P -= (A < 0 ? B : 0)
where the conditional is computed as (A >> 63) & B.
I think I get it.
LET A = 2 ** 64 - X
THEN
X = 2 ** 64 - A
SIGNED_P = -X * B
if (A * B == P) then
(2 ** 64 - X) * B == P
2 **64 * B - X * B == P
-X *B == P - 2**64*B
HI_P -= (A < 0 ? B :0)
It's confusing here. I paste the clearer code.
/*
* Let A = signed operand,
* B = unsigned operand
* P = mulu64(A, B), unsigned product
*
* LET X = 2 ** 64 - A, 2's complement of A
* SP = signed product
* THEN
* IF A < 0
* SP = -X * B
* = -(2 ** 64 - A) * B
* = A * B - 2 ** 64 * B
* = P - 2 ** 64 * B
* ELSE
* SP = P
* THEN
* HI_P -= (A < 0 ? B : 0)
*/
static int64_t do_mulhsu_d(int64_t s2, uint64_t s1)
{
uint64_t hi_64, lo_64;
mulu64(&lo_64, &hi_64, s2, s1);
hi_64 -= s2 < 0 ? s1 : 0;
return hi_64;
}
Zhiwei
Zhiwei
r~
- [PATCH v6 16/61] target/riscv: vector integer comparison instructions, (continued)
[PATCH v6 19/61] target/riscv: vector integer divide instructions, LIU Zhiwei, 2020/03/17
[PATCH v6 20/61] target/riscv: vector widening integer multiply instructions, LIU Zhiwei, 2020/03/17
[PATCH v6 21/61] target/riscv: vector single-width integer multiply-add instructions, LIU Zhiwei, 2020/03/17
[PATCH v6 22/61] target/riscv: vector widening integer multiply-add instructions, LIU Zhiwei, 2020/03/17
[PATCH v6 23/61] target/riscv: vector integer merge and move instructions, LIU Zhiwei, 2020/03/17