[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-ppc] [PULL 23/44] target/ppc: Use vector variable shifts for V
From: |
Laurent Vivier |
Subject: |
Re: [Qemu-ppc] [PULL 23/44] target/ppc: Use vector variable shifts for VSL, VSR, VSRA |
Date: |
Fri, 7 Jun 2019 16:09:31 +0200 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.8.0 |
On 07/06/2019 11:29, Laurent Vivier wrote:
> On 29/05/2019 08:49, David Gibson wrote:
>> From: Richard Henderson <address@hidden>
>>
>> The gvec expanders take care of masking the shift amount
>> against the element width.
>>
>> Signed-off-by: Richard Henderson <address@hidden>
>> Message-Id: <address@hidden>
>> Signed-off-by: David Gibson <address@hidden>
>> ---
>> target/ppc/helper.h | 12 ----------
>> target/ppc/int_helper.c | 37 -----------------------------
>> target/ppc/translate/vmx-impl.inc.c | 24 +++++++++----------
>> 3 files changed, 12 insertions(+), 61 deletions(-)
>
> This patch introduces a regressions
> with Fedora 29 guest:
>
> - during kernel boot:
>
> [ 40.397876] crypto_register_alg 'aes' = 0
> [ 40.577517] crypto_register_alg 'cbc(aes)' = 0
> [ 40.743576] crypto_register_alg 'ctr(aes)' = 0
> [ 41.061379] alg: skcipher: Test 1 failed (invalid result) on encryption
> for p8_aes_xts
> [ 41.062054] 00000000: 91 7c f6 9e bd 68 b2 ec 9b 9f e9 a3 ea dd a6 92
> [ 41.062163] 00000010: 98 10 35 57 5e dc 36 1e 9a f7 bc ba 39 f2 5c eb
> [ 41.062834] crypto_register_alg 'xts(aes)' = 0
> [ 41.077358] alg: hash: Test 2 failed for p8_ghash
> [ 41.077553] 00000000: 5f 89 ab f7 20 57 20 57 20 57 20 57 20 57 20 57
>
> - with libssl:
>
> # curl -o /dev/null https://www.google.com
> % Total % Received % Xferd Average Speed Time Time Time
> Current
> Dload Upload Total Spent Left Speed
> 0 0 0 0 0 0 0 0 --:--:-- --:--:-- --:--:-- 0
> curl: (35) error:1408F119:SSL routines:ssl3_get_record:decryption failed or
> bad record mac
>
> [before, this one fails with:
> curl: (35) error:04091068:rsa routines:int_rsa_verify:bad signature ]
>
> If I revert this patch on top of 0d74f3b427 + "target/ppc: Fix lxvw4x, lxvh8x
> and lxvb16x", all works fine.
>
> Thanks,
> Laurent
>
This seems to fix the problem:
diff --git a/accel/tcg/tcg-runtime-gvec.c b/accel/tcg/tcg-runtime-gvec.c
index 3b6052fe97..6f0709b307 100644
--- a/accel/tcg/tcg-runtime-gvec.c
+++ b/accel/tcg/tcg-runtime-gvec.c
@@ -874,7 +874,7 @@ void HELPER(gvec_sar8v)(void *d, void *a, void *b,
uint32_t desc)
intptr_t oprsz = simd_oprsz(desc);
intptr_t i;
- for (i = 0; i < oprsz; i += sizeof(vec8)) {
+ for (i = 0; i < oprsz; i += sizeof(uint8_t)) {
uint8_t sh = *(uint8_t *)(b + i) & 7;
*(int8_t *)(d + i) = *(int8_t *)(a + i) >> sh;
}
Thanks,
Laurent
- Re: [Qemu-ppc] [PULL 23/44] target/ppc: Use vector variable shifts for VSL, VSR, VSRA, Laurent Vivier, 2019/06/07
- Re: [Qemu-ppc] [PULL 23/44] target/ppc: Use vector variable shifts for VSL, VSR, VSRA,
Laurent Vivier <=
- Re: [Qemu-ppc] [PULL 23/44] target/ppc: Use vector variable shifts for VSL, VSR, VSRA, Richard Henderson, 2019/06/07
- Re: [Qemu-ppc] [PULL 23/44] target/ppc: Use vector variable shifts for VSL, VSR, VSRA, David Gibson, 2019/06/10
- Re: [Qemu-ppc] [PULL 23/44] target/ppc: Use vector variable shifts for VSL, VSR, VSRA, Laurent Vivier, 2019/06/11
- Re: [Qemu-ppc] [PULL 23/44] target/ppc: Use vector variable shifts for VSL, VSR, VSRA, Laurent Vivier, 2019/06/11