Re: [Qemu-devel] [PULL 23/44] target/ppc: Use vector variable shifts for VSL, VSR, VSRA

[Laurent Vivier <lvivier@redhat.com>]

On 07/06/2019 11:29, Laurent Vivier wrote:
> On 29/05/2019 08:49, David Gibson wrote:
>> From: Richard Henderson <richard.henderson@linaro.org>
>>
>> The gvec expanders take care of masking the shift amount
>> against the element width.
>>
>> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
>> Message-Id: <20190518191430.21686-2-richard.henderson@linaro.org>
>> Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
>> ---
>>  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