Dmitry Belyavskiy

> I wonder if it's in any way related to the cause of this: [randombit/botan#2802](https://github.com/randombit/botan/issues/2802) (compiler bug)? I believe the test_keyexpimp failure may be related, but I thinkr @levitte's hypothesis...

Will try to play with the new clang on Friday

Using clang 12.0.1, I get segfault on digest-with-provider. Will try to investigate.

Test output: ``` Test md_gost12_512: M1 from RFC 6986 (10.1.1) and GOST R 34.11-2012 (А.1.1): [MD 64] success Test md_gost12_256: M1 from RFC 6986 (10.1.2) and GOST R 34.11-2012 (А.1.2):...

I have just patched the test that was failing on linux. Could you please retry?

I think I found the culprit: https://github.com/gost-engine/engine/blob/8efda3214b5592a5267db96cb0d0da2a2c747a69/gost_prov_mac.c#L148 It should be `EVP_DigestFinalXOF`, but naïve replacement does not help.

Could you please the corresponding fix? I still can't reproduce it but it should fix the issue. ```diff --git a/gost_prov_mac.c b/gost_prov_mac.c index ab04a9c..1bad120 100644 --- a/gost_prov_mac.c +++ b/gost_prov_mac.c @@ -145,7...

Do I correctly assume that you've updated the code and applied the patch afterwards?

Yes, it's a relevant state +patch from https://github.com/gost-engine/engine/issues/359#issuecomment-968343015

А диспетчер менее адским сделать не получается, да? :(