[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <b306f6d5701cafc52a0323b7c4fe3fcd@smtp.hushmail.com>
Date: Mon, 21 Jan 2013 19:39:23 +0100
From: magnum <john.magnum@...hmail.com>
To: john-dev@...ts.openwall.com
Subject: Re: Proposed optimizations to pwsafe
On 21 Jan, 2013, at 19:15 , magnum <john.magnum@...hmail.com> wrote:
> On 21 Jan, 2013, at 18:55 , Brian Wallace <nightstrike9809@...il.com> wrote:
>
>> Fixed the OpenSSL comment issue and made a pull request. I can look into doing the SIMD for SHA256.
>
> Thanks! I accepted the pull request for unstable-jumbo although it is mosty bugfix-only. SIMD would be excellent but please check out the bleeding-jumbo branch for that.
Our local sha2.h may end up using OpenSSL, CommonCrypto or homebrewn code. The use of an OpenSSL-internal struct member broke CommonCrypto builds but it was an easy fix:
@@ -487,12 +487,18 @@ static int crypt_all(int *pcount, struct db_salt *salt)
#endif
{
SHA256_CTX ctx;
+
SHA256_Init(&ctx);
SHA256_Update(&ctx, saved_key[index], strlen(saved_key[index]));
SHA256_Update(&ctx, cur_salt->salt, 32);
SHA256_Final((unsigned char*)crypt_out[index], &ctx);
+#ifdef COMMON_DIGEST_FOR_OPENSSL
+ pwsafe_sha256_iterate(ctx.hash, cur_salt->iterations);
+ memcpy(crypt_out[index], ctx.hash, 32);
+#else
pwsafe_sha256_iterate(ctx.h, cur_salt->iterations);
memcpy(crypt_out[index], ctx.h, 32);
+#endif
}
return count;
}
The patches are merged to bleeding-jumbo too now.
magnum
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
