On 11/17/11 19:42, Ian Beckwith wrote:
> Yes, that fixed it, thanks.
OK, thanks, I pushed it, with a couple of minor further fixes:
a URL in the ChangeLog was wrong (Jim Meyering pointed this out),
and I had forgotten to remove the dependency of
crypto/gc-arctwo on crypto/bitrotate.
On Thu, Nov 17, 2011 at 01:59:27PM -0800, Paul Eggert wrote:
> I introduce that problem when I added stdalign.
> Does the following patch fix things for you?
Yes, that fixed it, thanks.
Ian.
--
Ian Beckwith - i...@erislabs.net - http://erislabs.net/ianb/
GPG fingerprint: AF6C C0F1 1E74 424B BCD
crypto/gc-hmac-sha1 (Depends-on):
Depend conditionally on crypto/hmac-sha1.
* modules/crypto/gc-md2 (Depends-on): Remove stdint, minmax.
* modules/crypto/gc-md4 (Depends-on):
Depend conditionally on crypto/md4.
* modules/crypto/gc-md5 (Depends-on):
Depend conditionally on crypto/md5.
* modules/cry
Hi,
crypto/gc-hmac-md5, crypto/gc-hmac-sha1, and crypto/gc-pbkdf2-sha1 all fail
with:
gcc -DHAVE_CONFIG_H -I. -I.. -DGNULIB_STRICT_CHECKING=1 -g -O2 -MT sha1.o
-MD -MP -MF .deps/sha1.Tpo -c -o sha1.o sha1.c
sha1.c:29:22: fatal error: stdalign.h: No such file or directory
or
gcc
I have installed this.
2005-10-13 Simon Josefsson <[EMAIL PROTECTED]>
* modules/gc-hmac-md5-tests: New file.
* tests/test-gc-hmac-sha1.c: New file.
,
| Files:
| tests/test-gc-hmac-sha1.c
|
| Depends-on:
|
| configure.ac:
|
| Makefile.am:
| TESTS += test-gc-hma
12 Oct 2005 09:26:41 -
@@ -1,5 +1,7 @@
2005-10-12 Simon Josefsson <[EMAIL PROTECTED]>
+ * gc-hmac-sha1.m4: New file.
+
* gc-sha1: New file.
* hmac-sha1.m4: New file.
Index: m4/gc-hmac-sha1.m4
=
Simon Josefsson <[EMAIL PROTECTED]> writes:
> What do you think? This is very similar to the hmac-md5 module.
I installed the patch below. It didn't touch on any existing module,
so I thought the impact was low.
> The reason the modules are separate is that I don't wan
What do you think? This is very similar to the hmac-md5 module.
The reason the modules are separate is that I don't want to force
hmac-sha1 users to have a hmac-md5 and md5 implementation around.
Alas, the sha1 module depend on md5 for 'md5_uint32'. Couldn't we fix
it to
