On Thu, Feb 24, 2005 at 06:43:32PM +0100, Alfred M. Szmidt wrote:
> For glibc-2_3-branch I had to add '#include <stdint.h>' to
> libio/fmemopen.c. (You have fixed that in HEAD, but not in
> glibc-2_3-branch; I sent you an email about that some days ago but
> didn't get an answer, yet.)
>
> Could you file this in bugzilla for glibc?
Done: <URL:http://sources.redhat.com/bugzilla/show_bug.cgi?id=765>
Is it correct that this was automatically assigned to
<URL:mailto:[EMAIL PROTECTED]>?
At least, I'm not aware of having changed anything.
> [...] configuring glibc.
>
> [...] isn't needed. Nor is --with-headers
It _is_. For me.
:-P
> On the other hand, you must use
> --without-tls, or you will get a totally broken system that is
> incapable of working at all.
That one or replacing '-Os' with '-O2' finally made a fresh build
succeed (or both of them).
(I did not yet run the test suite, though, but the included
'bin/getent' etc. works.)
Maybe 'gcc -Os' produces invalid code that makes GNU Mach crash?
I'll have a look at that later.
I needed the attached patches to make it compile. They were (partly)
discussed on the relevant mailing lists, but didn't get applied.
Shall I open bugzilla requests for them?
What's the policy for such tiny changes?
(... and why aren't they noticed by other people? -- They make the build
fail completely for me. Perhaps everyone uses Debian's glibc... Or
everyone is too lazy to report those tiny changes... Or...)
Regards,
Thomas
--- elf/dl-load.c.orig 2005-02-25 10:40:00.000000000 +0100
+++ elf/dl-load.c 2005-02-25 11:59:46.000000000 +0100
@@ -110,6 +110,8 @@
= PROT_GROWSDOWN;
#elif _STACK_GROWS_UP && defined PROT_GROWSUP
= PROT_GROWSUP;
+#else
+ ;
#endif
2004-12-12 Andreas Schwab <[EMAIL PROTECTED]>
* malloc/arena.c [USE_ARENAS]: Include <stdbool.h>.
--- malloc/arena.c.~1.13.~ 2004-12-12 17:39:32.000000000 +0100
+++ malloc/arena.c 2004-12-12 17:46:55.167066564 +0100
@@ -539,6 +539,8 @@ thread_atfork_static(ptmalloc_lock_all,
#if USE_ARENAS
+#include <stdbool.h>
+
#if MALLOC_DEBUG > 1
/* Print the complete contents of a single heap to stderr. */
--
Andreas Schwab, SuSE Labs, [EMAIL PROTECTED]
SuSE Linux Products GmbH, Maxfeldstra�e 5, 90409 N�rnberg, Germany
Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
_______________________________________________
Bug-hurd mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-hurd
