Has anyone noticed that test-lock takes a long time to complete on some
systems? On my laptop it is fast:
jas@latte:~/src/gnutls/gl/tests master$ time ./test-lock
Starting test_lock ... OK
Starting test_rwlock ... OK
Starting test_recursive_lock ... OK
Starting test_once ... OK
real0m1.724s
Simon Josefsson wrote:
> Has anyone noticed that test-lock takes a long time to complete on some
> systems? On my laptop it is fast:
...
> However on a otherwise idle machine with 2xE5520's (resulting in 16
> virtual CPUs), it takes much longer:
>
> jas@leo:~/gnutls-3.0.12/gl/tests$ time ./test-l
Jim Meyering writes:
> Hi Simon,
> Yes, I have seen precisely that problem, and it prompted this change in iwhd:
Thanks for confirmation, I've installed the same work-around in GnuTLS
now.
/Simon
* lib/malloca.c: Include .
(mmalloca, freea): Use uintptr_t to convert pointers to integers.
* modules/malloca (Depends-on): Add stdint.
* modules/relocatable-prog-wrapper (Depends-on): Likewise.
With review from Bruno Haible .
---
ChangeLog|9 +
lib/malloca.c
Hi,
Here is a small patch seris to avoid a couple of
warnings on mingw64 x86_64.
The last patch is RFC, I am not satisifed with the
solution, but mingw64 use a #define stat _stat64,
and it's difficult to keep this type if stat() is
redefined by rpl_stat().
Marc-André Lureau (4):
malloca: Avoi
* lib/select.c (IsConsoleHandle): Use intptr_t to convert handle
pointer to an integer. Fix warnings with MinGW64 x64.
* lib/poll.c (IsConsoleHandle): Likewise.
* lib/isatty.c (IsConsoleHandle): Likewise.
* modules/select (Depends-on): Add stdint.
* modules/poll (Depends-on): Likewise.
* modules/is
* lib/accept4.c (accept4): _open_osfhandle() expect a
intptr_t.
* lib/fcntl.c (dupfd): Likewise.
* lib/w32sock.h (SOCKET_TO_FD): Likewise.
---
ChangeLog |8
lib/accept4.c |2 +-
lib/fcntl.c |2 +-
lib/w32sock.h |2 +-
4 files changed, 11 insertions(+), 3 deletions(-)
* lib/sys_stat.in.h: in order to keep mingw64 stat type, add a
orig_strut_stat typedef. A bunch of "struct stat" need to be
replaced to use that original type, because stat() is defined by
rpl_stat() instead.
* lib/canonicalize-lgpl.c (__realpath): use orig_struct_stat.
* lib/readlink.c (readlink)
At the end of this messiage is a patch I imported from a pending glibc patch.
By the way, can anybody explain the following comment
in m4/include_next.m4?
dnl gcc does not warn about some things, and on some systems (Solaris and
Interix)
dnl __STDC__ evaluates to 0 instead of to 1. The latte
Does the following gnulib patch fix things for Bison on OS X?
I'll CC: this to bug-gnulib@gnu.org, to give Bruno Haible
a heads-up about the localcharset problem.
localcharset: port to Mac OS X's C locale
* lib/localcharset.c (get_charset_aliases) [DARWIN7]:
Map "US-ASCII" to "ASCII". Problem rep
Hi Simon,
> Has anyone noticed that test-lock takes a long time to complete on some
> systems?
Yes, on Solaris 11 for example it is particularly slow.
> Is there some scaling in the test that makes it take longer for
> multi-cpu machines? I didn't see any from a quick look.
No, the test has a
[adding bug-gnulib]
On 01/25/2012 01:13 PM, Marc-André Lureau wrote:
> ---
> tools/virsh.c |4 ++--
> 1 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/tools/virsh.c b/tools/virsh.c
> index 41c..246e638 100644
> --- a/tools/virsh.c
> +++ b/tools/virsh.c
> @@ -3703,7 +3703
[dropping libvirt]
On 01/25/2012 06:05 PM, Eric Blake wrote:
> [adding bug-gnulib]
>
> On 01/25/2012 01:13 PM, Marc-André Lureau wrote:
>> ---
>> tools/virsh.c |4 ++--
>> 1 files changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/tools/virsh.c b/tools/virsh.c
>> index 41c..246e
As the replacement cannot be made thread-safe, we need to
document this to prevent users from having a false sense
of safety.
* doc/glibc-functions/pipe2.texi (pipe2): mention lack of thread-safety
Signed-off-by: Eric Wong
---
I took the line off the existing documentation for the
similar-in-p
Paul Eggert wrote:
> By the way, can anybody explain the following comment
> in m4/include_next.m4?
>
> dnl gcc does not warn about some things, and on some systems (Solaris and
> Interix)
> dnl __STDC__ evaluates to 0 instead of to 1. The latter is an undesired side
> dnl effect; we are th
15 matches
Mail list logo