Excess dependency noted by Eli Zaretskii (Bug#28023#17).
* lib/tempname.c (__secure_getenv) [!_LIBC]: Remove; unused.
* modules/tempname (Depends-on): Remove secure_getenv.
---
ChangeLog| 7 +++
lib/tempname.c | 1 -
modules/tempname | 1 -
3 files changed, 7 insertions(+), 2 deletio
On 08/09/2017 09:03 AM, Eric Blake wrote:
Are we interested in improving gnulib's fchmodat() module to implement > this
workaround?
It would be nice if fchmodat worked with the AT_SYMLINK_NOFOLLOW flag,
yes. I think GNU Tar would use it, in some cases. I'll CC bug-tar, as I
had forgotten abou
glibc's current implementation of fchmodat() is non-compliant when
AT_SYMLINK_NOFOLLOW is used. qemu just recently rediscovered what musl
libc has already known for three years: on kernels new enough to support
O_PATH, you can get a race-free implementation that supports
AT_SYMLINK_NOFOLLOW even w
On 25/07/17 09:28, Paul Eggert wrote:
* lib/fts.c (struct dev_type): New struct.
(DEV_TYPE_HT_INITIAL_SIZE): New constant.
(dev_type_hash, dev_type_compare, filesystem_type): New functions.
(dirent_inode_sort_may_be_useful, leaf_optimization_applies):
Now takes FTSENT const *, not int. All uses
You're right. I missed that part. Here's an updated version
* Paul Eggert [170809 11:58]:
Darshit Shah wrote:
+This function is missing on all non-glibc platforms: glibc 2.25
No, as I mentioned it's present in the three main BSD platforms.
Please change "all non-glibc" to "many", and updat
Darshit Shah wrote:
> +This function is missing on all non-glibc platforms: glibc 2.25
No, as I mentioned it's present in the three main BSD platforms. Please change
"all non-glibc" to "many", and update the version numbers for FreeBSD etc. to
match my earlier email.
Hi Bruno,
Thanks for all your help. I've attached an updated version of the patch
which should now be (almost) complete. I've taken into account
everything that Paul and you mentioned in the last mail.
* Bruno Haible [170809 10:59]:
Hi Darshit,
Looks quite complete now. Other than what Pau