If I remember correctly the glibc (2.5) was probably first configured against a 2.6.10 ish kernel, and might not have been updated for 2.6.25. Anyways, the patch seems to work, so no need to dive any deeper. Thanks for the quick patch.
-----Original Message----- From: tz <tz-bounces@iana.org> On Behalf Of Rune Torgersen via tz Sent: Monday, September 18, 2023 8:43 AM To: Paul Eggert <eggert@cs.ucla.edu> Cc: Time zone mailing list <tz@iana.org> Subject: Re: [tz] 2023c install issue
This email originated from outside Innovative Systems. Do not click links or open attachments unless you recognize the sender and know the content is safe.
It's a PowerPC system, so that might a reason.
-----Original Message----- From: Paul Eggert <eggert@cs.ucla.edu> Sent: Friday, September 15, 2023 7:58 PM To: Rune Torgersen <runet@innovsys.com> Cc: Time zone mailing list <tz@iana.org> Subject: Re: [tz] 2023c install issue
This email originated from outside Innovative Systems. Do not click links or open attachments unless you recognize the sender and know the content is safe.
On 9/15/23 08:33, Rune Torgersen wrote:
Kernel version is actually 2.6.24 and 2.6.25.
That's odd, as linkat's AT_SYMLINK_FOLLOW support was introduced by Ulrich Drepper in Linux 2.6.18; I just checked the Linux source code repository. Perhaps your glibc is configured to not use the *at kernel calls, or maybe you're using an unusual (not x86; or tailored) platform.
It's probably not worth our time spelunking this old kernel. Glad my patch fixed things for you. The patch should appear in the next TZDB release.