Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
[bug #58546] GSTLS certificate expiration test always fails on 32-bit architectures
1 view
Skip to first unread message
Yavor Doganov
Jun 11, 2020, 1:07:05 PM6/11/20
to Yavor Doganov, bug-g...@gnu.org
URL:
<https://savannah.gnu.org/bugs/?58546>
Summary: GSTLS certificate expiration test always fails on
32-bit architectures
Project: GNUstep
Submitted by: yavor
Submitted on: Thu 11 Jun 2020 08:07:00 PM EEST
Category: Base/Foundation
Severity: 3 - Normal
Item Group: Bug
Status: None
Privacy: Public
Assigned to: None
Open/Closed: Open
Discussion Lock: Any
_______________________________________________________
Details:
See the attached patch. Another solution is to replace the test certificate
with a one that expires before 2038.
_______________________________________________________
File Attachments:
-------------------------------------------------------
Date: Thu 11 Jun 2020 08:07:00 PM EEST Name: gstls-time_t-limitation.patch
Size: 1KiB By: yavor
<http://savannah.gnu.org/bugs/download.php?file_id=49243>
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?58546>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/
<https://savannah.gnu.org/bugs/?58546>
Summary: GSTLS certificate expiration test always fails on
32-bit architectures
Project: GNUstep
Submitted by: yavor
Submitted on: Thu 11 Jun 2020 08:07:00 PM EEST
Category: Base/Foundation
Severity: 3 - Normal
Item Group: Bug
Status: None
Privacy: Public
Assigned to: None
Open/Closed: Open
Discussion Lock: Any
_______________________________________________________
Details:
See the attached patch. Another solution is to replace the test certificate
with a one that expires before 2038.
_______________________________________________________
File Attachments:
-------------------------------------------------------
Date: Thu 11 Jun 2020 08:07:00 PM EEST Name: gstls-time_t-limitation.patch
Size: 1KiB By: yavor
<http://savannah.gnu.org/bugs/download.php?file_id=49243>
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?58546>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/
Richard Frith-Macdonald
Jun 12, 2020, 4:16:00 AM6/12/20
to Richard Frith-Macdonald, Yavor Doganov, bug-g...@gnu.org
Update of bug #58546 (project gnustep):
Open/Closed: Open => Closed
_______________________________________________________
Follow-up Comment #1:
Thanks for spotting that ... patch applied.
Open/Closed: Open => Closed
_______________________________________________________
Follow-up Comment #1:
Thanks for spotting that ... patch applied.
Fred Kiefer
Jun 12, 2020, 4:29:49 AM6/12/20
to Richard Frith-Macdonald, Fred Kiefer, Yavor Doganov, bug-g...@gnu.org
Update of bug #58546 (project gnustep):
Status: None => Confirmed
Open/Closed: Closed => Open
_______________________________________________________
Follow-up Comment #2:
As far as I understand this you mean that this test is bound to fail on some
32bit systems as the expiration date of the certificate in 2118 can not be
represented on that systems. This is due to time_t being 32 bits on some
systems. As this is not true on all 32 bit system I would prefer to just mark
this test as hopeful. This should have similar results as your change.
Fred Kiefer
Jun 12, 2020, 4:36:33 AM6/12/20
to Richard Frith-Macdonald, Fred Kiefer, Yavor Doganov, bug-g...@gnu.org
Update of bug #58546 (project gnustep):
Status: Confirmed => Fixed
Assigned to: None => CaS
Open/Closed: Open => Closed
_______________________________________________________
Follow-up Comment #3:
_______________________________________________________
Sorry for the confusion. Richard and I worked on this item at the same time so
I did not see he already accepted this patch.
Yavor Doganov
Jun 12, 2020, 4:36:42 AM6/12/20
to Richard Frith-Macdonald, Fred Kiefer, Yavor Doganov, bug-g...@gnu.org
Follow-up Comment #4, bug #58546 (project gnustep):
AFAIK, the time_t limitation is on all 32 bit systems, at least glibc-based
ones. IIRC, there were some discussions to address this issue but it turned
out that such a change would break the ABI.
Perhaps you mean *BSD systems or some Unix flavours?
AFAIK, the time_t limitation is on all 32 bit systems, at least glibc-based
ones. IIRC, there were some discussions to address this issue but it turned
out that such a change would break the ABI.
Perhaps you mean *BSD systems or some Unix flavours?
Richard Frith-Macdonald
Jun 12, 2020, 4:41:15 AM6/12/20
to Richard Frith-Macdonald, Fred Kiefer, Yavor Doganov, bug-g...@gnu.org
Follow-up Comment #5, bug #58546 (project gnustep):
In practice, I don't see testing this on 32bit systems as a priority. If the
tests pass on 64bit systems then the code is very unlikely to have been broken
(and now that the world has moved to 64bit, it doesn't seem important to have
a specific test for 32bit systems).
In practice, I don't see testing this on 32bit systems as a priority. If the
tests pass on 64bit systems then the code is very unlikely to have been broken
(and now that the world has moved to 64bit, it doesn't seem important to have
a specific test for 32bit systems).
0 new messages