238586 – dev-util/git with USE="subversion" should require that dev-util/subversion be built with USE="-dso"

Bug 238586 - dev-util/git with USE="subversion" should require that dev-util/subversion be built with USE="-dso"

Summary: dev-util/git with USE="subversion" should require that dev-util/subversion be...

Status:	RESOLVED FIXED

Alias:	None

Product:	Gentoo Linux
Classification:	Unclassified
Component:	New packages (show other bugs)
Hardware:	All All

Importance:	High normal (vote)
Assignee:	Fernando J. Pereda (RETIRED)

URL:
Whiteboard:
Keywords:

Depends on:	221931 223747
Blocks:
	Show dependency tree

Reported:	2008-09-24 17:19 UTC by Arfrever Frehtes Taifersar Arahesis (RETIRED)
Modified:	2009-06-12 21:31 UTC (History)
CC List:	8 users (show)

See Also:
Package list:
Runtime testing required:	---

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.

Description Arfrever Frehtes Taifersar Arahesis (RETIRED) gentoo-dev

2008-09-24 17:19:11 UTC

dev-util/git with USE="subversion" should require that dev-util/subversion be built with USE="-dso". Please add appropriate built_with_use check.

See bug #221931 and bug #223747.

Comment 1 Karl H 2008-10-08 13:01:36 UTC

Verifying that this workaround works for me. My system is x86

Comment 2 skolima 2008-10-11 11:17:14 UTC

Worked for me, x86 (pentium3 and prescott).

Comment 3 Robin Johnson archtester

2008-10-11 21:57:45 UTC

Arfrever: I'd actually prefer not to add that check. There are cases when the actual problem (best described in bug 223747) doesn't occur, because the unload order is different, due to interactions from prelink.

The upstream bug really does need fixing.

Comment 4 Robert Buchholz (RETIRED) gentoo-dev

2008-10-12 02:37:36 UTC

(In reply to comment #3)
> Arfrever: I'd actually prefer not to add that check. There are cases when the
> actual problem (best described in bug 223747) doesn't occur, because the unload
> order is different, due to interactions from prelink.
> 
> The upstream bug really does need fixing.

Upstream has not fixed the bug since they were made aware of it half a year ago, and has not responded to the lastest mails on the list or the bug report.
The fact that it might work in some cases doesn't mean we should accept the fact that a feature that merely enhances performance breaks for for others (with no obvious solution).
I'd vote for a built_with_use check in stable (with an ewarn), and a negative USE dep in ~arch.

Comment 5 Robin Johnson archtester

2008-11-24 01:17:50 UTC

Ok, it prints a warning now. I'm avoiding EAPI=[12] for the moment for git, until we have the 1.6 series in stable.

Comment 6 Robert Buchholz (RETIRED) gentoo-dev

2009-01-31 10:32:14 UTC

(In reply to comment #5)
> Ok, it prints a warning now. I'm avoiding EAPI=[12] for the moment for git,
> until we have the 1.6 series in stable.

1.6.0.6 is stable, any chances for this bug now? What's the reason for hesitation concerning an EAPI bump?

Comment 7 Robin Johnson archtester

2009-03-06 06:55:12 UTC

In CVS for 1.6.2.