Bug 8373

Summary:	[update candidate] ldetect-0.12.1.1-5.mga2.
Product:	Mageia	Reporter:	Thierry Vignaud <thierry.vignaud>
Component:	RPM Packages	Assignee:	QA Team <qa-bugs>
Status:	RESOLVED FIXED	QA Contact:
Severity:	normal
Priority:	Normal	CC:	Mertens.MPT, sysadmin-bugs, tmb, wassi
Version:	2	Keywords:	validated_update
Target Milestone:	---
Hardware:	All
OS:	Linux
Whiteboard:	has_procedure mga2-64-OK MGA2-32-OK
Source RPM:	ldetect-0.12.1.1-5.mga2.src.rpm	CVE:
Status comment:
Bug Depends on:
Bug Blocks:	8320, 8369

Description Thierry Vignaud 2012-12-12 15:44:08 CET

Suggested advisory:
===================
This ldetect update fixes a bug where ldetect would fails to detect some devices when there were more than 100 PCI devices (mga#8320).
It also fixed a bottleneck in ldetect making lspcidrake 3x faster.

How to reproduce:
==================
Got a machine with more than 100 PCI devices.
Eg one where the following command would be limited to 100:
lspcidrake |perl -pi -e 'undef $_ if /^hub|^usb/../eof/'|wc -l

With new ldetect, all PCI devices should be detected.

Comment 1 claire robinson 2013-01-05 15:17:17 CET

SRPM: ldetect-0.12.1.1-5.mga2.src.rpm
-------------------------------------
ldetect
lib64ldetect0.12
lib64ldetect-devel

Comment 2 claire robinson 2013-01-05 15:17:57 CET

$ lspcidrake |perl -pi -e 'undef $_ if /^hub|^usb/../eof/'|wc -l
32

Comment 3 claire robinson 2013-01-05 15:28:19 CET

Testing complete mga2 64

Same number of pci devices after update and harddrake lists everything the same as it did before.

Whiteboard: (none) => has_procedure mga2-64-OK

Comment 4 Markus Mertens 2013-01-15 07:33:53 CET

(In reply to comment #3)
> Testing complete mga2 64
> 
> Same number of pci devices after update and harddrake lists everything the same
> as it did before.

On my system with more than 100 PCI devices this update does not change anything. The limit of 100 devices still exists. See also the following bugs:


https://bugs.mageia.org/show_bug.cgi?id=8369
https://bugs.mageia.org/show_bug.cgi?id=8320

CC: (none) => Mertens.MPT

Comment 5 claire robinson 2013-01-15 10:17:42 CET

Thanks for testing Markus

Can you confirm you installed both ldetect-0.12.1.1-5.mga2 and lib(64)ldetect0.12-0.12.1.1-5.mga2 from Core Updates Testing please.

$ rpm -qa | grep ldetect

claire robinson 2013-01-15 10:21:40 CET

Depends on: (none) => 8369, 8320

claire robinson 2013-01-15 10:21:53 CET

Whiteboard: has_procedure mga2-64-OK => has_procedure

Comment 6 Markus Mertens 2013-01-15 10:46:23 CET

(In reply to comment #5)
> Thanks for testing Markus
> 
> Can you confirm you installed both ldetect-0.12.1.1-5.mga2 and
> lib(64)ldetect0.12-0.12.1.1-5.mga2 from Core Updates Testing please.
> 
> $ rpm -qa | grep ldetect

# rpm -qa | grep ldetect
ldetect-0.12.1.1-5.mga2
ldetect-lst-0.1.303.2-1.mga2
lib64ldetect0.12-0.12.1-5.mga2
#

Well, lib64ldetect0.12 does not have the correct version. I will try to update...


# rpm -qa | grep ldetect
ldetect-0.12.1.1-5.mga2
lib64ldetect0.12-0.12.1.1-5.mga2
ldetect-lst-0.1.303.2-1.mga2
#

OK. Version *0.12.1.1* seems to solve the problems on my system. I can finally see all build-in hardware in harddrake2. I mean everything is listed but I could not test the installation procedure (https://bugs.mageia.org/show_bug.cgi?id=8369). But next week I will prepare an extra partition on my system and try to install Mageia 3 beta*. Maybe dracut will create a usable initrd once all hardware is properly detected.

Comment 7 claire robinson 2013-01-15 10:58:37 CET

Thanks Markus.

Can you also try the line from comment 0 below

# lspcidrake |perl -pi -e 'undef $_ if /^hub|^usb/../eof/'|wc -l

It should no longer be limited to 100

Comment 8 Markus Mertens 2013-01-15 11:07:07 CET

(In reply to comment #7)
> Thanks Markus.
> 
> Can you also try the line from comment 0 below
> 
> # lspcidrake |perl -pi -e 'undef $_ if /^hub|^usb/../eof/'|wc -l
> 
> It should no longer be limited to 100

# lspcidrake |perl -pi -e 'undef $_ if /^hub|^usb/../eof/'|wc -l
137
#

Looks good!

Comment 9 claire robinson 2013-01-15 11:13:24 CET

That's great, thankyou :) 

Adding mga2 64 testing complete back again.

Needs testing mga2 32 to validate.

Whiteboard: has_procedure => has_procedure mga2-64-OK

claire robinson 2013-01-15 11:13:45 CET

Blocks: (none) => 8369, 8320
Depends on: 8369, 8320 => (none)

Comment 10 user7 2013-01-15 15:50:45 CET

Testing complete on mga2, i586.

Before update:
[wassi@localhost ~]$ lspcidrake |perl -pi -e 'undef $_ if /^hub|^usb/../eof/'|wc -l
28

After update:
-> Reboot
[wassi@localhost ~]$ lspcidrake |perl -pi -e 'undef $_ if /^hub|^usb/../eof/'|wc -l
28
Harddrake looks the same as before.


Validating
----
No linking needed according to depcheck.

See comment 0 for advisory, comment 1 for SRPM.

Could sysadmin please push from core/updates_testing to core/updates.

Thank you!

Keywords: (none) => validated_update
CC: (none) => sysadmin-bugs, wassi
Whiteboard: has_procedure mga2-64-OK => has_procedure mga2-64-OK MGA2-32-OK

Comment 11 Thomas Backlund 2013-01-18 01:44:56 CET

Update pushed:
https://wiki.mageia.org/en/Support/Advisories/MGAA-2013-0004

Status: NEW => RESOLVED
CC: (none) => tmb
Resolution: (none) => FIXED