[PATCH] Re: Battery-related regression between 2.6.17-git3 and 2.6.17-git6

[Danny Kukawka]

On Wednesday 05 July 2006 00:14, Kay Sievers wrote:
[...]
> Hmm,
>   ./hald/util.h:#define HAL_PATH_MAX 256
> looks suspicious. :)

HAL_PATH_MAX was not the reason for the crash. In fact Kay already fixed the 
problem with this commit:

2005-11-02  Kay Sievers  <kay.sievers at vrfy.org>

        * hald/linux2/coldplug.c: (coldplug_synthesize_events): Prepare for 
	     2.6.15. Check if we find a symlink instead of a class device pointing to 
		 a device below a class device. This kernel sysfs layout will change again 
        in the future, for now resolve the link to the "real" device path, 
        like real hotplug events devpath would have.

This patch was not part of the HAL version Rafael used (hal-0.5.4).

Btw: While investigating the problem I found two problems in 
hal_util_get_normalized_path (). The attached patch (against CVS HEAD, I port 
the patch later to git) fixes this issues.

Danny

 ChangeLog   |    6 ++++++
 hald/util.c |    8 +++++---
 2 files changed, 11 insertions(+), 3 deletions(-)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hal-fix-hal_util_get_normalized_path.diff
Type: text/x-diff
Size: 1596 bytes
Desc: not available
Url : http://lists.freedesktop.org/archives/hal/attachments/20060728/7525e0ef/hal-fix-hal_util_get_normalized_path.bin