After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 385444 - gpilotd won't work when porting to solaris
gpilotd won't work when porting to solaris
Status: RESOLVED FIXED
Product: gnome-pilot
Classification: Other
Component: gpilotd
2.0.15
Other All
: Normal blocker
: ---
Assigned To: Matt Davey
gnome-pilot Maintainers
Depends on:
Blocks:
 
 
Reported: 2006-12-13 13:25 UTC by Jerry Yu
Modified: 2007-01-11 08:57 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
applications patch (1.34 KB, patch)
2006-12-13 13:27 UTC, Jerry Yu
none Details | Review

Description Jerry Yu 2006-12-13 13:25:52 UTC
Please describe the problem:
When porting the latest gnome-pilot pkgs for Solaris USB port on X86 and SPARC
platform, Gpilotd won't work.

Steps to reproduce:


Actual results:


Expected results:


Does this happen every time?
yes

Other information:
Comment 1 Jerry Yu 2006-12-13 13:27:11 UTC
Created attachment 78287 [details] [review]
applications patch
Comment 2 Matt Davey 2006-12-13 14:20:23 UTC
I guess this bug is related to 385434.

If I read your patch right, this will attempt a sync every 2 seconds on Solaris (it bypasses the code that sees if a recognised device has appeared by querying usbfs).

You suggest this is a regression since gnome-pilot 2.0.13, but as far as I can see, 2.0.13 would have tried to read "/proc/bus/usb/devices", and then try reading "/proc/bus/usb/devices_please-use-sysfs-instead" and then give up.

Help me out here! :)
Comment 3 André Klapper 2006-12-14 02:52:17 UTC
NEEDINFO state.
Jerry, please reopen this bug report when answering Matt's question. Thanks.
Comment 4 Jerry Yu 2006-12-29 08:01:24 UTC
No, it's not related to 385434.

Because there are no usb file and directory on Solaris which are similiar with "/proc/bus/usb/devices" and "/sys/bus/usb/devices" on Linux, so it's not necessary to check if a recognised devise has appeared by querying usb file on Solaris.

Actually, the patch has already been patched into gnome-pilot 2.0.13 on Solaris. However, we keep it as an internal patch then and we tend to commit it into community CVS currently by adding "if defined(sun) && defined (__SVR4)".
Comment 5 Matt Davey 2007-01-02 10:09:42 UTC
Thanks for the update, Jerry.
I understand that there's no need to check the non-existent
/proc/bus/usb/devices or /sys/bus/usb/devices on solaris.

Do you make any attempt to detect whether a recognised device has
appeared, or do you try to sync every 2 seconds?

As I mentioned in my first comment, your patch would appear to attempt
a sync every 2 seconds.  Presumably there is some way to detect devices
on solaris, and it would be better to use that than to blindly attempt
a sync every timeout interval.

What do you think?
Comment 6 Jerry Yu 2007-01-09 08:02:59 UTC
Currently this patch try to sync every 2 seconds on Solaris. 

I totally agree with your ideas that the devices should be detected at first which  
HAL can do. However, HAL doesn't work well enough on Solaris and we're pushing HAL team to fix it. When HAL works well, I'll enable --with-hal option when building gnome-pilot, then the palm devices can be detected automatically. But currently, unfortunately, there's no better way to do it. 
Comment 7 Matt Davey 2007-01-09 09:03:05 UTC
Thanks for clarifying, Jerry.
If that's the current situation on Solaris, then your changes seem straightforward.

I'll commit your changes at my next check in.
Comment 8 Jerry Yu 2007-01-09 10:47:15 UTC
Ok, thanks.
Comment 9 Matt Davey 2007-01-11 08:57:09 UTC
Patch applied to head (i.e. subversion trunk).