Bug 30740

Summary:	r128 DRI no-go after a resume
Product:	[Retired] Red Hat Raw Hide	Reporter:	Jeremy Katz <katzj>
Component:	XFree86	Assignee:	Mike A. Harris <mharris>
Status:	CLOSED WONTFIX	QA Contact:	David Lawrence <dkl>
Severity:	high	Docs Contact:
Priority:	high
Version:	1.0
Target Milestone:	---
Target Release:	---
Hardware:	i386
OS:	Linux
Whiteboard:
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2001-03-10 00:32:46 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Jeremy Katz 2001-03-06 01:18:27 UTC

First of all, DRI is working on my r128 M3 mobility in my laptop now with
XFree86-4.0.2-11.4.0.  The bus-mastering fix was the relevant fix from
previous testing with stuff off of dri-devel.  Unfortunately, after a
suspend/resume cycle, bus-mastering is disabled again.  So, bus-mastering
also needs to be explicitly enabled on an APM resume (since logging out of
X and restarting it does get working DRI again, but that's not a workable
solution always due to the fun of glx screensavers...)

Comment 1 Mike A. Harris 2001-03-06 16:34:34 UTC

APM is the problem here I believe.  Disable APM.  Does this fix it?

Comment 2 Jeremy Katz 2001-03-06 16:44:29 UTC

Disabling apm will of course work because then you can't suspend/resume at all,
which is kind of a bummer.  Luckily, the base X server continues to work after a
resume just as long as you avoid any GL apps...  I've sent a message to
dri-devel asking if there's a way to hook into the suspend/resume cycle from
within the driver, since you could then theoretically reenable bus mastering on
resume.

Comment 3 Mike A. Harris 2001-03-10 00:32:42 UTC

If they find a way to do it, let me know and I will add in the necessary
patch for you. I agree the current situation sucks, but at least it works
now..  ;o)

There's no NEEDPATCH so NEEDINFO.  ;o)

Comment 4 Mike A. Harris 2001-03-14 07:06:14 UTC

After reconsidering this, there is no realistic way that we (Red Hat) can
provide a fix for this until the XFree86 team and/or people working on
APM support provide a solution.  Such a fix will appear in a future release
of Red Hat Linux whenever the upstream maintainers of XFree/apmd/kernel
get APM functioning in Linux.  I will gladly look into any potential fixes
people send me in the meantime however.