This problem was reported by Dell and affects RHEL3 and RHEL4. When using a monitor that can handle very large resolutions it is possible to configure the video output beyond what the ES1000 can handle. Specifically we are seeing resolution 1600x1200@85Hz which will display with corruption. This mode is beyond what the ES1000 can support when using 16bit memory at 200MHz. ATI has provided a patch that uses an emperical formula to limit the pixel clock thereby limiting the resolutions that are available. This is part of a larger RHEL4 patch which will be appended shortly. We are still working on a backport to RHEL3. This is also upstream - https://bugs.freedesktop.org/show_bug.cgi?id=5766
Created attachment 125069 [details] Patch including the fix to reduce pixel clock.
Ok, I've reviewed the patch in comment #1, and it seems ok except for: @@ -3524,7 +3526,11 @@ /* don't use RMX if we have a dual-tmds panels */ if (pRADEONEnt->MonType2 == MT_DFP) info->ddc_mode = TRUE; - + + if (info->IsDellServer) + { + info->ddc_mode = TRUE; + } xf86DrvMsg(pScrn->scrnIndex, X_INFO, "Validating modes on %s head ---------\n", info->IsSecondary ? "Secondary" : "Primary"); That seems to have accidentally gotten into the patch, as it has nothing to do with fixing the reported problem as far as I can tell. I've discussed this problem with Michel, who has described the details to me, and provided the following upstream bug report with the current patch attached: https://bugs.freedesktop.org/show_bug.cgi?id=5766
This is the patch that we'll probably be going with, from the above upstream bug: https://bugs.freedesktop.org/attachment.cgi?id=4636
I believe the following bug might be a dupe of this, but it was filed with zero useful information, so hard to tell: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=179886 I've requested details in that bug, and will close it as a dupe of this one if we can conclude it is the same issue they're requesting an update for.
*** Bug 180274 has been marked as a duplicate of this bug. ***
*** Bug 179886 has been marked as a duplicate of this bug. ***
Patch xorg-x11-6.8.2-ati-radeon-rn50-pixelclock-limit-bug182511.patch added to test candidate for U4. Please test the xorg-x11-6.8.2-1.EL.13.26 packages available for download via FTP at: ftp://people.redhat.com/mharris/testing/4E/xorg-x11
Mike, can you also provide a patch for RHEL3. Can this be a candidate for U8.
Mike, sorry for the delay. We have tested the patch on RHEL4 U3 and it works correctly, limiting the modes as follows: For 24bpp, Max: 1280x1024@60Hz For 16bpp, Max: 1600x1200@75Hz For 8bpp, Max: 1920x1440@60Hz Can you advise when this will be available in a RHEL4 Ux beta? Also can you provide a patch for RHEL3?
(In reply to comment #9) > Mike, can you also provide a patch for RHEL3. Can this be a candidate for U8. Theoretically the patch could be backported to RHEL-3, probably trivially. In order to include any fix into RHEL-3 or RHEL-4 in an update cycle however, an official request needs to be filed against each individual OS release that is relevant for a particular problem. This provides us with the ability to track every issue per-OS-release and to allocate developer resources to issues per-OS-release. In order to consider this for RHEL-3 at this late stage in the U8 cycle, we need to have an issue tracker request filed by Dell or ATI against RHEL-3, and if product management approves it through our exception process, there may still be time to consider this for RHEL-3-U8.
(In reply to comment #11) > Mike, sorry for the delay. We have tested the patch on RHEL4 U3 and it works > correctly, limiting the modes as follows: > > For 24bpp, Max: 1280x1024@60Hz > For 16bpp, Max: 1600x1200@75Hz > For 8bpp, Max: 1920x1440@60Hz Good to hear. Setting bug to "MODIFIED" state for QA testing. > Can you advise when this will be available in a RHEL4 Ux beta? It's present in build now, and will be in the RHEL4 U4 betas. The ATI partner manager may be able to provide the beta release schedule to you for more specific timeframes though. Hope this helps.
(In reply to comment #15) > Theoretically the patch could be backported to RHEL-3, probably trivially. > In order to include any fix into RHEL-3 or RHEL-4 in an update cycle > however, an official request needs to be filed against each individual > OS release ATI has completed and tested the backport. A separate bugzilla has now been opened which contains this patch for RHEL3: https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=189927 > In order to consider this for RHEL-3 at this late stage in the U8 cycle, > we need to have an issue tracker request filed by Dell or ATI against RHEL-3 I will ask our contact at Dell to open this issue tracker as soon as possible.
Bug report changed to from ON_QA to VERIFIED status by Errata System. Advisory RHBA-2006:0412-21. http://errata.devel.redhat.com/errata/errata/showrequest.cgi?advisory=3850 The RN50 works properly.