| Summary: | [855GM TTM] X failed to allocate framebuffer if specified large virtual fb | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Product: | xorg | Reporter: | WuNian <nian.wu> | ||||||||||
| Component: | Driver/intel | Assignee: | Eric Anholt <eric> | ||||||||||
| Status: | VERIFIED FIXED | QA Contact: | Xorg Project Team <xorg-team> | ||||||||||
| Severity: | normal | ||||||||||||
| Priority: | medium | CC: | michael.fu, zhenyu.z.wang | ||||||||||
| Version: | git | ||||||||||||
| Hardware: | x86 (IA32) | ||||||||||||
| OS: | Linux (All) | ||||||||||||
| Whiteboard: | |||||||||||||
| i915 platform: | i915 features: | ||||||||||||
| Attachments: |
|
||||||||||||
|
Description
WuNian
2007-09-26 23:55:53 UTC
Created attachment 11775 [details]
xorg config file
Created attachment 11776 [details]
xorg log file
This seems to be caused by i915-unification. Now using the latest git code On 855GM, if specified "Virtual 2048 1050" in xorg.conf, then start up X, it display following error messages and exit. Below is the error info: Fatal server error: Couldn't bind memory for BO front buffer Created attachment 12858 [details]
xorg log file
Your last log is corrupted. Created attachment 13873 [details] xorg log file (In reply to comment #6) > Your last log is corrupted. > Thanks,I update a new log which got on latest code from 855gm. Eric, comment? What would help is putting i830_describe_allocations at various points in the memory allocation path, but particularly at the just-failed-an-allocation point. There may also be dmesg output relevant to this particular bug. This whole mess will get easier to do right once we move to exa-ttm and remove one of the suballocators that fragments our memory. (In reply to comment #9) > What would help is putting i830_describe_allocations at various points in the > memory allocation path, but particularly at the just-failed-an-allocation > point. There may also be dmesg output relevant to this particular bug. Do you need us to upload the dmesg? > This whole mess will get easier to do right once we move to exa-ttm and remove > one of the suballocators that fragments our memory. Do you mean the bug will be fixed later when exa-ttm is ready? Yes, if you run dmesg, you may see relevant output from the failure. And with exa-ttm, chance of success without any other changes just goes from "medium" to "high". Now, after specify virtual size as 2048x1050, X failed up and reported: Fatal server error: Couldn't bind memory for BO front buffer I checked the dmesg which had info: [root@x-855gm ~]# dmesg |grep drm [drm] Initialized drm 1.1.0 20060810 [drm] Initialized i915 1.12.0 20070209 on minor 0 [drm] Initialized i915 1.12.0 20070209 on minor 1 [drm:drm_buffer_object_validate] *ERROR* Failed moving buffer. *** Bug 13758 has been marked as a duplicate of this bug. *** Eric said this problem would go away with intel-batchbuffer merged. With git tip, no ttm, there's no such issue, so close this one verified |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.