Bug 5482
Summary: | Black screen with radeonfb | ||
---|---|---|---|
Product: | Other | Reporter: | BogDan (taipanromania) |
Component: | Modules | Assignee: | other_modules |
Status: | REJECTED INSUFFICIENT_DATA | ||
Severity: | blocking | CC: | benh, protasnb |
Priority: | P2 | ||
Hardware: | i386 | ||
OS: | Linux | ||
Kernel Version: | 2.6.13 .. 2.6.14-rc4 | Subsystem: | |
Regression: | --- | Bisected commit-id: |
Description
BogDan
2005-10-22 01:57:22 UTC
Looks like you have one of those stupid video cards that expose 2 functions for the 2 heads (total bogosity mandated by Windows it seems). Vesafb is taking control fo the first one (I don't know why, it shouldn't if you use an explicit kernel command line argument) and radeonfb is trying to get to the second but that all messes up because they are the same bit of hardware. Well, I _think_ that's what's happening. What if you don't build vesafb in your kernel ? The dmesg command I type when I use vesafb in kernel. When I use only radeonfb I cant see anything. I will change my MB (in a week) because I took my video card to a friend and there works perfect so I think maybe is an intel-agp module problem, because when I try to use xorg with dri and glx module, xorg reset my system. My MB is very strange (http://www.asrock.com/product/product_775Dual-915GL.htm) it has: " - ASRock A.G.I. Express slot: Compaitble with PCI Express x16 VGA card* - Worldwide patent ASRock A.G.I.8X graphics upgrade slot (AGP8X/4X compatible)* " If someone want's to give it a try I will give him(her) full access to my computer. sorry, sorry, for my VERY BAD English Even if you use vesafb with radeonfb, radeonfb should get initialized before vesafb. In your case, vesafb loads and was grabbed by the framebuffer console. Later on, radeonfb loads. It looks like that radeonfb is being loaded as a module. So, can you send your dmesg output without vesafb and your kernel config file? No response for 3 months. It does look like that the problem is that vesafb gets loaded first before radeonfb (which is most probably compiled as a module) It should be fixed by disabling vesafb, but cannot confirmed. Ben, should we close this bug as rejected->insufficient data? Maybe... also the DRM problem he's reporting is completely separate from whatever radeonfb issue... Bogdan, Can you try #4 please - disabling vesafb? Maybe it also makes sense to try building radeonfb into the kernel monoliticly. Thanks. Closing the bug, no response from reporter... |