From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jon Smirl Date: Mon, 30 Mar 2009 13:09:30 -0400 Subject: [U-Boot] [PATCH v4] board support patch for phyCORE-MPC5200B-tiny In-Reply-To: <49D0EF8F.7050002@denx.de> References: <20090324155637.12279.60059.stgit@localhost> <9e4733910903251214y1058215l6f4b78ab0b2870a7@mail.gmail.com> <20090327201217.58C92832E406@gemini.denx.de> <9e4733910903271348r19317a71m1a7cea7edbe8ac50@mail.gmail.com> <9e4733910903290729n7f10f704xbd191729a99457ad@mail.gmail.com> <49D0EF8F.7050002@denx.de> Message-ID: <9e4733910903301009h4373cdb7y82d872b9285a6e18@mail.gmail.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de On Mon, Mar 30, 2009 at 12:13 PM, Anatolij Gustschin wrote: > Jon Smirl wrote: >> On Sun, Mar 29, 2009 at 10:08 AM, Markus Rathgeb >> wrote: >>> Hi! >>> >>> Have you read the BIOSEMU stuff in the mailing list? >>> Can you adjust the PCI section and perhaps you can include the video >>> example (set undef video). >> >> I have v1 of the pcm030 hardware and the baseboard doesn't have a PCI slot. >> >> I looked at some of your older emails. It looks like the framebuffer >> is not accessible. You are hanging on a write to the PCI bus. >> >> Forget about running the video BIOS initially, just use the memory >> commands from u-boot to make sure you can read and write the >> framebuffer. > > No, memory access to frame buffer will only work after the SD-RAM > controller of the GPU is properly initialized. Video BIOS is supposed > to do this among other things. As the boot log showed, video memory > size was not recognized by the radeon driver which indicates that That comment triggered a memory. There is a bug fix in the fbdev Radeon driver in Linux that pokes an undocumented register. That poke fixes this problem. The problem is only present one some cards, not all. It is a flaw in the ASIC. I'm looking at the the fbdev driver but I can't remember where the fix is. > the video BIOS run failed to initialize the graphic chip and memory > controller. Fixing the XLB Initiator Window configuration for bus > PCI IO space starting at 0x00000000 seemed to solve problems with > video BIOS run. > > Best regards, > Anatolij > > -- Jon Smirl jonsmirl at gmail.com