Date: Sun, 04 Apr 2010 06:20:27 -0500 From: Robert Noland <rnoland@FreeBSD.org> To: Peter Jeremy <peterjeremy@acm.org> Cc: x11@FreeBSD.org, Martin Wilke <miwi@FreeBSD.org> Subject: Re: [Call for Testing] X.org 7.5 for FreeBSD Message-ID: <1270380027.4235.8.camel@balrog.2hip.net> In-Reply-To: <20100404000033.GG86236@server.vk2pj.dyndns.org> References: <20100311134413.GH19740@bsdcrew.de> <20100404000033.GG86236@server.vk2pj.dyndns.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, 2010-04-04 at 10:00 +1000, Peter Jeremy wrote: > On 2010-Mar-11 14:44:14 +0100, Martin Wilke <miwi@FreeBSD.org> wrote: > >Please report any problems and issus to x11 (at) FreeBSD.org. > > Another data point (sorry for the delay): > > FreeBSD server.vk2pj.dyndns.org 8.0-STABLE FreeBSD 8.0-STABLE #1: Wed Jan 27 06:55:10 EST 2010 root@server.vk2pj.dyndns.org:/var/obj/usr/src/sys/server amd64 > > (--) PCI:*(0:1:0:0) 1002:94c3:1458:2192 ATI Technologies Inc RV610 video device [Radeon HD 2400 PRO] rev 0, Mem @ 0xd0000000/268435456, 0xfdfe0000/65536, I/O @ 0x0000de00/256, BIOS @ 0x????????/65536 > > Good news: X starts and stops cleanly and VTY switching works. > > Problems: > 1) DRI (and apparently EXA) report as disabled: > (EE) RADEON(0): [pci] Out of memory (-12) > (EE) RADEON(0): [pci] PCI failed to initialize. Disabling the DRI. > (WW) RADEON(0): Direct rendering disabled > (EE) RADEON(0): Acceleration initialization failed > (II) RADEON(0): Acceleration disabled > > There was a thread in mid-February suggesting that this is caused by > a lack of contiguous RAM. I've tried running a process that just > allocates 1GB RAM and then exits (leaving me with >900MB free space). > Whilst this doesn't guarantee any large contiguous RAM blocks, it > seems odd that there wouldn't be any. I thought that I had a patch posted that addressed both this issue as well as the mapping overlap, but it looks like i never posted the reworked scatter gather allocation code. What kernel are you running? The new code interacts directly with the VM system and eliminates the requirement to have large chunks of contiguous memory to back the GART. > 2) mplayer plays video at about 2-3 times the correct rate and firefox-3.5 > also plays HTML5 video excessively fast. This is usually due to your audio hardware. Many radeons have HDMI ports, but in order to actually set the hardware up properly, you have to use the radeonhd driver with the Audio option set. mplayer appears to sync it's framerate to the audio and when the codec isn't initialized properly, it just plays full steam. If you aren't actually using HDMI, then all you need to do is set hw.snd.default_unit to point to your correct codec. Once sound works, mplayer should play at the correct rate. robert. > I'm not certain if this is Xorg or something else and will need to > investigate further. > > Any suggestions on either of these? -- Robert Noland <rnoland@FreeBSD.org> FreeBSD
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1270380027.4235.8.camel>