From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pasi =?iso-8859-1?Q?K=E4rkk=E4inen?= Subject: Re: [GIT PULL] devel/pat + devel/kms.fixes-0.5 / dma-api check_unmap() warning Date: Mon, 16 Aug 2010 20:50:37 +0300 Message-ID: <20100816175037.GS2804@reaktio.net> References: <20100806140609.GA3367@phenom.dumpdata.com> <20100815204125.GS2804@reaktio.net> <20100815205216.GU2804@reaktio.net> <20100816163133.GH12543@phenom.dumpdata.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20100816163133.GH12543@phenom.dumpdata.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Konrad Rzeszutek Wilk Cc: Jeremy Fitzhardinge , xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org On Mon, Aug 16, 2010 at 12:31:33PM -0400, Konrad Rzeszutek Wilk wrote: > > > Hello, > > > > > > I just tried the latest 2.6.32.19 kernel from xen/stable-2.6.32.x branch, > > > which now includes the kms.fixes-0.5, on my radeon/supermicro testbox. > > > > > > .. And KMS modesetting works properly now in Xen dom0! > > > My onboard ATI Radeon is the following: > > > > > > # lspci -v | grep VGA > > > 11:04.0 VGA compatible controller: ATI Technologies Inc ES1000 (rev 02) (prog-if 00 [VGA controller]) > > > > > > It seems also X starts OK, and I can use the Gnome desktop. > > > I could even run multiple OpenGL 3D applications! (ok, glxgears :) > > > > > > Thanks, great work! > > > > > > > Hmm, actually it seems I got this in dom0 dmesg: > > Good. I had seen this a month ago when I upgraded to FC13 (when it was > rawhide) and it disappeared once I upgraded to the real FC13. So I > blamed it on the alpha-rawhide X org driver. I could not reproduce this on > FC13 with an ES1000 in a SuperMicro (some X8DBN+). > > So, can you send to me, please: > > - lspci -vvvv > - cat /var/log/Xorg.0.log > - rpm -qa > - dmidecode > - dmesg > - xl dmesg > > That way I should have enough data to see if I can track this errant > issue down. > All of that is now here: http://pasik.reaktio.net/xen/pv_ops-dom0-debug/radeon-info-for-konrad/ > > > > ------------[ cut here ]------------ > > WARNING: at lib/dma-debug.c:802 check_unmap+0x18f/0x615() > > Hardware name: X7SB4/E > > NULL NULL: DMA-API: device driver tries to free DMA memory it has not allocated [device address=0x000000002eecc000] [size=4096 bytes] > > Modules linked in: ipt_MASQUERADE iptable_nat nf_nat bridge stp llc sunrpc ip6t_REJECT nf_conntrack_ipv6 ip6table_filter ip6_tables ipv6 xen_gntdev xen_evtchn xenfs uinput e1000e shpchp iTCO_wdt serio_raw i2c_i801 iTCO_vendor_support joydev pcspkr floppy usb_storage video aic79xx output scsi_transport_spi radeon ttm drm_kms_helper drm i2c_algo_bit i2c_core [last unloaded: scsi_wait_scan] > > Pid: 15, comm: events/0 Not tainted 2.6.32.19 #1 > > Call Trace: > > [] warn_slowpath_common+0x7c/0x94 > > [] warn_slowpath_fmt+0x41/0x43 > > [] ? check_unmap+0x100/0x615 > > [] ? xen_virt_to_bus+0x11/0x13 > > [] check_unmap+0x18f/0x615 > > [] ? xen_force_evtchn_callback+0xd/0xf > > [] ? check_events+0x12/0x20 > > [] debug_dma_free_coherent+0x6c/0x7a > > [] ttm_tt_free_alloced_pages+0x129/0x180 [ttm] > > [] ttm_tt_destroy+0x4d/0x8f [ttm] > > [] ttm_bo_release_list+0x96/0xd5 [ttm] > > [] ? ttm_bo_release_list+0x0/0xd5 [ttm] > > [] kref_put+0x43/0x4d > > [] ttm_bo_delayed_delete+0xad/0x121 [ttm] > > [] ttm_bo_delayed_workqueue+0x1f/0x34 [ttm] > > [] worker_thread+0x257/0x350 > > [] ? worker_thread+0x1ff/0x350 > > [] ? ttm_bo_delayed_workqueue+0x0/0x34 [ttm] > > [] ? autoremove_wake_function+0x0/0x39 > > [] ? worker_thread+0x0/0x350 > > [] kthread+0x7f/0x87 > > [] child_rip+0xa/0x20 > > [] ? restore_args+0x0/0x30 > > [] ? child_rip+0x0/0x20 > > ---[ end trace 84e813abe3a8bd4d ]--- > > fuse init (API version 7.13) > > end_request: I/O error, dev fd0, sector 0 > > end_request: I/O error, dev fd0, sector 0 > > DMA-API: debugging out of memory - disabling > > > > > > The system is up-to-date Fedora 13, with xen/stable-2.6.32.x dom0 kernel. > > > > Not fatal, things still seem to work.. > > Is this during shutdown or startup? > That warning/calltrace happens when starting X .. -- Pasi