Bug 79912

Summary: [NVE6] Multiple issues with 3 monitors and a GTX 670
Product: xorg Reporter: Tiziano Bacocco <tiziano.bacocco>
Component: Driver/nouveauAssignee: Nouveau Project <nouveau>
Status: RESOLVED MOVED QA Contact: Xorg Project Team <xorg-team>
Severity: major    
Priority: medium    
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:

Description Tiziano Bacocco 2014-06-11 09:16:16 UTC 
If i connect a third monitor on HDMI , on that one i get a purple vertical line on the left , DVI-D one does not work, and framerate of gnome-shell is very very low

On the kernel logs at boot i get 

[   25.163082] WARNING: CPU: 1 PID: 1188 at drivers/gpu/drm/ttm/ttm_bo_vm.c:269 ttm_bo_vm_open+0x52/0x80 [ttm]()
[   25.163084] Modules linked in: ccm xt_CHECKSUM iptable_mangle ipt_MASQUERADE iptable_nat nf_nat_ipv4 nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_conntrack nf_conntrack tun bnep bridge stp llc snd_hda_codec_hdmi x86_pkg_temp_thermal coretemp snd_hda_codec_realtek snd_hda_codec_generic snd_hda_intel kvm_intel snd_hda_codec snd_hwdep arc4 ath9k ath9k_common ath9k_hw snd_seq ath ir_lirc_codec ir_mce_kbd_decoder lirc_dev ir_sony_decoder ir_nec_decoder ir_sanyo_decoder ir_rc6_decoder ir_jvc_decoder mac80211 ir_rc5_decoder nouveau r820t kvm snd_seq_device snd_pcm iTCO_wdt rtl2832 cfg80211 crct10dif_pclmul iTCO_vendor_support dvb_usb_rtl28xxu rtl2830 i2c_algo_bit ttm crc32_pclmul btusb crc32c_intel dvb_usb_v2 eeepc_wmi asus_wmi snd_timer drm_kms_helper sparse_keymap snd drm ghash_clmulni_intel dvb_core soundcore
[   25.163138]  bluetooth i2c_i801 radio_ma901 tpm_tis tpm_infineon joydev videodev mei_me 6lowpan_iphc rfkill rc_core media i2c_core serio_raw tpm lpc_ich mei shpchp mfd_core microcode nfsd auth_rpcgss nfs_acl lockd sunrpc raid0 mxm_wmi e1000e ptp pps_core video wmi
[   25.163164] CPU: 1 PID: 1188 Comm: Xorg Tainted: G        W    3.14.5-200.fc20.x86_64 #1
[   25.163166] Hardware name: ASUS All Series/Z87-PRO, BIOS 0903 05/06/2013
[   25.163168]  0000000000000000 00000000c1415f90 ffff88030476bd90 ffffffff816efa42
[   25.163173]  0000000000000000 ffff88030476bdc8 ffffffff8108a1cd ffff880309ca6000
[   25.163179]  ffff880309d23ac8 ffff88030bd69f18 0000000000000000 0000000000000000
[   25.163183] Call Trace:
[   25.163189]  [<ffffffff816efa42>] dump_stack+0x45/0x56
[   25.163194]  [<ffffffff8108a1cd>] warn_slowpath_common+0x7d/0xa0
[   25.163199]  [<ffffffff8108a2fa>] warn_slowpath_null+0x1a/0x20
[   25.163209]  [<ffffffffa034b132>] ttm_bo_vm_open+0x52/0x80 [ttm]
[   25.163214]  [<ffffffff81088f17>] copy_process.part.26+0x18f7/0x1b90
[   25.163219]  [<ffffffff81089373>] do_fork+0xc3/0x370
[   25.163225]  [<ffffffff810896a6>] SyS_clone+0x16/0x20
[   25.163229]  [<ffffffff81700189>] stub_clone+0x69/0x90
[   25.163234]  [<ffffffff816ffe29>] ? system_call_fastpath+0x16/0x1b
[   25.163237] ---[ end trace da86bd3f343ad454 ]---
[   25.163241] ------------[ cut here ]------------


and when using it thousands of lines with
[  342.347721] nouveau E[   PDISP][0000:01:00.0] chid 0 mthd 0x0080 data 0x00000000 0x10005080 0x0000400d
[  342.360259] nouveau E[   PDISP][0000:01:00.0] chid 0 mthd 0x0080 data 0x00000000 0x10005080 0x0000400d
[  342.373450] nouveau E[   PDISP][0000:01:00.0] chid 0 mthd 0x0080 data 0x00000000 0x10005080 0x0000400d
[  342.386710] nouveau E[   PDISP][0000:01:00.0] chid 0 mthd 0x0080 data 0x00000000 0x10005080 0x0000400d
[  342.400035] nouveau E[   PDISP][0000:01:00.0] chid 0 mthd 0x0080 data 0x00000000 0x10005080 0x0000400d
[  342.412854] nouveau E[   PDISP][0000:01:00.0] chid 0 mthd 0x0080 data 0x00000000 0x10005080 0x0000400d
[  342.426119] nouveau E[   PDISP][0000:01:00.0] chid 0 mthd 0x0080 data 0x00000000 0x10005080 0x0000400d
[  343.290435] nouveau E[   PDISP][0000:01:00.0] chid 0 mthd 0x0080 data 0x00000000 0x10005080 0x0000400d
Comment 1 Tiziano Bacocco 2014-06-11 09:17:38 UTC 
Forgot to mention , i'm running Fedora 20 x86_64 , Kernel version 3.14.5-200
Comment 2 Martin Peres 2014-06-11 11:21:33 UTC 
Purple screen + extremely slow? You simply hit the memory bandwidth limit :s

It will be fixed by reclocking memory, efforts towards this goal are underway but won't be finished any time soon :s You may want to stick with the proprietary driver :s
Comment 3 Ilia Mirkin 2014-06-11 15:56:39 UTC 
If it's just memory that's too slow, you could try http://cgit.freedesktop.org/~airlied/linux/log/?h=drm-next which should allow you to try to change pstates via /sys/class/drm/card0/device/pstate. Note that this is very experimental, and is more than likely just going to hang your card. But worth a shot.
Comment 4 Ilia Mirkin 2014-11-20 20:32:28 UTC 
It seems like there are a bunch of different and unrelated issues here. HDMI + purple line has been fixed (in 3.18-rcX by commit a522946174b5ae389af0856c422d865a63c70316). That warning is weird... feels like perhaps you're running on a kernel with CMA enabled? If so, disable it.

Please test this out on 3.18-rc5 or later.
Comment 5 Martin Peres 2019-12-04 08:45:53 UTC 
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/driver/xf86-video-nouveau/issues/112.

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.