Xorg crashes randomly because of memory leak

Karol Herbst @karolherbst said:

this is already fixed upstream per xorg/xserver@74479a99

please ask your distribution to backport it.

Please reopen if this is indeed a different bug (but I highly doubt that)

Ilia Mirkin @imirkin said:

Additionally you're not using xf86-video-nouveau. (If you were, you wouldn't be using glamor.)

Karol Herbst @karolherbst said:

fyi: the patch was merged into the 1.20 xorg-server branch and should be in for 1.20.5

Peter Draganov said:

The problem persists in xorg-x11-server-Xorg-1.20.5-3.fc30.x86_64:

[113930.196] (EE) glamor0: GL error: GL_OUT_OF_MEMORY in glTexSubImage
[113930.198] (WW) glamor: Failed to allocate 5x1 FBO due to GL_OUT_OF_MEMORY.
[113930.198] (WW) glamor: Expect reduced performance.
[113930.198] (EE)
[113930.198] (EE) Backtrace:
[113930.235] (EE) 0: /usr/libexec/Xorg (OsLookupColor+0x13c) [0x5637893b8adc]
[113930.238] (EE) unw_get_proc_name failed: no unwind info found [-10]
[113930.238] (EE) 1: /lib64/libpthread.so.0 (?+0x0) [0x7f264909ee7f]
[113930.239] (EE) 2: /usr/lib64/xorg/modules/libglamoregl.so (glamor_get_pixmap_texture+0xa1) [0x7f2648726181]
[113930.240] (EE) 3: /usr/lib64/xorg/modules/libglamoregl.so (glamor_create_gc+0x7b84) [0x7f2648738814]
[113930.241] (EE) 4: /usr/lib64/xorg/modules/libglamoregl.so (glamor_create_gc+0x7f64) [0x7f2648738f84]
[113930.241] (EE) 5: /usr/lib64/xorg/modules/libglamoregl.so (glamor_create_gc+0x85bf) [0x7f264873975f]
[113930.242] (EE) 6: /usr/lib64/xorg/modules/libglamoregl.so (glamor_create_gc+0xa57a) [0x7f264873d34a]
[113930.242] (EE) 7: /usr/libexec/Xorg (DamageRegionAppend+0x6c0) [0x56378933a670]
[113930.243] (EE) 8: /usr/lib64/xorg/modules/libglamoregl.so (glamor_create_gc+0x1150a) [0x7f264874b92a]
[113930.243] (EE) 9: /usr/libexec/Xorg (AddTraps+0x4238) [0x563789332eb8]
[113930.243] (EE) 10: /usr/libexec/Xorg (SendErrorToClient+0x354) [0x563789259d54]
[113930.243] (EE) 11: /usr/libexec/Xorg (InitFonts+0x3b4) [0x56378925de14]
[113930.244] (EE) unw_get_proc_name failed: no unwind info found [-10]
[113930.244] (EE) 12: /lib/libc.so.6 (?+0x0) [0x7f2648ee9e40]
[113930.244] (EE) 13: /usr/libexec/Xorg (_start+0x2e) [0x56378924786e]
[113930.244] (EE)
[113930.244] (EE) Segmentation fault at address 0x0
[113930.244] (EE)
Fatal server error:
[113930.244] (EE) Caught signal 11 (Segmentation fault). Server aborting

Peter Draganov said:

@Ilia Mirkin
Additionally you're not using xf86-video-nouveau. (If you were, you wouldn't be using glamor.)
I am not sure about this, but here is the log for nouveau:
[ 148.426] (II) modeset(0): [DRI2] DRI driver: nouveau
[ 148.426] (II) modeset(0): [DRI2] VDPAU driver: nouveau
...
[ 148.441] (II) Initializing extension GLX
[ 148.473] (II) AIGLX: Loaded and initialized nouveau
[ 148.473] (II) GLX: Initialized DRI2 GL provider for screen 0

Karol Herbst @karolherbst said:

(In reply to Peter Draganov from comment 4)

The problem persists in xorg-x11-server-Xorg-1.20.5-3.fc30.x86_64:

[113930.196] (EE) glamor0: GL error: GL_OUT_OF_MEMORY in glTexSubImage
[113930.198] (WW) glamor: Failed to allocate 5x1 FBO due to GL_OUT_OF_MEMORY.
[113930.198] (WW) glamor: Expect reduced performance.
[113930.198] (EE)
[113930.198] (EE) Backtrace:
[113930.235] (EE) 0: /usr/libexec/Xorg (OsLookupColor+0x13c) [0x5637893b8adc]
[113930.238] (EE) unw_get_proc_name failed: no unwind info found [-10]
[113930.238] (EE) 1: /lib64/libpthread.so.0 (?+0x0) [0x7f264909ee7f]
[113930.239] (EE) 2: /usr/lib64/xorg/modules/libglamoregl.so
(glamor_get_pixmap_texture+0xa1) [0x7f2648726181]
[113930.240] (EE) 3: /usr/lib64/xorg/modules/libglamoregl.so
(glamor_create_gc+0x7b84) [0x7f2648738814]
[113930.241] (EE) 4: /usr/lib64/xorg/modules/libglamoregl.so
(glamor_create_gc+0x7f64) [0x7f2648738f84]
[113930.241] (EE) 5: /usr/lib64/xorg/modules/libglamoregl.so
(glamor_create_gc+0x85bf) [0x7f264873975f]
[113930.242] (EE) 6: /usr/lib64/xorg/modules/libglamoregl.so
(glamor_create_gc+0xa57a) [0x7f264873d34a]
[113930.242] (EE) 7: /usr/libexec/Xorg (DamageRegionAppend+0x6c0)
[0x56378933a670]
[113930.243] (EE) 8: /usr/lib64/xorg/modules/libglamoregl.so
(glamor_create_gc+0x1150a) [0x7f264874b92a]
[113930.243] (EE) 9: /usr/libexec/Xorg (AddTraps+0x4238) [0x563789332eb8]
[113930.243] (EE) 10: /usr/libexec/Xorg (SendErrorToClient+0x354)
[0x563789259d54]
[113930.243] (EE) 11: /usr/libexec/Xorg (InitFonts+0x3b4) [0x56378925de14]
[113930.244] (EE) unw_get_proc_name failed: no unwind info found [-10]
[113930.244] (EE) 12: /lib/libc.so.6 (?+0x0) [0x7f2648ee9e40]
[113930.244] (EE) 13: /usr/libexec/Xorg (_start+0x2e) [0x56378924786e]
[113930.244] (EE)
[113930.244] (EE) Segmentation fault at address 0x0
[113930.244] (EE)
Fatal server error:
[113930.244] (EE) Caught signal 11 (Segmentation fault). Server aborting

huh, that's weird. Will try to look into it.

Ilia Mirkin @imirkin said:

(In reply to Peter Draganov from comment 5)

@Ilia Mirkin
Additionally you're not using xf86-video-nouveau. (If you were, you wouldn't
be using glamor.)
I am not sure about this, but here is the log for nouveau:
[ 148.426] (II) modeset(0): [DRI2] DRI driver: nouveau
[ 148.426] (II) modeset(0): [DRI2] VDPAU driver: nouveau
...
[ 148.441] (II) Initializing extension GLX
[ 148.473] (II) AIGLX: Loaded and initialized nouveau
[ 148.473] (II) GLX: Initialized DRI2 GL provider for screen 0

See how it says "modeset(0)" and not "NOUVEAU(0)"? That means you're not using xf86-video-nouveau.

Olivier Fourdan @ofourdan said:

Out of curiosity, any hint on how to reproduce the issue?

What clients were running at the time the issue occurred
Any special x11 client that can reproduce?
Any particular action that triggers the issue?
which desktop environment/window manager?

Basically, anything that could help reproducing...

Peter Draganov said:

What clients were running at the time the issue occurred
Skype, System Monitor, Firefox, Mate Terminal, Libre Office, Caja, TigerVNC, Virtualbox, Pluma
Any special x11 client that can reproduce?
have no idea
Any particular action that triggers the issue?
Usually crashes during the night, when a black screen saver is running, but it happened several times during the day, when I worked
which desktop environment/window manager?
LightDM/MATE

Basically, anything that could help reproducing...
Just leave X running long enough time - usually after 2-3 days without restart

Karol Herbst @karolherbst said:

(In reply to Peter Draganov from comment 9)

What clients were running at the time the issue occurred
Skype, System Monitor, Firefox, Mate Terminal, Libre Office, Caja, TigerVNC,
Virtualbox, Pluma

Any special x11 client that can reproduce?
have no idea

Any particular action that triggers the issue?
Usually crashes during the night, when a black screen saver is running, but
it happened several times during the day, when I worked

which desktop environment/window manager?
LightDM/MATE

Basically, anything that could help reproducing...
Just leave X running long enough time - usually after 2-3 days without
restart

what GPU do you have? I am especially interested in the amount of VRAM it has.

Peter Draganov said:

what GPU do you have? I am especially interested in the amount of VRAM it has.
01:00.0 VGA compatible controller: NVIDIA Corporation GT218 [GeForce 210] (rev a2) (prog-if 00 [VGA controller])
Subsystem: Micro-Star International Co., Ltd. [MSI] N210 [Geforce 210] PCIe graphics adapter
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR+ FastB2B- DisINTx+
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
Latency: 0, Cache Line Size: 64 bytes
Interrupt: pin A routed to IRQ 31
NUMA node: 0
Region 0: Memory at fa000000 (32-bit, non-prefetchable) [size=16M]
Region 1: Memory at d0000000 (64-bit, prefetchable) [size=256M]
Region 3: Memory at ce000000 (64-bit, prefetchable) [size=32M]
Region 5: I/O ports at ec00 [size=128]
Expansion ROM at 000c0000 [disabled] [size=128K]
Capabilities: [60] Power Management version 3
Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME-
Capabilities: [68] MSI: Enable+ Count=1/1 Maskable- 64bit+
Address: 00000000fee02004 Data: 4026
Capabilities: [78] Express (v2) Endpoint, MSI 00
DevCap: MaxPayload 128 bytes, PhantFunc 0, Latency L0s unlimited, L1 <64us
ExtTag+ AttnBtn- AttnInd- PwrInd- RBE+ FLReset- SlotPowerLimit 75.000W
DevCtl: CorrErr- NonFatalErr- FatalErr- UnsupReq-
RlxdOrd- ExtTag+ PhantFunc- AuxPwr- NoSnoop+
MaxPayload 128 bytes, MaxReadReq 512 bytes
DevSta: CorrErr- NonFatalErr- FatalErr- UnsupReq- AuxPwr- TransPend-
LnkCap: Port #0, Speed 2.5GT/s, Width x16, ASPM L0s L1, Exit Latency L0s <256ns, L1 <4us
ClockPM+ Surprise- LLActRep- BwNot- ASPMOptComp-
LnkCtl: ASPM Disabled; RCB 128 bytes Disabled- CommClk+
ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt-
LnkSta: Speed 2.5GT/s (ok), Width x16 (ok)
TrErr- Train- SlotClk+ DLActive- BWMgmt- ABWMgmt-
DevCap2: Completion Timeout: Not Supported, TimeoutDis+, LTR-, OBFF Not Supported
AtomicOpsCap: 32bit- 64bit- 128bitCAS-
DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis-, LTR-, OBFF Disabled
AtomicOpsCtl: ReqEn-
LnkCtl2: Target Link Speed: 2.5GT/s, EnterCompliance- SpeedDis-
Transmit Margin: Normal Operating Range, EnterModifiedCompliance- ComplianceSOS-
Compliance De-emphasis: -6dB
LnkSta2: Current De-emphasis Level: -6dB, EqualizationComplete-, EqualizationPhase1-
EqualizationPhase2-, EqualizationPhase3-, LinkEqualizationRequest-
Capabilities: [b4] Vendor Specific Information: Len=14 >
Capabilities: [100 v1] Virtual Channel
Caps: LPEVC=0 RefClk=100ns PATEntryBits=1
Arb: Fixed- WRR32- WRR64- WRR128-
Ctrl: ArbSelect=Fixed
Status: InProgress-
VC0: Caps: PATOffset=00 MaxTimeSlots=1 RejSnoopTrans-
Arb: Fixed- WRR32- WRR64- WRR128- TWRR128- WRR256-
Ctrl: Enable+ ID=0 ArbSelect=Fixed TC/VC=ff
Status: NegoPending- InProgress-
Capabilities: [128 v1] Power Budgeting >
Capabilities: [600 v1] Vendor Specific Information: ID=0001 Rev=1 Len=024 <?>
Kernel driver in use: nouveau
Kernel modules: nouveau

I was with another old NVidia card when crashes started, so I decided to replace it, but this did not help. I was with Fedora 22 intil April and started upgrading it every day. Crashes started when I reached Fedora 24-25-26 probably and continues until last Fedora 30, if this can help.

Olivier Fourdan @ofourdan said:

Out of curiosity, can you also check with xrestop, after leaving your session running for some time, if any X11 client tend to be leaking or using an insane amount of X11 resources?

Peter Draganov said:

After nearly 1 day and 3 hours:
xrestop - Display: localhost:0
HMonitoring 51 clients. XErrors: 0
Pixmaps: 1180312K total, Other: 221K total, All: 1180534K total

res-base Wins GCs Fnts Pxms Misc Pxm mem Other Total PID Identifier
6200000 227 4 1 1242 1801 800547K 48K 800596K 9592 Linux1.ods - LibreOffice Calc
0c00000 43 4 2 94 4175 204916K 100K 205017K 2275 marco
3200000 106 74 1 86 325 117024K 12K 117037K 3072 phpPgAdmin - Mozilla Firefox
5a00000 17 3 1 6 81 17307K 3K 17310K 3891 pdraganov@bgva-f-00:~
1000000 8 3 1 27 59 10417K 2K 10420K 2303 x-caja-desktop
4600000 30 2 1 3 93 9990K 3K 9993K 3008 7 [Running] - Oracle VM VirtualBox
3e00000 2 1 0 3 28 9400K 744B 9401K ? <unknown>
0000000 2 0 2 0 277 3413K 8K 3421K ? <unknown>
6000000 0 0 0 1 0 3413K 0B 3413K ? <unknown>

Probably LibreOffice is the problem? I have 18 documents opened at the moment.

Olivier Fourdan @ofourdan said:

(In reply to Peter Draganov from comment 13)

Probably LibreOffice is the problem? I have 18 documents opened at the
moment.

Not necessarily the problem, no, unless it increases over time and never decreases... The values do not seem out of proportions either, I mean, I can get higher pixmap usage opening a few documents in LibreOffice here as well...

I'm just trying to figure if something is leaking, because maybe we look in the wrong place, fixing the root cause of GL_OUT_OF_MEMORY might be more efficient than trying to deal with the aftermaths.

Joel Polowin said:

FWIW, I'm seeing the same symptoms -- I found this report while searching for
info about what was going on. The only notable apps running, so far as I'm
aware, are VirtualBox and Firefox (along with various terminal and file windows).
I use OpenOffice, but infrequently. I definitely didn't have any documents
open at the time of the most recent crash.

I'm running 5.1.15-300.fc30.x86_64 . I'm willing to send info if it will help,
but I must warn you that I'm anything but a power user -- I'd need some coaching
to find files.

Joel Polowin said:

(In reply to Joel Polowin from comment 15)

I use OpenOffice, but infrequently.

Sorry -- LibreOffice. Old habits.

Peter Draganov uploaded an attachment:

Thank you for your e-mail. I am out of the office and will return on 16th of August 2019. I will have no access to my e-mails during this time. If your e-mail requires urgent attention please e-mail service@taxback.com. Alternatively I will reply to your e-mail when I am back. Kind Regards, Peter Draganov ICT Systems Administrator Email: pdraganov@taxback.com Web: www.taxback.com http://www.taxback.com<www.taxback.com%3chttp:/www.taxback.com%3e> Phone: +359 52 919190 Fax: +359 5268 6883 Mobile: +359 889 530589 Contact Us: Taxback.com is an ISO 9001 certified company with 25 offices in 20 countries. This email is confidential and intended solely for the use of the individual to whom it is addressed. Any views or opinions presented are solely those of the author and do not necessarily represent those of taxback.com. If you are not the intended recipient, be advised that you have received this email in error and that any use, dissemination, forwarding, printing, or copying of this email is strictly prohibited. If you have received this email in error please forward it to info@taxback.com.

Attachment 144844, "attachment-29996-0.html":
attachment-29996-0.html

Joel Polowin said:

(In reply to Joel Polowin from comment 15)

The only notable apps running, so far as I'm aware, are VirtualBox and Firefox
(along with various terminal and file windows).

Hmm. And Chromium, now that I think of it.

John Lindgren said:

I just saw this same crash (Xorg 1.20.4 here still though). I also had Firefox, VirtualBox, Evolution, and a few smaller applications running. The computer had been running for most of the week already, so maybe a memory leak somewhere?

My card is:

01:00.0 VGA compatible controller: NVIDIA Corporation GM107GLM [Quadro M2000M] (rev a2)

I am using xf86-video-modesetting. Should I be using xf86-video-nouveau instead?

John Lindgren uploaded an attachment:

Attachment 145281, "My Xorg log":
Xorg.0.log.old

Ilia Mirkin @imirkin said:

(In reply to John Lindgren from comment 19)

I am using xf86-video-modesetting. Should I be using xf86-video-nouveau
instead?

I would definitely recommend -nouveau over -modesetting.

Andrey Sychev uploaded an attachment:

Attachment 145675, "My Xorg.0.log.old":
Xorg.0.log.old

Andrey Sychev said:

I can confirm that bug still persists.

My OS: Debian (Buster)
Desktop: Xfce
Apps on which crash happens: Meld (frequently, but not every time). Libre office (rare)

In attepmt to fix crush
I have downloaded and installed

https://launchpad.net/ubuntu/+source/xorg-server/2:1.20.5+git20190820-0ubuntu3

but this does not help.

I can say that in my case crash does not happens after first message:

glamor0: GL error: GL_OUT_OF_MEMORY in glTexSubImage

Crash happens only after several similar messages.

mentioned in issue xorg/xserver#647 (closed)

Are there any updates on this? I've been having the same issue. It seems to be triggered by something in glamor. I created this bug report with a stack trace of the xorg crash: xorg/xserver#957. It seems to be different from the ones reported here (glamor_bind_texture).

mentioned in issue xorg/xserver#1329 (moved)

mentioned in issue mesa/mesa#6197

Fixed in linux 5.14

closed

Xorg crashes randomly because of memory leak

Submitted by Peter Draganov

Description

Designs

Child items ...

Activity

Admin message

Admin message

Xorg crashes randomly because of memory leak

Submitted by Peter Draganov

Description

Activity