Hi-
After updating one of my test systems to v5.15-rc, I found that it becomes unresponsive during the later part of the boot process. A power-on reset is necessary to recover.
I bisected to this commit:
572994bf18ff ("drm/ast: Zero is missing in detect function")
Checking out v5.15-rc3 and reverting this commit enables the system to boot again.
0b:00.0 VGA compatible controller: ASPEED Technology, Inc. ASPEED Graphics Family (rev 30) (prog-if 00 [VGA controller]) DeviceName: ASPEED Video AST2400 Subsystem: Super Micro Computer Inc X10SRL-F Control: I/O+ Mem+ BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx- Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx- Interrupt: pin A routed to IRQ 18 Region 0: Memory at fa000000 (32-bit, non-prefetchable) [size=16M] Region 1: Memory at fb000000 (32-bit, non-prefetchable) [size=128K] Region 2: I/O ports at c000 [size=128] Expansion ROM at 000c0000 [virtual] [disabled] [size=128K] Capabilities: [40] Power Management version 3 Flags: PMEClk- DSI- D1+ D2+ AuxCurrent=375mA PME(D0+,D1+,D2+,D3hot+,D3cold+) Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME- Capabilities: [50] MSI: Enable- Count=1/4 Maskable- 64bit+ Address: 0000000000000000 Data: 0000 Kernel driver in use: ast Kernel modules: ast
-- Chuck Lever
(cc: ainux.wang@gmail.com)
Hi
Am 03.10.21 um 20:09 schrieb Chuck Lever III:
You don't have a monitor connected, I guess? In that case, we now trigger the helpers that poll for connected monitors. However, the overhead seems rather extreme.
I'll have to try to reproduce this, or otherwise we can revert the commit.
Best regards Thomas
Correct, my lab systems use IPMI and a browser-attached console.
In that case, we now trigger the helpers that poll for connected monitors. However, the overhead seems rather extreme.
I'll have to try to reproduce this, or otherwise we can revert the commit.
It's strange, only that system in my lab seems to have a problem. The others work fine.
Thanks for having a look!
-- Chuck Lever
Hi
Am 04.10.21 um 15:34 schrieb Chuck Lever III:
Is it a HW or FW problem? Maybe a different revision?
I'm asking because the problematic commit does the correct thing. If there is no VGA cable connected, the driver should poll until it detects one. The overhead should be minimal.
But I'll try to reproduce anyway.
Best regards Thomas
Hi
Am 04.10.21 um 16:11 schrieb Chuck Lever III:
v5.15-rc3 works well on my test machine.
For getting the firmware revisions, run
sudo dmidecode
on the machine. It will print a long list of devices with related information. Running
sudo lspci -v
will give information about the PCI devices. There's an entry for the VGA device somewhere. Maybe you can find some difference between the different systems
If you think the machine got stuck, try to plug-in the VGA cable during the boot and see if it makes the machine come up.
Best regards Thomas
Fwiw, here's a working system:
DMI: Supermicro X10SRA-F/X10SRA-F, BIOS 2.0b 06/12/2017
Handle 0x0054, DMI type 41, 11 bytes Onboard Device Reference Designation: Aspeed Video AST2400 Type: Video Status: Enabled Type Instance: 1 Bus Address: 0000:08:00.0
08:00.0 VGA compatible controller: ASPEED Technology, Inc. ASPEED Graphics Family (rev 30) (prog-if 00 [VGA controller]) DeviceName: Aspeed Video AST2400 Subsystem: Super Micro Computer Inc Device 0857
The system that hangs reports:
DMI: Supermicro Super Server/X10SRL-F, BIOS 3.3 10/28/2020
Handle 0x0027, DMI type 41, 11 bytes Onboard Device Reference Designation: ASPEED Video AST2400 Type: Video Status: Enabled Type Instance: 1 Bus Address: 0000:0b:00.0
0b:00.0 VGA compatible controller: ASPEED Technology, Inc. ASPEED Graphics Family (rev 30) (prog-if 00 [VGA controller]) DeviceName: ASPEED Video AST2400 Subsystem: Super Micro Computer Inc X10SRL-F
If you think the machine got stuck, try to plug-in the VGA cable during the boot and see if it makes the machine come up.
I will try to make time to give this a shot.
-- Chuck Lever
Yes, plugging in a physical monitor unsticks the machine and booting continues normally.
However, after that, having a monitor present does not seem to be necessary. The machine has been rebooted several times with v5.15-rc5 and no monitor attached, without any delays.
I'll note this is Fedora 32, in case you suspect there is a user space interaction involved. The system is going to be updated very soon to a more recent release of Fedora.
-- Chuck Lever
dri-devel@lists.freedesktop.org
-
Chuck Lever III -
Thomas Zimmermann