RE: SMP+nosmp=hang [was: Re: 2.6.15-rc5-mm2]

RE: SMP+nosmp=hang [was: Re: 2.6.15-rc5-mm2]

[Brown, Len]

I think nosmp has been broken for a long time:

http://bugzilla.kernel.org/show_bug.cgi?id=1641

try maxcpus=1 instead of nosmp.

-Len

Re: SMP+nosmp=hang [was: Re: 2.6.15-rc5-mm2]

[Randy.Dunlap]

On Sun, 22 Jan 2006 16:47:24 -0500 Brown, Len wrote:

> I think nosmp has been broken for a long time:
> 
> http://bugzilla.kernel.org/show_bug.cgi?id=1641
> 
> try maxcpus=1 instead of nosmp.

Yes, I know that it's not recent.

Thanks for the bugz #.  I had already searched for that
but didn't find it.

---
~Randy

Re: SMP+nosmp=hang [was: Re: 2.6.15-rc5-mm2]

[Randy.Dunlap]

On Sun, 22 Jan 2006 00:46:36 +0100 J.A. Magallon wrote:

> On Fri, 20 Jan 2006 19:22:59 -0800, Andrew Morton <akpm@osdl.org> wrote:
> 
> > "J.A. Magallon" <jamagallon@able.es> wrote:
> > >
> > > On Sun, 11 Dec 2005 04:13:08 -0800, Andrew Morton <akpm@osdl.org> wrote:
> > > 
> > > > 
> > > > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.15-rc5/2.6.15-rc5-mm2/
> > > > 
> > > 
> > > Booting a SMP built kernel with 'nosmp' just hangs at the VFS layer, with
> > > the message about 'not being able to find root device sda1'.
> > > sda is a SATA drive on an Intel ICH5 controller:
> > > 
> > > libata version 1.20 loaded.
> > > ata_piix 0000:00:1f.2: version 1.05
> > > ACPI: PCI Interrupt 0000:00:1f.2[A] -> GSI 18 (level, low) -> IRQ 16
> > > PCI: Setting latency timer of device 0000:00:1f.2 to 64
> > > ata1: SATA max UDMA/133 cmd 0xC000 ctl 0xC402 bmdma 0xD000 irq 16
> > > ata2: SATA max UDMA/133 cmd 0xC800 ctl 0xCC02 bmdma 0xD008 irq 16
> > > ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4003 85:3469 86:3c01 87:4003 88:407f
> > > ata1: dev 0 ATA-6, max UDMA/133, 390721968 sectors: LBA48
> > > ata1: dev 0 configured for UDMA/133
> > > scsi0 : ata_piix
> > > ATA: abnormal status 0x7F on port 0xC807
> > > scsi1 : ata_piix
> > >   Vendor: ATA       Model: ST3200822AS       Rev: 3.01
> > >   Type:   Direct-Access                      ANSI SCSI revision: 05
> > > SCSI device sda: 390721968 512-byte hdwr sectors (200050 MB)
> > > SCSI device sda: drive cache: write back
> > > SCSI device sda: 390721968 512-byte hdwr sectors (200050 MB)
> > > SCSI device sda: drive cache: write back
> > >  sda: sda1 sda2 sda3
> > > sd 0:0:0:0: Attached scsi disk sda
> > > 
> > > I would have to double check, but I think it even missed the USB keyboard.
> > > 
> > 
> > Is this still happening?
> 
> Yes. I have just tried with 2.6.16-rc1-mm2, and the result is the same.
> No root device.
> 
> The nosmp-booted kernel looks much much slow than the SMP one, it takes
> ages to detect devices like usb ones, and even spent about 20 seconds here:
> 
> [    0.431601] libata version 1.20 loaded.
> [    0.431652] ata_piix 0000:00:1f.2: version 1.05
> [    0.431670] ACPI: PCI Interrupt 0000:00:1f.2[A] -> GSI 18 (level, low) -> IRQ 16
> [    0.431792] PCI: Setting latency timer of device 0000:00:1f.2 to 64
> [    0.431852] ata1: SATA max UDMA/133 cmd 0xC000 ctl 0xC402 bmdma 0xD000 irq 16
> [    0.431950] ata2: SATA max UDMA/133 cmd 0xC800 ctl 0xCC02 bmdma 0xD008 irq 16
>    >>>>>>>>>>>>> here <<<<<<<<<<<<<<<<<<
> [    0.690451] ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4003 85:3469 86:3c01 87:4003 88:407f
> [    0.690456] ata1: dev 0 ATA-6, max UDMA/133, 390721968 sectors: LBA48
> [    0.696278] ata1: dev 0 configured for UDMA/133
> [    0.701627] scsi0 : ata_piix
> [    1.937443] scsi1 : ata_piix
> 
> and it did not detect any drive, so the no-root-device error. This one can be
> a detection timeout, but as I say also the USB detection is dog slooow.
> 
> Need some info ? .config, or the like ?

Hi,
I've been testing this also.  I also see the problem.
I don't yet know what to do about it, but it's easy to
reproduce.  I'll continue to look at it.

For a temporary workaround, booting with "irqpoll" might help,
but that's not the real solution IMO.

---
~Randy

Re: SMP+nosmp=hang [was: Re: 2.6.15-rc5-mm2]

[J.A. Magallon]

[-- Attachment #1: Type: text/plain, Size: 3193 bytes --]

On Fri, 20 Jan 2006 19:22:59 -0800, Andrew Morton <akpm@osdl.org> wrote:

> "J.A. Magallon" <jamagallon@able.es> wrote:
> >
> > On Sun, 11 Dec 2005 04:13:08 -0800, Andrew Morton <akpm@osdl.org> wrote:
> > 
> > > 
> > > ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.15-rc5/2.6.15-rc5-mm2/
> > > 
> > 
> > Booting a SMP built kernel with 'nosmp' just hangs at the VFS layer, with
> > the message about 'not being able to find root device sda1'.
> > sda is a SATA drive on an Intel ICH5 controller:
> > 
> > libata version 1.20 loaded.
> > ata_piix 0000:00:1f.2: version 1.05
> > ACPI: PCI Interrupt 0000:00:1f.2[A] -> GSI 18 (level, low) -> IRQ 16
> > PCI: Setting latency timer of device 0000:00:1f.2 to 64
> > ata1: SATA max UDMA/133 cmd 0xC000 ctl 0xC402 bmdma 0xD000 irq 16
> > ata2: SATA max UDMA/133 cmd 0xC800 ctl 0xCC02 bmdma 0xD008 irq 16
> > ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4003 85:3469 86:3c01 87:4003 88:407f
> > ata1: dev 0 ATA-6, max UDMA/133, 390721968 sectors: LBA48
> > ata1: dev 0 configured for UDMA/133
> > scsi0 : ata_piix
> > ATA: abnormal status 0x7F on port 0xC807
> > scsi1 : ata_piix
> >   Vendor: ATA       Model: ST3200822AS       Rev: 3.01
> >   Type:   Direct-Access                      ANSI SCSI revision: 05
> > SCSI device sda: 390721968 512-byte hdwr sectors (200050 MB)
> > SCSI device sda: drive cache: write back
> > SCSI device sda: 390721968 512-byte hdwr sectors (200050 MB)
> > SCSI device sda: drive cache: write back
> >  sda: sda1 sda2 sda3
> > sd 0:0:0:0: Attached scsi disk sda
> > 
> > I would have to double check, but I think it even missed the USB keyboard.
> > 
> 
> Is this still happening?

Yes. I have just tried with 2.6.16-rc1-mm2, and the result is the same.
No root device.

The nosmp-booted kernel looks much much slow than the SMP one, it takes
ages to detect devices like usb ones, and even spent about 20 seconds here:

[    0.431601] libata version 1.20 loaded.
[    0.431652] ata_piix 0000:00:1f.2: version 1.05
[    0.431670] ACPI: PCI Interrupt 0000:00:1f.2[A] -> GSI 18 (level, low) -> IRQ 16
[    0.431792] PCI: Setting latency timer of device 0000:00:1f.2 to 64
[    0.431852] ata1: SATA max UDMA/133 cmd 0xC000 ctl 0xC402 bmdma 0xD000 irq 16
[    0.431950] ata2: SATA max UDMA/133 cmd 0xC800 ctl 0xCC02 bmdma 0xD008 irq 16
   >>>>>>>>>>>>> here <<<<<<<<<<<<<<<<<<
[    0.690451] ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4003 85:3469 86:3c01 87:4003 88:407f
[    0.690456] ata1: dev 0 ATA-6, max UDMA/133, 390721968 sectors: LBA48
[    0.696278] ata1: dev 0 configured for UDMA/133
[    0.701627] scsi0 : ata_piix
[    1.937443] scsi1 : ata_piix

and it did not detect any drive, so the no-root-device error. This one can be
a detection timeout, but as I say also the USB detection is dog slooow.

Need some info ? .config, or the like ?

--
J.A. Magallon <jamagallon()able!es>     \               Software is like sex:
werewolf!able!es                         \         It's better when it's free
Mandriva Linux release 2006.1 (Cooker) for i586
Linux 2.6.15-jam5 (gcc 4.0.2 (4.0.2-1mdk for Mandriva Linux release 2006.1))

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

SMP+nosmp=hang [was: Re: 2.6.15-rc5-mm2]

[J.A. Magallon]

[-- Attachment #1: Type: text/plain, Size: 1740 bytes --]

On Sun, 11 Dec 2005 04:13:08 -0800, Andrew Morton <akpm@osdl.org> wrote:

> 
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.15-rc5/2.6.15-rc5-mm2/
> 

Booting a SMP built kernel with 'nosmp' just hangs at the VFS layer, with
the message about 'not being able to find root device sda1'.
sda is a SATA drive on an Intel ICH5 controller:

libata version 1.20 loaded.
ata_piix 0000:00:1f.2: version 1.05
ACPI: PCI Interrupt 0000:00:1f.2[A] -> GSI 18 (level, low) -> IRQ 16
PCI: Setting latency timer of device 0000:00:1f.2 to 64
ata1: SATA max UDMA/133 cmd 0xC000 ctl 0xC402 bmdma 0xD000 irq 16
ata2: SATA max UDMA/133 cmd 0xC800 ctl 0xCC02 bmdma 0xD008 irq 16
ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4003 85:3469 86:3c01 87:4003 88:407f
ata1: dev 0 ATA-6, max UDMA/133, 390721968 sectors: LBA48
ata1: dev 0 configured for UDMA/133
scsi0 : ata_piix
ATA: abnormal status 0x7F on port 0xC807
scsi1 : ata_piix
  Vendor: ATA       Model: ST3200822AS       Rev: 3.01
  Type:   Direct-Access                      ANSI SCSI revision: 05
SCSI device sda: 390721968 512-byte hdwr sectors (200050 MB)
SCSI device sda: drive cache: write back
SCSI device sda: 390721968 512-byte hdwr sectors (200050 MB)
SCSI device sda: drive cache: write back
 sda: sda1 sda2 sda3
sd 0:0:0:0: Attached scsi disk sda

I would have to double check, but I think it even missed the USB keyboard.

Something really strange...

--
J.A. Magallon <jamagallon()able!es>     \               Software is like sex:
werewolf!able!es                         \         It's better when it's free
Mandriva Linux release 2006.1 (Cooker) for i586
Linux 2.6.14-jam4 (gcc 4.0.2 (4.0.2-1mdk for Mandriva Linux release 2006.1))

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]