On Wed, 26 Oct 2016, Bjorn Helgaas wrote:
> Hi Meelis,
>
> On Wed, Oct 07, 2015 at 12:25:19PM +0300, Meelis Roos wrote:
> > First I noticed that on Sun Netra t1-150 (sparc64 CPU), PCI allocation
> > was failing for qla1280 scsi driver. Reading dmesg uncovered "can't
> > claim BAR" errors.
On Wed, 26 Oct 2016, Bjorn Helgaas wrote:
> Hi Meelis,
>
> On Wed, Oct 07, 2015 at 12:25:19PM +0300, Meelis Roos wrote:
> > First I noticed that on Sun Netra t1-150 (sparc64 CPU), PCI allocation
> > was failing for qla1280 scsi driver. Reading dmesg uncovered "can't
> > claim BAR" errors.
Hi Meelis,
On Wed, Oct 07, 2015 at 12:25:19PM +0300, Meelis Roos wrote:
> First I noticed that on Sun Netra t1-150 (sparc64 CPU), PCI allocation
> was failing for qla1280 scsi driver. Reading dmesg uncovered "can't
> claim BAR" errors. Looking for this in my other test machines I see more
> of
Hi Meelis,
On Wed, Oct 07, 2015 at 12:25:19PM +0300, Meelis Roos wrote:
> First I noticed that on Sun Netra t1-150 (sparc64 CPU), PCI allocation
> was failing for qla1280 scsi driver. Reading dmesg uncovered "can't
> claim BAR" errors. Looking for this in my other test machines I see more
> of
On Wed, Oct 14, 2015 at 12:34 AM, Meelis Roos wrote:
>> > 0001:00:07.0 ISA bridge: ULi Electronics Inc. M1533/M1535/M1543 PCI to ISA
>> > Bridge [Aladdin IV/V/V+]
>> > Control: I/O+ Mem+ BusMaster+ SpecCycle+ MemWINV- VGASnoop-
>> > ParErr- Stepping- SERR- FastB2B- DisINTx-
>> >
> > 0001:00:07.0 ISA bridge: ULi Electronics Inc. M1533/M1535/M1543 PCI to ISA
> > Bridge [Aladdin IV/V/V+]
> > Control: I/O+ Mem+ BusMaster+ SpecCycle+ MemWINV- VGASnoop- ParErr-
> > Stepping- SERR- FastB2B- DisINTx-
> > Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium
> > 0001:00:07.0 ISA bridge: ULi Electronics Inc. M1533/M1535/M1543 PCI to ISA
> > Bridge [Aladdin IV/V/V+]
> > Control: I/O+ Mem+ BusMaster+ SpecCycle+ MemWINV- VGASnoop- ParErr-
> > Stepping- SERR- FastB2B- DisINTx-
> > Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium
On Wed, Oct 14, 2015 at 12:34 AM, Meelis Roos wrote:
>> > 0001:00:07.0 ISA bridge: ULi Electronics Inc. M1533/M1535/M1543 PCI to ISA
>> > Bridge [Aladdin IV/V/V+]
>> > Control: I/O+ Mem+ BusMaster+ SpecCycle+ MemWINV- VGASnoop-
>> > ParErr- Stepping- SERR- FastB2B-
On Tue, Oct 13, 2015 at 1:39 PM, Meelis Roos wrote:
>
> 0001:00:07.0 ISA bridge: ULi Electronics Inc. M1533/M1535/M1543 PCI to ISA
> Bridge [Aladdin IV/V/V+]
> Control: I/O+ Mem+ BusMaster+ SpecCycle+ MemWINV- VGASnoop- ParErr-
> Stepping- SERR- FastB2B- DisINTx-
> Status: Cap+
> >> Can you send boot log after booting with "debug ignore_loglevel
> >> ofpci_debug=1" ?
> >
> > http://kodu.ut.ee/~mroos/dm/dm.v210+patch+debug
>
> Firmware does reported overlapped resources.
[...]
> can you send out output for
> lspci -vvxxx
> lspci -tv
Here you are.
:00:02.0
On Tue, Oct 13, 2015 at 12:32 AM, Meelis Roos wrote:
>>
>> Can you send boot log after booting with "debug ignore_loglevel
>> ofpci_debug=1" ?
>
> http://kodu.ut.ee/~mroos/dm/dm.v210+patch+debug
Firmware does reported overlapped resources.
[ 34.150035] PCI: scan_bus[/pci@1e,60] bus no 0
On Tue, Oct 13, 2015 at 12:32 AM, Meelis Roos wrote:
> Applied the fix_mem_mem64_overlapping.patch and retested. The BAR
> allocation problems are still there:
>
> http://kodu.ut.ee/~mroos/dm/dm.v210+patch+mem64patch+debug
Good, now MEM64 is gone.
[ 34.145275] /pci@1f,70: TOMATILLO PCI
> > Updated on V210 - the others were more recent. Now I get more BAR
> > warning there too:
> >
> > http://kodu.ut.ee/~mroos/dm/dm.v210+patches
> >
>
> interesting, mem and mem64 is overlapping.
>
> [ 38.824625] /pci@1f,70: TOMATILLO PCI Bus Module ver[4:0]
> [ 38.824654]
On Mon, Oct 12, 2015 at 10:43 PM, Yinghai Lu wrote:
> On Mon, Oct 12, 2015 at 10:06 PM, Meelis Roos wrote:
>>> >> > sparc64 machines:
>>
>> Updated on V210 - the others were more recent. Now I get more BAR
>> warning there too:
>>
>> http://kodu.ut.ee/~mroos/dm/dm.v210+patches
>>
>
>
On Mon, Oct 12, 2015 at 10:06 PM, Meelis Roos wrote:
>> >> > sparc64 machines:
>
> Updated on V210 - the others were more recent. Now I get more BAR
> warning there too:
>
> http://kodu.ut.ee/~mroos/dm/dm.v210+patches
other warning:
[ 38.829890] pci 0001:00:07.0: can't claim BAR 2 [mem
> > Updated on V210 - the others were more recent. Now I get more BAR
> > warning there too:
> >
> > http://kodu.ut.ee/~mroos/dm/dm.v210+patches
> >
>
> interesting, mem and mem64 is overlapping.
>
> [ 38.824625] /pci@1f,70: TOMATILLO PCI Bus Module ver[4:0]
> [ 38.824654]
> >> Can you send boot log after booting with "debug ignore_loglevel
> >> ofpci_debug=1" ?
> >
> > http://kodu.ut.ee/~mroos/dm/dm.v210+patch+debug
>
> Firmware does reported overlapped resources.
[...]
> can you send out output for
> lspci -vvxxx
> lspci -tv
Here you are.
:00:02.0
On Tue, Oct 13, 2015 at 12:32 AM, Meelis Roos wrote:
>>
>> Can you send boot log after booting with "debug ignore_loglevel
>> ofpci_debug=1" ?
>
> http://kodu.ut.ee/~mroos/dm/dm.v210+patch+debug
Firmware does reported overlapped resources.
[ 34.150035] PCI:
On Tue, Oct 13, 2015 at 12:32 AM, Meelis Roos wrote:
> Applied the fix_mem_mem64_overlapping.patch and retested. The BAR
> allocation problems are still there:
>
> http://kodu.ut.ee/~mroos/dm/dm.v210+patch+mem64patch+debug
Good, now MEM64 is gone.
[ 34.145275] /pci@1f,70:
On Tue, Oct 13, 2015 at 1:39 PM, Meelis Roos wrote:
>
> 0001:00:07.0 ISA bridge: ULi Electronics Inc. M1533/M1535/M1543 PCI to ISA
> Bridge [Aladdin IV/V/V+]
> Control: I/O+ Mem+ BusMaster+ SpecCycle+ MemWINV- VGASnoop- ParErr-
> Stepping- SERR- FastB2B- DisINTx-
>
On Mon, Oct 12, 2015 at 10:06 PM, Meelis Roos wrote:
>> >> > sparc64 machines:
>
> Updated on V210 - the others were more recent. Now I get more BAR
> warning there too:
>
> http://kodu.ut.ee/~mroos/dm/dm.v210+patches
other warning:
[ 38.829890] pci 0001:00:07.0: can't claim
On Mon, Oct 12, 2015 at 10:43 PM, Yinghai Lu wrote:
> On Mon, Oct 12, 2015 at 10:06 PM, Meelis Roos wrote:
>>> >> > sparc64 machines:
>>
>> Updated on V210 - the others were more recent. Now I get more BAR
>> warning there too:
>>
>>
On Mon, Oct 12, 2015 at 10:06 PM, Meelis Roos wrote:
>> >> > sparc64 machines:
>> >>
>> >> git://git.kernel.org/pub/scm/linux/kernel/git/yinghai/linux-yinghai.git
>> >> for-pci-v4.4-next
>> >>
>> >> It should fix some "no compatible bridge window"
>> >
>> > Blade 100: still has 2 address
> >> > sparc64 machines:
> >>
> >> git://git.kernel.org/pub/scm/linux/kernel/git/yinghai/linux-yinghai.git
> >> for-pci-v4.4-next
> >>
> >> It should fix some "no compatible bridge window"
> >
> > Blade 100: still has 2 address conflicts:
> > http://kodu.ut.ee/~mroos/dm/dm.sb100+patch
> >
> >
On Mon, Oct 12, 2015 at 6:04 AM, Meelis Roos wrote:
>> > sparc64 machines:
>>
>> git://git.kernel.org/pub/scm/linux/kernel/git/yinghai/linux-yinghai.git
>> for-pci-v4.4-next
>>
>> It should fix some "no compatible bridge window"
>
> Blade 100: still has 2 address conflicts:
>
> > sparc64 machines:
> >
> > http://kodu.ut.ee/~mroos/dm/dm.t1-105
> > http://kodu.ut.ee/~mroos/dm/dm.t1-200
> > http://kodu.ut.ee/~mroos/dm/dm.v120
> > http://kodu.ut.ee/~mroos/dm/dm.v210
> > http://kodu.ut.ee/~mroos/dm/dm.v240
> > http://kodu.ut.ee/~mroos/dm/dm.sb100
>
> Please try
>
>
> > sparc64 machines:
> >
> > http://kodu.ut.ee/~mroos/dm/dm.t1-105
> > http://kodu.ut.ee/~mroos/dm/dm.t1-200
> > http://kodu.ut.ee/~mroos/dm/dm.v120
> > http://kodu.ut.ee/~mroos/dm/dm.v210
> > http://kodu.ut.ee/~mroos/dm/dm.v240
> > http://kodu.ut.ee/~mroos/dm/dm.sb100
>
> Please try
>
>
On Mon, Oct 12, 2015 at 6:04 AM, Meelis Roos wrote:
>> > sparc64 machines:
>>
>> git://git.kernel.org/pub/scm/linux/kernel/git/yinghai/linux-yinghai.git
>> for-pci-v4.4-next
>>
>> It should fix some "no compatible bridge window"
>
> Blade 100: still has 2 address conflicts:
>
> >> > sparc64 machines:
> >>
> >> git://git.kernel.org/pub/scm/linux/kernel/git/yinghai/linux-yinghai.git
> >> for-pci-v4.4-next
> >>
> >> It should fix some "no compatible bridge window"
> >
> > Blade 100: still has 2 address conflicts:
> > http://kodu.ut.ee/~mroos/dm/dm.sb100+patch
> >
> >
On Mon, Oct 12, 2015 at 10:06 PM, Meelis Roos wrote:
>> >> > sparc64 machines:
>> >>
>> >> git://git.kernel.org/pub/scm/linux/kernel/git/yinghai/linux-yinghai.git
>> >> for-pci-v4.4-next
>> >>
>> >> It should fix some "no compatible bridge window"
>> >
>> > Blade 100: still has 2
On Wed, Oct 07, 2015 at 04:16:37PM -0700, Yinghai Lu wrote:
> On Wed, Oct 7, 2015 at 2:25 AM, Meelis Roos wrote:
> > amd64 machine:
> >
> > http://kodu.ut.ee/~mroos/dm/dm.x2100
>
> [0.156360] bus: [bus 00-05] on node 0 link 0
> [0.156362] bus: 00 [io 0x-0x]
> [0.156364] bus:
On Wed, Oct 07, 2015 at 04:16:37PM -0700, Yinghai Lu wrote:
> On Wed, Oct 7, 2015 at 2:25 AM, Meelis Roos wrote:
> > amd64 machine:
> >
> > http://kodu.ut.ee/~mroos/dm/dm.x2100
>
> [0.156360] bus: [bus 00-05] on node 0 link 0
> [0.156362] bus: 00 [io 0x-0x]
> [
On Wed, Oct 7, 2015 at 2:25 AM, Meelis Roos wrote:
> amd64 machine:
>
> http://kodu.ut.ee/~mroos/dm/dm.x2100
[0.156360] bus: [bus 00-05] on node 0 link 0
[0.156362] bus: 00 [io 0x-0x]
[0.156364] bus: 00 [mem 0x000a-0x000b]
[0.156365] bus: 00 [mem
On Wed, Oct 7, 2015 at 2:25 AM, Meelis Roos wrote:
> First I noticed that on Sun Netra t1-150 (sparc64 CPU), PCI allocation
> was failing for qla1280 scsi driver. Reading dmesg uncovered "can't
> claim BAR" errors. Looking for this in my other test machines I see more
> of them - on many sparc64
First I noticed that on Sun Netra t1-150 (sparc64 CPU), PCI allocation
was failing for qla1280 scsi driver. Reading dmesg uncovered "can't
claim BAR" errors. Looking for this in my other test machines I see more
of them - on many sparc64 machines and some on x86.
In another thread for
On Wed, Oct 7, 2015 at 2:25 AM, Meelis Roos wrote:
> First I noticed that on Sun Netra t1-150 (sparc64 CPU), PCI allocation
> was failing for qla1280 scsi driver. Reading dmesg uncovered "can't
> claim BAR" errors. Looking for this in my other test machines I see more
> of them -
On Wed, Oct 7, 2015 at 2:25 AM, Meelis Roos wrote:
> amd64 machine:
>
> http://kodu.ut.ee/~mroos/dm/dm.x2100
[0.156360] bus: [bus 00-05] on node 0 link 0
[0.156362] bus: 00 [io 0x-0x]
[0.156364] bus: 00 [mem 0x000a-0x000b]
[0.156365] bus: 00 [mem
First I noticed that on Sun Netra t1-150 (sparc64 CPU), PCI allocation
was failing for qla1280 scsi driver. Reading dmesg uncovered "can't
claim BAR" errors. Looking for this in my other test machines I see more
of them - on many sparc64 machines and some on x86.
In another thread for
38 matches
Mail list logo