Re: inconsistent pci probing

SourceForge Headquarters 1320 Columbia Street Suite 310 San Diego, CA 92101 +1 (858) 422-6466

Hi Michel

Am 10-Aug-02 schrieb Michel D=E4nzer:

> On Sat, 2002-08-10 at 01:35, Andreas W=FCst wrote:
>>=20
>> Now running quite a long time with kernel 2.4.17-3 I noticed some
>> interesting kernel logging concerning the pci probes during boot. Here=
's
>> what I saw:
>>=20
>>=20
>> Aug 8 21:41:04 andrewI kernel: PCI: Probing PCI hardware Aug 8 21:41:0=
4
>> andrewI kernel: Memory resource not set for host bridge 0 Aug 8 21:41:=
04
>> andrewI kernel: apus_pcibios_fixup: PCI mem resource requested Aug 8
>> 21:41:04 andrewI kernel: PCI: Cannot allocate resource region 1 of PCI
>> bridge 0 Aug 8 21:41:04 andrewI kernel: PCI: resource is e0000000..fff=
c0000
>> (200), parent c023fc1c Aug 8 21:41:04 andrewI kernel: PCI:00:01.0: Res=
ource
>> 0: ef000000-ef01ffff (f=3D200) Aug 8 21:41:04 andrewI kernel: PCI:00:0=
1.0:
>> Resource 1: e0000000-e07fffff (f=3D200) Aug 8 21:41:04 andrewI kernel:
>> PCI:00:01.0: Resource 2: e1000000-e17fffff (f=3D200) Aug 8 21:41:04 an=
drewI
>> kernel: PCI: Switching off ROM of 00:01.0 Aug 8 21:41:04 andrewI kerne=
l:
>> Zorro: Probing AutoConfig expansion devices: 2 devices
>>=20
>> Aug 8 23:17:10 andrewI kernel: PCI: Probing PCI hardware Aug 8 23:17:1=
0
>> andrewI kernel: Memory resource not set for host bridge 0 Aug 8 23:17:=
10
>> andrewI kernel: apus_pcibios_fixup: PCI mem resource requested Aug 8
>> 23:17:10 andrewI kernel: PCI: Cannot allocate resource region 1 of PCI
>> bridge 0 Aug 8 23:17:10 andrewI kernel: PCI: resource is e0000000..fff=
c0000
>> (200), parent c023fc1c Aug 8 23:17:10 andrewI kernel: PCI:00:01.0: Res=
ource
>> 0: ef000000-ef01ffff (f=3D200) Aug 8 23:17:10 andrewI kernel: PCI:00:0=
1.0:
>> Resource 1: e0000000-e07fffff (f=3D200) Aug 8 23:17:10 andrewI kernel:
>> PCI:00:01.0: Resource 2: e1000000-e17fffff (f=3D200) Aug 8 23:17:10 an=
drewI
>> kernel: PCI:00:01.0: Resource 3: 3d071040-3d07107f (f=3D120c) Aug 8 23=
:17:10
>> andrewI kernel: PCI: Cannot allocate resource region 3 of device 00:01=
.0
>> Aug 8 23:17:10 andrewI kernel: PCI: Switching off ROM of 00:01.0 Aug 8
>> 23:17:10 andrewI kernel: PCI: Error while updating region 00:01.0/3
>> (e080000c !=3D 00000000) Aug 8 23:17:10 andrewI kernel: Zorro: Probing
>> AutoConfig expansion devices: 2 devices
>>=20
>> Aug 9 11:46:01 andrewI kernel: PCI: Probing PCI hardware Aug 9 11:46:0=
1
>> andrewI kernel: Memory resource not set for host bridge 0 Aug 9 11:46:=
01
>> andrewI kernel: apus_pcibios_fixup: PCI mem resource requested Aug 9
>> 11:46:01 andrewI kernel: PCI: Cannot allocate resource region 1 of PCI
>> bridge 0 Aug 9 11:46:01 andrewI kernel: PCI: resource is e0000000..fff=
c0000
>> (200), parent c023fc1c Aug 9 11:46:01 andrewI kernel: PCI:00:01.0: Res=
ource
>> 0: ef000000-ef01ffff (f=3D200) Aug 9 11:46:01 andrewI kernel: PCI:00:0=
1.0:
>> Resource 1: e0000000-e07fffff (f=3D200) Aug 9 11:46:01 andrewI kernel:
>> PCI:00:01.0: Resource 2: e1000000-e17fffff (f=3D200) Aug 9 11:46:01 an=
drewI
>> kernel: PCI:00:01.0: Resource 3: 3d071040-3d07107f (f=3D120c) Aug 9 11=
:46:01
>> andrewI kernel: PCI: Cannot allocate resource region 3 of device 00:01=
.0
>> Aug 9 11:46:01 andrewI kernel: PCI: Switching off ROM of 00:01.0 Aug 9
>> 11:46:01 andrewI kernel: PCI: Error while updating region 00:01.0/3
>> (e080000c !=3D 00000000) Aug 9 11:46:01 andrewI kernel: Zorro: Probing
>> AutoConfig expansion devices: 2 devices
>>=20
>> Aug 9 21:25:48 andrewI kernel: PCI: Probing PCI hardware Aug 9 21:25:4=
8
>> andrewI kernel: Memory resource not set for host bridge 0 Aug 9 21:25:=
48
>> andrewI kernel: apus_pcibios_fixup: PCI mem resource requested Aug 9
>> 21:25:48 andrewI kernel: PCI: Cannot allocate resource region 1 of PCI
>> bridge 0 Aug 9 21:25:48 andrewI kernel: PCI: resource is e0000000..fff=
c0000
>> (200), parent c023fc1c Aug 9 21:25:48 andrewI kernel: PCI:00:01.0: Res=
ource
>> 0: ef000000-ef01ffff (f=3D200) Aug 9 21:25:48 andrewI kernel: PCI:00:0=
1.0:
>> Resource 1: e0000000-e07fffff (f=3D200) Aug 9 21:25:48 andrewI kernel:
>> PCI:00:01.0: Resource 2: e1000000-e17fffff (f=3D200) Aug 9 21:25:48 an=
drewI
>> kernel: PCI: Switching off ROM of 00:01.0 Aug 9 21:25:48 andrewI kerne=
l:
>> Zorro: Probing AutoConfig expansion devices: 2 devices
>>=20
>>=20
>> Sometimes it sees 3 resources, sometimes two.
>=20
> Count again ;)

Not only are you swiss, you are even funny..

>> Anything to be worried about?
>=20
> I don't think so, as long as everything works...

Well, it works.

> What does lspci -vv say
> about the resources in both cases?

Aug 11 13:34:06 andrewI kernel: PCI: Probing PCI hardware
Aug 11 13:34:06 andrewI kernel: Memory resource not set for host bridge 0
Aug 11 13:34:06 andrewI kernel: apus_pcibios_fixup: PCI mem resource requ=
ested
Aug 11 13:34:06 andrewI kernel: PCI: Cannot allocate resource region 1 of=
 PCI bridge 0
Aug 11 13:34:06 andrewI kernel: PCI: resource is e0000000..fffc0000 (200)=
, parent c023fc1c
Aug 11 13:34:06 andrewI kernel: PCI:00:01.0: Resource 0: ef000000-ef01fff=
f (f=3D200)
Aug 11 13:34:06 andrewI kernel: PCI:00:01.0: Resource 1: e0000000-e0fffff=
f (f=3D200)
Aug 11 13:34:06 andrewI kernel: PCI:00:01.0: Resource 2: e1000000-e17ffff=
f (f=3D200)
Aug 11 13:34:06 andrewI kernel: PCI:00:01.0: Resource 3: 3d071040-3d07107=
f (f=3D120c)
Aug 11 13:34:06 andrewI kernel: PCI: Cannot allocate resource region 3 of=
 device 00:01.0
Aug 11 13:34:06 andrewI kernel: PCI: Switching off ROM of 00:01.0
Aug 11 13:34:06 andrewI kernel: PCI: Error while updating region 00:01.0/=
3 (e180000c !=3D 3d07104c)
Aug 11 13:34:06 andrewI kernel: Zorro: Probing AutoConfig expansion devic=
es: 2 devices

lspci -vv:

00:01.0 Display controller: Texas Instruments TVP4020 [Permedia 2] (rev 0=
1)
    Subsystem: Unknown device e4e4:e4e4
    Control: I/O+ Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- ParErr- S=
tepping- SERR- FastB2B-
    Status: Cap- 66Mhz- UDF- FastB2B+ ParErr+ DEVSEL=3Dmedium >TAbort- <T=
Abort- <MAbort- >SERR- <PERR-
    Interrupt: pin A routed to IRQ 0
    Region 0: Memory at ef000000 (32-bit, non-prefetchable) [disabled] [s=
ize=3D128K]
    Region 1: Memory at e0000000 (32-bit, non-prefetchable) [disabled] [s=
ize=3D16M]
    Region 2: Memory at e1000000 (32-bit, non-prefetchable) [disabled] [s=
ize=3D8M]
    Region 3: [virtual] Memory at e1800000 (64-bit, prefetchable) [disabl=
ed] [size=3D64]
    Expansion ROM at e2000000 [disabled] [size=3D64K]

Aug 11 22:25:23 andrewI kernel: PCI: Probing PCI hardware
Aug 11 22:25:23 andrewI kernel: Memory resource not set for host bridge 0
Aug 11 22:25:23 andrewI kernel: apus_pcibios_fixup: PCI mem resource requ=
ested
Aug 11 22:25:23 andrewI kernel: PCI: Cannot allocate resource region 1 of=
 PCI bridge 0
Aug 11 22:25:23 andrewI kernel: PCI: resource is e0000000..fffc0000 (200)=
, parent c023fc1c
Aug 11 22:25:23 andrewI kernel: PCI:00:01.0: Resource 0: ef000000-ef01fff=
f (f=3D200)
Aug 11 22:25:23 andrewI kernel: PCI:00:01.0: Resource 1: e0000000-e07ffff=
f (f=3D200)
Aug 11 22:25:23 andrewI kernel: PCI:00:01.0: Resource 2: e1000000-e17ffff=
f (f=3D200)
Aug 11 22:25:23 andrewI kernel: PCI: Switching off ROM of 00:01.0
Aug 11 22:25:23 andrewI kernel: PCI: Error while updating region 00:01.0/=
3 (e0800000 !=3D 00000000)
Aug 11 22:25:23 andrewI kernel: Zorro: Probing AutoConfig expansion devic=
es: 2 devices

lspci -vv:

00:01.0 Display controller: Texas Instruments TVP4020 [Permedia 2] (rev 0=
1)
    Subsystem: Unknown device e4e4:e4e4
    Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- S=
tepping- SERR- FastB2B-
    Status: Cap- 66Mhz- UDF- FastB2B+ ParErr- DEVSEL=3Dmedium >TAbort- <T=
Abort- <MAbort- >SERR- <PERR-
    Latency: 255 (48000ns min, 48000ns max)
    Interrupt: pin A routed to IRQ 0
    Region 0: Memory at ef000000 (32-bit, non-prefetchable) [size=3D128K]
    Region 1: Memory at e0000000 (32-bit, non-prefetchable) [size=3D8M]
    Region 2: Memory at e1000000 (32-bit, non-prefetchable) [size=3D8M]
    Region 3: [virtual] Memory at e0800000 (32-bit, non-prefetchable) [si=
ze=3D64]
    Expansion ROM at e2000000 [disabled] [size=3D64K]

There seem to be differencies between the Control and the Status line, an=
d
there's even a Latency line on the second one. Also, the Region 1 show
different sizes and Region 3 has an other address for the virtual Memory =
(and
an other bit size) and everything seems to be disabled on the first one. =
What
the heck does that say?? Hardware's hardware and software's software, isn=
't it??

> Could the difference somehow be
> related to something you do in AmigaOS?

I've already asked me that question, but I do mainly the same: I boot up =
into AmigaOS, then lock the drives (call me
whatever you want), then start up the script which fires up bootstrap. In=
 the
background I have a system checker running, as ever.

--=20
Best wishes,
Andi