Hello;
Well I couldn't really see what was happening
until I could experience this myself;
It seems the macbook pro ati chipset
is having the same problem as jack was with mmconfig yesterday.
I'm confused as to why my system was able to run this kernel(2.6.26.git1)
without a problem, now after doing a reinstall of the operating system
I am confronted with this same issue. How do I find the 0x3f number on
my system,
so I could plug that into the patch? to see if I can boot up
Now keep in mind the only difference with this install is the
installer took a dump
therefore you have to manually install grub(could this be the source
of the issue?)
regards;
--
Justin P. Mattock
On Fri, Jul 18, 2008 at 8:39 PM, Justin Mattock <[email protected]> wrote:
> Hello;
> Well I couldn't really see what was happening
> until I could experience this myself;
> It seems the macbook pro ati chipset
> is having the same problem as jack was with mmconfig yesterday.
> I'm confused as to why my system was able to run this kernel(2.6.26.git1)
> without a problem, now after doing a reinstall of the operating system
> I am confronted with this same issue. How do I find the 0x3f number on
> my system,
> so I could plug that into the patch? to see if I can boot up
> Now keep in mind the only difference with this install is the
> installer took a dump
> therefore you have to manually install grub(could this be the source
> of the issue?)
> regards;
can you disable ASPM in your config.?
YH
On Fri, Jul 18, 2008 at 9:45 PM, Yinghai Lu <[email protected]> wrote:
> On Fri, Jul 18, 2008 at 8:39 PM, Justin Mattock <[email protected]> wrote:
>> Hello;
>> Well I couldn't really see what was happening
>> until I could experience this myself;
>> It seems the macbook pro ati chipset
>> is having the same problem as jack was with mmconfig yesterday.
>> I'm confused as to why my system was able to run this kernel(2.6.26.git1)
>> without a problem, now after doing a reinstall of the operating system
>> I am confronted with this same issue. How do I find the 0x3f number on
>> my system,
>> so I could plug that into the patch? to see if I can boot up
>> Now keep in mind the only difference with this install is the
>> installer took a dump
>> therefore you have to manually install grub(could this be the source
>> of the issue?)
>> regards;
>
> can you disable ASPM in your config.?
>
> YH
>
I enabled and disabled in the pci section
PCIEASPM still the same results.
applied the patch as well i.g.
mmconf_dmi_table
going to try detect end_bus_number patch to
see if it works
--
Justin P. Mattock
On Fri, Jul 18, 2008 at 10:54 PM, Justin Mattock
<[email protected]> wrote:
> On Fri, Jul 18, 2008 at 9:45 PM, Yinghai Lu <[email protected]> wrote:
>> On Fri, Jul 18, 2008 at 8:39 PM, Justin Mattock <[email protected]> wrote:
>>> Hello;
>>> Well I couldn't really see what was happening
>>> until I could experience this myself;
>>> It seems the macbook pro ati chipset
>>> is having the same problem as jack was with mmconfig yesterday.
>>> I'm confused as to why my system was able to run this kernel(2.6.26.git1)
>>> without a problem, now after doing a reinstall of the operating system
>>> I am confronted with this same issue. How do I find the 0x3f number on
>>> my system,
>>> so I could plug that into the patch? to see if I can boot up
>>> Now keep in mind the only difference with this install is the
>>> installer took a dump
>>> therefore you have to manually install grub(could this be the source
>>> of the issue?)
>>> regards;
>>
>> can you disable ASPM in your config.?
>>
>> YH
>>
>
> I enabled and disabled in the pci section
> PCIEASPM still the same results.
> applied the patch as well i.g.
> mmconf_dmi_table
> going to try detect end_bus_number patch to
> see if it works
> --
> Justin P. Mattock
>
applied both patches and messed with boot options
but no go:
this is what I'm seeing
PCI: Fatal no config space access function found
ata-piix 0000:00:1f11 device not available
because of BAR0 [0xf0-0x1f7] collisions
interesting how this kernel was running fine, and then
BAM all of a sudden f*&^ed up, This tells me that this was something
happening a while ago, or always happening just not
noticed i.g. I plugged in 2.6.25-rc9(my favorite)
and it too was giving the same response.
after tracing back my steps the only application that might be causing
this or hiding this is grub but could be wrong.
any ideas or patches would be appreciated, until further
notice I'll be using osx until there is a
regards; : (
--
Justin P. Mattock
On Fri, Jul 18, 2008 at 11:54 PM, Justin Mattock
<[email protected]> wrote:
> On Fri, Jul 18, 2008 at 10:54 PM, Justin Mattock
> <[email protected]> wrote:
>> On Fri, Jul 18, 2008 at 9:45 PM, Yinghai Lu <[email protected]> wrote:
>>> On Fri, Jul 18, 2008 at 8:39 PM, Justin Mattock <[email protected]> wrote:
>>>> Hello;
>>>> Well I couldn't really see what was happening
>>>> until I could experience this myself;
>>>> It seems the macbook pro ati chipset
>>>> is having the same problem as jack was with mmconfig yesterday.
>>>> I'm confused as to why my system was able to run this kernel(2.6.26.git1)
>>>> without a problem, now after doing a reinstall of the operating system
>>>> I am confronted with this same issue. How do I find the 0x3f number on
>>>> my system,
>>>> so I could plug that into the patch? to see if I can boot up
>>>> Now keep in mind the only difference with this install is the
>>>> installer took a dump
>>>> therefore you have to manually install grub(could this be the source
>>>> of the issue?)
>>>> regards;
>>>
>>> can you disable ASPM in your config.?
>>>
>>> YH
>>>
>>
>> I enabled and disabled in the pci section
>> PCIEASPM still the same results.
>> applied the patch as well i.g.
>> mmconf_dmi_table
>> going to try detect end_bus_number patch to
>> see if it works
>> --
>> Justin P. Mattock
>>
>
> applied both patches and messed with boot options
> but no go:
> this is what I'm seeing
> PCI: Fatal no config space access function found
> ata-piix 0000:00:1f11 device not available
> because of BAR0 [0xf0-0x1f7] collisions
can you send out your .config?
YH
On Fri, Jul 18, 2008 at 11:58 PM, Yinghai Lu <[email protected]> wrote:
> On Fri, Jul 18, 2008 at 11:54 PM, Justin Mattock
> <[email protected]> wrote:
>> On Fri, Jul 18, 2008 at 10:54 PM, Justin Mattock
>> <[email protected]> wrote:
>>> On Fri, Jul 18, 2008 at 9:45 PM, Yinghai Lu <[email protected]> wrote:
>>>> On Fri, Jul 18, 2008 at 8:39 PM, Justin Mattock <[email protected]> wrote:
>>>>> Hello;
>>>>> Well I couldn't really see what was happening
>>>>> until I could experience this myself;
>>>>> It seems the macbook pro ati chipset
>>>>> is having the same problem as jack was with mmconfig yesterday.
>>>>> I'm confused as to why my system was able to run this kernel(2.6.26.git1)
>>>>> without a problem, now after doing a reinstall of the operating system
>>>>> I am confronted with this same issue. How do I find the 0x3f number on
>>>>> my system,
>>>>> so I could plug that into the patch? to see if I can boot up
>>>>> Now keep in mind the only difference with this install is the
>>>>> installer took a dump
>>>>> therefore you have to manually install grub(could this be the source
>>>>> of the issue?)
>>>>> regards;
>>>>
>>>> can you disable ASPM in your config.?
>>>>
>>>> YH
>>>>
>>>
>>> I enabled and disabled in the pci section
>>> PCIEASPM still the same results.
>>> applied the patch as well i.g.
>>> mmconf_dmi_table
>>> going to try detect end_bus_number patch to
>>> see if it works
>>> --
>>> Justin P. Mattock
>>>
>>
>> applied both patches and messed with boot options
>> but no go:
>> this is what I'm seeing
>> PCI: Fatal no config space access function found
>> ata-piix 0000:00:1f11 device not available
>> because of BAR0 [0xf0-0x1f7] collisions
>
> can you send out your .config?
>
> YH
>
here you go,
you might need to chmod because for some reason
mounting sda1 and then making a copy to my usb stick
changed the permission on this(.config)
as soon as I figure out how to chmod with mac osx I can send
you another if need be
regards;
--
Justin P. Mattock
Justin,
I never saw any error messages like that. If you are seeing the
same problem under 2.6.25 suddenly it might suggest a hardware
failure, no? That or something else has corrupted on your linux
volume. What happens if you boot from a linux rescue DVD? Does that
give the same error?
Jack
On Fri, Jul 18, 2008 at 11:54:38PM -0700, Justin Mattock wrote:
> applied both patches and messed with boot options
> but no go:
> this is what I'm seeing
> PCI: Fatal no config space access function found
> ata-piix 0000:00:1f11 device not available
> because of BAR0 [0xf0-0x1f7] collisions
>
>
> interesting how this kernel was running fine, and then
> BAM all of a sudden f*&^ed up, This tells me that this was something
> happening a while ago, or always happening just not
> noticed i.g. I plugged in 2.6.25-rc9(my favorite)
> and it too was giving the same response.
> after tracing back my steps the only application that might be causing
> this or hiding this is grub but could be wrong.
>
> any ideas or patches would be appreciated, until further
> notice I'll be using osx until there is a
> regards; : (
>
> --
> Justin P. Mattock
On Sat, Jul 19, 2008 at 6:10 AM, Jack Howarth <[email protected]> wrote:
> Justin,
> I never saw any error messages like that. If you are seeing the
> same problem under 2.6.25 suddenly it might suggest a hardware
> failure, no? That or something else has corrupted on your linux
> volume. What happens if you boot from a linux rescue DVD? Does that
> give the same error?
> Jack
>
> On Fri, Jul 18, 2008 at 11:54:38PM -0700, Justin Mattock wrote:
>> applied both patches and messed with boot options
>> but no go:
>> this is what I'm seeing
>> PCI: Fatal no config space access function found
>> ata-piix 0000:00:1f11 device not available
>> because of BAR0 [0xf0-0x1f7] collisions
>>
>>
>> interesting how this kernel was running fine, and then
>> BAM all of a sudden f*&^ed up, This tells me that this was something
>> happening a while ago, or always happening just not
>> noticed i.g. I plugged in 2.6.25-rc9(my favorite)
>> and it too was giving the same response.
>> after tracing back my steps the only application that might be causing
>> this or hiding this is grub but could be wrong.
>>
>> any ideas or patches would be appreciated, until further
>> notice I'll be using osx until there is a
>> regards; : (
>>
>> --
>> Justin P. Mattock
>
could be; but loading an iso works i.g.
after installing it and starting up.
these two messages came from after applying
the two patches(keep in mind I was tired so maybe
plugged them in wrong) before the two patches
the MMCONFIG disable message with f0000000
that you where receiving. maybe the manufacture
on mine needs to be MacBookPro2,2 and the 0x3f
needs to be 0x1f. Where do I find these numbers
and names on the system? I've done dmidecode
but can't seem to make out any of the info
--
Justin P. Mattock
On Sat, Jul 19, 2008 at 07:53:34AM -0700, Justin Mattock wrote:
> >
>
> could be; but loading an iso works i.g.
> after installing it and starting up.
> these two messages came from after applying
> the two patches(keep in mind I was tired so maybe
> plugged them in wrong) before the two patches
> the MMCONFIG disable message with f0000000
> that you where receiving. maybe the manufacture
> on mine needs to be MacBookPro2,2 and the 0x3f
> needs to be 0x1f. Where do I find these numbers
> and names on the system? I've done dmidecode
> but can't seem to make out any of the info
>
> --
> Justin P. Mattock
Justin,
I am attempting to build kernel rpms with tip right now.
My first effort would boot because somehow the modules weren't
being found (paths were being used with -tip in them). I've removed
the localversion file from the top level of tip and hope that fixes
the packaging issues. This issue didn't occur when I used a git
from the linus kernel branch.
Jack
On Sat, Jul 19, 2008 at 8:14 AM, Jack Howarth <[email protected]> wrote:
> On Sat, Jul 19, 2008 at 07:53:34AM -0700, Justin Mattock wrote:
>> >
>>
>> could be; but loading an iso works i.g.
>> after installing it and starting up.
>> these two messages came from after applying
>> the two patches(keep in mind I was tired so maybe
>> plugged them in wrong) before the two patches
>> the MMCONFIG disable message with f0000000
>> that you where receiving. maybe the manufacture
>> on mine needs to be MacBookPro2,2 and the 0x3f
>> needs to be 0x1f. Where do I find these numbers
>> and names on the system? I've done dmidecode
>> but can't seem to make out any of the info
>>
>> --
>> Justin P. Mattock
>
> Justin,
> I am attempting to build kernel rpms with tip right now.
> My first effort would boot because somehow the modules weren't
> being found (paths were being used with -tip in them). I've removed
> the localversion file from the top level of tip and hope that fixes
> the packaging issues. This issue didn't occur when I used a git
> from the linus kernel branch.
> Jack
>
>From over here; do you think grub
has something to do with this error?
why: when installing the intaller was broken
thus having to download the latest grub and manually
plug in the partitions and so forth.
when using the latest git(a day ago) I was using
an older version it seems of grub. But this wouldn't explain
why I received the PCI: not using MMCONFIG error
I need right now to make sure the name is correct
"MacBook2" and "Apple", and make sure 0x3f
is right also. You never know maybe these get changed with
each version of the mac's. Overall this is a pretty nasty bug
leaving completely out of business.
--
Justin P. Mattock
On Sat, Jul 19, 2008 at 08:54:58AM -0700, Justin Mattock wrote:
>
> >From over here; do you think grub
> has something to do with this error?
> why: when installing the intaller was broken
> thus having to download the latest grub and manually
> plug in the partitions and so forth.
> when using the latest git(a day ago) I was using
> an older version it seems of grub. But this wouldn't explain
> why I received the PCI: not using MMCONFIG error
> I need right now to make sure the name is correct
> "MacBook2" and "Apple", and make sure 0x3f
> is right also. You never know maybe these get changed with
> each version of the mac's. Overall this is a pretty nasty bug
> leaving completely out of business.
> --
> Justin P. Mattock
Justin,
The current tip from earlier this morning is booting fine under
Fedora 9 here on my MacBook Pro v2 (build with CONFIG_PCIEASPM unset).
I am rebuilding tip with CONFIG_PCIEASPM=y in case the issue is fixed
or has gone latent on tip. If not, I'll update my bugzilla and try
to get Jesse to take a look at it.
Jack
On Sat, Jul 19, 2008 at 9:28 AM, Jack Howarth <[email protected]> wrote:
> On Sat, Jul 19, 2008 at 08:54:58AM -0700, Justin Mattock wrote:
>>
>> >From over here; do you think grub
>> has something to do with this error?
>> why: when installing the intaller was broken
>> thus having to download the latest grub and manually
>> plug in the partitions and so forth.
>> when using the latest git(a day ago) I was using
>> an older version it seems of grub. But this wouldn't explain
>> why I received the PCI: not using MMCONFIG error
>> I need right now to make sure the name is correct
>> "MacBook2" and "Apple", and make sure 0x3f
>> is right also. You never know maybe these get changed with
>> each version of the mac's. Overall this is a pretty nasty bug
>> leaving completely out of business.
>> --
>> Justin P. Mattock
>
> Justin,
> The current tip from earlier this morning is booting fine under
> Fedora 9 here on my MacBook Pro v2 (build with CONFIG_PCIEASPM unset).
> I am rebuilding tip with CONFIG_PCIEASPM=y in case the issue is fixed
> or has gone latent on tip. If not, I'll update my bugzilla and try
> to get Jesse to take a look at it.
> Jack
>
>
when using the boot options
debug initctl_debug vga=790
I'm able to see the hardisk is being mounted alsa
is activated netfiler is being activated, but then
just stops giving me the:
VFS cannot mount on root filesystem error
keep in mind this is when using "any" option
in PCI, when using MMCONFIG
im getting what you where getting
I'm just confused on how the system was working
without any errors with "Direct" in pci
and now none of the options are working.
The posotive side is better work this issue now,
rather than later. what I'll try is loading a fresh install
of nubuntu, and then compile the kernel with there .config
to see if I'm able to start. Also they added apparmor so I'm not sure if
this is causing problems as well.
--
Justin P. Mattock
On Sat, Jul 19, 2008 at 10:15:52AM -0700, Justin Mattock wrote:
>
> when using the boot options
> debug initctl_debug vga=790
> I'm able to see the hardisk is being mounted alsa
> is activated netfiler is being activated, but then
> just stops giving me the:
> VFS cannot mount on root filesystem error
> keep in mind this is when using "any" option
> in PCI, when using MMCONFIG
> im getting what you where getting
> I'm just confused on how the system was working
> without any errors with "Direct" in pci
> and now none of the options are working.
> The posotive side is better work this issue now,
> rather than later. what I'll try is loading a fresh install
> of nubuntu, and then compile the kernel with there .config
> to see if I'm able to start. Also they added apparmor so I'm not sure if
> this is causing problems as well.
> --
> Justin P. Mattock
Justin,
Just to be clear, you do have CONFIG_PCIEASPM unset in your .config,
right? All 2.6.26 kernels I have tried (including tip) hang during the
boot when built with that option.
Jack
On Sat, Jul 19, 2008 at 10:58 AM, Jack Howarth
<[email protected]> wrote:
> On Sat, Jul 19, 2008 at 10:15:52AM -0700, Justin Mattock wrote:
>>
>> when using the boot options
>> debug initctl_debug vga=790
>> I'm able to see the hardisk is being mounted alsa
>> is activated netfiler is being activated, but then
>> just stops giving me the:
>> VFS cannot mount on root filesystem error
>> keep in mind this is when using "any" option
>> in PCI, when using MMCONFIG
>> im getting what you where getting
>> I'm just confused on how the system was working
>> without any errors with "Direct" in pci
>> and now none of the options are working.
>> The posotive side is better work this issue now,
>> rather than later. what I'll try is loading a fresh install
>> of nubuntu, and then compile the kernel with there .config
>> to see if I'm able to start. Also they added apparmor so I'm not sure if
>> this is causing problems as well.
>> --
>> Justin P. Mattock
>
> Justin,
> Just to be clear, you do have CONFIG_PCIEASPM unset in your .config,
> right? All 2.6.26 kernels I have tried (including tip) hang during the
> boot when built with that option.
> Jack
>
yep; didn't choose this option.
I'll check again to make sure.
just loaded a fresh 2.6.26 without any patches
just to be sure and still unable to mount VFS error
where can I find the patches you used that worked
to see if this works over here. Is there some place I should look in userspace
as well to see if I'm missing something?
--
Justin P. Mattock