Re: [v3.11][v3.12][v3.13][Regression] EISA: Initialize device before its resources

[Joseph Salisbury <joseph.salisbury@canonical.com>]

On 01/18/2014 09:37 AM, Bjorn Helgaas wrote:
> On Fri, Jan 17, 2014 at 6:35 PM, Joseph Salisbury
> <joseph.salisbury@canonical.com> wrote:
>> On 01/17/2014 05:19 PM, Bjorn Helgaas wrote:
>>> On Fri, Jan 17, 2014 at 02:26:23PM -0500, Joseph Salisbury wrote:
>>>> On 01/17/2014 12:02 PM, Bjorn Helgaas wrote:
>>>>> On Thu, Jan 16, 2014 at 01:14:01PM -0500, Joseph Salisbury wrote:
>>>>>> On 01/16/2014 01:12 PM, Bjorn Helgaas wrote:
>>>>>>> On Thu, Jan 16, 2014 at 10:53 AM, Joseph Salisbury
>>>>>>> <joseph.salisbury@canonical.com> wrote:
>>>>>>>> Hi Bjorn,
>>>>>>>>
>>>>>>>> A kernel bug was opened against Ubuntu [0].  After a kernel bisect, it
>>>>>>>> was found the following commit introduced this bug:
>>>>>>> Sorry about that, and thanks for the report.  Did you mean to include
>>>>>>> URL for the bug?
>>>>>> Yes, sorry about that:
>>>>>> http://pad.lv/1251816
>>>>> Hi Joseph,
>>>>>
>>>>> Can you attach the 3.8.0-32-generic config (the one matching the successful
>>>>> boot at https://launchpadlibrarian.net/156685076/BootDmesg.txt) to the bug?
>>>> I attached the config file to the bug:
>>>> https://launchpadlibrarian.net/162754666/config.common.ubuntu
>>>>
>>>> I also attached a tar file with the complete config directory for that
>>>> kernel version.
>>>> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1251816/+attachment/3951156/+files/raring-config.tar
>>> Thanks again.  I attached the following reverts to launchpad.  I screwed up
>>> when doing those EISA changes.  I'd like to squeeze these into my v3.14
>>> merge request (probably early next week), so please test and let me know
>>> if this fixes the problem.  I'm really sorry for the inconvenience.
>>>
>>> Bjorn
>>>
>>>
>>> Revert "EISA: Log device resources in dmesg"
>>>
>>> From: Bjorn Helgaas <bhelgaas@google.com>
>>>
>>> This reverts commit a2080d0c561c546d73cb8b296d4b7ca414e6860b.
>>>
>>> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
>>> ---
>>>  drivers/eisa/eisa-bus.c |    1 -
>>>  1 file changed, 1 deletion(-)
>>>
>>> diff --git a/drivers/eisa/eisa-bus.c b/drivers/eisa/eisa-bus.c
>>> index 8842cde69177..1b86fe0c2e80 100644
>>> --- a/drivers/eisa/eisa-bus.c
>>> +++ b/drivers/eisa/eisa-bus.c
>>> @@ -288,7 +288,6 @@ static int __init eisa_request_resources(struct eisa_root_device *root,
>>>                       edev->res[i].flags = IORESOURCE_IO | IORESOURCE_BUSY;
>>>               }
>>>
>>> -             dev_printk(KERN_DEBUG, &edev->dev, "%pR\n", &edev->res[i]);
>>>               if (request_resource(root->res, &edev->res[i]))
>>>                       goto failed;
>>>       }
>>> Revert "EISA: Initialize device before its resources"
>>>
>>> From: Bjorn Helgaas <bhelgaas@google.com>
>>>
>>> This reverts commit 26abfeed4341872364386c6a52b9acef8c81a81a.
>>>
>>> In the eisa_probe() force_probe path, if we were unable to request slot
>>> resources (e.g., [io 0x800-0x8ff]), we skipped the slot with "Cannot
>>> allocate resource for EISA slot %d" before reading the EISA signature in
>>> eisa_init_device().
>>>
>>> Commit 26abfeed4341 moved eisa_init_device() earlier, so we tried to read
>>> the EISA signature before requesting the slot resources, and this caused
>>> hangs during boot.
>>>
>>> Link: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1251816
>>> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
>>> ---
>>>  drivers/eisa/eisa-bus.c |   26 +++++++++++++++-----------
>>>  1 file changed, 15 insertions(+), 11 deletions(-)
>>>
>>> diff --git a/drivers/eisa/eisa-bus.c b/drivers/eisa/eisa-bus.c
>>> index 1b86fe0c2e80..612afeaec3cb 100644
>>> --- a/drivers/eisa/eisa-bus.c
>>> +++ b/drivers/eisa/eisa-bus.c
>>> @@ -277,11 +277,13 @@ static int __init eisa_request_resources(struct eisa_root_device *root,
>>>               }
>>>
>>>               if (slot) {
>>> +                     edev->res[i].name  = NULL;
>>>                       edev->res[i].start = SLOT_ADDRESS(root, slot)
>>>                                            + (i * 0x400);
>>>                       edev->res[i].end   = edev->res[i].start + 0xff;
>>>                       edev->res[i].flags = IORESOURCE_IO;
>>>               } else {
>>> +                     edev->res[i].name  = NULL;
>>>                       edev->res[i].start = SLOT_ADDRESS(root, slot)
>>>                                            + EISA_VENDOR_ID_OFFSET;
>>>                       edev->res[i].end   = edev->res[i].start + 3;
>>> @@ -327,19 +329,20 @@ static int __init eisa_probe(struct eisa_root_device *root)
>>>               return -ENOMEM;
>>>       }
>>>
>>> -     if (eisa_init_device(root, edev, 0)) {
>>> +     if (eisa_request_resources(root, edev, 0)) {
>>> +             dev_warn(root->dev,
>>> +                      "EISA: Cannot allocate resource for mainboard\n");
>>>               kfree(edev);
>>>               if (!root->force_probe)
>>> -                     return -ENODEV;
>>> +                     return -EBUSY;
>>>               goto force_probe;
>>>       }
>>>
>>> -     if (eisa_request_resources(root, edev, 0)) {
>>> -             dev_warn(root->dev,
>>> -                      "EISA: Cannot allocate resource for mainboard\n");
>>> +     if (eisa_init_device(root, edev, 0)) {
>>> +             eisa_release_resources(edev);
>>>               kfree(edev);
>>>               if (!root->force_probe)
>>> -                     return -EBUSY;
>>> +                     return -ENODEV;
>>>               goto force_probe;
>>>       }
>>>
>>> @@ -362,11 +365,6 @@ static int __init eisa_probe(struct eisa_root_device *root)
>>>                       continue;
>>>               }
>>>
>>> -             if (eisa_init_device(root, edev, i)) {
>>> -                     kfree(edev);
>>> -                     continue;
>>> -             }
>>> -
>>>               if (eisa_request_resources(root, edev, i)) {
>>>                       dev_warn(root->dev,
>>>                                "Cannot allocate resource for EISA slot %d\n",
>>> @@ -375,6 +373,12 @@ static int __init eisa_probe(struct eisa_root_device *root)
>>>                       continue;
>>>               }
>>>
>>> +             if (eisa_init_device(root, edev, i)) {
>>> +                     eisa_release_resources(edev);
>>> +                     kfree(edev);
>>> +                     continue;
>>> +             }
>>> +
>>>               if (edev->state == (EISA_CONFIG_ENABLED | EISA_CONFIG_FORCED))
>>>                       enabled_str = " (forced enabled)";
>>>               else if (edev->state == EISA_CONFIG_FORCED)
>> Thanks again for looking at this, Bjorn!
>>
>> Yes, reverting commit 26abfeed4 does in fact resolve the bug.  I already
>> built a test kernel, noted in comment #34:
>> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1251816/comments/34
>>
>> The original bug reporter tested the kernel to confirm it fixes the bug:
>> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1251816/comments/36
> OK.  I don't know exactly what patch you applied to revert this.  I
> assume that you reverted both a2080d0c561 and 26abfeed43.

Right, my original test kernel only had commit 26abfeed43 reverted.  I
built one more kernel with both a2080d0c561 and 26abfeed43 reverted, and
that does resolve the bug:
 https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1251816/comments/44

Thanks again for the help, Bjorn!

>
>> Would it be also possible for you to request the revert in the stable
>> trees, in addition to 3.14?
> Oh, of course, thanks for the reminder.  I added a stable tag and re-pushed.
>
> Bjorn