Re: [PATCH v4] acpi,pci: warn about duplicate IRQ routing entries returned from _PRT

[Mateusz Jończyk]

W dniu 16.02.2024 o 19:49, Bjorn Helgaas pisze:
> On Fri, Feb 16, 2024 at 07:26:06PM +0100, Rafael J. Wysocki wrote:
>> On Tue, Dec 26, 2023 at 1:50 PM Mateusz Jończyk <mat.jonczyk@xxxxx> wrote:
>>> On some platforms, the ACPI _PRT function returns duplicate interrupt
>>> routing entries. Linux uses the first matching entry, but sometimes the
>>> second matching entry contains the correct interrupt vector.
>>>
>>> As a debugging aid, print a warning to dmesg if duplicate interrupt
>>> routing entries are present. This way, we could check how many models
>>> are affected.
>>>
>>> This happens on a Dell Latitude E6500 laptop with the i2c-i801 Intel
>>> SMBus controller. This controller is nonfunctional unless its interrupt
>>> usage is disabled (using the "disable_features=0x10" module parameter).
>>>
>>> After investigation, it turned out that the driver was using an
>>> incorrect interrupt vector: in lspci output for this device there was:
>>>         Interrupt: pin B routed to IRQ 19
>>> but after running i2cdetect (without using any i2c-i801 module
>>> parameters) the following was logged to dmesg:
>>>
>>>         [...]
>>>         i801_smbus 0000:00:1f.3: Timeout waiting for interrupt!
>>>         i801_smbus 0000:00:1f.3: Transaction timeout
>>>         i801_smbus 0000:00:1f.3: Timeout waiting for interrupt!
>>>         i801_smbus 0000:00:1f.3: Transaction timeout
>>>         irq 17: nobody cared (try booting with the "irqpoll" option)
>>>
>>> Existence of duplicate entries in a table returned by the _PRT method
>>> was confirmed by disassembling the ACPI DSDT table.
>>>
>>> Windows XP is using IRQ3 (as reported by HWiNFO32 and in the Device
>>> Manager), which is neither of the two vectors returned by _PRT.
>>> As HWiNFO32 decoded contents of the SPD EEPROMs, the i2c-i801 device is
>>> working under Windows. It appears that Windows has reconfigured the
>>> chipset independently to use another interrupt vector for the device.
>>> This is possible, according to the chipset datasheet [1], page 436 for
>>> example (PIRQ[n]_ROUT—PIRQ[A,B,C,D] Routing Control Register).
>>>
>>> [1] https://www.intel.com/content/dam/doc/datasheet/io-controller-hub-9-datasheet.pdf
>>>
>>> Signed-off-by: Mateusz Jończyk <mat.jonczyk@xxxxx>
>>> Cc: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
>>> Cc: "Rafael J. Wysocki" <rafael@xxxxxxxxxx>
>>> Cc: Len Brown <lenb@xxxxxxxxxx>
>>> Cc: Borislav Petkov <bp@xxxxxxx>
>>> Cc: Jean Delvare <jdelvare@xxxxxxx>
>>> Previously-reviewed-by: Jean Delvare <jdelvare@xxxxxxx>
>>> Previously-tested-by: Jean Delvare <jdelvare@xxxxxxx>
>>>
>>> ---
>>> Hello,
>>>
>>> I'm resurrecting an older patch that was discussed back in January:
>>>
>>> https://lore.kernel.org/lkml/20230121153314.6109-1-mat.jonczyk@xxxxx/T/#u
>>>
>>> To consider: should we print a warning or an error in case of duplicate
>>> entries? This may not be serious enough to disturb the user with an
>>> error message at boot.
>>>
>>> I'm also looking into modifying the i2c-i801 driver to disable its usage
>>> of interrupts if one did not fire.
>>>
>>> v2: - add a newline at the end of the kernel log message,
>>>     - replace: "if (match == NULL)" -> "if (!match)"
>>>     - patch description tweaks.
>>> v3: - fix C style issues pointed by Jean Delvare,
>>>     - switch severity from warning to error.
>>> v3 RESEND: retested on top of v6.2-rc4
>>> v4: - rebase and retest on top of v6.7-rc7
>>>     - switch severity back to warning,
>>>     - change pr_err() to dev_warn() and simplify the code,
>>>     - modify patch description (describe Windows behaviour etc.)
>>> ---
>>>  drivers/acpi/pci_irq.c | 25 ++++++++++++++++++++++---
>>>  1 file changed, 22 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c
>>> index ff30ceca2203..1fcf72e335b0 100644
>>> --- a/drivers/acpi/pci_irq.c
>>> +++ b/drivers/acpi/pci_irq.c
>>> @@ -203,6 +203,8 @@ static int acpi_pci_irq_find_prt_entry(struct pci_dev *dev,
>>>         struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL };
>>>         struct acpi_pci_routing_table *entry;
>>>         acpi_handle handle = NULL;
>>> +       struct acpi_prt_entry *match = NULL;
>>> +       const char *match_int_source = NULL;
>>>
>>>         if (dev->bus->bridge)
>>>                 handle = ACPI_HANDLE(dev->bus->bridge);
>>> @@ -219,13 +221,30 @@ static int acpi_pci_irq_find_prt_entry(struct pci_dev *dev,
>>>
>>>         entry = buffer.pointer;
>>>         while (entry && (entry->length > 0)) {
>>> -               if (!acpi_pci_irq_check_entry(handle, dev, pin,
>>> -                                                entry, entry_ptr))
>>> -                       break;
>>> +               struct acpi_prt_entry *curr;
>>> +
>>> +               if (!acpi_pci_irq_check_entry(handle, dev, pin, entry, &curr)) {
>>> +                       if (!match) {
>>> +                               match = curr;
>>> +                               match_int_source = entry->source;
>>> +                        } else {
>>> +                               dev_warn(&dev->dev, FW_BUG
>> dev_info() would be sufficient here IMV.
>>
>>> +                                      "ACPI _PRT returned duplicate IRQ routing entries for INT%c: %s[%d] and %s[%d]\n",
>>> +                                      pin_name(curr->pin),
>>> +                                      match_int_source, match->index,
>>> +                                      entry->source, curr->index);
>>> +                               /* We use the first matching entry nonetheless,
>>> +                                * for compatibility with older kernels.
> The usual comment style in this file is:
>
>   /*
>    * We use ...
>    */
>
>>> +                                */
>>> +                       }
>>> +               }
>>> +
>>>                 entry = (struct acpi_pci_routing_table *)
>>>                     ((unsigned long)entry + entry->length);
>>>         }
>>>
>>> +       *entry_ptr = match;
>>> +
>>>         kfree(buffer.pointer);
>>>         return 0;
>>>  }
>>>
>>> base-commit: 861deac3b092f37b2c5e6871732f3e11486f7082
>>> --
>> Bjorn, any concerns regarding this one?
> No concerns from me.  
>
> I guess this only adds a message, right?  It doesn't actually fix
> anything or change any behavior?
Exactly.
> This talks about "duplicate" entries, which suggests to me that they
> are identical, but I don't think they are.  It sounds like it's two
> "matching" entries, i.e., two entries for the same (device, pin)?

Right.

> And neither of the two _PRT entries yields a working i801 device?

Unpatched Linux uses the first matching entry, but the second one gives
a working i801 device. The point is to print a warning message to see
how many devices are affected and whether it is safe to switch the code
to use the last matching entry in all instances.

Therefore I used dev_warn().

> Bjorn

Greetings,

Mateusz