[PATCH] PCI: Fix bit definitions of PCI_EXP_LNKCAP2 register

[Bjorn Helgaas]

On Fri, Nov 9, 2012 at 10:37 AM, Alex Deucher <alexdeucher at gmail.com> wrote:
> On Fri, Nov 9, 2012 at 12:15 PM, Bjorn Helgaas <bhelgaas at google.com> wrote:
>> On Thu, Nov 8, 2012 at 11:56 PM, Jingoo Han <jg1.han at samsung.com> wrote:
>>> According to the PCIe 3.0 spec, PCI_EXP_LNKCAP2_SLS_2_5GB is
>>> 1st bit of PCI_EXP_LNKCAP2 register, not 0th bit. So, the bit
>>> definition of supported link speed vector should be fixed.
>>>
>>> Signed-off-by: Jingoo Han <jg1.han at samsung.com>
>>> ---
>>>  include/uapi/linux/pci_regs.h |    6 +++---
>>>  1 files changed, 3 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/include/uapi/linux/pci_regs.h b/include/uapi/linux/pci_regs.h
>>> index 20ae747..14a3184 100644
>>> --- a/include/uapi/linux/pci_regs.h
>>> +++ b/include/uapi/linux/pci_regs.h
>>> @@ -544,9 +544,9 @@
>>>  #define  PCI_EXP_OBFF_WAKE_EN  0x6000  /* OBFF using WAKE# signaling */
>>>  #define PCI_CAP_EXP_ENDPOINT_SIZEOF_V2 44      /* v2 endpoints end here */
>>>  #define PCI_EXP_LNKCAP2                44      /* Link Capability 2 */
>>> -#define  PCI_EXP_LNKCAP2_SLS_2_5GB 0x01        /* Current Link Speed 2.5GT/s */
>>> -#define  PCI_EXP_LNKCAP2_SLS_5_0GB 0x02        /* Current Link Speed 5.0GT/s */
>>> -#define  PCI_EXP_LNKCAP2_SLS_8_0GB 0x04        /* Current Link Speed 8.0GT/s */
>>> +#define  PCI_EXP_LNKCAP2_SLS_2_5GB 0x02        /* Current Link Speed 2.5GT/s */
>>> +#define  PCI_EXP_LNKCAP2_SLS_5_0GB 0x04        /* Current Link Speed 5.0GT/s */
>>> +#define  PCI_EXP_LNKCAP2_SLS_8_0GB 0x08        /* Current Link Speed 8.0GT/s */
>>>  #define  PCI_EXP_LNKCAP2_CROSSLINK 0x100 /* Crosslink supported */
>>>  #define PCI_EXP_LNKCTL2                48      /* Link Control 2 */
>>>  #define PCI_EXP_LNKSTA2                50      /* Link Status 2 */
>>
>> I think this patch is correct, per spec sec 7.8.18.  If I apply it, I
>> think the comments should also be changed to "Supported Link Speed"
>> instead of "Current."
>
> Correct.
>
>>
>> The only in-tree user of these symbols is
>> drm_pcie_get_speed_cap_mask().  Dave, can you ack/nack this?  I don't
>> want to apply this if it's going to break something there.
>
> The patch is fine and shouldn't break anything.

Thanks for checking this out, Alex.

I applied this to my pci/misc branch as v3.8 material.

Bjorn