[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-ppc] [PATCH v5 4/4] spapr_pci: populate ibm,loc-code
From: |
Nikunj A Dadhania |
Subject: |
Re: [Qemu-ppc] [PATCH v5 4/4] spapr_pci: populate ibm,loc-code |
Date: |
Mon, 25 May 2015 15:57:41 +0530 |
User-agent: |
Notmuch/0.17+27~gae47d61 (http://notmuchmail.org) Emacs/24.3.1 (x86_64-redhat-linux-gnu) |
Alexey Kardashevskiy <address@hidden> writes:
> On 05/25/2015 02:58 PM, Nikunj A Dadhania wrote:
>> Alexey Kardashevskiy <address@hidden> writes:
>>
>>>
>>>> + }
>>>> + /*
>>>> + * For emulated devices and VFIO-failure case, make up
>>>> + * the loc-code.
>>>> + */
>>>> + buf = g_strdup_printf("%s_%s:%02d:%02d.%1d",
>>>
>>> How is "1d" different here from just "d"? If @devfn>10, all digits will be
>>> printed.
>>
>> A device can have only 8 functions, right ?
>
>
> Yes. It is "%d" (decimal) so "1" is a minimum field length anyway ("1" only
> makes sense for strings imho). And if a function number is bigger than 9,
> print() will print all digits, not just the last one. So %d==%1d but I
> would never grep for "%02d:%02d.%1d", I would grep for "%02d:%02d.%d".
>
> [few minutes later] And I just did. And I found that most places actually
> use "%02x:%02x.%x" (trace-events, hw/pci/pci.c, hw/vfio/pci.c) and only one
> (get_pci_host_devaddr) uses "%d" for a function (and even there "%02x" is
> used for a bus and a slot, not "%02d").
Using "%02x:%02x.%x" should be fine then, I will change the code
accordingly.
Regards
Nikunj
[Qemu-ppc] [PATCH v5 4/4] spapr_pci: populate ibm,loc-code, Nikunj A Dadhania, 2015/05/19
[Qemu-ppc] [PATCH v5 2/4] spapr_pci: encode class code including Prog IF register, Nikunj A Dadhania, 2015/05/19
[Qemu-ppc] [PATCH v5 1/4] spapr_pci: encode missing 64-bit memory address space, Nikunj A Dadhania, 2015/05/19