Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1307419 > unrolled thread

Re: [PATCH V2 11/23] arm64, pci: Remove useless boot time IRQ assignment when booting with DT.

Started byLorenzo Pieralisi <lorenzo.pieralisi@arm.com>
First post2016-01-12 14:50 +0100
Last post2016-01-13 10:50 +0100
Articles 3 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH V2 11/23] arm64, pci: Remove useless boot time IRQ  assignment when booting with DT. Lorenzo Pieralisi <lorenzo.pieralisi@arm.com> - 2016-01-12 14:50 +0100
    Re: [PATCH V2 11/23] arm64, pci: Remove useless boot time IRQ  assignment when booting with DT. Tomasz Nowicki <tn@semihalf.com> - 2016-01-12 17:20 +0100
      Re: [PATCH V2 11/23] arm64, pci: Remove useless boot time IRQ  assignment when booting with DT. Tomasz Nowicki <tn@semihalf.com> - 2016-01-13 10:50 +0100

#1307419 — Re: [PATCH V2 11/23] arm64, pci: Remove useless boot time IRQ assignment when booting with DT.

FromLorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Date2016-01-12 14:50 +0100
SubjectRe: [PATCH V2 11/23] arm64, pci: Remove useless boot time IRQ assignment when booting with DT.
Message-ID<qQ7uH-1zT-27@gated-at.bofh.it>
On Wed, Dec 16, 2015 at 04:16:21PM +0100, Tomasz Nowicki wrote:
> In order to probe PCIe host controller when booting with DT,
> ARM64 is using drivers which defer IRQ assignment to device
> enable time. It means that boot time DT specific irq map initialization
> is always overridden, so lets remove that code.
> 
> Signed-off-by: Tomasz Nowicki <tn@semihalf.com>
> ---
>  arch/arm64/kernel/pci.c | 10 ----------
>  1 file changed, 10 deletions(-)
> 
> diff --git a/arch/arm64/kernel/pci.c b/arch/arm64/kernel/pci.c
> index 023b983..f7948f5 100644
> --- a/arch/arm64/kernel/pci.c
> +++ b/arch/arm64/kernel/pci.c
> @@ -51,16 +51,6 @@ int pcibios_enable_device(struct pci_dev *dev, int mask)
>  	return pci_enable_resources(dev, mask);
>  }
>  
> -/*
> - * Try to assign the IRQ number from DT when adding a new device
> - */
> -int pcibios_add_device(struct pci_dev *dev)
> -{
> -	dev->irq = of_irq_parse_and_map_pci(dev, 0, 0);
> -
> -	return 0;
> -}
> -

Actually, this patch should be part of Matthew's series:

http://comments.gmane.org/gmane.linux.kernel.pci/46461

Lorenzo

[toc] | [next] | [standalone]


#1307596

FromTomasz Nowicki <tn@semihalf.com>
Date2016-01-12 17:20 +0100
Message-ID<qQ9PR-3lA-23@gated-at.bofh.it>
In reply to#1307419
On 12.01.2016 14:50, Lorenzo Pieralisi wrote:
> On Wed, Dec 16, 2015 at 04:16:21PM +0100, Tomasz Nowicki wrote:
>> In order to probe PCIe host controller when booting with DT,
>> ARM64 is using drivers which defer IRQ assignment to device
>> enable time. It means that boot time DT specific irq map initialization
>> is always overridden, so lets remove that code.
>>
>> Signed-off-by: Tomasz Nowicki <tn@semihalf.com>
>> ---
>>   arch/arm64/kernel/pci.c | 10 ----------
>>   1 file changed, 10 deletions(-)
>>
>> diff --git a/arch/arm64/kernel/pci.c b/arch/arm64/kernel/pci.c
>> index 023b983..f7948f5 100644
>> --- a/arch/arm64/kernel/pci.c
>> +++ b/arch/arm64/kernel/pci.c
>> @@ -51,16 +51,6 @@ int pcibios_enable_device(struct pci_dev *dev, int mask)
>>   	return pci_enable_resources(dev, mask);
>>   }
>>
>> -/*
>> - * Try to assign the IRQ number from DT when adding a new device
>> - */
>> -int pcibios_add_device(struct pci_dev *dev)
>> -{
>> -	dev->irq = of_irq_parse_and_map_pci(dev, 0, 0);
>> -
>> -	return 0;
>> -}
>> -
>
> Actually, this patch should be part of Matthew's series:
>

Agree.

Matthew can you please add this patch to your series?

Thanks,
Tomasz

[toc] | [prev] | [next] | [standalone]


#1308225

FromTomasz Nowicki <tn@semihalf.com>
Date2016-01-13 10:50 +0100
Message-ID<qQqdZ-6f4-21@gated-at.bofh.it>
In reply to#1307596
On 12.01.2016 18:56, David Daney wrote:
> On 01/12/2016 08:13 AM, Tomasz Nowicki wrote:
>> On 12.01.2016 14:50, Lorenzo Pieralisi wrote:
>>> On Wed, Dec 16, 2015 at 04:16:21PM +0100, Tomasz Nowicki wrote:
>>>> In order to probe PCIe host controller when booting with DT,
>>>> ARM64 is using drivers which defer IRQ assignment to device
>>>> enable time. It means that boot time DT specific irq map initialization
>>>> is always overridden, so lets remove that code.
>>>>
>>>> Signed-off-by: Tomasz Nowicki <tn@semihalf.com>
>>>> ---
>>>>   arch/arm64/kernel/pci.c | 10 ----------
>>>>   1 file changed, 10 deletions(-)
>>>>
>>>> diff --git a/arch/arm64/kernel/pci.c b/arch/arm64/kernel/pci.c
>>>> index 023b983..f7948f5 100644
>>>> --- a/arch/arm64/kernel/pci.c
>>>> +++ b/arch/arm64/kernel/pci.c
>>>> @@ -51,16 +51,6 @@ int pcibios_enable_device(struct pci_dev *dev, int
>>>> mask)
>>>>       return pci_enable_resources(dev, mask);
>>>>   }
>>>>
>>>> -/*
>>>> - * Try to assign the IRQ number from DT when adding a new device
>>>> - */
>>>> -int pcibios_add_device(struct pci_dev *dev)
>>>> -{
>>>> -    dev->irq = of_irq_parse_and_map_pci(dev, 0, 0);
>>>> -
>>>> -    return 0;
>>>> -}
>>>> -
>
> How was this tested?  Or in other words, what PCI devices that use
> legacy INT{A,B,C,D} interrupts were used in testing this patch?
>

I used QEMU and e1000 NIC:

sudo ./qemu/aarch64-softmmu/qemu-system-aarch64 -smp 1 -m 1024 -M virt 
-cpu cortex-a57 -nographic -device 
e1000,netdev=net1,mac=52:54:00:12:34:56 -netdev 
type=tap,ifname=tun1,id=net1,script=no,downscript=no -drive 
file=qemu/ubuntu.img,id=root,if=none,format=raw -device 
virtio-blk-device,drive=root -kernel linux-aarch64/arch/arm64/boot/Image 
-append "console=ttyAMA0 earlycon=pl011,0x9000000 rw root=/dev/vda"

root@ubuntu:~# ifconfig
eth0      Link encap:Ethernet  HWaddr 52:54:00:12:34:56
           inet addr:10.0.0.48  Bcast:10.0.255.255  Mask:255.255.0.0
           UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
           RX packets:107 errors:0 dropped:0 overruns:0 frame:0
           TX packets:32 errors:0 dropped:0 overruns:0 carrier:0
           collisions:0 txqueuelen:1000
           RX bytes:18087 (18.0 KB)  TX bytes:3818 (3.8 KB)

root@ubuntu:~# cat /proc/interrupts
            CPU0
  39:        280       GIC  36 Level     eth0

root@ubuntu:~# lspci -vvv
00:00.0 Host bridge: Red Hat, Inc. Device 0008
         Subsystem: Red Hat, Inc Device 1100
         Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- 
ParErr- Stepping- SERR- FastB2B- DisINTx-
         Status: Cap- 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- 
<TAbort- <MAbort- >SERR- <PERR- INTx-

00:01.0 Ethernet controller: Intel Corporation 82540EM Gigabit Ethernet 
Controller (rev 03)
         Subsystem: Red Hat, Inc QEMU Virtual Machine
         Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- 
ParErr- Stepping- SERR- FastB2B- DisINTx-
         Status: Cap- 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- 
<TAbort- <MAbort- >SERR- <PERR- INTx-
         Latency: 0
         Interrupt: pin A routed to IRQ 39
         Region 0: Memory at 10040000 (32-bit, non-prefetchable) [size=128K]
         Region 1: I/O ports at 1000 [size=64]
         [virtual] Expansion ROM at 10000000 [disabled] [size=256K]
         Kernel driver in use: e1000

Any specific concern w.r.t. this patch?

Tomasz

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web