diff mbox

[RFC,1/2] pci: determine CLS more intelligently

Message ID 20090625.020507.203808859.davem@davemloft.net (mailing list archive)
State Superseded, archived
Headers show

Commit Message

David Miller June 25, 2009, 9:05 a.m. UTC
From: Tejun Heo <tj@kernel.org>
Date: Thu, 25 Jun 2009 16:12:13 +0900

> sparc64 is currently the only one using PCI_CACHE_LINE_BYTES.

Feel free to add the patch below and:

Acked-by: David S. Miller <davem@davemloft.net>

Although I think it's better to declare pci_dfl_cache_line_size in
linux/pci.h instead of making every arch do the extern decl.

--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Greg KH June 25, 2009, 5:33 p.m. UTC | #1
On Thu, Jun 25, 2009 at 02:05:07AM -0700, David Miller wrote:
> From: Tejun Heo <tj@kernel.org>
> Date: Thu, 25 Jun 2009 16:12:13 +0900
> 
> > sparc64 is currently the only one using PCI_CACHE_LINE_BYTES.
> 
> Feel free to add the patch below and:
> 
> Acked-by: David S. Miller <davem@davemloft.net>
> 
> Although I think it's better to declare pci_dfl_cache_line_size in
> linux/pci.h instead of making every arch do the extern decl.

I agree, that would make more sense.

Otherwise it looks good to me:
	Acked-by: Greg Kroah-Hartman <gregkh@suse.de>

thanks,

greg k-h
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Tejun Heo June 30, 2009, 3:07 a.m. UTC | #2
Greg KH wrote:
> On Thu, Jun 25, 2009 at 02:05:07AM -0700, David Miller wrote:
>> From: Tejun Heo <tj@kernel.org>
>> Date: Thu, 25 Jun 2009 16:12:13 +0900
>>
>>> sparc64 is currently the only one using PCI_CACHE_LINE_BYTES.
>> Feel free to add the patch below and:
>>
>> Acked-by: David S. Miller <davem@davemloft.net>
>>
>> Although I think it's better to declare pci_dfl_cache_line_size in
>> linux/pci.h instead of making every arch do the extern decl.
> 
> I agree, that would make more sense.
> 
> Otherwise it looks good to me:
> 	Acked-by: Greg Kroah-Hartman <gregkh@suse.de>

Great, I'll incorporate both comments and repost the patchset after
Benjamin's testing or before the end of this week.

Thanks.
Benjamin Herrenschmidt June 30, 2009, 10:22 a.m. UTC | #3
On Tue, 2009-06-30 at 12:07 +0900, Tejun Heo wrote:
> 
> Great, I'll incorporate both comments and repost the patchset after
> Benjamin's testing or before the end of this week.

Just go ahead if you can test build powerpc. I'm on vacation on the
coast and won't be able to do much testing :-)

In any case, it shouldn't affect us (hopefully).

Cheers,
Ben.


--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/arch/sparc/kernel/pci.c b/arch/sparc/kernel/pci.c
index 57859ad..511a592 100644
--- a/arch/sparc/kernel/pci.c
+++ b/arch/sparc/kernel/pci.c
@@ -1081,3 +1081,12 @@  void pci_resource_to_user(const struct pci_dev *pdev, int bar,
 	*start = rp->start - offset;
 	*end = rp->end - offset;
 }
+
+extern u8 pci_dfl_cache_line_size;
+
+static int __init pcibios_init(void)
+{
+	pci_dfl_cache_line_size = 64 >> 2;
+	return 0;
+}
+subsys_initcall(pcibios_init);