On Sep 6, 2013, at 10:36 AM, Scott Wood wrote:

On Fri, 2013-09-06 at 10:01 -0500, Kumar Gala wrote:

quoted

On Sep 5, 2013, at 1:37 PM, Scott Wood wrote:
=20

quoted

On Thu, 2013-09-05 at 13:34 -0500, Kumar Gala wrote:

quoted

On Apr 2, 2013, at 9:03 PM, Jia Hongtao wrote:

quoted

+			msi->feature |=3D MSI_HW_ERRATA_ENDIAN;
+	}
+
	/*
	 * Remember the phandle, so that we can match with any PCI nodes
	 * that have an "fsl,msi" property.

diff --git a/arch/powerpc/sysdev/fsl_msi.h =

b/arch/powerpc/sysdev/fsl_msi.h

quoted

quoted

quoted

quoted

index 8225f86..7389e8e 100644

--- a/arch/powerpc/sysdev/fsl_msi.h
+++ b/arch/powerpc/sysdev/fsl_msi.h

@@ -25,6 +25,8 @@

#define FSL_PIC_IP_IPIC   0x00000002
#define FSL_PIC_IP_VMPIC  0x00000003
=20
+#define MSI_HW_ERRATA_ENDIAN 0x00000010
+

=20
Why does this need to be in the header, why not just have it in the =

.c only

quoted

quoted

=20
Didn't you ask this last time around? :-)
=20
This flag is part of the same numberspace as FSL_PIC_IP_xxx and thus
should be defined in the same place.

=20
I probably did, if its part of the FSL_PIC_IP_xxx namespace, than =

lets remove blank line between things to make that a bit more clear

=20
It's not part of the FSL_PIC_IP_MASK subnumberspace though, just the
overall msi->features numberspace.
=20
It would be nice if these symbols could have some sort of prefix in
common, though.
=20
-Scott

Maybe we should do something like MSI_FTR_ as a prefix

- k