On Wed, 20 Apr 2005 17:37:11 -0700
Tom Duffy <[EMAIL PROTECTED]> wrote:
> This breaks building on sparc64:
...
> This is ugly, but fixes the build. Perhaps sparc needs
> pgprot_noncached() to be a noop?
No, it should actually do something, like so:
include/asm-sparc64/pgtable.h: af9bf175a223cf44310293287d50302e0fd3f9e9
--- a/include/asm-sparc64/pgtable.h
+++ b/include/asm-sparc64/pgtable.h
@@ -416,6 +416,11 @@ extern int io_remap_pfn_range(struct vm_
unsigned long pfn,
unsigned long size, pgprot_t prot);
+/* Clear virtual and physical cachability, set side-effect bit. */
+#define pgprot_noncached(prot) \
+ (__pgprot((pgprot_val(prot) & ~(_PAGE_CP | _PAGE_CV)) | \
+ _PAGE_E))
+
/*
* For sparc32&64, the pfn in io_remap_pfn_range() carries <iospace> in
* its high 4 bits. These macros/functions put it there or get it from there.
_______________________________________________
openib-general mailing list
[email protected]
http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
