unify pfn_to_page [25/25] sparc64 pfn_to_page/page_to_pfn
RFC PATCH 1/1 unify pfn_to_page [25/25] sparc64 pfn_to_page/page_to_pfn
sparc64 can use generic ones by defining ARCH_PFN_OFFSET as pfn_base.
include/asm-sparc64/page.h +1 -2
@@ -129,8 +129,7 @@ typedef unsigned long pgprot_t;* the first physical page in the machine is at some huge physical address, * such as 4GB. This is common on a partitioned E10000, for example. */ -extern struct page *pfn_to_page (unsigned long pfn); -extern unsigned long page_to_pfn (struct page *); +#define ARCH_PFN_OFFSET (pfn_base)#define virt_to_page (kaddr) pfn_to_page(__pa(kaddr)>>PAGE_SHIFT)
arch/sparc64/mm/init.c +0 -10
@@ -320,16 +320,6 @@ void __kprobes flush_icache_range(unsign } } -unsigned long page_to_pfn (struct page *page) -{ -return (unsigned long ) ((page - mem_map) + pfn_base); -} - -struct page *pfn_to_page (unsigned long pfn) -{ -return (mem_map + (pfn - pfn_base)); -} -void show_mem (void ) {printk ("Mem-info:\n" );
From: KAMEZAWA Hiroyuki <[email protected]>
Date: Mon, 06 Feb 2006 20:23:33 +0900
Quoted 9 lines
arch/sparc64/mm/init.c +0
-3@@ -320,16 +320,6 @@ void __kprobes flush_icache_range(unsign } } -unsigned long page_to_pfn (struct page *page) -{ -return (unsigned long ) ((page - mem_map) + pfn_base);
arch/sparc64/mm/init.c void __kprobes flush_icache_range(unsign · line 325
Quoted 10 lines
arch/sparc64/mm/init.c +0
-7-} - -struct page *pfn_to_page (unsigned long pfn) -{ -return (mem_map + (pfn - pfn_base)); -} -void show_mem (void ) {printk ("Mem-info:\n" );
We did not want these inlined on sparc64 for a good reason. The pointer arithmatic gets expanded to many additions, subtractions, and shifts, and I felt it too much to inline.
If you want to consolidate all of the implementations, that's fine, but please keep the option of not inlining these two routines.
Thank you.
Quoted 4 lines · David S. Miller
We did not want these inlined on sparc64 for a good reason. The pointer arithmatic gets expanded to many additions, subtractions, and shifts, and I felt it too much to inline.
Okay.
Quoted 4 lines
If you want to consolidate all of the implementations, that's fine, but please keep the option of not inlining these two routines.
I'll do (and CC to each arch's maintainer in next post)
Thanks,
-- Kame