Re: [RFC][PATCH 1/3] radix priority search tree - objrmap complexity fix

[Andrea Arcangeli <andrea@suse.de>]

can you try this potential fix too? (maybe you want to try this first
thing)

this is from Hugh's anobjramp patches.

I merged it once, then I got a crash report, so I backed it out since it
was working anyways, but it was due a merging error that it didn't work
correctly, the below version should be fine and it seems really needed.

I'll upload a new kernel with this applied.

--- x/arch/ppc/mm/pgtable.c.~1~	2004-02-20 17:26:33.000000000 +0100
+++ x/arch/ppc/mm/pgtable.c	2004-04-03 18:51:35.072468040 +0200
@@ -86,9 +86,14 @@ pte_t *pte_alloc_one_kernel(struct mm_st
 	extern int mem_init_done;
 	extern void *early_get_page(void);
 
-	if (mem_init_done)
+	if (mem_init_done) {
 		pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT);
-	else
+		if (pte) {
+			struct page *ptepage = virt_to_page(pte);
+			ptepage->mapping = (void *) mm;
+			ptepage->index = address & PMD_MASK;
+		}
+	} else
 		pte = (pte_t *)early_get_page();
 	if (pte)
 		clear_page(pte);
@@ -106,8 +111,11 @@ struct page *pte_alloc_one(struct mm_str
 #endif
 
 	pte = alloc_pages(flags, 0);
-	if (pte)
+	if (pte) {
+		pte->mapping = (void *) mm;
+		pte->index = address & PMD_MASK;
 		clear_highpage(pte);
+	}
 	return pte;
 }
 
@@ -116,6 +124,7 @@ void pte_free_kernel(pte_t *pte)
 #ifdef CONFIG_SMP
 	hash_page_sync();
 #endif
+	virt_to_page(pte)->mapping = NULL;
 	free_page((unsigned long)pte);
 }
 
@@ -124,6 +133,7 @@ void pte_free(struct page *pte)
 #ifdef CONFIG_SMP
 	hash_page_sync();
 #endif
+	pte->mapping = NULL;
 	__free_page(pte);
 }
 
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"aart@kvack.org"> aart@kvack.org </a>