bitkeeper revision 1.1389.1.13 (42722e78tJ9dMvMOlJBiS68RNPTvfw)

[PATCH] [patch] pgtable.c cleanups

  Hi guys,

The new mm_(un)pin stuff is certainly a nice optimization.

But can we take care a bit more about coding style issues
and write code not *that* hackish please?  Casts, code
duplication, "unsigned long" for page table entries *again*.
And a few more variables also make the code more readable
(and gcc should be clever enougth that this isn't a
performance hit).  We want to have that merged some day in
mainline, don't we?

The patch below cleans up the pin/unpin code in pgtable.c
I've created two helper functions to factor out common code,
the page table walk is basically the same for both pin and
unpin, it's just different page flags.  Also fixed up the
walk code to correcly walk through all 4 levels linux has,
so it works correctly for the PAE case.  For x86_64 it
should work fine as well (untested though).

please apply,

  Gerd

BitKeeper/etc/logging_ok

@@ -56,6 +56,7 @@ kaf24@striker.cl.cam.ac.uk
 kaf24@viper.(none)
 katzj@redhat.com
 kmacy@shemp.lab.netapp.com
+kraxel@bytesex.org
 laudney@eclipse.(none)
 leendert@watson.ibm.com
 lynx@idefix.cl.cam.ac.uk

linux-2.6.11-xen-sparse/arch/xen/i386/mm/pgtable.c

@@ -407,30 +407,58 @@ void make_pages_writable(void *va, unsigned int nr)
 }
 #endif /* CONFIG_XEN_SHADOW_MODE */
 
+static inline void mm_walk_set_prot(void *pt, pgprot_t flags)
+{
+	struct page *page = virt_to_page(pt);
+	unsigned long pfn = page_to_pfn(page);
+
+	if (PageHighMem(page))
+		return;
+	HYPERVISOR_update_va_mapping(
+		(unsigned long)__va(pfn << PAGE_SHIFT),
+		pfn_pte(pfn, flags), 0);
+}
+
+static void mm_walk(struct mm_struct *mm, pgprot_t flags)
+{
+	pgd_t       *pgd;
+	pud_t       *pud;
+	pmd_t       *pmd;
+	pte_t       *pte;
+	int          g,u,m;
+
+	pgd = mm->pgd;
+	for (g = 0; g < USER_PTRS_PER_PGD; g++, pgd++) {
+		if (pgd_none(*pgd))
+			continue;
+		pud = pud_offset(pgd, 0);
+		if (PTRS_PER_PUD > 1) /* not folded */
+			mm_walk_set_prot(pud,flags);
+		for (u = 0; u < PTRS_PER_PUD; u++, pud++) {
+			if (pud_none(*pud))
+				continue;
+			pmd = pmd_offset(pud, 0);
+			if (PTRS_PER_PMD > 1) /* not folded */
+				mm_walk_set_prot(pmd,flags);
+			for (m = 0; m < PTRS_PER_PMD; m++, pmd++) {
+				if (pmd_none(*pmd))
+					continue;
+				pte = pte_offset_kernel(pmd,0);
+				mm_walk_set_prot(pte,flags);
+			}
+		}
+	}
+}
+
 void mm_pin(struct mm_struct *mm)
 {
-    pgd_t       *pgd;
-    struct page *page;
-    int          i;
-
     spin_lock(&mm->page_table_lock);
 
-    for ( i = 0, pgd = mm->pgd; i < USER_PTRS_PER_PGD; i++, pgd++ )
-    {
-        if ( *(unsigned long *)pgd == 0 )
-            continue;
-        page = pmd_page(*(pmd_t *)pgd);
-        if ( !PageHighMem(page) )
-            HYPERVISOR_update_va_mapping(
-                (unsigned long)__va(page_to_pfn(page)<<PAGE_SHIFT),
-                pfn_pte(page_to_pfn(page), PAGE_KERNEL_RO), 0);
-    }
-
+    mm_walk(mm, PAGE_KERNEL_RO);
     HYPERVISOR_update_va_mapping(
         (unsigned long)mm->pgd,
         pfn_pte(virt_to_phys(mm->pgd)>>PAGE_SHIFT, PAGE_KERNEL_RO), 0);
     xen_pgd_pin(__pa(mm->pgd));
-
     mm->context.pinned = 1;
 
     spin_unlock(&mm->page_table_lock);
@@ -438,28 +466,13 @@ void mm_pin(struct mm_struct *mm)
 
 void mm_unpin(struct mm_struct *mm)
 {
-    pgd_t       *pgd;
-    struct page *page;
-    int          i;
-
     spin_lock(&mm->page_table_lock);
 
     xen_pgd_unpin(__pa(mm->pgd));
     HYPERVISOR_update_va_mapping(
         (unsigned long)mm->pgd,
         pfn_pte(virt_to_phys(mm->pgd)>>PAGE_SHIFT, PAGE_KERNEL), 0);
-
-    for ( i = 0, pgd = mm->pgd; i < USER_PTRS_PER_PGD; i++, pgd++ )
-    {
-        if ( *(unsigned long *)pgd == 0 )
-            continue;
-        page = pmd_page(*(pmd_t *)pgd);
-        if ( !PageHighMem(page) )
-            HYPERVISOR_update_va_mapping(
-                (unsigned long)__va(page_to_pfn(page)<<PAGE_SHIFT),
-                pfn_pte(page_to_pfn(page), PAGE_KERNEL), 0);
-    }
-
+    mm_walk(mm, PAGE_KERNEL);
     mm->context.pinned = 0;
 
     spin_unlock(&mm->page_table_lock);