mm: stack based kmap_atomic()

Keep the current interface but ignore the KM_type and use a stack based
approach.

The advantage is that we get rid of crappy code like:

	#define __KM_PTE			\
		(in_nmi() ? KM_NMI_PTE : 	\
		 in_irq() ? KM_IRQ_PTE :	\
		 KM_PTE0)

and in general can stop worrying about what context we're in and what kmap
slots might be appropriate for that.

The downside is that FRV kmap_atomic() gets more expensive.

For now we use a CPP trick suggested by Andrew:

  #define kmap_atomic(page, args...) __kmap_atomic(page)

to avoid having to touch all kmap_atomic() users in a single patch.

[ not compiled on:
  - mn10300: the arch doesn't actually build with highmem to begin with ]

[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: fix up drivers/gpu/drm/i915/intel_overlay.c]
Acked-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Chris Metcalf <cmetcalf@tilera.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: David Miller <davem@davemloft.net>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

2 files changed, 66 insertions, 51 deletions

diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
index 5e8fa12ef861..d723e369003c 100644
--- a/arch/x86/mm/highmem_32.c
+++ b/arch/x86/mm/highmem_32.c
@@ -9,6 +9,7 @@ void *kmap(struct page *page)
                 return page_address(page);
         return kmap_high(page);
 }
+EXPORT_SYMBOL(kmap);
 
 void kunmap(struct page *page)
 {
@@ -18,6 +19,7 @@ void kunmap(struct page *page)
                 return;
         kunmap_high(page);
 }
+EXPORT_SYMBOL(kunmap);
 
 /*
  * kmap_atomic/kunmap_atomic is significantly faster than kmap/kunmap because
@@ -27,10 +29,10 @@ void kunmap(struct page *page)
  * However when holding an atomic kmap it is not legal to sleep, so atomic
  * kmaps are appropriate for short, tight code paths only.
  */
-void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
+void *kmap_atomic_prot(struct page *page, pgprot_t prot)
 {
-        enum fixed_addresses idx;
         unsigned long vaddr;
+        int idx, type;
 
         /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
         pagefault_disable();
@@ -38,8 +40,7 @@ void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
         if (!PageHighMem(page))
                 return page_address(page);
 
-        debug_kmap_atomic(type);
-
+        type = kmap_atomic_idx_push();
         idx = type + KM_TYPE_NR*smp_processor_id();
         vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
         BUG_ON(!pte_none(*(kmap_pte-idx)));
@@ -47,44 +48,56 @@ void *kmap_atomic_prot(struct page *page, enum km_type type, pgprot_t prot)
 
         return (void *)vaddr;
 }
+EXPORT_SYMBOL(kmap_atomic_prot);
+
+void *__kmap_atomic(struct page *page)
+{
+        return kmap_atomic_prot(page, kmap_prot);
+}
+EXPORT_SYMBOL(__kmap_atomic);
 
-void *kmap_atomic(struct page *page, enum km_type type)
+/*
+ * This is the same as kmap_atomic() but can map memory that doesn't
+ * have a struct page associated with it.
+ */
+void *kmap_atomic_pfn(unsigned long pfn)
 {
-        return kmap_atomic_prot(page, type, kmap_prot);
+        return kmap_atomic_prot_pfn(pfn, kmap_prot);
 }
+EXPORT_SYMBOL_GPL(kmap_atomic_pfn);
 
-void kunmap_atomic_notypecheck(void *kvaddr, enum km_type type)
+void __kunmap_atomic(void *kvaddr)
 {
         unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
-        enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
-
-        /*
-         * Force other mappings to Oops if they'll try to access this pte
-         * without first remap it.  Keeping stale mappings around is a bad idea
-         * also, in case the page changes cacheability attributes or becomes
-         * a protected page in a hypervisor.
-         */
-        if (vaddr == __fix_to_virt(FIX_KMAP_BEGIN+idx))
+
+        if (vaddr >= __fix_to_virt(FIX_KMAP_END) &&
+            vaddr <= __fix_to_virt(FIX_KMAP_BEGIN)) {
+                int idx, type;
+
+                type = kmap_atomic_idx_pop();
+                idx = type + KM_TYPE_NR * smp_processor_id();
+
+#ifdef CONFIG_DEBUG_HIGHMEM
+                WARN_ON_ONCE(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
+#endif
+                /*
+                 * Force other mappings to Oops if they'll try to access this
+                 * pte without first remap it.  Keeping stale mappings around
+                 * is a bad idea also, in case the page changes cacheability
+                 * attributes or becomes a protected page in a hypervisor.
+                 */
                 kpte_clear_flush(kmap_pte-idx, vaddr);
-        else {
+        }
 #ifdef CONFIG_DEBUG_HIGHMEM
+        else {
                 BUG_ON(vaddr < PAGE_OFFSET);
                 BUG_ON(vaddr >= (unsigned long)high_memory);
-#endif
         }
+#endif
 
         pagefault_enable();
 }
-
-/*
- * This is the same as kmap_atomic() but can map memory that doesn't
- * have a struct page associated with it.
- */
-void *kmap_atomic_pfn(unsigned long pfn, enum km_type type)
-{
-        return kmap_atomic_prot_pfn(pfn, type, kmap_prot);
-}
-EXPORT_SYMBOL_GPL(kmap_atomic_pfn); /* temporarily in use by i915 GEM until vmap */
+EXPORT_SYMBOL(__kunmap_atomic);
 
 struct page *kmap_atomic_to_page(void *ptr)
 {
@@ -98,12 +111,6 @@ struct page *kmap_atomic_to_page(void *ptr)
         pte = kmap_pte - (idx - FIX_KMAP_BEGIN);
         return pte_page(*pte);
 }
-
-EXPORT_SYMBOL(kmap);
-EXPORT_SYMBOL(kunmap);
-EXPORT_SYMBOL(kmap_atomic);
-EXPORT_SYMBOL(kunmap_atomic_notypecheck);
-EXPORT_SYMBOL(kmap_atomic_prot);
 EXPORT_SYMBOL(kmap_atomic_to_page);
 
 void __init set_highmem_pages_init(void)
diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c
index 72fc70cf6184..75a3d7f24a2c 100644
--- a/arch/x86/mm/iomap_32.c
+++ b/arch/x86/mm/iomap_32.c
@@ -48,21 +48,20 @@ int iomap_create_wc(resource_size_t base, unsigned long size, pgprot_t *prot)
 }
 EXPORT_SYMBOL_GPL(iomap_create_wc);
 
-void
-iomap_free(resource_size_t base, unsigned long size)
+void iomap_free(resource_size_t base, unsigned long size)
 {
         io_free_memtype(base, base + size);
 }
 EXPORT_SYMBOL_GPL(iomap_free);
 
-void *kmap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
+void *kmap_atomic_prot_pfn(unsigned long pfn, pgprot_t prot)
 {
-        enum fixed_addresses idx;
         unsigned long vaddr;
+        int idx, type;
 
         pagefault_disable();
 
-        debug_kmap_atomic(type);
+        type = kmap_atomic_idx_push();
         idx = type + KM_TYPE_NR * smp_processor_id();
         vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
         set_pte(kmap_pte - idx, pfn_pte(pfn, prot));
@@ -72,10 +71,10 @@ void *kmap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
 }
 
 /*
- * Map 'pfn' using fixed map 'type' and protections 'prot'
+ * Map 'pfn' using protections 'prot'
  */
 void __iomem *
-iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
+iomap_atomic_prot_pfn(unsigned long pfn, pgprot_t prot)
 {
         /*
          * For non-PAT systems, promote PAGE_KERNEL_WC to PAGE_KERNEL_UC_MINUS.
@@ -86,24 +85,33 @@ iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot)
         if (!pat_enabled && pgprot_val(prot) == pgprot_val(PAGE_KERNEL_WC))
                 prot = PAGE_KERNEL_UC_MINUS;
 
-        return (void __force __iomem *) kmap_atomic_prot_pfn(pfn, type, prot);
+        return (void __force __iomem *) kmap_atomic_prot_pfn(pfn, prot);
 }
 EXPORT_SYMBOL_GPL(iomap_atomic_prot_pfn);
 
 void
-iounmap_atomic(void __iomem *kvaddr, enum km_type type)
+iounmap_atomic(void __iomem *kvaddr)
 {
         unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
-        enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
 
-        /*
-         * Force other mappings to Oops if they'll try to access this pte
-         * without first remap it.  Keeping stale mappings around is a bad idea
-         * also, in case the page changes cacheability attributes or becomes
-         * a protected page in a hypervisor.
-         */
-        if (vaddr == __fix_to_virt(FIX_KMAP_BEGIN+idx))
+        if (vaddr >= __fix_to_virt(FIX_KMAP_END) &&
+            vaddr <= __fix_to_virt(FIX_KMAP_BEGIN)) {
+                int idx, type;
+
+                type = kmap_atomic_idx_pop();
+                idx = type + KM_TYPE_NR * smp_processor_id();
+
+#ifdef CONFIG_DEBUG_HIGHMEM
+                WARN_ON_ONCE(vaddr != __fix_to_virt(FIX_KMAP_BEGIN + idx));
+#endif
+                /*
+                 * Force other mappings to Oops if they'll try to access this
+                 * pte without first remap it.  Keeping stale mappings around
+                 * is a bad idea also, in case the page changes cacheability
+                 * attributes or becomes a protected page in a hypervisor.
+                 */
                 kpte_clear_flush(kmap_pte-idx, vaddr);
+        }
 
         pagefault_enable();
 }