From f00a75c094c340c4e7435665816c3273c870e849 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Mon, 5 Oct 2009 15:17:45 +0100
Subject: ARM: Pass VMA to copy_user_highpage() implementations

Our copy_user_highpage() implementations may require cache maintainence.
Ensure that implementations have all necessary details to perform this
maintainence.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/copypage-feroceon.c | 2 +-
 arch/arm/mm/copypage-v3.c       | 2 +-
 arch/arm/mm/copypage-v4mc.c     | 2 +-
 arch/arm/mm/copypage-v4wb.c     | 2 +-
 arch/arm/mm/copypage-v4wt.c     | 2 +-
 arch/arm/mm/copypage-v6.c       | 4 ++--
 arch/arm/mm/copypage-xsc3.c     | 2 +-
 arch/arm/mm/copypage-xscale.c   | 2 +-
 8 files changed, 9 insertions(+), 9 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/copypage-feroceon.c b/arch/arm/mm/copypage-feroceon.c
index 70997d5bee2d..e2e8d29c548b 100644
--- a/arch/arm/mm/copypage-feroceon.c
+++ b/arch/arm/mm/copypage-feroceon.c
@@ -68,7 +68,7 @@ feroceon_copy_user_page(void *kto, const void *kfrom)
 }
 
 void feroceon_copy_user_highpage(struct page *to, struct page *from,
-	unsigned long vaddr)
+	unsigned long vaddr, struct vm_area_struct *vma)
 {
 	void *kto, *kfrom;
 
diff --git a/arch/arm/mm/copypage-v3.c b/arch/arm/mm/copypage-v3.c
index de9c06854ad7..f72303e1d804 100644
--- a/arch/arm/mm/copypage-v3.c
+++ b/arch/arm/mm/copypage-v3.c
@@ -38,7 +38,7 @@ v3_copy_user_page(void *kto, const void *kfrom)
 }
 
 void v3_copy_user_highpage(struct page *to, struct page *from,
-	unsigned long vaddr)
+	unsigned long vaddr, struct vm_area_struct *vma)
 {
 	void *kto, *kfrom;
 
diff --git a/arch/arm/mm/copypage-v4mc.c b/arch/arm/mm/copypage-v4mc.c
index 7370a7142b04..598c51ad5071 100644
--- a/arch/arm/mm/copypage-v4mc.c
+++ b/arch/arm/mm/copypage-v4mc.c
@@ -69,7 +69,7 @@ mc_copy_user_page(void *from, void *to)
 }
 
 void v4_mc_copy_user_highpage(struct page *to, struct page *from,
-	unsigned long vaddr)
+	unsigned long vaddr, struct vm_area_struct *vma)
 {
 	void *kto = kmap_atomic(to, KM_USER1);
 
diff --git a/arch/arm/mm/copypage-v4wb.c b/arch/arm/mm/copypage-v4wb.c
index 9ab098414227..e9920f68b76b 100644
--- a/arch/arm/mm/copypage-v4wb.c
+++ b/arch/arm/mm/copypage-v4wb.c
@@ -48,7 +48,7 @@ v4wb_copy_user_page(void *kto, const void *kfrom)
 }
 
 void v4wb_copy_user_highpage(struct page *to, struct page *from,
-	unsigned long vaddr)
+	unsigned long vaddr, struct vm_area_struct *vma)
 {
 	void *kto, *kfrom;
 
diff --git a/arch/arm/mm/copypage-v4wt.c b/arch/arm/mm/copypage-v4wt.c
index 300efafd6643..172e6a55458e 100644
--- a/arch/arm/mm/copypage-v4wt.c
+++ b/arch/arm/mm/copypage-v4wt.c
@@ -44,7 +44,7 @@ v4wt_copy_user_page(void *kto, const void *kfrom)
 }
 
 void v4wt_copy_user_highpage(struct page *to, struct page *from,
-	unsigned long vaddr)
+	unsigned long vaddr, struct vm_area_struct *vma)
 {
 	void *kto, *kfrom;
 
diff --git a/arch/arm/mm/copypage-v6.c b/arch/arm/mm/copypage-v6.c
index 4127a7bddfe5..334d5602770e 100644
--- a/arch/arm/mm/copypage-v6.c
+++ b/arch/arm/mm/copypage-v6.c
@@ -34,7 +34,7 @@ static DEFINE_SPINLOCK(v6_lock);
  * attack the kernel's existing mapping of these pages.
  */
 static void v6_copy_user_highpage_nonaliasing(struct page *to,
-	struct page *from, unsigned long vaddr)
+	struct page *from, unsigned long vaddr, struct vm_area_struct *vma)
 {
 	void *kto, *kfrom;
 
@@ -73,7 +73,7 @@ static void discard_old_kernel_data(void *kto)
  * Copy the page, taking account of the cache colour.
  */
 static void v6_copy_user_highpage_aliasing(struct page *to,
-	struct page *from, unsigned long vaddr)
+	struct page *from, unsigned long vaddr, struct vm_area_struct *vma)
 {
 	unsigned int offset = CACHE_COLOUR(vaddr);
 	unsigned long kfrom, kto;
diff --git a/arch/arm/mm/copypage-xsc3.c b/arch/arm/mm/copypage-xsc3.c
index bc4525f5ab23..18ae05d5829c 100644
--- a/arch/arm/mm/copypage-xsc3.c
+++ b/arch/arm/mm/copypage-xsc3.c
@@ -71,7 +71,7 @@ xsc3_mc_copy_user_page(void *kto, const void *kfrom)
 }
 
 void xsc3_mc_copy_user_highpage(struct page *to, struct page *from,
-	unsigned long vaddr)
+	unsigned long vaddr, struct vm_area_struct *vma)
 {
 	void *kto, *kfrom;
 
diff --git a/arch/arm/mm/copypage-xscale.c b/arch/arm/mm/copypage-xscale.c
index 76824d3e966a..9920c0ae2096 100644
--- a/arch/arm/mm/copypage-xscale.c
+++ b/arch/arm/mm/copypage-xscale.c
@@ -91,7 +91,7 @@ mc_copy_user_page(void *from, void *to)
 }
 
 void xscale_mc_copy_user_highpage(struct page *to, struct page *from,
-	unsigned long vaddr)
+	unsigned long vaddr, struct vm_area_struct *vma)
 {
 	void *kto = kmap_atomic(to, KM_USER1);
 
-- 
cgit v1.2.2


From 2725898fc9bb2121ac0fb1b5e4faf4fc09014729 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Mon, 5 Oct 2009 15:34:22 +0100
Subject: ARM: Flush user mapping on VIVT processors when copying a page

Steven Walter <stevenrwalter@gmail.com> writes:
> I've been tracking down an instance of userspace data corruption,
> and I believe I have found a window during fork where data can be
> lost.  The corruption is occurring on an ARMv5 system with VIVT
> caches.  Here's the scenario in question.  Thread A is forking,
> Thread B is running in userspace:
>
> Thread A: flush_cache_mm() (dup_mmap)
> Thread B: writes to a page in the above mm
> Thread A: pte_wrprotect() the above page (copy_one_pte)
> Thread B: writes to the same page again
>
> During thread B's second write, he'll take a fault and enter the
> do_wp_page() case.  We'll end up calling copy_page(), which notably
> uses the kernel virtual addresses for the old and new pages.  This
> means that the new page does not necessarily have the data from the
> first write.  Now there are two conflicting copies of the same
> cache-line in dcache.  If the userspace cache-line flushes before
> the kernel cache-line, we lose the changes made during the first
> write.  do_wp_page does call flush_dcache_page on the newly-copied
> page, but there's still a window where the CPU could flush the
> userspace cache-line before then.

Resolve this by flushing the user mapping before copying the page
on processors with a writeback VIVT cache.

Note: this does have a performance impact, and so needs further
consideration before being merged - can we optimize out some of
the cache flushes if, eg, we know that the page isn't yet mapped?

Thread: <e06498070903061426o5875ad13hc6328aa0d3f08ed7@mail.gmail.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/copypage-feroceon.c | 1 +
 arch/arm/mm/copypage-v4wb.c     | 1 +
 arch/arm/mm/copypage-xsc3.c     | 1 +
 3 files changed, 3 insertions(+)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/copypage-feroceon.c b/arch/arm/mm/copypage-feroceon.c
index e2e8d29c548b..5eb4fd93893d 100644
--- a/arch/arm/mm/copypage-feroceon.c
+++ b/arch/arm/mm/copypage-feroceon.c
@@ -74,6 +74,7 @@ void feroceon_copy_user_highpage(struct page *to, struct page *from,
 
 	kto = kmap_atomic(to, KM_USER0);
 	kfrom = kmap_atomic(from, KM_USER1);
+	flush_cache_page(vma, vaddr, page_to_pfn(from));
 	feroceon_copy_user_page(kto, kfrom);
 	kunmap_atomic(kfrom, KM_USER1);
 	kunmap_atomic(kto, KM_USER0);
diff --git a/arch/arm/mm/copypage-v4wb.c b/arch/arm/mm/copypage-v4wb.c
index e9920f68b76b..7c2eb55cd4a9 100644
--- a/arch/arm/mm/copypage-v4wb.c
+++ b/arch/arm/mm/copypage-v4wb.c
@@ -54,6 +54,7 @@ void v4wb_copy_user_highpage(struct page *to, struct page *from,
 
 	kto = kmap_atomic(to, KM_USER0);
 	kfrom = kmap_atomic(from, KM_USER1);
+	flush_cache_page(vma, vaddr, page_to_pfn(from));
 	v4wb_copy_user_page(kto, kfrom);
 	kunmap_atomic(kfrom, KM_USER1);
 	kunmap_atomic(kto, KM_USER0);
diff --git a/arch/arm/mm/copypage-xsc3.c b/arch/arm/mm/copypage-xsc3.c
index 18ae05d5829c..747ad4140fc7 100644
--- a/arch/arm/mm/copypage-xsc3.c
+++ b/arch/arm/mm/copypage-xsc3.c
@@ -77,6 +77,7 @@ void xsc3_mc_copy_user_highpage(struct page *to, struct page *from,
 
 	kto = kmap_atomic(to, KM_USER0);
 	kfrom = kmap_atomic(from, KM_USER1);
+	flush_cache_page(vma, vaddr, page_to_pfn(from));
 	xsc3_mc_copy_user_page(kto, kfrom);
 	kunmap_atomic(kfrom, KM_USER1);
 	kunmap_atomic(kto, KM_USER0);
-- 
cgit v1.2.2


From 2ef7f3dbd7a70a48c3f09b498df528cb00ea03a4 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 5 Nov 2009 13:29:36 +0000
Subject: ARM: Fix ptrace accesses

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/flush.c | 51 +++++++++++++++++++++++++++++++++++++++++++--------
 1 file changed, 43 insertions(+), 8 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c
index 6f3a4b7a3b82..e34f095e2090 100644
--- a/arch/arm/mm/flush.c
+++ b/arch/arm/mm/flush.c
@@ -13,6 +13,7 @@
 
 #include <asm/cacheflush.h>
 #include <asm/cachetype.h>
+#include <asm/smp_plat.h>
 #include <asm/system.h>
 #include <asm/tlbflush.h>
 
@@ -87,13 +88,26 @@ void flush_cache_page(struct vm_area_struct *vma, unsigned long user_addr, unsig
 	if (vma->vm_flags & VM_EXEC && icache_is_vivt_asid_tagged())
 		__flush_icache_all();
 }
+#else
+#define flush_pfn_alias(pfn,vaddr)	do { } while (0)
+#endif
 
+#ifdef CONFIG_SMP
+static void flush_ptrace_access_other(void *args)
+{
+	__flush_icache_all();
+}
+#endif
+
+static
 void flush_ptrace_access(struct vm_area_struct *vma, struct page *page,
-			 unsigned long uaddr, void *kaddr,
-			 unsigned long len, int write)
+			 unsigned long uaddr, void *kaddr, unsigned long len)
 {
 	if (cache_is_vivt()) {
-		vivt_flush_ptrace_access(vma, page, uaddr, kaddr, len, write);
+		if (cpumask_test_cpu(smp_processor_id(), mm_cpumask(vma->vm_mm))) {
+			unsigned long addr = (unsigned long)kaddr;
+			__cpuc_coherent_kern_range(addr, addr + len);
+		}
 		return;
 	}
 
@@ -104,16 +118,37 @@ void flush_ptrace_access(struct vm_area_struct *vma, struct page *page,
 	}
 
 	/* VIPT non-aliasing cache */
-	if (cpumask_test_cpu(smp_processor_id(), mm_cpumask(vma->vm_mm)) &&
-	    vma->vm_flags & VM_EXEC) {
+	if (vma->vm_flags & VM_EXEC) {
 		unsigned long addr = (unsigned long)kaddr;
-		/* only flushing the kernel mapping on non-aliasing VIPT */
 		__cpuc_coherent_kern_range(addr, addr + len);
+#ifdef CONFIG_SMP
+		if (cache_ops_need_broadcast())
+			smp_call_function(flush_ptrace_access_other,
+					  NULL, 1);
+#endif
 	}
 }
-#else
-#define flush_pfn_alias(pfn,vaddr)	do { } while (0)
+
+/*
+ * Copy user data from/to a page which is mapped into a different
+ * processes address space.  Really, we want to allow our "user
+ * space" model to handle this.
+ *
+ * Note that this code needs to run on the current CPU.
+ */
+void copy_to_user_page(struct vm_area_struct *vma, struct page *page,
+		       unsigned long uaddr, void *dst, const void *src,
+		       unsigned long len)
+{
+#ifdef CONFIG_SMP
+	preempt_disable();
 #endif
+	memcpy(dst, src, len);
+	flush_ptrace_access(vma, page, uaddr, dst, len);
+#ifdef CONFIG_SMP
+	preempt_enable();
+#endif
+}
 
 void __flush_dcache_page(struct address_space *mapping, struct page *page)
 {
-- 
cgit v1.2.2


From c26c20b823d48addbde9cb5709d80655c6fadf18 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 18 Dec 2009 16:21:35 +0000
Subject: ARM: make_coherent: split adjust_pte() in two

adjust_pte() walks the page tables, and do_adjust_pte() does the
page table manipulation.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/fault-armv.c | 52 +++++++++++++++++++++++++++++-------------------
 1 file changed, 32 insertions(+), 20 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 56ee15321b00..074e6bb54eb3 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -36,28 +36,12 @@ static unsigned long shared_pte_mask = L_PTE_MT_BUFFERABLE;
  * Therefore those configurations which might call adjust_pte (those
  * without CONFIG_CPU_CACHE_VIPT) cannot support split page_table_lock.
  */
-static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
+static int do_adjust_pte(struct vm_area_struct *vma, unsigned long address,
+	pte_t *ptep)
 {
-	pgd_t *pgd;
-	pmd_t *pmd;
-	pte_t *pte, entry;
+	pte_t entry = *ptep;
 	int ret;
 
-	pgd = pgd_offset(vma->vm_mm, address);
-	if (pgd_none(*pgd))
-		goto no_pgd;
-	if (pgd_bad(*pgd))
-		goto bad_pgd;
-
-	pmd = pmd_offset(pgd, address);
-	if (pmd_none(*pmd))
-		goto no_pmd;
-	if (pmd_bad(*pmd))
-		goto bad_pmd;
-
-	pte = pte_offset_map(pmd, address);
-	entry = *pte;
-
 	/*
 	 * If this page is present, it's actually being shared.
 	 */
@@ -74,10 +58,38 @@ static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
 				  (pfn << PAGE_SHIFT) + PAGE_SIZE);
 		pte_val(entry) &= ~L_PTE_MT_MASK;
 		pte_val(entry) |= shared_pte_mask;
-		set_pte_at(vma->vm_mm, address, pte, entry);
+		set_pte_at(vma->vm_mm, address, ptep, entry);
 		flush_tlb_page(vma, address);
 	}
+
+	return ret;
+}
+
+static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
+{
+	pgd_t *pgd;
+	pmd_t *pmd;
+	pte_t *pte;
+	int ret;
+
+	pgd = pgd_offset(vma->vm_mm, address);
+	if (pgd_none(*pgd))
+		goto no_pgd;
+	if (pgd_bad(*pgd))
+		goto bad_pgd;
+
+	pmd = pmd_offset(pgd, address);
+	if (pmd_none(*pmd))
+		goto no_pmd;
+	if (pmd_bad(*pmd))
+		goto bad_pmd;
+
+	pte = pte_offset_map(pmd, address);
+
+	ret = do_adjust_pte(vma, address, pte);
+
 	pte_unmap(pte);
+
 	return ret;
 
 bad_pgd:
-- 
cgit v1.2.2


From f8a85f1164a33e3eb5b421b137ced793ed53ee33 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 18 Dec 2009 16:23:44 +0000
Subject: ARM: make_coherent: convert adjust_pte() to use
 p*d_none_or_clear_bad()

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/fault-armv.c | 24 ++++--------------------
 1 file changed, 4 insertions(+), 20 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 074e6bb54eb3..7a8efe1b37d8 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -73,16 +73,12 @@ static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
 	int ret;
 
 	pgd = pgd_offset(vma->vm_mm, address);
-	if (pgd_none(*pgd))
-		goto no_pgd;
-	if (pgd_bad(*pgd))
-		goto bad_pgd;
+	if (pgd_none_or_clear_bad(pgd))
+		return 0;
 
 	pmd = pmd_offset(pgd, address);
-	if (pmd_none(*pmd))
-		goto no_pmd;
-	if (pmd_bad(*pmd))
-		goto bad_pmd;
+	if (pmd_none_or_clear_bad(pmd))
+		return 0;
 
 	pte = pte_offset_map(pmd, address);
 
@@ -91,18 +87,6 @@ static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
 	pte_unmap(pte);
 
 	return ret;
-
-bad_pgd:
-	pgd_ERROR(*pgd);
-	pgd_clear(pgd);
-no_pgd:
-	return 0;
-
-bad_pmd:
-	pmd_ERROR(*pmd);
-	pmd_clear(pmd);
-no_pmd:
-	return 0;
 }
 
 static void
-- 
cgit v1.2.2


From 56dd47098abe1fdde598a8d8b7c04d775506f456 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 18 Dec 2009 16:24:34 +0000
Subject: ARM: make_coherent: fix problems with highpte, part 1

update_mmu_cache() is called with a page table already mapped.  We
call make_coherent(), which then calls adjust_pte() which wants to
map other page tables.  This causes kmap_atomic() to BUG() because
the slot its trying to use is already taken.

Since do_adjust_pte() modifies the page tables, we are also missing
any form of locking, so we're risking corrupting the page tables.

Fix this by using pte_offset_map_nested(), and taking the pte page
table lock around do_adjust_pte().

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/fault-armv.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 7a8efe1b37d8..8e9bc517132e 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -67,6 +67,7 @@ static int do_adjust_pte(struct vm_area_struct *vma, unsigned long address,
 
 static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
 {
+	spinlock_t *ptl;
 	pgd_t *pgd;
 	pmd_t *pmd;
 	pte_t *pte;
@@ -80,11 +81,19 @@ static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
 	if (pmd_none_or_clear_bad(pmd))
 		return 0;
 
-	pte = pte_offset_map(pmd, address);
+	/*
+	 * This is called while another page table is mapped, so we
+	 * must use the nested version.  This also means we need to
+	 * open-code the spin-locking.
+	 */
+	ptl = pte_lockptr(vma->vm_mm, pmd);
+	pte = pte_offset_map_nested(pmd, address);
+	spin_lock(ptl);
 
 	ret = do_adjust_pte(vma, address, pte);
 
-	pte_unmap(pte);
+	spin_unlock(ptl);
+	pte_unmap_nested(pte);
 
 	return ret;
 }
-- 
cgit v1.2.2


From ed42acaef1a9d51631a31b55e9ed52d400430492 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 18 Dec 2009 16:31:38 +0000
Subject: ARM: make_coherent: avoid recalculating the pfn for the modified page

We already know the pfn for the page to be modified in make_coherent,
so let's stop recalculating it unnecessarily.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/fault-armv.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 8e9bc517132e..ae88f2c3a6df 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -37,7 +37,7 @@ static unsigned long shared_pte_mask = L_PTE_MT_BUFFERABLE;
  * without CONFIG_CPU_CACHE_VIPT) cannot support split page_table_lock.
  */
 static int do_adjust_pte(struct vm_area_struct *vma, unsigned long address,
-	pte_t *ptep)
+	unsigned long pfn, pte_t *ptep)
 {
 	pte_t entry = *ptep;
 	int ret;
@@ -52,7 +52,6 @@ static int do_adjust_pte(struct vm_area_struct *vma, unsigned long address,
 	 * fault (ie, is old), we can safely ignore any issues.
 	 */
 	if (ret && (pte_val(entry) & L_PTE_MT_MASK) != shared_pte_mask) {
-		unsigned long pfn = pte_pfn(entry);
 		flush_cache_page(vma, address, pfn);
 		outer_flush_range((pfn << PAGE_SHIFT),
 				  (pfn << PAGE_SHIFT) + PAGE_SIZE);
@@ -65,7 +64,8 @@ static int do_adjust_pte(struct vm_area_struct *vma, unsigned long address,
 	return ret;
 }
 
-static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
+static int adjust_pte(struct vm_area_struct *vma, unsigned long address,
+	unsigned long pfn)
 {
 	spinlock_t *ptl;
 	pgd_t *pgd;
@@ -90,7 +90,7 @@ static int adjust_pte(struct vm_area_struct *vma, unsigned long address)
 	pte = pte_offset_map_nested(pmd, address);
 	spin_lock(ptl);
 
-	ret = do_adjust_pte(vma, address, pte);
+	ret = do_adjust_pte(vma, address, pfn, pte);
 
 	spin_unlock(ptl);
 	pte_unmap_nested(pte);
@@ -127,11 +127,11 @@ make_coherent(struct address_space *mapping, struct vm_area_struct *vma, unsigne
 		if (!(mpnt->vm_flags & VM_MAYSHARE))
 			continue;
 		offset = (pgoff - mpnt->vm_pgoff) << PAGE_SHIFT;
-		aliases += adjust_pte(mpnt, mpnt->vm_start + offset);
+		aliases += adjust_pte(mpnt, mpnt->vm_start + offset, pfn);
 	}
 	flush_dcache_mmap_unlock(mapping);
 	if (aliases)
-		adjust_pte(vma, addr);
+		adjust_pte(vma, addr, pfn);
 	else
 		flush_cache_page(vma, addr, pfn);
 }
-- 
cgit v1.2.2


From 7ada189f5c8627662c23f49b3e68463f86fc511e Mon Sep 17 00:00:00 2001
From: Jamie Iles <jamie.iles@picochip.com>
Date: Tue, 2 Feb 2010 20:24:58 +0100
Subject: ARM: 5900/2: arm: enable support for software perf events

The perf events subsystem allows counting of both hardware and
software events. This patch implements the bare minimum for software
performance events.

Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Jamie Iles <jamie.iles@picochip.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/fault.c | 7 +++++++
 1 file changed, 7 insertions(+)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index 10e06801afb3..9d40c341e07e 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -18,6 +18,7 @@
 #include <linux/page-flags.h>
 #include <linux/sched.h>
 #include <linux/highmem.h>
+#include <linux/perf_event.h>
 
 #include <asm/system.h>
 #include <asm/pgtable.h>
@@ -302,6 +303,12 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 	fault = __do_page_fault(mm, addr, fsr, tsk);
 	up_read(&mm->mmap_sem);
 
+	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, 0, regs, addr);
+	if (fault & VM_FAULT_MAJOR)
+		perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MAJ, 1, 0, regs, addr);
+	else if (fault & VM_FAULT_MINOR)
+		perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS_MIN, 1, 0, regs, addr);
+
 	/*
 	 * Handle the "normal" case first - VM_FAULT_MAJOR / VM_FAULT_MINOR
 	 */
-- 
cgit v1.2.2


From 18eabe2347ae7a11b3db768695913724166dfb0e Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sat, 31 Oct 2009 16:52:16 +0000
Subject: ARM: dma-mapping: introduce the idea of buffer ownership

The DMA API has the notion of buffer ownership; make it explicit in the
ARM implementation of this API.  This gives us a set of hooks to allow
us to deal with CPU cache issues arising from non-cache coherent DMA.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-By: Santosh Shilimkar <santosh.shilimkar@ti.com>
Tested-By: Jamie Iles <jamie@jamieiles.com>
---
 arch/arm/mm/dma-mapping.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 26325cb5d368..a316c9459526 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -573,8 +573,12 @@ void dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg,
 	int i;
 
 	for_each_sg(sg, s, nents, i) {
-		dmabounce_sync_for_cpu(dev, sg_dma_address(s), 0,
-					sg_dma_len(s), dir);
+		if (!dmabounce_sync_for_cpu(dev, sg_dma_address(s), 0,
+					    sg_dma_len(s), dir))
+			continue;
+
+		__dma_page_dev_to_cpu(sg_page(s), s->offset,
+				      s->length, dir);
 	}
 }
 EXPORT_SYMBOL(dma_sync_sg_for_cpu);
@@ -597,9 +601,8 @@ void dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg,
 					sg_dma_len(s), dir))
 			continue;
 
-		if (!arch_is_coherent())
-			dma_cache_maint_page(sg_page(s), s->offset,
-					     s->length, dir);
+		__dma_page_cpu_to_dev(sg_page(s), s->offset,
+				      s->length, dir);
 	}
 }
 EXPORT_SYMBOL(dma_sync_sg_for_device);
-- 
cgit v1.2.2


From 4ea0d7371e808628d11154b0d44140b70f05b998 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Tue, 24 Nov 2009 16:27:17 +0000
Subject: ARM: dma-mapping: push buffer ownership down into dma-mapping.c

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-By: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
 arch/arm/mm/dma-mapping.c | 34 ++++++++++++++++++++++++++++++----
 1 file changed, 30 insertions(+), 4 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index a316c9459526..bbf87880b915 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -404,7 +404,7 @@ EXPORT_SYMBOL(dma_free_coherent);
  * platforms with CONFIG_DMABOUNCE.
  * Use the driver DMA support - see dma-mapping.h (dma_sync_*)
  */
-void dma_cache_maint(const void *start, size_t size, int direction)
+static void dma_cache_maint(const void *start, size_t size, int direction)
 {
 	void (*inner_op)(const void *, const void *);
 	void (*outer_op)(unsigned long, unsigned long);
@@ -431,7 +431,20 @@ void dma_cache_maint(const void *start, size_t size, int direction)
 	inner_op(start, start + size);
 	outer_op(__pa(start), __pa(start) + size);
 }
-EXPORT_SYMBOL(dma_cache_maint);
+
+void ___dma_single_cpu_to_dev(const void *kaddr, size_t size,
+	enum dma_data_direction dir)
+{
+	dma_cache_maint(kaddr, size, dir);
+}
+EXPORT_SYMBOL(___dma_single_cpu_to_dev);
+
+void ___dma_single_dev_to_cpu(const void *kaddr, size_t size,
+	enum dma_data_direction dir)
+{
+	/* nothing to do */
+}
+EXPORT_SYMBOL(___dma_single_dev_to_cpu);
 
 static void dma_cache_maint_contiguous(struct page *page, unsigned long offset,
 				       size_t size, int direction)
@@ -474,7 +487,7 @@ static void dma_cache_maint_contiguous(struct page *page, unsigned long offset,
 	outer_op(paddr, paddr + size);
 }
 
-void dma_cache_maint_page(struct page *page, unsigned long offset,
+static void dma_cache_maint_page(struct page *page, unsigned long offset,
 			  size_t size, int dir)
 {
 	/*
@@ -499,7 +512,20 @@ void dma_cache_maint_page(struct page *page, unsigned long offset,
 		left -= len;
 	} while (left);
 }
-EXPORT_SYMBOL(dma_cache_maint_page);
+
+void ___dma_page_cpu_to_dev(struct page *page, unsigned long off,
+	size_t size, enum dma_data_direction dir)
+{
+	dma_cache_maint_page(page, off, size, dir);
+}
+EXPORT_SYMBOL(___dma_page_cpu_to_dev);
+
+void ___dma_page_dev_to_cpu(struct page *page, unsigned long off,
+	size_t size, enum dma_data_direction dir)
+{
+	/* nothing to do */
+}
+EXPORT_SYMBOL(___dma_page_dev_to_cpu);
 
 /**
  * dma_map_sg - map a set of SG buffers for streaming mode DMA
-- 
cgit v1.2.2


From 65af191a0414d0e1145f67c153e1b63d122dfbb4 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Tue, 24 Nov 2009 17:53:33 +0000
Subject: ARM: dma-mapping: move selection of page ops out of
 dma_cache_maint_contiguous

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-By: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
 arch/arm/mm/dma-mapping.c | 59 ++++++++++++++++++++++++-----------------------
 1 file changed, 30 insertions(+), 29 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index bbf87880b915..77dc483e64c1 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -447,48 +447,25 @@ void ___dma_single_dev_to_cpu(const void *kaddr, size_t size,
 EXPORT_SYMBOL(___dma_single_dev_to_cpu);
 
 static void dma_cache_maint_contiguous(struct page *page, unsigned long offset,
-				       size_t size, int direction)
+		       size_t size, void (*op)(const void *, const void *))
 {
 	void *vaddr;
-	unsigned long paddr;
-	void (*inner_op)(const void *, const void *);
-	void (*outer_op)(unsigned long, unsigned long);
-
-	switch (direction) {
-	case DMA_FROM_DEVICE:		/* invalidate only */
-		inner_op = dmac_inv_range;
-		outer_op = outer_inv_range;
-		break;
-	case DMA_TO_DEVICE:		/* writeback only */
-		inner_op = dmac_clean_range;
-		outer_op = outer_clean_range;
-		break;
-	case DMA_BIDIRECTIONAL:		/* writeback and invalidate */
-		inner_op = dmac_flush_range;
-		outer_op = outer_flush_range;
-		break;
-	default:
-		BUG();
-	}
 
 	if (!PageHighMem(page)) {
 		vaddr = page_address(page) + offset;
-		inner_op(vaddr, vaddr + size);
+		op(vaddr, vaddr + size);
 	} else {
 		vaddr = kmap_high_get(page);
 		if (vaddr) {
 			vaddr += offset;
-			inner_op(vaddr, vaddr + size);
+			op(vaddr, vaddr + size);
 			kunmap_high(page);
 		}
 	}
-
-	paddr = page_to_phys(page) + offset;
-	outer_op(paddr, paddr + size);
 }
 
 static void dma_cache_maint_page(struct page *page, unsigned long offset,
-			  size_t size, int dir)
+	size_t size, void (*op)(const void *, const void *))
 {
 	/*
 	 * A single sg entry may refer to multiple physically contiguous
@@ -506,7 +483,7 @@ static void dma_cache_maint_page(struct page *page, unsigned long offset,
 			}
 			len = PAGE_SIZE - offset;
 		}
-		dma_cache_maint_contiguous(page, offset, len, dir);
+		dma_cache_maint_contiguous(page, offset, len, op);
 		offset = 0;
 		page++;
 		left -= len;
@@ -516,7 +493,31 @@ static void dma_cache_maint_page(struct page *page, unsigned long offset,
 void ___dma_page_cpu_to_dev(struct page *page, unsigned long off,
 	size_t size, enum dma_data_direction dir)
 {
-	dma_cache_maint_page(page, off, size, dir);
+	unsigned long paddr;
+	void (*inner_op)(const void *, const void *);
+	void (*outer_op)(unsigned long, unsigned long);
+
+	switch (direction) {
+	case DMA_FROM_DEVICE:		/* invalidate only */
+		inner_op = dmac_inv_range;
+		outer_op = outer_inv_range;
+		break;
+	case DMA_TO_DEVICE:		/* writeback only */
+		inner_op = dmac_clean_range;
+		outer_op = outer_clean_range;
+		break;
+	case DMA_BIDIRECTIONAL:		/* writeback and invalidate */
+		inner_op = dmac_flush_range;
+		outer_op = outer_flush_range;
+		break;
+	default:
+		BUG();
+	}
+
+	dma_cache_maint_page(page, off, size, inner_op);
+
+	paddr = page_to_phys(page) + off;
+	outer_op(paddr, paddr + size);
 }
 EXPORT_SYMBOL(___dma_page_cpu_to_dev);
 
-- 
cgit v1.2.2


From 93f1d629e22b08642eb713ad96ac2cb9ade0641c Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Tue, 24 Nov 2009 14:41:01 +0000
Subject: ARM: dma-mapping: simplify dma_cache_maint_page

dma_cache_maint_contiguous is now simple enough to live inside
dma_cache_maint_page, so move it there.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-By: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
 arch/arm/mm/dma-mapping.c | 42 ++++++++++++++++++------------------------
 1 file changed, 18 insertions(+), 24 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 77dc483e64c1..0d68d2c83cda 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -446,24 +446,6 @@ void ___dma_single_dev_to_cpu(const void *kaddr, size_t size,
 }
 EXPORT_SYMBOL(___dma_single_dev_to_cpu);
 
-static void dma_cache_maint_contiguous(struct page *page, unsigned long offset,
-		       size_t size, void (*op)(const void *, const void *))
-{
-	void *vaddr;
-
-	if (!PageHighMem(page)) {
-		vaddr = page_address(page) + offset;
-		op(vaddr, vaddr + size);
-	} else {
-		vaddr = kmap_high_get(page);
-		if (vaddr) {
-			vaddr += offset;
-			op(vaddr, vaddr + size);
-			kunmap_high(page);
-		}
-	}
-}
-
 static void dma_cache_maint_page(struct page *page, unsigned long offset,
 	size_t size, void (*op)(const void *, const void *))
 {
@@ -476,14 +458,26 @@ static void dma_cache_maint_page(struct page *page, unsigned long offset,
 	size_t left = size;
 	do {
 		size_t len = left;
-		if (PageHighMem(page) && len + offset > PAGE_SIZE) {
-			if (offset >= PAGE_SIZE) {
-				page += offset / PAGE_SIZE;
-				offset %= PAGE_SIZE;
+		void *vaddr;
+
+		if (PageHighMem(page)) {
+			if (len + offset > PAGE_SIZE) {
+				if (offset >= PAGE_SIZE) {
+					page += offset / PAGE_SIZE;
+					offset %= PAGE_SIZE;
+				}
+				len = PAGE_SIZE - offset;
+			}
+			vaddr = kmap_high_get(page);
+			if (vaddr) {
+				vaddr += offset;
+				op(vaddr, vaddr + len);
+				kunmap_high(page);
 			}
-			len = PAGE_SIZE - offset;
+		} else {
+			vaddr = page_address(page) + offset;
+			op(vaddr, vaddr + len);
 		}
-		dma_cache_maint_contiguous(page, offset, len, op);
 		offset = 0;
 		page++;
 		left -= len;
-- 
cgit v1.2.2


From a9c9147eb9b1dba0ce567a41897c7773b4d1b0bc Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 26 Nov 2009 16:19:58 +0000
Subject: ARM: dma-mapping: provide per-cpu type map/unmap functions

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-By: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
 arch/arm/mm/cache-fa.S      | 26 ++++++++++++++++++++++++++
 arch/arm/mm/cache-v3.S      | 24 ++++++++++++++++++++++++
 arch/arm/mm/cache-v4.S      | 24 ++++++++++++++++++++++++
 arch/arm/mm/cache-v4wb.S    | 26 ++++++++++++++++++++++++++
 arch/arm/mm/cache-v4wt.S    | 25 +++++++++++++++++++++++++
 arch/arm/mm/cache-v6.S      | 26 ++++++++++++++++++++++++++
 arch/arm/mm/cache-v7.S      | 26 ++++++++++++++++++++++++++
 arch/arm/mm/dma-mapping.c   | 29 ++++++++++++-----------------
 arch/arm/mm/proc-arm1020.S  | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm1020e.S | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm1022.S  | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm1026.S  | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm920.S   | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm922.S   | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm925.S   | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm926.S   | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm940.S   | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-arm946.S   | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-feroceon.S | 42 ++++++++++++++++++++++++++++++++++++++++++
 arch/arm/mm/proc-mohawk.S   | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-xsc3.S     | 26 ++++++++++++++++++++++++++
 arch/arm/mm/proc-xscale.S   | 41 +++++++++++++++++++++++++++++++++++++++++
 22 files changed, 584 insertions(+), 17 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/cache-fa.S b/arch/arm/mm/cache-fa.S
index a89444a3c016..8ebffdd6fcff 100644
--- a/arch/arm/mm/cache-fa.S
+++ b/arch/arm/mm/cache-fa.S
@@ -205,6 +205,30 @@ ENTRY(fa_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain write buffer
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(fa_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	fa_dma_clean_range
+	bcs	fa_dma_inv_range
+	b	fa_dma_flush_range
+ENDPROC(fa_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(fa_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(fa_dma_unmap_area)
+
 	__INITDATA
 
 	.type	fa_cache_fns, #object
@@ -215,6 +239,8 @@ ENTRY(fa_cache_fns)
 	.long	fa_coherent_kern_range
 	.long	fa_coherent_user_range
 	.long	fa_flush_kern_dcache_area
+	.long	fa_dma_map_area
+	.long	fa_dma_unmap_area
 	.long	fa_dma_inv_range
 	.long	fa_dma_clean_range
 	.long	fa_dma_flush_range
diff --git a/arch/arm/mm/cache-v3.S b/arch/arm/mm/cache-v3.S
index 2a482731ea36..6df52dc014be 100644
--- a/arch/arm/mm/cache-v3.S
+++ b/arch/arm/mm/cache-v3.S
@@ -121,6 +121,28 @@ ENTRY(v3_dma_flush_range)
 ENTRY(v3_dma_clean_range)
 	mov	pc, lr
 
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v3_dma_unmap_area)
+	teq	r2, #DMA_TO_DEVICE
+	bne	v3_dma_inv_range
+	/* FALLTHROUGH */
+
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v3_dma_map_area)
+	mov	pc, lr
+ENDPROC(v3_dma_unmap_area)
+ENDPROC(v3_dma_map_area)
+
 	__INITDATA
 
 	.type	v3_cache_fns, #object
@@ -131,6 +153,8 @@ ENTRY(v3_cache_fns)
 	.long	v3_coherent_kern_range
 	.long	v3_coherent_user_range
 	.long	v3_flush_kern_dcache_area
+	.long	v3_dma_map_area
+	.long	v3_dma_unmap_area
 	.long	v3_dma_inv_range
 	.long	v3_dma_clean_range
 	.long	v3_dma_flush_range
diff --git a/arch/arm/mm/cache-v4.S b/arch/arm/mm/cache-v4.S
index 5c7da3e372e9..df3b423713b9 100644
--- a/arch/arm/mm/cache-v4.S
+++ b/arch/arm/mm/cache-v4.S
@@ -133,6 +133,28 @@ ENTRY(v4_dma_flush_range)
 ENTRY(v4_dma_clean_range)
 	mov	pc, lr
 
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v4_dma_unmap_area)
+	teq	r2, #DMA_TO_DEVICE
+	bne	v4_dma_inv_range
+	/* FALLTHROUGH */
+
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v4_dma_map_area)
+	mov	pc, lr
+ENDPROC(v4_dma_unmap_area)
+ENDPROC(v4_dma_map_area)
+
 	__INITDATA
 
 	.type	v4_cache_fns, #object
@@ -143,6 +165,8 @@ ENTRY(v4_cache_fns)
 	.long	v4_coherent_kern_range
 	.long	v4_coherent_user_range
 	.long	v4_flush_kern_dcache_area
+	.long	v4_dma_map_area
+	.long	v4_dma_unmap_area
 	.long	v4_dma_inv_range
 	.long	v4_dma_clean_range
 	.long	v4_dma_flush_range
diff --git a/arch/arm/mm/cache-v4wb.S b/arch/arm/mm/cache-v4wb.S
index 3dbedf1ec0e7..32e7a7448496 100644
--- a/arch/arm/mm/cache-v4wb.S
+++ b/arch/arm/mm/cache-v4wb.S
@@ -216,6 +216,30 @@ ENTRY(v4wb_dma_clean_range)
 	.globl	v4wb_dma_flush_range
 	.set	v4wb_dma_flush_range, v4wb_coherent_kern_range
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v4wb_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	v4wb_dma_clean_range
+	bcs	v4wb_dma_inv_range
+	b	v4wb_dma_flush_range
+ENDPROC(v4wb_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v4wb_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(v4wb_dma_unmap_area)
+
 	__INITDATA
 
 	.type	v4wb_cache_fns, #object
@@ -226,6 +250,8 @@ ENTRY(v4wb_cache_fns)
 	.long	v4wb_coherent_kern_range
 	.long	v4wb_coherent_user_range
 	.long	v4wb_flush_kern_dcache_area
+	.long	v4wb_dma_map_area
+	.long	v4wb_dma_unmap_area
 	.long	v4wb_dma_inv_range
 	.long	v4wb_dma_clean_range
 	.long	v4wb_dma_flush_range
diff --git a/arch/arm/mm/cache-v4wt.S b/arch/arm/mm/cache-v4wt.S
index b3b7410270b4..3d8dad5b2650 100644
--- a/arch/arm/mm/cache-v4wt.S
+++ b/arch/arm/mm/cache-v4wt.S
@@ -172,6 +172,29 @@ ENTRY(v4wt_dma_clean_range)
 	.globl	v4wt_dma_flush_range
 	.equ	v4wt_dma_flush_range, v4wt_dma_inv_range
 
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v4wt_dma_unmap_area)
+	add	r1, r1, r0
+	teq	r2, #DMA_TO_DEVICE
+	bne	v4wt_dma_inv_range
+	/* FALLTHROUGH */
+
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v4wt_dma_map_area)
+	mov	pc, lr
+ENDPROC(v4wt_dma_unmap_area)
+ENDPROC(v4wt_dma_map_area)
+
 	__INITDATA
 
 	.type	v4wt_cache_fns, #object
@@ -182,6 +205,8 @@ ENTRY(v4wt_cache_fns)
 	.long	v4wt_coherent_kern_range
 	.long	v4wt_coherent_user_range
 	.long	v4wt_flush_kern_dcache_area
+	.long	v4wt_dma_map_area
+	.long	v4wt_dma_unmap_area
 	.long	v4wt_dma_inv_range
 	.long	v4wt_dma_clean_range
 	.long	v4wt_dma_flush_range
diff --git a/arch/arm/mm/cache-v6.S b/arch/arm/mm/cache-v6.S
index 4ba0a24ce6f5..6f926dd0e0f7 100644
--- a/arch/arm/mm/cache-v6.S
+++ b/arch/arm/mm/cache-v6.S
@@ -263,6 +263,30 @@ ENTRY(v6_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain write buffer
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v6_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	v6_dma_clean_range
+	bcs	v6_dma_inv_range
+	b	v6_dma_flush_range
+ENDPROC(v6_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v6_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(v6_dma_unmap_area)
+
 	__INITDATA
 
 	.type	v6_cache_fns, #object
@@ -273,6 +297,8 @@ ENTRY(v6_cache_fns)
 	.long	v6_coherent_kern_range
 	.long	v6_coherent_user_range
 	.long	v6_flush_kern_dcache_area
+	.long	v6_dma_map_area
+	.long	v6_dma_unmap_area
 	.long	v6_dma_inv_range
 	.long	v6_dma_clean_range
 	.long	v6_dma_flush_range
diff --git a/arch/arm/mm/cache-v7.S b/arch/arm/mm/cache-v7.S
index 9073db849fb4..e30d8bc67182 100644
--- a/arch/arm/mm/cache-v7.S
+++ b/arch/arm/mm/cache-v7.S
@@ -271,6 +271,30 @@ ENTRY(v7_dma_flush_range)
 	mov	pc, lr
 ENDPROC(v7_dma_flush_range)
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v7_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	v7_dma_clean_range
+	bcs	v7_dma_inv_range
+	b	v7_dma_flush_range
+ENDPROC(v7_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(v7_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(v7_dma_unmap_area)
+
 	__INITDATA
 
 	.type	v7_cache_fns, #object
@@ -281,6 +305,8 @@ ENTRY(v7_cache_fns)
 	.long	v7_coherent_kern_range
 	.long	v7_coherent_user_range
 	.long	v7_flush_kern_dcache_area
+	.long	v7_dma_map_area
+	.long	v7_dma_unmap_area
 	.long	v7_dma_inv_range
 	.long	v7_dma_clean_range
 	.long	v7_dma_flush_range
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 0d68d2c83cda..efa8efa33f5e 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -406,35 +406,31 @@ EXPORT_SYMBOL(dma_free_coherent);
  */
 static void dma_cache_maint(const void *start, size_t size, int direction)
 {
-	void (*inner_op)(const void *, const void *);
 	void (*outer_op)(unsigned long, unsigned long);
 
-	BUG_ON(!virt_addr_valid(start) || !virt_addr_valid(start + size - 1));
-
 	switch (direction) {
 	case DMA_FROM_DEVICE:		/* invalidate only */
-		inner_op = dmac_inv_range;
 		outer_op = outer_inv_range;
 		break;
 	case DMA_TO_DEVICE:		/* writeback only */
-		inner_op = dmac_clean_range;
 		outer_op = outer_clean_range;
 		break;
 	case DMA_BIDIRECTIONAL:		/* writeback and invalidate */
-		inner_op = dmac_flush_range;
 		outer_op = outer_flush_range;
 		break;
 	default:
 		BUG();
 	}
 
-	inner_op(start, start + size);
 	outer_op(__pa(start), __pa(start) + size);
 }
 
 void ___dma_single_cpu_to_dev(const void *kaddr, size_t size,
 	enum dma_data_direction dir)
 {
+	BUG_ON(!virt_addr_valid(kaddr) || !virt_addr_valid(kaddr + size - 1));
+
+	dmac_map_area(kaddr, size, dir);
 	dma_cache_maint(kaddr, size, dir);
 }
 EXPORT_SYMBOL(___dma_single_cpu_to_dev);
@@ -442,12 +438,15 @@ EXPORT_SYMBOL(___dma_single_cpu_to_dev);
 void ___dma_single_dev_to_cpu(const void *kaddr, size_t size,
 	enum dma_data_direction dir)
 {
-	/* nothing to do */
+	BUG_ON(!virt_addr_valid(kaddr) || !virt_addr_valid(kaddr + size - 1));
+
+	dmac_unmap_area(kaddr, size, dir);
 }
 EXPORT_SYMBOL(___dma_single_dev_to_cpu);
 
 static void dma_cache_maint_page(struct page *page, unsigned long offset,
-	size_t size, void (*op)(const void *, const void *))
+	size_t size, enum dma_data_direction dir,
+	void (*op)(const void *, size_t, int))
 {
 	/*
 	 * A single sg entry may refer to multiple physically contiguous
@@ -471,12 +470,12 @@ static void dma_cache_maint_page(struct page *page, unsigned long offset,
 			vaddr = kmap_high_get(page);
 			if (vaddr) {
 				vaddr += offset;
-				op(vaddr, vaddr + len);
+				op(vaddr, len, dir);
 				kunmap_high(page);
 			}
 		} else {
 			vaddr = page_address(page) + offset;
-			op(vaddr, vaddr + len);
+			op(vaddr, len, dir);
 		}
 		offset = 0;
 		page++;
@@ -488,27 +487,23 @@ void ___dma_page_cpu_to_dev(struct page *page, unsigned long off,
 	size_t size, enum dma_data_direction dir)
 {
 	unsigned long paddr;
-	void (*inner_op)(const void *, const void *);
 	void (*outer_op)(unsigned long, unsigned long);
 
 	switch (direction) {
 	case DMA_FROM_DEVICE:		/* invalidate only */
-		inner_op = dmac_inv_range;
 		outer_op = outer_inv_range;
 		break;
 	case DMA_TO_DEVICE:		/* writeback only */
-		inner_op = dmac_clean_range;
 		outer_op = outer_clean_range;
 		break;
 	case DMA_BIDIRECTIONAL:		/* writeback and invalidate */
-		inner_op = dmac_flush_range;
 		outer_op = outer_flush_range;
 		break;
 	default:
 		BUG();
 	}
 
-	dma_cache_maint_page(page, off, size, inner_op);
+	dma_cache_maint_page(page, off, size, dir, dmac_map_area);
 
 	paddr = page_to_phys(page) + off;
 	outer_op(paddr, paddr + size);
@@ -518,7 +513,7 @@ EXPORT_SYMBOL(___dma_page_cpu_to_dev);
 void ___dma_page_dev_to_cpu(struct page *page, unsigned long off,
 	size_t size, enum dma_data_direction dir)
 {
-	/* nothing to do */
+	dma_cache_maint_page(page, off, size, dir, dmac_unmap_area);
 }
 EXPORT_SYMBOL(___dma_page_dev_to_cpu);
 
diff --git a/arch/arm/mm/proc-arm1020.S b/arch/arm/mm/proc-arm1020.S
index 8012e24282b2..c85f5eb42634 100644
--- a/arch/arm/mm/proc-arm1020.S
+++ b/arch/arm/mm/proc-arm1020.S
@@ -330,6 +330,30 @@ ENTRY(arm1020_dma_flush_range)
 	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm1020_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm1020_dma_clean_range
+	bcs	arm1020_dma_inv_range
+	b	arm1020_dma_flush_range
+ENDPROC(arm1020_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm1020_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm1020_dma_unmap_area)
+
 ENTRY(arm1020_cache_fns)
 	.long	arm1020_flush_kern_cache_all
 	.long	arm1020_flush_user_cache_all
@@ -337,6 +361,8 @@ ENTRY(arm1020_cache_fns)
 	.long	arm1020_coherent_kern_range
 	.long	arm1020_coherent_user_range
 	.long	arm1020_flush_kern_dcache_area
+	.long	arm1020_dma_map_area
+	.long	arm1020_dma_unmap_area
 	.long	arm1020_dma_inv_range
 	.long	arm1020_dma_clean_range
 	.long	arm1020_dma_flush_range
diff --git a/arch/arm/mm/proc-arm1020e.S b/arch/arm/mm/proc-arm1020e.S
index 41fe25d234f5..5a3cf7620a2c 100644
--- a/arch/arm/mm/proc-arm1020e.S
+++ b/arch/arm/mm/proc-arm1020e.S
@@ -316,6 +316,30 @@ ENTRY(arm1020e_dma_flush_range)
 	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm1020e_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm1020e_dma_clean_range
+	bcs	arm1020e_dma_inv_range
+	b	arm1020e_dma_flush_range
+ENDPROC(arm1020e_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm1020e_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm1020e_dma_unmap_area)
+
 ENTRY(arm1020e_cache_fns)
 	.long	arm1020e_flush_kern_cache_all
 	.long	arm1020e_flush_user_cache_all
@@ -323,6 +347,8 @@ ENTRY(arm1020e_cache_fns)
 	.long	arm1020e_coherent_kern_range
 	.long	arm1020e_coherent_user_range
 	.long	arm1020e_flush_kern_dcache_area
+	.long	arm1020e_dma_map_area
+	.long	arm1020e_dma_unmap_area
 	.long	arm1020e_dma_inv_range
 	.long	arm1020e_dma_clean_range
 	.long	arm1020e_dma_flush_range
diff --git a/arch/arm/mm/proc-arm1022.S b/arch/arm/mm/proc-arm1022.S
index 20a5b1b31a70..fec8f5878438 100644
--- a/arch/arm/mm/proc-arm1022.S
+++ b/arch/arm/mm/proc-arm1022.S
@@ -305,6 +305,30 @@ ENTRY(arm1022_dma_flush_range)
 	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm1022_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm1022_dma_clean_range
+	bcs	arm1022_dma_inv_range
+	b	arm1022_dma_flush_range
+ENDPROC(arm1022_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm1022_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm1022_dma_unmap_area)
+
 ENTRY(arm1022_cache_fns)
 	.long	arm1022_flush_kern_cache_all
 	.long	arm1022_flush_user_cache_all
@@ -312,6 +336,8 @@ ENTRY(arm1022_cache_fns)
 	.long	arm1022_coherent_kern_range
 	.long	arm1022_coherent_user_range
 	.long	arm1022_flush_kern_dcache_area
+	.long	arm1022_dma_map_area
+	.long	arm1022_dma_unmap_area
 	.long	arm1022_dma_inv_range
 	.long	arm1022_dma_clean_range
 	.long	arm1022_dma_flush_range
diff --git a/arch/arm/mm/proc-arm1026.S b/arch/arm/mm/proc-arm1026.S
index 96aedb10fcc4..9ece6f666497 100644
--- a/arch/arm/mm/proc-arm1026.S
+++ b/arch/arm/mm/proc-arm1026.S
@@ -299,6 +299,30 @@ ENTRY(arm1026_dma_flush_range)
 	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm1026_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm1026_dma_clean_range
+	bcs	arm1026_dma_inv_range
+	b	arm1026_dma_flush_range
+ENDPROC(arm1026_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm1026_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm1026_dma_unmap_area)
+
 ENTRY(arm1026_cache_fns)
 	.long	arm1026_flush_kern_cache_all
 	.long	arm1026_flush_user_cache_all
@@ -306,6 +330,8 @@ ENTRY(arm1026_cache_fns)
 	.long	arm1026_coherent_kern_range
 	.long	arm1026_coherent_user_range
 	.long	arm1026_flush_kern_dcache_area
+	.long	arm1026_dma_map_area
+	.long	arm1026_dma_unmap_area
 	.long	arm1026_dma_inv_range
 	.long	arm1026_dma_clean_range
 	.long	arm1026_dma_flush_range
diff --git a/arch/arm/mm/proc-arm920.S b/arch/arm/mm/proc-arm920.S
index 471669e2d7cb..6f6ab2747da6 100644
--- a/arch/arm/mm/proc-arm920.S
+++ b/arch/arm/mm/proc-arm920.S
@@ -288,6 +288,30 @@ ENTRY(arm920_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm920_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm920_dma_clean_range
+	bcs	arm920_dma_inv_range
+	b	arm920_dma_flush_range
+ENDPROC(arm920_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm920_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm920_dma_unmap_area)
+
 ENTRY(arm920_cache_fns)
 	.long	arm920_flush_kern_cache_all
 	.long	arm920_flush_user_cache_all
@@ -295,6 +319,8 @@ ENTRY(arm920_cache_fns)
 	.long	arm920_coherent_kern_range
 	.long	arm920_coherent_user_range
 	.long	arm920_flush_kern_dcache_area
+	.long	arm920_dma_map_area
+	.long	arm920_dma_unmap_area
 	.long	arm920_dma_inv_range
 	.long	arm920_dma_clean_range
 	.long	arm920_dma_flush_range
diff --git a/arch/arm/mm/proc-arm922.S b/arch/arm/mm/proc-arm922.S
index ee111b00fa41..4e4396b121ca 100644
--- a/arch/arm/mm/proc-arm922.S
+++ b/arch/arm/mm/proc-arm922.S
@@ -290,6 +290,30 @@ ENTRY(arm922_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm922_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm922_dma_clean_range
+	bcs	arm922_dma_inv_range
+	b	arm922_dma_flush_range
+ENDPROC(arm922_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm922_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm922_dma_unmap_area)
+
 ENTRY(arm922_cache_fns)
 	.long	arm922_flush_kern_cache_all
 	.long	arm922_flush_user_cache_all
@@ -297,6 +321,8 @@ ENTRY(arm922_cache_fns)
 	.long	arm922_coherent_kern_range
 	.long	arm922_coherent_user_range
 	.long	arm922_flush_kern_dcache_area
+	.long	arm922_dma_map_area
+	.long	arm922_dma_unmap_area
 	.long	arm922_dma_inv_range
 	.long	arm922_dma_clean_range
 	.long	arm922_dma_flush_range
diff --git a/arch/arm/mm/proc-arm925.S b/arch/arm/mm/proc-arm925.S
index 8deb5bde58e4..7c01c5d1108c 100644
--- a/arch/arm/mm/proc-arm925.S
+++ b/arch/arm/mm/proc-arm925.S
@@ -341,6 +341,30 @@ ENTRY(arm925_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm925_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm925_dma_clean_range
+	bcs	arm925_dma_inv_range
+	b	arm925_dma_flush_range
+ENDPROC(arm925_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm925_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm925_dma_unmap_area)
+
 ENTRY(arm925_cache_fns)
 	.long	arm925_flush_kern_cache_all
 	.long	arm925_flush_user_cache_all
@@ -348,6 +372,8 @@ ENTRY(arm925_cache_fns)
 	.long	arm925_coherent_kern_range
 	.long	arm925_coherent_user_range
 	.long	arm925_flush_kern_dcache_area
+	.long	arm925_dma_map_area
+	.long	arm925_dma_unmap_area
 	.long	arm925_dma_inv_range
 	.long	arm925_dma_clean_range
 	.long	arm925_dma_flush_range
diff --git a/arch/arm/mm/proc-arm926.S b/arch/arm/mm/proc-arm926.S
index 64db6e275a44..72a01a4b80ab 100644
--- a/arch/arm/mm/proc-arm926.S
+++ b/arch/arm/mm/proc-arm926.S
@@ -304,6 +304,30 @@ ENTRY(arm926_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm926_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm926_dma_clean_range
+	bcs	arm926_dma_inv_range
+	b	arm926_dma_flush_range
+ENDPROC(arm926_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm926_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm926_dma_unmap_area)
+
 ENTRY(arm926_cache_fns)
 	.long	arm926_flush_kern_cache_all
 	.long	arm926_flush_user_cache_all
@@ -311,6 +335,8 @@ ENTRY(arm926_cache_fns)
 	.long	arm926_coherent_kern_range
 	.long	arm926_coherent_user_range
 	.long	arm926_flush_kern_dcache_area
+	.long	arm926_dma_map_area
+	.long	arm926_dma_unmap_area
 	.long	arm926_dma_inv_range
 	.long	arm926_dma_clean_range
 	.long	arm926_dma_flush_range
diff --git a/arch/arm/mm/proc-arm940.S b/arch/arm/mm/proc-arm940.S
index 8196b9f401fb..6bb58fca7270 100644
--- a/arch/arm/mm/proc-arm940.S
+++ b/arch/arm/mm/proc-arm940.S
@@ -233,6 +233,30 @@ ENTRY(arm940_dma_flush_range)
 	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm940_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm940_dma_clean_range
+	bcs	arm940_dma_inv_range
+	b	arm940_dma_flush_range
+ENDPROC(arm940_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm940_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm940_dma_unmap_area)
+
 ENTRY(arm940_cache_fns)
 	.long	arm940_flush_kern_cache_all
 	.long	arm940_flush_user_cache_all
@@ -240,6 +264,8 @@ ENTRY(arm940_cache_fns)
 	.long	arm940_coherent_kern_range
 	.long	arm940_coherent_user_range
 	.long	arm940_flush_kern_dcache_area
+	.long	arm940_dma_map_area
+	.long	arm940_dma_unmap_area
 	.long	arm940_dma_inv_range
 	.long	arm940_dma_clean_range
 	.long	arm940_dma_flush_range
diff --git a/arch/arm/mm/proc-arm946.S b/arch/arm/mm/proc-arm946.S
index 9a951239c86c..ac0f9ba719d7 100644
--- a/arch/arm/mm/proc-arm946.S
+++ b/arch/arm/mm/proc-arm946.S
@@ -275,6 +275,30 @@ ENTRY(arm946_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm946_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	arm946_dma_clean_range
+	bcs	arm946_dma_inv_range
+	b	arm946_dma_flush_range
+ENDPROC(arm946_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(arm946_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(arm946_dma_unmap_area)
+
 ENTRY(arm946_cache_fns)
 	.long	arm946_flush_kern_cache_all
 	.long	arm946_flush_user_cache_all
@@ -282,6 +306,8 @@ ENTRY(arm946_cache_fns)
 	.long	arm946_coherent_kern_range
 	.long	arm946_coherent_user_range
 	.long	arm946_flush_kern_dcache_area
+	.long	arm946_dma_map_area
+	.long	arm946_dma_unmap_area
 	.long	arm946_dma_inv_range
 	.long	arm946_dma_clean_range
 	.long	arm946_dma_flush_range
diff --git a/arch/arm/mm/proc-feroceon.S b/arch/arm/mm/proc-feroceon.S
index dbc39383e66a..97e1d784f152 100644
--- a/arch/arm/mm/proc-feroceon.S
+++ b/arch/arm/mm/proc-feroceon.S
@@ -367,6 +367,44 @@ ENTRY(feroceon_range_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(feroceon_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	feroceon_dma_clean_range
+	bcs	feroceon_dma_inv_range
+	b	feroceon_dma_flush_range
+ENDPROC(feroceon_dma_map_area)
+
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(feroceon_range_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	feroceon_range_dma_clean_range
+	bcs	feroceon_range_dma_inv_range
+	b	feroceon_range_dma_flush_range
+ENDPROC(feroceon_range_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(feroceon_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(feroceon_dma_unmap_area)
+
 ENTRY(feroceon_cache_fns)
 	.long	feroceon_flush_kern_cache_all
 	.long	feroceon_flush_user_cache_all
@@ -374,6 +412,8 @@ ENTRY(feroceon_cache_fns)
 	.long	feroceon_coherent_kern_range
 	.long	feroceon_coherent_user_range
 	.long	feroceon_flush_kern_dcache_area
+	.long	feroceon_dma_map_area
+	.long	feroceon_dma_unmap_area
 	.long	feroceon_dma_inv_range
 	.long	feroceon_dma_clean_range
 	.long	feroceon_dma_flush_range
@@ -385,6 +425,8 @@ ENTRY(feroceon_range_cache_fns)
 	.long	feroceon_coherent_kern_range
 	.long	feroceon_coherent_user_range
 	.long	feroceon_range_flush_kern_dcache_area
+	.long	feroceon_range_dma_map_area
+	.long	feroceon_dma_unmap_area
 	.long	feroceon_range_dma_inv_range
 	.long	feroceon_range_dma_clean_range
 	.long	feroceon_range_dma_flush_range
diff --git a/arch/arm/mm/proc-mohawk.S b/arch/arm/mm/proc-mohawk.S
index 9674d36cc97d..55b7fbec6548 100644
--- a/arch/arm/mm/proc-mohawk.S
+++ b/arch/arm/mm/proc-mohawk.S
@@ -268,6 +268,30 @@ ENTRY(mohawk_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(mohawk_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	mohawk_dma_clean_range
+	bcs	mohawk_dma_inv_range
+	b	mohawk_dma_flush_range
+ENDPROC(mohawk_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(mohawk_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(mohawk_dma_unmap_area)
+
 ENTRY(mohawk_cache_fns)
 	.long	mohawk_flush_kern_cache_all
 	.long	mohawk_flush_user_cache_all
@@ -275,6 +299,8 @@ ENTRY(mohawk_cache_fns)
 	.long	mohawk_coherent_kern_range
 	.long	mohawk_coherent_user_range
 	.long	mohawk_flush_kern_dcache_area
+	.long	mohawk_dma_map_area
+	.long	mohawk_dma_unmap_area
 	.long	mohawk_dma_inv_range
 	.long	mohawk_dma_clean_range
 	.long	mohawk_dma_flush_range
diff --git a/arch/arm/mm/proc-xsc3.S b/arch/arm/mm/proc-xsc3.S
index 96456f548798..4e4ce889b3e6 100644
--- a/arch/arm/mm/proc-xsc3.S
+++ b/arch/arm/mm/proc-xsc3.S
@@ -304,6 +304,30 @@ ENTRY(xsc3_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ data write barrier
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(xsc3_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	xsc3_dma_clean_range
+	bcs	xsc3_dma_inv_range
+	b	xsc3_dma_flush_range
+ENDPROC(xsc3_dma_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(xsc3_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(xsc3_dma_unmap_area)
+
 ENTRY(xsc3_cache_fns)
 	.long	xsc3_flush_kern_cache_all
 	.long	xsc3_flush_user_cache_all
@@ -311,6 +335,8 @@ ENTRY(xsc3_cache_fns)
 	.long	xsc3_coherent_kern_range
 	.long	xsc3_coherent_user_range
 	.long	xsc3_flush_kern_dcache_area
+	.long	xsc3_dma_map_area
+	.long	xsc3_dma_unmap_area
 	.long	xsc3_dma_inv_range
 	.long	xsc3_dma_clean_range
 	.long	xsc3_dma_flush_range
diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S
index 93df47265f2d..a7999f94bf27 100644
--- a/arch/arm/mm/proc-xscale.S
+++ b/arch/arm/mm/proc-xscale.S
@@ -363,6 +363,43 @@ ENTRY(xscale_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ Drain Write (& Fill) Buffer
 	mov	pc, lr
 
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(xscale_dma_map_area)
+	add	r1, r1, r0
+	cmp	r2, #DMA_TO_DEVICE
+	beq	xscale_dma_clean_range
+	bcs	xscale_dma_inv_range
+	b	xscale_dma_flush_range
+ENDPROC(xscale_dma_map_area)
+
+/*
+ *	dma_map_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(xscale_dma_a0_map_area)
+	add	r1, r1, r0
+	teq	r2, #DMA_TO_DEVICE
+	beq	xscale_dma_clean_range
+	b	xscale_dma_flush_range
+ENDPROC(xscsale_dma_a0_map_area)
+
+/*
+ *	dma_unmap_area(start, size, dir)
+ *	- start	- kernel virtual start address
+ *	- size	- size of region
+ *	- dir	- DMA direction
+ */
+ENTRY(xscale_dma_unmap_area)
+	mov	pc, lr
+ENDPROC(xscale_dma_unmap_area)
+
 ENTRY(xscale_cache_fns)
 	.long	xscale_flush_kern_cache_all
 	.long	xscale_flush_user_cache_all
@@ -370,6 +407,8 @@ ENTRY(xscale_cache_fns)
 	.long	xscale_coherent_kern_range
 	.long	xscale_coherent_user_range
 	.long	xscale_flush_kern_dcache_area
+	.long	xscale_dma_map_area
+	.long	xscale_dma_unmap_area
 	.long	xscale_dma_inv_range
 	.long	xscale_dma_clean_range
 	.long	xscale_dma_flush_range
@@ -394,6 +433,8 @@ ENTRY(xscale_80200_A0_A1_cache_fns)
 	.long	xscale_coherent_kern_range
 	.long	xscale_coherent_user_range
 	.long	xscale_flush_kern_dcache_area
+	.long	xscale_dma_a0_map_area
+	.long	xscale_dma_unmap_area
 	.long	xscale_dma_flush_range
 	.long	xscale_dma_clean_range
 	.long	xscale_dma_flush_range
-- 
cgit v1.2.2


From 702b94bff3c50542a6e4ab9a4f4cef093262fe65 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Thu, 26 Nov 2009 16:24:19 +0000
Subject: ARM: dma-mapping: remove dmac_clean_range and dmac_inv_range

These are now unused, and so can be removed.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-By: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
 arch/arm/mm/cache-fa.S      |  6 ++----
 arch/arm/mm/cache-v3.S      | 29 +----------------------------
 arch/arm/mm/cache-v4.S      | 29 +----------------------------
 arch/arm/mm/cache-v4wb.S    |  6 ++----
 arch/arm/mm/cache-v4wt.S    | 15 +--------------
 arch/arm/mm/cache-v6.S      |  6 ++----
 arch/arm/mm/cache-v7.S      |  6 ++----
 arch/arm/mm/proc-arm1020.S  |  6 ++----
 arch/arm/mm/proc-arm1020e.S |  6 ++----
 arch/arm/mm/proc-arm1022.S  |  6 ++----
 arch/arm/mm/proc-arm1026.S  |  6 ++----
 arch/arm/mm/proc-arm920.S   |  6 ++----
 arch/arm/mm/proc-arm922.S   |  6 ++----
 arch/arm/mm/proc-arm925.S   |  6 ++----
 arch/arm/mm/proc-arm926.S   |  6 ++----
 arch/arm/mm/proc-arm940.S   |  6 ++----
 arch/arm/mm/proc-arm946.S   |  6 ++----
 arch/arm/mm/proc-feroceon.S | 12 ++++--------
 arch/arm/mm/proc-mohawk.S   |  6 ++----
 arch/arm/mm/proc-xsc3.S     |  6 ++----
 arch/arm/mm/proc-xscale.S   |  8 ++------
 21 files changed, 41 insertions(+), 148 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/cache-fa.S b/arch/arm/mm/cache-fa.S
index 8ebffdd6fcff..7148e53e6078 100644
--- a/arch/arm/mm/cache-fa.S
+++ b/arch/arm/mm/cache-fa.S
@@ -157,7 +157,7 @@ ENTRY(fa_flush_kern_dcache_area)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(fa_dma_inv_range)
+fa_dma_inv_range:
 	tst	r0, #CACHE_DLINESIZE - 1
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c14, 1		@ clean & invalidate D entry
@@ -180,7 +180,7 @@ ENTRY(fa_dma_inv_range)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(fa_dma_clean_range)
+fa_dma_clean_range:
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	add	r0, r0, #CACHE_DLINESIZE
@@ -241,7 +241,5 @@ ENTRY(fa_cache_fns)
 	.long	fa_flush_kern_dcache_area
 	.long	fa_dma_map_area
 	.long	fa_dma_unmap_area
-	.long	fa_dma_inv_range
-	.long	fa_dma_clean_range
 	.long	fa_dma_flush_range
 	.size	fa_cache_fns, . - fa_cache_fns
diff --git a/arch/arm/mm/cache-v3.S b/arch/arm/mm/cache-v3.S
index 6df52dc014be..c2ff3c599fee 100644
--- a/arch/arm/mm/cache-v3.S
+++ b/arch/arm/mm/cache-v3.S
@@ -83,20 +83,6 @@ ENTRY(v3_coherent_user_range)
 ENTRY(v3_flush_kern_dcache_area)
 	/* FALLTHROUGH */
 
-/*
- *	dma_inv_range(start, end)
- *
- *	Invalidate (discard) the specified virtual address range.
- *	May not write back any entries.  If 'start' or 'end'
- *	are not cache line aligned, those lines must be written
- *	back.
- *
- *	- start  - virtual start address
- *	- end	 - virtual end address
- */
-ENTRY(v3_dma_inv_range)
-	/* FALLTHROUGH */
-
 /*
  *	dma_flush_range(start, end)
  *
@@ -108,17 +94,6 @@ ENTRY(v3_dma_inv_range)
 ENTRY(v3_dma_flush_range)
 	mov	r0, #0
 	mcr	p15, 0, r0, c7, c0, 0		@ flush ID cache
-	/* FALLTHROUGH */
-
-/*
- *	dma_clean_range(start, end)
- *
- *	Clean (write back) the specified virtual address range.
- *
- *	- start  - virtual start address
- *	- end	 - virtual end address
- */
-ENTRY(v3_dma_clean_range)
 	mov	pc, lr
 
 /*
@@ -129,7 +104,7 @@ ENTRY(v3_dma_clean_range)
  */
 ENTRY(v3_dma_unmap_area)
 	teq	r2, #DMA_TO_DEVICE
-	bne	v3_dma_inv_range
+	bne	v3_dma_flush_range
 	/* FALLTHROUGH */
 
 /*
@@ -155,7 +130,5 @@ ENTRY(v3_cache_fns)
 	.long	v3_flush_kern_dcache_area
 	.long	v3_dma_map_area
 	.long	v3_dma_unmap_area
-	.long	v3_dma_inv_range
-	.long	v3_dma_clean_range
 	.long	v3_dma_flush_range
 	.size	v3_cache_fns, . - v3_cache_fns
diff --git a/arch/arm/mm/cache-v4.S b/arch/arm/mm/cache-v4.S
index df3b423713b9..4810f7e3e813 100644
--- a/arch/arm/mm/cache-v4.S
+++ b/arch/arm/mm/cache-v4.S
@@ -93,20 +93,6 @@ ENTRY(v4_coherent_user_range)
 ENTRY(v4_flush_kern_dcache_area)
 	/* FALLTHROUGH */
 
-/*
- *	dma_inv_range(start, end)
- *
- *	Invalidate (discard) the specified virtual address range.
- *	May not write back any entries.  If 'start' or 'end'
- *	are not cache line aligned, those lines must be written
- *	back.
- *
- *	- start  - virtual start address
- *	- end	 - virtual end address
- */
-ENTRY(v4_dma_inv_range)
-	/* FALLTHROUGH */
-
 /*
  *	dma_flush_range(start, end)
  *
@@ -120,17 +106,6 @@ ENTRY(v4_dma_flush_range)
 	mov	r0, #0
 	mcr	p15, 0, r0, c7, c7, 0		@ flush ID cache
 #endif
-	/* FALLTHROUGH */
-
-/*
- *	dma_clean_range(start, end)
- *
- *	Clean (write back) the specified virtual address range.
- *
- *	- start  - virtual start address
- *	- end	 - virtual end address
- */
-ENTRY(v4_dma_clean_range)
 	mov	pc, lr
 
 /*
@@ -141,7 +116,7 @@ ENTRY(v4_dma_clean_range)
  */
 ENTRY(v4_dma_unmap_area)
 	teq	r2, #DMA_TO_DEVICE
-	bne	v4_dma_inv_range
+	bne	v4_dma_flush_range
 	/* FALLTHROUGH */
 
 /*
@@ -167,7 +142,5 @@ ENTRY(v4_cache_fns)
 	.long	v4_flush_kern_dcache_area
 	.long	v4_dma_map_area
 	.long	v4_dma_unmap_area
-	.long	v4_dma_inv_range
-	.long	v4_dma_clean_range
 	.long	v4_dma_flush_range
 	.size	v4_cache_fns, . - v4_cache_fns
diff --git a/arch/arm/mm/cache-v4wb.S b/arch/arm/mm/cache-v4wb.S
index 32e7a7448496..df8368afa102 100644
--- a/arch/arm/mm/cache-v4wb.S
+++ b/arch/arm/mm/cache-v4wb.S
@@ -173,7 +173,7 @@ ENTRY(v4wb_coherent_user_range)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(v4wb_dma_inv_range)
+v4wb_dma_inv_range:
 	tst	r0, #CACHE_DLINESIZE - 1
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -194,7 +194,7 @@ ENTRY(v4wb_dma_inv_range)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(v4wb_dma_clean_range)
+v4wb_dma_clean_range:
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	add	r0, r0, #CACHE_DLINESIZE
@@ -252,7 +252,5 @@ ENTRY(v4wb_cache_fns)
 	.long	v4wb_flush_kern_dcache_area
 	.long	v4wb_dma_map_area
 	.long	v4wb_dma_unmap_area
-	.long	v4wb_dma_inv_range
-	.long	v4wb_dma_clean_range
 	.long	v4wb_dma_flush_range
 	.size	v4wb_cache_fns, . - v4wb_cache_fns
diff --git a/arch/arm/mm/cache-v4wt.S b/arch/arm/mm/cache-v4wt.S
index 3d8dad5b2650..45c70312f43b 100644
--- a/arch/arm/mm/cache-v4wt.S
+++ b/arch/arm/mm/cache-v4wt.S
@@ -142,23 +142,12 @@ ENTRY(v4wt_flush_kern_dcache_area)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(v4wt_dma_inv_range)
+v4wt_dma_inv_range:
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
 	add	r0, r0, #CACHE_DLINESIZE
 	cmp	r0, r1
 	blo	1b
-	/* FALLTHROUGH */
-
-/*
- *	dma_clean_range(start, end)
- *
- *	Clean the specified virtual address range.
- *
- *	- start  - virtual start address
- *	- end	 - virtual end address
- */
-ENTRY(v4wt_dma_clean_range)
 	mov	pc, lr
 
 /*
@@ -207,7 +196,5 @@ ENTRY(v4wt_cache_fns)
 	.long	v4wt_flush_kern_dcache_area
 	.long	v4wt_dma_map_area
 	.long	v4wt_dma_unmap_area
-	.long	v4wt_dma_inv_range
-	.long	v4wt_dma_clean_range
 	.long	v4wt_dma_flush_range
 	.size	v4wt_cache_fns, . - v4wt_cache_fns
diff --git a/arch/arm/mm/cache-v6.S b/arch/arm/mm/cache-v6.S
index 6f926dd0e0f7..a11934e53fbd 100644
--- a/arch/arm/mm/cache-v6.S
+++ b/arch/arm/mm/cache-v6.S
@@ -195,7 +195,7 @@ ENTRY(v6_flush_kern_dcache_area)
  *	- start   - virtual start address of region
  *	- end     - virtual end address of region
  */
-ENTRY(v6_dma_inv_range)
+v6_dma_inv_range:
 	tst	r0, #D_CACHE_LINE_SIZE - 1
 	bic	r0, r0, #D_CACHE_LINE_SIZE - 1
 #ifdef HARVARD_CACHE
@@ -228,7 +228,7 @@ ENTRY(v6_dma_inv_range)
  *	- start   - virtual start address of region
  *	- end     - virtual end address of region
  */
-ENTRY(v6_dma_clean_range)
+v6_dma_clean_range:
 	bic	r0, r0, #D_CACHE_LINE_SIZE - 1
 1:
 #ifdef HARVARD_CACHE
@@ -299,7 +299,5 @@ ENTRY(v6_cache_fns)
 	.long	v6_flush_kern_dcache_area
 	.long	v6_dma_map_area
 	.long	v6_dma_unmap_area
-	.long	v6_dma_inv_range
-	.long	v6_dma_clean_range
 	.long	v6_dma_flush_range
 	.size	v6_cache_fns, . - v6_cache_fns
diff --git a/arch/arm/mm/cache-v7.S b/arch/arm/mm/cache-v7.S
index e30d8bc67182..b1cd0fd91207 100644
--- a/arch/arm/mm/cache-v7.S
+++ b/arch/arm/mm/cache-v7.S
@@ -216,7 +216,7 @@ ENDPROC(v7_flush_kern_dcache_area)
  *	- start   - virtual start address of region
  *	- end     - virtual end address of region
  */
-ENTRY(v7_dma_inv_range)
+v7_dma_inv_range:
 	dcache_line_size r2, r3
 	sub	r3, r2, #1
 	tst	r0, r3
@@ -240,7 +240,7 @@ ENDPROC(v7_dma_inv_range)
  *	- start   - virtual start address of region
  *	- end     - virtual end address of region
  */
-ENTRY(v7_dma_clean_range)
+v7_dma_clean_range:
 	dcache_line_size r2, r3
 	sub	r3, r2, #1
 	bic	r0, r0, r3
@@ -307,7 +307,5 @@ ENTRY(v7_cache_fns)
 	.long	v7_flush_kern_dcache_area
 	.long	v7_dma_map_area
 	.long	v7_dma_unmap_area
-	.long	v7_dma_inv_range
-	.long	v7_dma_clean_range
 	.long	v7_dma_flush_range
 	.size	v7_cache_fns, . - v7_cache_fns
diff --git a/arch/arm/mm/proc-arm1020.S b/arch/arm/mm/proc-arm1020.S
index c85f5eb42634..72507c630ceb 100644
--- a/arch/arm/mm/proc-arm1020.S
+++ b/arch/arm/mm/proc-arm1020.S
@@ -265,7 +265,7 @@ ENTRY(arm1020_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(arm1020_dma_inv_range)
+arm1020_dma_inv_range:
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
 	tst	r0, #CACHE_DLINESIZE - 1
@@ -295,7 +295,7 @@ ENTRY(arm1020_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(arm1020_dma_clean_range)
+arm1020_dma_clean_range:
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
 	bic	r0, r0, #CACHE_DLINESIZE - 1
@@ -363,8 +363,6 @@ ENTRY(arm1020_cache_fns)
 	.long	arm1020_flush_kern_dcache_area
 	.long	arm1020_dma_map_area
 	.long	arm1020_dma_unmap_area
-	.long	arm1020_dma_inv_range
-	.long	arm1020_dma_clean_range
 	.long	arm1020_dma_flush_range
 
 	.align	5
diff --git a/arch/arm/mm/proc-arm1020e.S b/arch/arm/mm/proc-arm1020e.S
index 5a3cf7620a2c..d27829805609 100644
--- a/arch/arm/mm/proc-arm1020e.S
+++ b/arch/arm/mm/proc-arm1020e.S
@@ -258,7 +258,7 @@ ENTRY(arm1020e_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(arm1020e_dma_inv_range)
+arm1020e_dma_inv_range:
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
 	tst	r0, #CACHE_DLINESIZE - 1
@@ -284,7 +284,7 @@ ENTRY(arm1020e_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(arm1020e_dma_clean_range)
+arm1020e_dma_clean_range:
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
 	bic	r0, r0, #CACHE_DLINESIZE - 1
@@ -349,8 +349,6 @@ ENTRY(arm1020e_cache_fns)
 	.long	arm1020e_flush_kern_dcache_area
 	.long	arm1020e_dma_map_area
 	.long	arm1020e_dma_unmap_area
-	.long	arm1020e_dma_inv_range
-	.long	arm1020e_dma_clean_range
 	.long	arm1020e_dma_flush_range
 
 	.align	5
diff --git a/arch/arm/mm/proc-arm1022.S b/arch/arm/mm/proc-arm1022.S
index fec8f5878438..ce13e4a827de 100644
--- a/arch/arm/mm/proc-arm1022.S
+++ b/arch/arm/mm/proc-arm1022.S
@@ -247,7 +247,7 @@ ENTRY(arm1022_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(arm1022_dma_inv_range)
+arm1022_dma_inv_range:
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
 	tst	r0, #CACHE_DLINESIZE - 1
@@ -273,7 +273,7 @@ ENTRY(arm1022_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(arm1022_dma_clean_range)
+arm1022_dma_clean_range:
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
 	bic	r0, r0, #CACHE_DLINESIZE - 1
@@ -338,8 +338,6 @@ ENTRY(arm1022_cache_fns)
 	.long	arm1022_flush_kern_dcache_area
 	.long	arm1022_dma_map_area
 	.long	arm1022_dma_unmap_area
-	.long	arm1022_dma_inv_range
-	.long	arm1022_dma_clean_range
 	.long	arm1022_dma_flush_range
 
 	.align	5
diff --git a/arch/arm/mm/proc-arm1026.S b/arch/arm/mm/proc-arm1026.S
index 9ece6f666497..636672a29c6d 100644
--- a/arch/arm/mm/proc-arm1026.S
+++ b/arch/arm/mm/proc-arm1026.S
@@ -241,7 +241,7 @@ ENTRY(arm1026_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(arm1026_dma_inv_range)
+arm1026_dma_inv_range:
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
 	tst	r0, #CACHE_DLINESIZE - 1
@@ -267,7 +267,7 @@ ENTRY(arm1026_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(arm1026_dma_clean_range)
+arm1026_dma_clean_range:
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_DISABLE
 	bic	r0, r0, #CACHE_DLINESIZE - 1
@@ -332,8 +332,6 @@ ENTRY(arm1026_cache_fns)
 	.long	arm1026_flush_kern_dcache_area
 	.long	arm1026_dma_map_area
 	.long	arm1026_dma_unmap_area
-	.long	arm1026_dma_inv_range
-	.long	arm1026_dma_clean_range
 	.long	arm1026_dma_flush_range
 
 	.align	5
diff --git a/arch/arm/mm/proc-arm920.S b/arch/arm/mm/proc-arm920.S
index 6f6ab2747da6..8be81992645d 100644
--- a/arch/arm/mm/proc-arm920.S
+++ b/arch/arm/mm/proc-arm920.S
@@ -239,7 +239,7 @@ ENTRY(arm920_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(arm920_dma_inv_range)
+arm920_dma_inv_range:
 	tst	r0, #CACHE_DLINESIZE - 1
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -262,7 +262,7 @@ ENTRY(arm920_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(arm920_dma_clean_range)
+arm920_dma_clean_range:
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	add	r0, r0, #CACHE_DLINESIZE
@@ -321,8 +321,6 @@ ENTRY(arm920_cache_fns)
 	.long	arm920_flush_kern_dcache_area
 	.long	arm920_dma_map_area
 	.long	arm920_dma_unmap_area
-	.long	arm920_dma_inv_range
-	.long	arm920_dma_clean_range
 	.long	arm920_dma_flush_range
 
 #endif
diff --git a/arch/arm/mm/proc-arm922.S b/arch/arm/mm/proc-arm922.S
index 4e4396b121ca..c0ff8e4b1074 100644
--- a/arch/arm/mm/proc-arm922.S
+++ b/arch/arm/mm/proc-arm922.S
@@ -241,7 +241,7 @@ ENTRY(arm922_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(arm922_dma_inv_range)
+arm922_dma_inv_range:
 	tst	r0, #CACHE_DLINESIZE - 1
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -264,7 +264,7 @@ ENTRY(arm922_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(arm922_dma_clean_range)
+arm922_dma_clean_range:
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	add	r0, r0, #CACHE_DLINESIZE
@@ -323,8 +323,6 @@ ENTRY(arm922_cache_fns)
 	.long	arm922_flush_kern_dcache_area
 	.long	arm922_dma_map_area
 	.long	arm922_dma_unmap_area
-	.long	arm922_dma_inv_range
-	.long	arm922_dma_clean_range
 	.long	arm922_dma_flush_range
 
 #endif
diff --git a/arch/arm/mm/proc-arm925.S b/arch/arm/mm/proc-arm925.S
index 7c01c5d1108c..3c6cffe400f6 100644
--- a/arch/arm/mm/proc-arm925.S
+++ b/arch/arm/mm/proc-arm925.S
@@ -283,7 +283,7 @@ ENTRY(arm925_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(arm925_dma_inv_range)
+arm925_dma_inv_range:
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
 	tst	r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -308,7 +308,7 @@ ENTRY(arm925_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(arm925_dma_clean_range)
+arm925_dma_clean_range:
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -374,8 +374,6 @@ ENTRY(arm925_cache_fns)
 	.long	arm925_flush_kern_dcache_area
 	.long	arm925_dma_map_area
 	.long	arm925_dma_unmap_area
-	.long	arm925_dma_inv_range
-	.long	arm925_dma_clean_range
 	.long	arm925_dma_flush_range
 
 ENTRY(cpu_arm925_dcache_clean_area)
diff --git a/arch/arm/mm/proc-arm926.S b/arch/arm/mm/proc-arm926.S
index 72a01a4b80ab..75b707c9cce1 100644
--- a/arch/arm/mm/proc-arm926.S
+++ b/arch/arm/mm/proc-arm926.S
@@ -246,7 +246,7 @@ ENTRY(arm926_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(arm926_dma_inv_range)
+arm926_dma_inv_range:
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
 	tst	r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -271,7 +271,7 @@ ENTRY(arm926_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(arm926_dma_clean_range)
+arm926_dma_clean_range:
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -337,8 +337,6 @@ ENTRY(arm926_cache_fns)
 	.long	arm926_flush_kern_dcache_area
 	.long	arm926_dma_map_area
 	.long	arm926_dma_unmap_area
-	.long	arm926_dma_inv_range
-	.long	arm926_dma_clean_range
 	.long	arm926_dma_flush_range
 
 ENTRY(cpu_arm926_dcache_clean_area)
diff --git a/arch/arm/mm/proc-arm940.S b/arch/arm/mm/proc-arm940.S
index 6bb58fca7270..1af1657819eb 100644
--- a/arch/arm/mm/proc-arm940.S
+++ b/arch/arm/mm/proc-arm940.S
@@ -171,7 +171,7 @@ ENTRY(arm940_flush_kern_dcache_area)
  *	- start	- virtual start address
  *	- end	- virtual end address
  */
-ENTRY(arm940_dma_inv_range)
+arm940_dma_inv_range:
 	mov	ip, #0
 	mov	r1, #(CACHE_DSEGMENTS - 1) << 4	@ 4 segments
 1:	orr	r3, r1, #(CACHE_DENTRIES - 1) << 26 @ 64 entries
@@ -192,7 +192,7 @@ ENTRY(arm940_dma_inv_range)
  *	- start	- virtual start address
  *	- end	- virtual end address
  */
-ENTRY(arm940_dma_clean_range)
+arm940_dma_clean_range:
 ENTRY(cpu_arm940_dcache_clean_area)
 	mov	ip, #0
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
@@ -266,8 +266,6 @@ ENTRY(arm940_cache_fns)
 	.long	arm940_flush_kern_dcache_area
 	.long	arm940_dma_map_area
 	.long	arm940_dma_unmap_area
-	.long	arm940_dma_inv_range
-	.long	arm940_dma_clean_range
 	.long	arm940_dma_flush_range
 
 	__INIT
diff --git a/arch/arm/mm/proc-arm946.S b/arch/arm/mm/proc-arm946.S
index ac0f9ba719d7..1664b6aaff79 100644
--- a/arch/arm/mm/proc-arm946.S
+++ b/arch/arm/mm/proc-arm946.S
@@ -215,7 +215,7 @@ ENTRY(arm946_flush_kern_dcache_area)
  *	- end	- virtual end address
  * (same as arm926)
  */
-ENTRY(arm946_dma_inv_range)
+arm946_dma_inv_range:
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
 	tst	r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -240,7 +240,7 @@ ENTRY(arm946_dma_inv_range)
  *
  * (same as arm926)
  */
-ENTRY(arm946_dma_clean_range)
+arm946_dma_clean_range:
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -308,8 +308,6 @@ ENTRY(arm946_cache_fns)
 	.long	arm946_flush_kern_dcache_area
 	.long	arm946_dma_map_area
 	.long	arm946_dma_unmap_area
-	.long	arm946_dma_inv_range
-	.long	arm946_dma_clean_range
 	.long	arm946_dma_flush_range
 
 
diff --git a/arch/arm/mm/proc-feroceon.S b/arch/arm/mm/proc-feroceon.S
index 97e1d784f152..53e632343849 100644
--- a/arch/arm/mm/proc-feroceon.S
+++ b/arch/arm/mm/proc-feroceon.S
@@ -274,7 +274,7 @@ ENTRY(feroceon_range_flush_kern_dcache_area)
  * (same as v4wb)
  */
 	.align	5
-ENTRY(feroceon_dma_inv_range)
+feroceon_dma_inv_range:
 	tst	r0, #CACHE_DLINESIZE - 1
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -288,7 +288,7 @@ ENTRY(feroceon_dma_inv_range)
 	mov	pc, lr
 
 	.align	5
-ENTRY(feroceon_range_dma_inv_range)
+feroceon_range_dma_inv_range:
 	mrs	r2, cpsr
 	tst	r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -314,7 +314,7 @@ ENTRY(feroceon_range_dma_inv_range)
  * (same as v4wb)
  */
 	.align	5
-ENTRY(feroceon_dma_clean_range)
+feroceon_dma_clean_range:
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	add	r0, r0, #CACHE_DLINESIZE
@@ -324,7 +324,7 @@ ENTRY(feroceon_dma_clean_range)
 	mov	pc, lr
 
 	.align	5
-ENTRY(feroceon_range_dma_clean_range)
+feroceon_range_dma_clean_range:
 	mrs	r2, cpsr
 	cmp	r1, r0
 	subne	r1, r1, #1			@ top address is inclusive
@@ -414,8 +414,6 @@ ENTRY(feroceon_cache_fns)
 	.long	feroceon_flush_kern_dcache_area
 	.long	feroceon_dma_map_area
 	.long	feroceon_dma_unmap_area
-	.long	feroceon_dma_inv_range
-	.long	feroceon_dma_clean_range
 	.long	feroceon_dma_flush_range
 
 ENTRY(feroceon_range_cache_fns)
@@ -427,8 +425,6 @@ ENTRY(feroceon_range_cache_fns)
 	.long	feroceon_range_flush_kern_dcache_area
 	.long	feroceon_range_dma_map_area
 	.long	feroceon_dma_unmap_area
-	.long	feroceon_range_dma_inv_range
-	.long	feroceon_range_dma_clean_range
 	.long	feroceon_range_dma_flush_range
 
 	.align	5
diff --git a/arch/arm/mm/proc-mohawk.S b/arch/arm/mm/proc-mohawk.S
index 55b7fbec6548..caa31154e7db 100644
--- a/arch/arm/mm/proc-mohawk.S
+++ b/arch/arm/mm/proc-mohawk.S
@@ -218,7 +218,7 @@ ENTRY(mohawk_flush_kern_dcache_area)
  *
  * (same as v4wb)
  */
-ENTRY(mohawk_dma_inv_range)
+mohawk_dma_inv_range:
 	tst	r0, #CACHE_DLINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
 	tst	r1, #CACHE_DLINESIZE - 1
@@ -241,7 +241,7 @@ ENTRY(mohawk_dma_inv_range)
  *
  * (same as v4wb)
  */
-ENTRY(mohawk_dma_clean_range)
+mohawk_dma_clean_range:
 	bic	r0, r0, #CACHE_DLINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	add	r0, r0, #CACHE_DLINESIZE
@@ -301,8 +301,6 @@ ENTRY(mohawk_cache_fns)
 	.long	mohawk_flush_kern_dcache_area
 	.long	mohawk_dma_map_area
 	.long	mohawk_dma_unmap_area
-	.long	mohawk_dma_inv_range
-	.long	mohawk_dma_clean_range
 	.long	mohawk_dma_flush_range
 
 ENTRY(cpu_mohawk_dcache_clean_area)
diff --git a/arch/arm/mm/proc-xsc3.S b/arch/arm/mm/proc-xsc3.S
index 4e4ce889b3e6..046b3d88955e 100644
--- a/arch/arm/mm/proc-xsc3.S
+++ b/arch/arm/mm/proc-xsc3.S
@@ -257,7 +257,7 @@ ENTRY(xsc3_flush_kern_dcache_area)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(xsc3_dma_inv_range)
+xsc3_dma_inv_range:
 	tst	r0, #CACHELINESIZE - 1
 	bic	r0, r0, #CACHELINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean L1 D line
@@ -278,7 +278,7 @@ ENTRY(xsc3_dma_inv_range)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(xsc3_dma_clean_range)
+xsc3_dma_clean_range:
 	bic	r0, r0, #CACHELINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean L1 D line
 	add	r0, r0, #CACHELINESIZE
@@ -337,8 +337,6 @@ ENTRY(xsc3_cache_fns)
 	.long	xsc3_flush_kern_dcache_area
 	.long	xsc3_dma_map_area
 	.long	xsc3_dma_unmap_area
-	.long	xsc3_dma_inv_range
-	.long	xsc3_dma_clean_range
 	.long	xsc3_dma_flush_range
 
 ENTRY(cpu_xsc3_dcache_clean_area)
diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S
index a7999f94bf27..63037e2162f2 100644
--- a/arch/arm/mm/proc-xscale.S
+++ b/arch/arm/mm/proc-xscale.S
@@ -315,7 +315,7 @@ ENTRY(xscale_flush_kern_dcache_area)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(xscale_dma_inv_range)
+xscale_dma_inv_range:
 	tst	r0, #CACHELINESIZE - 1
 	bic	r0, r0, #CACHELINESIZE - 1
 	mcrne	p15, 0, r0, c7, c10, 1		@ clean D entry
@@ -336,7 +336,7 @@ ENTRY(xscale_dma_inv_range)
  *	- start  - virtual start address
  *	- end	 - virtual end address
  */
-ENTRY(xscale_dma_clean_range)
+xscale_dma_clean_range:
 	bic	r0, r0, #CACHELINESIZE - 1
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
 	add	r0, r0, #CACHELINESIZE
@@ -409,8 +409,6 @@ ENTRY(xscale_cache_fns)
 	.long	xscale_flush_kern_dcache_area
 	.long	xscale_dma_map_area
 	.long	xscale_dma_unmap_area
-	.long	xscale_dma_inv_range
-	.long	xscale_dma_clean_range
 	.long	xscale_dma_flush_range
 
 /*
@@ -436,8 +434,6 @@ ENTRY(xscale_80200_A0_A1_cache_fns)
 	.long	xscale_dma_a0_map_area
 	.long	xscale_dma_unmap_area
 	.long	xscale_dma_flush_range
-	.long	xscale_dma_clean_range
-	.long	xscale_dma_flush_range
 
 ENTRY(cpu_xscale_dcache_clean_area)
 1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
-- 
cgit v1.2.2


From 2ffe2da3e71652d4f4cae19539b5c78c2a239136 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sat, 31 Oct 2009 16:52:16 +0000
Subject: ARM: dma-mapping: fix for speculative prefetching

ARMv6 and ARMv7 CPUs can perform speculative prefetching, which makes
DMA cache coherency handling slightly more interesting.  Rather than
being able to rely upon the CPU not accessing the DMA buffer until DMA
has completed, we now must expect that the cache could be loaded with
possibly stale data from the DMA buffer.

Where DMA involves data being transferred to the device, we clean the
cache before handing it over for DMA, otherwise we invalidate the buffer
to get rid of potential writebacks.  On DMA Completion, if data was
transferred from the device, we invalidate the buffer to get rid of
any stale speculative prefetches.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Tested-By: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
 arch/arm/mm/cache-v6.S    | 10 ++++---
 arch/arm/mm/cache-v7.S    | 10 ++++---
 arch/arm/mm/dma-mapping.c | 68 +++++++++++++++++++++--------------------------
 3 files changed, 42 insertions(+), 46 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/cache-v6.S b/arch/arm/mm/cache-v6.S
index a11934e53fbd..9d89c67a1cc3 100644
--- a/arch/arm/mm/cache-v6.S
+++ b/arch/arm/mm/cache-v6.S
@@ -271,10 +271,9 @@ ENTRY(v6_dma_flush_range)
  */
 ENTRY(v6_dma_map_area)
 	add	r1, r1, r0
-	cmp	r2, #DMA_TO_DEVICE
-	beq	v6_dma_clean_range
-	bcs	v6_dma_inv_range
-	b	v6_dma_flush_range
+	teq	r2, #DMA_FROM_DEVICE
+	beq	v6_dma_inv_range
+	b	v6_dma_clean_range
 ENDPROC(v6_dma_map_area)
 
 /*
@@ -284,6 +283,9 @@ ENDPROC(v6_dma_map_area)
  *	- dir	- DMA direction
  */
 ENTRY(v6_dma_unmap_area)
+	add	r1, r1, r0
+	teq	r2, #DMA_TO_DEVICE
+	bne	v6_dma_inv_range
 	mov	pc, lr
 ENDPROC(v6_dma_unmap_area)
 
diff --git a/arch/arm/mm/cache-v7.S b/arch/arm/mm/cache-v7.S
index b1cd0fd91207..bcd64f265870 100644
--- a/arch/arm/mm/cache-v7.S
+++ b/arch/arm/mm/cache-v7.S
@@ -279,10 +279,9 @@ ENDPROC(v7_dma_flush_range)
  */
 ENTRY(v7_dma_map_area)
 	add	r1, r1, r0
-	cmp	r2, #DMA_TO_DEVICE
-	beq	v7_dma_clean_range
-	bcs	v7_dma_inv_range
-	b	v7_dma_flush_range
+	teq	r2, #DMA_FROM_DEVICE
+	beq	v7_dma_inv_range
+	b	v7_dma_clean_range
 ENDPROC(v7_dma_map_area)
 
 /*
@@ -292,6 +291,9 @@ ENDPROC(v7_dma_map_area)
  *	- dir	- DMA direction
  */
 ENTRY(v7_dma_unmap_area)
+	add	r1, r1, r0
+	teq	r2, #DMA_TO_DEVICE
+	bne	v7_dma_inv_range
 	mov	pc, lr
 ENDPROC(v7_dma_unmap_area)
 
diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index efa8efa33f5e..64daef2173bd 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -404,34 +404,22 @@ EXPORT_SYMBOL(dma_free_coherent);
  * platforms with CONFIG_DMABOUNCE.
  * Use the driver DMA support - see dma-mapping.h (dma_sync_*)
  */
-static void dma_cache_maint(const void *start, size_t size, int direction)
-{
-	void (*outer_op)(unsigned long, unsigned long);
-
-	switch (direction) {
-	case DMA_FROM_DEVICE:		/* invalidate only */
-		outer_op = outer_inv_range;
-		break;
-	case DMA_TO_DEVICE:		/* writeback only */
-		outer_op = outer_clean_range;
-		break;
-	case DMA_BIDIRECTIONAL:		/* writeback and invalidate */
-		outer_op = outer_flush_range;
-		break;
-	default:
-		BUG();
-	}
-
-	outer_op(__pa(start), __pa(start) + size);
-}
-
 void ___dma_single_cpu_to_dev(const void *kaddr, size_t size,
 	enum dma_data_direction dir)
 {
+	unsigned long paddr;
+
 	BUG_ON(!virt_addr_valid(kaddr) || !virt_addr_valid(kaddr + size - 1));
 
 	dmac_map_area(kaddr, size, dir);
-	dma_cache_maint(kaddr, size, dir);
+
+	paddr = __pa(kaddr);
+	if (dir == DMA_FROM_DEVICE) {
+		outer_inv_range(paddr, paddr + size);
+	} else {
+		outer_clean_range(paddr, paddr + size);
+	}
+	/* FIXME: non-speculating: flush on bidirectional mappings? */
 }
 EXPORT_SYMBOL(___dma_single_cpu_to_dev);
 
@@ -440,6 +428,13 @@ void ___dma_single_dev_to_cpu(const void *kaddr, size_t size,
 {
 	BUG_ON(!virt_addr_valid(kaddr) || !virt_addr_valid(kaddr + size - 1));
 
+	/* FIXME: non-speculating: not required */
+	/* don't bother invalidating if DMA to device */
+	if (dir != DMA_TO_DEVICE) {
+		unsigned long paddr = __pa(kaddr);
+		outer_inv_range(paddr, paddr + size);
+	}
+
 	dmac_unmap_area(kaddr, size, dir);
 }
 EXPORT_SYMBOL(___dma_single_dev_to_cpu);
@@ -487,32 +482,29 @@ void ___dma_page_cpu_to_dev(struct page *page, unsigned long off,
 	size_t size, enum dma_data_direction dir)
 {
 	unsigned long paddr;
-	void (*outer_op)(unsigned long, unsigned long);
-
-	switch (direction) {
-	case DMA_FROM_DEVICE:		/* invalidate only */
-		outer_op = outer_inv_range;
-		break;
-	case DMA_TO_DEVICE:		/* writeback only */
-		outer_op = outer_clean_range;
-		break;
-	case DMA_BIDIRECTIONAL:		/* writeback and invalidate */
-		outer_op = outer_flush_range;
-		break;
-	default:
-		BUG();
-	}
 
 	dma_cache_maint_page(page, off, size, dir, dmac_map_area);
 
 	paddr = page_to_phys(page) + off;
-	outer_op(paddr, paddr + size);
+	if (dir == DMA_FROM_DEVICE) {
+		outer_inv_range(paddr, paddr + size);
+	} else {
+		outer_clean_range(paddr, paddr + size);
+	}
+	/* FIXME: non-speculating: flush on bidirectional mappings? */
 }
 EXPORT_SYMBOL(___dma_page_cpu_to_dev);
 
 void ___dma_page_dev_to_cpu(struct page *page, unsigned long off,
 	size_t size, enum dma_data_direction dir)
 {
+	unsigned long paddr = page_to_phys(page) + off;
+
+	/* FIXME: non-speculating: not required */
+	/* don't bother invalidating if DMA to device */
+	if (dir != DMA_TO_DEVICE)
+		outer_inv_range(paddr, paddr + size);
+
 	dma_cache_maint_page(page, off, size, dir, dmac_unmap_area);
 }
 EXPORT_SYMBOL(___dma_page_dev_to_cpu);
-- 
cgit v1.2.2


From 31aa8fd6fd30b0f36416df7d09619768d26b4332 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 18 Dec 2009 11:10:03 +0000
Subject: ARM: Add caller information to ioremap

This allows the procfs vmallocinfo file to show who created the ioremap
regions.  Note: __builtin_return_address(0) doesn't do what's expected
if its used in an inline function, so we leave __arm_ioremap callers
in such places alone.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/ioremap.c | 57 ++++++++++++++++++++++++++++++---------------------
 arch/arm/mm/nommu.c   | 12 +++++++++++
 2 files changed, 46 insertions(+), 23 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/ioremap.c b/arch/arm/mm/ioremap.c
index 0ab75c60f7cf..28c8b950ef04 100644
--- a/arch/arm/mm/ioremap.c
+++ b/arch/arm/mm/ioremap.c
@@ -139,8 +139,8 @@ void __check_kvm_seq(struct mm_struct *mm)
  * which requires the new ioremap'd region to be referenced, the CPU will
  * reference the _old_ region.
  *
- * Note that get_vm_area() allocates a guard 4K page, so we need to mask
- * the size back to 1MB aligned or we will overflow in the loop below.
+ * Note that get_vm_area_caller() allocates a guard 4K page, so we need to
+ * mask the size back to 1MB aligned or we will overflow in the loop below.
  */
 static void unmap_area_sections(unsigned long virt, unsigned long size)
 {
@@ -254,22 +254,8 @@ remap_area_supersections(unsigned long virt, unsigned long pfn,
 }
 #endif
 
-
-/*
- * Remap an arbitrary physical address space into the kernel virtual
- * address space. Needed when the kernel wants to access high addresses
- * directly.
- *
- * NOTE! We need to allow non-page-aligned mappings too: we will obviously
- * have to convert them into an offset in a page-aligned mapping, but the
- * caller shouldn't need to know that small detail.
- *
- * 'flags' are the extra L_PTE_ flags that you want to specify for this
- * mapping.  See <asm/pgtable.h> for more information.
- */
-void __iomem *
-__arm_ioremap_pfn(unsigned long pfn, unsigned long offset, size_t size,
-		  unsigned int mtype)
+void __iomem * __arm_ioremap_pfn_caller(unsigned long pfn,
+	unsigned long offset, size_t size, unsigned int mtype, void *caller)
 {
 	const struct mem_type *type;
 	int err;
@@ -291,7 +277,7 @@ __arm_ioremap_pfn(unsigned long pfn, unsigned long offset, size_t size,
 	 */
 	size = PAGE_ALIGN(offset + size);
 
- 	area = get_vm_area(size, VM_IOREMAP);
+	area = get_vm_area_caller(size, VM_IOREMAP, caller);
  	if (!area)
  		return NULL;
  	addr = (unsigned long)area->addr;
@@ -318,10 +304,9 @@ __arm_ioremap_pfn(unsigned long pfn, unsigned long offset, size_t size,
 	flush_cache_vmap(addr, addr + size);
 	return (void __iomem *) (offset + addr);
 }
-EXPORT_SYMBOL(__arm_ioremap_pfn);
 
-void __iomem *
-__arm_ioremap(unsigned long phys_addr, size_t size, unsigned int mtype)
+void __iomem *__arm_ioremap_caller(unsigned long phys_addr, size_t size,
+	unsigned int mtype, void *caller)
 {
 	unsigned long last_addr;
  	unsigned long offset = phys_addr & ~PAGE_MASK;
@@ -334,7 +319,33 @@ __arm_ioremap(unsigned long phys_addr, size_t size, unsigned int mtype)
 	if (!size || last_addr < phys_addr)
 		return NULL;
 
- 	return __arm_ioremap_pfn(pfn, offset, size, mtype);
+	return __arm_ioremap_pfn_caller(pfn, offset, size, mtype,
+			caller);
+}
+
+/*
+ * Remap an arbitrary physical address space into the kernel virtual
+ * address space. Needed when the kernel wants to access high addresses
+ * directly.
+ *
+ * NOTE! We need to allow non-page-aligned mappings too: we will obviously
+ * have to convert them into an offset in a page-aligned mapping, but the
+ * caller shouldn't need to know that small detail.
+ */
+void __iomem *
+__arm_ioremap_pfn(unsigned long pfn, unsigned long offset, size_t size,
+		  unsigned int mtype)
+{
+	return __arm_ioremap_pfn_caller(pfn, offset, size, mtype,
+			__builtin_return_address(0));
+}
+EXPORT_SYMBOL(__arm_ioremap_pfn);
+
+void __iomem *
+__arm_ioremap(unsigned long phys_addr, size_t size, unsigned int mtype)
+{
+	return __arm_ioremap_caller(phys_addr, size, mtype,
+			__builtin_return_address(0));
 }
 EXPORT_SYMBOL(__arm_ioremap);
 
diff --git a/arch/arm/mm/nommu.c b/arch/arm/mm/nommu.c
index 374a8311bc84..9bfeb6b9509a 100644
--- a/arch/arm/mm/nommu.c
+++ b/arch/arm/mm/nommu.c
@@ -74,6 +74,12 @@ void __iomem *__arm_ioremap_pfn(unsigned long pfn, unsigned long offset,
 }
 EXPORT_SYMBOL(__arm_ioremap_pfn);
 
+void __iomem *__arm_ioremap_pfn_caller(unsigned long pfn, unsigned long offset,
+			   size_t size, unsigned int mtype, void *caller)
+{
+	return __arm_ioremap_pfn(pfn, offset, size, mtype);
+}
+
 void __iomem *__arm_ioremap(unsigned long phys_addr, size_t size,
 			    unsigned int mtype)
 {
@@ -81,6 +87,12 @@ void __iomem *__arm_ioremap(unsigned long phys_addr, size_t size,
 }
 EXPORT_SYMBOL(__arm_ioremap);
 
+void __iomem *__arm_ioremap(unsigned long phys_addr, size_t size,
+			    unsigned int mtype, void *caller)
+{
+	return __arm_ioremap(phys_addr, size, mtype);
+}
+
 void __iounmap(volatile void __iomem *addr)
 {
 }
-- 
cgit v1.2.2


From e119bfff1f102f8d1505910cd6c09df55c776b43 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Sun, 10 Jan 2010 17:23:29 +0000
Subject: ARM: Move creation of /proc/cpu out of alignment.c

Always creating this directory avoids other users having to jump
through silly hoops when they want to share this directory.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/alignment.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/alignment.c b/arch/arm/mm/alignment.c
index b270d6228fe2..0c5eb6983cef 100644
--- a/arch/arm/mm/alignment.c
+++ b/arch/arm/mm/alignment.c
@@ -898,11 +898,7 @@ static int __init alignment_init(void)
 #ifdef CONFIG_PROC_FS
 	struct proc_dir_entry *res;
 
-	res = proc_mkdir("cpu", NULL);
-	if (!res)
-		return -ENOMEM;
-
-	res = create_proc_entry("alignment", S_IWUSR | S_IRUGO, res);
+	res = create_proc_entry("cpu/alignment", S_IWUSR | S_IRUGO, NULL);
 	if (!res)
 		return -ENOMEM;
 
-- 
cgit v1.2.2


From 2b0d8c251b8876d530a6bf671eb5425838fa698a Mon Sep 17 00:00:00 2001
From: Jeremy Kerr <jeremy.kerr@canonical.com>
Date: Mon, 11 Jan 2010 23:17:34 +0100
Subject: ARM: 5880/1: arm: use generic infrastructure for early params

The ARM setup code includes its own parser for early params, there's
also one in the generic init code.

This patch removes __early_init (and related code) from
arch/arm/kernel/setup.c, and changes users to the generic early_init
macro instead.

The generic macro takes a char * argument, rather than char **, so we
need to update the parser functions a little.

Signed-off-by: Jeremy Kerr <jeremy.kerr@canonical.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c | 12 +++++++-----
 arch/arm/mm/mmu.c  | 41 +++++++++++++++++++++--------------------
 2 files changed, 28 insertions(+), 25 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index a04ffbbbe253..a340569b991e 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -32,19 +32,21 @@
 static unsigned long phys_initrd_start __initdata = 0;
 static unsigned long phys_initrd_size __initdata = 0;
 
-static void __init early_initrd(char **p)
+static int __init early_initrd(char *p)
 {
 	unsigned long start, size;
+	char *endp;
 
-	start = memparse(*p, p);
-	if (**p == ',') {
-		size = memparse((*p) + 1, p);
+	start = memparse(p, &endp);
+	if (*endp == ',') {
+		size = memparse(endp + 1, NULL);
 
 		phys_initrd_start = start;
 		phys_initrd_size = size;
 	}
+	return 0;
 }
-__early_param("initrd=", early_initrd);
+early_param("initrd", early_initrd);
 
 static int __init parse_tag_initrd(const struct tag *tag)
 {
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index 1708da82da96..88f5d71248d9 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -100,18 +100,17 @@ static struct cachepolicy cache_policies[] __initdata = {
  * writebuffer to be turned off.  (Note: the write
  * buffer should not be on and the cache off).
  */
-static void __init early_cachepolicy(char **p)
+static int __init early_cachepolicy(char *p)
 {
 	int i;
 
 	for (i = 0; i < ARRAY_SIZE(cache_policies); i++) {
 		int len = strlen(cache_policies[i].policy);
 
-		if (memcmp(*p, cache_policies[i].policy, len) == 0) {
+		if (memcmp(p, cache_policies[i].policy, len) == 0) {
 			cachepolicy = i;
 			cr_alignment &= ~cache_policies[i].cr_mask;
 			cr_no_alignment &= ~cache_policies[i].cr_mask;
-			*p += len;
 			break;
 		}
 	}
@@ -130,36 +129,37 @@ static void __init early_cachepolicy(char **p)
 	}
 	flush_cache_all();
 	set_cr(cr_alignment);
+	return 0;
 }
-__early_param("cachepolicy=", early_cachepolicy);
+early_param("cachepolicy", early_cachepolicy);
 
-static void __init early_nocache(char **__unused)
+static int __init early_nocache(char *__unused)
 {
 	char *p = "buffered";
 	printk(KERN_WARNING "nocache is deprecated; use cachepolicy=%s\n", p);
-	early_cachepolicy(&p);
+	early_cachepolicy(p);
+	return 0;
 }
-__early_param("nocache", early_nocache);
+early_param("nocache", early_nocache);
 
-static void __init early_nowrite(char **__unused)
+static int __init early_nowrite(char *__unused)
 {
 	char *p = "uncached";
 	printk(KERN_WARNING "nowb is deprecated; use cachepolicy=%s\n", p);
-	early_cachepolicy(&p);
+	early_cachepolicy(p);
+	return 0;
 }
-__early_param("nowb", early_nowrite);
+early_param("nowb", early_nowrite);
 
-static void __init early_ecc(char **p)
+static int __init early_ecc(char *p)
 {
-	if (memcmp(*p, "on", 2) == 0) {
+	if (memcmp(p, "on", 2) == 0)
 		ecc_mask = PMD_PROTECTION;
-		*p += 2;
-	} else if (memcmp(*p, "off", 3) == 0) {
+	else if (memcmp(p, "off", 3) == 0)
 		ecc_mask = 0;
-		*p += 3;
-	}
+	return 0;
 }
-__early_param("ecc=", early_ecc);
+early_param("ecc", early_ecc);
 
 static int __init noalign_setup(char *__unused)
 {
@@ -670,9 +670,9 @@ static unsigned long __initdata vmalloc_reserve = SZ_128M;
  * bytes. This can be used to increase (or decrease) the vmalloc
  * area - the default is 128m.
  */
-static void __init early_vmalloc(char **arg)
+static int __init early_vmalloc(char *arg)
 {
-	vmalloc_reserve = memparse(*arg, arg);
+	vmalloc_reserve = memparse(arg, NULL);
 
 	if (vmalloc_reserve < SZ_16M) {
 		vmalloc_reserve = SZ_16M;
@@ -687,8 +687,9 @@ static void __init early_vmalloc(char **arg)
 			"vmalloc area is too big, limiting to %luMB\n",
 			vmalloc_reserve >> 20);
 	}
+	return 0;
 }
-__early_param("vmalloc=", early_vmalloc);
+early_param("vmalloc", early_vmalloc);
 
 #define VMALLOC_MIN	(void *)(VMALLOC_END - vmalloc_reserve)
 
-- 
cgit v1.2.2


From 11805bcfa411c816b7c76fc40724be6733c74ffc Mon Sep 17 00:00:00 2001
From: Catalin Marinas <catalin.marinas@arm.com>
Date: Tue, 26 Jan 2010 19:09:42 +0100
Subject: ARM: 5905/1: ARM: Global ASID allocation on SMP

The current ASID allocation algorithm doesn't ensure the notification
of the other CPUs when the ASID rolls over. This may lead to two
processes using the same ASID (but different generation) or multiple
threads of the same process using different ASIDs.

This patch adds the broadcasting of the ASID rollover event to the
other CPUs. To avoid a race on multiple CPUs modifying "cpu_last_asid"
during the handling of the broadcast, the ASID numbering now starts at
"smp_processor_id() + 1". At rollover, the cpu_last_asid will be set
to NR_CPUS.

Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/context.c | 124 ++++++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 110 insertions(+), 14 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/context.c b/arch/arm/mm/context.c
index a9e22e31eaa1..b0ee9ba3cfab 100644
--- a/arch/arm/mm/context.c
+++ b/arch/arm/mm/context.c
@@ -10,12 +10,17 @@
 #include <linux/init.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
+#include <linux/smp.h>
+#include <linux/percpu.h>
 
 #include <asm/mmu_context.h>
 #include <asm/tlbflush.h>
 
 static DEFINE_SPINLOCK(cpu_asid_lock);
 unsigned int cpu_last_asid = ASID_FIRST_VERSION;
+#ifdef CONFIG_SMP
+DEFINE_PER_CPU(struct mm_struct *, current_mm);
+#endif
 
 /*
  * We fork()ed a process, and we need a new context for the child
@@ -26,13 +31,109 @@ unsigned int cpu_last_asid = ASID_FIRST_VERSION;
 void __init_new_context(struct task_struct *tsk, struct mm_struct *mm)
 {
 	mm->context.id = 0;
+	spin_lock_init(&mm->context.id_lock);
 }
 
+static void flush_context(void)
+{
+	/* set the reserved ASID before flushing the TLB */
+	asm("mcr	p15, 0, %0, c13, c0, 1\n" : : "r" (0));
+	isb();
+	local_flush_tlb_all();
+	if (icache_is_vivt_asid_tagged()) {
+		__flush_icache_all();
+		dsb();
+	}
+}
+
+#ifdef CONFIG_SMP
+
+static void set_mm_context(struct mm_struct *mm, unsigned int asid)
+{
+	unsigned long flags;
+
+	/*
+	 * Locking needed for multi-threaded applications where the
+	 * same mm->context.id could be set from different CPUs during
+	 * the broadcast. This function is also called via IPI so the
+	 * mm->context.id_lock has to be IRQ-safe.
+	 */
+	spin_lock_irqsave(&mm->context.id_lock, flags);
+	if (likely((mm->context.id ^ cpu_last_asid) >> ASID_BITS)) {
+		/*
+		 * Old version of ASID found. Set the new one and
+		 * reset mm_cpumask(mm).
+		 */
+		mm->context.id = asid;
+		cpumask_clear(mm_cpumask(mm));
+	}
+	spin_unlock_irqrestore(&mm->context.id_lock, flags);
+
+	/*
+	 * Set the mm_cpumask(mm) bit for the current CPU.
+	 */
+	cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm));
+}
+
+/*
+ * Reset the ASID on the current CPU. This function call is broadcast
+ * from the CPU handling the ASID rollover and holding cpu_asid_lock.
+ */
+static void reset_context(void *info)
+{
+	unsigned int asid;
+	unsigned int cpu = smp_processor_id();
+	struct mm_struct *mm = per_cpu(current_mm, cpu);
+
+	/*
+	 * Check if a current_mm was set on this CPU as it might still
+	 * be in the early booting stages and using the reserved ASID.
+	 */
+	if (!mm)
+		return;
+
+	smp_rmb();
+	asid = cpu_last_asid + cpu + 1;
+
+	flush_context();
+	set_mm_context(mm, asid);
+
+	/* set the new ASID */
+	asm("mcr	p15, 0, %0, c13, c0, 1\n" : : "r" (mm->context.id));
+	isb();
+}
+
+#else
+
+static inline void set_mm_context(struct mm_struct *mm, unsigned int asid)
+{
+	mm->context.id = asid;
+	cpumask_copy(mm_cpumask(mm), cpumask_of(smp_processor_id()));
+}
+
+#endif
+
 void __new_context(struct mm_struct *mm)
 {
 	unsigned int asid;
 
 	spin_lock(&cpu_asid_lock);
+#ifdef CONFIG_SMP
+	/*
+	 * Check the ASID again, in case the change was broadcast from
+	 * another CPU before we acquired the lock.
+	 */
+	if (unlikely(((mm->context.id ^ cpu_last_asid) >> ASID_BITS) == 0)) {
+		cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm));
+		spin_unlock(&cpu_asid_lock);
+		return;
+	}
+#endif
+	/*
+	 * At this point, it is guaranteed that the current mm (with
+	 * an old ASID) isn't active on any other CPU since the ASIDs
+	 * are changed simultaneously via IPI.
+	 */
 	asid = ++cpu_last_asid;
 	if (asid == 0)
 		asid = cpu_last_asid = ASID_FIRST_VERSION;
@@ -42,20 +143,15 @@ void __new_context(struct mm_struct *mm)
 	 * to start a new version and flush the TLB.
 	 */
 	if (unlikely((asid & ~ASID_MASK) == 0)) {
-		asid = ++cpu_last_asid;
-		/* set the reserved ASID before flushing the TLB */
-		asm("mcr	p15, 0, %0, c13, c0, 1	@ set reserved context ID\n"
-		    :
-		    : "r" (0));
-		isb();
-		flush_tlb_all();
-		if (icache_is_vivt_asid_tagged()) {
-			__flush_icache_all();
-			dsb();
-		}
+		asid = cpu_last_asid + smp_processor_id() + 1;
+		flush_context();
+#ifdef CONFIG_SMP
+		smp_wmb();
+		smp_call_function(reset_context, NULL, 1);
+#endif
+		cpu_last_asid += NR_CPUS;
 	}
-	spin_unlock(&cpu_asid_lock);
 
-	cpumask_copy(mm_cpumask(mm), cpumask_of(smp_processor_id()));
-	mm->context.id = asid;
+	set_mm_context(mm, asid);
+	spin_unlock(&cpu_asid_lock);
 }
-- 
cgit v1.2.2


From 1a28e3d977860dc760909083df625b300f695680 Mon Sep 17 00:00:00 2001
From: Tony Lindgren <tony@atomide.com>
Date: Mon, 1 Feb 2010 23:30:26 +0100
Subject: ARM: 5911/1: ARM: Select CPU_32v6K for CPU_V7 only if ARCH_OMAP2 is
 not selected

Otherwise the kernel built with both CPU_V6 and CPU_V7 will not
boot on omap2.

Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/Kconfig | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index baf638487a2d..4c2e90dc549c 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -399,7 +399,7 @@ config CPU_V6
 config CPU_32v6K
 	bool "Support ARM V6K processor extensions" if !SMP
 	depends on CPU_V6
-	default y if SMP && !ARCH_MX3
+	default y if SMP && !(ARCH_MX3 || ARCH_OMAP2)
 	help
 	  Say Y here if your ARMv6 processor supports the 'K' extension.
 	  This enables the kernel to use some instructions not present
@@ -410,7 +410,7 @@ config CPU_32v6K
 # ARMv7
 config CPU_V7
 	bool "Support ARM V7 processor" if ARCH_INTEGRATOR || MACH_REALVIEW_EB || MACH_REALVIEW_PBX
-	select CPU_32v6K
+	select CPU_32v6K if !ARCH_OMAP2
 	select CPU_32v7
 	select CPU_ABRT_EV7
 	select CPU_PABRT_V7
-- 
cgit v1.2.2


From 424d6b145f863d012c540082d0c1afb5bb4dea48 Mon Sep 17 00:00:00 2001
From: Santosh Shilimkar <santosh.shilimkar@ti.com>
Date: Thu, 4 Feb 2010 19:35:06 +0100
Subject: ARM: 5916/1: ARM: L2 : Add maintainace by line helper functions

This patch adds the cache maintainance by line helper functions.

Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/cache-l2x0.c | 36 ++++++++++++++++++++++++++----------
 1 file changed, 26 insertions(+), 10 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c
index cb8fc6573b1b..1a14d18e5713 100644
--- a/arch/arm/mm/cache-l2x0.c
+++ b/arch/arm/mm/cache-l2x0.c
@@ -42,6 +42,27 @@ static inline void cache_sync(void)
 	cache_wait(base + L2X0_CACHE_SYNC, 1);
 }
 
+static inline void l2x0_clean_line(unsigned long addr)
+{
+	void __iomem *base = l2x0_base;
+	cache_wait(base + L2X0_CLEAN_LINE_PA, 1);
+	writel(addr, base + L2X0_CLEAN_LINE_PA);
+}
+
+static inline void l2x0_inv_line(unsigned long addr)
+{
+	void __iomem *base = l2x0_base;
+	cache_wait(base + L2X0_INV_LINE_PA, 1);
+	writel(addr, base + L2X0_INV_LINE_PA);
+}
+
+static inline void l2x0_flush_line(unsigned long addr)
+{
+	void __iomem *base = l2x0_base;
+	cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
+	writel(addr, base + L2X0_CLEAN_INV_LINE_PA);
+}
+
 static inline void l2x0_inv_all(void)
 {
 	unsigned long flags;
@@ -62,23 +83,20 @@ static void l2x0_inv_range(unsigned long start, unsigned long end)
 	spin_lock_irqsave(&l2x0_lock, flags);
 	if (start & (CACHE_LINE_SIZE - 1)) {
 		start &= ~(CACHE_LINE_SIZE - 1);
-		cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
-		writel(start, base + L2X0_CLEAN_INV_LINE_PA);
+		l2x0_flush_line(start);
 		start += CACHE_LINE_SIZE;
 	}
 
 	if (end & (CACHE_LINE_SIZE - 1)) {
 		end &= ~(CACHE_LINE_SIZE - 1);
-		cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
-		writel(end, base + L2X0_CLEAN_INV_LINE_PA);
+		l2x0_flush_line(end);
 	}
 
 	while (start < end) {
 		unsigned long blk_end = start + min(end - start, 4096UL);
 
 		while (start < blk_end) {
-			cache_wait(base + L2X0_INV_LINE_PA, 1);
-			writel(start, base + L2X0_INV_LINE_PA);
+			l2x0_inv_line(start);
 			start += CACHE_LINE_SIZE;
 		}
 
@@ -103,8 +121,7 @@ static void l2x0_clean_range(unsigned long start, unsigned long end)
 		unsigned long blk_end = start + min(end - start, 4096UL);
 
 		while (start < blk_end) {
-			cache_wait(base + L2X0_CLEAN_LINE_PA, 1);
-			writel(start, base + L2X0_CLEAN_LINE_PA);
+			l2x0_clean_line(start);
 			start += CACHE_LINE_SIZE;
 		}
 
@@ -129,8 +146,7 @@ static void l2x0_flush_range(unsigned long start, unsigned long end)
 		unsigned long blk_end = start + min(end - start, 4096UL);
 
 		while (start < blk_end) {
-			cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
-			writel(start, base + L2X0_CLEAN_INV_LINE_PA);
+			l2x0_flush_line(start);
 			start += CACHE_LINE_SIZE;
 		}
 
-- 
cgit v1.2.2


From d309427e792ea750cdd312e7a92cf6047ae44962 Mon Sep 17 00:00:00 2001
From: Santosh Shilimkar <santosh.shilimkar@ti.com>
Date: Thu, 4 Feb 2010 19:37:09 +0100
Subject: ARM: 5917/1: OMAP4: Add L2 Cache support

This patch adds L2 Cache support for OMAP4. External L2 cache
is used in OMAP4

CC: Catalin Marinas <catalin.marinas@arm.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 4c2e90dc549c..e859743024a0 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -754,7 +754,7 @@ config CACHE_FEROCEON_L2_WRITETHROUGH
 config CACHE_L2X0
 	bool "Enable the L2x0 outer cache controller"
 	depends on REALVIEW_EB_ARM11MP || MACH_REALVIEW_PB11MP || MACH_REALVIEW_PB1176 || \
-		   REALVIEW_EB_A9MP || ARCH_MX35 || ARCH_MX31 || MACH_REALVIEW_PBX || ARCH_NOMADIK
+		   REALVIEW_EB_A9MP || ARCH_MX35 || ARCH_MX31 || MACH_REALVIEW_PBX || ARCH_NOMADIK || ARCH_OMAP4
 	default y
 	select OUTER_CACHE
 	help
-- 
cgit v1.2.2


From 9e65582a8e8715f883a34eea66e0643778ce878d Mon Sep 17 00:00:00 2001
From: Santosh Shilimkar <santosh.shilimkar@ti.com>
Date: Thu, 4 Feb 2010 19:42:42 +0100
Subject: ARM: 5919/1: ARM: L2 : Errata 588369: Clean & Invalidate do not
 invalidate clean lines

This patch implements the work-around for the errata 588369.The secure
API is used to alter L2 debug register because of trust-zone.

This version updated with comments from Russell and Catalin and
generated against 2.6.33-rc6 mainline kernel. Detail
comments can be found:
http://www.spinics.net/lists/linux-omap/msg23431.html

Signed-off-by: Woodruff Richard <r-woodruff2@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/cache-l2x0.c | 36 ++++++++++++++++++++++++++++++++++++
 1 file changed, 36 insertions(+)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c
index 1a14d18e5713..07334632d3e2 100644
--- a/arch/arm/mm/cache-l2x0.c
+++ b/arch/arm/mm/cache-l2x0.c
@@ -56,12 +56,42 @@ static inline void l2x0_inv_line(unsigned long addr)
 	writel(addr, base + L2X0_INV_LINE_PA);
 }
 
+#ifdef CONFIG_PL310_ERRATA_588369
+static void debug_writel(unsigned long val)
+{
+	extern void omap_smc1(u32 fn, u32 arg);
+
+	/*
+	 * Texas Instrument secure monitor api to modify the
+	 * PL310 Debug Control Register.
+	 */
+	omap_smc1(0x100, val);
+}
+
+static inline void l2x0_flush_line(unsigned long addr)
+{
+	void __iomem *base = l2x0_base;
+
+	/* Clean by PA followed by Invalidate by PA */
+	cache_wait(base + L2X0_CLEAN_LINE_PA, 1);
+	writel(addr, base + L2X0_CLEAN_LINE_PA);
+	cache_wait(base + L2X0_INV_LINE_PA, 1);
+	writel(addr, base + L2X0_INV_LINE_PA);
+}
+#else
+
+/* Optimised out for non-errata case */
+static inline void debug_writel(unsigned long val)
+{
+}
+
 static inline void l2x0_flush_line(unsigned long addr)
 {
 	void __iomem *base = l2x0_base;
 	cache_wait(base + L2X0_CLEAN_INV_LINE_PA, 1);
 	writel(addr, base + L2X0_CLEAN_INV_LINE_PA);
 }
+#endif
 
 static inline void l2x0_inv_all(void)
 {
@@ -83,13 +113,17 @@ static void l2x0_inv_range(unsigned long start, unsigned long end)
 	spin_lock_irqsave(&l2x0_lock, flags);
 	if (start & (CACHE_LINE_SIZE - 1)) {
 		start &= ~(CACHE_LINE_SIZE - 1);
+		debug_writel(0x03);
 		l2x0_flush_line(start);
+		debug_writel(0x00);
 		start += CACHE_LINE_SIZE;
 	}
 
 	if (end & (CACHE_LINE_SIZE - 1)) {
 		end &= ~(CACHE_LINE_SIZE - 1);
+		debug_writel(0x03);
 		l2x0_flush_line(end);
+		debug_writel(0x00);
 	}
 
 	while (start < end) {
@@ -145,10 +179,12 @@ static void l2x0_flush_range(unsigned long start, unsigned long end)
 	while (start < end) {
 		unsigned long blk_end = start + min(end - start, 4096UL);
 
+		debug_writel(0x03);
 		while (start < blk_end) {
 			l2x0_flush_line(start);
 			start += CACHE_LINE_SIZE;
 		}
+		debug_writel(0x00);
 
 		if (blk_end < end) {
 			spin_unlock_irqrestore(&l2x0_lock, flags);
-- 
cgit v1.2.2


From db9ef1af4879c121c354ad2f653f185f1d50fd89 Mon Sep 17 00:00:00 2001
From: Fenkart/Bostandzhyan <andreas.fenkart@streamunlimited.com>
Date: Sun, 7 Feb 2010 21:45:47 +0100
Subject: ARM: 5926/1: Add "Virtual kernel memory..." printout.

Code based on parisc and x86_32.

Tested-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Andreas Fenkart <andreas.fenkart@streamunlimited.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c | 78 +++++++++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 69 insertions(+), 9 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index a340569b991e..e8e3a74ac5b5 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -23,6 +23,7 @@
 #include <asm/setup.h>
 #include <asm/sizes.h>
 #include <asm/tlb.h>
+#include <asm/fixmap.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
@@ -562,7 +563,7 @@ static void __init free_unused_memmap_node(int node, struct meminfo *mi)
  */
 void __init mem_init(void)
 {
-	unsigned int codesize, datasize, initsize;
+	unsigned long reserved_pages, free_pages;
 	int i, node;
 
 #ifndef CONFIG_DISCONTIGMEM
@@ -598,6 +599,33 @@ void __init mem_init(void)
 	totalram_pages += totalhigh_pages;
 #endif
 
+	reserved_pages = free_pages = 0;
+
+	for_each_online_node(node) {
+		pg_data_t *n = NODE_DATA(node);
+		struct page *map = pgdat_page_nr(n, 0) - n->node_start_pfn;
+
+		for_each_nodebank(i, &meminfo, node) {
+			struct membank *bank = &meminfo.bank[i];
+			unsigned int pfn1, pfn2;
+			struct page *page, *end;
+
+			pfn1 = bank_pfn_start(bank);
+			pfn2 = bank_pfn_end(bank);
+
+			page = map + pfn1;
+			end  = map + pfn2;
+
+			do {
+				if (PageReserved(page))
+					reserved_pages++;
+				else if (!page_count(page))
+					free_pages++;
+				page++;
+			} while (page < end);
+		}
+	}
+
 	/*
 	 * Since our memory may not be contiguous, calculate the
 	 * real number of pages we have in this system
@@ -610,16 +638,48 @@ void __init mem_init(void)
 	}
 	printk(" = %luMB total\n", num_physpages >> (20 - PAGE_SHIFT));
 
-	codesize = _etext - _text;
-	datasize = _end - _data;
-	initsize = __init_end - __init_begin;
-
-	printk(KERN_NOTICE "Memory: %luKB available (%dK code, "
-		"%dK data, %dK init, %luK highmem)\n",
-		nr_free_pages() << (PAGE_SHIFT-10), codesize >> 10,
-		datasize >> 10, initsize >> 10,
+	printk(KERN_NOTICE "Memory: %luk/%luk available, %luk reserved, %luK highmem\n",
+		nr_free_pages() << (PAGE_SHIFT-10),
+		free_pages << (PAGE_SHIFT-10),
+		reserved_pages << (PAGE_SHIFT-10),
 		totalhigh_pages << (PAGE_SHIFT-10));
 
+#define MLK(b, t) b, t, ((t) - (b)) >> 10
+#define MLM(b, t) b, t, ((t) - (b)) >> 20
+#define MLK_ROUNDUP(b, t) b, t, DIV_ROUND_UP(((t) - (b)), SZ_1K)
+
+	printk(KERN_NOTICE "Virtual kernel memory layout:\n"
+			"    vector  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+			"    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+			"    vmalloc : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+			"    lowmem  : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+#ifdef CONFIG_HIGHMEM
+			"    pkmap   : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+#endif
+			"    modules : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+			"      .init : 0x%p" " - 0x%p" "   (%4d kB)\n"
+			"      .text : 0x%p" " - 0x%p" "   (%4d kB)\n"
+			"      .data : 0x%p" " - 0x%p" "   (%4d kB)\n",
+
+			MLK(UL(CONFIG_VECTORS_BASE), UL(CONFIG_VECTORS_BASE) +
+				(PAGE_SIZE)),
+			MLK(FIXADDR_START, FIXADDR_TOP),
+			MLM(VMALLOC_START, (unsigned long)VMALLOC_END),
+			MLM(PAGE_OFFSET, (unsigned long)high_memory),
+#ifdef CONFIG_HIGHMEM
+			MLM(PKMAP_BASE, (PKMAP_BASE) + (LAST_PKMAP) *
+				(PAGE_SIZE)),
+#endif
+			MLM(MODULES_VADDR, MODULES_END),
+
+			MLK_ROUNDUP(__init_begin, __init_end),
+			MLK_ROUNDUP(_text, _etext),
+			MLK_ROUNDUP(_data, _edata));
+
+#undef MLK
+#undef MLM
+#undef MLK_ROUNDUP
+
 	if (PAGE_SIZE >= 16384 && num_physpages <= 128) {
 		extern int sysctl_overcommit_memory;
 		/*
-- 
cgit v1.2.2


From a7bd08c82e4f74387a39eeebb942712f23967420 Mon Sep 17 00:00:00 2001
From: Fenkart/Bostandzhyan <andreas.fenkart@streamunlimited.com>
Date: Sun, 7 Feb 2010 21:46:33 +0100
Subject: ARM: 5927/1: Make delimiters of DMA area globally visibly.

Adds DMA area to 'virtual memory map' startup message

Tested-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Andreas Fenkart <andreas.fenkart@streamunlimited.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/dma-mapping.c | 3 ---
 arch/arm/mm/init.c        | 6 ++++++
 2 files changed, 6 insertions(+), 3 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c
index 26325cb5d368..48eedab1609b 100644
--- a/arch/arm/mm/dma-mapping.c
+++ b/arch/arm/mm/dma-mapping.c
@@ -29,9 +29,6 @@
 #error "CONSISTENT_DMA_SIZE must be multiple of 2MiB"
 #endif
 
-#define CONSISTENT_END	(0xffe00000)
-#define CONSISTENT_BASE	(CONSISTENT_END - CONSISTENT_DMA_SIZE)
-
 #define CONSISTENT_OFFSET(x)	(((unsigned long)(x) - CONSISTENT_BASE) >> PAGE_SHIFT)
 #define CONSISTENT_PTE_INDEX(x) (((unsigned long)(x) - CONSISTENT_BASE) >> PGDIR_SHIFT)
 #define NUM_CONSISTENT_PTES (CONSISTENT_DMA_SIZE >> PGDIR_SHIFT)
diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index e8e3a74ac5b5..bda481e6bc0f 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -651,6 +651,9 @@ void __init mem_init(void)
 	printk(KERN_NOTICE "Virtual kernel memory layout:\n"
 			"    vector  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
 			"    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
+#ifdef CONFIG_MMU
+			"    DMA     : 0x%08lx - 0x%08lx   (%4ld MB)\n"
+#endif
 			"    vmalloc : 0x%08lx - 0x%08lx   (%4ld MB)\n"
 			"    lowmem  : 0x%08lx - 0x%08lx   (%4ld MB)\n"
 #ifdef CONFIG_HIGHMEM
@@ -664,6 +667,9 @@ void __init mem_init(void)
 			MLK(UL(CONFIG_VECTORS_BASE), UL(CONFIG_VECTORS_BASE) +
 				(PAGE_SIZE)),
 			MLK(FIXADDR_START, FIXADDR_TOP),
+#ifdef CONFIG_MMU
+			MLM(CONSISTENT_BASE, CONSISTENT_END),
+#endif
 			MLM(VMALLOC_START, (unsigned long)VMALLOC_END),
 			MLM(PAGE_OFFSET, (unsigned long)high_memory),
 #ifdef CONFIG_HIGHMEM
-- 
cgit v1.2.2


From c931b4f655a1b86c929384e674eb8c31795f3bd7 Mon Sep 17 00:00:00 2001
From: Fenkart/Bostandzhyan <andreas.fenkart@streamunlimited.com>
Date: Sun, 7 Feb 2010 21:47:17 +0100
Subject: ARM: 5928/1: Change type of VMALLOC_END to unsigned long.

Makes it consistent with VMALLOC_START

Tested-by: H Hartley Sweeten <hsweeten@visionengravers.com>
Signed-off-by: Andreas Fenkart <andreas.fenkart@streamunlimited.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index bda481e6bc0f..3a2077239474 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -670,7 +670,7 @@ void __init mem_init(void)
 #ifdef CONFIG_MMU
 			MLM(CONSISTENT_BASE, CONSISTENT_END),
 #endif
-			MLM(VMALLOC_START, (unsigned long)VMALLOC_END),
+			MLM(VMALLOC_START, VMALLOC_END),
 			MLM(PAGE_OFFSET, (unsigned long)high_memory),
 #ifdef CONFIG_HIGHMEM
 			MLM(PKMAP_BASE, (PKMAP_BASE) + (LAST_PKMAP) *
-- 
cgit v1.2.2


From a183927213df225bd93d21857b6aaafbb95e590d Mon Sep 17 00:00:00 2001
From: Fenkart/Bostandzhyan <andreas.fenkart@streamunlimited.com>
Date: Sun, 7 Feb 2010 21:47:58 +0100
Subject: ARM: 5929/1: Add checks to detect overlap of memory regions.

Tested-by: H Hartley Sweeten <hsweeten@visionengravers.com>

Signed-off-by: Andreas Fenkart <andreas.fenkart@streamunlimited.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/init.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/init.c b/arch/arm/mm/init.c
index 3a2077239474..7829cb5425f5 100644
--- a/arch/arm/mm/init.c
+++ b/arch/arm/mm/init.c
@@ -686,6 +686,23 @@ void __init mem_init(void)
 #undef MLM
 #undef MLK_ROUNDUP
 
+	/*
+	 * Check boundaries twice: Some fundamental inconsistencies can
+	 * be detected at build time already.
+	 */
+#ifdef CONFIG_MMU
+	BUILD_BUG_ON(VMALLOC_END			> CONSISTENT_BASE);
+	BUG_ON(VMALLOC_END				> CONSISTENT_BASE);
+
+	BUILD_BUG_ON(TASK_SIZE				> MODULES_VADDR);
+	BUG_ON(TASK_SIZE 				> MODULES_VADDR);
+#endif
+
+#ifdef CONFIG_HIGHMEM
+	BUILD_BUG_ON(PKMAP_BASE + LAST_PKMAP * PAGE_SIZE > PAGE_OFFSET);
+	BUG_ON(PKMAP_BASE + LAST_PKMAP * PAGE_SIZE	> PAGE_OFFSET);
+#endif
+
 	if (PAGE_SIZE >= 16384 && num_physpages <= 128) {
 		extern int sysctl_overcommit_memory;
 		/*
-- 
cgit v1.2.2


From 4b3073e1c53a256275f1079c0fbfbe85883d9275 Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 18 Dec 2009 16:40:18 +0000
Subject: MM: Pass a PTE pointer to update_mmu_cache() rather than the PTE
 itself

On VIVT ARM, when we have multiple shared mappings of the same file
in the same MM, we need to ensure that we have coherency across all
copies.  We do this via make_coherent() by making the pages
uncacheable.

This used to work fine, until we allowed highmem with highpte - we
now have a page table which is mapped as required, and is not available
for modification via update_mmu_cache().

Ralf Beache suggested getting rid of the PTE value passed to
update_mmu_cache():

  On MIPS update_mmu_cache() calls __update_tlb() which walks pagetables
  to construct a pointer to the pte again.  Passing a pte_t * is much
  more elegant.  Maybe we might even replace the pte argument with the
  pte_t?

Ben Herrenschmidt would also like the pte pointer for PowerPC:

  Passing the ptep in there is exactly what I want.  I want that
  -instead- of the PTE value, because I have issue on some ppc cases,
  for I$/D$ coherency, where set_pte_at() may decide to mask out the
  _PAGE_EXEC.

So, pass in the mapped page table pointer into update_mmu_cache(), and
remove the PTE value, updating all implementations and call sites to
suit.

Includes a fix from Stephen Rothwell:

  sparc: fix fallout from update_mmu_cache API change

  Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>

Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/fault-armv.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index ae88f2c3a6df..c45f9bb318ad 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -149,9 +149,10 @@ make_coherent(struct address_space *mapping, struct vm_area_struct *vma, unsigne
  *
  * Note that the pte lock will be held.
  */
-void update_mmu_cache(struct vm_area_struct *vma, unsigned long addr, pte_t pte)
+void update_mmu_cache(struct vm_area_struct *vma, unsigned long addr,
+	pte_t *ptep)
 {
-	unsigned long pfn = pte_pfn(pte);
+	unsigned long pfn = pte_pfn(*ptep);
 	struct address_space *mapping;
 	struct page *page;
 
-- 
cgit v1.2.2


From ae1402022edbeef3991f1e4bae8fa99558be291b Mon Sep 17 00:00:00 2001
From: Russell King <rmk+kernel@arm.linux.org.uk>
Date: Fri, 18 Dec 2009 16:43:57 +0000
Subject: ARM: make_coherent(): fix problems with highpte, part 2

update_mmu_cache() is called with the page table for the faulted-in
page still mapped.  We need to modify the PTE for this page to ensure
coherency with other shared mappings when multiple shared mappings
exist within a MM.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/fault-armv.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index c45f9bb318ad..c9b97e9836a2 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -99,7 +99,8 @@ static int adjust_pte(struct vm_area_struct *vma, unsigned long address,
 }
 
 static void
-make_coherent(struct address_space *mapping, struct vm_area_struct *vma, unsigned long addr, unsigned long pfn)
+make_coherent(struct address_space *mapping, struct vm_area_struct *vma,
+	unsigned long addr, pte_t *ptep, unsigned long pfn)
 {
 	struct mm_struct *mm = vma->vm_mm;
 	struct vm_area_struct *mpnt;
@@ -131,7 +132,7 @@ make_coherent(struct address_space *mapping, struct vm_area_struct *vma, unsigne
 	}
 	flush_dcache_mmap_unlock(mapping);
 	if (aliases)
-		adjust_pte(vma, addr, pfn);
+		do_adjust_pte(vma, addr, pfn, ptep);
 	else
 		flush_cache_page(vma, addr, pfn);
 }
@@ -174,7 +175,7 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long addr,
 #endif
 	if (mapping) {
 		if (cache_is_vivt())
-			make_coherent(mapping, vma, addr, pfn);
+			make_coherent(mapping, vma, addr, ptep, pfn);
 		else if (vma->vm_flags & VM_EXEC)
 			__flush_icache_all();
 	}
-- 
cgit v1.2.2


From d6d502fa4be1acd01971476fc732c95a4da16d90 Mon Sep 17 00:00:00 2001
From: Kukjin Kim <kgene.kim@samsung.com>
Date: Mon, 22 Feb 2010 00:02:59 +0100
Subject: ARM: 5952/1: ARM: MM: Add ARM_L1_CACHE_SHIFT_6 for handle inside each
 ARCH Kconfig

Add ARM_L1_CACHE_SHIFT_6 to arch/arm/Kconfig to allow CPUs with
L1 cache lines which are 64bytes to indicate this without having to
alter the arch/arm/mm/Kconfig entry each time.

Update the mm Kconfig so that ARM_L1_CACHE_SHIFT default value
uses this and change OMAP3 and S5PC1XX to select ARM_L1_CACHE_SHIFT_6.

Acked-by: Ben Dooks <ben-linux@fluff.org>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Kukjin Kim <kgene.kim@samsung.com>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mm/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

(limited to 'arch/arm/mm')

diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index e859743024a0..c4ed9f93f646 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -779,5 +779,5 @@ config CACHE_XSC3L2
 
 config ARM_L1_CACHE_SHIFT
 	int
-	default 6 if ARCH_OMAP3 || ARCH_S5PC1XX
+	default 6 if ARM_L1_CACHE_SHIFT_6
 	default 5
-- 
cgit v1.2.2