Merge branch 'for-linus' of git://ftp.arm.linux.org.uk/~rmk/linux-arm

Pull ARM updates from Russell King:
 "In this set, we have:
   - Refactoring of some of the old StrongARM-1100 GPIO code to make
     things simpler by Dmitry Eremin-Solenikov
   - Read-only and non-executable support for modules on ARM from Laura
     Abbot
   - Removal of unnecessary set_drvdata() calls in AMBA code
   - Some non-executable support for kernel lowmem mappings at the 1MB
     section granularity, and dumping of kernel page tables via debugfs
   - Some improvements for the timer/clock code on Footbridge platforms,
     and cleanup some of the LED code there
   - Fix fls/ffs() signatures to match x86 to prevent build warnings,
     particularly where these are used with min/max() macros
   - Avoid using the bootmem allocator on ARM (patches from Santosh
     Shilimkar)
   - Various asid/unaligned access updates from Will Deacon"

* 'for-linus' of git://ftp.arm.linux.org.uk/~rmk/linux-arm: (51 commits)
  ARM: SMP implementations are not supposed to return from smp_ops.cpu_die()
  ARM: ignore memory below PHYS_OFFSET
  Fix select-induced Kconfig warning for ZBOOT_ROM
  ARM: fix ffs/fls implementations to match x86
  ARM: 7935/1: sa1100: collie: add gpio-keys configuration
  ARM: 7932/1: bcm: Add DEBUG_LL console support
  ARM: 7929/1: Remove duplicate SCHED_HRTICK config option
  ARM: 7928/1: kconfig: select HAVE_EFFICIENT_UNALIGNED_ACCESS for CPUv6+ && MMU
  ARM: 7927/1: dcache: select DCACHE_WORD_ACCESS for big-endian CPUs
  ARM: 7926/1: mm: flesh out and fix the comments in the ASID allocator
  ARM: 7925/1: mm: keep track of last ASID allocation to improve bitmap searching
  ARM: 7924/1: mm: don't bother with reserved ttbr0 when running with LPAE
  ARM: PCI: add legacy IDE IRQ implementation
  ARM: footbridge: cleanup LEDs code
  ARM: pgd allocation: retry on failure
  ARM: footbridge: add one-shot mode for DC21285 timer
  ARM: footbridge: add sched_clock implementation
  ARM: 7922/1: l2x0: add Marvell Tauros3 support
  ARM: 7877/1: use built-in byte swap function
  ARM: 7921/1: mcpm: remove redundant dsb instructions prior to sev
  ...

10 files changed, 104 insertions, 44 deletions

diff --git a/arch/arm/include/asm/bitops.h b/arch/arm/include/asm/bitops.h
index e691ec91e4d3..b2e298a90d76 100644
--- a/arch/arm/include/asm/bitops.h
+++ b/arch/arm/include/asm/bitops.h
@@ -254,25 +254,59 @@ static inline int constant_fls(int x)
 }
 
 /*
- * On ARMv5 and above those functions can be implemented around
- * the clz instruction for much better code efficiency.
+ * On ARMv5 and above those functions can be implemented around the
+ * clz instruction for much better code efficiency.  __clz returns
+ * the number of leading zeros, zero input will return 32, and
+ * 0x80000000 will return 0.
  */
+static inline unsigned int __clz(unsigned int x)
+{
+        unsigned int ret;
+
+        asm("clz\t%0, %1" : "=r" (ret) : "r" (x));
 
+        return ret;
+}
+
+/*
+ * fls() returns zero if the input is zero, otherwise returns the bit
+ * position of the last set bit, where the LSB is 1 and MSB is 32.
+ */
 static inline int fls(int x)
 {
-        int ret;
-
         if (__builtin_constant_p(x))
                return constant_fls(x);
 
-        asm("clz\t%0, %1" : "=r" (ret) : "r" (x));
-        ret = 32 - ret;
-        return ret;
+        return 32 - __clz(x);
+}
+
+/*
+ * __fls() returns the bit position of the last bit set, where the
+ * LSB is 0 and MSB is 31.  Zero input is undefined.
+ */
+static inline unsigned long __fls(unsigned long x)
+{
+        return fls(x) - 1;
+}
+
+/*
+ * ffs() returns zero if the input was zero, otherwise returns the bit
+ * position of the first set bit, where the LSB is 1 and MSB is 32.
+ */
+static inline int ffs(int x)
+{
+        return fls(x & -x);
+}
+
+/*
+ * __ffs() returns the bit position of the first bit set, where the
+ * LSB is 0 and MSB is 31.  Zero input is undefined.
+ */
+static inline unsigned long __ffs(unsigned long x)
+{
+        return ffs(x) - 1;
 }
 
-#define __fls(x) (fls(x) - 1)
-#define ffs(x) ({ unsigned long __t = (x); fls(__t & -__t); })
-#define __ffs(x) (ffs(x) - 1)
 #define ffz(x) __ffs( ~(x) )
 
 #endif
diff --git a/arch/arm/include/asm/cacheflush.h b/arch/arm/include/asm/cacheflush.h
index ee753f1749cd..e9a49fe0284e 100644
--- a/arch/arm/include/asm/cacheflush.h
+++ b/arch/arm/include/asm/cacheflush.h
@@ -481,4 +481,9 @@ static inline void __sync_cache_range_r(volatile void *p, size_t size)
         : : : "r0","r1","r2","r3","r4","r5","r6","r7", \
               "r9","r10","lr","memory" )
 
+int set_memory_ro(unsigned long addr, int numpages);
+int set_memory_rw(unsigned long addr, int numpages);
+int set_memory_x(unsigned long addr, int numpages);
+int set_memory_nx(unsigned long addr, int numpages);
+
 #endif
diff --git a/arch/arm/include/asm/checksum.h b/arch/arm/include/asm/checksum.h
index 6dcc16430868..523315115478 100644
--- a/arch/arm/include/asm/checksum.h
+++ b/arch/arm/include/asm/checksum.h
@@ -87,19 +87,33 @@ static inline __wsum
 csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
                    unsigned short proto, __wsum sum)
 {
-        __asm__(
-        "adds   %0, %1, %2              @ csum_tcpudp_nofold    \n\
-        adcs    %0, %0, %3                                      \n"
+        u32 lenprot = len | proto << 16;
+        if (__builtin_constant_p(sum) && sum == 0) {
+                __asm__(
+                "adds   %0, %1, %2      @ csum_tcpudp_nofold0   \n\t"
 #ifdef __ARMEB__
-        "adcs   %0, %0, %4                                      \n"
+                "adcs   %0, %0, %3                              \n\t"
 #else
-        "adcs   %0, %0, %4, lsl #8                              \n"
+                "adcs   %0, %0, %3, ror #8                      \n\t"
 #endif
-        "adcs   %0, %0, %5                                      \n\
-        adc     %0, %0, #0"
-        : "=&r"(sum)
-        : "r" (sum), "r" (daddr), "r" (saddr), "r" (len), "Ir" (htons(proto))
-        : "cc");
+                "adc    %0, %0, #0"
+                : "=&r" (sum)
+                : "r" (daddr), "r" (saddr), "r" (lenprot)
+                : "cc");
+        } else {
+                __asm__(
+                "adds   %0, %1, %2      @ csum_tcpudp_nofold    \n\t"
+                "adcs   %0, %0, %3                              \n\t"
+#ifdef __ARMEB__
+                "adcs   %0, %0, %4                              \n\t"
+#else
+                "adcs   %0, %0, %4, ror #8                      \n\t"
+#endif
+                "adc    %0, %0, #0"
+                : "=&r"(sum)
+                : "r" (sum), "r" (daddr), "r" (saddr), "r" (lenprot)
+                : "cc");
+        }
         return sum;
 }       
 /*
diff --git a/arch/arm/include/asm/hardware/cache-l2x0.h b/arch/arm/include/asm/hardware/cache-l2x0.h
index 3b2c40b5bfa2..6795ff743b3d 100644
--- a/arch/arm/include/asm/hardware/cache-l2x0.h
+++ b/arch/arm/include/asm/hardware/cache-l2x0.h
@@ -131,6 +131,7 @@ struct l2x0_regs {
         unsigned long prefetch_ctrl;
         unsigned long pwr_ctrl;
         unsigned long ctrl;
+        unsigned long aux2_ctrl;
 };
 
 extern struct l2x0_regs l2x0_saved_regs;
diff --git a/arch/arm/include/asm/mach/map.h b/arch/arm/include/asm/mach/map.h
index 2fe141fcc8d6..f98c7f32c9c8 100644
--- a/arch/arm/include/asm/mach/map.h
+++ b/arch/arm/include/asm/mach/map.h
@@ -22,18 +22,21 @@ struct map_desc {
 };
 
 /* types 0-3 are defined in asm/io.h */
-#define MT_UNCACHED             4
-#define MT_CACHECLEAN           5
-#define MT_MINICLEAN            6
-#define MT_LOW_VECTORS          7
-#define MT_HIGH_VECTORS         8
-#define MT_MEMORY               9
-#define MT_ROM                  10
-#define MT_MEMORY_NONCACHED     11
-#define MT_MEMORY_DTCM          12
-#define MT_MEMORY_ITCM          13
-#define MT_MEMORY_SO            14
-#define MT_MEMORY_DMA_READY     15
+enum {
+        MT_UNCACHED = 4,
+        MT_CACHECLEAN,
+        MT_MINICLEAN,
+        MT_LOW_VECTORS,
+        MT_HIGH_VECTORS,
+        MT_MEMORY_RWX,
+        MT_MEMORY_RW,
+        MT_ROM,
+        MT_MEMORY_RWX_NONCACHED,
+        MT_MEMORY_RW_DTCM,
+        MT_MEMORY_RWX_ITCM,
+        MT_MEMORY_RW_SO,
+        MT_MEMORY_DMA_READY,
+};
 
 #ifdef CONFIG_MMU
 extern void iotable_init(struct map_desc *, int);
diff --git a/arch/arm/include/asm/pci.h b/arch/arm/include/asm/pci.h
index a98a2e112fae..680a83e94467 100644
--- a/arch/arm/include/asm/pci.h
+++ b/arch/arm/include/asm/pci.h
@@ -57,12 +57,9 @@ static inline void pci_dma_burst_advice(struct pci_dev *pdev,
 extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
                                enum pci_mmap_state mmap_state, int write_combine);
 
-/*
- * Dummy implementation; always return 0.
- */
 static inline int pci_get_legacy_ide_irq(struct pci_dev *dev, int channel)
 {
-        return 0;
+        return channel ? 15 : 14;
 }
 
 #endif /* __KERNEL__ */
diff --git a/arch/arm/include/asm/pgtable-2level.h b/arch/arm/include/asm/pgtable-2level.h
index 86a659a19526..dfff709fda3c 100644
--- a/arch/arm/include/asm/pgtable-2level.h
+++ b/arch/arm/include/asm/pgtable-2level.h
@@ -160,6 +160,7 @@ static inline pmd_t *pmd_offset(pud_t *pud, unsigned long addr)
         return (pmd_t *)pud;
 }
 
+#define pmd_large(pmd)          (pmd_val(pmd) & 2)
 #define pmd_bad(pmd)            (pmd_val(pmd) & 2)
 
 #define copy_pmd(pmdpd,pmdps)           \
diff --git a/arch/arm/include/asm/pgtable-3level.h b/arch/arm/include/asm/pgtable-3level.h
index 4f9503908dca..03243f7eeddf 100644
--- a/arch/arm/include/asm/pgtable-3level.h
+++ b/arch/arm/include/asm/pgtable-3level.h
@@ -142,6 +142,7 @@
                                                  PMD_TYPE_TABLE)
 #define pmd_sect(pmd)           ((pmd_val(pmd) & PMD_TYPE_MASK) == \
                                                  PMD_TYPE_SECT)
+#define pmd_large(pmd)          pmd_sect(pmd)
 
 #define pud_clear(pudp)                 \
         do {                            \
diff --git a/arch/arm/include/asm/pgtable.h b/arch/arm/include/asm/pgtable.h
index 1571d126e9dd..7d59b524f2af 100644
--- a/arch/arm/include/asm/pgtable.h
+++ b/arch/arm/include/asm/pgtable.h
@@ -254,6 +254,8 @@ PTE_BIT_FUNC(mkclean,   &= ~L_PTE_DIRTY);
 PTE_BIT_FUNC(mkdirty,   |= L_PTE_DIRTY);
 PTE_BIT_FUNC(mkold,     &= ~L_PTE_YOUNG);
 PTE_BIT_FUNC(mkyoung,   |= L_PTE_YOUNG);
+PTE_BIT_FUNC(mkexec,   &= ~L_PTE_XN);
+PTE_BIT_FUNC(mknexec,   |= L_PTE_XN);
 
 static inline pte_t pte_mkspecial(pte_t pte) { return pte; }
 
diff --git a/arch/arm/include/asm/word-at-a-time.h b/arch/arm/include/asm/word-at-a-time.h
index 4d52f92967a6..a6d0a29861e7 100644
--- a/arch/arm/include/asm/word-at-a-time.h
+++ b/arch/arm/include/asm/word-at-a-time.h
@@ -48,10 +48,14 @@ static inline unsigned long find_zero(unsigned long mask)
         return ret;
 }
 
-#ifdef CONFIG_DCACHE_WORD_ACCESS
-
 #define zero_bytemask(mask) (mask)
 
+#else   /* __ARMEB__ */
+#include <asm-generic/word-at-a-time.h>
+#endif
+
+#ifdef CONFIG_DCACHE_WORD_ACCESS
+
 /*
  * Load an unaligned word from kernel space.
  *
@@ -73,7 +77,11 @@ static inline unsigned long load_unaligned_zeropad(const void *addr)
         "       bic     %2, %2, #0x3\n"
         "       ldr     %0, [%2]\n"
         "       lsl     %1, %1, #0x3\n"
+#ifndef __ARMEB__
         "       lsr     %0, %0, %1\n"
+#else
+        "       lsl     %0, %0, %1\n"
+#endif
         "       b       2b\n"
         "       .popsection\n"
         "       .pushsection __ex_table,\"a\"\n"
@@ -86,11 +94,5 @@ static inline unsigned long load_unaligned_zeropad(const void *addr)
         return ret;
 }
 
-
 #endif  /* DCACHE_WORD_ACCESS */
-
-#else   /* __ARMEB__ */
-#include <asm-generic/word-at-a-time.h>
-#endif
-
 #endif /* __ASM_ARM_WORD_AT_A_TIME_H */