10 files changed, 563 insertions, 34 deletions

diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 33ed048502a3..a2c8b006d710 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -180,8 +180,8 @@ config CPU_ARM925T
 # ARM926T
 config CPU_ARM926T
         bool "Support ARM926T processor"
-        depends on ARCH_INTEGRATOR || ARCH_VERSATILE_PB || MACH_VERSATILE_AB || ARCH_OMAP730 || ARCH_OMAP16XX || MACH_REALVIEW_EB || ARCH_PNX4008 || ARCH_NETX || CPU_S3C2412 || ARCH_AT91SAM9260 || ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91CAP9 || ARCH_NS9XXX || ARCH_DAVINCI
-        default y if ARCH_VERSATILE_PB || MACH_VERSATILE_AB || ARCH_OMAP730 || ARCH_OMAP16XX || ARCH_PNX4008 || ARCH_NETX || CPU_S3C2412 || ARCH_AT91SAM9260 || ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91CAP9 || ARCH_NS9XXX || ARCH_DAVINCI
+        depends on ARCH_INTEGRATOR || ARCH_VERSATILE_PB || MACH_VERSATILE_AB || ARCH_OMAP730 || ARCH_OMAP16XX || MACH_REALVIEW_EB || ARCH_PNX4008 || ARCH_NETX || CPU_S3C2412 || ARCH_AT91SAM9260 || ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91SAM9G20 || ARCH_AT91CAP9 || ARCH_NS9XXX || ARCH_DAVINCI
+        default y if ARCH_VERSATILE_PB || MACH_VERSATILE_AB || ARCH_OMAP730 || ARCH_OMAP16XX || ARCH_PNX4008 || ARCH_NETX || CPU_S3C2412 || ARCH_AT91SAM9260 || ARCH_AT91SAM9261 || ARCH_AT91SAM9263 || ARCH_AT91SAM9RL || ARCH_AT91SAM9G20 || ARCH_AT91CAP9 || ARCH_NS9XXX || ARCH_DAVINCI
         select CPU_32v5
         select CPU_ABRT_EV5TJ
         select CPU_PABRT_NOIFAR
@@ -365,7 +365,7 @@ config CPU_XSC3
 # Feroceon
 config CPU_FEROCEON
         bool
-        depends on ARCH_ORION5X
+        depends on ARCH_ORION5X || ARCH_LOKI || ARCH_KIRKWOOD || ARCH_MV78XX0
         default y
         select CPU_32v5
         select CPU_ABRT_EV5T
@@ -373,7 +373,7 @@ config CPU_FEROCEON
         select CPU_CACHE_VIVT
         select CPU_CP15_MMU
         select CPU_COPY_FEROCEON if MMU
-        select CPU_TLB_V4WBI if MMU
+        select CPU_TLB_FEROCEON if MMU
 
 config CPU_FEROCEON_OLD_ID
         bool "Accept early Feroceon cores with an ARM926 ID"
@@ -551,6 +551,11 @@ config CPU_TLB_V4WBI
           ARM Architecture Version 4 TLB with writeback cache and invalidate
           instruction cache entry.
 
+config CPU_TLB_FEROCEON
+        bool
+        help
+          Feroceon TLB (v4wbi with non-outer-cachable page table walks).
+
 config CPU_TLB_V6
         bool
 
@@ -709,6 +714,14 @@ config OUTER_CACHE
         bool
         default n
 
+config CACHE_FEROCEON_L2
+        bool "Enable the Feroceon L2 cache controller"
+        depends on ARCH_KIRKWOOD || ARCH_MV78XX0
+        default y
+        select OUTER_CACHE
+        help
+          This option enables the Feroceon L2 cache controller.
+
 config CACHE_L2X0
         bool "Enable the L2x0 outer cache controller"
         depends on REALVIEW_EB_ARM11MP || MACH_REALVIEW_PB11MP || MACH_REALVIEW_PB1176
diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile
index 32b2d2d213a6..f64b92557b11 100644
--- a/arch/arm/mm/Makefile
+++ b/arch/arm/mm/Makefile
@@ -46,6 +46,7 @@ obj-$(CONFIG_CPU_TLB_V3)	+= tlb-v3.o
 obj-$(CONFIG_CPU_TLB_V4WT)      += tlb-v4.o
 obj-$(CONFIG_CPU_TLB_V4WB)      += tlb-v4wb.o
 obj-$(CONFIG_CPU_TLB_V4WBI)     += tlb-v4wbi.o
+obj-$(CONFIG_CPU_TLB_FEROCEON)  += tlb-v4wbi.o  # reuse v4wbi TLB functions
 obj-$(CONFIG_CPU_TLB_V6)        += tlb-v6.o
 obj-$(CONFIG_CPU_TLB_V7)        += tlb-v7.o
 
@@ -73,4 +74,5 @@ obj-$(CONFIG_CPU_FEROCEON)	+= proc-feroceon.o
 obj-$(CONFIG_CPU_V6)            += proc-v6.o
 obj-$(CONFIG_CPU_V7)            += proc-v7.o
 
+obj-$(CONFIG_CACHE_FEROCEON_L2) += cache-feroceon-l2.o
 obj-$(CONFIG_CACHE_L2X0)        += cache-l2x0.o
diff --git a/arch/arm/mm/cache-feroceon-l2.c b/arch/arm/mm/cache-feroceon-l2.c
new file mode 100644
index 000000000000..20eec4ba173f
--- /dev/null
+++ b/arch/arm/mm/cache-feroceon-l2.c
@@ -0,0 +1,318 @@
+/*
+ * arch/arm/mm/cache-feroceon-l2.c - Feroceon L2 cache controller support
+ *
+ * Copyright (C) 2008 Marvell Semiconductor
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2.  This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ *
+ * References:
+ * - Unified Layer 2 Cache for Feroceon CPU Cores,
+ *   Document ID MV-S104858-00, Rev. A, October 23 2007.
+ */
+
+#include <linux/init.h>
+#include <asm/cacheflush.h>
+#include <asm/plat-orion/cache-feroceon-l2.h>
+
+
+/*
+ * Low-level cache maintenance operations.
+ *
+ * As well as the regular 'clean/invalidate/flush L2 cache line by
+ * MVA' instructions, the Feroceon L2 cache controller also features
+ * 'clean/invalidate L2 range by MVA' operations.
+ *
+ * Cache range operations are initiated by writing the start and
+ * end addresses to successive cp15 registers, and process every
+ * cache line whose first byte address lies in the inclusive range
+ * [start:end].
+ *
+ * The cache range operations stall the CPU pipeline until completion.
+ *
+ * The range operations require two successive cp15 writes, in
+ * between which we don't want to be preempted.
+ */
+static inline void l2_clean_pa(unsigned long addr)
+{
+        __asm__("mcr p15, 1, %0, c15, c9, 3" : : "r" (addr));
+}
+
+static inline void l2_clean_mva_range(unsigned long start, unsigned long end)
+{
+        unsigned long flags;
+
+        /*
+         * Make sure 'start' and 'end' reference the same page, as
+         * L2 is PIPT and range operations only do a TLB lookup on
+         * the start address.
+         */
+        BUG_ON((start ^ end) & ~(PAGE_SIZE - 1));
+
+        raw_local_irq_save(flags);
+        __asm__("mcr p15, 1, %0, c15, c9, 4" : : "r" (start));
+        __asm__("mcr p15, 1, %0, c15, c9, 5" : : "r" (end));
+        raw_local_irq_restore(flags);
+}
+
+static inline void l2_clean_pa_range(unsigned long start, unsigned long end)
+{
+        l2_clean_mva_range(__phys_to_virt(start), __phys_to_virt(end));
+}
+
+static inline void l2_clean_inv_pa(unsigned long addr)
+{
+        __asm__("mcr p15, 1, %0, c15, c10, 3" : : "r" (addr));
+}
+
+static inline void l2_inv_pa(unsigned long addr)
+{
+        __asm__("mcr p15, 1, %0, c15, c11, 3" : : "r" (addr));
+}
+
+static inline void l2_inv_mva_range(unsigned long start, unsigned long end)
+{
+        unsigned long flags;
+
+        /*
+         * Make sure 'start' and 'end' reference the same page, as
+         * L2 is PIPT and range operations only do a TLB lookup on
+         * the start address.
+         */
+        BUG_ON((start ^ end) & ~(PAGE_SIZE - 1));
+
+        raw_local_irq_save(flags);
+        __asm__("mcr p15, 1, %0, c15, c11, 4" : : "r" (start));
+        __asm__("mcr p15, 1, %0, c15, c11, 5" : : "r" (end));
+        raw_local_irq_restore(flags);
+}
+
+static inline void l2_inv_pa_range(unsigned long start, unsigned long end)
+{
+        l2_inv_mva_range(__phys_to_virt(start), __phys_to_virt(end));
+}
+
+
+/*
+ * Linux primitives.
+ *
+ * Note that the end addresses passed to Linux primitives are
+ * noninclusive, while the hardware cache range operations use
+ * inclusive start and end addresses.
+ */
+#define CACHE_LINE_SIZE         32
+#define MAX_RANGE_SIZE          1024
+
+static int l2_wt_override;
+
+static unsigned long calc_range_end(unsigned long start, unsigned long end)
+{
+        unsigned long range_end;
+
+        BUG_ON(start & (CACHE_LINE_SIZE - 1));
+        BUG_ON(end & (CACHE_LINE_SIZE - 1));
+
+        /*
+         * Try to process all cache lines between 'start' and 'end'.
+         */
+        range_end = end;
+
+        /*
+         * Limit the number of cache lines processed at once,
+         * since cache range operations stall the CPU pipeline
+         * until completion.
+         */
+        if (range_end > start + MAX_RANGE_SIZE)
+                range_end = start + MAX_RANGE_SIZE;
+
+        /*
+         * Cache range operations can't straddle a page boundary.
+         */
+        if (range_end > (start | (PAGE_SIZE - 1)) + 1)
+                range_end = (start | (PAGE_SIZE - 1)) + 1;
+
+        return range_end;
+}
+
+static void feroceon_l2_inv_range(unsigned long start, unsigned long end)
+{
+        /*
+         * Clean and invalidate partial first cache line.
+         */
+        if (start & (CACHE_LINE_SIZE - 1)) {
+                l2_clean_inv_pa(start & ~(CACHE_LINE_SIZE - 1));
+                start = (start | (CACHE_LINE_SIZE - 1)) + 1;
+        }
+
+        /*
+         * Clean and invalidate partial last cache line.
+         */
+        if (end & (CACHE_LINE_SIZE - 1)) {
+                l2_clean_inv_pa(end & ~(CACHE_LINE_SIZE - 1));
+                end &= ~(CACHE_LINE_SIZE - 1);
+        }
+
+        /*
+         * Invalidate all full cache lines between 'start' and 'end'.
+         */
+        while (start != end) {
+                unsigned long range_end = calc_range_end(start, end);
+                l2_inv_pa_range(start, range_end - CACHE_LINE_SIZE);
+                start = range_end;
+        }
+
+        dsb();
+}
+
+static void feroceon_l2_clean_range(unsigned long start, unsigned long end)
+{
+        /*
+         * If L2 is forced to WT, the L2 will always be clean and we
+         * don't need to do anything here.
+         */
+        if (!l2_wt_override) {
+                start &= ~(CACHE_LINE_SIZE - 1);
+                end = (end + CACHE_LINE_SIZE - 1) & ~(CACHE_LINE_SIZE - 1);
+                while (start != end) {
+                        unsigned long range_end = calc_range_end(start, end);
+                        l2_clean_pa_range(start, range_end - CACHE_LINE_SIZE);
+                        start = range_end;
+                }
+        }
+
+        dsb();
+}
+
+static void feroceon_l2_flush_range(unsigned long start, unsigned long end)
+{
+        start &= ~(CACHE_LINE_SIZE - 1);
+        end = (end + CACHE_LINE_SIZE - 1) & ~(CACHE_LINE_SIZE - 1);
+        while (start != end) {
+                unsigned long range_end = calc_range_end(start, end);
+                if (!l2_wt_override)
+                        l2_clean_pa_range(start, range_end - CACHE_LINE_SIZE);
+                l2_inv_pa_range(start, range_end - CACHE_LINE_SIZE);
+                start = range_end;
+        }
+
+        dsb();
+}
+
+
+/*
+ * Routines to disable and re-enable the D-cache and I-cache at run
+ * time.  These are necessary because the L2 cache can only be enabled
+ * or disabled while the L1 Dcache and Icache are both disabled.
+ */
+static void __init invalidate_and_disable_dcache(void)
+{
+        u32 cr;
+
+        cr = get_cr();
+        if (cr & CR_C) {
+                unsigned long flags;
+
+                raw_local_irq_save(flags);
+                flush_cache_all();
+                set_cr(cr & ~CR_C);
+                raw_local_irq_restore(flags);
+        }
+}
+
+static void __init enable_dcache(void)
+{
+        u32 cr;
+
+        cr = get_cr();
+        if (!(cr & CR_C))
+                set_cr(cr | CR_C);
+}
+
+static void __init __invalidate_icache(void)
+{
+        int dummy;
+
+        __asm__ __volatile__("mcr p15, 0, %0, c7, c5, 0\n" : "=r" (dummy));
+}
+
+static void __init invalidate_and_disable_icache(void)
+{
+        u32 cr;
+
+        cr = get_cr();
+        if (cr & CR_I) {
+                set_cr(cr & ~CR_I);
+                __invalidate_icache();
+        }
+}
+
+static void __init enable_icache(void)
+{
+        u32 cr;
+
+        cr = get_cr();
+        if (!(cr & CR_I))
+                set_cr(cr | CR_I);
+}
+
+static inline u32 read_extra_features(void)
+{
+        u32 u;
+
+        __asm__("mrc p15, 1, %0, c15, c1, 0" : "=r" (u));
+
+        return u;
+}
+
+static inline void write_extra_features(u32 u)
+{
+        __asm__("mcr p15, 1, %0, c15, c1, 0" : : "r" (u));
+}
+
+static void __init disable_l2_prefetch(void)
+{
+        u32 u;
+
+        /*
+         * Read the CPU Extra Features register and verify that the
+         * Disable L2 Prefetch bit is set.
+         */
+        u = read_extra_features();
+        if (!(u & 0x01000000)) {
+                printk(KERN_INFO "Feroceon L2: Disabling L2 prefetch.\n");
+                write_extra_features(u | 0x01000000);
+        }
+}
+
+static void __init enable_l2(void)
+{
+        u32 u;
+
+        u = read_extra_features();
+        if (!(u & 0x00400000)) {
+                printk(KERN_INFO "Feroceon L2: Enabling L2\n");
+
+                invalidate_and_disable_dcache();
+                invalidate_and_disable_icache();
+                write_extra_features(u | 0x00400000);
+                enable_icache();
+                enable_dcache();
+        }
+}
+
+void __init feroceon_l2_init(int __l2_wt_override)
+{
+        l2_wt_override = __l2_wt_override;
+
+        disable_l2_prefetch();
+
+        outer_cache.inv_range = feroceon_l2_inv_range;
+        outer_cache.clean_range = feroceon_l2_clean_range;
+        outer_cache.flush_range = feroceon_l2_flush_range;
+
+        enable_l2();
+
+        printk(KERN_INFO "Feroceon L2: Cache support initialised%s.\n",
+                         l2_wt_override ? ", in WT override mode" : "");
+}
diff --git a/arch/arm/mm/fault-armv.c b/arch/arm/mm/fault-armv.c
index 44558d5f9313..fbfa26058442 100644
--- a/arch/arm/mm/fault-armv.c
+++ b/arch/arm/mm/fault-armv.c
@@ -144,13 +144,17 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long addr, pte_t pte)
         page = pfn_to_page(pfn);
         mapping = page_mapping(page);
         if (mapping) {
+#ifndef CONFIG_SMP
                 int dirty = test_and_clear_bit(PG_dcache_dirty, &page->flags);
 
                 if (dirty)
                         __flush_dcache_page(mapping, page);
+#endif
 
                 if (cache_is_vivt())
                         make_coherent(mapping, vma, addr, pfn);
+                else if (vma->vm_flags & VM_EXEC)
+                        __flush_icache_all();
         }
 }
 
diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c
index 9df507d36e0b..029ee65fda2b 100644
--- a/arch/arm/mm/flush.c
+++ b/arch/arm/mm/flush.c
@@ -199,6 +199,8 @@ void flush_dcache_page(struct page *page)
                 __flush_dcache_page(mapping, page);
                 if (mapping && cache_is_vivt())
                         __flush_dcache_aliases(mapping, page);
+                else if (mapping)
+                        __flush_icache_all();
         }
 }
 EXPORT_SYMBOL(flush_dcache_page);
diff --git a/arch/arm/mm/proc-arm925.S b/arch/arm/mm/proc-arm925.S
index 065087afb772..d045812f3399 100644
--- a/arch/arm/mm/proc-arm925.S
+++ b/arch/arm/mm/proc-arm925.S
@@ -332,7 +332,7 @@ ENTRY(arm925_dma_flush_range)
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
         mcr     p15, 0, r0, c7, c14, 1          @ clean+invalidate D entry
 #else
-        mcr     p15, 0, r0, c7, c10, 1          @ clean D entry
+        mcr     p15, 0, r0, c7, c6, 1           @ invalidate D entry
 #endif
         add     r0, r0, #CACHE_DLINESIZE
         cmp     r0, r1
diff --git a/arch/arm/mm/proc-arm926.S b/arch/arm/mm/proc-arm926.S
index 997db8472b5c..4cd33169a7c9 100644
--- a/arch/arm/mm/proc-arm926.S
+++ b/arch/arm/mm/proc-arm926.S
@@ -295,7 +295,7 @@ ENTRY(arm926_dma_flush_range)
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
         mcr     p15, 0, r0, c7, c14, 1          @ clean+invalidate D entry
 #else
-        mcr     p15, 0, r0, c7, c10, 1          @ clean D entry
+        mcr     p15, 0, r0, c7, c6, 1           @ invalidate D entry
 #endif
         add     r0, r0, #CACHE_DLINESIZE
         cmp     r0, r1
diff --git a/arch/arm/mm/proc-arm940.S b/arch/arm/mm/proc-arm940.S
index 44ead902bd54..1a3d63df8e90 100644
--- a/arch/arm/mm/proc-arm940.S
+++ b/arch/arm/mm/proc-arm940.S
@@ -222,7 +222,7 @@ ENTRY(arm940_dma_flush_range)
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
         mcr     p15, 0, r3, c7, c14, 2          @ clean/flush D entry
 #else
-        mcr     p15, 0, r3, c7, c10, 2          @ clean D entry
+        mcr     p15, 0, r3, c7, c6, 2           @ invalidate D entry
 #endif
         subs    r3, r3, #1 << 26
         bcs     2b                              @ entries 63 to 0
diff --git a/arch/arm/mm/proc-arm946.S b/arch/arm/mm/proc-arm946.S
index 2218b0c01330..82d579ac9b98 100644
--- a/arch/arm/mm/proc-arm946.S
+++ b/arch/arm/mm/proc-arm946.S
@@ -265,7 +265,7 @@ ENTRY(arm946_dma_flush_range)
 #ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
         mcr     p15, 0, r0, c7, c14, 1          @ clean+invalidate D entry
 #else
-        mcr     p15, 0, r0, c7, c10, 1          @ clean D entry
+        mcr     p15, 0, r0, c7, c6, 1           @ invalidate D entry
 #endif
         add     r0, r0, #CACHE_DLINESIZE
         cmp     r0, r1
diff --git a/arch/arm/mm/proc-feroceon.S b/arch/arm/mm/proc-feroceon.S
index a02c1712b52d..f2e5884c513a 100644
--- a/arch/arm/mm/proc-feroceon.S
+++ b/arch/arm/mm/proc-feroceon.S
@@ -44,11 +44,31 @@
  */
 #define CACHE_DLINESIZE 32
 
+        .bss
+        .align 3
+__cache_params_loc:
+        .space  8
+
         .text
+__cache_params:
+        .word   __cache_params_loc
+
 /*
  * cpu_feroceon_proc_init()
  */
 ENTRY(cpu_feroceon_proc_init)
+        mrc     p15, 0, r0, c0, c0, 1           @ read cache type register
+        ldr     r1, __cache_params
+        mov     r2, #(16 << 5)
+        tst     r0, #(1 << 16)                  @ get way
+        mov     r0, r0, lsr #18                 @ get cache size order
+        movne   r3, #((4 - 1) << 30)            @ 4-way
+        and     r0, r0, #0xf
+        moveq   r3, #0                          @ 1-way
+        mov     r2, r2, lsl r0                  @ actual cache size
+        movne   r2, r2, lsr #2                  @ turned into # of sets
+        sub     r2, r2, #(1 << 5)
+        stmia   r1, {r2, r3}
         mov     pc, lr
 
 /*
@@ -59,6 +79,13 @@ ENTRY(cpu_feroceon_proc_fin)
         mov     ip, #PSR_F_BIT | PSR_I_BIT | SVC_MODE
         msr     cpsr_c, ip
         bl      feroceon_flush_kern_cache_all
+
+#if defined(CONFIG_CACHE_FEROCEON_L2) && !defined(CONFIG_L2_CACHE_WRITETHROUGH)
+        mov     r0, #0
+        mcr     p15, 1, r0, c15, c9, 0          @ clean L2
+        mcr     p15, 0, r0, c7, c10, 4          @ drain WB
+#endif
+
         mrc     p15, 0, r0, c1, c0, 0           @ ctrl register
         bic     r0, r0, #0x1000                 @ ...i............
         bic     r0, r0, #0x000e                 @ ............wca.
@@ -117,11 +144,19 @@ ENTRY(feroceon_flush_user_cache_all)
  */
 ENTRY(feroceon_flush_kern_cache_all)
         mov     r2, #VM_EXEC
-        mov     ip, #0
+
 __flush_whole_cache:
-1:      mrc     p15, 0, r15, c7, c14, 3         @ test,clean,invalidate
-        bne     1b
+        ldr     r1, __cache_params
+        ldmia   r1, {r1, r3}
+1:      orr     ip, r1, r3
+2:      mcr     p15, 0, ip, c7, c14, 2          @ clean + invalidate D set/way
+        subs    ip, ip, #(1 << 30)              @ next way
+        bcs     2b
+        subs    r1, r1, #(1 << 5)               @ next set
+        bcs     1b
+
         tst     r2, #VM_EXEC
+        mov     ip, #0
         mcrne   p15, 0, ip, c7, c5, 0           @ invalidate I cache
         mcrne   p15, 0, ip, c7, c10, 4          @ drain WB
         mov     pc, lr
@@ -138,7 +173,6 @@ __flush_whole_cache:
  */
         .align  5
 ENTRY(feroceon_flush_user_cache_range)
-        mov     ip, #0
         sub     r3, r1, r0                      @ calculate total size
         cmp     r3, #CACHE_DLIMIT
         bgt     __flush_whole_cache
@@ -152,6 +186,7 @@ ENTRY(feroceon_flush_user_cache_range)
         cmp     r0, r1
         blo     1b
         tst     r2, #VM_EXEC
+        mov     ip, #0
         mcrne   p15, 0, ip, c7, c10, 4          @ drain WB
         mov     pc, lr
 
@@ -209,6 +244,20 @@ ENTRY(feroceon_flush_kern_dcache_page)
         mcr     p15, 0, r0, c7, c10, 4          @ drain WB
         mov     pc, lr
 
+        .align  5
+ENTRY(feroceon_range_flush_kern_dcache_page)
+        mrs     r2, cpsr
+        add     r1, r0, #PAGE_SZ - CACHE_DLINESIZE      @ top addr is inclusive
+        orr     r3, r2, #PSR_I_BIT
+        msr     cpsr_c, r3                      @ disable interrupts
+        mcr     p15, 5, r0, c15, c15, 0         @ D clean/inv range start
+        mcr     p15, 5, r1, c15, c15, 1         @ D clean/inv range top
+        msr     cpsr_c, r2                      @ restore interrupts
+        mov     r0, #0
+        mcr     p15, 0, r0, c7, c5, 0           @ invalidate I cache
+        mcr     p15, 0, r0, c7, c10, 4          @ drain WB
+        mov     pc, lr
+
 /*
  *      dma_inv_range(start, end)
  *
@@ -225,10 +274,10 @@ ENTRY(feroceon_flush_kern_dcache_page)
         .align  5
 ENTRY(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
         tst     r1, #CACHE_DLINESIZE - 1
         mcrne   p15, 0, r1, c7, c10, 1          @ clean D entry
-        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
@@ -236,6 +285,22 @@ ENTRY(feroceon_dma_inv_range)
         mcr     p15, 0, r0, c7, c10, 4          @ drain WB
         mov     pc, lr
 
+        .align  5
+ENTRY(feroceon_range_dma_inv_range)
+        mrs     r2, cpsr
+        tst     r0, #CACHE_DLINESIZE - 1
+        mcrne   p15, 0, r0, c7, c10, 1          @ clean D entry
+        tst     r1, #CACHE_DLINESIZE - 1
+        mcrne   p15, 0, r1, c7, c10, 1          @ clean D entry
+        cmp     r1, r0
+        subne   r1, r1, #1                      @ top address is inclusive
+        orr     r3, r2, #PSR_I_BIT
+        msr     cpsr_c, r3                      @ disable interrupts
+        mcr     p15, 5, r0, c15, c14, 0         @ D inv range start
+        mcr     p15, 5, r1, c15, c14, 1         @ D inv range top
+        msr     cpsr_c, r2                      @ restore interrupts
+        mov     pc, lr
+
 /*
  *      dma_clean_range(start, end)
  *
@@ -256,6 +321,19 @@ ENTRY(feroceon_dma_clean_range)
         mcr     p15, 0, r0, c7, c10, 4          @ drain WB
         mov     pc, lr
 
+        .align  5
+ENTRY(feroceon_range_dma_clean_range)
+        mrs     r2, cpsr
+        cmp     r1, r0
+        subne   r1, r1, #1                      @ top address is inclusive
+        orr     r3, r2, #PSR_I_BIT
+        msr     cpsr_c, r3                      @ disable interrupts
+        mcr     p15, 5, r0, c15, c13, 0         @ D clean range start
+        mcr     p15, 5, r1, c15, c13, 1         @ D clean range top
+        msr     cpsr_c, r2                      @ restore interrupts
+        mcr     p15, 0, r0, c7, c10, 4          @ drain WB
+        mov     pc, lr
+
 /*
  *      dma_flush_range(start, end)
  *
@@ -274,6 +352,19 @@ ENTRY(feroceon_dma_flush_range)
         mcr     p15, 0, r0, c7, c10, 4          @ drain WB
         mov     pc, lr
 
+        .align  5
+ENTRY(feroceon_range_dma_flush_range)
+        mrs     r2, cpsr
+        cmp     r1, r0
+        subne   r1, r1, #1                      @ top address is inclusive
+        orr     r3, r2, #PSR_I_BIT
+        msr     cpsr_c, r3                      @ disable interrupts
+        mcr     p15, 5, r0, c15, c15, 0         @ D clean/inv range start
+        mcr     p15, 5, r1, c15, c15, 1         @ D clean/inv range top
+        msr     cpsr_c, r2                      @ restore interrupts
+        mcr     p15, 0, r0, c7, c10, 4          @ drain WB
+        mov     pc, lr
+
 ENTRY(feroceon_cache_fns)
         .long   feroceon_flush_kern_cache_all
         .long   feroceon_flush_user_cache_all
@@ -285,12 +376,33 @@ ENTRY(feroceon_cache_fns)
         .long   feroceon_dma_clean_range
         .long   feroceon_dma_flush_range
 
+ENTRY(feroceon_range_cache_fns)
+        .long   feroceon_flush_kern_cache_all
+        .long   feroceon_flush_user_cache_all
+        .long   feroceon_flush_user_cache_range
+        .long   feroceon_coherent_kern_range
+        .long   feroceon_coherent_user_range
+        .long   feroceon_range_flush_kern_dcache_page
+        .long   feroceon_range_dma_inv_range
+        .long   feroceon_range_dma_clean_range
+        .long   feroceon_range_dma_flush_range
+
         .align  5
 ENTRY(cpu_feroceon_dcache_clean_area)
+#if defined(CONFIG_CACHE_FEROCEON_L2) && !defined(CONFIG_L2_CACHE_WRITETHROUGH)
+        mov     r2, r0
+        mov     r3, r1
+#endif
 1:      mcr     p15, 0, r0, c7, c10, 1          @ clean D entry
         add     r0, r0, #CACHE_DLINESIZE
         subs    r1, r1, #CACHE_DLINESIZE
         bhi     1b
+#if defined(CONFIG_CACHE_FEROCEON_L2) && !defined(CONFIG_L2_CACHE_WRITETHROUGH)
+1:      mcr     p15, 1, r2, c15, c9, 1          @ clean L2 entry
+        add     r2, r2, #CACHE_DLINESIZE
+        subs    r3, r3, #CACHE_DLINESIZE
+        bhi     1b
+#endif
         mcr     p15, 0, r0, c7, c10, 4          @ drain WB
         mov     pc, lr
 
@@ -306,16 +418,25 @@ ENTRY(cpu_feroceon_dcache_clean_area)
         .align  5
 ENTRY(cpu_feroceon_switch_mm)
 #ifdef CONFIG_MMU
-        mov     ip, #0
-@ && 'Clean & Invalidate whole DCache'
-1:      mrc     p15, 0, r15, c7, c14, 3         @ test,clean,invalidate
-        bne     1b
-        mcr     p15, 0, ip, c7, c5, 0           @ invalidate I cache
-        mcr     p15, 0, ip, c7, c10, 4          @ drain WB
+        /*
+         * Note: we wish to call __flush_whole_cache but we need to preserve
+         * lr to do so.  The only way without touching main memory is to
+         * use r2 which is normally used to test the VM_EXEC flag, and
+         * compensate locally for the skipped ops if it is not set.
+         */
+        mov     r2, lr                          @ abuse r2 to preserve lr
+        bl      __flush_whole_cache
+        @ if r2 contains the VM_EXEC bit then the next 2 ops are done already
+        tst     r2, #VM_EXEC
+        mcreq   p15, 0, ip, c7, c5, 0           @ invalidate I cache
+        mcreq   p15, 0, ip, c7, c10, 4          @ drain WB
+
         mcr     p15, 0, r0, c2, c0, 0           @ load page table pointer
         mcr     p15, 0, ip, c8, c7, 0           @ invalidate I & D TLBs
-#endif
+        mov     pc, r2
+#else
         mov     pc, lr
+#endif
 
 /*
  * cpu_feroceon_set_pte_ext(ptep, pte, ext)
@@ -345,6 +466,9 @@ ENTRY(cpu_feroceon_set_pte_ext)
         str     r2, [r0]                        @ hardware version
         mov     r0, r0
         mcr     p15, 0, r0, c7, c10, 1          @ clean D entry
+#if defined(CONFIG_CACHE_FEROCEON_L2) && !defined(CONFIG_L2_CACHE_WRITETHROUGH)
+        mcr     p15, 1, r0, c15, c9, 1          @ clean L2 entry
+#endif
         mcr     p15, 0, r0, c7, c10, 4          @ drain WB
 #endif
         mov     pc, lr
@@ -369,14 +493,15 @@ __feroceon_setup:
         .size   __feroceon_setup, . - __feroceon_setup
 
         /*
-         *  R
-         * .RVI ZFRS BLDP WCAM
-         * .011 0001 ..11 0101
+         *      B
+         *  R   P
+         * .RVI UFRS BLDP WCAM
+         * .011 .001 ..11 0101
          *
          */
         .type   feroceon_crval, #object
 feroceon_crval:
-        crval   clear=0x00007f3f, mmuset=0x00003135, ucset=0x00001134
+        crval   clear=0x0000773f, mmuset=0x00003135, ucset=0x00001134
 
         __INITDATA
 
@@ -414,6 +539,21 @@ cpu_feroceon_name:
         .asciz  "Feroceon"
         .size   cpu_feroceon_name, . - cpu_feroceon_name
 
+        .type   cpu_88fr531_name, #object
+cpu_88fr531_name:
+        .asciz  "Feroceon 88FR531-vd"
+        .size   cpu_88fr531_name, . - cpu_88fr531_name
+
+        .type   cpu_88fr571_name, #object
+cpu_88fr571_name:
+        .asciz  "Feroceon 88FR571-vd"
+        .size   cpu_88fr571_name, . - cpu_88fr571_name
+
+        .type   cpu_88fr131_name, #object
+cpu_88fr131_name:
+        .asciz  "Feroceon 88FR131"
+        .size   cpu_88fr131_name, . - cpu_88fr131_name
+
         .align
 
         .section ".proc.info.init", #alloc, #execinstr
@@ -421,15 +561,15 @@ cpu_feroceon_name:
 #ifdef CONFIG_CPU_FEROCEON_OLD_ID
         .type   __feroceon_old_id_proc_info,#object
 __feroceon_old_id_proc_info:
-        .long   0x41069260
-        .long   0xfffffff0
-        .long   PMD_TYPE_SECT | \
+        .long   0x41009260
+        .long   0xff00fff0
+        .long   PMD_TYPE_SECT | \
                 PMD_SECT_BUFFERABLE | \
                 PMD_SECT_CACHEABLE | \
                 PMD_BIT4 | \
                 PMD_SECT_AP_WRITE | \
                 PMD_SECT_AP_READ
-        .long   PMD_TYPE_SECT | \
+        .long   PMD_TYPE_SECT | \
                 PMD_BIT4 | \
                 PMD_SECT_AP_WRITE | \
                 PMD_SECT_AP_READ
@@ -445,17 +585,17 @@ __feroceon_old_id_proc_info:
         .size   __feroceon_old_id_proc_info, . - __feroceon_old_id_proc_info
 #endif
 
-        .type   __feroceon_proc_info,#object
-__feroceon_proc_info:
+        .type   __88fr531_proc_info,#object
+__88fr531_proc_info:
         .long   0x56055310
         .long   0xfffffff0
-        .long   PMD_TYPE_SECT | \
+        .long   PMD_TYPE_SECT | \
                 PMD_SECT_BUFFERABLE | \
                 PMD_SECT_CACHEABLE | \
                 PMD_BIT4 | \
                 PMD_SECT_AP_WRITE | \
                 PMD_SECT_AP_READ
-        .long   PMD_TYPE_SECT | \
+        .long   PMD_TYPE_SECT | \
                 PMD_BIT4 | \
                 PMD_SECT_AP_WRITE | \
                 PMD_SECT_AP_READ
@@ -463,9 +603,59 @@ __feroceon_proc_info:
         .long   cpu_arch_name
         .long   cpu_elf_name
         .long   HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP
-        .long   cpu_feroceon_name
+        .long   cpu_88fr531_name
         .long   feroceon_processor_functions
         .long   v4wbi_tlb_fns
         .long   feroceon_user_fns
         .long   feroceon_cache_fns
-        .size   __feroceon_proc_info, . - __feroceon_proc_info
+        .size   __88fr531_proc_info, . - __88fr531_proc_info
+
+        .type   __88fr571_proc_info,#object
+__88fr571_proc_info:
+        .long   0x56155710
+        .long   0xfffffff0
+        .long   PMD_TYPE_SECT | \
+                PMD_SECT_BUFFERABLE | \
+                PMD_SECT_CACHEABLE | \
+                PMD_BIT4 | \
+                PMD_SECT_AP_WRITE | \
+                PMD_SECT_AP_READ
+        .long   PMD_TYPE_SECT | \
+                PMD_BIT4 | \
+                PMD_SECT_AP_WRITE | \
+                PMD_SECT_AP_READ
+        b       __feroceon_setup
+        .long   cpu_arch_name
+        .long   cpu_elf_name
+        .long   HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP
+        .long   cpu_88fr571_name
+        .long   feroceon_processor_functions
+        .long   v4wbi_tlb_fns
+        .long   feroceon_user_fns
+        .long   feroceon_range_cache_fns
+        .size   __88fr571_proc_info, . - __88fr571_proc_info
+
+        .type   __88fr131_proc_info,#object
+__88fr131_proc_info:
+        .long   0x56251310
+        .long   0xfffffff0
+        .long   PMD_TYPE_SECT | \
+                PMD_SECT_BUFFERABLE | \
+                PMD_SECT_CACHEABLE | \
+                PMD_BIT4 | \
+                PMD_SECT_AP_WRITE | \
+                PMD_SECT_AP_READ
+        .long   PMD_TYPE_SECT | \
+                PMD_BIT4 | \
+                PMD_SECT_AP_WRITE | \
+                PMD_SECT_AP_READ
+        b       __feroceon_setup
+        .long   cpu_arch_name
+        .long   cpu_elf_name
+        .long   HWCAP_SWP|HWCAP_HALF|HWCAP_THUMB|HWCAP_FAST_MULT|HWCAP_EDSP
+        .long   cpu_88fr131_name
+        .long   feroceon_processor_functions
+        .long   v4wbi_tlb_fns
+        .long   feroceon_user_fns
+        .long   feroceon_range_cache_fns
+        .size   __88fr131_proc_info, . - __88fr131_proc_info