--- a/arch/mips/include/asm/r4kcache.h
+++ b/arch/mips/include/asm/r4kcache.h
@@ -17,6 +17,20 @@
 #include <asm/cpu-features.h>
 #include <asm/mipsmtregs.h>
 
+#ifdef CONFIG_BCM47XX
+#include <asm/paccess.h>
+#include <linux/ssb/ssb.h>
+#define BCM4710_DUMMY_RREG() ((void) *((u8 *) KSEG1ADDR(SSB_ENUM_BASE)))
+
+#define BCM4710_FILL_TLB(addr) (*(volatile unsigned long *)(addr))
+#define BCM4710_PROTECTED_FILL_TLB(addr) ({ unsigned long x; get_dbe(x, (volatile unsigned long *)(addr)); })
+#else
+#define BCM4710_DUMMY_RREG()
+
+#define BCM4710_FILL_TLB(addr)
+#define BCM4710_PROTECTED_FILL_TLB(addr)
+#endif
+
 /*
  * This macro return a properly sign-extended address suitable as base address
  * for indexed cache operations.  Two issues here:
@@ -150,6 +164,7 @@ static inline void flush_icache_line_ind
 static inline void flush_dcache_line_indexed(unsigned long addr)
 {
 	__dflush_prologue
+	BCM4710_DUMMY_RREG();
 	cache_op(Index_Writeback_Inv_D, addr);
 	__dflush_epilogue
 }
@@ -169,6 +184,7 @@ static inline void flush_icache_line(uns
 static inline void flush_dcache_line(unsigned long addr)
 {
 	__dflush_prologue
+	BCM4710_DUMMY_RREG();
 	cache_op(Hit_Writeback_Inv_D, addr);
 	__dflush_epilogue
 }
@@ -176,6 +192,7 @@ static inline void flush_dcache_line(uns
 static inline void invalidate_dcache_line(unsigned long addr)
 {
 	__dflush_prologue
+	BCM4710_DUMMY_RREG();
 	cache_op(Hit_Invalidate_D, addr);
 	__dflush_epilogue
 }
@@ -208,6 +225,7 @@ static inline void flush_scache_line(uns
  */
 static inline void protected_flush_icache_line(unsigned long addr)
 {
+	BCM4710_DUMMY_RREG();
 	protected_cache_op(Hit_Invalidate_I, addr);
 }
 
@@ -219,6 +237,7 @@ static inline void protected_flush_icach
  */
 static inline void protected_writeback_dcache_line(unsigned long addr)
 {
+	BCM4710_DUMMY_RREG();
 	protected_cache_op(Hit_Writeback_Inv_D, addr);
 }
 
@@ -339,8 +358,52 @@ static inline void invalidate_tcache_pag
 		: "r" (base),						\
 		  "i" (op));
 
+static inline void blast_dcache(void)
+{
+	unsigned long start = KSEG0;
+	unsigned long dcache_size = current_cpu_data.dcache.waysize * current_cpu_data.dcache.ways;
+	unsigned long end = (start + dcache_size);
+
+	do {
+		BCM4710_DUMMY_RREG();
+		cache_op(Index_Writeback_Inv_D, start);
+		start += current_cpu_data.dcache.linesz;
+	} while(start < end);
+}
+
+static inline void blast_dcache_page(unsigned long page)
+{
+	unsigned long start = page;
+	unsigned long end = start + PAGE_SIZE;
+
+	BCM4710_FILL_TLB(start);
+	do {
+		BCM4710_DUMMY_RREG();
+		cache_op(Hit_Writeback_Inv_D, start);
+		start += current_cpu_data.dcache.linesz;
+	} while(start < end);
+}
+
+static inline void blast_dcache_page_indexed(unsigned long page)
+{
+	unsigned long start = page;
+	unsigned long end = start + PAGE_SIZE;
+	unsigned long ws_inc = 1UL << current_cpu_data.dcache.waybit;
+	unsigned long ws_end = current_cpu_data.dcache.ways <<
+	                       current_cpu_data.dcache.waybit;
+	unsigned long ws, addr;
+	for (ws = 0; ws < ws_end; ws += ws_inc) {
+		start = page + ws;
+		for (addr = start; addr < end; addr += current_cpu_data.dcache.linesz) {
+			BCM4710_DUMMY_RREG();
+			cache_op(Index_Writeback_Inv_D, addr);
+		}
+	}
+}
+
+
 /* build blast_xxx, blast_xxx_page, blast_xxx_page_indexed */
-#define __BUILD_BLAST_CACHE(pfx, desc, indexop, hitop, lsize) \
+#define __BUILD_BLAST_CACHE(pfx, desc, indexop, hitop, lsize, war) \
 static inline void blast_##pfx##cache##lsize(void)			\
 {									\
 	unsigned long start = INDEX_BASE;				\
@@ -352,6 +415,7 @@ static inline void blast_##pfx##cache##l
 									\
 	__##pfx##flush_prologue						\
 									\
+	war								\
 	for (ws = 0; ws < ws_end; ws += ws_inc)				\
 		for (addr = start; addr < end; addr += lsize * 32)	\
 			cache##lsize##_unroll32(addr|ws, indexop);	\
@@ -366,6 +430,7 @@ static inline void blast_##pfx##cache##l
 									\
 	__##pfx##flush_prologue						\
 									\
+	war								\
 	do {								\
 		cache##lsize##_unroll32(start, hitop);			\
 		start += lsize * 32;					\
@@ -384,6 +449,8 @@ static inline void blast_##pfx##cache##l
 	                       current_cpu_data.desc.waybit;		\
 	unsigned long ws, addr;						\
 									\
+	war								\
+									\
 	__##pfx##flush_prologue						\
 									\
 	for (ws = 0; ws < ws_end; ws += ws_inc)				\
@@ -393,36 +460,38 @@ static inline void blast_##pfx##cache##l
 	__##pfx##flush_epilogue						\
 }
 
-__BUILD_BLAST_CACHE(d, dcache, Index_Writeback_Inv_D, Hit_Writeback_Inv_D, 16)
-__BUILD_BLAST_CACHE(i, icache, Index_Invalidate_I, Hit_Invalidate_I, 16)
-__BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 16)
-__BUILD_BLAST_CACHE(d, dcache, Index_Writeback_Inv_D, Hit_Writeback_Inv_D, 32)
-__BUILD_BLAST_CACHE(i, icache, Index_Invalidate_I, Hit_Invalidate_I, 32)
-__BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 32)
-__BUILD_BLAST_CACHE(d, dcache, Index_Writeback_Inv_D, Hit_Writeback_Inv_D, 64)
-__BUILD_BLAST_CACHE(i, icache, Index_Invalidate_I, Hit_Invalidate_I, 64)
-__BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 64)
-__BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 128)
-
-__BUILD_BLAST_CACHE(inv_d, dcache, Index_Writeback_Inv_D, Hit_Invalidate_D, 16)
-__BUILD_BLAST_CACHE(inv_d, dcache, Index_Writeback_Inv_D, Hit_Invalidate_D, 32)
-__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 16)
-__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 32)
-__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 64)
-__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 128)
+__BUILD_BLAST_CACHE(d, dcache, Index_Writeback_Inv_D, Hit_Writeback_Inv_D, 16, )
+__BUILD_BLAST_CACHE(i, icache, Index_Invalidate_I, Hit_Invalidate_I, 16, BCM4710_FILL_TLB(start);)
+__BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 16, )
+__BUILD_BLAST_CACHE(d, dcache, Index_Writeback_Inv_D, Hit_Writeback_Inv_D, 32, )
+__BUILD_BLAST_CACHE(i, icache, Index_Invalidate_I, Hit_Invalidate_I, 32, BCM4710_FILL_TLB(start);)
+__BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 32, )
+__BUILD_BLAST_CACHE(d, dcache, Index_Writeback_Inv_D, Hit_Writeback_Inv_D, 64, )
+__BUILD_BLAST_CACHE(i, icache, Index_Invalidate_I, Hit_Invalidate_I, 64, BCM4710_FILL_TLB(start);)
+__BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 64, )
+__BUILD_BLAST_CACHE(s, scache, Index_Writeback_Inv_SD, Hit_Writeback_Inv_SD, 128, )
+
+__BUILD_BLAST_CACHE(inv_d, dcache, Index_Writeback_Inv_D, Hit_Invalidate_D, 16, )
+__BUILD_BLAST_CACHE(inv_d, dcache, Index_Writeback_Inv_D, Hit_Invalidate_D, 32, )
+__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 16, )
+__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 32, )
+__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 64, )
+__BUILD_BLAST_CACHE(inv_s, scache, Index_Writeback_Inv_SD, Hit_Invalidate_SD, 128, )
 
 /* build blast_xxx_range, protected_blast_xxx_range */
-#define __BUILD_BLAST_CACHE_RANGE(pfx, desc, hitop, prot) \
+#define __BUILD_BLAST_CACHE_RANGE(pfx, desc, hitop, prot, war, war2) \
 static inline void prot##blast_##pfx##cache##_range(unsigned long start, \
 						    unsigned long end)	\
 {									\
 	unsigned long lsize = cpu_##desc##_line_size();			\
 	unsigned long addr = start & ~(lsize - 1);			\
 	unsigned long aend = (end - 1) & ~(lsize - 1);			\
+	war								\
 									\
 	__##pfx##flush_prologue						\
 									\
 	while (1) {							\
+		war2						\
 		prot##cache_op(hitop, addr);				\
 		if (addr == aend)					\
 			break;						\
@@ -432,13 +501,13 @@ static inline void prot##blast_##pfx##ca
 	__##pfx##flush_epilogue						\
 }
 
-__BUILD_BLAST_CACHE_RANGE(d, dcache, Hit_Writeback_Inv_D, protected_)
-__BUILD_BLAST_CACHE_RANGE(s, scache, Hit_Writeback_Inv_SD, protected_)
-__BUILD_BLAST_CACHE_RANGE(i, icache, Hit_Invalidate_I, protected_)
-__BUILD_BLAST_CACHE_RANGE(d, dcache, Hit_Writeback_Inv_D, )
-__BUILD_BLAST_CACHE_RANGE(s, scache, Hit_Writeback_Inv_SD, )
+__BUILD_BLAST_CACHE_RANGE(d, dcache, Hit_Writeback_Inv_D, protected_, BCM4710_PROTECTED_FILL_TLB(addr); BCM4710_PROTECTED_FILL_TLB(aend);, BCM4710_DUMMY_RREG();)
+__BUILD_BLAST_CACHE_RANGE(s, scache, Hit_Writeback_Inv_SD, protected_,, )
+__BUILD_BLAST_CACHE_RANGE(i, icache, Hit_Invalidate_I, protected_,, )
+__BUILD_BLAST_CACHE_RANGE(d, dcache, Hit_Writeback_Inv_D,, BCM4710_FILL_TLB(addr); BCM4710_FILL_TLB(aend);, BCM4710_DUMMY_RREG();)
+__BUILD_BLAST_CACHE_RANGE(s, scache, Hit_Writeback_Inv_SD,,, )
 /* blast_inv_dcache_range */
-__BUILD_BLAST_CACHE_RANGE(inv_d, dcache, Hit_Invalidate_D, )
-__BUILD_BLAST_CACHE_RANGE(inv_s, scache, Hit_Invalidate_SD, )
+__BUILD_BLAST_CACHE_RANGE(inv_d, dcache, Hit_Invalidate_D,,,BCM4710_DUMMY_RREG();)
+__BUILD_BLAST_CACHE_RANGE(inv_s, scache, Hit_Invalidate_SD,,, )
 
 #endif /* _ASM_R4KCACHE_H */
--- a/arch/mips/include/asm/stackframe.h
+++ b/arch/mips/include/asm/stackframe.h
@@ -449,6 +449,10 @@
 		.macro	RESTORE_SP_AND_RET
 		LONG_L	sp, PT_R29(sp)
 		.set	mips3
+#ifdef CONFIG_BCM47XX
+		nop
+		nop
+#endif
 		eret
 		.set	mips0
 		.endm
--- a/arch/mips/kernel/genex.S
+++ b/arch/mips/kernel/genex.S
@@ -52,6 +52,10 @@ NESTED(except_vec1_generic, 0, sp)
 NESTED(except_vec3_generic, 0, sp)
 	.set	push
 	.set	noat
+#ifdef CONFIG_BCM47XX
+	nop
+	nop
+#endif
 #if R5432_CP0_INTERRUPT_WAR
 	mfc0	k0, CP0_INDEX
 #endif
--- a/arch/mips/mm/c-r4k.c
+++ b/arch/mips/mm/c-r4k.c
@@ -35,6 +35,9 @@
 #include <asm/cacheflush.h> /* for run_uncached() */
 
 
+/* For enabling BCM4710 cache workarounds */
+int bcm4710 = 0;
+
 /*
  * Special Variant of smp_call_function for use by cache functions:
  *
@@ -111,6 +114,9 @@ static void __cpuinit r4k_blast_dcache_p
 {
 	unsigned long  dc_lsize = cpu_dcache_line_size();
 
+	if (bcm4710)
+		r4k_blast_dcache_page = blast_dcache_page;
+	else
 	if (dc_lsize == 0)
 		r4k_blast_dcache_page = (void *)cache_noop;
 	else if (dc_lsize == 16)
@@ -127,6 +133,9 @@ static void __cpuinit r4k_blast_dcache_p
 {
 	unsigned long dc_lsize = cpu_dcache_line_size();
 
+	if (bcm4710)
+		r4k_blast_dcache_page_indexed = blast_dcache_page_indexed;
+	else
 	if (dc_lsize == 0)
 		r4k_blast_dcache_page_indexed = (void *)cache_noop;
 	else if (dc_lsize == 16)
@@ -143,6 +152,9 @@ static void __cpuinit r4k_blast_dcache_s
 {
 	unsigned long dc_lsize = cpu_dcache_line_size();
 
+	if (bcm4710)
+		r4k_blast_dcache = blast_dcache;
+	else
 	if (dc_lsize == 0)
 		r4k_blast_dcache = (void *)cache_noop;
 	else if (dc_lsize == 16)
@@ -680,6 +692,8 @@ static void local_r4k_flush_cache_sigtra
 	unsigned long addr = (unsigned long) arg;
 
 	R4600_HIT_CACHEOP_WAR_IMPL;
+	BCM4710_PROTECTED_FILL_TLB(addr);
+	BCM4710_PROTECTED_FILL_TLB(addr + 4);
 	if (dc_lsize)
 		protected_writeback_dcache_line(addr & ~(dc_lsize - 1));
 	if (!cpu_icache_snoops_remote_store && scache_size)
@@ -1311,6 +1325,17 @@ static void __cpuinit coherency_setup(vo
 	 * silly idea of putting something else there ...
 	 */
 	switch (current_cpu_type()) {
+	case CPU_BCM3302:
+		{
+			u32 cm;
+			cm = read_c0_diag();
+			/* Enable icache */
+			cm |= (1 << 31);
+			/* Enable dcache */
+			cm |= (1 << 30);
+			write_c0_diag(cm);
+		}
+		break;
 	case CPU_R4000PC:
 	case CPU_R4000SC:
 	case CPU_R4000MC:
@@ -1367,6 +1392,15 @@ void __cpuinit r4k_cache_init(void)
 		break;
 	}
 
+	/* Check if special workarounds are required */
+#ifdef CONFIG_BCM47XX
+	if (current_cpu_data.cputype == CPU_BCM4710 && (current_cpu_data.processor_id & 0xff) == 0) {
+		printk("Enabling BCM4710A0 cache workarounds.\n");
+		bcm4710 = 1;
+	} else
+#endif
+		bcm4710 = 0;
+
 	probe_pcache();
 	setup_scache();
 
@@ -1425,5 +1459,13 @@ void __cpuinit r4k_cache_init(void)
 #if !defined(CONFIG_MIPS_CMP)
 	local_r4k___flush_cache_all(NULL);
 #endif
+#ifdef CONFIG_BCM47XX
+	{
+		static void (*_coherency_setup)(void);
+		_coherency_setup = (void (*)(void)) KSEG1ADDR(coherency_setup);
+		_coherency_setup();
+	}
+#else
 	coherency_setup();
+#endif
 }
--- a/arch/mips/mm/tlbex.c
+++ b/arch/mips/mm/tlbex.c
@@ -873,6 +873,9 @@ static void __cpuinit build_r4000_tlb_re
 		/* No need for uasm_i_nop */
 	}
 
+#ifdef CONFIG_BCM47XX
+	uasm_i_nop(&p);
+#endif
 #ifdef CONFIG_64BIT
 	build_get_pmde64(&p, &l, &r, K0, K1); /* get pmd in K1 */
 #else
@@ -1323,6 +1326,9 @@ build_r4000_tlbchange_handler_head(u32 *
 				   struct uasm_reloc **r, unsigned int pte,
 				   unsigned int ptr)
 {
+#ifdef CONFIG_BCM47XX
+	uasm_i_nop(p);
+#endif
 #ifdef CONFIG_64BIT
 	build_get_pmde64(p, l, r, pte, ptr); /* get pmd in ptr */
 #else