[ARM] Marvell Feroceon CPU core support

The Feroceon is a family of independent ARMv5TE compliant CPU core
implementations, supporting a variable depth pipeline and out-of-order
execution.  The Feroceon is configurable with VFP support, and the
later models in the series are superscalar with up to two instructions
per clock cycle.

This patch adds the initial low-level cache/TLB handling for this core.
Signed-off-by: Assaf Hoffman <hoffman@marvell.com>
Reviewed-by: Tzachi Perelstein <tzachi@marvell.com>
Reviewed-by: Nicolas Pitre <nico@marvell.com>
Reviewed-by: Lennert Buytenhek <buytenh@marvell.com>
Acked-by: Russell King <rmk+kernel@arm.linux.org.uk>

arch/arm/Kconfig

@@ -951,7 +951,7 @@ config FPE_FASTFPE
 
 config VFP
 	bool "VFP-format floating point maths"
-	depends on CPU_V6 || CPU_ARM926T
+	depends on CPU_V6 || CPU_ARM926T || CPU_FEROCEON
 	help
 	  Say Y to include VFP support code in the kernel. This is needed
 	  if your hardware includes a VFP unit.

arch/arm/mm/Kconfig

@@ -342,6 +342,18 @@ config CPU_XSC3
 	select CPU_TLB_V4WBI if MMU
 	select IO_36
 
+# Feroceon
+config CPU_FEROCEON
+	bool
+	depends on ARCH_ORION
+	default y
+	select CPU_32v5
+	select CPU_ABRT_EV5T
+	select CPU_CACHE_VIVT
+	select CPU_CP15_MMU
+	select CPU_COPY_V4WB if MMU
+	select CPU_TLB_V4WBI if MMU
+
 # ARMv6
 config CPU_V6
 	bool "Support ARM V6 processor"
@@ -538,7 +550,7 @@ comment "Processor Features"
 
 config ARM_THUMB
 	bool "Support Thumb user binaries"
-	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6 || CPU_V7
+	depends on CPU_ARM720T || CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020 || CPU_ARM1020E || CPU_ARM1022 || CPU_ARM1026 || CPU_XSCALE || CPU_XSC3 || CPU_V6 || CPU_V7 || CPU_FEROCEON
 	default y
 	help
 	  Say Y if you want to include kernel support for running user space
@@ -600,7 +612,7 @@ config CPU_DCACHE_SIZE
 
 config CPU_DCACHE_WRITETHROUGH
 	bool "Force write through D-cache"
-	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020) && !CPU_DCACHE_DISABLE
+	depends on (CPU_ARM740T || CPU_ARM920T || CPU_ARM922T || CPU_ARM925T || CPU_ARM926T || CPU_ARM940T || CPU_ARM946E || CPU_ARM1020 || CPU_FEROCEON) && !CPU_DCACHE_DISABLE
 	default y if CPU_ARM925T
 	help
 	  Say Y here to use the data cache in writethrough mode. Unless you

arch/arm/mm/Makefile

@@ -68,6 +68,7 @@ obj-$(CONFIG_CPU_SA110)		+= proc-sa110.o
 obj-$(CONFIG_CPU_SA1100)	+= proc-sa1100.o
 obj-$(CONFIG_CPU_XSCALE)	+= proc-xscale.o
 obj-$(CONFIG_CPU_XSC3)		+= proc-xsc3.o
+obj-$(CONFIG_CPU_FEROCEON)	+= proc-feroceon.o
 obj-$(CONFIG_CPU_V6)		+= proc-v6.o
 obj-$(CONFIG_CPU_V7)		+= proc-v7.o
 

arch/arm/mm/proc-feroceon.S

+/*
+ *  linux/arch/arm/mm/proc-feroceon.S: MMU functions for Feroceon
+ *
+ *  Heavily based on proc-arm926.S
+ *  Maintainer: Assaf Hoffman <hoffman@marvell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+#include <asm/elf.h>
+#include <asm/pgtable-hwdef.h>
+#include <asm/pgtable.h>
+#include <asm/page.h>
+#include <asm/ptrace.h>
+#include "proc-macros.S"
+
+/*
+ * This is the maximum size of an area which will be invalidated
+ * using the single invalidate entry instructions.  Anything larger
+ * than this, and we go for the whole cache.
+ *
+ * This value should be chosen such that we choose the cheapest
+ * alternative.
+ */
+#define CACHE_DLIMIT	16384
+
+/*
+ * the cache line size of the I and D cache
+ */
+#define CACHE_DLINESIZE	32
+
+	.text
+/*
+ * cpu_feroceon_proc_init()
+ */
+ENTRY(cpu_feroceon_proc_init)
+	mov	pc, lr
+
+/*
+ * cpu_feroceon_proc_fin()
+ */
+ENTRY(cpu_feroceon_proc_fin)
+	stmfd	sp!, {lr}
+	mov	ip, #PSR_F_BIT | PSR_I_BIT | SVC_MODE
+	msr	cpsr_c, ip
+	bl	feroceon_flush_kern_cache_all
+	mrc	p15, 0, r0, c1, c0, 0		@ ctrl register
+	bic	r0, r0, #0x1000			@ ...i............
+	bic	r0, r0, #0x000e			@ ............wca.
+	mcr	p15, 0, r0, c1, c0, 0		@ disable caches
+	ldmfd	sp!, {pc}
+
+/*
+ * cpu_feroceon_reset(loc)
+ *
+ * Perform a soft reset of the system.  Put the CPU into the
+ * same state as it would be if it had been reset, and branch
+ * to what would be the reset vector.
+ *
+ * loc: location to jump to for soft reset
+ */
+	.align	5
+ENTRY(cpu_feroceon_reset)
+	mov	ip, #0
+	mcr	p15, 0, ip, c7, c7, 0		@ invalidate I,D caches
+	mcr	p15, 0, ip, c7, c10, 4		@ drain WB
+#ifdef CONFIG_MMU
+	mcr	p15, 0, ip, c8, c7, 0		@ invalidate I & D TLBs
+#endif
+	mrc	p15, 0, ip, c1, c0, 0		@ ctrl register
+	bic	ip, ip, #0x000f			@ ............wcam
+	bic	ip, ip, #0x1100			@ ...i...s........
+	mcr	p15, 0, ip, c1, c0, 0		@ ctrl register
+	mov	pc, r0
+
+/*
+ * cpu_feroceon_do_idle()
+ *
+ * Called with IRQs disabled
+ */
+	.align	10
+ENTRY(cpu_feroceon_do_idle)
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c10, 4		@ Drain write buffer
+	mcr	p15, 0, r0, c7, c0, 4		@ Wait for interrupt
+	mov	pc, lr
+
+/*
+ *	flush_user_cache_all()
+ *
+ *	Clean and invalidate all cache entries in a particular
+ *	address space.
+ */
+ENTRY(feroceon_flush_user_cache_all)
+	/* FALLTHROUGH */
+
+/*
+ *	flush_kern_cache_all()
+ *
+ *	Clean and invalidate the entire cache.
+ */
+ENTRY(feroceon_flush_kern_cache_all)
+	mov	r2, #VM_EXEC
+	mov	ip, #0
+__flush_whole_cache:
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, ip, c7, c6, 0		@ invalidate D cache
+#else
+1:	mrc	p15, 0, r15, c7, c14, 3 	@ test,clean,invalidate
+	bne	1b
+#endif
+	tst	r2, #VM_EXEC
+	mcrne	p15, 0, ip, c7, c5, 0		@ invalidate I cache
+	mcrne	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	flush_user_cache_range(start, end, flags)
+ *
+ *	Clean and invalidate a range of cache entries in the
+ *	specified address range.
+ *
+ *	- start	- start address (inclusive)
+ *	- end	- end address (exclusive)
+ *	- flags	- vm_flags describing address space
+ */
+ENTRY(feroceon_flush_user_cache_range)
+	mov	ip, #0
+	sub	r3, r1, r0			@ calculate total size
+	cmp	r3, #CACHE_DLIMIT
+	bgt	__flush_whole_cache
+1:	tst	r2, #VM_EXEC
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	mcr	p15, 0, r0, c7, c6, 1		@ invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+#else
+	mcr	p15, 0, r0, c7, c14, 1		@ clean and invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	mcr	p15, 0, r0, c7, c14, 1		@ clean and invalidate D entry
+	mcrne	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+#endif
+	cmp	r0, r1
+	blo	1b
+	tst	r2, #VM_EXEC
+	mcrne	p15, 0, ip, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	coherent_kern_range(start, end)
+ *
+ *	Ensure coherency between the Icache and the Dcache in the
+ *	region described by start, end.  If you have non-snooping
+ *	Harvard caches, you need to implement this function.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(feroceon_coherent_kern_range)
+	/* FALLTHROUGH */
+
+/*
+ *	coherent_user_range(start, end)
+ *
+ *	Ensure coherency between the Icache and the Dcache in the
+ *	region described by start, end.  If you have non-snooping
+ *	Harvard caches, you need to implement this function.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(feroceon_coherent_user_range)
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	mcr	p15, 0, r0, c7, c5, 1		@ invalidate I entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	flush_kern_dcache_page(void *page)
+ *
+ *	Ensure no D cache aliasing occurs, either with itself or
+ *	the I cache
+ *
+ *	- addr	- page aligned address
+ */
+ENTRY(feroceon_flush_kern_dcache_page)
+	add	r1, r0, #PAGE_SZ
+1:	mcr	p15, 0, r0, c7, c14, 1		@ clean+invalidate D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	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)
+ *
+ *	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
+ *
+ * (same as v4wb)
+ */
+ENTRY(feroceon_dma_inv_range)
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	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
+#endif
+	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
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_clean_range(start, end)
+ *
+ *	Clean the specified virtual address range.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ *
+ * (same as v4wb)
+ */
+ENTRY(feroceon_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
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/*
+ *	dma_flush_range(start, end)
+ *
+ *	Clean and invalidate the specified virtual address range.
+ *
+ *	- start	- virtual start address
+ *	- end	- virtual end address
+ */
+ENTRY(feroceon_dma_flush_range)
+	bic	r0, r0, #CACHE_DLINESIZE - 1
+1:
+#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
+#endif
+	add	r0, r0, #CACHE_DLINESIZE
+	cmp	r0, r1
+	blo	1b
+	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
+	.long	feroceon_flush_user_cache_range
+	.long	feroceon_coherent_kern_range
+	.long	feroceon_coherent_user_range
+	.long	feroceon_flush_kern_dcache_page
+	.long	feroceon_dma_inv_range
+	.long	feroceon_dma_clean_range
+	.long	feroceon_dma_flush_range
+
+ENTRY(cpu_feroceon_dcache_clean_area)
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+1:	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+	add	r0, r0, #CACHE_DLINESIZE
+	subs	r1, r1, #CACHE_DLINESIZE
+	bhi	1b
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+	mov	pc, lr
+
+/* =============================== PageTable ============================== */
+
+/*
+ * cpu_feroceon_switch_mm(pgd)
+ *
+ * Set the translation base pointer to be as described by pgd.
+ *
+ * pgd: new page tables
+ */
+	.align	5
+ENTRY(cpu_feroceon_switch_mm)
+#ifdef CONFIG_MMU
+	mov	ip, #0
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, ip, c7, c6, 0		@ invalidate D cache
+#else
+@ && 'Clean & Invalidate whole DCache'
+1:	mrc	p15, 0, r15, c7, c14, 3 	@ test,clean,invalidate
+	bne	1b
+#endif
+	mcr	p15, 0, ip, c7, c5, 0		@ invalidate I cache
+	mcr	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, lr
+
+/*
+ * cpu_feroceon_set_pte_ext(ptep, pte, ext)
+ *
+ * Set a PTE and flush it out
+ */
+	.align	5
+ENTRY(cpu_feroceon_set_pte_ext)
+#ifdef CONFIG_MMU
+	str	r1, [r0], #-2048		@ linux version
+
+	eor	r1, r1, #L_PTE_PRESENT | L_PTE_YOUNG | L_PTE_WRITE | L_PTE_DIRTY
+
+	bic	r2, r1, #PTE_SMALL_AP_MASK
+	bic	r2, r2, #PTE_TYPE_MASK
+	orr	r2, r2, #PTE_TYPE_SMALL
+
+	tst	r1, #L_PTE_USER			@ User?
+	orrne	r2, r2, #PTE_SMALL_AP_URO_SRW
+
+	tst	r1, #L_PTE_WRITE | L_PTE_DIRTY	@ Write and Dirty?
+	orreq	r2, r2, #PTE_SMALL_AP_UNO_SRW
+
+	tst	r1, #L_PTE_PRESENT | L_PTE_YOUNG	@ Present and Young?
+	movne	r2, #0
+
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	eor	r3, r2, #0x0a			@ C & small page?
+	tst	r3, #0x0b
+	biceq	r2, r2, #4
+#endif
+	str	r2, [r0]			@ hardware version
+	mov	r0, r0
+#ifndef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mcr	p15, 0, r0, c7, c10, 1		@ clean D entry
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain WB
+#endif
+	mov	pc, lr
+
+	__INIT
+
+	.type	__feroceon_setup, #function
+__feroceon_setup:
+	mov	r0, #0
+	mcr	p15, 0, r0, c7, c7		@ invalidate I,D caches on v4
+	mcr	p15, 0, r0, c7, c10, 4		@ drain write buffer on v4
+#ifdef CONFIG_MMU
+	mcr	p15, 0, r0, c8, c7		@ invalidate I,D TLBs on v4
+#endif
+
+
+#ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
+	mov	r0, #4				@ disable write-back on caches explicitly
+	mcr	p15, 7, r0, c15, c0, 0
+#endif
+
+	adr	r5, feroceon_crval
+	ldmia	r5, {r5, r6}
+	mrc	p15, 0, r0, c1, c0		@ get control register v4
+	bic	r0, r0, r5
+	orr	r0, r0, r6
+#ifdef CONFIG_CPU_CACHE_ROUND_ROBIN
+	orr	r0, r0, #0x4000			@ .1.. .... .... ....
+#endif
+	mov	pc, lr
+	.size	__feroceon_setup, . - __feroceon_setup
+
+	/*
+	 *  R
+	 * .RVI ZFRS BLDP WCAM
+	 * .011 0001 ..11 0101
+	 *
+	 */
+	.type	feroceon_crval, #object
+feroceon_crval:
+	crval	clear=0x00007f3f, mmuset=0x00003135, ucset=0x00001134
+
+	__INITDATA
+
+/*
+ * Purpose : Function pointers used to access above functions - all calls
+ *	     come through these
+ */
+	.type	feroceon_processor_functions, #object
+feroceon_processor_functions:
+	.word	v5t_early_abort
+	.word	cpu_feroceon_proc_init
+	.word	cpu_feroceon_proc_fin
+	.word	cpu_feroceon_reset
+	.word	cpu_feroceon_do_idle
+	.word	cpu_feroceon_dcache_clean_area
+	.word	cpu_feroceon_switch_mm
+	.word	cpu_feroceon_set_pte_ext
+	.size	feroceon_processor_functions, . - feroceon_processor_functions
+
+	.section ".rodata"
+
+	.type	cpu_arch_name, #object
+cpu_arch_name:
+	.asciz	"armv5te"
+	.size	cpu_arch_name, . - cpu_arch_name
+
+	.type	cpu_elf_name, #object
+cpu_elf_name:
+	.asciz	"v5"
+	.size	cpu_elf_name, . - cpu_elf_name
+
+	.type	cpu_feroceon_name, #object
+cpu_feroceon_name:
+	.asciz	"Feroceon"
+	.size	cpu_feroceon_name, . - cpu_feroceon_name
+
+	.align
+
+	.section ".proc.info.init", #alloc, #execinstr
+
+	.type	__feroceon_proc_info,#object
+__feroceon_proc_info:
+	.long	0x56055310
+	.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_feroceon_name
+	.long	feroceon_processor_functions
+	.long	v4wbi_tlb_fns
+	.long	v4wb_user_fns
+	.long	feroceon_cache_fns
+	.size	__feroceon_proc_info, . - __feroceon_proc_info

include/asm-arm/cacheflush.h

@@ -94,6 +94,14 @@
 # endif
 #endif
 
+#if defined(CONFIG_CPU_FEROCEON)
+# ifdef _CACHE
+#  define MULTI_CACHE 1
+# else
+#  define _CACHE feroceon
+# endif
+#endif
+
 #if defined(CONFIG_CPU_V6)
 //# ifdef _CACHE
 #  define MULTI_CACHE 1

include/asm-arm/proc-fns.h

@@ -185,6 +185,14 @@
 #   define CPU_NAME cpu_xsc3
 #  endif
 # endif
+# ifdef CONFIG_CPU_FEROCEON
+#  ifdef CPU_NAME
+#   undef  MULTI_CPU
+#   define MULTI_CPU
+#  else
+#   define CPU_NAME cpu_feroceon
+#  endif
+# endif
 # ifdef CONFIG_CPU_V6
 #  ifdef CPU_NAME
 #   undef  MULTI_CPU