From e651eab0af88aa7a281fe9e8c36c0846552aa7fc Mon Sep 17 00:00:00 2001 From: Sricharan R Date: Mon, 18 Mar 2013 12:24:04 +0100 Subject: ARM: 7677/1: LPAE: Fix mapping in alloc_init_section for unaligned addresses With LPAE enabled, alloc_init_section() does not map the entire address space for unaligned addresses. The issue also reproduced with CMA + LPAE. CMA tries to map 16MB with page granularity mappings during boot. alloc_init_pte() is called and out of 16MB, only 2MB gets mapped and rest remains unaccessible. Because of this OMAP5 boot is broken with CMA + LPAE enabled. Fix the issue by ensuring that the entire addresses are mapped. Signed-off-by: R Sricharan Cc: Catalin Marinas Cc: Christoffer Dall Cc: Santosh Shilimkar Tested-by: Laura Abbott Acked-by: Catalin Marinas Acked-by: Christoffer Dall Signed-off-by: Russell King --- arch/arm/mm/mmu.c | 73 +++++++++++++++++++++++++++++++++++-------------------- 1 file changed, 47 insertions(+), 26 deletions(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c index e95a996ab78f..78978945492a 100644 --- a/arch/arm/mm/mmu.c +++ b/arch/arm/mm/mmu.c @@ -598,39 +598,60 @@ static void __init alloc_init_pte(pmd_t *pmd, unsigned long addr, } while (pte++, addr += PAGE_SIZE, addr != end); } -static void __init alloc_init_section(pud_t *pud, unsigned long addr, - unsigned long end, phys_addr_t phys, - const struct mem_type *type) +static void __init map_init_section(pmd_t *pmd, unsigned long addr, + unsigned long end, phys_addr_t phys, + const struct mem_type *type) { - pmd_t *pmd = pmd_offset(pud, addr); - +#ifndef CONFIG_ARM_LPAE /* - * Try a section mapping - end, addr and phys must all be aligned - * to a section boundary. Note that PMDs refer to the individual - * L1 entries, whereas PGDs refer to a group of L1 entries making - * up one logical pointer to an L2 table. + * In classic MMU format, puds and pmds are folded in to + * the pgds. pmd_offset gives the PGD entry. PGDs refer to a + * group of L1 entries making up one logical pointer to + * an L2 table (2MB), where as PMDs refer to the individual + * L1 entries (1MB). Hence increment to get the correct + * offset for odd 1MB sections. + * (See arch/arm/include/asm/pgtable-2level.h) */ - if (type->prot_sect && ((addr | end | phys) & ~SECTION_MASK) == 0) { - pmd_t *p = pmd; - -#ifndef CONFIG_ARM_LPAE - if (addr & SECTION_SIZE) - pmd++; + if (addr & SECTION_SIZE) + pmd++; #endif + do { + *pmd = __pmd(phys | type->prot_sect); + phys += SECTION_SIZE; + } while (pmd++, addr += SECTION_SIZE, addr != end); - do { - *pmd = __pmd(phys | type->prot_sect); - phys += SECTION_SIZE; - } while (pmd++, addr += SECTION_SIZE, addr != end); + flush_pmd_entry(pmd); +} - flush_pmd_entry(p); - } else { +static void __init alloc_init_pmd(pud_t *pud, unsigned long addr, + unsigned long end, phys_addr_t phys, + const struct mem_type *type) +{ + pmd_t *pmd = pmd_offset(pud, addr); + unsigned long next; + + do { /* - * No need to loop; pte's aren't interested in the - * individual L1 entries. + * With LPAE, we must loop over to map + * all the pmds for the given range. */ - alloc_init_pte(pmd, addr, end, __phys_to_pfn(phys), type); - } + next = pmd_addr_end(addr, end); + + /* + * Try a section mapping - addr, next and phys must all be + * aligned to a section boundary. + */ + if (type->prot_sect && + ((addr | next | phys) & ~SECTION_MASK) == 0) { + map_init_section(pmd, addr, next, phys, type); + } else { + alloc_init_pte(pmd, addr, next, + __phys_to_pfn(phys), type); + } + + phys += next - addr; + + } while (pmd++, addr = next, addr != end); } static void __init alloc_init_pud(pgd_t *pgd, unsigned long addr, @@ -641,7 +662,7 @@ static void __init alloc_init_pud(pgd_t *pgd, unsigned long addr, do { next = pud_addr_end(addr, end); - alloc_init_section(pud, addr, next, phys, type); + alloc_init_pmd(pud, addr, next, phys, type); phys += next - addr; } while (pud++, addr = next, addr != end); } -- cgit v1.2.1 From 8164f7af88d9ad3a757bd14f634b23997ee77f6b Mon Sep 17 00:00:00 2001 From: Stephen Boyd Date: Mon, 18 Mar 2013 19:44:15 +0100 Subject: ARM: 7680/1: Detect support for SDIV/UDIV from ISAR0 register The ISAR0 register indicates support for the SDIV and UDIV instructions in both the Thumb and ARM instruction set. Read the register to detect the supported instructions and update the elf_hwcap mask as appropriate. This is better than adding more and more cpuid checks in proc-v7.S for each new cpu variant that supports these instructions. Acked-by: Will Deacon Cc: Stepan Moskovchenko Signed-off-by: Stephen Boyd Signed-off-by: Russell King --- arch/arm/mm/proc-v7.S | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S index 3a3c015f8d5c..bcd3d48922fb 100644 --- a/arch/arm/mm/proc-v7.S +++ b/arch/arm/mm/proc-v7.S @@ -420,7 +420,7 @@ __v7_pj4b_proc_info: __v7_ca7mp_proc_info: .long 0x410fc070 .long 0xff0ffff0 - __v7_proc __v7_ca7mp_setup, hwcaps = HWCAP_IDIV + __v7_proc __v7_ca7mp_setup .size __v7_ca7mp_proc_info, . - __v7_ca7mp_proc_info /* @@ -430,7 +430,7 @@ __v7_ca7mp_proc_info: __v7_ca15mp_proc_info: .long 0x410fc0f0 .long 0xff0ffff0 - __v7_proc __v7_ca15mp_setup, hwcaps = HWCAP_IDIV + __v7_proc __v7_ca15mp_setup .size __v7_ca15mp_proc_info, . - __v7_ca15mp_proc_info /* -- cgit v1.2.1 From 120ecfafabec382c4feb79ff159ef42a39b6d33b Mon Sep 17 00:00:00 2001 From: Stepan Moskovchenko Date: Mon, 18 Mar 2013 19:44:16 +0100 Subject: ARM: 7678/1: Work around faulty ISAR0 register in some Krait CPUs Some early versions of the Krait CPU design incorrectly indicate that they only support the UDIV and SDIV instructions in Thumb mode when they actually support them in ARM and Thumb mode. It seems that these CPUs follow the DDI0406B ARM ARM which has two possible values for the divide instructions field, instead of the DDI0406C document which has three possible values. Work around this problem by checking the MIDR against Krait CPUs with this faulty ISAR0 register and force the hwcaps to indicate support in both modes. [sboyd: Rewrote commit text to reflect real reasoning now that we autodetect udiv/sdiv] Signed-off-by: Stepan Moskovchenko Acked-by: Will Deacon Signed-off-by: Stephen Boyd Signed-off-by: Russell King --- arch/arm/mm/proc-v7.S | 15 +++++++++++++++ 1 file changed, 15 insertions(+) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S index bcd3d48922fb..f584d3f5b37c 100644 --- a/arch/arm/mm/proc-v7.S +++ b/arch/arm/mm/proc-v7.S @@ -433,6 +433,21 @@ __v7_ca15mp_proc_info: __v7_proc __v7_ca15mp_setup .size __v7_ca15mp_proc_info, . - __v7_ca15mp_proc_info + /* + * Qualcomm Inc. Krait processors. + */ + .type __krait_proc_info, #object +__krait_proc_info: + .long 0x510f0400 @ Required ID value + .long 0xff0ffc00 @ Mask for ID + /* + * Some Krait processors don't indicate support for SDIV and UDIV + * instructions in the ARM instruction set, even though they actually + * do support them. + */ + __v7_proc __v7_setup, hwcaps = HWCAP_IDIV + .size __krait_proc_info, . - __krait_proc_info + /* * Match any ARMv7 processor core. */ -- cgit v1.2.1 From 82d9b0d0c6c4cbd5d0022d7df5e6bf071a9eb6c7 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 15 Jan 2013 12:07:40 +0000 Subject: ARM: cache: remove ARMv3 support code This is only used by 740t, which is a v4 core and (by my reading of the datasheet for the CPU) ignores CRm for the cp15 cache flush operation, making the v4 cache implementation in cache-v4.S sufficient for this CPU. Tested with 740T core-tile on Integrator/AP baseboard. Acked-by: Hyok S. Choi Acked-by: Greg Ungerer Signed-off-by: Will Deacon --- arch/arm/mm/Kconfig | 5 +- arch/arm/mm/Makefile | 1 - arch/arm/mm/cache-v3.S | 137 ---------------------------------------------- arch/arm/mm/proc-arm740.S | 2 +- 4 files changed, 2 insertions(+), 143 deletions(-) delete mode 100644 arch/arm/mm/cache-v3.S (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig index 025d17328730..4045c4931a30 100644 --- a/arch/arm/mm/Kconfig +++ b/arch/arm/mm/Kconfig @@ -43,7 +43,7 @@ config CPU_ARM740T depends on !MMU select CPU_32v4T select CPU_ABRT_LV4T - select CPU_CACHE_V3 # although the core is v4t + select CPU_CACHE_V4 select CPU_CP15_MPU select CPU_PABRT_LEGACY help @@ -469,9 +469,6 @@ config CPU_PABRT_V7 bool # The cache model -config CPU_CACHE_V3 - bool - config CPU_CACHE_V4 bool diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile index 4e333fa2756f..9e51be96f635 100644 --- a/arch/arm/mm/Makefile +++ b/arch/arm/mm/Makefile @@ -33,7 +33,6 @@ obj-$(CONFIG_CPU_PABRT_LEGACY) += pabort-legacy.o obj-$(CONFIG_CPU_PABRT_V6) += pabort-v6.o obj-$(CONFIG_CPU_PABRT_V7) += pabort-v7.o -obj-$(CONFIG_CPU_CACHE_V3) += cache-v3.o obj-$(CONFIG_CPU_CACHE_V4) += cache-v4.o obj-$(CONFIG_CPU_CACHE_V4WT) += cache-v4wt.o obj-$(CONFIG_CPU_CACHE_V4WB) += cache-v4wb.o diff --git a/arch/arm/mm/cache-v3.S b/arch/arm/mm/cache-v3.S deleted file mode 100644 index 8a3fadece8d3..000000000000 --- a/arch/arm/mm/cache-v3.S +++ /dev/null @@ -1,137 +0,0 @@ -/* - * linux/arch/arm/mm/cache-v3.S - * - * Copyright (C) 1997-2002 Russell king - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License version 2 as - * published by the Free Software Foundation. - */ -#include -#include -#include -#include "proc-macros.S" - -/* - * flush_icache_all() - * - * Unconditionally clean and invalidate the entire icache. - */ -ENTRY(v3_flush_icache_all) - mov pc, lr -ENDPROC(v3_flush_icache_all) - -/* - * flush_user_cache_all() - * - * Invalidate all cache entries in a particular address - * space. - * - * - mm - mm_struct describing address space - */ -ENTRY(v3_flush_user_cache_all) - /* FALLTHROUGH */ -/* - * flush_kern_cache_all() - * - * Clean and invalidate the entire cache. - */ -ENTRY(v3_flush_kern_cache_all) - /* FALLTHROUGH */ - -/* - * flush_user_cache_range(start, end, flags) - * - * Invalidate a range of cache entries in the specified - * address space. - * - * - start - start address (may not be aligned) - * - end - end address (exclusive, may not be aligned) - * - flags - vma_area_struct flags describing address space - */ -ENTRY(v3_flush_user_cache_range) - mov ip, #0 - mcreq p15, 0, ip, c7, c0, 0 @ flush ID cache - mov pc, lr - -/* - * coherent_kern_range(start, end) - * - * Ensure coherency between the Icache and the Dcache in the - * region described by start. If you have non-snooping - * Harvard caches, you need to implement this function. - * - * - start - virtual start address - * - end - virtual end address - */ -ENTRY(v3_coherent_kern_range) - /* FALLTHROUGH */ - -/* - * coherent_user_range(start, end) - * - * Ensure coherency between the Icache and the Dcache in the - * region described by start. If you have non-snooping - * Harvard caches, you need to implement this function. - * - * - start - virtual start address - * - end - virtual end address - */ -ENTRY(v3_coherent_user_range) - mov r0, #0 - mov pc, lr - -/* - * flush_kern_dcache_area(void *page, size_t size) - * - * Ensure no D cache aliasing occurs, either with itself or - * the I cache - * - * - addr - kernel address - * - size - region size - */ -ENTRY(v3_flush_kern_dcache_area) - /* FALLTHROUGH */ - -/* - * dma_flush_range(start, end) - * - * Clean and invalidate the specified virtual address range. - * - * - start - virtual start address - * - end - virtual end address - */ -ENTRY(v3_dma_flush_range) - mov r0, #0 - mcr p15, 0, r0, c7, c0, 0 @ flush ID cache - 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_flush_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) - - .globl v3_flush_kern_cache_louis - .equ v3_flush_kern_cache_louis, v3_flush_kern_cache_all - - __INITDATA - - @ define struct cpu_cache_fns (see and proc-macros.S) - define_cache_functions v3 diff --git a/arch/arm/mm/proc-arm740.S b/arch/arm/mm/proc-arm740.S index dc5de5d53f20..2088234978c4 100644 --- a/arch/arm/mm/proc-arm740.S +++ b/arch/arm/mm/proc-arm740.S @@ -145,5 +145,5 @@ __arm740_proc_info: .long arm740_processor_functions .long 0 .long 0 - .long v3_cache_fns @ cache model + .long v4_cache_fns @ cache model .size __arm740_proc_info, . - __arm740_proc_info -- cgit v1.2.1 From 3ef52f2a00efc5f83ae6d40e55cae96ce275893f Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 22 Jan 2013 10:37:51 +0000 Subject: ARM: mm: fix numerous hideous errors in proc-arm740.S The setup code in proc-arm740.S is completely broken and, as far as I can tell, always has been. I was >this< close to ripping it out, when a 740t core-tile materialised in the office, so I've had a crack at fixing things up: - Fix the ram/flash area calculations so that we actually set the condition flags before testing them... - Fix the proc_info structure so that __cpu_io_mmu_flags are defined as 0, placing the __cpu_flush pointer at the correct offset - Re-number the registers used during __arm740_setup so that we don't clobber the machine ID et al - Advertise Thumb support via the hwcaps, since 740T is the only 740 implementation. Acked-by: Hyok S. Choi Signed-off-by: Will Deacon --- arch/arm/mm/proc-arm740.S | 28 ++++++++++++++++------------ 1 file changed, 16 insertions(+), 12 deletions(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/proc-arm740.S b/arch/arm/mm/proc-arm740.S index 2088234978c4..fde2d2a794cf 100644 --- a/arch/arm/mm/proc-arm740.S +++ b/arch/arm/mm/proc-arm740.S @@ -77,24 +77,27 @@ __arm740_setup: mcr p15, 0, r0, c6, c0 @ set area 0, default ldr r0, =(CONFIG_DRAM_BASE & 0xFFFFF000) @ base[31:12] of RAM - ldr r1, =(CONFIG_DRAM_SIZE >> 12) @ size of RAM (must be >= 4KB) - mov r2, #10 @ 11 is the minimum (4KB) -1: add r2, r2, #1 @ area size *= 2 - mov r1, r1, lsr #1 + ldr r3, =(CONFIG_DRAM_SIZE >> 12) @ size of RAM (must be >= 4KB) + mov r4, #10 @ 11 is the minimum (4KB) +1: add r4, r4, #1 @ area size *= 2 + movs r3, r3, lsr #1 bne 1b @ count not zero r-shift - orr r0, r0, r2, lsl #1 @ the area register value + orr r0, r0, r4, lsl #1 @ the area register value orr r0, r0, #1 @ set enable bit mcr p15, 0, r0, c6, c1 @ set area 1, RAM ldr r0, =(CONFIG_FLASH_MEM_BASE & 0xFFFFF000) @ base[31:12] of FLASH - ldr r1, =(CONFIG_FLASH_SIZE >> 12) @ size of FLASH (must be >= 4KB) - mov r2, #10 @ 11 is the minimum (4KB) -1: add r2, r2, #1 @ area size *= 2 - mov r1, r1, lsr #1 + ldr r3, =(CONFIG_FLASH_SIZE >> 12) @ size of FLASH (must be >= 4KB) + cmp r3, #0 + moveq r0, #0 + beq 2f + mov r4, #10 @ 11 is the minimum (4KB) +1: add r4, r4, #1 @ area size *= 2 + movs r3, r3, lsr #1 bne 1b @ count not zero r-shift - orr r0, r0, r2, lsl #1 @ the area register value + orr r0, r0, r4, lsl #1 @ the area register value orr r0, r0, #1 @ set enable bit - mcr p15, 0, r0, c6, c2 @ set area 2, ROM/FLASH +2: mcr p15, 0, r0, c6, c2 @ set area 2, ROM/FLASH mov r0, #0x06 mcr p15, 0, r0, c2, c0 @ Region 1&2 cacheable @@ -137,10 +140,11 @@ __arm740_proc_info: .long 0x41807400 .long 0xfffffff0 .long 0 + .long 0 b __arm740_setup .long cpu_arch_name .long cpu_elf_name - .long HWCAP_SWP | HWCAP_HALF | HWCAP_26BIT + .long HWCAP_SWP | HWCAP_HALF | HWCAP_THUMB | HWCAP_26BIT .long cpu_arm740_name .long arm740_processor_functions .long 0 -- cgit v1.2.1 From 794fe85da484353dedcb5dc6da14d923d0645fc3 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 22 Jan 2013 19:11:38 +0000 Subject: ARM: mm: remove broken condition check for v4 flushing There's no point having a conditional cache flush if we don't know the state of the condition beforehand. This patch makes the cacheflush in v4_flush_user_cache_range unconditional. signed-off-by: will deacon --- arch/arm/mm/cache-v4.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/cache-v4.S b/arch/arm/mm/cache-v4.S index 43e5d77be677..a7ba68f59f0c 100644 --- a/arch/arm/mm/cache-v4.S +++ b/arch/arm/mm/cache-v4.S @@ -58,7 +58,7 @@ ENTRY(v4_flush_kern_cache_all) ENTRY(v4_flush_user_cache_range) #ifdef CONFIG_CPU_CP15 mov ip, #0 - mcreq p15, 0, ip, c7, c7, 0 @ flush ID cache + mcr p15, 0, ip, c7, c7, 0 @ flush ID cache mov pc, lr #else /* FALLTHROUGH */ -- cgit v1.2.1 From d455bac223d6f9858b29a29272756243ec3d3904 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Tue, 22 Jan 2013 11:00:54 +0000 Subject: ARM: modules: don't export cpu_set_pte_ext when !MMU cpu_set_pte_ext is only guaranteed to be defined when CONFIG_MMU, so don't export it to modules otherwise. Signed-off-by: Will Deacon --- arch/arm/mm/proc-syms.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/proc-syms.c b/arch/arm/mm/proc-syms.c index 3e6210b4d6d4..054b491ff764 100644 --- a/arch/arm/mm/proc-syms.c +++ b/arch/arm/mm/proc-syms.c @@ -17,7 +17,9 @@ #ifndef MULTI_CPU EXPORT_SYMBOL(cpu_dcache_clean_area); +#ifdef CONFIG_MMU EXPORT_SYMBOL(cpu_set_pte_ext); +#endif #else EXPORT_SYMBOL(processor); #endif -- cgit v1.2.1 From 6e7aceeb7c70b9ebad79bcfe91fcf738826e8e6d Mon Sep 17 00:00:00 2001 From: Rob Herring Date: Mon, 25 Mar 2013 17:02:48 +0100 Subject: ARM: 7682/1: cache-l2x0: fix masking of RTL revision numbering and set_debug init Commit b8db6b8 (ARM: 7547/4: cache-l2x0: add support for Aurora L2 cache ctrl) moved the masking of the part ID which caused the RTL version to be lost. Commit 6248d06 (ARM: 7545/1: cache-l2x0: make outer_cache_fns a field of l2x0_of_data) changed how .set_debug is initialized. Both commits break commit 74ddcdb (ARM: 7608/1: l2x0: Only set .set_debug on PL310 r3p0 and earlier) which uses the RTL version to conditionally set .set_debug function pointer. Commit b8db6b8 also caused the printed cache ID to be missing the version information. Fix this by reverting how the part number is masked so the RTL version info is maintained. The cache-id-part DT property does not set the RTL bits so masking them should have no effect. Also, re-arrange the order of the function pointer init so the .set_debug function can be overridden. Reported-by: Paolo Pisati Signed-off-by: Rob Herring Cc: Gregory CLEMENT Cc: Yehuda Yitschak Signed-off-by: Russell King --- arch/arm/mm/cache-l2x0.c | 11 ++++------- 1 file changed, 4 insertions(+), 7 deletions(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c index c2f37390308a..c465faca51b0 100644 --- a/arch/arm/mm/cache-l2x0.c +++ b/arch/arm/mm/cache-l2x0.c @@ -299,7 +299,7 @@ static void l2x0_unlock(u32 cache_id) int lockregs; int i; - switch (cache_id) { + switch (cache_id & L2X0_CACHE_ID_PART_MASK) { case L2X0_CACHE_ID_PART_L310: lockregs = 8; break; @@ -333,15 +333,14 @@ void __init l2x0_init(void __iomem *base, u32 aux_val, u32 aux_mask) if (cache_id_part_number_from_dt) cache_id = cache_id_part_number_from_dt; else - cache_id = readl_relaxed(l2x0_base + L2X0_CACHE_ID) - & L2X0_CACHE_ID_PART_MASK; + cache_id = readl_relaxed(l2x0_base + L2X0_CACHE_ID); aux = readl_relaxed(l2x0_base + L2X0_AUX_CTRL); aux &= aux_mask; aux |= aux_val; /* Determine the number of ways */ - switch (cache_id) { + switch (cache_id & L2X0_CACHE_ID_PART_MASK) { case L2X0_CACHE_ID_PART_L310: if (aux & (1 << 16)) ways = 16; @@ -725,7 +724,6 @@ static const struct l2x0_of_data pl310_data = { .flush_all = l2x0_flush_all, .inv_all = l2x0_inv_all, .disable = l2x0_disable, - .set_debug = pl310_set_debug, }, }; @@ -814,9 +812,8 @@ int __init l2x0_of_init(u32 aux_val, u32 aux_mask) data->save(); of_init = true; - l2x0_init(l2x0_base, aux_val, aux_mask); - memcpy(&outer_cache, &data->outer_cache, sizeof(outer_cache)); + l2x0_init(l2x0_base, aux_val, aux_mask); return 0; } -- cgit v1.2.1 From 93dc68876b608da041fe40ed39424b0fcd5aa2fb Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Tue, 26 Mar 2013 23:35:04 +0100 Subject: ARM: 7684/1: errata: Workaround for Cortex-A15 erratum 798181 (TLBI/DSB operations) On Cortex-A15 (r0p0..r3p2) the TLBI/DSB are not adequately shooting down all use of the old entries. This patch implements the erratum workaround which consists of: 1. Dummy TLBIMVAIS and DSB on the CPU doing the TLBI operation. 2. Send IPI to the CPUs that are running the same mm (and ASID) as the one being invalidated (or all the online CPUs for global pages). 3. CPU receiving the IPI executes a DMB and CLREX (part of the exception return code already). Signed-off-by: Catalin Marinas Signed-off-by: Russell King --- arch/arm/mm/context.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/context.c b/arch/arm/mm/context.c index a5a4b2bc42ba..2ac37372ef52 100644 --- a/arch/arm/mm/context.c +++ b/arch/arm/mm/context.c @@ -48,7 +48,7 @@ static DEFINE_RAW_SPINLOCK(cpu_asid_lock); static atomic64_t asid_generation = ATOMIC64_INIT(ASID_FIRST_VERSION); static DECLARE_BITMAP(asid_map, NUM_USER_ASIDS); -static DEFINE_PER_CPU(atomic64_t, active_asids); +DEFINE_PER_CPU(atomic64_t, active_asids); static DEFINE_PER_CPU(u64, reserved_asids); static cpumask_t tlb_flush_pending; @@ -215,6 +215,7 @@ void check_and_switch_context(struct mm_struct *mm, struct task_struct *tsk) if (cpumask_test_and_clear_cpu(cpu, &tlb_flush_pending)) { local_flush_bp_all(); local_flush_tlb_all(); + dummy_flush_tlb_a15_erratum(); } atomic64_set(&per_cpu(active_asids, cpu), asid); -- cgit v1.2.1 From ae8a8b9553bd3906af74ff4e8d763904d20ab4e5 Mon Sep 17 00:00:00 2001 From: Will Deacon Date: Wed, 3 Apr 2013 17:16:57 +0100 Subject: ARM: 7691/1: mm: kill unused TLB_CAN_READ_FROM_L1_CACHE and use ALT_SMP instead Many ARMv7 cores have hardware page table walkers that can read the L1 cache. This is discoverable from the ID_MMFR3 register, although this can be expensive to access from the low-level set_pte functions and is a pain to cache, particularly with multi-cluster systems. A useful observation is that the multi-processing extensions for ARMv7 require coherent table walks, meaning that we can make use of ALT_SMP patching in proc-v7-* to patch away the cache flush safely for these cores. Reported-by: Albin Tonnerre Reviewed-by: Catalin Marinas Signed-off-by: Will Deacon Signed-off-by: Russell King --- arch/arm/mm/proc-v6.S | 2 -- arch/arm/mm/proc-v7-2level.S | 3 ++- arch/arm/mm/proc-v7-3level.S | 3 ++- arch/arm/mm/proc-v7.S | 4 ++-- 4 files changed, 6 insertions(+), 6 deletions(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/proc-v6.S b/arch/arm/mm/proc-v6.S index bcaaa8de9325..a286d4712b57 100644 --- a/arch/arm/mm/proc-v6.S +++ b/arch/arm/mm/proc-v6.S @@ -80,12 +80,10 @@ ENTRY(cpu_v6_do_idle) mov pc, lr ENTRY(cpu_v6_dcache_clean_area) -#ifndef TLB_CAN_READ_FROM_L1_CACHE 1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry add r0, r0, #D_CACHE_LINE_SIZE subs r1, r1, #D_CACHE_LINE_SIZE bhi 1b -#endif mov pc, lr /* diff --git a/arch/arm/mm/proc-v7-2level.S b/arch/arm/mm/proc-v7-2level.S index 78f520bc0e99..9704097c450e 100644 --- a/arch/arm/mm/proc-v7-2level.S +++ b/arch/arm/mm/proc-v7-2level.S @@ -110,7 +110,8 @@ ENTRY(cpu_v7_set_pte_ext) ARM( str r3, [r0, #2048]! ) THUMB( add r0, r0, #2048 ) THUMB( str r3, [r0] ) - mcr p15, 0, r0, c7, c10, 1 @ flush_pte + ALT_SMP(mov pc,lr) + ALT_UP (mcr p15, 0, r0, c7, c10, 1) @ flush_pte #endif mov pc, lr ENDPROC(cpu_v7_set_pte_ext) diff --git a/arch/arm/mm/proc-v7-3level.S b/arch/arm/mm/proc-v7-3level.S index 6ffd78c0f9ab..363027e811d6 100644 --- a/arch/arm/mm/proc-v7-3level.S +++ b/arch/arm/mm/proc-v7-3level.S @@ -73,7 +73,8 @@ ENTRY(cpu_v7_set_pte_ext) tst r3, #1 << (55 - 32) @ L_PTE_DIRTY orreq r2, #L_PTE_RDONLY 1: strd r2, r3, [r0] - mcr p15, 0, r0, c7, c10, 1 @ flush_pte + ALT_SMP(mov pc, lr) + ALT_UP (mcr p15, 0, r0, c7, c10, 1) @ flush_pte #endif mov pc, lr ENDPROC(cpu_v7_set_pte_ext) diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S index 3a3c015f8d5c..37716b0508e1 100644 --- a/arch/arm/mm/proc-v7.S +++ b/arch/arm/mm/proc-v7.S @@ -75,14 +75,14 @@ ENTRY(cpu_v7_do_idle) ENDPROC(cpu_v7_do_idle) ENTRY(cpu_v7_dcache_clean_area) -#ifndef TLB_CAN_READ_FROM_L1_CACHE + ALT_SMP(mov pc, lr) @ MP extensions imply L1 PTW + ALT_UP(W(nop)) dcache_line_size r2, r3 1: mcr p15, 0, r0, c7, c10, 1 @ clean D entry add r0, r0, r2 subs r1, r1, r2 bhi 1b dsb -#endif mov pc, lr ENDPROC(cpu_v7_dcache_clean_area) -- cgit v1.2.1 From b6c7aabd923a17af993c5a5d5d7995f0b27c000a Mon Sep 17 00:00:00 2001 From: Russell King Date: Mon, 8 Apr 2013 11:44:57 +0100 Subject: ARM: Do 15e0d9e37c (ARM: pm: let platforms select cpu_suspend support) properly Let's do the changes properly and fix the same problem everywhere, not just for one case. Cc: # kernels containing 15e0d9e37c or equivalent Signed-off-by: Russell King --- arch/arm/mm/proc-arm920.S | 2 +- arch/arm/mm/proc-arm926.S | 2 +- arch/arm/mm/proc-mohawk.S | 2 +- arch/arm/mm/proc-sa1100.S | 2 +- arch/arm/mm/proc-v6.S | 2 +- arch/arm/mm/proc-xsc3.S | 2 +- arch/arm/mm/proc-xscale.S | 2 +- 7 files changed, 7 insertions(+), 7 deletions(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/proc-arm920.S b/arch/arm/mm/proc-arm920.S index 2c3b9421ab5e..2556cf1c2da1 100644 --- a/arch/arm/mm/proc-arm920.S +++ b/arch/arm/mm/proc-arm920.S @@ -387,7 +387,7 @@ ENTRY(cpu_arm920_set_pte_ext) /* Suspend/resume support: taken from arch/arm/plat-s3c24xx/sleep.S */ .globl cpu_arm920_suspend_size .equ cpu_arm920_suspend_size, 4 * 3 -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_ARM_CPU_SUSPEND ENTRY(cpu_arm920_do_suspend) stmfd sp!, {r4 - r6, lr} mrc p15, 0, r4, c13, c0, 0 @ PID diff --git a/arch/arm/mm/proc-arm926.S b/arch/arm/mm/proc-arm926.S index f1803f7e2972..344c8a548cc0 100644 --- a/arch/arm/mm/proc-arm926.S +++ b/arch/arm/mm/proc-arm926.S @@ -402,7 +402,7 @@ ENTRY(cpu_arm926_set_pte_ext) /* Suspend/resume support: taken from arch/arm/plat-s3c24xx/sleep.S */ .globl cpu_arm926_suspend_size .equ cpu_arm926_suspend_size, 4 * 3 -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_ARM_CPU_SUSPEND ENTRY(cpu_arm926_do_suspend) stmfd sp!, {r4 - r6, lr} mrc p15, 0, r4, c13, c0, 0 @ PID diff --git a/arch/arm/mm/proc-mohawk.S b/arch/arm/mm/proc-mohawk.S index 82f9cdc751d6..0b60dd3d742a 100644 --- a/arch/arm/mm/proc-mohawk.S +++ b/arch/arm/mm/proc-mohawk.S @@ -350,7 +350,7 @@ ENTRY(cpu_mohawk_set_pte_ext) .globl cpu_mohawk_suspend_size .equ cpu_mohawk_suspend_size, 4 * 6 -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_ARM_CPU_SUSPEND ENTRY(cpu_mohawk_do_suspend) stmfd sp!, {r4 - r9, lr} mrc p14, 0, r4, c6, c0, 0 @ clock configuration, for turbo mode diff --git a/arch/arm/mm/proc-sa1100.S b/arch/arm/mm/proc-sa1100.S index 3aa0da11fd84..d92dfd081429 100644 --- a/arch/arm/mm/proc-sa1100.S +++ b/arch/arm/mm/proc-sa1100.S @@ -172,7 +172,7 @@ ENTRY(cpu_sa1100_set_pte_ext) .globl cpu_sa1100_suspend_size .equ cpu_sa1100_suspend_size, 4 * 3 -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_ARM_CPU_SUSPEND ENTRY(cpu_sa1100_do_suspend) stmfd sp!, {r4 - r6, lr} mrc p15, 0, r4, c3, c0, 0 @ domain ID diff --git a/arch/arm/mm/proc-v6.S b/arch/arm/mm/proc-v6.S index bcaaa8de9325..5c07ee4fe3eb 100644 --- a/arch/arm/mm/proc-v6.S +++ b/arch/arm/mm/proc-v6.S @@ -138,7 +138,7 @@ ENTRY(cpu_v6_set_pte_ext) /* Suspend/resume support: taken from arch/arm/mach-s3c64xx/sleep.S */ .globl cpu_v6_suspend_size .equ cpu_v6_suspend_size, 4 * 6 -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_ARM_CPU_SUSPEND ENTRY(cpu_v6_do_suspend) stmfd sp!, {r4 - r9, lr} mrc p15, 0, r4, c13, c0, 0 @ FCSE/PID diff --git a/arch/arm/mm/proc-xsc3.S b/arch/arm/mm/proc-xsc3.S index eb93d6487f35..e8efd83b6f25 100644 --- a/arch/arm/mm/proc-xsc3.S +++ b/arch/arm/mm/proc-xsc3.S @@ -413,7 +413,7 @@ ENTRY(cpu_xsc3_set_pte_ext) .globl cpu_xsc3_suspend_size .equ cpu_xsc3_suspend_size, 4 * 6 -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_ARM_CPU_SUSPEND ENTRY(cpu_xsc3_do_suspend) stmfd sp!, {r4 - r9, lr} mrc p14, 0, r4, c6, c0, 0 @ clock configuration, for turbo mode diff --git a/arch/arm/mm/proc-xscale.S b/arch/arm/mm/proc-xscale.S index 25510361aa18..e766f889bfd6 100644 --- a/arch/arm/mm/proc-xscale.S +++ b/arch/arm/mm/proc-xscale.S @@ -528,7 +528,7 @@ ENTRY(cpu_xscale_set_pte_ext) .globl cpu_xscale_suspend_size .equ cpu_xscale_suspend_size, 4 * 6 -#ifdef CONFIG_PM_SLEEP +#ifdef CONFIG_ARM_CPU_SUSPEND ENTRY(cpu_xscale_do_suspend) stmfd sp!, {r4 - r9, lr} mrc p14, 0, r4, c6, c0, 0 @ clock configuration, for turbo mode -- cgit v1.2.1 From de40614e92bf1b0308d953387b0cb9d3a5710186 Mon Sep 17 00:00:00 2001 From: Joonsoo Kim Date: Fri, 5 Apr 2013 03:16:51 +0100 Subject: ARM: 7694/1: ARM, TCM: initialize TCM in paging_init(), instead of setup_arch() tcm_init() call iotable_init() and it use early_alloc variants which do memblock allocation. Directly using memblock allocation after initializing bootmem should not permitted, because bootmem can't know where are additinally reserved. So move tcm_init() to a safe place before initalizing bootmem. (On the U300) Tested-by: Linus Walleij Signed-off-by: Joonsoo Kim Signed-off-by: Russell King --- arch/arm/mm/mmu.c | 2 ++ arch/arm/mm/tcm.h | 17 +++++++++++++++++ 2 files changed, 19 insertions(+) create mode 100644 arch/arm/mm/tcm.h (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c index 78978945492a..a84ff763ac39 100644 --- a/arch/arm/mm/mmu.c +++ b/arch/arm/mm/mmu.c @@ -34,6 +34,7 @@ #include #include "mm.h" +#include "tcm.h" /* * empty_zero_page is a special page that is used for @@ -1277,6 +1278,7 @@ void __init paging_init(struct machine_desc *mdesc) dma_contiguous_remap(); devicemaps_init(mdesc); kmap_init(); + tcm_init(); top_pmd = pmd_off_k(0xffff0000); diff --git a/arch/arm/mm/tcm.h b/arch/arm/mm/tcm.h new file mode 100644 index 000000000000..8015ad434a40 --- /dev/null +++ b/arch/arm/mm/tcm.h @@ -0,0 +1,17 @@ +/* + * Copyright (C) 2008-2009 ST-Ericsson AB + * License terms: GNU General Public License (GPL) version 2 + * TCM memory handling for ARM systems + * + * Author: Linus Walleij + * Author: Rickard Andersson + */ + +#ifdef CONFIG_HAVE_TCM +void __init tcm_init(void); +#else +/* No TCM support, just blank inlines to be optimized out */ +inline void tcm_init(void) +{ +} +#endif -- cgit v1.2.1 From cd272d1ea71583170e95dde02c76166c7f9017e6 Mon Sep 17 00:00:00 2001 From: Illia Ragozin Date: Wed, 10 Apr 2013 19:43:34 +0100 Subject: ARM: 7696/1: Fix kexec by setting outer_cache.inv_all for Feroceon On Feroceon the L2 cache becomes non-coherent with the CPU when the L1 caches are disabled. Thus the L2 needs to be invalidated after both L1 caches are disabled. On kexec before the starting the code for relocation the kernel, the L1 caches are disabled in cpu_froc_fin (cpu_v7_proc_fin for Feroceon), but after L2 cache is never invalidated, because inv_all is not set in cache-feroceon-l2.c. So kernel relocation and decompression may has (and usually has) errors. Setting the function enables L2 invalidation and fixes the issue. Cc: Signed-off-by: Illia Ragozin Acked-by: Jason Cooper Signed-off-by: Russell King --- arch/arm/mm/cache-feroceon-l2.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/cache-feroceon-l2.c b/arch/arm/mm/cache-feroceon-l2.c index dd3d59122cc3..48bc3c0a87ce 100644 --- a/arch/arm/mm/cache-feroceon-l2.c +++ b/arch/arm/mm/cache-feroceon-l2.c @@ -343,6 +343,7 @@ void __init feroceon_l2_init(int __l2_wt_override) outer_cache.inv_range = feroceon_l2_inv_range; outer_cache.clean_range = feroceon_l2_clean_range; outer_cache.flush_range = feroceon_l2_flush_range; + outer_cache.inv_all = l2_inv_all; enable_l2(); -- cgit v1.2.1 From dd0f67f4747797f36f0c6bab7fed6a1f2448476d Mon Sep 17 00:00:00 2001 From: Joonsoo Kim Date: Fri, 5 Apr 2013 03:16:14 +0100 Subject: ARM: 7693/1: mm: clean-up in order to reduce to call kmap_high_get() In kmap_atomic(), kmap_high_get() is invoked for checking already mapped area. In __flush_dcache_page() and dma_cache_maint_page(), we explicitly call kmap_high_get() before kmap_atomic() when cache_is_vipt(), so kmap_high_get() can be invoked twice. This is useless operation, so remove one. v2: change cache_is_vipt() to cache_is_vipt_nonaliasing() in order to be self-documented Acked-by: Nicolas Pitre Signed-off-by: Joonsoo Kim Signed-off-by: Russell King --- arch/arm/mm/dma-mapping.c | 15 ++++++++------- arch/arm/mm/flush.c | 15 +++++++++------ 2 files changed, 17 insertions(+), 13 deletions(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c index c7e3759f16d3..b47dd48d8634 100644 --- a/arch/arm/mm/dma-mapping.c +++ b/arch/arm/mm/dma-mapping.c @@ -822,16 +822,17 @@ static void dma_cache_maint_page(struct page *page, unsigned long offset, if (PageHighMem(page)) { if (len + offset > PAGE_SIZE) len = PAGE_SIZE - offset; - vaddr = kmap_high_get(page); - if (vaddr) { - vaddr += offset; - op(vaddr, len, dir); - kunmap_high(page); - } else if (cache_is_vipt()) { - /* unmapped pages might still be cached */ + + if (cache_is_vipt_nonaliasing()) { vaddr = kmap_atomic(page); op(vaddr + offset, len, dir); kunmap_atomic(vaddr); + } else { + vaddr = kmap_high_get(page); + if (vaddr) { + op(vaddr + offset, len, dir); + kunmap_high(page); + } } } else { vaddr = page_address(page) + offset; diff --git a/arch/arm/mm/flush.c b/arch/arm/mm/flush.c index 1c8f7f564175..0d473cce501c 100644 --- a/arch/arm/mm/flush.c +++ b/arch/arm/mm/flush.c @@ -170,15 +170,18 @@ void __flush_dcache_page(struct address_space *mapping, struct page *page) if (!PageHighMem(page)) { __cpuc_flush_dcache_area(page_address(page), PAGE_SIZE); } else { - void *addr = kmap_high_get(page); - if (addr) { - __cpuc_flush_dcache_area(addr, PAGE_SIZE); - kunmap_high(page); - } else if (cache_is_vipt()) { - /* unmapped pages might still be cached */ + void *addr; + + if (cache_is_vipt_nonaliasing()) { addr = kmap_atomic(page); __cpuc_flush_dcache_area(addr, PAGE_SIZE); kunmap_atomic(addr); + } else { + addr = kmap_high_get(page); + if (addr) { + __cpuc_flush_dcache_area(addr, PAGE_SIZE); + kunmap_high(page); + } } } -- cgit v1.2.1 From b361d61dc1aaa9bdac0a0995e443c12146d916fd Mon Sep 17 00:00:00 2001 From: Gregory CLEMENT Date: Tue, 9 Apr 2013 13:37:20 +0100 Subject: ARM: 7695/1: mvebu: Enable pj4b on LPAE compilations pj4b cpus are LPAE capable so enable them on LPAE compilations Signed-off-by: Lior Amsalem Tested-by: Franklin Signed-off-by: Gregory CLEMENT Signed-off-by: Russell King --- arch/arm/mm/proc-v7.S | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/arm/mm') diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S index 37716b0508e1..4fa28acaf7f9 100644 --- a/arch/arm/mm/proc-v7.S +++ b/arch/arm/mm/proc-v7.S @@ -402,6 +402,8 @@ __v7_ca9mp_proc_info: __v7_proc __v7_ca9mp_setup .size __v7_ca9mp_proc_info, . - __v7_ca9mp_proc_info +#endif /* CONFIG_ARM_LPAE */ + /* * Marvell PJ4B processor. */ @@ -411,7 +413,6 @@ __v7_pj4b_proc_info: .long 0xfffffff0 __v7_proc __v7_pj4b_setup .size __v7_pj4b_proc_info, . - __v7_pj4b_proc_info -#endif /* CONFIG_ARM_LPAE */ /* * ARM Ltd. Cortex A7 processor. -- cgit v1.2.1