From: Mike Rapoport <rppt@xxxxxxxxxxxxx> Subject: powerpc: prefer memblock APIs returning virtual address Patch series "memblock: simplify several early memory allocation", v4. These patches simplify some of the early memory allocations by replacing usage of older memblock APIs with newer and shinier ones. Quite a few places in the arch/ code allocated memory using a memblock API that returns a physical address of the allocated area, then converted this physical address to a virtual one and then used memset(0) to clear the allocated range. More recent memblock APIs do all the three steps in one call and their usage simplifies the code. It's important to note that regardless of API used, the core allocation is nearly identical for any set of memblock allocators: first it tries to find a free memory with all the constraints specified by the caller and then falls back to the allocation with some or all constraints disabled. The first three patches perform the conversion of call sites that have exact requirements for the node and the possible memory range. The fourth patch is a bit one-off as it simplifies openrisc's implementation of pte_alloc_one_kernel(), and not only the memblock usage. The fifth patch takes care of simpler cases when the allocation can be satisfied with a simple call to memblock_alloc(). The sixth patch removes one-liner wrappers for memblock_alloc on arm and unicore32, as suggested by Christoph. This patch (of 6): There are a several places that allocate memory using memblock APIs that return a physical address, convert the returned address to the virtual address and frequently also memset(0) the allocated range. Update these places to use memblock allocators already returning a virtual address. Use memblock functions that clear the allocated memory instead of calling memset(0) where appropriate. The calls to memblock_alloc_base() that were not followed by memset(0) are replaced with memblock_alloc_try_nid_raw(). Since the latter does not panic() when the allocation fails, the appropriate panic() calls are added to the call sites. Link: http://lkml.kernel.org/r/1546248566-14910-2-git-send-email-rppt@xxxxxxxxxxxxx Signed-off-by: Mike Rapoport <rppt@xxxxxxxxxxxxx> Cc: Arnd Bergmann <arnd@xxxxxxxx> Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx> Cc: "David S. Miller" <davem@xxxxxxxxxxxxx> Cc: Guan Xuetao <gxt@xxxxxxxxxx> Cc: Greentime Hu <green.hu@xxxxxxxxx> Cc: Heiko Carstens <heiko.carstens@xxxxxxxxxx> Cc: Jonas Bonn <jonas@xxxxxxxxxxxx> Cc: Martin Schwidefsky <schwidefsky@xxxxxxxxxx> Cc: Michael Ellerman <mpe@xxxxxxxxxxxxxx> Cc: Michal Hocko <mhocko@xxxxxxxx> Cc: Michal Simek <monstr@xxxxxxxxx> Cc: Mark Salter <msalter@xxxxxxxxxx> Cc: Paul Mackerras <paulus@xxxxxxxxx> Cc: Rich Felker <dalias@xxxxxxxx> Cc: Russell King <linux@xxxxxxxxxxxxxxx> Cc: Stefan Kristiansson <stefan.kristiansson@xxxxxxxxxxxxx> Cc: Stafford Horne <shorne@xxxxxxxxx> Cc: Vincent Chen <deanbo422@xxxxxxxxx> Cc: Yoshinori Sato <ysato@xxxxxxxxxxxxxxxxxxxx> Cc: Christoph Hellwig <hch@xxxxxxxxxxxxx> Cc: Michal Simek <michal.simek@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- --- a/arch/powerpc/kernel/paca.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/kernel/paca.c @@ -28,7 +28,7 @@ static void *__init alloc_paca_data(unsigned long size, unsigned long align, unsigned long limit, int cpu) { - unsigned long pa; + void *ptr; int nid; /* @@ -43,17 +43,15 @@ static void *__init alloc_paca_data(unsi nid = early_cpu_to_node(cpu); } - pa = memblock_alloc_base_nid(size, align, limit, nid, MEMBLOCK_NONE); - if (!pa) { - pa = memblock_alloc_base(size, align, limit); - if (!pa) - panic("cannot allocate paca data"); - } + ptr = memblock_alloc_try_nid(size, align, MEMBLOCK_LOW_LIMIT, + limit, nid); + if (!ptr) + panic("cannot allocate paca data"); if (cpu == boot_cpuid) memblock_set_bottom_up(false); - return __va(pa); + return ptr; } #ifdef CONFIG_PPC_PSERIES @@ -119,7 +117,6 @@ static struct slb_shadow * __init new_sl } s = alloc_paca_data(sizeof(*s), L1_CACHE_BYTES, limit, cpu); - memset(s, 0, sizeof(*s)); s->persistent = cpu_to_be32(SLB_NUM_BOLTED); s->buffer_length = cpu_to_be32(sizeof(*s)); @@ -223,7 +220,6 @@ void __init allocate_paca(int cpu) paca = alloc_paca_data(sizeof(struct paca_struct), L1_CACHE_BYTES, limit, cpu); paca_ptrs[cpu] = paca; - memset(paca, 0, sizeof(struct paca_struct)); initialise_paca(paca, cpu); #ifdef CONFIG_PPC_PSERIES --- a/arch/powerpc/kernel/setup_64.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/kernel/setup_64.c @@ -933,8 +933,9 @@ static void __ref init_fallback_flush(vo * hardware prefetch runoff. We don't have a recipe for load patterns to * reliably avoid the prefetcher. */ - l1d_flush_fallback_area = __va(memblock_alloc_base(l1d_size * 2, l1d_size, limit)); - memset(l1d_flush_fallback_area, 0, l1d_size * 2); + l1d_flush_fallback_area = memblock_alloc_try_nid(l1d_size * 2, + l1d_size, MEMBLOCK_LOW_LIMIT, + limit, NUMA_NO_NODE); for_each_possible_cpu(cpu) { struct paca_struct *paca = paca_ptrs[cpu]; --- a/arch/powerpc/mm/hash_utils_64.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/mm/hash_utils_64.c @@ -908,9 +908,9 @@ static void __init htab_initialize(void) #ifdef CONFIG_DEBUG_PAGEALLOC if (debug_pagealloc_enabled()) { linear_map_hash_count = memblock_end_of_DRAM() >> PAGE_SHIFT; - linear_map_hash_slots = __va(memblock_alloc_base( - linear_map_hash_count, 1, ppc64_rma_size)); - memset(linear_map_hash_slots, 0, linear_map_hash_count); + linear_map_hash_slots = memblock_alloc_try_nid( + linear_map_hash_count, 1, MEMBLOCK_LOW_LIMIT, + ppc64_rma_size, NUMA_NO_NODE); } #endif /* CONFIG_DEBUG_PAGEALLOC */ --- a/arch/powerpc/mm/pgtable-book3e.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/mm/pgtable-book3e.c @@ -57,12 +57,8 @@ void vmemmap_remove_mapping(unsigned lon static __ref void *early_alloc_pgtable(unsigned long size) { - void *pt; - - pt = __va(memblock_alloc_base(size, size, __pa(MAX_DMA_ADDRESS))); - memset(pt, 0, size); - - return pt; + return memblock_alloc_try_nid(size, size, MEMBLOCK_LOW_LIMIT, + __pa(MAX_DMA_ADDRESS), NUMA_NO_NODE); } /* --- a/arch/powerpc/mm/pgtable-book3s64.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/mm/pgtable-book3s64.c @@ -195,11 +195,8 @@ void __init mmu_partition_table_init(voi unsigned long ptcr; BUILD_BUG_ON_MSG((PATB_SIZE_SHIFT > 36), "Partition table size too large."); - partition_tb = __va(memblock_alloc_base(patb_size, patb_size, - MEMBLOCK_ALLOC_ANYWHERE)); - /* Initialize the Partition Table with no entries */ - memset((void *)partition_tb, 0, patb_size); + partition_tb = memblock_alloc(patb_size, patb_size); /* * update partition table control register, --- a/arch/powerpc/mm/pgtable-radix.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/mm/pgtable-radix.c @@ -51,26 +51,15 @@ static int native_register_process_table static __ref void *early_alloc_pgtable(unsigned long size, int nid, unsigned long region_start, unsigned long region_end) { - unsigned long pa = 0; - void *pt; + phys_addr_t min_addr = MEMBLOCK_LOW_LIMIT; + phys_addr_t max_addr = MEMBLOCK_ALLOC_ANYWHERE; - if (region_start || region_end) /* has region hint */ - pa = memblock_alloc_range(size, size, region_start, region_end, - MEMBLOCK_NONE); - else if (nid != -1) /* has node hint */ - pa = memblock_alloc_base_nid(size, size, - MEMBLOCK_ALLOC_ANYWHERE, - nid, MEMBLOCK_NONE); + if (region_start) + min_addr = region_start; + if (region_end) + max_addr = region_end; - if (!pa) - pa = memblock_alloc_base(size, size, MEMBLOCK_ALLOC_ANYWHERE); - - BUG_ON(!pa); - - pt = __va(pa); - memset(pt, 0, size); - - return pt; + return memblock_alloc_try_nid(size, size, min_addr, max_addr, nid); } static int early_map_kernel_page(unsigned long ea, unsigned long pa, --- a/arch/powerpc/platforms/pasemi/iommu.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/platforms/pasemi/iommu.c @@ -208,7 +208,9 @@ static int __init iob_init(struct device pr_debug(" -> %s\n", __func__); /* For 2G space, 8x64 pages (2^21 bytes) is max total l2 size */ - iob_l2_base = (u32 *)__va(memblock_alloc_base(1UL<<21, 1UL<<21, 0x80000000)); + iob_l2_base = memblock_alloc_try_nid_raw(1UL << 21, 1UL << 21, + MEMBLOCK_LOW_LIMIT, 0x80000000, + NUMA_NO_NODE); pr_info("IOBMAP L2 allocated at: %p\n", iob_l2_base); @@ -269,4 +271,3 @@ void __init iommu_init_early_pasemi(void pasemi_pci_controller_ops.dma_bus_setup = pci_dma_bus_setup_pasemi; set_pci_dma_ops(&dma_iommu_ops); } - --- a/arch/powerpc/platforms/pseries/setup.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/platforms/pseries/setup.c @@ -130,8 +130,13 @@ static void __init fwnmi_init(void) * It will be used in real mode mce handler, hence it needs to be * below RMA. */ - mce_data_buf = __va(memblock_alloc_base(RTAS_ERROR_LOG_MAX * nr_cpus, - RTAS_ERROR_LOG_MAX, ppc64_rma_size)); + mce_data_buf = memblock_alloc_try_nid_raw(RTAS_ERROR_LOG_MAX * nr_cpus, + RTAS_ERROR_LOG_MAX, MEMBLOCK_LOW_LIMIT, + ppc64_rma_size, NUMA_NO_NODE); + if (!mce_data_buf) + panic("Failed to allocate %d bytes below %pa for MCE buffer\n", + RTAS_ERROR_LOG_MAX * nr_cpus, &ppc64_rma_size); + for_each_possible_cpu(i) { paca_ptrs[i]->mce_data_buf = mce_data_buf + (RTAS_ERROR_LOG_MAX * i); @@ -140,8 +145,13 @@ static void __init fwnmi_init(void) #ifdef CONFIG_PPC_BOOK3S_64 /* Allocate per cpu slb area to save old slb contents during MCE */ size = sizeof(struct slb_entry) * mmu_slb_size * nr_cpus; - slb_ptr = __va(memblock_alloc_base(size, sizeof(struct slb_entry), - ppc64_rma_size)); + slb_ptr = memblock_alloc_try_nid_raw(size, sizeof(struct slb_entry), + MEMBLOCK_LOW_LIMIT, ppc64_rma_size, + NUMA_NO_NODE); + if (!slb_ptr) + panic("Failed to allocate %zu bytes below %pa for slb area\n", + size, &ppc64_rma_size); + for_each_possible_cpu(i) paca_ptrs[i]->mce_faulty_slbs = slb_ptr + (mmu_slb_size * i); #endif --- a/arch/powerpc/sysdev/dart_iommu.c~powerpc-prefer-memblock-apis-returning-virtual-address +++ a/arch/powerpc/sysdev/dart_iommu.c @@ -251,8 +251,11 @@ static void allocate_dart(void) * 16MB (1 << 24) alignment. We allocate a full 16Mb chuck since we * will blow up an entire large page anyway in the kernel mapping. */ - dart_tablebase = __va(memblock_alloc_base(1UL<<24, - 1UL<<24, 0x80000000L)); + dart_tablebase = memblock_alloc_try_nid_raw(SZ_16M, SZ_16M, + MEMBLOCK_LOW_LIMIT, SZ_2G, + NUMA_NO_NODE); + if (!dart_tablebase) + panic("Failed to allocate 16MB below 2GB for DART table\n"); /* There is no point scanning the DART space for leaks*/ kmemleak_no_scan((void *)dart_tablebase); _
