memblock: stop using implicit alignment to SMP_CACHE_BYTES

When a memblock allocation APIs are called with align = 0, the alignment
is implicitly set to SMP_CACHE_BYTES.

Implicit alignment is done deep in the memblock allocator and it can
come as a surprise. Not that such an alignment would be wrong even
when used incorrectly but it is better to be explicit for the sake of
clarity and the prinicple of the least surprise.

Replace all such uses of memblock APIs with the 'align' parameter
explicitly set to SMP_CACHE_BYTES and stop implicit alignment assignment
in the memblock internal allocation functions.

For the case when memblock APIs are used via helper functions, e.g. like
iommu_arena_new_node() in Alpha, the helper functions were detected with
Coccinelle's help and then manually examined and updated where
appropriate.

The direct memblock APIs users were updated using the semantic patch below:

@@
expression size, min_addr, max_addr, nid;
@@
(
|
- memblock_alloc_try_nid_raw(size, 0, min_addr, max_addr, nid)
+ memblock_alloc_try_nid_raw(size, SMP_CACHE_BYTES, min_addr, max_addr,
nid)
|
- memblock_alloc_try_nid_nopanic(size, 0, min_addr, max_addr, nid)
+ memblock_alloc_try_nid_nopanic(size, SMP_CACHE_BYTES, min_addr, max_addr,
nid)
|
- memblock_alloc_try_nid(size, 0, min_addr, max_addr, nid)
+ memblock_alloc_try_nid(size, SMP_CACHE_BYTES, min_addr, max_addr, nid)
|
- memblock_alloc(size, 0)
+ memblock_alloc(size, SMP_CACHE_BYTES)
|
- memblock_alloc_raw(size, 0)
+ memblock_alloc_raw(size, SMP_CACHE_BYTES)
|
- memblock_alloc_from(size, 0, min_addr)
+ memblock_alloc_from(size, SMP_CACHE_BYTES, min_addr)
|
- memblock_alloc_nopanic(size, 0)
+ memblock_alloc_nopanic(size, SMP_CACHE_BYTES)
|
- memblock_alloc_low(size, 0)
+ memblock_alloc_low(size, SMP_CACHE_BYTES)
|
- memblock_alloc_low_nopanic(size, 0)
+ memblock_alloc_low_nopanic(size, SMP_CACHE_BYTES)
|
- memblock_alloc_from_nopanic(size, 0, min_addr)
+ memblock_alloc_from_nopanic(size, SMP_CACHE_BYTES, min_addr)
|
- memblock_alloc_node(size, 0, nid)
+ memblock_alloc_node(size, SMP_CACHE_BYTES, nid)
)

[[e-mail]: changelog update]
[[e-mail]: coding-style fixes]
[[e-mail]: fix missed uses of implicit alignment]
Link: http://lkml.kernel.org/r/20181016133656.GA10925@rapoport-lnx
Link: http://lkml.kernel.org/r/[e-mail]
Signed-off-by: Mike Rapoport <[e-mail]>
Suggested-by: Michal Hocko <[e-mail]>
Acked-by: Paul Burton <[e-mail]> [MIPS]
Acked-by: Michael Ellerman <[e-mail]> [powerpc]
Acked-by: Michal Hocko <[e-mail]>
Cc: Catalin Marinas <[e-mail]>
Cc: Chris Zankel <[e-mail]>
Cc: Geert Uytterhoeven <[e-mail]>
Cc: Guan Xuetao <[e-mail]>
Cc: Ingo Molnar <[e-mail]>
Cc: Matt Turner <[e-mail]>
Cc: Michal Simek <[e-mail]>
Cc: Richard Weinberger <[e-mail]>
Cc: Russell King <[e-mail]>
Cc: Thomas Gleixner <[e-mail]>
Cc: Tony Luck <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>
Signed-off-by: Linus Torvalds <[e-mail]>

arch/alpha/kernel/core_apecs.c

···
 	 * Window 1 is direct access 1GB at 1GB
 	 * Window 2 is scatter-gather 8MB at 8MB (for isa)
 	 */
-	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000, 0);
+	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000,
+				       SMP_CACHE_BYTES);
 	hose->sg_pci = NULL;
 	__direct_map_base = 0x40000000;
 	__direct_map_size = 0x40000000;

arch/alpha/kernel/core_lca.c

···
 	 * Note that we do not try to save any of the DMA window CSRs
 	 * before setting them, since we cannot read those CSRs on LCA.
 	 */
-	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000, 0);
+	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000,
+				       SMP_CACHE_BYTES);
 	hose->sg_pci = NULL;
 	__direct_map_base = 0x40000000;
 	__direct_map_size = 0x40000000;

arch/alpha/kernel/core_marvel.c

···
 	char *name;
 	
 	sprintf(tmp, "PCI %s PE %d PORT %d", str, pe, port);
-	name = memblock_alloc(strlen(tmp) + 1, 0);
+	name = memblock_alloc(strlen(tmp) + 1, SMP_CACHE_BYTES);
 	strcpy(name, tmp);
 
 	return name;
···
 		return NULL;
 	}
 
-	io7 = memblock_alloc(sizeof(*io7), 0);
+	io7 = memblock_alloc(sizeof(*io7), SMP_CACHE_BYTES);
 	io7->pe = pe;
 	raw_spin_lock_init(&io7->irq_lock);
 

arch/alpha/kernel/core_mcpcia.c

···
 	 * Window 1 is scatter-gather (up to) 1GB at 1GB (for pci)
 	 * Window 2 is direct access 2GB at 2GB
 	 */
-	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000, 0);
+	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000,
+				       SMP_CACHE_BYTES);
 	hose->sg_pci = iommu_arena_new(hose, 0x40000000,
-				       size_for_memory(0x40000000), 0);
+				       size_for_memory(0x40000000),
+				       SMP_CACHE_BYTES);
 
 	__direct_map_base = 0x80000000;
 	__direct_map_size = 0x80000000;

arch/alpha/kernel/core_t2.c

···
 
 	/* Note we can only do 1 SG window, as the other is for direct, so
 	   do an ISA SG area, especially for the floppy. */
-	hose->sg_isa = iommu_arena_new(hose, base, length, 0);
+	hose->sg_isa = iommu_arena_new(hose, base, length, SMP_CACHE_BYTES);
 	hose->sg_pci = NULL;
 
 	temp = (base & 0xfff00000UL) | ((base + length - 1) >> 20);

arch/alpha/kernel/core_titan.c

···
 	 * Window 1 is direct access 1GB at 2GB
 	 * Window 2 is scatter-gather 1GB at 3GB
 	 */
-	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000, 0);
+	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000,
+				       SMP_CACHE_BYTES);
 	hose->sg_isa->align_entry = 8; /* 64KB for ISA */
 
-	hose->sg_pci = iommu_arena_new(hose, 0xc0000000, 0x40000000, 0);
+	hose->sg_pci = iommu_arena_new(hose, 0xc0000000, 0x40000000,
+				       SMP_CACHE_BYTES);
 	hose->sg_pci->align_entry = 4; /* Titan caches 4 PTEs at a time */
 
 	port->wsba[0].csr = hose->sg_isa->dma_base | 3;

arch/alpha/kernel/core_tsunami.c

···
 	 * NOTE: we need the align_entry settings for Acer devices on ES40,
 	 * specifically floppy and IDE when memory is larger than 2GB.
 	 */
-	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000, 0);
+	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000,
+				       SMP_CACHE_BYTES);
 	/* Initially set for 4 PTEs, but will be overridden to 64K for ISA. */
         hose->sg_isa->align_entry = 4;
 
 	hose->sg_pci = iommu_arena_new(hose, 0x40000000,
-				       size_for_memory(0x40000000), 0);
+				       size_for_memory(0x40000000),
+				       SMP_CACHE_BYTES);
         hose->sg_pci->align_entry = 4; /* Tsunami caches 4 PTEs at a time */
 
 	__direct_map_base = 0x80000000;

arch/alpha/kernel/core_wildfire.c

···
          * ??? We ought to scale window 3 memory.
          *
          */
-        hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000, 0);
-        hose->sg_pci = iommu_arena_new(hose, 0xc0000000, 0x08000000, 0);
+	hose->sg_isa = iommu_arena_new(hose, 0x00800000, 0x00800000,
+				       SMP_CACHE_BYTES);
+	hose->sg_pci = iommu_arena_new(hose, 0xc0000000, 0x08000000,
+				       SMP_CACHE_BYTES);
 
 	pci = WILDFIRE_pci(qbbno, hoseno);
 

arch/alpha/kernel/pci-noop.c

···
 {
 	struct pci_controller *hose;
 
-	hose = memblock_alloc(sizeof(*hose), 0);
+	hose = memblock_alloc(sizeof(*hose), SMP_CACHE_BYTES);
 
 	*hose_tail = hose;
 	hose_tail = &hose->next;
···
 struct resource * __init
 alloc_resource(void)
 {
-	return memblock_alloc(sizeof(struct resource), 0);
+	return memblock_alloc(sizeof(struct resource), SMP_CACHE_BYTES);
 }
 
 SYSCALL_DEFINE3(pciconfig_iobase, long, which, unsigned long, bus,

arch/alpha/kernel/pci.c

···
 {
 	struct pci_controller *hose;
 
-	hose = memblock_alloc(sizeof(*hose), 0);
+	hose = memblock_alloc(sizeof(*hose), SMP_CACHE_BYTES);
 
 	*hose_tail = hose;
 	hose_tail = &hose->next;
···
 struct resource * __init
 alloc_resource(void)
 {
-	return memblock_alloc(sizeof(struct resource), 0);
+	return memblock_alloc(sizeof(struct resource), SMP_CACHE_BYTES);
 }
 
 

arch/alpha/kernel/pci_iommu.c

···
 		printk("%s: couldn't allocate arena from node %d\n"
 		       "    falling back to system-wide allocation\n",
 		       __func__, nid);
-		arena = memblock_alloc(sizeof(*arena), 0);
+		arena = memblock_alloc(sizeof(*arena), SMP_CACHE_BYTES);
 	}
 
 	arena->ptes = memblock_alloc_node(sizeof(*arena), align, nid);
···
 
 #else /* CONFIG_DISCONTIGMEM */
 
-	arena = memblock_alloc(sizeof(*arena), 0);
+	arena = memblock_alloc(sizeof(*arena), SMP_CACHE_BYTES);
 	arena->ptes = memblock_alloc_from(mem_size, align, 0);
 
 #endif /* CONFIG_DISCONTIGMEM */

arch/arm/kernel/setup.c

···
 		 */
 		boot_alias_start = phys_to_idmap(start);
 		if (arm_has_idmap_alias() && boot_alias_start != IDMAP_INVALID_ADDR) {
-			res = memblock_alloc(sizeof(*res), 0);
+			res = memblock_alloc(sizeof(*res), SMP_CACHE_BYTES);
 			res->name = "System RAM (boot alias)";
 			res->start = boot_alias_start;
 			res->end = phys_to_idmap(end);
···
 			request_resource(&iomem_resource, res);
 		}
 
-		res = memblock_alloc(sizeof(*res), 0);
+		res = memblock_alloc(sizeof(*res), SMP_CACHE_BYTES);
 		res->name  = "System RAM";
 		res->start = start;
 		res->end = end;

arch/arm/mach-omap2/omap_hwmod.c

···
 	u64 size;
 	int i;
 
-	provider = memblock_alloc(sizeof(*provider), 0);
+	provider = memblock_alloc(sizeof(*provider), SMP_CACHE_BYTES);
 	if (!provider)
 		return -ENOMEM;
 
···
 		of_property_count_elems_of_size(np, "reg", sizeof(u32)) / 2;
 
 	provider->addr =
-		memblock_alloc(sizeof(void *) * provider->num_addrs, 0);
+		memblock_alloc(sizeof(void *) * provider->num_addrs,
+			       SMP_CACHE_BYTES);
 	if (!provider->addr)
 		return -ENOMEM;
 
 	provider->size =
-		memblock_alloc(sizeof(u32) * provider->num_addrs, 0);
+		memblock_alloc(sizeof(u32) * provider->num_addrs,
+			       SMP_CACHE_BYTES);
 	if (!provider->size)
 		return -ENOMEM;
 

arch/arm64/kernel/setup.c

···
 	num_standard_resources = memblock.memory.cnt;
 	standard_resources = memblock_alloc_low(num_standard_resources *
 					        sizeof(*standard_resources),
-					        0);
+					        SMP_CACHE_BYTES);
 
 	for_each_memblock(memory, region) {
 		res = &standard_resources[i++];

arch/ia64/kernel/mca.c

···
 
 #define IA64_LOG_ALLOCATE(it, size) \
 	{ia64_state_log[it].isl_log[IA64_LOG_CURR_INDEX(it)] = \
-		(ia64_err_rec_t *)memblock_alloc(size, 0); \
+		(ia64_err_rec_t *)memblock_alloc(size, SMP_CACHE_BYTES); \
 	ia64_state_log[it].isl_log[IA64_LOG_NEXT_INDEX(it)] = \
-		(ia64_err_rec_t *)memblock_alloc(size, 0);}
+		(ia64_err_rec_t *)memblock_alloc(size, SMP_CACHE_BYTES);}
 #define IA64_LOG_LOCK_INIT(it) spin_lock_init(&ia64_state_log[it].isl_lock)
 #define IA64_LOG_LOCK(it)      spin_lock_irqsave(&ia64_state_log[it].isl_lock, s)
 #define IA64_LOG_UNLOCK(it)    spin_unlock_irqrestore(&ia64_state_log[it].isl_lock,s)

arch/ia64/mm/tlb.c

···
 void __init
 mmu_context_init (void)
 {
-	ia64_ctx.bitmap = memblock_alloc((ia64_ctx.max_ctx + 1) >> 3, 0);
-	ia64_ctx.flushmap = memblock_alloc((ia64_ctx.max_ctx + 1) >> 3, 0);
+	ia64_ctx.bitmap = memblock_alloc((ia64_ctx.max_ctx + 1) >> 3,
+					 SMP_CACHE_BYTES);
+	ia64_ctx.flushmap = memblock_alloc((ia64_ctx.max_ctx + 1) >> 3,
+					   SMP_CACHE_BYTES);
 }
 
 /*

arch/ia64/sn/kernel/io_common.c

···
 	if (node >= num_online_nodes())	/* Headless/memless IO nodes */
 		node = 0;
 
-	hubdev_info = (struct hubdev_info *)memblock_alloc_node(size, 0, node);
+	hubdev_info = (struct hubdev_info *)memblock_alloc_node(size,
+								SMP_CACHE_BYTES,
+								node);
 
 	npda->pdinfo = (void *)hubdev_info;
 }

arch/ia64/sn/kernel/setup.c

···
 	 */
 	for_each_online_node(cnode) {
 		nodepdaindr[cnode] =
-		    memblock_alloc_node(sizeof(nodepda_t), 0, cnode);
+		    memblock_alloc_node(sizeof(nodepda_t), SMP_CACHE_BYTES,
+					cnode);
 		memset(nodepdaindr[cnode]->phys_cpuid, -1,
 		    sizeof(nodepdaindr[cnode]->phys_cpuid));
 		spin_lock_init(&nodepdaindr[cnode]->ptc_lock);
···
 	 */
 	for (cnode = num_online_nodes(); cnode < num_cnodes; cnode++)
 		nodepdaindr[cnode] =
-		    memblock_alloc_node(sizeof(nodepda_t), 0, 0);
+		    memblock_alloc_node(sizeof(nodepda_t), SMP_CACHE_BYTES, 0);
 
 	/*
 	 * Now copy the array of nodepda pointers to each nodepda.

arch/m68k/sun3/sun3dvma.c

···
 	list_add(&(hole->list), &hole_list);
 
 	iommu_use = memblock_alloc(IOMMU_TOTAL_ENTRIES * sizeof(unsigned long),
-				   0);
+				   SMP_CACHE_BYTES);
 
 	dvma_unmap_iommu(DVMA_START, DVMA_SIZE);
 

arch/microblaze/mm/init.c

···
 	if (mem_init_done)
 		p = kzalloc(size, mask);
 	else {
-		p = memblock_alloc(size, 0);
+		p = memblock_alloc(size, SMP_CACHE_BYTES);
 		if (p)
 			memset(p, 0, size);
 	}

arch/mips/kernel/setup.c

···
 		if (end >= HIGHMEM_START)
 			end = HIGHMEM_START - 1;
 
-		res = memblock_alloc(sizeof(struct resource), 0);
+		res = memblock_alloc(sizeof(struct resource), SMP_CACHE_BYTES);
 
 		res->start = start;
 		res->end = end;

arch/powerpc/kernel/paca.c

···
 	paca_nr_cpu_ids = nr_cpu_ids;
 
 	paca_ptrs_size = sizeof(struct paca_struct *) * nr_cpu_ids;
-	paca_ptrs = __va(memblock_phys_alloc(paca_ptrs_size, 0));
+	paca_ptrs = __va(memblock_phys_alloc(paca_ptrs_size, SMP_CACHE_BYTES));
 	memset(paca_ptrs, 0x88, paca_ptrs_size);
 }
 

arch/powerpc/kernel/pci_32.c

···
 	struct property* of_prop;
 	struct device_node *dn;
 
-	of_prop = memblock_alloc(sizeof(struct property) + 256, 0);
+	of_prop = memblock_alloc(sizeof(struct property) + 256,
+				 SMP_CACHE_BYTES);
 	dn = of_find_node_by_path("/");
 	if (dn) {
 		memset(of_prop, -1, sizeof(struct property) + 256);

arch/powerpc/lib/alloc.c

···
 	if (slab_is_available())
 		p = kzalloc(size, mask);
 	else {
-		p = memblock_alloc(size, 0);
+		p = memblock_alloc(size, SMP_CACHE_BYTES);
 	}
 	return p;
 }

arch/powerpc/mm/mmu_context_nohash.c

···
 	/*
 	 * Allocate the maps used by context management
 	 */
-	context_map = memblock_alloc(CTX_MAP_SIZE, 0);
-	context_mm = memblock_alloc(sizeof(void *) * (LAST_CONTEXT + 1), 0);
+	context_map = memblock_alloc(CTX_MAP_SIZE, SMP_CACHE_BYTES);
+	context_mm = memblock_alloc(sizeof(void *) * (LAST_CONTEXT + 1),
+				    SMP_CACHE_BYTES);
 #ifdef CONFIG_SMP
-	stale_map[boot_cpuid] = memblock_alloc(CTX_MAP_SIZE, 0);
+	stale_map[boot_cpuid] = memblock_alloc(CTX_MAP_SIZE, SMP_CACHE_BYTES);
 
 	cpuhp_setup_state_nocalls(CPUHP_POWERPC_MMU_CTX_PREPARE,
 				  "powerpc/mmu/ctx:prepare",

arch/powerpc/platforms/powermac/nvram.c

···
 		printk(KERN_ERR "nvram: no address\n");
 		return -EINVAL;
 	}
-	nvram_image = memblock_alloc(NVRAM_SIZE, 0);
+	nvram_image = memblock_alloc(NVRAM_SIZE, SMP_CACHE_BYTES);
 	nvram_data = ioremap(addr, NVRAM_SIZE*2);
 	nvram_naddrs = 1; /* Make sure we get the correct case */
 

arch/powerpc/platforms/powernv/pci-ioda.c

···
 	phb_id = be64_to_cpup(prop64);
 	pr_debug("  PHB-ID  : 0x%016llx\n", phb_id);
 
-	phb = memblock_alloc(sizeof(*phb), 0);
+	phb = memblock_alloc(sizeof(*phb), SMP_CACHE_BYTES);
 
 	/* Allocate PCI controller */
 	phb->hose = hose = pcibios_alloc_controller(np);
···
 	else
 		phb->diag_data_size = PNV_PCI_DIAG_BUF_SIZE;
 
-	phb->diag_data = memblock_alloc(phb->diag_data_size, 0);
+	phb->diag_data = memblock_alloc(phb->diag_data_size, SMP_CACHE_BYTES);
 
 	/* Parse 32-bit and IO ranges (if any) */
 	pci_process_bridge_OF_ranges(hose, np, !hose->global_number);
···
 	}
 	pemap_off = size;
 	size += phb->ioda.total_pe_num * sizeof(struct pnv_ioda_pe);
-	aux = memblock_alloc(size, 0);
+	aux = memblock_alloc(size, SMP_CACHE_BYTES);
 	phb->ioda.pe_alloc = aux;
 	phb->ioda.m64_segmap = aux + m64map_off;
 	phb->ioda.m32_segmap = aux + m32map_off;

arch/powerpc/sysdev/msi_bitmap.c

···
 	if (bmp->bitmap_from_slab)
 		bmp->bitmap = kzalloc(size, GFP_KERNEL);
 	else {
-		bmp->bitmap = memblock_alloc(size, 0);
+		bmp->bitmap = memblock_alloc(size, SMP_CACHE_BYTES);
 		/* the bitmap won't be freed from memblock allocator */
 		kmemleak_not_leak(bmp->bitmap);
 	}

arch/um/drivers/net_kern.c

···
 		return 1;
 	}
 
-	new = memblock_alloc(sizeof(*new), 0);
+	new = memblock_alloc(sizeof(*new), SMP_CACHE_BYTES);
 
 	INIT_LIST_HEAD(&new->list);
 	new->index = n;

arch/um/drivers/vector_kern.c

···
 				 str, error);
 		return 1;
 	}
-	new = memblock_alloc(sizeof(*new), 0);
+	new = memblock_alloc(sizeof(*new), SMP_CACHE_BYTES);
 	INIT_LIST_HEAD(&new->list);
 	new->unit = n;
 	new->arguments = str;

arch/um/kernel/initrd.c

···
 		return 0;
 	}
 
-	area = memblock_alloc(size, 0);
+	area = memblock_alloc(size, SMP_CACHE_BYTES);
 
 	if (load_initrd(initrd, area, size) == -1)
 		return 0;

arch/unicore32/kernel/setup.c

···
 		if (mi->bank[i].size == 0)
 			continue;
 
-		res = memblock_alloc_low(sizeof(*res), 0);
+		res = memblock_alloc_low(sizeof(*res), SMP_CACHE_BYTES);
 		res->name  = "System RAM";
 		res->start = mi->bank[i].start;
 		res->end   = mi->bank[i].start + mi->bank[i].size - 1;

arch/x86/kernel/acpi/boot.c

···
 	 */
 #define HPET_RESOURCE_NAME_SIZE 9
 	hpet_res = memblock_alloc(sizeof(*hpet_res) + HPET_RESOURCE_NAME_SIZE,
-				  0);
+				  SMP_CACHE_BYTES);
 
 	hpet_res->name = (void *)&hpet_res[1];
 	hpet_res->flags = IORESOURCE_MEM;

arch/x86/kernel/apic/io_apic.c

···
 	n = IOAPIC_RESOURCE_NAME_SIZE + sizeof(struct resource);
 	n *= nr_ioapics;
 
-	mem = memblock_alloc(n, 0);
+	mem = memblock_alloc(n, SMP_CACHE_BYTES);
 	res = (void *)mem;
 
 	mem += sizeof(struct resource) * nr_ioapics;

arch/x86/kernel/e820.c

···
 	struct resource *res;
 	u64 end;
 
-	res = memblock_alloc(sizeof(*res) * e820_table->nr_entries, 0);
+	res = memblock_alloc(sizeof(*res) * e820_table->nr_entries,
+			     SMP_CACHE_BYTES);
 	e820_res = res;
 
 	for (i = 0; i < e820_table->nr_entries; i++) {

arch/x86/platform/olpc/olpc_dt.c

···
 		 * fast enough on the platforms we care about while minimizing
 		 * wasted bootmem) and hand off chunks of it to callers.
 		 */
-		res = memblock_alloc(chunk_size, 0);
+		res = memblock_alloc(chunk_size, SMP_CACHE_BYTES);
 		BUG_ON(!res);
 		prom_early_allocated += chunk_size;
 		memset(res, 0, chunk_size);

arch/xtensa/platforms/iss/network.c

···
 		return 1;
 	}
 
-	new = memblock_alloc(sizeof(*new), 0);
+	new = memblock_alloc(sizeof(*new), SMP_CACHE_BYTES);
 	if (new == NULL) {
 		pr_err("Alloc_bootmem failed\n");
 		return 1;

drivers/clk/ti/clk.c

···
 {
 	struct clk_iomap *io;
 
-	io = memblock_alloc(sizeof(*io), 0);
+	io = memblock_alloc(sizeof(*io), SMP_CACHE_BYTES);
 
 	io->mem = mem;
 

drivers/firmware/efi/memmap.c

···
 
 static phys_addr_t __init __efi_memmap_alloc_early(unsigned long size)
 {
-	return memblock_phys_alloc(size, 0);
+	return memblock_phys_alloc(size, SMP_CACHE_BYTES);
 }
 
 static phys_addr_t __init __efi_memmap_alloc_late(unsigned long size)

drivers/firmware/memmap.c

···
 {
 	struct firmware_map_entry *entry;
 
-	entry = memblock_alloc(sizeof(struct firmware_map_entry), 0);
+	entry = memblock_alloc(sizeof(struct firmware_map_entry),
+			       SMP_CACHE_BYTES);
 	if (WARN_ON(!entry))
 		return -ENOMEM;
 

drivers/macintosh/smu.c

···
 		goto fail_np;
 	}
 
-	smu = memblock_alloc(sizeof(struct smu_device), 0);
+	smu = memblock_alloc(sizeof(struct smu_device), SMP_CACHE_BYTES);
 
 	spin_lock_init(&smu->lock);
 	INIT_LIST_HEAD(&smu->cmd_list);

drivers/of/of_reserved_mem.c

···
 	 * panic()s on allocation failure.
 	 */
 	end = !end ? MEMBLOCK_ALLOC_ANYWHERE : end;
+	align = !align ? SMP_CACHE_BYTES : align;
 	base = __memblock_alloc_base(size, align, end);
 	if (!base)
 		return -ENOMEM;

include/linux/memblock.h

···
 static inline void * __init memblock_alloc_node_nopanic(phys_addr_t size,
 							int nid)
 {
-	return memblock_alloc_try_nid_nopanic(size, 0, MEMBLOCK_LOW_LIMIT,
+	return memblock_alloc_try_nid_nopanic(size, SMP_CACHE_BYTES,
+					      MEMBLOCK_LOW_LIMIT,
 					      MEMBLOCK_ALLOC_ACCESSIBLE, nid);
 }
 

init/main.c

···
 static void __init setup_command_line(char *command_line)
 {
 	saved_command_line =
-		memblock_alloc(strlen(boot_command_line) + 1, 0);
+		memblock_alloc(strlen(boot_command_line) + 1, SMP_CACHE_BYTES);
 	initcall_command_line =
-		memblock_alloc(strlen(boot_command_line) + 1, 0);
-	static_command_line = memblock_alloc(strlen(command_line) + 1, 0);
+		memblock_alloc(strlen(boot_command_line) + 1, SMP_CACHE_BYTES);
+	static_command_line = memblock_alloc(strlen(command_line) + 1,
+					     SMP_CACHE_BYTES);
 	strcpy(saved_command_line, boot_command_line);
 	strcpy(static_command_line, command_line);
 }
···
 		str_entry = strsep(&str, ",");
 		if (str_entry) {
 			pr_debug("blacklisting initcall %s\n", str_entry);
-			entry = memblock_alloc(sizeof(*entry), 0);
-			entry->buf = memblock_alloc(strlen(str_entry) + 1, 0);
+			entry = memblock_alloc(sizeof(*entry),
+					       SMP_CACHE_BYTES);
+			entry->buf = memblock_alloc(strlen(str_entry) + 1,
+						    SMP_CACHE_BYTES);
 			strcpy(entry->buf, str_entry);
 			list_add(&entry->next, &blacklisted_initcalls);
 		}

kernel/power/snapshot.c

···
 		BUG_ON(!region);
 	} else {
 		/* This allocation cannot fail */
-		region = memblock_alloc(sizeof(struct nosave_region), 0);
+		region = memblock_alloc(sizeof(struct nosave_region),
+					SMP_CACHE_BYTES);
 	}
 	region->start_pfn = start_pfn;
 	region->end_pfn = end_pfn;

lib/cpumask.c

···
  */
 void __init alloc_bootmem_cpumask_var(cpumask_var_t *mask)
 {
-	*mask = memblock_alloc(cpumask_size(), 0);
+	*mask = memblock_alloc(cpumask_size(), SMP_CACHE_BYTES);
 }
 
 /**

mm/memblock.c

···
 {
 	phys_addr_t found;
 
-	if (!align)
-		align = SMP_CACHE_BYTES;
-
 	found = memblock_find_in_range_node(size, align, start, end, nid,
 					    flags);
 	if (found && !memblock_reserve(found, size)) {
···
  * The allocation is performed from memory region limited by
  * memblock.current_limit if @max_addr == %MEMBLOCK_ALLOC_ACCESSIBLE.
  *
- * The memory block is aligned on %SMP_CACHE_BYTES if @align == 0.
- *
  * The phys address of allocated boot memory block is converted to virtual and
  * allocated memory is reset to 0.
  *
···
 	 */
 	if (WARN_ON_ONCE(slab_is_available()))
 		return kzalloc_node(size, GFP_NOWAIT, nid);
-
-	if (!align)
-		align = SMP_CACHE_BYTES;
 
 	if (max_addr > memblock.current_limit)
 		max_addr = memblock.current_limit;

mm/page_alloc.c

···
 		size = bucketsize << log2qty;
 		if (flags & HASH_EARLY) {
 			if (flags & HASH_ZERO)
-				table = memblock_alloc_nopanic(size, 0);
+				table = memblock_alloc_nopanic(size,
+							       SMP_CACHE_BYTES);
 			else
-				table = memblock_alloc_raw(size, 0);
+				table = memblock_alloc_raw(size,
+							   SMP_CACHE_BYTES);
 		} else if (hashdist) {
 			table = __vmalloc(size, gfp_flags, PAGE_KERNEL);
 		} else {

mm/percpu.c

···
 
 	/* allocate chunk */
 	chunk = memblock_alloc(sizeof(struct pcpu_chunk) +
-				    BITS_TO_LONGS(region_size >> PAGE_SHIFT),
-				    0);
+			       BITS_TO_LONGS(region_size >> PAGE_SHIFT),
+			       SMP_CACHE_BYTES);
 
 	INIT_LIST_HEAD(&chunk->list);
 
···
 	chunk->nr_pages = region_size >> PAGE_SHIFT;
 	region_bits = pcpu_chunk_map_bits(chunk);
 
-	chunk->alloc_map = memblock_alloc(BITS_TO_LONGS(region_bits) *
-					       sizeof(chunk->alloc_map[0]), 0);
-	chunk->bound_map = memblock_alloc(BITS_TO_LONGS(region_bits + 1) *
-					       sizeof(chunk->bound_map[0]), 0);
-	chunk->md_blocks = memblock_alloc(pcpu_chunk_nr_blocks(chunk) *
-					       sizeof(chunk->md_blocks[0]), 0);
+	chunk->alloc_map = memblock_alloc(BITS_TO_LONGS(region_bits) * sizeof(chunk->alloc_map[0]),
+					  SMP_CACHE_BYTES);
+	chunk->bound_map = memblock_alloc(BITS_TO_LONGS(region_bits + 1) * sizeof(chunk->bound_map[0]),
+					  SMP_CACHE_BYTES);
+	chunk->md_blocks = memblock_alloc(pcpu_chunk_nr_blocks(chunk) * sizeof(chunk->md_blocks[0]),
+					  SMP_CACHE_BYTES);
 	pcpu_init_md_blocks(chunk);
 
 	/* manage populated page bitmap */
···
 	PCPU_SETUP_BUG_ON(pcpu_verify_alloc_info(ai) < 0);
 
 	/* process group information and build config tables accordingly */
-	group_offsets = memblock_alloc(ai->nr_groups *
-					     sizeof(group_offsets[0]), 0);
-	group_sizes = memblock_alloc(ai->nr_groups *
-					   sizeof(group_sizes[0]), 0);
-	unit_map = memblock_alloc(nr_cpu_ids * sizeof(unit_map[0]), 0);
-	unit_off = memblock_alloc(nr_cpu_ids * sizeof(unit_off[0]), 0);
+	group_offsets = memblock_alloc(ai->nr_groups * sizeof(group_offsets[0]),
+				       SMP_CACHE_BYTES);
+	group_sizes = memblock_alloc(ai->nr_groups * sizeof(group_sizes[0]),
+				     SMP_CACHE_BYTES);
+	unit_map = memblock_alloc(nr_cpu_ids * sizeof(unit_map[0]),
+				  SMP_CACHE_BYTES);
+	unit_off = memblock_alloc(nr_cpu_ids * sizeof(unit_off[0]),
+				  SMP_CACHE_BYTES);
 
 	for (cpu = 0; cpu < nr_cpu_ids; cpu++)
 		unit_map[cpu] = UINT_MAX;
···
 	 * empty chunks.
 	 */
 	pcpu_nr_slots = __pcpu_size_to_slot(pcpu_unit_size) + 2;
-	pcpu_slot = memblock_alloc(
-			pcpu_nr_slots * sizeof(pcpu_slot[0]), 0);
+	pcpu_slot = memblock_alloc(pcpu_nr_slots * sizeof(pcpu_slot[0]),
+				   SMP_CACHE_BYTES);
 	for (i = 0; i < pcpu_nr_slots; i++)
 		INIT_LIST_HEAD(&pcpu_slot[i]);
 
···
 	size_sum = ai->static_size + ai->reserved_size + ai->dyn_size;
 	areas_size = PFN_ALIGN(ai->nr_groups * sizeof(void *));
 
-	areas = memblock_alloc_nopanic(areas_size, 0);
+	areas = memblock_alloc_nopanic(areas_size, SMP_CACHE_BYTES);
 	if (!areas) {
 		rc = -ENOMEM;
 		goto out_free;
···
 	/* unaligned allocations can't be freed, round up to page size */
 	pages_size = PFN_ALIGN(unit_pages * num_possible_cpus() *
 			       sizeof(pages[0]));
-	pages = memblock_alloc(pages_size, 0);
+	pages = memblock_alloc(pages_size, SMP_CACHE_BYTES);
 
 	/* allocate pages */
 	j = 0;

mm/sparse.c

···
 	if (slab_is_available())
 		section = kzalloc_node(array_size, GFP_KERNEL, nid);
 	else
-		section = memblock_alloc_node(array_size, 0, nid);
+		section = memblock_alloc_node(array_size, SMP_CACHE_BYTES,
+					      nid);
 
 	return section;
 }