From cde14bbfb3aa79b479db35bd29e6c083513d8614 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 5 Feb 2007 18:46:40 -0800 Subject: [IA64] swiotlb bug fixes This patch fixes - marking I-cache clean of pages DMAed to now only done for IA64 - broken multiple inclusion in include/asm-x86_64/swiotlb.h - missing call to mark_clean in swiotlb_sync_sg() - a (perhaps only theoretical) issue in swiotlb_dma_supported() when io_tlb_end is exactly at the end of memory Signed-off-by: Jan Beulich Signed-off-by: Andrew Morton Signed-off-by: Tony Luck --- lib/swiotlb.c | 33 ++++++++------------------------- 1 file changed, 8 insertions(+), 25 deletions(-) (limited to 'lib') diff --git a/lib/swiotlb.c b/lib/swiotlb.c index 10625785eefd..34278338aad0 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -557,25 +557,6 @@ swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir) return dev_addr; } -/* - * Since DMA is i-cache coherent, any (complete) pages that were written via - * DMA can be marked as "clean" so that lazy_mmu_prot_update() doesn't have to - * flush them when they get mapped into an executable vm-area. - */ -static void -mark_clean(void *addr, size_t size) -{ - unsigned long pg_addr, end; - - pg_addr = PAGE_ALIGN((unsigned long) addr); - end = (unsigned long) addr + size; - while (pg_addr + PAGE_SIZE <= end) { - struct page *page = virt_to_page(pg_addr); - set_bit(PG_arch_1, &page->flags); - pg_addr += PAGE_SIZE; - } -} - /* * Unmap a single streaming mode DMA translation. The dma_addr and size must * match what was provided for in a previous swiotlb_map_single call. All @@ -594,7 +575,7 @@ swiotlb_unmap_single(struct device *hwdev, dma_addr_t dev_addr, size_t size, if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) unmap_single(hwdev, dma_addr, size, dir); else if (dir == DMA_FROM_DEVICE) - mark_clean(dma_addr, size); + dma_mark_clean(dma_addr, size); } /* @@ -617,7 +598,7 @@ swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr, if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) sync_single(hwdev, dma_addr, size, dir, target); else if (dir == DMA_FROM_DEVICE) - mark_clean(dma_addr, size); + dma_mark_clean(dma_addr, size); } void @@ -648,7 +629,7 @@ swiotlb_sync_single_range(struct device *hwdev, dma_addr_t dev_addr, if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) sync_single(hwdev, dma_addr, size, dir, target); else if (dir == DMA_FROM_DEVICE) - mark_clean(dma_addr, size); + dma_mark_clean(dma_addr, size); } void @@ -698,7 +679,6 @@ swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems, dev_addr = virt_to_phys(addr); if (swiotlb_force || address_needs_mapping(hwdev, dev_addr)) { void *map = map_single(hwdev, addr, sg->length, dir); - sg->dma_address = virt_to_bus(map); if (!map) { /* Don't panic here, we expect map_sg users to do proper error handling. */ @@ -707,6 +687,7 @@ swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems, sg[0].dma_length = 0; return 0; } + sg->dma_address = virt_to_bus(map); } else sg->dma_address = dev_addr; sg->dma_length = sg->length; @@ -730,7 +711,7 @@ swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nelems, if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg)) unmap_single(hwdev, (void *) phys_to_virt(sg->dma_address), sg->dma_length, dir); else if (dir == DMA_FROM_DEVICE) - mark_clean(SG_ENT_VIRT_ADDRESS(sg), sg->dma_length); + dma_mark_clean(SG_ENT_VIRT_ADDRESS(sg), sg->dma_length); } /* @@ -752,6 +733,8 @@ swiotlb_sync_sg(struct device *hwdev, struct scatterlist *sg, if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg)) sync_single(hwdev, (void *) sg->dma_address, sg->dma_length, dir, target); + else if (dir == DMA_FROM_DEVICE) + dma_mark_clean(SG_ENT_VIRT_ADDRESS(sg), sg->dma_length); } void @@ -783,7 +766,7 @@ swiotlb_dma_mapping_error(dma_addr_t dma_addr) int swiotlb_dma_supported (struct device *hwdev, u64 mask) { - return (virt_to_phys (io_tlb_end) - 1) <= mask; + return virt_to_phys(io_tlb_end - 1) <= mask; } EXPORT_SYMBOL(swiotlb_init); -- cgit v1.2.3 From 93fbff63e62b87fe450814db41f859d60b048fb8 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 5 Feb 2007 18:49:45 -0800 Subject: [IA64] make swiotlb use bus_to_virt/virt_to_bus Convert all phys_to_virt/virt_to_phys uses to bus_to_virt/virt_to_bus, as is what is meant and what is needed in (at least) some virtualized environments like Xen. Signed-off-by: Jan Beulich Acked-by: Muli Ben-Yehuda Signed-off-by: Andrew Morton Signed-off-by: Tony Luck --- lib/swiotlb.c | 33 +++++++++++++++++---------------- 1 file changed, 17 insertions(+), 16 deletions(-) (limited to 'lib') diff --git a/lib/swiotlb.c b/lib/swiotlb.c index 34278338aad0..bc684d1fd426 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -36,7 +36,7 @@ ( (val) & ( (align) - 1))) #define SG_ENT_VIRT_ADDRESS(sg) (page_address((sg)->page) + (sg)->offset) -#define SG_ENT_PHYS_ADDRESS(SG) virt_to_phys(SG_ENT_VIRT_ADDRESS(SG)) +#define SG_ENT_PHYS_ADDRESS(sg) virt_to_bus(SG_ENT_VIRT_ADDRESS(sg)) /* * Maximum allowable number of contiguous slabs to map, @@ -163,7 +163,7 @@ swiotlb_init_with_default_size (size_t default_size) */ io_tlb_overflow_buffer = alloc_bootmem_low(io_tlb_overflow); printk(KERN_INFO "Placing software IO TLB between 0x%lx - 0x%lx\n", - virt_to_phys(io_tlb_start), virt_to_phys(io_tlb_end)); + virt_to_bus(io_tlb_start), virt_to_bus(io_tlb_end)); } void @@ -244,7 +244,7 @@ swiotlb_late_init_with_default_size (size_t default_size) printk(KERN_INFO "Placing %ldMB software IO TLB between 0x%lx - " "0x%lx\n", (io_tlb_nslabs * (1 << IO_TLB_SHIFT)) >> 20, - virt_to_phys(io_tlb_start), virt_to_phys(io_tlb_end)); + virt_to_bus(io_tlb_start), virt_to_bus(io_tlb_end)); return 0; @@ -445,7 +445,7 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, flags |= GFP_DMA; ret = (void *)__get_free_pages(flags, order); - if (ret && address_needs_mapping(hwdev, virt_to_phys(ret))) { + if (ret && address_needs_mapping(hwdev, virt_to_bus(ret))) { /* * The allocated memory isn't reachable by the device. * Fall back on swiotlb_map_single(). @@ -465,11 +465,11 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, if (swiotlb_dma_mapping_error(handle)) return NULL; - ret = phys_to_virt(handle); + ret = bus_to_virt(handle); } memset(ret, 0, size); - dev_addr = virt_to_phys(ret); + dev_addr = virt_to_bus(ret); /* Confirm address can be DMA'd by device */ if (address_needs_mapping(hwdev, dev_addr)) { @@ -525,7 +525,7 @@ swiotlb_full(struct device *dev, size_t size, int dir, int do_panic) dma_addr_t swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir) { - unsigned long dev_addr = virt_to_phys(ptr); + unsigned long dev_addr = virt_to_bus(ptr); void *map; BUG_ON(dir == DMA_NONE); @@ -546,7 +546,7 @@ swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir) map = io_tlb_overflow_buffer; } - dev_addr = virt_to_phys(map); + dev_addr = virt_to_bus(map); /* * Ensure that the address returned is DMA'ble @@ -569,7 +569,7 @@ void swiotlb_unmap_single(struct device *hwdev, dma_addr_t dev_addr, size_t size, int dir) { - char *dma_addr = phys_to_virt(dev_addr); + char *dma_addr = bus_to_virt(dev_addr); BUG_ON(dir == DMA_NONE); if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) @@ -592,7 +592,7 @@ static inline void swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr, size_t size, int dir, int target) { - char *dma_addr = phys_to_virt(dev_addr); + char *dma_addr = bus_to_virt(dev_addr); BUG_ON(dir == DMA_NONE); if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) @@ -623,7 +623,7 @@ swiotlb_sync_single_range(struct device *hwdev, dma_addr_t dev_addr, unsigned long offset, size_t size, int dir, int target) { - char *dma_addr = phys_to_virt(dev_addr) + offset; + char *dma_addr = bus_to_virt(dev_addr) + offset; BUG_ON(dir == DMA_NONE); if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) @@ -676,7 +676,7 @@ swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems, for (i = 0; i < nelems; i++, sg++) { addr = SG_ENT_VIRT_ADDRESS(sg); - dev_addr = virt_to_phys(addr); + dev_addr = virt_to_bus(addr); if (swiotlb_force || address_needs_mapping(hwdev, dev_addr)) { void *map = map_single(hwdev, addr, sg->length, dir); if (!map) { @@ -709,7 +709,8 @@ swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nelems, for (i = 0; i < nelems; i++, sg++) if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg)) - unmap_single(hwdev, (void *) phys_to_virt(sg->dma_address), sg->dma_length, dir); + unmap_single(hwdev, bus_to_virt(sg->dma_address), + sg->dma_length, dir); else if (dir == DMA_FROM_DEVICE) dma_mark_clean(SG_ENT_VIRT_ADDRESS(sg), sg->dma_length); } @@ -731,7 +732,7 @@ swiotlb_sync_sg(struct device *hwdev, struct scatterlist *sg, for (i = 0; i < nelems; i++, sg++) if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg)) - sync_single(hwdev, (void *) sg->dma_address, + sync_single(hwdev, bus_to_virt(sg->dma_address), sg->dma_length, dir, target); else if (dir == DMA_FROM_DEVICE) dma_mark_clean(SG_ENT_VIRT_ADDRESS(sg), sg->dma_length); @@ -754,7 +755,7 @@ swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg, int swiotlb_dma_mapping_error(dma_addr_t dma_addr) { - return (dma_addr == virt_to_phys(io_tlb_overflow_buffer)); + return (dma_addr == virt_to_bus(io_tlb_overflow_buffer)); } /* @@ -766,7 +767,7 @@ swiotlb_dma_mapping_error(dma_addr_t dma_addr) int swiotlb_dma_supported (struct device *hwdev, u64 mask) { - return virt_to_phys(io_tlb_end - 1) <= mask; + return virt_to_bus(io_tlb_end - 1) <= mask; } EXPORT_SYMBOL(swiotlb_init); -- cgit v1.2.3 From 563aaf064f3776623ff5e7aef511ac2eb7e5f0bb Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 5 Feb 2007 18:51:25 -0800 Subject: [IA64] swiotlb cleanup - add proper __init decoration to swiotlb's init code (and the code calling it, where not already the case) - replace uses of 'unsigned long' with dma_addr_t where appropriate - do miscellaneous simplicfication and cleanup Signed-off-by: Jan Beulich Signed-off-by: Andrew Morton Signed-off-by: Tony Luck --- arch/x86_64/kernel/pci-swiotlb.c | 2 +- lib/swiotlb.c | 56 +++++++++++++++++++++++----------------- 2 files changed, 33 insertions(+), 25 deletions(-) (limited to 'lib') diff --git a/arch/x86_64/kernel/pci-swiotlb.c b/arch/x86_64/kernel/pci-swiotlb.c index 697f0aa794b9..eb18be5a6569 100644 --- a/arch/x86_64/kernel/pci-swiotlb.c +++ b/arch/x86_64/kernel/pci-swiotlb.c @@ -29,7 +29,7 @@ struct dma_mapping_ops swiotlb_dma_ops = { .dma_supported = NULL, }; -void pci_swiotlb_init(void) +void __init pci_swiotlb_init(void) { /* don't initialize swiotlb if iommu=off (no_iommu=1) */ if (!iommu_detected && !no_iommu && end_pfn > MAX_DMA32_PFN) diff --git a/lib/swiotlb.c b/lib/swiotlb.c index bc684d1fd426..067eed5b2758 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -1,7 +1,7 @@ /* * Dynamic DMA mapping support. * - * This implementation is for IA-64 and EM64T platforms that do not support + * This implementation is a fallback for platforms that do not support * I/O TLBs (aka DMA address translation hardware). * Copyright (C) 2000 Asit Mallick * Copyright (C) 2000 Goutham Rao @@ -129,23 +129,25 @@ __setup("swiotlb=", setup_io_tlb_npages); * Statically reserve bounce buffer space and initialize bounce buffer data * structures for the software IO TLB used to implement the DMA API. */ -void -swiotlb_init_with_default_size (size_t default_size) +void __init +swiotlb_init_with_default_size(size_t default_size) { - unsigned long i; + unsigned long i, bytes; if (!io_tlb_nslabs) { io_tlb_nslabs = (default_size >> IO_TLB_SHIFT); io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE); } + bytes = io_tlb_nslabs << IO_TLB_SHIFT; + /* * Get IO TLB memory from the low pages */ - io_tlb_start = alloc_bootmem_low_pages(io_tlb_nslabs * (1 << IO_TLB_SHIFT)); + io_tlb_start = alloc_bootmem_low_pages(bytes); if (!io_tlb_start) panic("Cannot allocate SWIOTLB buffer"); - io_tlb_end = io_tlb_start + io_tlb_nslabs * (1 << IO_TLB_SHIFT); + io_tlb_end = io_tlb_start + bytes; /* * Allocate and initialize the free list array. This array is used @@ -162,12 +164,15 @@ swiotlb_init_with_default_size (size_t default_size) * Get the overflow emergency buffer */ io_tlb_overflow_buffer = alloc_bootmem_low(io_tlb_overflow); + if (!io_tlb_overflow_buffer) + panic("Cannot allocate SWIOTLB overflow buffer!\n"); + printk(KERN_INFO "Placing software IO TLB between 0x%lx - 0x%lx\n", virt_to_bus(io_tlb_start), virt_to_bus(io_tlb_end)); } -void -swiotlb_init (void) +void __init +swiotlb_init(void) { swiotlb_init_with_default_size(64 * (1<<20)); /* default to 64MB */ } @@ -178,9 +183,9 @@ swiotlb_init (void) * This should be just like above, but with some error catching. */ int -swiotlb_late_init_with_default_size (size_t default_size) +swiotlb_late_init_with_default_size(size_t default_size) { - unsigned long i, req_nslabs = io_tlb_nslabs; + unsigned long i, bytes, req_nslabs = io_tlb_nslabs; unsigned int order; if (!io_tlb_nslabs) { @@ -191,8 +196,9 @@ swiotlb_late_init_with_default_size (size_t default_size) /* * Get IO TLB memory from the low pages */ - order = get_order(io_tlb_nslabs * (1 << IO_TLB_SHIFT)); + order = get_order(io_tlb_nslabs << IO_TLB_SHIFT); io_tlb_nslabs = SLABS_PER_PAGE << order; + bytes = io_tlb_nslabs << IO_TLB_SHIFT; while ((SLABS_PER_PAGE << order) > IO_TLB_MIN_SLABS) { io_tlb_start = (char *)__get_free_pages(GFP_DMA | __GFP_NOWARN, @@ -205,13 +211,14 @@ swiotlb_late_init_with_default_size (size_t default_size) if (!io_tlb_start) goto cleanup1; - if (order != get_order(io_tlb_nslabs * (1 << IO_TLB_SHIFT))) { + if (order != get_order(bytes)) { printk(KERN_WARNING "Warning: only able to allocate %ld MB " "for software IO TLB\n", (PAGE_SIZE << order) >> 20); io_tlb_nslabs = SLABS_PER_PAGE << order; + bytes = io_tlb_nslabs << IO_TLB_SHIFT; } - io_tlb_end = io_tlb_start + io_tlb_nslabs * (1 << IO_TLB_SHIFT); - memset(io_tlb_start, 0, io_tlb_nslabs * (1 << IO_TLB_SHIFT)); + io_tlb_end = io_tlb_start + bytes; + memset(io_tlb_start, 0, bytes); /* * Allocate and initialize the free list array. This array is used @@ -242,8 +249,8 @@ swiotlb_late_init_with_default_size (size_t default_size) if (!io_tlb_overflow_buffer) goto cleanup4; - printk(KERN_INFO "Placing %ldMB software IO TLB between 0x%lx - " - "0x%lx\n", (io_tlb_nslabs * (1 << IO_TLB_SHIFT)) >> 20, + printk(KERN_INFO "Placing %luMB software IO TLB between 0x%lx - " + "0x%lx\n", bytes >> 20, virt_to_bus(io_tlb_start), virt_to_bus(io_tlb_end)); return 0; @@ -256,8 +263,8 @@ cleanup3: free_pages((unsigned long)io_tlb_list, get_order(io_tlb_nslabs * sizeof(int))); io_tlb_list = NULL; - io_tlb_end = NULL; cleanup2: + io_tlb_end = NULL; free_pages((unsigned long)io_tlb_start, order); io_tlb_start = NULL; cleanup1: @@ -433,7 +440,7 @@ void * swiotlb_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t flags) { - unsigned long dev_addr; + dma_addr_t dev_addr; void *ret; int order = get_order(size); @@ -473,8 +480,9 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, /* Confirm address can be DMA'd by device */ if (address_needs_mapping(hwdev, dev_addr)) { - printk("hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016lx\n", - (unsigned long long)*hwdev->dma_mask, dev_addr); + printk("hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016Lx\n", + (unsigned long long)*hwdev->dma_mask, + (unsigned long long)dev_addr); panic("swiotlb_alloc_coherent: allocated memory is out of " "range for device"); } @@ -504,7 +512,7 @@ swiotlb_full(struct device *dev, size_t size, int dir, int do_panic) * When the mapping is small enough return a static buffer to limit * the damage, or panic when the transfer is too big. */ - printk(KERN_ERR "DMA: Out of SW-IOMMU space for %lu bytes at " + printk(KERN_ERR "DMA: Out of SW-IOMMU space for %zu bytes at " "device %s\n", size, dev ? dev->bus_id : "?"); if (size > io_tlb_overflow && do_panic) { @@ -525,7 +533,7 @@ swiotlb_full(struct device *dev, size_t size, int dir, int do_panic) dma_addr_t swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir) { - unsigned long dev_addr = virt_to_bus(ptr); + dma_addr_t dev_addr = virt_to_bus(ptr); void *map; BUG_ON(dir == DMA_NONE); @@ -669,7 +677,7 @@ swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems, int dir) { void *addr; - unsigned long dev_addr; + dma_addr_t dev_addr; int i; BUG_ON(dir == DMA_NONE); @@ -765,7 +773,7 @@ swiotlb_dma_mapping_error(dma_addr_t dma_addr) * this function. */ int -swiotlb_dma_supported (struct device *hwdev, u64 mask) +swiotlb_dma_supported(struct device *hwdev, u64 mask) { return virt_to_bus(io_tlb_end - 1) <= mask; } -- cgit v1.2.3 From 51099005ab8e09d68a13fea8d55bc739c1040ca6 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 5 Feb 2007 18:53:04 -0800 Subject: [IA64] swiotlb abstraction (e.g. for Xen) Add abstraction so that the file can be used by environments other than IA64 and EM64T, namely for Xen. Signed-off-by: Jan Beulich Signed-off-by: Andrew Morton Signed-off-by: Tony Luck --- include/asm-ia64/swiotlb.h | 9 +++ include/asm-x86_64/swiotlb.h | 1 + lib/swiotlb.c | 184 +++++++++++++++++++++++++++++++++++-------- 3 files changed, 159 insertions(+), 35 deletions(-) create mode 100644 include/asm-ia64/swiotlb.h (limited to 'lib') diff --git a/include/asm-ia64/swiotlb.h b/include/asm-ia64/swiotlb.h new file mode 100644 index 000000000000..452c162dee4e --- /dev/null +++ b/include/asm-ia64/swiotlb.h @@ -0,0 +1,9 @@ +#ifndef _ASM_SWIOTLB_H +#define _ASM_SWIOTLB_H 1 + +#include + +#define SWIOTLB_ARCH_NEED_LATE_INIT +#define SWIOTLB_ARCH_NEED_ALLOC + +#endif /* _ASM_SWIOTLB_H */ diff --git a/include/asm-x86_64/swiotlb.h b/include/asm-x86_64/swiotlb.h index f9c589539a82..ab913ffcad56 100644 --- a/include/asm-x86_64/swiotlb.h +++ b/include/asm-x86_64/swiotlb.h @@ -44,6 +44,7 @@ extern void swiotlb_init(void); extern int swiotlb_force; #ifdef CONFIG_SWIOTLB +#define SWIOTLB_ARCH_NEED_ALLOC extern int swiotlb; #else #define swiotlb 0 diff --git a/lib/swiotlb.c b/lib/swiotlb.c index 067eed5b2758..50a438010182 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include @@ -35,8 +36,10 @@ #define OFFSET(val,align) ((unsigned long) \ ( (val) & ( (align) - 1))) +#ifndef SG_ENT_VIRT_ADDRESS #define SG_ENT_VIRT_ADDRESS(sg) (page_address((sg)->page) + (sg)->offset) #define SG_ENT_PHYS_ADDRESS(sg) virt_to_bus(SG_ENT_VIRT_ADDRESS(sg)) +#endif /* * Maximum allowable number of contiguous slabs to map, @@ -101,13 +104,25 @@ static unsigned int io_tlb_index; * We need to save away the original address corresponding to a mapped entry * for the sync operations. */ -static unsigned char **io_tlb_orig_addr; +#ifndef SWIOTLB_ARCH_HAS_IO_TLB_ADDR_T +typedef char *io_tlb_addr_t; +#define swiotlb_orig_addr_null(buffer) (!(buffer)) +#define ptr_to_io_tlb_addr(ptr) (ptr) +#define page_to_io_tlb_addr(pg, off) (page_address(pg) + (off)) +#define sg_to_io_tlb_addr(sg) SG_ENT_VIRT_ADDRESS(sg) +#endif +static io_tlb_addr_t *io_tlb_orig_addr; /* * Protect the above data structures in the map and unmap calls */ static DEFINE_SPINLOCK(io_tlb_lock); +#ifdef SWIOTLB_EXTRA_VARIABLES +SWIOTLB_EXTRA_VARIABLES; +#endif + +#ifndef SWIOTLB_ARCH_HAS_SETUP_IO_TLB_NPAGES static int __init setup_io_tlb_npages(char *str) { @@ -122,9 +137,25 @@ setup_io_tlb_npages(char *str) swiotlb_force = 1; return 1; } +#endif __setup("swiotlb=", setup_io_tlb_npages); /* make io_tlb_overflow tunable too? */ +#ifndef swiotlb_adjust_size +#define swiotlb_adjust_size(size) ((void)0) +#endif + +#ifndef swiotlb_adjust_seg +#define swiotlb_adjust_seg(start, size) ((void)0) +#endif + +#ifndef swiotlb_print_info +#define swiotlb_print_info(bytes) \ + printk(KERN_INFO "Placing %luMB software IO TLB between 0x%lx - " \ + "0x%lx\n", bytes >> 20, \ + virt_to_bus(io_tlb_start), virt_to_bus(io_tlb_end)) +#endif + /* * Statically reserve bounce buffer space and initialize bounce buffer data * structures for the software IO TLB used to implement the DMA API. @@ -138,6 +169,8 @@ swiotlb_init_with_default_size(size_t default_size) io_tlb_nslabs = (default_size >> IO_TLB_SHIFT); io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE); } + swiotlb_adjust_size(io_tlb_nslabs); + swiotlb_adjust_size(io_tlb_overflow); bytes = io_tlb_nslabs << IO_TLB_SHIFT; @@ -155,10 +188,14 @@ swiotlb_init_with_default_size(size_t default_size) * between io_tlb_start and io_tlb_end. */ io_tlb_list = alloc_bootmem(io_tlb_nslabs * sizeof(int)); - for (i = 0; i < io_tlb_nslabs; i++) + for (i = 0; i < io_tlb_nslabs; i++) { + if ( !(i % IO_TLB_SEGSIZE) ) + swiotlb_adjust_seg(io_tlb_start + (i << IO_TLB_SHIFT), + IO_TLB_SEGSIZE << IO_TLB_SHIFT); io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE); + } io_tlb_index = 0; - io_tlb_orig_addr = alloc_bootmem(io_tlb_nslabs * sizeof(char *)); + io_tlb_orig_addr = alloc_bootmem(io_tlb_nslabs * sizeof(io_tlb_addr_t)); /* * Get the overflow emergency buffer @@ -166,17 +203,21 @@ swiotlb_init_with_default_size(size_t default_size) io_tlb_overflow_buffer = alloc_bootmem_low(io_tlb_overflow); if (!io_tlb_overflow_buffer) panic("Cannot allocate SWIOTLB overflow buffer!\n"); + swiotlb_adjust_seg(io_tlb_overflow_buffer, io_tlb_overflow); - printk(KERN_INFO "Placing software IO TLB between 0x%lx - 0x%lx\n", - virt_to_bus(io_tlb_start), virt_to_bus(io_tlb_end)); + swiotlb_print_info(bytes); } +#ifndef __swiotlb_init_with_default_size +#define __swiotlb_init_with_default_size swiotlb_init_with_default_size +#endif void __init swiotlb_init(void) { - swiotlb_init_with_default_size(64 * (1<<20)); /* default to 64MB */ + __swiotlb_init_with_default_size(64 * (1<<20)); /* default to 64MB */ } +#ifdef SWIOTLB_ARCH_NEED_LATE_INIT /* * Systems with larger DMA zones (those that don't support ISA) can * initialize the swiotlb later using the slab allocator if needed. @@ -234,12 +275,12 @@ swiotlb_late_init_with_default_size(size_t default_size) io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE); io_tlb_index = 0; - io_tlb_orig_addr = (unsigned char **)__get_free_pages(GFP_KERNEL, - get_order(io_tlb_nslabs * sizeof(char *))); + io_tlb_orig_addr = (io_tlb_addr_t *)__get_free_pages(GFP_KERNEL, + get_order(io_tlb_nslabs * sizeof(io_tlb_addr_t))); if (!io_tlb_orig_addr) goto cleanup3; - memset(io_tlb_orig_addr, 0, io_tlb_nslabs * sizeof(char *)); + memset(io_tlb_orig_addr, 0, io_tlb_nslabs * sizeof(io_tlb_addr_t)); /* * Get the overflow emergency buffer @@ -249,19 +290,17 @@ swiotlb_late_init_with_default_size(size_t default_size) if (!io_tlb_overflow_buffer) goto cleanup4; - printk(KERN_INFO "Placing %luMB software IO TLB between 0x%lx - " - "0x%lx\n", bytes >> 20, - virt_to_bus(io_tlb_start), virt_to_bus(io_tlb_end)); + swiotlb_print_info(bytes); return 0; cleanup4: - free_pages((unsigned long)io_tlb_orig_addr, get_order(io_tlb_nslabs * - sizeof(char *))); + free_pages((unsigned long)io_tlb_orig_addr, + get_order(io_tlb_nslabs * sizeof(io_tlb_addr_t))); io_tlb_orig_addr = NULL; cleanup3: - free_pages((unsigned long)io_tlb_list, get_order(io_tlb_nslabs * - sizeof(int))); + free_pages((unsigned long)io_tlb_list, + get_order(io_tlb_nslabs * sizeof(int))); io_tlb_list = NULL; cleanup2: io_tlb_end = NULL; @@ -271,7 +310,9 @@ cleanup1: io_tlb_nslabs = req_nslabs; return -ENOMEM; } +#endif +#ifndef SWIOTLB_ARCH_HAS_NEEDS_MAPPING static inline int address_needs_mapping(struct device *hwdev, dma_addr_t addr) { @@ -282,11 +323,35 @@ address_needs_mapping(struct device *hwdev, dma_addr_t addr) return (addr & ~mask) != 0; } +static inline int range_needs_mapping(const void *ptr, size_t size) +{ + return swiotlb_force; +} + +static inline int order_needs_mapping(unsigned int order) +{ + return 0; +} +#endif + +static void +__sync_single(io_tlb_addr_t buffer, char *dma_addr, size_t size, int dir) +{ +#ifndef SWIOTLB_ARCH_HAS_SYNC_SINGLE + if (dir == DMA_TO_DEVICE) + memcpy(dma_addr, buffer, size); + else + memcpy(buffer, dma_addr, size); +#else + __swiotlb_arch_sync_single(buffer, dma_addr, size, dir); +#endif +} + /* * Allocates bounce buffer and returns its kernel virtual address. */ static void * -map_single(struct device *hwdev, char *buffer, size_t size, int dir) +map_single(struct device *hwdev, io_tlb_addr_t buffer, size_t size, int dir) { unsigned long flags; char *dma_addr; @@ -359,7 +424,7 @@ map_single(struct device *hwdev, char *buffer, size_t size, int dir) */ io_tlb_orig_addr[index] = buffer; if (dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL) - memcpy(dma_addr, buffer, size); + __sync_single(buffer, dma_addr, size, DMA_TO_DEVICE); return dma_addr; } @@ -373,17 +438,18 @@ unmap_single(struct device *hwdev, char *dma_addr, size_t size, int dir) unsigned long flags; int i, count, nslots = ALIGN(size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT; int index = (dma_addr - io_tlb_start) >> IO_TLB_SHIFT; - char *buffer = io_tlb_orig_addr[index]; + io_tlb_addr_t buffer = io_tlb_orig_addr[index]; /* * First, sync the memory before unmapping the entry */ - if (buffer && ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL))) + if (!swiotlb_orig_addr_null(buffer) + && ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL))) /* * bounce... copy the data back into the original buffer * and * delete the bounce buffer. */ - memcpy(buffer, dma_addr, size); + __sync_single(buffer, dma_addr, size, DMA_FROM_DEVICE); /* * Return the buffer to the free list by setting the corresponding @@ -416,18 +482,18 @@ sync_single(struct device *hwdev, char *dma_addr, size_t size, int dir, int target) { int index = (dma_addr - io_tlb_start) >> IO_TLB_SHIFT; - char *buffer = io_tlb_orig_addr[index]; + io_tlb_addr_t buffer = io_tlb_orig_addr[index]; switch (target) { case SYNC_FOR_CPU: if (likely(dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL)) - memcpy(buffer, dma_addr, size); + __sync_single(buffer, dma_addr, size, DMA_FROM_DEVICE); else BUG_ON(dir != DMA_TO_DEVICE); break; case SYNC_FOR_DEVICE: if (likely(dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL)) - memcpy(dma_addr, buffer, size); + __sync_single(buffer, dma_addr, size, DMA_TO_DEVICE); else BUG_ON(dir != DMA_FROM_DEVICE); break; @@ -436,6 +502,8 @@ sync_single(struct device *hwdev, char *dma_addr, size_t size, } } +#ifdef SWIOTLB_ARCH_NEED_ALLOC + void * swiotlb_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t flags) @@ -451,7 +519,10 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, */ flags |= GFP_DMA; - ret = (void *)__get_free_pages(flags, order); + if (!order_needs_mapping(order)) + ret = (void *)__get_free_pages(flags, order); + else + ret = NULL; if (ret && address_needs_mapping(hwdev, virt_to_bus(ret))) { /* * The allocated memory isn't reachable by the device. @@ -489,6 +560,7 @@ swiotlb_alloc_coherent(struct device *hwdev, size_t size, *dma_handle = dev_addr; return ret; } +EXPORT_SYMBOL(swiotlb_alloc_coherent); void swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr, @@ -501,6 +573,9 @@ swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr, /* DMA_TO_DEVICE to avoid memcpy in unmap_single */ swiotlb_unmap_single (hwdev, dma_handle, size, DMA_TO_DEVICE); } +EXPORT_SYMBOL(swiotlb_free_coherent); + +#endif static void swiotlb_full(struct device *dev, size_t size, int dir, int do_panic) @@ -542,13 +617,14 @@ swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir) * we can safely return the device addr and not worry about bounce * buffering it. */ - if (!address_needs_mapping(hwdev, dev_addr) && !swiotlb_force) + if (!range_needs_mapping(ptr, size) + && !address_needs_mapping(hwdev, dev_addr)) return dev_addr; /* * Oh well, have to allocate and map a bounce buffer. */ - map = map_single(hwdev, ptr, size, dir); + map = map_single(hwdev, ptr_to_io_tlb_addr(ptr), size, dir); if (!map) { swiotlb_full(hwdev, size, dir, 1); map = io_tlb_overflow_buffer; @@ -676,17 +752,16 @@ int swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems, int dir) { - void *addr; dma_addr_t dev_addr; int i; BUG_ON(dir == DMA_NONE); for (i = 0; i < nelems; i++, sg++) { - addr = SG_ENT_VIRT_ADDRESS(sg); - dev_addr = virt_to_bus(addr); - if (swiotlb_force || address_needs_mapping(hwdev, dev_addr)) { - void *map = map_single(hwdev, addr, sg->length, dir); + dev_addr = SG_ENT_PHYS_ADDRESS(sg); + if (range_needs_mapping(SG_ENT_VIRT_ADDRESS(sg), sg->length) + || address_needs_mapping(hwdev, dev_addr)) { + void *map = map_single(hwdev, sg_to_io_tlb_addr(sg), sg->length, dir); if (!map) { /* Don't panic here, we expect map_sg users to do proper error handling. */ @@ -760,6 +835,44 @@ swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg, swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_DEVICE); } +#ifdef SWIOTLB_ARCH_NEED_MAP_PAGE + +dma_addr_t +swiotlb_map_page(struct device *hwdev, struct page *page, + unsigned long offset, size_t size, + enum dma_data_direction direction) +{ + dma_addr_t dev_addr; + char *map; + + dev_addr = page_to_bus(page) + offset; + if (address_needs_mapping(hwdev, dev_addr)) { + map = map_single(hwdev, page_to_io_tlb_addr(page, offset), size, direction); + if (!map) { + swiotlb_full(hwdev, size, direction, 1); + map = io_tlb_overflow_buffer; + } + dev_addr = virt_to_bus(map); + } + + return dev_addr; +} + +void +swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr, + size_t size, enum dma_data_direction direction) +{ + char *dma_addr = bus_to_virt(dev_addr); + + BUG_ON(direction == DMA_NONE); + if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end) + unmap_single(hwdev, dma_addr, size, direction); + else if (direction == DMA_FROM_DEVICE) + dma_mark_clean(dma_addr, size); +} + +#endif + int swiotlb_dma_mapping_error(dma_addr_t dma_addr) { @@ -772,10 +885,13 @@ swiotlb_dma_mapping_error(dma_addr_t dma_addr) * during bus mastering, then you would pass 0x00ffffff as the mask to * this function. */ +#ifndef __swiotlb_dma_supported +#define __swiotlb_dma_supported(hwdev, mask) (virt_to_bus(io_tlb_end - 1) <= (mask)) +#endif int swiotlb_dma_supported(struct device *hwdev, u64 mask) { - return virt_to_bus(io_tlb_end - 1) <= mask; + return __swiotlb_dma_supported(hwdev, mask); } EXPORT_SYMBOL(swiotlb_init); @@ -790,6 +906,4 @@ EXPORT_SYMBOL_GPL(swiotlb_sync_single_range_for_device); EXPORT_SYMBOL(swiotlb_sync_sg_for_cpu); EXPORT_SYMBOL(swiotlb_sync_sg_for_device); EXPORT_SYMBOL(swiotlb_dma_mapping_error); -EXPORT_SYMBOL(swiotlb_alloc_coherent); -EXPORT_SYMBOL(swiotlb_free_coherent); EXPORT_SYMBOL(swiotlb_dma_supported); -- cgit v1.2.3 From 31b9025aa0f89b392077db3f87458fd46bcc4f58 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 18 Jan 2007 12:23:51 -0800 Subject: Kobject: make kobject apis more robust in handling NULL pointers It should be ok to pass in NULL for some kobject functions, so add error checking for all exported kobject functions to be more robust. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- lib/kobject.c | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) (limited to 'lib') diff --git a/lib/kobject.c b/lib/kobject.c index 7ce6dc138e90..9aed594bfcac 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -126,6 +126,8 @@ EXPORT_SYMBOL_GPL(kobject_get_path); */ void kobject_init(struct kobject * kobj) { + if (!kobj) + return; kref_init(&kobj->kref); INIT_LIST_HEAD(&kobj->entry); init_waitqueue_head(&kobj->poll); @@ -366,6 +368,8 @@ out: void kobject_del(struct kobject * kobj) { + if (!kobj) + return; sysfs_remove_dir(kobj); unlink(kobj); } @@ -377,6 +381,8 @@ void kobject_del(struct kobject * kobj) void kobject_unregister(struct kobject * kobj) { + if (!kobj) + return; pr_debug("kobject %s: unregistering\n",kobject_name(kobj)); kobject_uevent(kobj, KOBJ_REMOVE); kobject_del(kobj); @@ -523,6 +529,8 @@ int kset_add(struct kset * k) int kset_register(struct kset * k) { + if (!k) + return -EINVAL; kset_init(k); return kset_add(k); } @@ -535,6 +543,8 @@ int kset_register(struct kset * k) void kset_unregister(struct kset * k) { + if (!k) + return; kobject_unregister(&k->kobj); } @@ -586,6 +596,9 @@ int subsystem_register(struct subsystem * s) { int error; + if (!s) + return -EINVAL; + subsystem_init(s); pr_debug("subsystem %s: registering\n",s->kset.kobj.name); @@ -598,6 +611,8 @@ int subsystem_register(struct subsystem * s) void subsystem_unregister(struct subsystem * s) { + if (!s) + return; pr_debug("subsystem %s: unregistering\n",s->kset.kobj.name); kset_unregister(&s->kset); } @@ -612,6 +627,10 @@ void subsystem_unregister(struct subsystem * s) int subsys_create_file(struct subsystem * s, struct subsys_attribute * a) { int error = 0; + + if (!s || !a) + return -EINVAL; + if (subsys_get(s)) { error = sysfs_create_file(&s->kset.kobj,&a->attr); subsys_put(s); -- cgit v1.2.3 From c744aeae9d173a953b771a7ad5c872f91fa99dec Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Mon, 8 Jan 2007 20:16:44 +0100 Subject: driver core: Allow device_move(dev, NULL). If we allow NULL as the new parent in device_move(), we need to make sure that the device is placed into the same place as it would if it was newly registered: - Consider the device virtual tree. In order to be able to reuse code, setup_parent() has been tweaked a bit. - kobject_move() can fall back to the kset's kobject. - sysfs_move_dir() uses the sysfs root dir as fallback. Signed-off-by: Cornelia Huck Cc: Marcel Holtmann Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 76 +++++++++++++++++++++++++++++++++-------------------- fs/sysfs/dir.c | 6 ++--- lib/kobject.c | 6 ++--- 3 files changed, 52 insertions(+), 36 deletions(-) (limited to 'lib') diff --git a/drivers/base/core.c b/drivers/base/core.c index 36eedfdeb547..c60114dbe5f5 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -390,22 +390,23 @@ void device_initialize(struct device *dev) } #ifdef CONFIG_SYSFS_DEPRECATED -static int setup_parent(struct device *dev, struct device *parent) +static struct kobject * get_device_parent(struct device *dev, + struct device *parent) { /* Set the parent to the class, not the parent device */ /* this keeps sysfs from having a symlink to make old udevs happy */ if (dev->class) - dev->kobj.parent = &dev->class->subsys.kset.kobj; + return &dev->class->subsys.kset.kobj; else if (parent) - dev->kobj.parent = &parent->kobj; + return &parent->kobj; - return 0; + return NULL; } #else -static int virtual_device_parent(struct device *dev) +static struct kobject * virtual_device_parent(struct device *dev) { if (!dev->class) - return -ENODEV; + return ERR_PTR(-ENODEV); if (!dev->class->virtual_dir) { static struct kobject *virtual_dir = NULL; @@ -415,25 +416,31 @@ static int virtual_device_parent(struct device *dev) dev->class->virtual_dir = kobject_add_dir(virtual_dir, dev->class->name); } - dev->kobj.parent = dev->class->virtual_dir; - return 0; + return dev->class->virtual_dir; } -static int setup_parent(struct device *dev, struct device *parent) +static struct kobject * get_device_parent(struct device *dev, + struct device *parent) { - int error; - /* if this is a class device, and has no parent, create one */ if ((dev->class) && (parent == NULL)) { - error = virtual_device_parent(dev); - if (error) - return error; + return virtual_device_parent(dev); } else if (parent) - dev->kobj.parent = &parent->kobj; + return &parent->kobj; + return NULL; +} +#endif +static int setup_parent(struct device *dev, struct device *parent) +{ + struct kobject *kobj; + kobj = get_device_parent(dev, parent); + if (IS_ERR(kobj)) + return PTR_ERR(kobj); + if (kobj) + dev->kobj.parent = kobj; return 0; } -#endif /** * device_add - add device to device hierarchy. @@ -976,12 +983,18 @@ static int device_move_class_links(struct device *dev, sysfs_remove_link(&dev->kobj, "device"); sysfs_remove_link(&old_parent->kobj, class_name); } - error = sysfs_create_link(&dev->kobj, &new_parent->kobj, "device"); - if (error) - goto out; - error = sysfs_create_link(&new_parent->kobj, &dev->kobj, class_name); - if (error) - sysfs_remove_link(&dev->kobj, "device"); + if (new_parent) { + error = sysfs_create_link(&dev->kobj, &new_parent->kobj, + "device"); + if (error) + goto out; + error = sysfs_create_link(&new_parent->kobj, &dev->kobj, + class_name); + if (error) + sysfs_remove_link(&dev->kobj, "device"); + } + else + error = 0; out: kfree(class_name); return error; @@ -993,25 +1006,28 @@ out: /** * device_move - moves a device to a new parent * @dev: the pointer to the struct device to be moved - * @new_parent: the new parent of the device + * @new_parent: the new parent of the device (can by NULL) */ int device_move(struct device *dev, struct device *new_parent) { int error; struct device *old_parent; + struct kobject *new_parent_kobj; dev = get_device(dev); if (!dev) return -EINVAL; new_parent = get_device(new_parent); - if (!new_parent) { - error = -EINVAL; + new_parent_kobj = get_device_parent (dev, new_parent); + if (IS_ERR(new_parent_kobj)) { + error = PTR_ERR(new_parent_kobj); + put_device(new_parent); goto out; } pr_debug("DEVICE: moving '%s' to '%s'\n", dev->bus_id, - new_parent->bus_id); - error = kobject_move(&dev->kobj, &new_parent->kobj); + new_parent ? new_parent->bus_id : ""); + error = kobject_move(&dev->kobj, new_parent_kobj); if (error) { put_device(new_parent); goto out; @@ -1020,7 +1036,8 @@ int device_move(struct device *dev, struct device *new_parent) dev->parent = new_parent; if (old_parent) klist_remove(&dev->knode_parent); - klist_add_tail(&dev->knode_parent, &new_parent->klist_children); + if (new_parent) + klist_add_tail(&dev->knode_parent, &new_parent->klist_children); if (!dev->class) goto out_put; error = device_move_class_links(dev, old_parent, new_parent); @@ -1028,7 +1045,8 @@ int device_move(struct device *dev, struct device *new_parent) /* We ignore errors on cleanup since we're hosed anyway... */ device_move_class_links(dev, new_parent, old_parent); if (!kobject_move(&dev->kobj, &old_parent->kobj)) { - klist_remove(&dev->knode_parent); + if (new_parent) + klist_remove(&dev->knode_parent); if (old_parent) klist_add_tail(&dev->knode_parent, &old_parent->klist_children); diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 511edef8b321..2bab1b4ddf5a 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -378,12 +378,10 @@ int sysfs_move_dir(struct kobject *kobj, struct kobject *new_parent) struct sysfs_dirent *new_parent_sd, *sd; int error; - if (!new_parent) - return -EINVAL; - old_parent_dentry = kobj->parent ? kobj->parent->dentry : sysfs_mount->mnt_sb->s_root; - new_parent_dentry = new_parent->dentry; + new_parent_dentry = new_parent ? + new_parent->dentry : sysfs_mount->mnt_sb->s_root; again: mutex_lock(&old_parent_dentry->d_inode->i_mutex); diff --git a/lib/kobject.c b/lib/kobject.c index 9aed594bfcac..c033dc8fa9af 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -314,7 +314,7 @@ int kobject_rename(struct kobject * kobj, const char *new_name) /** * kobject_move - move object to another parent * @kobj: object in question. - * @new_parent: object's new parent + * @new_parent: object's new parent (can be NULL) */ int kobject_move(struct kobject *kobj, struct kobject *new_parent) @@ -330,8 +330,8 @@ int kobject_move(struct kobject *kobj, struct kobject *new_parent) return -EINVAL; new_parent = kobject_get(new_parent); if (!new_parent) { - error = -EINVAL; - goto out; + if (kobj->kset) + new_parent = kobject_get(&kobj->kset->kobj); } /* old object path */ devpath = kobject_get_path(kobj, GFP_KERNEL); -- cgit v1.2.3 From b067db49e1f4013ef02ef68845701b600e88a722 Mon Sep 17 00:00:00 2001 From: Mariusz Kozlowski Date: Tue, 2 Jan 2007 13:44:44 +0100 Subject: kobject: kobject_put cleanup This patch removes redundant argument checks for kobject_put(). Signed-off-by: Mariusz Kozlowski Signed-off-by: Greg Kroah-Hartman --- lib/kobject.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'lib') diff --git a/lib/kobject.c b/lib/kobject.c index c033dc8fa9af..74b8dbca150e 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -195,8 +195,7 @@ int kobject_add(struct kobject * kobj) if (error) { /* unlink does the kobject_put() for us */ unlink(kobj); - if (parent) - kobject_put(parent); + kobject_put(parent); /* be noisy on error issues */ if (error == -EEXIST) @@ -420,8 +419,7 @@ void kobject_cleanup(struct kobject * kobj) t->release(kobj); if (s) kset_put(s); - if (parent) - kobject_put(parent); + kobject_put(parent); } static void kobject_release(struct kref *kref) -- cgit v1.2.3 From b592fcfe7f06c15ec11774b5be7ce0de3aa86e73 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 24 Jan 2007 12:35:52 -0700 Subject: sysfs: Shadow directory support The problem. When implementing a network namespace I need to be able to have multiple network devices with the same name. Currently this is a problem for /sys/class/net/*. What I want is a separate /sys/class/net directory in sysfs for each network namespace, and I want to name each of them /sys/class/net. I looked and the VFS actually allows that. All that is needed is for /sys/class/net to implement a follow link method to redirect lookups to the real directory you want. Implementing a follow link method that is sensitive to the current network namespace turns out to be 3 lines of code so it looks like a clean approach. Modifying sysfs so it doesn't get in my was is a bit trickier. I am calling the concept of multiple directories all at the same path in the filesystem shadow directories. With the directory entry really at that location the shadow master. The following patch modifies sysfs so it can handle a directory structure slightly different from the kobject tree so I can implement the shadow directories for handling /sys/class/net/. Signed-off-by: Eric W. Biederman Cc: Maneesh Soni Signed-off-by: Greg Kroah-Hartman --- fs/sysfs/dir.c | 207 ++++++++++++++++++++++++++++++++++++++++-------- fs/sysfs/group.c | 1 + fs/sysfs/inode.c | 10 +++ fs/sysfs/mount.c | 2 +- fs/sysfs/sysfs.h | 5 ++ include/linux/kobject.h | 4 + include/linux/sysfs.h | 23 +++++- lib/kobject.c | 42 ++++++++-- 8 files changed, 249 insertions(+), 45 deletions(-) (limited to 'lib') diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 9ff04491e3ea..9dcdf556c99c 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -33,8 +33,7 @@ static struct dentry_operations sysfs_dentry_ops = { /* * Allocates a new sysfs_dirent and links it to the parent sysfs_dirent */ -static struct sysfs_dirent * sysfs_new_dirent(struct sysfs_dirent * parent_sd, - void * element) +static struct sysfs_dirent * __sysfs_new_dirent(void * element) { struct sysfs_dirent * sd; @@ -46,12 +45,28 @@ static struct sysfs_dirent * sysfs_new_dirent(struct sysfs_dirent * parent_sd, atomic_set(&sd->s_count, 1); atomic_set(&sd->s_event, 1); INIT_LIST_HEAD(&sd->s_children); - list_add(&sd->s_sibling, &parent_sd->s_children); + INIT_LIST_HEAD(&sd->s_sibling); sd->s_element = element; return sd; } +static void __sysfs_list_dirent(struct sysfs_dirent *parent_sd, + struct sysfs_dirent *sd) +{ + if (sd) + list_add(&sd->s_sibling, &parent_sd->s_children); +} + +static struct sysfs_dirent * sysfs_new_dirent(struct sysfs_dirent *parent_sd, + void * element) +{ + struct sysfs_dirent *sd; + sd = __sysfs_new_dirent(element); + __sysfs_list_dirent(parent_sd, sd); + return sd; +} + /* * * Return -EEXIST if there is already a sysfs element with the same name for @@ -78,14 +93,14 @@ int sysfs_dirent_exist(struct sysfs_dirent *parent_sd, } -int sysfs_make_dirent(struct sysfs_dirent * parent_sd, struct dentry * dentry, - void * element, umode_t mode, int type) +static struct sysfs_dirent * +__sysfs_make_dirent(struct dentry *dentry, void *element, mode_t mode, int type) { struct sysfs_dirent * sd; - sd = sysfs_new_dirent(parent_sd, element); + sd = __sysfs_new_dirent(element); if (!sd) - return -ENOMEM; + goto out; sd->s_mode = mode; sd->s_type = type; @@ -95,7 +110,19 @@ int sysfs_make_dirent(struct sysfs_dirent * parent_sd, struct dentry * dentry, dentry->d_op = &sysfs_dentry_ops; } - return 0; +out: + return sd; +} + +int sysfs_make_dirent(struct sysfs_dirent * parent_sd, struct dentry * dentry, + void * element, umode_t mode, int type) +{ + struct sysfs_dirent *sd; + + sd = __sysfs_make_dirent(dentry, element, mode, type); + __sysfs_list_dirent(parent_sd, sd); + + return sd ? 0 : -ENOMEM; } static int init_dir(struct inode * inode) @@ -166,11 +193,11 @@ int sysfs_create_subdir(struct kobject * k, const char * n, struct dentry ** d) /** * sysfs_create_dir - create a directory for an object. - * @parent: parent parent object. * @kobj: object we're creating directory for. + * @shadow_parent: parent parent object. */ -int sysfs_create_dir(struct kobject * kobj) +int sysfs_create_dir(struct kobject * kobj, struct dentry *shadow_parent) { struct dentry * dentry = NULL; struct dentry * parent; @@ -178,7 +205,9 @@ int sysfs_create_dir(struct kobject * kobj) BUG_ON(!kobj); - if (kobj->parent) + if (shadow_parent) + parent = shadow_parent; + else if (kobj->parent) parent = kobj->parent->dentry; else if (sysfs_mount && sysfs_mount->mnt_sb) parent = sysfs_mount->mnt_sb->s_root; @@ -299,21 +328,12 @@ void sysfs_remove_subdir(struct dentry * d) } -/** - * sysfs_remove_dir - remove an object's directory. - * @kobj: object. - * - * The only thing special about this is that we remove any files in - * the directory before we remove the directory, and we've inlined - * what used to be sysfs_rmdir() below, instead of calling separately. - */ - -void sysfs_remove_dir(struct kobject * kobj) +static void __sysfs_remove_dir(struct dentry *dentry) { - struct dentry * dentry = dget(kobj->dentry); struct sysfs_dirent * parent_sd; struct sysfs_dirent * sd, * tmp; + dget(dentry); if (!dentry) return; @@ -334,32 +354,60 @@ void sysfs_remove_dir(struct kobject * kobj) * Drop reference from dget() on entrance. */ dput(dentry); +} + +/** + * sysfs_remove_dir - remove an object's directory. + * @kobj: object. + * + * The only thing special about this is that we remove any files in + * the directory before we remove the directory, and we've inlined + * what used to be sysfs_rmdir() below, instead of calling separately. + */ + +void sysfs_remove_dir(struct kobject * kobj) +{ + __sysfs_remove_dir(kobj->dentry); kobj->dentry = NULL; } -int sysfs_rename_dir(struct kobject * kobj, const char *new_name) +int sysfs_rename_dir(struct kobject * kobj, struct dentry *new_parent, + const char *new_name) { int error = 0; - struct dentry * new_dentry, * parent; + struct dentry * new_dentry; - if (!strcmp(kobject_name(kobj), new_name)) - return -EINVAL; - - if (!kobj->parent) - return -EINVAL; + if (!new_parent) + return -EFAULT; down_write(&sysfs_rename_sem); - parent = kobj->parent->dentry; + mutex_lock(&new_parent->d_inode->i_mutex); - mutex_lock(&parent->d_inode->i_mutex); - - new_dentry = lookup_one_len(new_name, parent, strlen(new_name)); + new_dentry = lookup_one_len(new_name, new_parent, strlen(new_name)); if (!IS_ERR(new_dentry)) { - if (!new_dentry->d_inode) { + /* By allowing two different directories with the + * same d_parent we allow this routine to move + * between different shadows of the same directory + */ + if (kobj->dentry->d_parent->d_inode != new_parent->d_inode) + return -EINVAL; + else if (new_dentry->d_parent->d_inode != new_parent->d_inode) + error = -EINVAL; + else if (new_dentry == kobj->dentry) + error = -EINVAL; + else if (!new_dentry->d_inode) { error = kobject_set_name(kobj, "%s", new_name); if (!error) { + struct sysfs_dirent *sd, *parent_sd; + d_add(new_dentry, NULL); d_move(kobj->dentry, new_dentry); + + sd = kobj->dentry->d_fsdata; + parent_sd = new_parent->d_fsdata; + + list_del_init(&sd->s_sibling); + list_add(&sd->s_sibling, &parent_sd->s_children); } else d_drop(new_dentry); @@ -367,7 +415,7 @@ int sysfs_rename_dir(struct kobject * kobj, const char *new_name) error = -EEXIST; dput(new_dentry); } - mutex_unlock(&parent->d_inode->i_mutex); + mutex_unlock(&new_parent->d_inode->i_mutex); up_write(&sysfs_rename_sem); return error; @@ -546,6 +594,95 @@ static loff_t sysfs_dir_lseek(struct file * file, loff_t offset, int origin) return offset; } + +/** + * sysfs_make_shadowed_dir - Setup so a directory can be shadowed + * @kobj: object we're creating shadow of. + */ + +int sysfs_make_shadowed_dir(struct kobject *kobj, + void * (*follow_link)(struct dentry *, struct nameidata *)) +{ + struct inode *inode; + struct inode_operations *i_op; + + inode = kobj->dentry->d_inode; + if (inode->i_op != &sysfs_dir_inode_operations) + return -EINVAL; + + i_op = kmalloc(sizeof(*i_op), GFP_KERNEL); + if (!i_op) + return -ENOMEM; + + memcpy(i_op, &sysfs_dir_inode_operations, sizeof(*i_op)); + i_op->follow_link = follow_link; + + /* Locking of inode->i_op? + * Since setting i_op is a single word write and they + * are atomic we should be ok here. + */ + inode->i_op = i_op; + return 0; +} + +/** + * sysfs_create_shadow_dir - create a shadow directory for an object. + * @kobj: object we're creating directory for. + * + * sysfs_make_shadowed_dir must already have been called on this + * directory. + */ + +struct dentry *sysfs_create_shadow_dir(struct kobject *kobj) +{ + struct sysfs_dirent *sd; + struct dentry *parent, *dir, *shadow; + struct inode *inode; + + dir = kobj->dentry; + inode = dir->d_inode; + parent = dir->d_parent; + shadow = ERR_PTR(-EINVAL); + if (!sysfs_is_shadowed_inode(inode)) + goto out; + + shadow = d_alloc(parent, &dir->d_name); + if (!shadow) + goto nomem; + + sd = __sysfs_make_dirent(shadow, kobj, inode->i_mode, SYSFS_DIR); + if (!sd) + goto nomem; + + d_instantiate(shadow, igrab(inode)); + inc_nlink(inode); + inc_nlink(parent->d_inode); + shadow->d_op = &sysfs_dentry_ops; + + dget(shadow); /* Extra count - pin the dentry in core */ + +out: + return shadow; +nomem: + dput(shadow); + shadow = ERR_PTR(-ENOMEM); + goto out; +} + +/** + * sysfs_remove_shadow_dir - remove an object's directory. + * @shadow: dentry of shadow directory + * + * The only thing special about this is that we remove any files in + * the directory before we remove the directory, and we've inlined + * what used to be sysfs_rmdir() below, instead of calling separately. + */ + +void sysfs_remove_shadow_dir(struct dentry *shadow) +{ + __sysfs_remove_dir(shadow); +} + const struct file_operations sysfs_dir_operations = { .open = sysfs_dir_open, .release = sysfs_dir_close, diff --git a/fs/sysfs/group.c b/fs/sysfs/group.c index 46a277b0838e..b20951c93761 100644 --- a/fs/sysfs/group.c +++ b/fs/sysfs/group.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include "sysfs.h" diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index dbd820f9aeed..542d2bcc73df 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -33,6 +33,16 @@ static struct inode_operations sysfs_inode_operations ={ .setattr = sysfs_setattr, }; +void sysfs_delete_inode(struct inode *inode) +{ + /* Free the shadowed directory inode operations */ + if (sysfs_is_shadowed_inode(inode)) { + kfree(inode->i_op); + inode->i_op = NULL; + } + return generic_delete_inode(inode); +} + int sysfs_setattr(struct dentry * dentry, struct iattr * iattr) { struct inode * inode = dentry->d_inode; diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index a1a58b97f322..f6a87a824883 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -23,7 +23,7 @@ static void sysfs_clear_inode(struct inode *inode); static struct super_operations sysfs_ops = { .statfs = simple_statfs, - .drop_inode = generic_delete_inode, + .drop_inode = sysfs_delete_inode, .clear_inode = sysfs_clear_inode, }; diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index 39c623fdc277..fe1cbfd208ed 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -2,6 +2,7 @@ extern struct vfsmount * sysfs_mount; extern struct kmem_cache *sysfs_dir_cachep; +extern void sysfs_delete_inode(struct inode *inode); extern struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent *); extern int sysfs_create(struct dentry *, int mode, int (*init)(struct inode *)); @@ -112,3 +113,7 @@ static inline void sysfs_put(struct sysfs_dirent * sd) release_sysfs_dirent(sd); } +static inline int sysfs_is_shadowed_inode(struct inode *inode) +{ + return S_ISDIR(inode->i_mode) && inode->i_op->follow_link; +} diff --git a/include/linux/kobject.h b/include/linux/kobject.h index 76538fcf2c4e..b850e0310538 100644 --- a/include/linux/kobject.h +++ b/include/linux/kobject.h @@ -74,9 +74,13 @@ extern void kobject_init(struct kobject *); extern void kobject_cleanup(struct kobject *); extern int __must_check kobject_add(struct kobject *); +extern int __must_check kobject_shadow_add(struct kobject *, struct dentry *); extern void kobject_del(struct kobject *); extern int __must_check kobject_rename(struct kobject *, const char *new_name); +extern int __must_check kobject_shadow_rename(struct kobject *kobj, + struct dentry *new_parent, + const char *new_name); extern int __must_check kobject_move(struct kobject *, struct kobject *); extern int __must_check kobject_register(struct kobject *); diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index eee485957c0c..192de3afa96b 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -16,6 +16,7 @@ struct kobject; struct module; +struct nameidata; struct attribute { const char * name; @@ -89,13 +90,13 @@ struct sysfs_dirent { #ifdef CONFIG_SYSFS extern int __must_check -sysfs_create_dir(struct kobject *); +sysfs_create_dir(struct kobject *, struct dentry *); extern void sysfs_remove_dir(struct kobject *); extern int __must_check -sysfs_rename_dir(struct kobject *, const char *new_name); +sysfs_rename_dir(struct kobject *, struct dentry *, const char *new_name); extern int __must_check sysfs_move_dir(struct kobject *, struct kobject *); @@ -127,11 +128,17 @@ int __must_check sysfs_create_group(struct kobject *, void sysfs_remove_group(struct kobject *, const struct attribute_group *); void sysfs_notify(struct kobject * k, char *dir, char *attr); + +extern int sysfs_make_shadowed_dir(struct kobject *kobj, + void * (*follow_link)(struct dentry *, struct nameidata *)); +extern struct dentry *sysfs_create_shadow_dir(struct kobject *kobj); +extern void sysfs_remove_shadow_dir(struct dentry *dir); + extern int __must_check sysfs_init(void); #else /* CONFIG_SYSFS */ -static inline int sysfs_create_dir(struct kobject * k) +static inline int sysfs_create_dir(struct kobject * k, struct dentry *shadow) { return 0; } @@ -141,7 +148,9 @@ static inline void sysfs_remove_dir(struct kobject * k) ; } -static inline int sysfs_rename_dir(struct kobject * k, const char *new_name) +static inline int sysfs_rename_dir(struct kobject * k, + struct dentry *new_parent, + const char *new_name) { return 0; } @@ -205,6 +214,12 @@ static inline void sysfs_notify(struct kobject * k, char *dir, char *attr) { } +static inline int sysfs_make_shadowed_dir(struct kobject *kobj, + void * (*follow_link)(struct dentry *, struct nameidata *)) +{ + return 0; +} + static inline int __must_check sysfs_init(void) { return 0; diff --git a/lib/kobject.c b/lib/kobject.c index 74b8dbca150e..c2917ffe8bf1 100644 --- a/lib/kobject.c +++ b/lib/kobject.c @@ -44,11 +44,11 @@ static int populate_dir(struct kobject * kobj) return error; } -static int create_dir(struct kobject * kobj) +static int create_dir(struct kobject * kobj, struct dentry *shadow_parent) { int error = 0; if (kobject_name(kobj)) { - error = sysfs_create_dir(kobj); + error = sysfs_create_dir(kobj, shadow_parent); if (!error) { if ((error = populate_dir(kobj))) sysfs_remove_dir(kobj); @@ -158,9 +158,10 @@ static void unlink(struct kobject * kobj) /** * kobject_add - add an object to the hierarchy. * @kobj: object. + * @shadow_parent: sysfs directory to add to. */ -int kobject_add(struct kobject * kobj) +int kobject_shadow_add(struct kobject * kobj, struct dentry *shadow_parent) { int error = 0; struct kobject * parent; @@ -191,7 +192,7 @@ int kobject_add(struct kobject * kobj) } kobj->parent = parent; - error = create_dir(kobj); + error = create_dir(kobj, shadow_parent); if (error) { /* unlink does the kobject_put() for us */ unlink(kobj); @@ -212,6 +213,15 @@ int kobject_add(struct kobject * kobj) return error; } +/** + * kobject_add - add an object to the hierarchy. + * @kobj: object. + */ +int kobject_add(struct kobject * kobj) +{ + return kobject_shadow_add(kobj, NULL); +} + /** * kobject_register - initialize and add an object. @@ -304,7 +314,29 @@ int kobject_rename(struct kobject * kobj, const char *new_name) kobj = kobject_get(kobj); if (!kobj) return -EINVAL; - error = sysfs_rename_dir(kobj, new_name); + if (!kobj->parent) + return -EINVAL; + error = sysfs_rename_dir(kobj, kobj->parent->dentry, new_name); + kobject_put(kobj); + + return error; +} + +/** + * kobject_rename - change the name of an object + * @kobj: object in question. + * @new_name: object's new name + */ + +int kobject_shadow_rename(struct kobject * kobj, struct dentry *new_parent, + const char *new_name) +{ + int error = 0; + + kobj = kobject_get(kobj); + if (!kobj) + return -EINVAL; + error = sysfs_rename_dir(kobj, new_parent, new_name); kobject_put(kobj); return error; -- cgit v1.2.3 From 9ac7849e35f705830f7b016ff272b0ff1f7ff759 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 20 Jan 2007 16:00:26 +0900 Subject: devres: device resource management Implement device resource management, in short, devres. A device driver can allocate arbirary size of devres data which is associated with a release function. On driver detach, release function is invoked on the devres data, then, devres data is freed. devreses are typed by associated release functions. Some devreses are better represented by single instance of the type while others need multiple instances sharing the same release function. Both usages are supported. devreses can be grouped using devres group such that a device driver can easily release acquired resources halfway through initialization or selectively release resources (e.g. resources for port 1 out of 4 ports). This patch adds devres core including documentation and the following managed interfaces. * alloc/free : devm_kzalloc(), devm_kzfree() * IO region : devm_request_region(), devm_release_region() * IRQ : devm_request_irq(), devm_free_irq() * DMA : dmam_alloc_coherent(), dmam_free_coherent(), dmam_declare_coherent_memory(), dmam_pool_create(), dmam_pool_destroy() * PCI : pcim_enable_device(), pcim_pin_device(), pci_is_managed() * iomap : devm_ioport_map(), devm_ioport_unmap(), devm_ioremap(), devm_ioremap_nocache(), devm_iounmap(), pcim_iomap_table(), pcim_iomap(), pcim_iounmap() Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- Documentation/driver-model/devres.txt | 268 ++++++++++++++ drivers/base/Kconfig | 12 + drivers/base/Makefile | 1 + drivers/base/base.h | 1 + drivers/base/core.c | 2 + drivers/base/dd.c | 3 + drivers/base/devres.c | 644 ++++++++++++++++++++++++++++++++++ drivers/base/dma-mapping.c | 218 ++++++++++++ drivers/base/dmapool.c | 59 ++++ drivers/pci/pci.c | 127 ++++++- include/linux/device.h | 38 ++ include/linux/dma-mapping.h | 29 +- include/linux/dmapool.h | 7 + include/linux/interrupt.h | 6 + include/linux/io.h | 17 + include/linux/ioport.h | 20 ++ include/linux/pci.h | 9 + kernel/irq/manage.c | 86 +++++ kernel/resource.c | 62 ++++ lib/Makefile | 3 +- lib/iomap.c | 246 ++++++++++++- 21 files changed, 1853 insertions(+), 5 deletions(-) create mode 100644 Documentation/driver-model/devres.txt create mode 100644 drivers/base/devres.c create mode 100644 drivers/base/dma-mapping.c (limited to 'lib') diff --git a/Documentation/driver-model/devres.txt b/Documentation/driver-model/devres.txt new file mode 100644 index 000000000000..5163b85308f5 --- /dev/null +++ b/Documentation/driver-model/devres.txt @@ -0,0 +1,268 @@ +Devres - Managed Device Resource +================================ + +Tejun Heo + +First draft 10 January 2007 + + +1. Intro : Huh? Devres? +2. Devres : Devres in a nutshell +3. Devres Group : Group devres'es and release them together +4. Details : Life time rules, calling context, ... +5. Overhead : How much do we have to pay for this? +6. List of managed interfaces : Currently implemented managed interfaces + + + 1. Intro + -------- + +devres came up while trying to convert libata to use iomap. Each +iomapped address should be kept and unmapped on driver detach. For +example, a plain SFF ATA controller (that is, good old PCI IDE) in +native mode makes use of 5 PCI BARs and all of them should be +maintained. + +As with many other device drivers, libata low level drivers have +sufficient bugs in ->remove and ->probe failure path. Well, yes, +that's probably because libata low level driver developers are lazy +bunch, but aren't all low level driver developers? After spending a +day fiddling with braindamaged hardware with no document or +braindamaged document, if it's finally working, well, it's working. + +For one reason or another, low level drivers don't receive as much +attention or testing as core code, and bugs on driver detach or +initilaization failure doesn't happen often enough to be noticeable. +Init failure path is worse because it's much less travelled while +needs to handle multiple entry points. + +So, many low level drivers end up leaking resources on driver detach +and having half broken failure path implementation in ->probe() which +would leak resources or even cause oops when failure occurs. iomap +adds more to this mix. So do msi and msix. + + + 2. Devres + --------- + +devres is basically linked list of arbitrarily sized memory areas +associated with a struct device. Each devres entry is associated with +a release function. A devres can be released in several ways. No +matter what, all devres entries are released on driver detach. On +release, the associated release function is invoked and then the +devres entry is freed. + +Managed interface is created for resources commonly used by device +drivers using devres. For example, coherent DMA memory is acquired +using dma_alloc_coherent(). The managed version is called +dmam_alloc_coherent(). It is identical to dma_alloc_coherent() except +for the DMA memory allocated using it is managed and will be +automatically released on driver detach. Implementation looks like +the following. + + struct dma_devres { + size_t size; + void *vaddr; + dma_addr_t dma_handle; + }; + + static void dmam_coherent_release(struct device *dev, void *res) + { + struct dma_devres *this = res; + + dma_free_coherent(dev, this->size, this->vaddr, this->dma_handle); + } + + dmam_alloc_coherent(dev, size, dma_handle, gfp) + { + struct dma_devres *dr; + void *vaddr; + + dr = devres_alloc(dmam_coherent_release, sizeof(*dr), gfp); + ... + + /* alloc DMA memory as usual */ + vaddr = dma_alloc_coherent(...); + ... + + /* record size, vaddr, dma_handle in dr */ + dr->vaddr = vaddr; + ... + + devres_add(dev, dr); + + return vaddr; + } + +If a driver uses dmam_alloc_coherent(), the area is guaranteed to be +freed whether initialization fails half-way or the device gets +detached. If most resources are acquired using managed interface, a +driver can have much simpler init and exit code. Init path basically +looks like the following. + + my_init_one() + { + struct mydev *d; + + d = devm_kzalloc(dev, sizeof(*d), GFP_KERNEL); + if (!d) + return -ENOMEM; + + d->ring = dmam_alloc_coherent(...); + if (!d->ring) + return -ENOMEM; + + if (check something) + return -EINVAL; + ... + + return register_to_upper_layer(d); + } + +And exit path, + + my_remove_one() + { + unregister_from_upper_layer(d); + shutdown_my_hardware(); + } + +As shown above, low level drivers can be simplified a lot by using +devres. Complexity is shifted from less maintained low level drivers +to better maintained higher layer. Also, as init failure path is +shared with exit path, both can get more testing. + + + 3. Devres group + --------------- + +Devres entries can be grouped using devres group. When a group is +released, all contained normal devres entries and properly nested +groups are released. One usage is to rollback series of acquired +resources on failure. For example, + + if (!devres_open_group(dev, NULL, GFP_KERNEL)) + return -ENOMEM; + + acquire A; + if (failed) + goto err; + + acquire B; + if (failed) + goto err; + ... + + devres_remove_group(dev, NULL); + return 0; + + err: + devres_release_group(dev, NULL); + return err_code; + +As resource acquision failure usually means probe failure, constructs +like above are usually useful in midlayer driver (e.g. libata core +layer) where interface function shouldn't have side effect on failure. +For LLDs, just returning error code suffices in most cases. + +Each group is identified by void *id. It can either be explicitly +specified by @id argument to devres_open_group() or automatically +created by passing NULL as @id as in the above example. In both +cases, devres_open_group() returns the group's id. The returned id +can be passed to other devres functions to select the target group. +If NULL is given to those functions, the latest open group is +selected. + +For example, you can do something like the following. + + int my_midlayer_create_something() + { + if (!devres_open_group(dev, my_midlayer_create_something, GFP_KERNEL)) + return -ENOMEM; + + ... + + devres_close_group(dev, my_midlayer_something); + return 0; + } + + void my_midlayer_destroy_something() + { + devres_release_group(dev, my_midlayer_create_soemthing); + } + + + 4. Details + ---------- + +Lifetime of a devres entry begins on devres allocation and finishes +when it is released or destroyed (removed and freed) - no reference +counting. + +devres core guarantees atomicity to all basic devres operations and +has support for single-instance devres types (atomic +lookup-and-add-if-not-found). Other than that, synchronizing +concurrent accesses to allocated devres data is caller's +responsibility. This is usually non-issue because bus ops and +resource allocations already do the job. + +For an example of single-instance devres type, read pcim_iomap_table() +in lib/iomap.c. + +All devres interface functions can be called without context if the +right gfp mask is given. + + + 5. Overhead + ----------- + +Each devres bookkeeping info is allocated together with requested data +area. With debug option turned off, bookkeeping info occupies 16 +bytes on 32bit machines and 24 bytes on 64bit (three pointers rounded +up to ull alignment). If singly linked list is used, it can be +reduced to two pointers (8 bytes on 32bit, 16 bytes on 64bit). + +Each devres group occupies 8 pointers. It can be reduced to 6 if +singly linked list is used. + +Memory space overhead on ahci controller with two ports is between 300 +and 400 bytes on 32bit machine after naive conversion (we can +certainly invest a bit more effort into libata core layer). + + + 6. List of managed interfaces + ----------------------------- + +IO region + devm_request_region() + devm_request_mem_region() + devm_release_region() + devm_release_mem_region() + +IRQ + devm_request_irq() + devm_free_irq() + +DMA + dmam_alloc_coherent() + dmam_free_coherent() + dmam_alloc_noncoherent() + dmam_free_noncoherent() + dmam_declare_coherent_memory() + dmam_pool_create() + dmam_pool_destroy() + +PCI + pcim_enable_device() : after success, all PCI ops become managed + pcim_pin_device() : keep PCI device enabled after release + +IOMAP + devm_ioport_map() + devm_ioport_unmap() + devm_ioremap() + devm_ioremap_nocache() + devm_iounmap() + pcim_iomap() + pcim_iounmap() + pcim_iomap_table() : array of mapped addresses indexed by BAR + pcim_iomap_regions() : do request_region() and iomap() on multiple BARs diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index 1429f3a2629e..5d6312e33490 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -37,6 +37,18 @@ config DEBUG_DRIVER If you are unsure about this, say N here. +config DEBUG_DEVRES + bool "Managed device resources verbose debug messages" + depends on DEBUG_KERNEL + help + This option enables kernel parameter devres.log. If set to + non-zero, devres debug messages are printed. Select this if + you are having a problem with devres or want to debug + resource management for a managed device. devres.log can be + switched on and off from sysfs node. + + If you are unsure about this, Say N here. + config SYS_HYPERVISOR bool default n diff --git a/drivers/base/Makefile b/drivers/base/Makefile index 7bbb9eeda235..e9eb7382ac3a 100644 --- a/drivers/base/Makefile +++ b/drivers/base/Makefile @@ -3,6 +3,7 @@ obj-y := core.o sys.o bus.o dd.o \ driver.o class.o platform.o \ cpu.o firmware.o init.o map.o dmapool.o \ + dma-mapping.o devres.o \ attribute_container.o transport_class.o obj-y += power/ obj-$(CONFIG_ISA) += isa.o diff --git a/drivers/base/base.h b/drivers/base/base.h index d26644a59537..de7e1442ce60 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -44,3 +44,4 @@ struct class_device_attribute *to_class_dev_attr(struct attribute *_attr) extern char *make_class_name(const char *name, struct kobject *kobj); +extern void devres_release_all(struct device *dev); diff --git a/drivers/base/core.c b/drivers/base/core.c index e13614241c9e..a8ac34ba6107 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -428,6 +428,8 @@ void device_initialize(struct device *dev) INIT_LIST_HEAD(&dev->dma_pools); INIT_LIST_HEAD(&dev->node); init_MUTEX(&dev->sem); + spin_lock_init(&dev->devres_lock); + INIT_LIST_HEAD(&dev->devres_head); device_init_wakeup(dev, 0); set_dev_node(dev, -1); } diff --git a/drivers/base/dd.c b/drivers/base/dd.c index b5bf243d9cd6..6a48824e43ff 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -112,6 +112,7 @@ static int really_probe(void *void_data) atomic_inc(&probe_count); pr_debug("%s: Probing driver %s with device %s\n", drv->bus->name, drv->name, dev->bus_id); + WARN_ON(!list_empty(&dev->devres_head)); dev->driver = drv; if (driver_sysfs_add(dev)) { @@ -137,6 +138,7 @@ static int really_probe(void *void_data) goto done; probe_failed: + devres_release_all(dev); driver_sysfs_remove(dev); dev->driver = NULL; @@ -327,6 +329,7 @@ static void __device_release_driver(struct device * dev) dev->bus->remove(dev); else if (drv->remove) drv->remove(dev); + devres_release_all(dev); dev->driver = NULL; put_driver(drv); } diff --git a/drivers/base/devres.c b/drivers/base/devres.c new file mode 100644 index 000000000000..e177c9533b6c --- /dev/null +++ b/drivers/base/devres.c @@ -0,0 +1,644 @@ +/* + * drivers/base/devres.c - device resource management + * + * Copyright (c) 2006 SUSE Linux Products GmbH + * Copyright (c) 2006 Tejun Heo + * + * This file is released under the GPLv2. + */ + +#include +#include + +struct devres_node { + struct list_head entry; + dr_release_t release; +#ifdef CONFIG_DEBUG_DEVRES + const char *name; + size_t size; +#endif +}; + +struct devres { + struct devres_node node; + /* -- 3 pointers */ + unsigned long long data[]; /* guarantee ull alignment */ +}; + +struct devres_group { + struct devres_node node[2]; + void *id; + int color; + /* -- 8 pointers */ +}; + +#ifdef CONFIG_DEBUG_DEVRES +static int log_devres = 0; +module_param_named(log, log_devres, int, S_IRUGO | S_IWUSR); + +static void set_node_dbginfo(struct devres_node *node, const char *name, + size_t size) +{ + node->name = name; + node->size = size; +} + +static void devres_log(struct device *dev, struct devres_node *node, + const char *op) +{ + if (unlikely(log_devres)) + dev_printk(KERN_ERR, dev, "DEVRES %3s %p %s (%lu bytes)\n", + op, node, node->name, (unsigned long)node->size); +} +#else /* CONFIG_DEBUG_DEVRES */ +#define set_node_dbginfo(node, n, s) do {} while (0) +#define devres_log(dev, node, op) do {} while (0) +#endif /* CONFIG_DEBUG_DEVRES */ + +/* + * Release functions for devres group. These callbacks are used only + * for identification. + */ +static void group_open_release(struct device *dev, void *res) +{ + /* noop */ +} + +static void group_close_release(struct device *dev, void *res) +{ + /* noop */ +} + +static struct devres_group * node_to_group(struct devres_node *node) +{ + if (node->release == &group_open_release) + return container_of(node, struct devres_group, node[0]); + if (node->release == &group_close_release) + return container_of(node, struct devres_group, node[1]); + return NULL; +} + +static __always_inline struct devres * alloc_dr(dr_release_t release, + size_t size, gfp_t gfp) +{ + size_t tot_size = sizeof(struct devres) + size; + struct devres *dr; + + dr = kmalloc_track_caller(tot_size, gfp); + if (unlikely(!dr)) + return NULL; + + memset(dr, 0, tot_size); + INIT_LIST_HEAD(&dr->node.entry); + dr->node.release = release; + return dr; +} + +static void add_dr(struct device *dev, struct devres_node *node) +{ + devres_log(dev, node, "ADD"); + BUG_ON(!list_empty(&node->entry)); + list_add_tail(&node->entry, &dev->devres_head); +} + +/** + * devres_alloc - Allocate device resource data + * @release: Release function devres will be associated with + * @size: Allocation size + * @gfp: Allocation flags + * + * allocate devres of @size bytes. The allocated area is zeroed, then + * associated with @release. The returned pointer can be passed to + * other devres_*() functions. + * + * RETURNS: + * Pointer to allocated devres on success, NULL on failure. + */ +#ifdef CONFIG_DEBUG_DEVRES +void * __devres_alloc(dr_release_t release, size_t size, gfp_t gfp, + const char *name) +{ + struct devres *dr; + + dr = alloc_dr(release, size, gfp); + if (unlikely(!dr)) + return NULL; + set_node_dbginfo(&dr->node, name, size); + return dr->data; +} +EXPORT_SYMBOL_GPL(__devres_alloc); +#else +void * devres_alloc(dr_release_t release, size_t size, gfp_t gfp) +{ + struct devres *dr; + + dr = alloc_dr(release, size, gfp); + if (unlikely(!dr)) + return NULL; + return dr->data; +} +EXPORT_SYMBOL_GPL(devres_alloc); +#endif + +/** + * devres_free - Free device resource data + * @res: Pointer to devres data to free + * + * Free devres created with devres_alloc(). + */ +void devres_free(void *res) +{ + if (res) { + struct devres *dr = container_of(res, struct devres, data); + + BUG_ON(!list_empty(&dr->node.entry)); + kfree(dr); + } +} +EXPORT_SYMBOL_GPL(devres_free); + +/** + * devres_add - Register device resource + * @dev: Device to add resource to + * @res: Resource to register + * + * Register devres @res to @dev. @res should have been allocated + * using devres_alloc(). On driver detach, the associated release + * function will be invoked and devres will be freed automatically. + */ +void devres_add(struct device *dev, void *res) +{ + struct devres *dr = container_of(res, struct devres, data); + unsigned long flags; + + spin_lock_irqsave(&dev->devres_lock, flags); + add_dr(dev, &dr->node); + spin_unlock_irqrestore(&dev->devres_lock, flags); +} +EXPORT_SYMBOL_GPL(devres_add); + +static struct devres *find_dr(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data) +{ + struct devres_node *node; + + list_for_each_entry_reverse(node, &dev->devres_head, entry) { + struct devres *dr = container_of(node, struct devres, node); + + if (node->release != release) + continue; + if (match && !match(dev, dr->data, match_data)) + continue; + return dr; + } + + return NULL; +} + +/** + * devres_find - Find device resource + * @dev: Device to lookup resource from + * @release: Look for resources associated with this release function + * @match: Match function (optional) + * @match_data: Data for the match function + * + * Find the latest devres of @dev which is associated with @release + * and for which @match returns 1. If @match is NULL, it's considered + * to match all. + * + * RETURNS: + * Pointer to found devres, NULL if not found. + */ +void * devres_find(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data) +{ + struct devres *dr; + unsigned long flags; + + spin_lock_irqsave(&dev->devres_lock, flags); + dr = find_dr(dev, release, match, match_data); + spin_unlock_irqrestore(&dev->devres_lock, flags); + + if (dr) + return dr->data; + return NULL; +} +EXPORT_SYMBOL_GPL(devres_find); + +/** + * devres_get - Find devres, if non-existent, add one atomically + * @dev: Device to lookup or add devres for + * @new_res: Pointer to new initialized devres to add if not found + * @match: Match function (optional) + * @match_data: Data for the match function + * + * Find the latest devres of @dev which has the same release function + * as @new_res and for which @match return 1. If found, @new_res is + * freed; otherwise, @new_res is added atomically. + * + * RETURNS: + * Pointer to found or added devres. + */ +void * devres_get(struct device *dev, void *new_res, + dr_match_t match, void *match_data) +{ + struct devres *new_dr = container_of(new_res, struct devres, data); + struct devres *dr; + unsigned long flags; + + spin_lock_irqsave(&dev->devres_lock, flags); + dr = find_dr(dev, new_dr->node.release, match, match_data); + if (!dr) { + add_dr(dev, &new_dr->node); + dr = new_dr; + new_dr = NULL; + } + spin_unlock_irqrestore(&dev->devres_lock, flags); + devres_free(new_dr); + + return dr->data; +} +EXPORT_SYMBOL_GPL(devres_get); + +/** + * devres_remove - Find a device resource and remove it + * @dev: Device to find resource from + * @release: Look for resources associated with this release function + * @match: Match function (optional) + * @match_data: Data for the match function + * + * Find the latest devres of @dev associated with @release and for + * which @match returns 1. If @match is NULL, it's considered to + * match all. If found, the resource is removed atomically and + * returned. + * + * RETURNS: + * Pointer to removed devres on success, NULL if not found. + */ +void * devres_remove(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data) +{ + struct devres *dr; + unsigned long flags; + + spin_lock_irqsave(&dev->devres_lock, flags); + dr = find_dr(dev, release, match, match_data); + if (dr) { + list_del_init(&dr->node.entry); + devres_log(dev, &dr->node, "REM"); + } + spin_unlock_irqrestore(&dev->devres_lock, flags); + + if (dr) + return dr->data; + return NULL; +} +EXPORT_SYMBOL_GPL(devres_remove); + +/** + * devres_destroy - Find a device resource and destroy it + * @dev: Device to find resource from + * @release: Look for resources associated with this release function + * @match: Match function (optional) + * @match_data: Data for the match function + * + * Find the latest devres of @dev associated with @release and for + * which @match returns 1. If @match is NULL, it's considered to + * match all. If found, the resource is removed atomically and freed. + * + * RETURNS: + * 0 if devres is found and freed, -ENOENT if not found. + */ +int devres_destroy(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data) +{ + void *res; + + res = devres_remove(dev, release, match, match_data); + if (unlikely(!res)) + return -ENOENT; + + devres_free(res); + return 0; +} +EXPORT_SYMBOL_GPL(devres_destroy); + +static int remove_nodes(struct device *dev, + struct list_head *first, struct list_head *end, + struct list_head *todo) +{ + int cnt = 0, nr_groups = 0; + struct list_head *cur; + + /* First pass - move normal devres entries to @todo and clear + * devres_group colors. + */ + cur = first; + while (cur != end) { + struct devres_node *node; + struct devres_group *grp; + + node = list_entry(cur, struct devres_node, entry); + cur = cur->next; + + grp = node_to_group(node); + if (grp) { + /* clear color of group markers in the first pass */ + grp->color = 0; + nr_groups++; + } else { + /* regular devres entry */ + if (&node->entry == first) + first = first->next; + list_move_tail(&node->entry, todo); + cnt++; + } + } + + if (!nr_groups) + return cnt; + + /* Second pass - Scan groups and color them. A group gets + * color value of two iff the group is wholly contained in + * [cur, end). That is, for a closed group, both opening and + * closing markers should be in the range, while just the + * opening marker is enough for an open group. + */ + cur = first; + while (cur != end) { + struct devres_node *node; + struct devres_group *grp; + + node = list_entry(cur, struct devres_node, entry); + cur = cur->next; + + grp = node_to_group(node); + BUG_ON(!grp || list_empty(&grp->node[0].entry)); + + grp->color++; + if (list_empty(&grp->node[1].entry)) + grp->color++; + + BUG_ON(grp->color <= 0 || grp->color > 2); + if (grp->color == 2) { + /* No need to update cur or end. The removed + * nodes are always before both. + */ + list_move_tail(&grp->node[0].entry, todo); + list_del_init(&grp->node[1].entry); + } + } + + return cnt; +} + +static int release_nodes(struct device *dev, struct list_head *first, + struct list_head *end, unsigned long flags) +{ + LIST_HEAD(todo); + int cnt; + struct devres *dr, *tmp; + + cnt = remove_nodes(dev, first, end, &todo); + + spin_unlock_irqrestore(&dev->devres_lock, flags); + + /* Release. Note that both devres and devres_group are + * handled as devres in the following loop. This is safe. + */ + list_for_each_entry_safe_reverse(dr, tmp, &todo, node.entry) { + devres_log(dev, &dr->node, "REL"); + dr->node.release(dev, dr->data); + kfree(dr); + } + + return cnt; +} + +/** + * devres_release_all - Release all resources + * @dev: Device to release resources for + * + * Release all resources associated with @dev. This function is + * called on driver detach. + */ +int devres_release_all(struct device *dev) +{ + unsigned long flags; + + spin_lock_irqsave(&dev->devres_lock, flags); + return release_nodes(dev, dev->devres_head.next, &dev->devres_head, + flags); +} + +/** + * devres_open_group - Open a new devres group + * @dev: Device to open devres group for + * @id: Separator ID + * @gfp: Allocation flags + * + * Open a new devres group for @dev with @id. For @id, using a + * pointer to an object which won't be used for another group is + * recommended. If @id is NULL, address-wise unique ID is created. + * + * RETURNS: + * ID of the new group, NULL on failure. + */ +void * devres_open_group(struct device *dev, void *id, gfp_t gfp) +{ + struct devres_group *grp; + unsigned long flags; + + grp = kmalloc(sizeof(*grp), gfp); + if (unlikely(!grp)) + return NULL; + + grp->node[0].release = &group_open_release; + grp->node[1].release = &group_close_release; + INIT_LIST_HEAD(&grp->node[0].entry); + INIT_LIST_HEAD(&grp->node[1].entry); + set_node_dbginfo(&grp->node[0], "grp<", 0); + set_node_dbginfo(&grp->node[1], "grp>", 0); + grp->id = grp; + if (id) + grp->id = id; + + spin_lock_irqsave(&dev->devres_lock, flags); + add_dr(dev, &grp->node[0]); + spin_unlock_irqrestore(&dev->devres_lock, flags); + return grp->id; +} +EXPORT_SYMBOL_GPL(devres_open_group); + +/* Find devres group with ID @id. If @id is NULL, look for the latest. */ +static struct devres_group * find_group(struct device *dev, void *id) +{ + struct devres_node *node; + + list_for_each_entry_reverse(node, &dev->devres_head, entry) { + struct devres_group *grp; + + if (node->release != &group_open_release) + continue; + + grp = container_of(node, struct devres_group, node[0]); + + if (id) { + if (grp->id == id) + return grp; + } else if (list_empty(&grp->node[1].entry)) + return grp; + } + + return NULL; +} + +/** + * devres_close_group - Close a devres group + * @dev: Device to close devres group for + * @id: ID of target group, can be NULL + * + * Close the group identified by @id. If @id is NULL, the latest open + * group is selected. + */ +void devres_close_group(struct device *dev, void *id) +{ + struct devres_group *grp; + unsigned long flags; + + spin_lock_irqsave(&dev->devres_lock, flags); + + grp = find_group(dev, id); + if (grp) + add_dr(dev, &grp->node[1]); + else + WARN_ON(1); + + spin_unlock_irqrestore(&dev->devres_lock, flags); +} +EXPORT_SYMBOL_GPL(devres_close_group); + +/** + * devres_remove_group - Remove a devres group + * @dev: Device to remove group for + * @id: ID of target group, can be NULL + * + * Remove the group identified by @id. If @id is NULL, the latest + * open group is selected. Note that removing a group doesn't affect + * any other resources. + */ +void devres_remove_group(struct device *dev, void *id) +{ + struct devres_group *grp; + unsigned long flags; + + spin_lock_irqsave(&dev->devres_lock, flags); + + grp = find_group(dev, id); + if (grp) { + list_del_init(&grp->node[0].entry); + list_del_init(&grp->node[1].entry); + devres_log(dev, &grp->node[0], "REM"); + } else + WARN_ON(1); + + spin_unlock_irqrestore(&dev->devres_lock, flags); + + kfree(grp); +} +EXPORT_SYMBOL_GPL(devres_remove_group); + +/** + * devres_release_group - Release resources in a devres group + * @dev: Device to release group for + * @id: ID of target group, can be NULL + * + * Release all resources in the group identified by @id. If @id is + * NULL, the latest open group is selected. The selected group and + * groups properly nested inside the selected group are removed. + * + * RETURNS: + * The number of released non-group resources. + */ +int devres_release_group(struct device *dev, void *id) +{ + struct devres_group *grp; + unsigned long flags; + int cnt = 0; + + spin_lock_irqsave(&dev->devres_lock, flags); + + grp = find_group(dev, id); + if (grp) { + struct list_head *first = &grp->node[0].entry; + struct list_head *end = &dev->devres_head; + + if (!list_empty(&grp->node[1].entry)) + end = grp->node[1].entry.next; + + cnt = release_nodes(dev, first, end, flags); + } else { + WARN_ON(1); + spin_unlock_irqrestore(&dev->devres_lock, flags); + } + + return cnt; +} +EXPORT_SYMBOL_GPL(devres_release_group); + +/* + * Managed kzalloc/kfree + */ +static void devm_kzalloc_release(struct device *dev, void *res) +{ + /* noop */ +} + +static int devm_kzalloc_match(struct device *dev, void *res, void *data) +{ + return res == data; +} + +/** + * devm_kzalloc - Managed kzalloc + * @dev: Device to allocate memory for + * @size: Allocation size + * @gfp: Allocation gfp flags + * + * Managed kzalloc. Memory allocated with this function is + * automatically freed on driver detach. Like all other devres + * resources, guaranteed alignment is unsigned long long. + * + * RETURNS: + * Pointer to allocated memory on success, NULL on failure. + */ +void * devm_kzalloc(struct device *dev, size_t size, gfp_t gfp) +{ + struct devres *dr; + + /* use raw alloc_dr for kmalloc caller tracing */ + dr = alloc_dr(devm_kzalloc_release, size, gfp); + if (unlikely(!dr)) + return NULL; + + set_node_dbginfo(&dr->node, "devm_kzalloc_release", size); + devres_add(dev, dr->data); + return dr->data; +} +EXPORT_SYMBOL_GPL(devm_kzalloc); + +/** + * devm_kfree - Managed kfree + * @dev: Device this memory belongs to + * @p: Memory to free + * + * Free memory allocated with dev_kzalloc(). + */ +void devm_kfree(struct device *dev, void *p) +{ + int rc; + + rc = devres_destroy(dev, devm_kzalloc_release, devm_kzalloc_match, p); + WARN_ON(rc); +} +EXPORT_SYMBOL_GPL(devm_kfree); diff --git a/drivers/base/dma-mapping.c b/drivers/base/dma-mapping.c new file mode 100644 index 000000000000..ca9186f70a69 --- /dev/null +++ b/drivers/base/dma-mapping.c @@ -0,0 +1,218 @@ +/* + * drivers/base/dma-mapping.c - arch-independent dma-mapping routines + * + * Copyright (c) 2006 SUSE Linux Products GmbH + * Copyright (c) 2006 Tejun Heo + * + * This file is released under the GPLv2. + */ + +#include + +/* + * Managed DMA API + */ +struct dma_devres { + size_t size; + void *vaddr; + dma_addr_t dma_handle; +}; + +static void dmam_coherent_release(struct device *dev, void *res) +{ + struct dma_devres *this = res; + + dma_free_coherent(dev, this->size, this->vaddr, this->dma_handle); +} + +static void dmam_noncoherent_release(struct device *dev, void *res) +{ + struct dma_devres *this = res; + + dma_free_noncoherent(dev, this->size, this->vaddr, this->dma_handle); +} + +static int dmam_match(struct device *dev, void *res, void *match_data) +{ + struct dma_devres *this = res, *match = match_data; + + if (this->vaddr == match->vaddr) { + WARN_ON(this->size != match->size || + this->dma_handle != match->dma_handle); + return 1; + } + return 0; +} + +/** + * dmam_alloc_coherent - Managed dma_alloc_coherent() + * @dev: Device to allocate coherent memory for + * @size: Size of allocation + * @dma_handle: Out argument for allocated DMA handle + * @gfp: Allocation flags + * + * Managed dma_alloc_coherent(). Memory allocated using this function + * will be automatically released on driver detach. + * + * RETURNS: + * Pointer to allocated memory on success, NULL on failure. + */ +void * dmam_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t gfp) +{ + struct dma_devres *dr; + void *vaddr; + + dr = devres_alloc(dmam_coherent_release, sizeof(*dr), gfp); + if (!dr) + return NULL; + + vaddr = dma_alloc_coherent(dev, size, dma_handle, gfp); + if (!vaddr) { + devres_free(dr); + return NULL; + } + + dr->vaddr = vaddr; + dr->dma_handle = *dma_handle; + dr->size = size; + + devres_add(dev, dr); + + return vaddr; +} +EXPORT_SYMBOL(dmam_alloc_coherent); + +/** + * dmam_free_coherent - Managed dma_free_coherent() + * @dev: Device to free coherent memory for + * @size: Size of allocation + * @vaddr: Virtual address of the memory to free + * @dma_handle: DMA handle of the memory to free + * + * Managed dma_free_coherent(). + */ +void dmam_free_coherent(struct device *dev, size_t size, void *vaddr, + dma_addr_t dma_handle) +{ + struct dma_devres match_data = { size, vaddr, dma_handle }; + + dma_free_coherent(dev, size, vaddr, dma_handle); + WARN_ON(devres_destroy(dev, dmam_coherent_release, dmam_match, + &match_data)); +} +EXPORT_SYMBOL(dmam_free_coherent); + +/** + * dmam_alloc_non_coherent - Managed dma_alloc_non_coherent() + * @dev: Device to allocate non_coherent memory for + * @size: Size of allocation + * @dma_handle: Out argument for allocated DMA handle + * @gfp: Allocation flags + * + * Managed dma_alloc_non_coherent(). Memory allocated using this + * function will be automatically released on driver detach. + * + * RETURNS: + * Pointer to allocated memory on success, NULL on failure. + */ +void *dmam_alloc_noncoherent(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t gfp) +{ + struct dma_devres *dr; + void *vaddr; + + dr = devres_alloc(dmam_noncoherent_release, sizeof(*dr), gfp); + if (!dr) + return NULL; + + vaddr = dma_alloc_noncoherent(dev, size, dma_handle, gfp); + if (!vaddr) { + devres_free(dr); + return NULL; + } + + dr->vaddr = vaddr; + dr->dma_handle = *dma_handle; + dr->size = size; + + devres_add(dev, dr); + + return vaddr; +} +EXPORT_SYMBOL(dmam_alloc_noncoherent); + +/** + * dmam_free_coherent - Managed dma_free_noncoherent() + * @dev: Device to free noncoherent memory for + * @size: Size of allocation + * @vaddr: Virtual address of the memory to free + * @dma_handle: DMA handle of the memory to free + * + * Managed dma_free_noncoherent(). + */ +void dmam_free_noncoherent(struct device *dev, size_t size, void *vaddr, + dma_addr_t dma_handle) +{ + struct dma_devres match_data = { size, vaddr, dma_handle }; + + dma_free_noncoherent(dev, size, vaddr, dma_handle); + WARN_ON(!devres_destroy(dev, dmam_noncoherent_release, dmam_match, + &match_data)); +} +EXPORT_SYMBOL(dmam_free_noncoherent); + +#ifdef ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY + +static void dmam_coherent_decl_release(struct device *dev, void *res) +{ + dma_release_declared_memory(dev); +} + +/** + * dmam_declare_coherent_memory - Managed dma_declare_coherent_memory() + * @dev: Device to declare coherent memory for + * @bus_addr: Bus address of coherent memory to be declared + * @device_addr: Device address of coherent memory to be declared + * @size: Size of coherent memory to be declared + * @flags: Flags + * + * Managed dma_declare_coherent_memory(). + * + * RETURNS: + * 0 on success, -errno on failure. + */ +int dmam_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr, + dma_addr_t device_addr, size_t size, int flags) +{ + void *res; + int rc; + + res = devres_alloc(dmam_coherent_decl_release, 0, GFP_KERNEL); + if (!res) + return -ENOMEM; + + rc = dma_declare_coherent_memory(dev, bus_addr, device_addr, size, + flags); + if (rc == 0) + devres_add(dev, res); + else + devres_free(res); + + return rc; +} +EXPORT_SYMBOL(dmam_declare_coherent_memory); + +/** + * dmam_release_declared_memory - Managed dma_release_declared_memory(). + * @dev: Device to release declared coherent memory for + * + * Managed dmam_release_declared_memory(). + */ +void dmam_release_declared_memory(struct device *dev) +{ + WARN_ON(devres_destroy(dev, dmam_coherent_decl_release, NULL, NULL)); +} +EXPORT_SYMBOL(dmam_release_declared_memory); + +#endif diff --git a/drivers/base/dmapool.c b/drivers/base/dmapool.c index f95d50277274..cd467c9f33b3 100644 --- a/drivers/base/dmapool.c +++ b/drivers/base/dmapool.c @@ -415,8 +415,67 @@ dma_pool_free (struct dma_pool *pool, void *vaddr, dma_addr_t dma) spin_unlock_irqrestore (&pool->lock, flags); } +/* + * Managed DMA pool + */ +static void dmam_pool_release(struct device *dev, void *res) +{ + struct dma_pool *pool = *(struct dma_pool **)res; + + dma_pool_destroy(pool); +} + +static int dmam_pool_match(struct device *dev, void *res, void *match_data) +{ + return *(struct dma_pool **)res == match_data; +} + +/** + * dmam_pool_create - Managed dma_pool_create() + * @name: name of pool, for diagnostics + * @dev: device that will be doing the DMA + * @size: size of the blocks in this pool. + * @align: alignment requirement for blocks; must be a power of two + * @allocation: returned blocks won't cross this boundary (or zero) + * + * Managed dma_pool_create(). DMA pool created with this function is + * automatically destroyed on driver detach. + */ +struct dma_pool *dmam_pool_create(const char *name, struct device *dev, + size_t size, size_t align, size_t allocation) +{ + struct dma_pool **ptr, *pool; + + ptr = devres_alloc(dmam_pool_release, sizeof(*ptr), GFP_KERNEL); + if (!ptr) + return NULL; + + pool = *ptr = dma_pool_create(name, dev, size, align, allocation); + if (pool) + devres_add(dev, ptr); + else + devres_free(ptr); + + return pool; +} + +/** + * dmam_pool_destroy - Managed dma_pool_destroy() + * @pool: dma pool that will be destroyed + * + * Managed dma_pool_destroy(). + */ +void dmam_pool_destroy(struct dma_pool *pool) +{ + struct device *dev = pool->dev; + + dma_pool_destroy(pool); + WARN_ON(devres_destroy(dev, dmam_pool_release, dmam_pool_match, pool)); +} EXPORT_SYMBOL (dma_pool_create); EXPORT_SYMBOL (dma_pool_destroy); EXPORT_SYMBOL (dma_pool_alloc); EXPORT_SYMBOL (dma_pool_free); +EXPORT_SYMBOL (dmam_pool_create); +EXPORT_SYMBOL (dmam_pool_destroy); diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 84c757ba0664..8b44cff2c176 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -744,6 +744,104 @@ int pci_enable_device(struct pci_dev *dev) return pci_enable_device_bars(dev, (1 << PCI_NUM_RESOURCES) - 1); } +/* + * Managed PCI resources. This manages device on/off, intx/msi/msix + * on/off and BAR regions. pci_dev itself records msi/msix status, so + * there's no need to track it separately. pci_devres is initialized + * when a device is enabled using managed PCI device enable interface. + */ +struct pci_devres { + unsigned int disable:1; + unsigned int orig_intx:1; + unsigned int restore_intx:1; + u32 region_mask; +}; + +static void pcim_release(struct device *gendev, void *res) +{ + struct pci_dev *dev = container_of(gendev, struct pci_dev, dev); + struct pci_devres *this = res; + int i; + + if (dev->msi_enabled) + pci_disable_msi(dev); + if (dev->msix_enabled) + pci_disable_msix(dev); + + for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) + if (this->region_mask & (1 << i)) + pci_release_region(dev, i); + + if (this->restore_intx) + pci_intx(dev, this->orig_intx); + + if (this->disable) + pci_disable_device(dev); +} + +static struct pci_devres * get_pci_dr(struct pci_dev *pdev) +{ + struct pci_devres *dr, *new_dr; + + dr = devres_find(&pdev->dev, pcim_release, NULL, NULL); + if (dr) + return dr; + + new_dr = devres_alloc(pcim_release, sizeof(*new_dr), GFP_KERNEL); + if (!new_dr) + return NULL; + return devres_get(&pdev->dev, new_dr, NULL, NULL); +} + +static struct pci_devres * find_pci_dr(struct pci_dev *pdev) +{ + if (pci_is_managed(pdev)) + return devres_find(&pdev->dev, pcim_release, NULL, NULL); + return NULL; +} + +/** + * pcim_enable_device - Managed pci_enable_device() + * @pdev: PCI device to be initialized + * + * Managed pci_enable_device(). + */ +int pcim_enable_device(struct pci_dev *pdev) +{ + struct pci_devres *dr; + int rc; + + dr = get_pci_dr(pdev); + if (unlikely(!dr)) + return -ENOMEM; + WARN_ON(!!dr->disable); + + rc = pci_enable_device(pdev); + if (!rc) { + pdev->is_managed = 1; + dr->disable = 1; + } + return rc; +} + +/** + * pcim_pin_device - Pin managed PCI device + * @pdev: PCI device to pin + * + * Pin managed PCI device @pdev. Pinned device won't be disabled on + * driver detach. @pdev must have been enabled with + * pcim_enable_device(). + */ +void pcim_pin_device(struct pci_dev *pdev) +{ + struct pci_devres *dr; + + dr = find_pci_dr(pdev); + WARN_ON(!dr || !dr->disable); + if (dr) + dr->disable = 0; +} + /** * pcibios_disable_device - disable arch specific PCI resources for device dev * @dev: the PCI device to disable @@ -767,8 +865,13 @@ void __attribute__ ((weak)) pcibios_disable_device (struct pci_dev *dev) {} void pci_disable_device(struct pci_dev *dev) { + struct pci_devres *dr; u16 pci_command; + dr = find_pci_dr(dev); + if (dr) + dr->disable = 0; + if (atomic_sub_return(1, &dev->enable_cnt) != 0) return; @@ -867,6 +970,8 @@ pci_get_interrupt_pin(struct pci_dev *dev, struct pci_dev **bridge) */ void pci_release_region(struct pci_dev *pdev, int bar) { + struct pci_devres *dr; + if (pci_resource_len(pdev, bar) == 0) return; if (pci_resource_flags(pdev, bar) & IORESOURCE_IO) @@ -875,6 +980,10 @@ void pci_release_region(struct pci_dev *pdev, int bar) else if (pci_resource_flags(pdev, bar) & IORESOURCE_MEM) release_mem_region(pci_resource_start(pdev, bar), pci_resource_len(pdev, bar)); + + dr = find_pci_dr(pdev); + if (dr) + dr->region_mask &= ~(1 << bar); } /** @@ -893,6 +1002,8 @@ void pci_release_region(struct pci_dev *pdev, int bar) */ int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name) { + struct pci_devres *dr; + if (pci_resource_len(pdev, bar) == 0) return 0; @@ -906,7 +1017,11 @@ int pci_request_region(struct pci_dev *pdev, int bar, const char *res_name) pci_resource_len(pdev, bar), res_name)) goto err_out; } - + + dr = find_pci_dr(pdev); + if (dr) + dr->region_mask |= 1 << bar; + return 0; err_out: @@ -1144,7 +1259,15 @@ pci_intx(struct pci_dev *pdev, int enable) } if (new != pci_command) { + struct pci_devres *dr; + pci_write_config_word(pdev, PCI_COMMAND, new); + + dr = find_pci_dr(pdev); + if (dr && !dr->restore_intx) { + dr->restore_intx = 1; + dr->orig_intx = !enable; + } } } @@ -1226,6 +1349,8 @@ device_initcall(pci_init); EXPORT_SYMBOL_GPL(pci_restore_bars); EXPORT_SYMBOL(pci_enable_device_bars); EXPORT_SYMBOL(pci_enable_device); +EXPORT_SYMBOL(pcim_enable_device); +EXPORT_SYMBOL(pcim_pin_device); EXPORT_SYMBOL(pci_disable_device); EXPORT_SYMBOL(pci_find_capability); EXPORT_SYMBOL(pci_bus_find_capability); diff --git a/include/linux/device.h b/include/linux/device.h index 5ca1cdba563a..26e4692f2d1a 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -354,6 +354,41 @@ extern int __must_check device_create_bin_file(struct device *dev, struct bin_attribute *attr); extern void device_remove_bin_file(struct device *dev, struct bin_attribute *attr); + +/* device resource management */ +typedef void (*dr_release_t)(struct device *dev, void *res); +typedef int (*dr_match_t)(struct device *dev, void *res, void *match_data); + +#ifdef CONFIG_DEBUG_DEVRES +extern void * __devres_alloc(dr_release_t release, size_t size, gfp_t gfp, + const char *name); +#define devres_alloc(release, size, gfp) \ + __devres_alloc(release, size, gfp, #release) +#else +extern void * devres_alloc(dr_release_t release, size_t size, gfp_t gfp); +#endif +extern void devres_free(void *res); +extern void devres_add(struct device *dev, void *res); +extern void * devres_find(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data); +extern void * devres_get(struct device *dev, void *new_res, + dr_match_t match, void *match_data); +extern void * devres_remove(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data); +extern int devres_destroy(struct device *dev, dr_release_t release, + dr_match_t match, void *match_data); + +/* devres group */ +extern void * __must_check devres_open_group(struct device *dev, void *id, + gfp_t gfp); +extern void devres_close_group(struct device *dev, void *id); +extern void devres_remove_group(struct device *dev, void *id); +extern int devres_release_group(struct device *dev, void *id); + +/* managed kzalloc/kfree for device drivers, no kmalloc, always use kzalloc */ +extern void *devm_kzalloc(struct device *dev, size_t size, gfp_t gfp); +extern void devm_kfree(struct device *dev, void *p); + struct device { struct klist klist_children; struct klist_node knode_parent; /* node in sibling list */ @@ -397,6 +432,9 @@ struct device { /* arch specific additions */ struct dev_archdata archdata; + spinlock_t devres_lock; + struct list_head devres_head; + /* class_device migration path */ struct list_head node; struct class *class; diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index ff203c465fed..9a663c6db16a 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -66,6 +66,33 @@ dma_mark_declared_memory_occupied(struct device *dev, } #endif -#endif +/* + * Managed DMA API + */ +extern void *dmam_alloc_coherent(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t gfp); +extern void dmam_free_coherent(struct device *dev, size_t size, void *vaddr, + dma_addr_t dma_handle); +extern void *dmam_alloc_noncoherent(struct device *dev, size_t size, + dma_addr_t *dma_handle, gfp_t gfp); +extern void dmam_free_noncoherent(struct device *dev, size_t size, void *vaddr, + dma_addr_t dma_handle); +#ifdef ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY +extern int dmam_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr, + dma_addr_t device_addr, size_t size, + int flags); +extern void dmam_release_declared_memory(struct device *dev); +#else /* ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY */ +static inline int dmam_declare_coherent_memory(struct device *dev, + dma_addr_t bus_addr, dma_addr_t device_addr, + size_t size, gfp_t gfp) +{ + return 0; +} +static inline void dmam_release_declared_memory(struct device *dev) +{ +} +#endif /* ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY */ +#endif diff --git a/include/linux/dmapool.h b/include/linux/dmapool.h index 76f12f46db7f..022e34fcbd1b 100644 --- a/include/linux/dmapool.h +++ b/include/linux/dmapool.h @@ -24,5 +24,12 @@ void *dma_pool_alloc(struct dma_pool *pool, gfp_t mem_flags, void dma_pool_free(struct dma_pool *pool, void *vaddr, dma_addr_t addr); +/* + * Managed DMA pool + */ +struct dma_pool *dmam_pool_create(const char *name, struct device *dev, + size_t size, size_t align, size_t allocation); +void dmam_pool_destroy(struct dma_pool *pool); + #endif diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index e36e86c869fb..5a8ba0b8ccba 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -83,6 +84,11 @@ extern int request_irq(unsigned int, irq_handler_t handler, unsigned long, const char *, void *); extern void free_irq(unsigned int, void *); +extern int devm_request_irq(struct device *dev, unsigned int irq, + irq_handler_t handler, unsigned long irqflags, + const char *devname, void *dev_id); +extern void devm_free_irq(struct device *dev, unsigned int irq, void *dev_id); + /* * On lockdep we dont want to enable hardirqs in hardirq * context. Use local_irq_enable_in_hardirq() to annotate diff --git a/include/linux/io.h b/include/linux/io.h index 81877ea39309..f5edf9c5de0a 100644 --- a/include/linux/io.h +++ b/include/linux/io.h @@ -28,6 +28,23 @@ void __iowrite64_copy(void __iomem *to, const void *from, size_t count); int ioremap_page_range(unsigned long addr, unsigned long end, unsigned long phys_addr, pgprot_t prot); +/* + * Managed iomap interface + */ +void __iomem * devm_ioport_map(struct device *dev, unsigned long port, + unsigned int nr); +void devm_ioport_unmap(struct device *dev, void __iomem *addr); + +void __iomem * devm_ioremap(struct device *dev, unsigned long offset, + unsigned long size); +void __iomem * devm_ioremap_nocache(struct device *dev, unsigned long offset, + unsigned long size); +void devm_iounmap(struct device *dev, void __iomem *addr); + +void __iomem * pcim_iomap(struct pci_dev *pdev, int bar, unsigned long maxlen); +void pcim_iounmap(struct pci_dev *pdev, void __iomem *addr); +void __iomem * const * pcim_iomap_table(struct pci_dev *pdev); + /** * check_signature - find BIOS signatures * @io_addr: mmio address to check diff --git a/include/linux/ioport.h b/include/linux/ioport.h index 15228d79c5bc..6859a3b14088 100644 --- a/include/linux/ioport.h +++ b/include/linux/ioport.h @@ -137,4 +137,24 @@ static inline int __deprecated check_region(resource_size_t s, { return __check_region(&ioport_resource, s, n); } + +/* Wrappers for managed devices */ +struct device; +#define devm_request_region(dev,start,n,name) \ + __devm_request_region(dev, &ioport_resource, (start), (n), (name)) +#define devm_request_mem_region(dev,start,n,name) \ + __devm_request_region(dev, &iomem_resource, (start), (n), (name)) + +extern struct resource * __devm_request_region(struct device *dev, + struct resource *parent, resource_size_t start, + resource_size_t n, const char *name); + +#define devm_release_region(start,n) \ + __devm_release_region(dev, &ioport_resource, (start), (n)) +#define devm_release_mem_region(start,n) \ + __devm_release_region(dev, &iomem_resource, (start), (n)) + +extern void __devm_release_region(struct device *dev, struct resource *parent, + resource_size_t start, resource_size_t n); + #endif /* _LINUX_IOPORT_H */ diff --git a/include/linux/pci.h b/include/linux/pci.h index 805412cc6875..9e3042e7e1cc 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -167,6 +167,7 @@ struct pci_dev { unsigned int broken_parity_status:1; /* Device generates false positive parity */ unsigned int msi_enabled:1; unsigned int msix_enabled:1; + unsigned int is_managed:1; atomic_t enable_cnt; /* pci_enable_device has been called */ u32 saved_config_space[16]; /* config space saved at suspend time */ @@ -528,6 +529,14 @@ static inline int pci_write_config_dword(struct pci_dev *dev, int where, u32 val int __must_check pci_enable_device(struct pci_dev *dev); int __must_check pci_enable_device_bars(struct pci_dev *dev, int mask); +int __must_check pcim_enable_device(struct pci_dev *pdev); +void pcim_pin_device(struct pci_dev *pdev); + +static inline int pci_is_managed(struct pci_dev *pdev) +{ + return pdev->is_managed; +} + void pci_disable_device(struct pci_dev *dev); void pci_set_master(struct pci_dev *dev); #define HAVE_PCI_SET_MWI diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 8b961adc3bd2..c4b7ed1cebf7 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -482,3 +482,89 @@ int request_irq(unsigned int irq, irq_handler_t handler, return retval; } EXPORT_SYMBOL(request_irq); + +/* + * Device resource management aware IRQ request/free implementation. + */ +struct irq_devres { + unsigned int irq; + void *dev_id; +}; + +static void devm_irq_release(struct device *dev, void *res) +{ + struct irq_devres *this = res; + + free_irq(this->irq, this->dev_id); +} + +static int devm_irq_match(struct device *dev, void *res, void *data) +{ + struct irq_devres *this = res, *match = data; + + return this->irq == match->irq && this->dev_id == match->dev_id; +} + +/** + * devm_request_irq - allocate an interrupt line for a managed device + * @dev: device to request interrupt for + * @irq: Interrupt line to allocate + * @handler: Function to be called when the IRQ occurs + * @irqflags: Interrupt type flags + * @devname: An ascii name for the claiming device + * @dev_id: A cookie passed back to the handler function + * + * Except for the extra @dev argument, this function takes the + * same arguments and performs the same function as + * request_irq(). IRQs requested with this function will be + * automatically freed on driver detach. + * + * If an IRQ allocated with this function needs to be freed + * separately, dev_free_irq() must be used. + */ +int devm_request_irq(struct device *dev, unsigned int irq, + irq_handler_t handler, unsigned long irqflags, + const char *devname, void *dev_id) +{ + struct irq_devres *dr; + int rc; + + dr = devres_alloc(devm_irq_release, sizeof(struct irq_devres), + GFP_KERNEL); + if (!dr) + return -ENOMEM; + + rc = request_irq(irq, handler, irqflags, devname, dev_id); + if (rc) { + kfree(dr); + return rc; + } + + dr->irq = irq; + dr->dev_id = dev_id; + devres_add(dev, dr); + + return 0; +} +EXPORT_SYMBOL(devm_request_irq); + +/** + * devm_free_irq - free an interrupt + * @dev: device to free interrupt for + * @irq: Interrupt line to free + * @dev_id: Device identity to free + * + * Except for the extra @dev argument, this function takes the + * same arguments and performs the same function as free_irq(). + * This function instead of free_irq() should be used to manually + * free IRQs allocated with dev_request_irq(). + */ +void devm_free_irq(struct device *dev, unsigned int irq, void *dev_id) +{ + struct irq_devres match_data = { irq, dev_id }; + + free_irq(irq, dev_id); + WARN_ON(devres_destroy(dev, devm_irq_release, devm_irq_match, + &match_data)); +} +EXPORT_SYMBOL(devm_free_irq); diff --git a/kernel/resource.c b/kernel/resource.c index 7b9a497419d9..2a3f88636580 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -17,6 +17,7 @@ #include #include #include +#include #include @@ -617,6 +618,67 @@ void __release_region(struct resource *parent, resource_size_t start, } EXPORT_SYMBOL(__release_region); +/* + * Managed region resource + */ +struct region_devres { + struct resource *parent; + resource_size_t start; + resource_size_t n; +}; + +static void devm_region_release(struct device *dev, void *res) +{ + struct region_devres *this = res; + + __release_region(this->parent, this->start, this->n); +} + +static int devm_region_match(struct device *dev, void *res, void *match_data) +{ + struct region_devres *this = res, *match = match_data; + + return this->parent == match->parent && + this->start == match->start && this->n == match->n; +} + +struct resource * __devm_request_region(struct device *dev, + struct resource *parent, resource_size_t start, + resource_size_t n, const char *name) +{ + struct region_devres *dr = NULL; + struct resource *res; + + dr = devres_alloc(devm_region_release, sizeof(struct region_devres), + GFP_KERNEL); + if (!dr) + return NULL; + + dr->parent = parent; + dr->start = start; + dr->n = n; + + res = __request_region(parent, start, n, name); + if (res) + devres_add(dev, dr); + else + devres_free(dr); + + return res; +} +EXPORT_SYMBOL(__devm_request_region); + +void __devm_release_region(struct device *dev, struct resource *parent, + resource_size_t start, resource_size_t n) +{ + struct region_devres match_data = { parent, start, n }; + + __release_region(parent, start, n); + WARN_ON(devres_destroy(dev, devm_region_release, devm_region_match, + &match_data)); +} +EXPORT_SYMBOL(__devm_release_region); + /* * Called from init/main.c to reserve IO ports. */ diff --git a/lib/Makefile b/lib/Makefile index 77b4bad7d441..29b2e9912bbb 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -5,7 +5,7 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \ bust_spinlocks.o rbtree.o radix-tree.o dump_stack.o \ idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \ - sha1.o irq_regs.o reciprocal_div.o + sha1.o irq_regs.o reciprocal_div.o iomap.o lib-$(CONFIG_MMU) += ioremap.o lib-$(CONFIG_SMP) += cpumask.o @@ -41,7 +41,6 @@ obj-$(CONFIG_CRC_CCITT) += crc-ccitt.o obj-$(CONFIG_CRC16) += crc16.o obj-$(CONFIG_CRC32) += crc32.o obj-$(CONFIG_LIBCRC32C) += libcrc32c.o -obj-$(CONFIG_GENERIC_IOMAP) += iomap.o obj-$(CONFIG_GENERIC_ALLOCATOR) += genalloc.o obj-$(CONFIG_ZLIB_INFLATE) += zlib_inflate/ diff --git a/lib/iomap.c b/lib/iomap.c index d6ccdd85df53..3214028141da 100644 --- a/lib/iomap.c +++ b/lib/iomap.c @@ -4,8 +4,10 @@ * (C) Copyright 2004 Linus Torvalds */ #include +#include + +#ifdef CONFIG_GENERIC_IOMAP #include -#include /* * Read/write from/to an (offsettable) iomem cookie. It might be a PIO @@ -254,3 +256,245 @@ void pci_iounmap(struct pci_dev *dev, void __iomem * addr) } EXPORT_SYMBOL(pci_iomap); EXPORT_SYMBOL(pci_iounmap); + +#endif /* CONFIG_GENERIC_IOMAP */ + +/* + * Generic iomap devres + */ +static void devm_ioport_map_release(struct device *dev, void *res) +{ + ioport_unmap(*(void __iomem **)res); +} + +static int devm_ioport_map_match(struct device *dev, void *res, + void *match_data) +{ + return *(void **)res == match_data; +} + +/** + * devm_ioport_map - Managed ioport_map() + * @dev: Generic device to map ioport for + * @port: Port to map + * @nr: Number of ports to map + * + * Managed ioport_map(). Map is automatically unmapped on driver + * detach. + */ +void __iomem * devm_ioport_map(struct device *dev, unsigned long port, + unsigned int nr) +{ + void __iomem **ptr, *addr; + + ptr = devres_alloc(devm_ioport_map_release, sizeof(*ptr), GFP_KERNEL); + if (!ptr) + return NULL; + + addr = ioport_map(port, nr); + if (addr) { + *ptr = addr; + devres_add(dev, ptr); + } else + devres_free(ptr); + + return addr; +} +EXPORT_SYMBOL(devm_ioport_map); + +/** + * devm_ioport_unmap - Managed ioport_unmap() + * @dev: Generic device to unmap for + * @addr: Address to unmap + * + * Managed ioport_unmap(). @addr must have been mapped using + * devm_ioport_map(). + */ +void devm_ioport_unmap(struct device *dev, void __iomem *addr) +{ + ioport_unmap(addr); + WARN_ON(devres_destroy(dev, devm_ioport_map_release, + devm_ioport_map_match, (void *)addr)); +} +EXPORT_SYMBOL(devm_ioport_unmap); + +static void devm_ioremap_release(struct device *dev, void *res) +{ + iounmap(*(void __iomem **)res); +} + +static int devm_ioremap_match(struct device *dev, void *res, void *match_data) +{ + return *(void **)res == match_data; +} + +/** + * devm_ioremap - Managed ioremap() + * @dev: Generic device to remap IO address for + * @offset: BUS offset to map + * @size: Size of map + * + * Managed ioremap(). Map is automatically unmapped on driver detach. + */ +void __iomem *devm_ioremap(struct device *dev, unsigned long offset, + unsigned long size) +{ + void __iomem **ptr, *addr; + + ptr = devres_alloc(devm_ioremap_release, sizeof(*ptr), GFP_KERNEL); + if (!ptr) + return NULL; + + addr = ioremap(offset, size); + if (addr) { + *ptr = addr; + devres_add(dev, ptr); + } else + devres_free(ptr); + + return addr; +} +EXPORT_SYMBOL(devm_ioremap); + +/** + * devm_ioremap_nocache - Managed ioremap_nocache() + * @dev: Generic device to remap IO address for + * @offset: BUS offset to map + * @size: Size of map + * + * Managed ioremap_nocache(). Map is automatically unmapped on driver + * detach. + */ +void __iomem *devm_ioremap_nocache(struct device *dev, unsigned long offset, + unsigned long size) +{ + void __iomem **ptr, *addr; + + ptr = devres_alloc(devm_ioremap_release, sizeof(*ptr), GFP_KERNEL); + if (!ptr) + return NULL; + + addr = ioremap_nocache(offset, size); + if (addr) { + *ptr = addr; + devres_add(dev, ptr); + } else + devres_free(ptr); + + return addr; +} +EXPORT_SYMBOL(devm_ioremap_nocache); + +/** + * devm_iounmap - Managed iounmap() + * @dev: Generic device to unmap for + * @addr: Address to unmap + * + * Managed iounmap(). @addr must have been mapped using devm_ioremap*(). + */ +void devm_iounmap(struct device *dev, void __iomem *addr) +{ + iounmap(addr); + WARN_ON(devres_destroy(dev, devm_ioremap_release, devm_ioremap_match, + (void *)addr)); +} +EXPORT_SYMBOL(devm_iounmap); + +/* + * PCI iomap devres + */ +#define PCIM_IOMAP_MAX PCI_ROM_RESOURCE + +struct pcim_iomap_devres { + void __iomem *table[PCIM_IOMAP_MAX]; +}; + +static void pcim_iomap_release(struct device *gendev, void *res) +{ + struct pci_dev *dev = container_of(gendev, struct pci_dev, dev); + struct pcim_iomap_devres *this = res; + int i; + + for (i = 0; i < PCIM_IOMAP_MAX; i++) + if (this->table[i]) + pci_iounmap(dev, this->table[i]); +} + +/** + * pcim_iomap_table - access iomap allocation table + * @pdev: PCI device to access iomap table for + * + * Access iomap allocation table for @dev. If iomap table doesn't + * exist and @pdev is managed, it will be allocated. All iomaps + * recorded in the iomap table are automatically unmapped on driver + * detach. + * + * This function might sleep when the table is first allocated but can + * be safely called without context and guaranteed to succed once + * allocated. + */ +void __iomem * const * pcim_iomap_table(struct pci_dev *pdev) +{ + struct pcim_iomap_devres *dr, *new_dr; + + dr = devres_find(&pdev->dev, pcim_iomap_release, NULL, NULL); + if (dr) + return dr->table; + + new_dr = devres_alloc(pcim_iomap_release, sizeof(*new_dr), GFP_KERNEL); + if (!new_dr) + return NULL; + dr = devres_get(&pdev->dev, new_dr, NULL, NULL); + return dr->table; +} +EXPORT_SYMBOL(pcim_iomap_table); + +/** + * pcim_iomap - Managed pcim_iomap() + * @pdev: PCI device to iomap for + * @bar: BAR to iomap + * @maxlen: Maximum length of iomap + * + * Managed pci_iomap(). Map is automatically unmapped on driver + * detach. + */ +void __iomem * pcim_iomap(struct pci_dev *pdev, int bar, unsigned long maxlen) +{ + void __iomem **tbl; + + BUG_ON(bar >= PCIM_IOMAP_MAX); + + tbl = (void __iomem **)pcim_iomap_table(pdev); + if (!tbl || tbl[bar]) /* duplicate mappings not allowed */ + return NULL; + + tbl[bar] = pci_iomap(pdev, bar, maxlen); + return tbl[bar]; +} +EXPORT_SYMBOL(pcim_iomap); + +/** + * pcim_iounmap - Managed pci_iounmap() + * @pdev: PCI device to iounmap for + * @addr: Address to unmap + * + * Managed pci_iounmap(). @addr must have been mapped using pcim_iomap(). + */ +void pcim_iounmap(struct pci_dev *pdev, void __iomem *addr) +{ + void __iomem **tbl; + int i; + + pci_iounmap(pdev, addr); + + tbl = (void __iomem **)pcim_iomap_table(pdev); + BUG_ON(!tbl); + + for (i = 0; i < PCIM_IOMAP_MAX; i++) + if (tbl[i] == addr) { + tbl[i] = NULL; + return; + } + WARN_ON(1); +} +EXPORT_SYMBOL(pcim_iounmap); -- cgit v1.2.3 From d24bbbf251e70bf984cbaa9b1fcadc5f56fc3ae9 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 20 Jan 2007 16:00:28 +0900 Subject: devres: implement pcim_iomap_regions() Implement pcim_iomap_regions(). This function takes mask of BARs to request and iomap. No BAR should have length of zero. BARs are iomapped using pcim_iomap_table(). Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- include/linux/io.h | 2 ++ lib/iomap.c | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 55 insertions(+) (limited to 'lib') diff --git a/include/linux/io.h b/include/linux/io.h index f5edf9c5de0a..45a9c944cb0a 100644 --- a/include/linux/io.h +++ b/include/linux/io.h @@ -45,6 +45,8 @@ void __iomem * pcim_iomap(struct pci_dev *pdev, int bar, unsigned long maxlen); void pcim_iounmap(struct pci_dev *pdev, void __iomem *addr); void __iomem * const * pcim_iomap_table(struct pci_dev *pdev); +int pcim_iomap_regions(struct pci_dev *pdev, u16 mask, const char *name); + /** * check_signature - find BIOS signatures * @io_addr: mmio address to check diff --git a/lib/iomap.c b/lib/iomap.c index 3214028141da..4990c736bc4b 100644 --- a/lib/iomap.c +++ b/lib/iomap.c @@ -498,3 +498,56 @@ void pcim_iounmap(struct pci_dev *pdev, void __iomem *addr) WARN_ON(1); } EXPORT_SYMBOL(pcim_iounmap); + +/** + * pcim_iomap_regions - Request and iomap PCI BARs + * @pdev: PCI device to map IO resources for + * @mask: Mask of BARs to request and iomap + * @name: Name used when requesting regions + * + * Request and iomap regions specified by @mask. + */ +int pcim_iomap_regions(struct pci_dev *pdev, u16 mask, const char *name) +{ + void __iomem * const *iomap; + int i, rc; + + iomap = pcim_iomap_table(pdev); + if (!iomap) + return -ENOMEM; + + for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) { + unsigned long len; + + if (!(mask & (1 << i))) + continue; + + rc = -EINVAL; + len = pci_resource_len(pdev, i); + if (!len) + goto err_inval; + + rc = pci_request_region(pdev, i, name); + if (rc) + goto err_region; + + rc = -ENOMEM; + if (!pcim_iomap(pdev, i, 0)) + goto err_iomap; + } + + return 0; + + err_iomap: + pcim_iounmap(pdev, iomap[i]); + err_region: + pci_release_region(pdev, i); + err_inval: + while (--i >= 0) { + pcim_iounmap(pdev, iomap[i]); + pci_release_region(pdev, i); + } + + return rc; +} +EXPORT_SYMBOL(pcim_iomap_regions); -- cgit v1.2.3 From ca2997885219486cf91a369233c909fbd555bdf7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 31 Jan 2007 22:48:06 +0900 Subject: iomap: iomap should be in obj-y not in lib-y devres change moved iomap.o from obj-$(CONFIG_GENERIC_IOMAP) to lib-y making it not linked if no in-kernel driver uses it. Fix it. Signed-off-by: Tejun Heo Signed-off-by: Jeff Garzik --- lib/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'lib') diff --git a/lib/Makefile b/lib/Makefile index 29b2e9912bbb..3b605da448f4 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -5,14 +5,14 @@ lib-y := ctype.o string.o vsprintf.o cmdline.o \ bust_spinlocks.o rbtree.o radix-tree.o dump_stack.o \ idr.o div64.o int_sqrt.o bitmap.o extable.o prio_tree.o \ - sha1.o irq_regs.o reciprocal_div.o iomap.o + sha1.o irq_regs.o reciprocal_div.o lib-$(CONFIG_MMU) += ioremap.o lib-$(CONFIG_SMP) += cpumask.o lib-y += kobject.o kref.o kobject_uevent.o klist.o -obj-y += sort.o parser.o halfmd4.o debug_locks.o random32.o +obj-y += sort.o parser.o halfmd4.o debug_locks.o random32.o iomap.o ifeq ($(CONFIG_DEBUG_KOBJECT),y) CFLAGS_kobject.o += -DDEBUG -- cgit v1.2.3