diff options
Diffstat (limited to 'drivers/nvdimm')
| -rw-r--r-- | drivers/nvdimm/Kconfig | 19 | ||||
| -rw-r--r-- | drivers/nvdimm/btt.c | 16 | ||||
| -rw-r--r-- | drivers/nvdimm/bus.c | 4 | ||||
| -rw-r--r-- | drivers/nvdimm/dax_devs.c | 2 | ||||
| -rw-r--r-- | drivers/nvdimm/dimm_devs.c | 2 | ||||
| -rw-r--r-- | drivers/nvdimm/nd-core.h | 10 | ||||
| -rw-r--r-- | drivers/nvdimm/nd.h | 6 | ||||
| -rw-r--r-- | drivers/nvdimm/pfn_devs.c | 42 | ||||
| -rw-r--r-- | drivers/nvdimm/pmem.c | 24 | ||||
| -rw-r--r-- | drivers/nvdimm/region_devs.c | 4 |
10 files changed, 62 insertions, 67 deletions
diff --git a/drivers/nvdimm/Kconfig b/drivers/nvdimm/Kconfig index 79d93126453d..77b06d54cc62 100644 --- a/drivers/nvdimm/Kconfig +++ b/drivers/nvdimm/Kconfig @@ -102,6 +102,25 @@ config NVDIMM_KEYS depends on ENCRYPTED_KEYS depends on (LIBNVDIMM=ENCRYPTED_KEYS) || LIBNVDIMM=m +config NVDIMM_KMSAN + bool + depends on KMSAN + help + KMSAN, and other memory debug facilities, increase the size of + 'struct page' to contain extra metadata. This collides with + the NVDIMM capability to store a potentially + larger-than-"System RAM" size 'struct page' array in a + reservation of persistent memory rather than limited / + precious DRAM. However, that reservation needs to persist for + the life of the given NVDIMM namespace. If you are using KMSAN + to debug an issue unrelated to NVDIMMs or DAX then say N to this + option. Otherwise, say Y but understand that any namespaces + (with the page array stored pmem) created with this build of + the kernel will permanently reserve and strand excess + capacity compared to the CONFIG_KMSAN=n case. + + Select N if unsure. + config NVDIMM_TEST_BUILD tristate "Build the unit test core" depends on m diff --git a/drivers/nvdimm/btt.c b/drivers/nvdimm/btt.c index 0297b7882e33..d5593b0dc700 100644 --- a/drivers/nvdimm/btt.c +++ b/drivers/nvdimm/btt.c @@ -1482,20 +1482,6 @@ static void btt_submit_bio(struct bio *bio) bio_endio(bio); } -static int btt_rw_page(struct block_device *bdev, sector_t sector, - struct page *page, enum req_op op) -{ - struct btt *btt = bdev->bd_disk->private_data; - int rc; - - rc = btt_do_bvec(btt, NULL, page, thp_size(page), 0, op, sector); - if (rc == 0) - page_endio(page, op_is_write(op), 0); - - return rc; -} - - static int btt_getgeo(struct block_device *bd, struct hd_geometry *geo) { /* some standard values */ @@ -1508,7 +1494,6 @@ static int btt_getgeo(struct block_device *bd, struct hd_geometry *geo) static const struct block_device_operations btt_fops = { .owner = THIS_MODULE, .submit_bio = btt_submit_bio, - .rw_page = btt_rw_page, .getgeo = btt_getgeo, }; @@ -1530,6 +1515,7 @@ static int btt_blk_init(struct btt *btt) blk_queue_logical_block_size(btt->btt_disk->queue, btt->sector_size); blk_queue_max_hw_sectors(btt->btt_disk->queue, UINT_MAX); blk_queue_flag_set(QUEUE_FLAG_NONROT, btt->btt_disk->queue); + blk_queue_flag_set(QUEUE_FLAG_SYNCHRONOUS, btt->btt_disk->queue); if (btt_meta_size(btt)) { rc = nd_integrity_init(btt->btt_disk, btt_meta_size(btt)); diff --git a/drivers/nvdimm/bus.c b/drivers/nvdimm/bus.c index 5ad49056921b..4976a0069e9c 100644 --- a/drivers/nvdimm/bus.c +++ b/drivers/nvdimm/bus.c @@ -28,7 +28,7 @@ static int nvdimm_bus_major; struct class *nd_class; static DEFINE_IDA(nd_ida); -static int to_nd_device_type(struct device *dev) +static int to_nd_device_type(const struct device *dev) { if (is_nvdimm(dev)) return ND_DEVICE_DIMM; @@ -42,7 +42,7 @@ static int to_nd_device_type(struct device *dev) return 0; } -static int nvdimm_bus_uevent(struct device *dev, struct kobj_uevent_env *env) +static int nvdimm_bus_uevent(const struct device *dev, struct kobj_uevent_env *env) { return add_uevent_var(env, "MODALIAS=" ND_DEVICE_MODALIAS_FMT, to_nd_device_type(dev)); diff --git a/drivers/nvdimm/dax_devs.c b/drivers/nvdimm/dax_devs.c index 7f4a9d28b670..3bd61f245788 100644 --- a/drivers/nvdimm/dax_devs.c +++ b/drivers/nvdimm/dax_devs.c @@ -38,7 +38,7 @@ static const struct device_type nd_dax_device_type = { .groups = nd_pfn_attribute_groups, }; -bool is_nd_dax(struct device *dev) +bool is_nd_dax(const struct device *dev) { return dev ? dev->type == &nd_dax_device_type : false; } diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c index 6d3b03a9fa02..957f7c3d17ba 100644 --- a/drivers/nvdimm/dimm_devs.c +++ b/drivers/nvdimm/dimm_devs.c @@ -572,7 +572,7 @@ static const struct device_type nvdimm_device_type = { .groups = nvdimm_attribute_groups, }; -bool is_nvdimm(struct device *dev) +bool is_nvdimm(const struct device *dev) { return dev->type == &nvdimm_device_type; } diff --git a/drivers/nvdimm/nd-core.h b/drivers/nvdimm/nd-core.h index 845408f10655..86976a9e8a15 100644 --- a/drivers/nvdimm/nd-core.h +++ b/drivers/nvdimm/nd-core.h @@ -82,14 +82,14 @@ static inline void nvdimm_security_overwrite_query(struct work_struct *work) } #endif -bool is_nvdimm(struct device *dev); -bool is_nd_pmem(struct device *dev); -bool is_nd_volatile(struct device *dev); -static inline bool is_nd_region(struct device *dev) +bool is_nvdimm(const struct device *dev); +bool is_nd_pmem(const struct device *dev); +bool is_nd_volatile(const struct device *dev); +static inline bool is_nd_region(const struct device *dev) { return is_nd_pmem(dev) || is_nd_volatile(dev); } -static inline bool is_memory(struct device *dev) +static inline bool is_memory(const struct device *dev) { return is_nd_pmem(dev) || is_nd_volatile(dev); } diff --git a/drivers/nvdimm/nd.h b/drivers/nvdimm/nd.h index 85ca5b4da3cf..e8b9d27dbb3c 100644 --- a/drivers/nvdimm/nd.h +++ b/drivers/nvdimm/nd.h @@ -599,7 +599,7 @@ static inline int nd_pfn_validate(struct nd_pfn *nd_pfn, const char *sig) struct nd_dax *to_nd_dax(struct device *dev); #if IS_ENABLED(CONFIG_NVDIMM_DAX) int nd_dax_probe(struct device *dev, struct nd_namespace_common *ndns); -bool is_nd_dax(struct device *dev); +bool is_nd_dax(const struct device *dev); struct device *nd_dax_create(struct nd_region *nd_region); #else static inline int nd_dax_probe(struct device *dev, @@ -608,7 +608,7 @@ static inline int nd_dax_probe(struct device *dev, return -ENODEV; } -static inline bool is_nd_dax(struct device *dev) +static inline bool is_nd_dax(const struct device *dev) { return false; } @@ -652,7 +652,7 @@ void devm_namespace_disable(struct device *dev, struct nd_namespace_common *ndns); #if IS_ENABLED(CONFIG_ND_CLAIM) /* max struct page size independent of kernel config */ -#define MAX_STRUCT_PAGE_SIZE 128 +#define MAX_STRUCT_PAGE_SIZE 64 int nvdimm_setup_pfn(struct nd_pfn *nd_pfn, struct dev_pagemap *pgmap); #else static inline int nvdimm_setup_pfn(struct nd_pfn *nd_pfn, diff --git a/drivers/nvdimm/pfn_devs.c b/drivers/nvdimm/pfn_devs.c index 61af072ac98f..af7d9301520c 100644 --- a/drivers/nvdimm/pfn_devs.c +++ b/drivers/nvdimm/pfn_devs.c @@ -13,6 +13,8 @@ #include "pfn.h" #include "nd.h" +static const bool page_struct_override = IS_ENABLED(CONFIG_NVDIMM_KMSAN); + static void nd_pfn_release(struct device *dev) { struct nd_region *nd_region = to_nd_region(dev->parent); @@ -758,12 +760,6 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn) return -ENXIO; } - /* - * Note, we use 64 here for the standard size of struct page, - * debugging options may cause it to be larger in which case the - * implementation will limit the pfns advertised through - * ->direct_access() to those that are included in the memmap. - */ start = nsio->res.start; size = resource_size(&nsio->res); npfns = PHYS_PFN(size - SZ_8K); @@ -782,20 +778,33 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn) } end_trunc = start + size - ALIGN_DOWN(start + size, align); if (nd_pfn->mode == PFN_MODE_PMEM) { + unsigned long page_map_size = MAX_STRUCT_PAGE_SIZE * npfns; + /* * The altmap should be padded out to the block size used * when populating the vmemmap. This *should* be equal to * PMD_SIZE for most architectures. * - * Also make sure size of struct page is less than 128. We - * want to make sure we use large enough size here so that - * we don't have a dynamic reserve space depending on - * struct page size. But we also want to make sure we notice - * when we end up adding new elements to struct page. + * Also make sure size of struct page is less than + * MAX_STRUCT_PAGE_SIZE. The goal here is compatibility in the + * face of production kernel configurations that reduce the + * 'struct page' size below MAX_STRUCT_PAGE_SIZE. For debug + * kernel configurations that increase the 'struct page' size + * above MAX_STRUCT_PAGE_SIZE, the page_struct_override allows + * for continuing with the capacity that will be wasted when + * reverting to a production kernel configuration. Otherwise, + * those configurations are blocked by default. */ - BUILD_BUG_ON(sizeof(struct page) > MAX_STRUCT_PAGE_SIZE); - offset = ALIGN(start + SZ_8K + MAX_STRUCT_PAGE_SIZE * npfns, align) - - start; + if (sizeof(struct page) > MAX_STRUCT_PAGE_SIZE) { + if (page_struct_override) + page_map_size = sizeof(struct page) * npfns; + else { + dev_err(&nd_pfn->dev, + "Memory debug options prevent using pmem for the page map\n"); + return -EINVAL; + } + } + offset = ALIGN(start + SZ_8K + page_map_size, align) - start; } else if (nd_pfn->mode == PFN_MODE_RAM) offset = ALIGN(start + SZ_8K, align) - start; else @@ -818,7 +827,10 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn) pfn_sb->version_minor = cpu_to_le16(4); pfn_sb->end_trunc = cpu_to_le32(end_trunc); pfn_sb->align = cpu_to_le32(nd_pfn->align); - pfn_sb->page_struct_size = cpu_to_le16(MAX_STRUCT_PAGE_SIZE); + if (sizeof(struct page) > MAX_STRUCT_PAGE_SIZE && page_struct_override) + pfn_sb->page_struct_size = cpu_to_le16(sizeof(struct page)); + else + pfn_sb->page_struct_size = cpu_to_le16(MAX_STRUCT_PAGE_SIZE); pfn_sb->page_size = cpu_to_le32(PAGE_SIZE); checksum = nd_sb_checksum((struct nd_gen_sb *) pfn_sb); pfn_sb->checksum = cpu_to_le64(checksum); diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 96e6e9a5f235..ceea55f621cc 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -238,28 +238,6 @@ static void pmem_submit_bio(struct bio *bio) bio_endio(bio); } -static int pmem_rw_page(struct block_device *bdev, sector_t sector, - struct page *page, enum req_op op) -{ - struct pmem_device *pmem = bdev->bd_disk->private_data; - blk_status_t rc; - - if (op_is_write(op)) - rc = pmem_do_write(pmem, page, 0, sector, thp_size(page)); - else - rc = pmem_do_read(pmem, page, 0, sector, thp_size(page)); - /* - * The ->rw_page interface is subtle and tricky. The core - * retries on any error, so we can only invoke page_endio() in - * the successful completion case. Otherwise, we'll see crashes - * caused by double completion. - */ - if (rc == 0) - page_endio(page, op_is_write(op), 0); - - return blk_status_to_errno(rc); -} - /* see "strong" declaration in tools/testing/nvdimm/pmem-dax.c */ __weak long __pmem_direct_access(struct pmem_device *pmem, pgoff_t pgoff, long nr_pages, enum dax_access_mode mode, void **kaddr, @@ -310,7 +288,6 @@ __weak long __pmem_direct_access(struct pmem_device *pmem, pgoff_t pgoff, static const struct block_device_operations pmem_fops = { .owner = THIS_MODULE, .submit_bio = pmem_submit_bio, - .rw_page = pmem_rw_page, }; static int pmem_dax_zero_page_range(struct dax_device *dax_dev, pgoff_t pgoff, @@ -565,6 +542,7 @@ static int pmem_attach_disk(struct device *dev, blk_queue_logical_block_size(q, pmem_sector_size(ndns)); blk_queue_max_hw_sectors(q, UINT_MAX); blk_queue_flag_set(QUEUE_FLAG_NONROT, q); + blk_queue_flag_set(QUEUE_FLAG_SYNCHRONOUS, q); if (pmem->pfn_flags & PFN_MAP) blk_queue_flag_set(QUEUE_FLAG_DAX, q); diff --git a/drivers/nvdimm/region_devs.c b/drivers/nvdimm/region_devs.c index 83dbf398ea84..8f134d63af13 100644 --- a/drivers/nvdimm/region_devs.c +++ b/drivers/nvdimm/region_devs.c @@ -839,12 +839,12 @@ static const struct device_type nd_volatile_device_type = { .groups = nd_region_attribute_groups, }; -bool is_nd_pmem(struct device *dev) +bool is_nd_pmem(const struct device *dev) { return dev ? dev->type == &nd_pmem_device_type : false; } -bool is_nd_volatile(struct device *dev) +bool is_nd_volatile(const struct device *dev) { return dev ? dev->type == &nd_volatile_device_type : false; } |