From 782b86641e5d471e9eb1cf0072c012d2f758e568 Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Tue, 26 Jan 2016 21:57:24 +0100 Subject: xen, mm: Set IORESOURCE_SYSTEM_RAM to System RAM Set IORESOURCE_SYSTEM_RAM in struct resource.flags of "System RAM" entries. Signed-off-by: Toshi Kani Signed-off-by: Borislav Petkov Acked-by: David Vrabel # xen Cc: Andrew Banman Cc: Andrew Morton Cc: Andy Lutomirski Cc: Boris Ostrovsky Cc: Borislav Petkov Cc: Brian Gerst Cc: Dan Williams Cc: David Rientjes Cc: Denys Vlasenko Cc: Gu Zheng Cc: H. Peter Anvin Cc: Konrad Rzeszutek Wilk Cc: Linus Torvalds Cc: Luis R. Rodriguez Cc: Mel Gorman Cc: Naoya Horiguchi Cc: Peter Zijlstra Cc: Tang Chen Cc: Thomas Gleixner Cc: Toshi Kani Cc: linux-arch@vger.kernel.org Cc: linux-mm Cc: xen-devel@lists.xenproject.org Link: http://lkml.kernel.org/r/1453841853-11383-9-git-send-email-bp@alien8.de Signed-off-by: Ingo Molnar --- drivers/xen/balloon.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c index 12eab503efd1..dc4305b407bf 100644 --- a/drivers/xen/balloon.c +++ b/drivers/xen/balloon.c @@ -257,7 +257,7 @@ static struct resource *additional_memory_resource(phys_addr_t size) return NULL; res->name = "System RAM"; - res->flags = IORESOURCE_MEM | IORESOURCE_BUSY; + res->flags = IORESOURCE_SYSTEM_RAM | IORESOURCE_BUSY; ret = allocate_resource(&iomem_resource, res, size, 0, -1, -- cgit v1.2.3 From 9a975bee4b3945b271bcff18a520d4863c210f8b Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Tue, 26 Jan 2016 21:57:25 +0100 Subject: drivers: Initialize resource entry to zero I/O resource descriptor, 'desc' in struct resource, needs to be initialized to zero by default. Some drivers call kmalloc() to allocate a resource entry, but do not initialize it to zero by memset(). Change these drivers to call kzalloc(), instead. Signed-off-by: Toshi Kani Signed-off-by: Borislav Petkov Acked-by: Alexandre Bounine Acked-by: Helge Deller Acked-by: Rafael J. Wysocki Acked-by: Simon Horman Cc: Andrew Morton Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Linus Torvalds Cc: Luis R. Rodriguez Cc: Peter Zijlstra Cc: Thomas Gleixner Cc: Toshi Kani Cc: linux-acpi@vger.kernel.org Cc: linux-arch@vger.kernel.org Cc: linux-mm Cc: linux-parisc@vger.kernel.org Cc: linux-renesas-soc@vger.kernel.org Cc: linux-sh@vger.kernel.org Link: http://lkml.kernel.org/r/1453841853-11383-10-git-send-email-bp@alien8.de Signed-off-by: Ingo Molnar --- drivers/acpi/acpi_platform.c | 2 +- drivers/parisc/eisa_enumerator.c | 4 ++-- drivers/rapidio/rio.c | 8 ++++---- drivers/sh/superhyway/superhyway.c | 2 +- 4 files changed, 8 insertions(+), 8 deletions(-) (limited to 'drivers') diff --git a/drivers/acpi/acpi_platform.c b/drivers/acpi/acpi_platform.c index 296b7a14893a..b6f7fa3a1d40 100644 --- a/drivers/acpi/acpi_platform.c +++ b/drivers/acpi/acpi_platform.c @@ -62,7 +62,7 @@ struct platform_device *acpi_create_platform_device(struct acpi_device *adev) if (count < 0) { return NULL; } else if (count > 0) { - resources = kmalloc(count * sizeof(struct resource), + resources = kzalloc(count * sizeof(struct resource), GFP_KERNEL); if (!resources) { dev_err(&adev->dev, "No memory for resources\n"); diff --git a/drivers/parisc/eisa_enumerator.c b/drivers/parisc/eisa_enumerator.c index a656d9e83343..21905fef2cbf 100644 --- a/drivers/parisc/eisa_enumerator.c +++ b/drivers/parisc/eisa_enumerator.c @@ -91,7 +91,7 @@ static int configure_memory(const unsigned char *buf, for (i=0;iname = name; @@ -183,7 +183,7 @@ static int configure_port(const unsigned char *buf, struct resource *io_parent, for (i=0;iname = board; res->start = get_16(buf+len+1); res->end = get_16(buf+len+1)+(c&HPEE_PORT_SIZE_MASK)+1; diff --git a/drivers/rapidio/rio.c b/drivers/rapidio/rio.c index d7b87c64b7cd..e220edc85c68 100644 --- a/drivers/rapidio/rio.c +++ b/drivers/rapidio/rio.c @@ -117,7 +117,7 @@ int rio_request_inb_mbox(struct rio_mport *mport, if (mport->ops->open_inb_mbox == NULL) goto out; - res = kmalloc(sizeof(struct resource), GFP_KERNEL); + res = kzalloc(sizeof(struct resource), GFP_KERNEL); if (res) { rio_init_mbox_res(res, mbox, mbox); @@ -185,7 +185,7 @@ int rio_request_outb_mbox(struct rio_mport *mport, if (mport->ops->open_outb_mbox == NULL) goto out; - res = kmalloc(sizeof(struct resource), GFP_KERNEL); + res = kzalloc(sizeof(struct resource), GFP_KERNEL); if (res) { rio_init_mbox_res(res, mbox, mbox); @@ -285,7 +285,7 @@ int rio_request_inb_dbell(struct rio_mport *mport, { int rc = 0; - struct resource *res = kmalloc(sizeof(struct resource), GFP_KERNEL); + struct resource *res = kzalloc(sizeof(struct resource), GFP_KERNEL); if (res) { rio_init_dbell_res(res, start, end); @@ -360,7 +360,7 @@ int rio_release_inb_dbell(struct rio_mport *mport, u16 start, u16 end) struct resource *rio_request_outb_dbell(struct rio_dev *rdev, u16 start, u16 end) { - struct resource *res = kmalloc(sizeof(struct resource), GFP_KERNEL); + struct resource *res = kzalloc(sizeof(struct resource), GFP_KERNEL); if (res) { rio_init_dbell_res(res, start, end); diff --git a/drivers/sh/superhyway/superhyway.c b/drivers/sh/superhyway/superhyway.c index 2d9e7f3d5611..bb1fb7712134 100644 --- a/drivers/sh/superhyway/superhyway.c +++ b/drivers/sh/superhyway/superhyway.c @@ -66,7 +66,7 @@ int superhyway_add_device(unsigned long base, struct superhyway_device *sdev, superhyway_read_vcr(dev, base, &dev->vcr); if (!dev->resource) { - dev->resource = kmalloc(sizeof(struct resource), GFP_KERNEL); + dev->resource = kzalloc(sizeof(struct resource), GFP_KERNEL); if (!dev->resource) { kfree(dev); return -ENOMEM; -- cgit v1.2.3 From f0f4711aa16b82016c0b6e59871934bbd71258da Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Tue, 26 Jan 2016 21:57:30 +0100 Subject: x86, kexec, nvdimm: Use walk_iomem_res_desc() for iomem search Change the callers of walk_iomem_res() scanning for the following resources by name to use walk_iomem_res_desc() instead. "ACPI Tables" "ACPI Non-volatile Storage" "Persistent Memory (legacy)" "Crash kernel" Note, the caller of walk_iomem_res() with "GART" will be removed in a later patch. Signed-off-by: Toshi Kani Signed-off-by: Borislav Petkov Reviewed-by: Dave Young Cc: Andrew Morton Cc: Andy Lutomirski Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Chun-Yi Cc: Dan Williams Cc: Denys Vlasenko Cc: Don Zickus Cc: H. Peter Anvin Cc: Lee, Chun-Yi Cc: Linus Torvalds Cc: Luis R. Rodriguez Cc: Minfei Huang Cc: Peter Zijlstra (Intel) Cc: Ross Zwisler Cc: Stephen Rothwell Cc: Takao Indoh Cc: Thomas Gleixner Cc: Toshi Kani Cc: kexec@lists.infradead.org Cc: linux-arch@vger.kernel.org Cc: linux-mm Cc: linux-nvdimm@lists.01.org Link: http://lkml.kernel.org/r/1453841853-11383-15-git-send-email-bp@alien8.de Signed-off-by: Ingo Molnar --- drivers/nvdimm/e820.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'drivers') diff --git a/drivers/nvdimm/e820.c b/drivers/nvdimm/e820.c index b0045a505dc8..95825b38559a 100644 --- a/drivers/nvdimm/e820.c +++ b/drivers/nvdimm/e820.c @@ -55,7 +55,7 @@ static int e820_pmem_probe(struct platform_device *pdev) for (p = iomem_resource.child; p ; p = p->sibling) { struct nd_region_desc ndr_desc; - if (strncmp(p->name, "Persistent Memory (legacy)", 26) != 0) + if (p->desc != IORES_DESC_PERSISTENT_MEMORY_LEGACY) continue; memset(&ndr_desc, 0, sizeof(ndr_desc)); -- cgit v1.2.3 From 4650bac1fc45d64aef62ab99aa4db93d41dedbd9 Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Tue, 26 Jan 2016 21:57:33 +0100 Subject: ACPI/EINJ: Allow memory error injection to NVDIMM In the case of memory error injection, einj_error_inject() checks if a target address is System RAM. Change this check to allow injecting a memory error into NVDIMM memory by calling region_intersects() with IORES_DESC_PERSISTENT_MEMORY. This enables memory error testing on both System RAM and NVDIMM. In addition, page_is_ram() is replaced with region_intersects() with IORESOURCE_SYSTEM_RAM, so that it can verify a target address range with the requested size. Signed-off-by: Toshi Kani Signed-off-by: Borislav Petkov Reviewed-by: Dan Williams Acked-by: Tony Luck Cc: Andrew Morton Cc: Andy Lutomirski Cc: Borislav Petkov Cc: Brian Gerst Cc: Denys Vlasenko Cc: H. Peter Anvin Cc: Jarkko Nikula Cc: Len Brown Cc: Linus Torvalds Cc: Luis R. Rodriguez Cc: Peter Zijlstra Cc: Rafael J. Wysocki Cc: Thomas Gleixner Cc: Toshi Kani Cc: Vishal Verma Cc: linux-acpi@vger.kernel.org Cc: linux-arch@vger.kernel.org Cc: linux-mm Cc: linux-nvdimm@lists.01.org Link: http://lkml.kernel.org/r/1453841853-11383-18-git-send-email-bp@alien8.de Signed-off-by: Ingo Molnar --- drivers/acpi/apei/einj.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) (limited to 'drivers') diff --git a/drivers/acpi/apei/einj.c b/drivers/acpi/apei/einj.c index 0431883653be..559c1173de1c 100644 --- a/drivers/acpi/apei/einj.c +++ b/drivers/acpi/apei/einj.c @@ -519,7 +519,7 @@ static int einj_error_inject(u32 type, u32 flags, u64 param1, u64 param2, u64 param3, u64 param4) { int rc; - unsigned long pfn; + u64 base_addr, size; /* If user manually set "flags", make sure it is legal */ if (flags && (flags & @@ -545,10 +545,17 @@ static int einj_error_inject(u32 type, u32 flags, u64 param1, u64 param2, /* * Disallow crazy address masks that give BIOS leeway to pick * injection address almost anywhere. Insist on page or - * better granularity and that target address is normal RAM. + * better granularity and that target address is normal RAM or + * NVDIMM. */ - pfn = PFN_DOWN(param1 & param2); - if (!page_is_ram(pfn) || ((param2 & PAGE_MASK) != PAGE_MASK)) + base_addr = param1 & param2; + size = ~param2 + 1; + + if (((param2 & PAGE_MASK) != PAGE_MASK) || + ((region_intersects(base_addr, size, IORESOURCE_SYSTEM_RAM, IORES_DESC_NONE) + != REGION_INTERSECTS) && + (region_intersects(base_addr, size, IORESOURCE_MEM, IORES_DESC_PERSISTENT_MEMORY) + != REGION_INTERSECTS))) return -EINVAL; inject: -- cgit v1.2.3 From d9cbe09d39aa13f6924dc5fb88325de7ef01a72e Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Thu, 3 Mar 2016 09:14:36 -0800 Subject: libnvdimm, pmem: fix 'pfn' support for section-misaligned namespaces The altmap for a section-misaligned namespace needs to arrange for the base_pfn to be section-aligned. As a result the 'reserve' region (pfns from base that do not have a struct page) must be increased. Otherwise we trip the altmap validation check in __add_pages: if (altmap->base_pfn != phys_start_pfn || vmem_altmap_offset(altmap) > nr_pages) { pr_warn_once("memory add fail, invalid altmap\n"); return -EINVAL; } Signed-off-by: Dan Williams --- drivers/nvdimm/pfn.h | 13 +++++++++++++ drivers/nvdimm/pmem.c | 24 ++++++++++++++++++++++-- 2 files changed, 35 insertions(+), 2 deletions(-) (limited to 'drivers') diff --git a/drivers/nvdimm/pfn.h b/drivers/nvdimm/pfn.h index cc243754acef..6ee707e5b279 100644 --- a/drivers/nvdimm/pfn.h +++ b/drivers/nvdimm/pfn.h @@ -15,6 +15,7 @@ #define __NVDIMM_PFN_H #include +#include #define PFN_SIG_LEN 16 #define PFN_SIG "NVDIMM_PFN_INFO\0" @@ -32,4 +33,16 @@ struct nd_pfn_sb { u8 padding[4012]; __le64 checksum; }; + +#ifdef CONFIG_SPARSEMEM +#define PFN_SECTION_ALIGN_DOWN(x) SECTION_ALIGN_DOWN(x) +#define PFN_SECTION_ALIGN_UP(x) SECTION_ALIGN_UP(x) +#else +/* + * In this case ZONE_DEVICE=n and we will disable 'pfn' device support, + * but we still want pmem to compile. + */ +#define PFN_SECTION_ALIGN_DOWN(x) (x) +#define PFN_SECTION_ALIGN_UP(x) (x) +#endif #endif /* __NVDIMM_PFN_H */ diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 8d0b54670184..59d568ab7556 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -356,6 +356,26 @@ static int nvdimm_namespace_detach_pfn(struct nd_namespace_common *ndns) return 0; } +/* + * We hotplug memory at section granularity, pad the reserved area from + * the previous section base to the namespace base address. + */ +static unsigned long init_altmap_base(resource_size_t base) +{ + unsigned long base_pfn = __phys_to_pfn(base); + + return PFN_SECTION_ALIGN_DOWN(base_pfn); +} + +static unsigned long init_altmap_reserve(resource_size_t base) +{ + unsigned long reserve = __phys_to_pfn(SZ_8K); + unsigned long base_pfn = __phys_to_pfn(base); + + reserve += base_pfn - PFN_SECTION_ALIGN_DOWN(base_pfn); + return reserve; +} + static int nvdimm_namespace_attach_pfn(struct nd_namespace_common *ndns) { struct nd_namespace_io *nsio = to_nd_namespace_io(&ndns->dev); @@ -369,8 +389,8 @@ static int nvdimm_namespace_attach_pfn(struct nd_namespace_common *ndns) phys_addr_t offset; int rc; struct vmem_altmap __altmap = { - .base_pfn = __phys_to_pfn(nsio->res.start), - .reserve = __phys_to_pfn(SZ_8K), + .base_pfn = init_altmap_base(nsio->res.start), + .reserve = init_altmap_reserve(nsio->res.start), }; if (!nd_pfn->uuid || !nd_pfn->ndns) -- cgit v1.2.3 From cfe30b872058f211630eda7f65fb19d83beaaa3c Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Thu, 3 Mar 2016 09:38:00 -0800 Subject: libnvdimm, pmem: adjust for section collisions with 'System RAM' On a platform where 'Persistent Memory' and 'System RAM' are mixed within a given sparsemem section, trim the namespace and notify about the sub-optimal alignment. Cc: Toshi Kani Cc: Ross Zwisler Signed-off-by: Dan Williams --- drivers/nvdimm/namespace_devs.c | 7 +++ drivers/nvdimm/pfn.h | 10 +++- drivers/nvdimm/pfn_devs.c | 5 ++ drivers/nvdimm/pmem.c | 125 +++++++++++++++++++++++++++++----------- 4 files changed, 111 insertions(+), 36 deletions(-) (limited to 'drivers') diff --git a/drivers/nvdimm/namespace_devs.c b/drivers/nvdimm/namespace_devs.c index 9edf7eb7d17c..f5cb88601359 100644 --- a/drivers/nvdimm/namespace_devs.c +++ b/drivers/nvdimm/namespace_devs.c @@ -133,6 +133,7 @@ bool nd_is_uuid_unique(struct device *dev, u8 *uuid) bool pmem_should_map_pages(struct device *dev) { struct nd_region *nd_region = to_nd_region(dev->parent); + struct nd_namespace_io *nsio; if (!IS_ENABLED(CONFIG_ZONE_DEVICE)) return false; @@ -143,6 +144,12 @@ bool pmem_should_map_pages(struct device *dev) if (is_nd_pfn(dev) || is_nd_btt(dev)) return false; + nsio = to_nd_namespace_io(dev); + if (region_intersects(nsio->res.start, resource_size(&nsio->res), + IORESOURCE_SYSTEM_RAM, + IORES_DESC_NONE) == REGION_MIXED) + return false; + #ifdef ARCH_MEMREMAP_PMEM return ARCH_MEMREMAP_PMEM == MEMREMAP_WB; #else diff --git a/drivers/nvdimm/pfn.h b/drivers/nvdimm/pfn.h index 6ee707e5b279..8e343a3ca873 100644 --- a/drivers/nvdimm/pfn.h +++ b/drivers/nvdimm/pfn.h @@ -27,10 +27,13 @@ struct nd_pfn_sb { __le32 flags; __le16 version_major; __le16 version_minor; - __le64 dataoff; + __le64 dataoff; /* relative to namespace_base + start_pad */ __le64 npfns; __le32 mode; - u8 padding[4012]; + /* minor-version-1 additions for section alignment */ + __le32 start_pad; + __le32 end_trunc; + u8 padding[4004]; __le64 checksum; }; @@ -45,4 +48,7 @@ struct nd_pfn_sb { #define PFN_SECTION_ALIGN_DOWN(x) (x) #define PFN_SECTION_ALIGN_UP(x) (x) #endif + +#define PHYS_SECTION_ALIGN_DOWN(x) PFN_PHYS(PFN_SECTION_ALIGN_DOWN(PHYS_PFN(x))) +#define PHYS_SECTION_ALIGN_UP(x) PFN_PHYS(PFN_SECTION_ALIGN_UP(PHYS_PFN(x))) #endif /* __NVDIMM_PFN_H */ diff --git a/drivers/nvdimm/pfn_devs.c b/drivers/nvdimm/pfn_devs.c index ae81a2f1da50..75a31a7359fb 100644 --- a/drivers/nvdimm/pfn_devs.c +++ b/drivers/nvdimm/pfn_devs.c @@ -299,6 +299,11 @@ int nd_pfn_validate(struct nd_pfn *nd_pfn) if (memcmp(pfn_sb->parent_uuid, parent_uuid, 16) != 0) return -ENODEV; + if (__le16_to_cpu(pfn_sb->version_minor) < 1) { + pfn_sb->start_pad = 0; + pfn_sb->end_trunc = 0; + } + switch (le32_to_cpu(pfn_sb->mode)) { case PFN_MODE_RAM: case PFN_MODE_PMEM: diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 59d568ab7556..0cb450e1b400 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -43,7 +43,10 @@ struct pmem_device { phys_addr_t data_offset; u64 pfn_flags; void __pmem *virt_addr; + /* immutable base size of the namespace */ size_t size; + /* trim size when namespace capacity has been section aligned */ + u32 pfn_pad; struct badblocks bb; }; @@ -145,7 +148,7 @@ static long pmem_direct_access(struct block_device *bdev, sector_t sector, *kaddr = pmem->virt_addr + offset; *pfn = phys_to_pfn_t(pmem->phys_addr + offset, pmem->pfn_flags); - return pmem->size - offset; + return pmem->size - pmem->pfn_pad - offset; } static const struct block_device_operations pmem_fops = { @@ -236,7 +239,8 @@ static int pmem_attach_disk(struct device *dev, disk->flags = GENHD_FL_EXT_DEVT; nvdimm_namespace_disk_name(ndns, disk->disk_name); disk->driverfs_dev = dev; - set_capacity(disk, (pmem->size - pmem->data_offset) / 512); + set_capacity(disk, (pmem->size - pmem->pfn_pad - pmem->data_offset) + / 512); pmem->pmem_disk = disk; devm_exit_badblocks(dev, &pmem->bb); if (devm_init_badblocks(dev, &pmem->bb)) @@ -279,6 +283,9 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn) struct nd_pfn_sb *pfn_sb = kzalloc(sizeof(*pfn_sb), GFP_KERNEL); struct pmem_device *pmem = dev_get_drvdata(&nd_pfn->dev); struct nd_namespace_common *ndns = nd_pfn->ndns; + u32 start_pad = 0, end_trunc = 0; + resource_size_t start, size; + struct nd_namespace_io *nsio; struct nd_region *nd_region; unsigned long npfns; phys_addr_t offset; @@ -304,21 +311,56 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn) } memset(pfn_sb, 0, sizeof(*pfn_sb)); - npfns = (pmem->size - SZ_8K) / SZ_4K; + + /* + * Check if pmem collides with 'System RAM' when section aligned and + * trim it accordingly + */ + nsio = to_nd_namespace_io(&ndns->dev); + start = PHYS_SECTION_ALIGN_DOWN(nsio->res.start); + size = resource_size(&nsio->res); + if (region_intersects(start, size, IORESOURCE_SYSTEM_RAM, + IORES_DESC_NONE) == REGION_MIXED) { + + start = nsio->res.start; + start_pad = PHYS_SECTION_ALIGN_UP(start) - start; + } + + start = nsio->res.start; + size = PHYS_SECTION_ALIGN_UP(start + size) - start; + if (region_intersects(start, size, IORESOURCE_SYSTEM_RAM, + IORES_DESC_NONE) == REGION_MIXED) { + size = resource_size(&nsio->res); + end_trunc = start + size - PHYS_SECTION_ALIGN_DOWN(start + size); + } + + if (start_pad + end_trunc) + dev_info(&nd_pfn->dev, "%s section collision, truncate %d bytes\n", + dev_name(&ndns->dev), start_pad + end_trunc); + /* * 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 += start_pad; + npfns = (pmem->size - start_pad - end_trunc - SZ_8K) / SZ_4K; if (nd_pfn->mode == PFN_MODE_PMEM) - offset = ALIGN(SZ_8K + 64 * npfns, nd_pfn->align); + offset = ALIGN(start + SZ_8K + 64 * npfns, nd_pfn->align) + - start; else if (nd_pfn->mode == PFN_MODE_RAM) - offset = ALIGN(SZ_8K, nd_pfn->align); + offset = ALIGN(start + SZ_8K, nd_pfn->align) - start; else goto err; - npfns = (pmem->size - offset) / SZ_4K; + if (offset + start_pad + end_trunc >= pmem->size) { + dev_err(&nd_pfn->dev, "%s unable to satisfy requested alignment\n", + dev_name(&ndns->dev)); + goto err; + } + + npfns = (pmem->size - offset - start_pad - end_trunc) / SZ_4K; pfn_sb->mode = cpu_to_le32(nd_pfn->mode); pfn_sb->dataoff = cpu_to_le64(offset); pfn_sb->npfns = cpu_to_le64(npfns); @@ -326,6 +368,9 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn) memcpy(pfn_sb->uuid, nd_pfn->uuid, 16); memcpy(pfn_sb->parent_uuid, nd_dev_to_uuid(&ndns->dev), 16); pfn_sb->version_major = cpu_to_le16(1); + pfn_sb->version_minor = cpu_to_le16(1); + pfn_sb->start_pad = cpu_to_le32(start_pad); + pfn_sb->end_trunc = cpu_to_le32(end_trunc); checksum = nd_sb_checksum((struct nd_gen_sb *) pfn_sb); pfn_sb->checksum = cpu_to_le64(checksum); @@ -376,41 +421,36 @@ static unsigned long init_altmap_reserve(resource_size_t base) return reserve; } -static int nvdimm_namespace_attach_pfn(struct nd_namespace_common *ndns) +static int __nvdimm_namespace_attach_pfn(struct nd_pfn *nd_pfn) { - struct nd_namespace_io *nsio = to_nd_namespace_io(&ndns->dev); - struct nd_pfn *nd_pfn = to_nd_pfn(ndns->claim); - struct device *dev = &nd_pfn->dev; - struct nd_region *nd_region; - struct vmem_altmap *altmap; - struct nd_pfn_sb *pfn_sb; - struct pmem_device *pmem; - struct request_queue *q; - phys_addr_t offset; int rc; + struct resource res; + struct request_queue *q; + struct pmem_device *pmem; + struct vmem_altmap *altmap; + struct device *dev = &nd_pfn->dev; + struct nd_pfn_sb *pfn_sb = nd_pfn->pfn_sb; + struct nd_namespace_common *ndns = nd_pfn->ndns; + u32 start_pad = __le32_to_cpu(pfn_sb->start_pad); + u32 end_trunc = __le32_to_cpu(pfn_sb->end_trunc); + struct nd_namespace_io *nsio = to_nd_namespace_io(&ndns->dev); + resource_size_t base = nsio->res.start + start_pad; struct vmem_altmap __altmap = { - .base_pfn = init_altmap_base(nsio->res.start), - .reserve = init_altmap_reserve(nsio->res.start), + .base_pfn = init_altmap_base(base), + .reserve = init_altmap_reserve(base), }; - if (!nd_pfn->uuid || !nd_pfn->ndns) - return -ENODEV; - - nd_region = to_nd_region(dev->parent); - rc = nd_pfn_init(nd_pfn); - if (rc) - return rc; - - pfn_sb = nd_pfn->pfn_sb; - offset = le64_to_cpu(pfn_sb->dataoff); + pmem = dev_get_drvdata(dev); + pmem->data_offset = le64_to_cpu(pfn_sb->dataoff); + pmem->pfn_pad = start_pad + end_trunc; nd_pfn->mode = le32_to_cpu(nd_pfn->pfn_sb->mode); if (nd_pfn->mode == PFN_MODE_RAM) { - if (offset < SZ_8K) + if (pmem->data_offset < SZ_8K) return -EINVAL; nd_pfn->npfns = le64_to_cpu(pfn_sb->npfns); altmap = NULL; } else if (nd_pfn->mode == PFN_MODE_PMEM) { - nd_pfn->npfns = (resource_size(&nsio->res) - offset) + nd_pfn->npfns = (pmem->size - pmem->pfn_pad - pmem->data_offset) / PAGE_SIZE; if (le64_to_cpu(nd_pfn->pfn_sb->npfns) > nd_pfn->npfns) dev_info(&nd_pfn->dev, @@ -418,7 +458,7 @@ static int nvdimm_namespace_attach_pfn(struct nd_namespace_common *ndns) le64_to_cpu(nd_pfn->pfn_sb->npfns), nd_pfn->npfns); altmap = & __altmap; - altmap->free = __phys_to_pfn(offset - SZ_8K); + altmap->free = __phys_to_pfn(pmem->data_offset - SZ_8K); altmap->alloc = 0; } else { rc = -ENXIO; @@ -426,10 +466,12 @@ static int nvdimm_namespace_attach_pfn(struct nd_namespace_common *ndns) } /* establish pfn range for lookup, and switch to direct map */ - pmem = dev_get_drvdata(dev); q = pmem->pmem_queue; + memcpy(&res, &nsio->res, sizeof(res)); + res.start += start_pad; + res.end -= end_trunc; devm_memunmap(dev, (void __force *) pmem->virt_addr); - pmem->virt_addr = (void __pmem *) devm_memremap_pages(dev, &nsio->res, + pmem->virt_addr = (void __pmem *) devm_memremap_pages(dev, &res, &q->q_usage_counter, altmap); pmem->pfn_flags |= PFN_MAP; if (IS_ERR(pmem->virt_addr)) { @@ -438,7 +480,6 @@ static int nvdimm_namespace_attach_pfn(struct nd_namespace_common *ndns) } /* attach pmem disk in "pfn-mode" */ - pmem->data_offset = offset; rc = pmem_attach_disk(dev, ndns, pmem); if (rc) goto err; @@ -447,6 +488,22 @@ static int nvdimm_namespace_attach_pfn(struct nd_namespace_common *ndns) err: nvdimm_namespace_detach_pfn(ndns); return rc; + +} + +static int nvdimm_namespace_attach_pfn(struct nd_namespace_common *ndns) +{ + struct nd_pfn *nd_pfn = to_nd_pfn(ndns->claim); + int rc; + + if (!nd_pfn->uuid || !nd_pfn->ndns) + return -ENODEV; + + rc = nd_pfn_init(nd_pfn); + if (rc) + return rc; + /* we need a valid pfn_sb before we can init a vmem_altmap */ + return __nvdimm_namespace_attach_pfn(nd_pfn); } static int nd_pmem_probe(struct device *dev) -- cgit v1.2.3 From f6ed58c70d14572d0272ee129579dbfc97b97f50 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Thu, 3 Mar 2016 09:46:04 -0800 Subject: libnvdimm, pfn: 'resource'-address and 'size' attributes for pfn devices Currenty with a raw mode pmem namespace the physical memory address range for the device can be obtained via /sys/block/pmemX/device/{resource|size}. Add similar attributes for pfn instances that takes the struct page memmap and section padding into account. Reported-by: Haozhong Zhang Signed-off-by: Dan Williams --- drivers/nvdimm/pfn_devs.c | 56 +++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 56 insertions(+) (limited to 'drivers') diff --git a/drivers/nvdimm/pfn_devs.c b/drivers/nvdimm/pfn_devs.c index 75a31a7359fb..254d3bc13f70 100644 --- a/drivers/nvdimm/pfn_devs.c +++ b/drivers/nvdimm/pfn_devs.c @@ -205,11 +205,67 @@ static ssize_t namespace_store(struct device *dev, } static DEVICE_ATTR_RW(namespace); +static ssize_t resource_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nd_pfn *nd_pfn = to_nd_pfn(dev); + ssize_t rc; + + device_lock(dev); + if (dev->driver) { + struct nd_pfn_sb *pfn_sb = nd_pfn->pfn_sb; + u64 offset = __le64_to_cpu(pfn_sb->dataoff); + struct nd_namespace_common *ndns = nd_pfn->ndns; + u32 start_pad = __le32_to_cpu(pfn_sb->start_pad); + struct nd_namespace_io *nsio = to_nd_namespace_io(&ndns->dev); + + rc = sprintf(buf, "%#llx\n", (unsigned long long) nsio->res.start + + start_pad + offset); + } else { + /* no address to convey if the pfn instance is disabled */ + rc = -ENXIO; + } + device_unlock(dev); + + return rc; +} +static DEVICE_ATTR_RO(resource); + +static ssize_t size_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct nd_pfn *nd_pfn = to_nd_pfn(dev); + ssize_t rc; + + device_lock(dev); + if (dev->driver) { + struct nd_pfn_sb *pfn_sb = nd_pfn->pfn_sb; + u64 offset = __le64_to_cpu(pfn_sb->dataoff); + struct nd_namespace_common *ndns = nd_pfn->ndns; + u32 start_pad = __le32_to_cpu(pfn_sb->start_pad); + u32 end_trunc = __le32_to_cpu(pfn_sb->end_trunc); + struct nd_namespace_io *nsio = to_nd_namespace_io(&ndns->dev); + + rc = sprintf(buf, "%llu\n", (unsigned long long) + resource_size(&nsio->res) - start_pad + - end_trunc - offset); + } else { + /* no size to convey if the pfn instance is disabled */ + rc = -ENXIO; + } + device_unlock(dev); + + return rc; +} +static DEVICE_ATTR_RO(size); + static struct attribute *nd_pfn_attributes[] = { &dev_attr_mode.attr, &dev_attr_namespace.attr, &dev_attr_uuid.attr, &dev_attr_align.attr, + &dev_attr_resource.attr, + &dev_attr_size.attr, NULL, }; -- cgit v1.2.3 From 45f68802f2542a6ad1550dab9c07004de6e0df40 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Sun, 6 Mar 2016 08:04:12 -0800 Subject: libnvdimm, pmem: fix ia64 build, use PHYS_PFN drivers/nvdimm/pmem.c: In function 'nvdimm_namespace_attach_pfn': drivers/nvdimm/pmem.c:367:3: error: implicit declaration of function '__phys_to_pfn' [-Werror=implicit-function-declaration] .base_pfn = __phys_to_pfn(nsio->res.start), ia64 does not provide __phys_to_pfn(), just use the PHYS_PFN() alias. Cc: Guenter Roeck Reported-by: kbuild test robot Signed-off-by: Dan Williams --- drivers/nvdimm/pmem.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'drivers') diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 0cb450e1b400..74e2569910d8 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -407,15 +407,15 @@ static int nvdimm_namespace_detach_pfn(struct nd_namespace_common *ndns) */ static unsigned long init_altmap_base(resource_size_t base) { - unsigned long base_pfn = __phys_to_pfn(base); + unsigned long base_pfn = PHYS_PFN(base); return PFN_SECTION_ALIGN_DOWN(base_pfn); } static unsigned long init_altmap_reserve(resource_size_t base) { - unsigned long reserve = __phys_to_pfn(SZ_8K); - unsigned long base_pfn = __phys_to_pfn(base); + unsigned long reserve = PHYS_PFN(SZ_8K); + unsigned long base_pfn = PHYS_PFN(base); reserve += base_pfn - PFN_SECTION_ALIGN_DOWN(base_pfn); return reserve; @@ -458,7 +458,7 @@ static int __nvdimm_namespace_attach_pfn(struct nd_pfn *nd_pfn) le64_to_cpu(nd_pfn->pfn_sb->npfns), nd_pfn->npfns); altmap = & __altmap; - altmap->free = __phys_to_pfn(pmem->data_offset - SZ_8K); + altmap->free = PHYS_PFN(pmem->data_offset - SZ_8K); altmap->alloc = 0; } else { rc = -ENXIO; -- cgit v1.2.3 From af1996ef59dbcb36fe4878df7c717a02eb89d07a Mon Sep 17 00:00:00 2001 From: Toshi Kani Date: Wed, 9 Mar 2016 12:47:06 -0700 Subject: ACPI: Change NFIT driver to insert new resource ACPI 6 defines persistent memory (PMEM) ranges in multiple firmware interfaces, e820, EFI, and ACPI NFIT table. This EFI change, however, leads to hit a bug in the grub bootloader, which treats EFI_PERSISTENT_MEMORY type as regular memory and corrupts stored user data [1]. Therefore, BIOS may set generic reserved type in e820 and EFI to cover PMEM ranges. The kernel can initialize PMEM ranges from ACPI NFIT table alone. This scheme causes a problem in the iomem table, though. On x86, for instance, e820_reserve_resources() initializes top-level entries (iomem_resource.child) from the e820 table at early boot-time. This creates "reserved" entry for a PMEM range, which does not allow region_intersects() to check with PMEM type. Change acpi_nfit_register_region() to call acpi_nfit_insert_resource(), which calls insert_resource() to insert a PMEM entry from NFIT when the iomem table does not have a PMEM entry already. That is, when a PMEM range is marked as reserved type in e820, it inserts "Persistent Memory" entry, which results as follows. + "Persistent Memory" + "reserved" This allows the EINJ driver, which calls region_intersects() to check PMEM ranges, to work continuously even if BIOS sets reserved type (or sets nothing) to PMEM ranges in e820 and EFI. [1]: https://lists.gnu.org/archive/html/grub-devel/2015-11/msg00209.html Signed-off-by: Toshi Kani Cc: Rafael J. Wysocki Cc: Dan Williams Cc: Ingo Molnar Cc: Borislav Petkov Cc: Andrew Morton Signed-off-by: Dan Williams --- drivers/acpi/nfit.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 48 insertions(+) (limited to 'drivers') diff --git a/drivers/acpi/nfit.c b/drivers/acpi/nfit.c index fb53db187854..269de5f75d98 100644 --- a/drivers/acpi/nfit.c +++ b/drivers/acpi/nfit.c @@ -1571,6 +1571,48 @@ static int ars_status_process_records(struct nvdimm_bus *nvdimm_bus, return 0; } +static void acpi_nfit_remove_resource(void *data) +{ + struct resource *res = data; + + remove_resource(res); +} + +static int acpi_nfit_insert_resource(struct acpi_nfit_desc *acpi_desc, + struct nd_region_desc *ndr_desc) +{ + struct resource *res, *nd_res = ndr_desc->res; + int is_pmem, ret; + + /* No operation if the region is already registered as PMEM */ + is_pmem = region_intersects(nd_res->start, resource_size(nd_res), + IORESOURCE_MEM, IORES_DESC_PERSISTENT_MEMORY); + if (is_pmem == REGION_INTERSECTS) + return 0; + + res = devm_kzalloc(acpi_desc->dev, sizeof(*res), GFP_KERNEL); + if (!res) + return -ENOMEM; + + res->name = "Persistent Memory"; + res->start = nd_res->start; + res->end = nd_res->end; + res->flags = IORESOURCE_MEM; + res->desc = IORES_DESC_PERSISTENT_MEMORY; + + ret = insert_resource(&iomem_resource, res); + if (ret) + return ret; + + ret = devm_add_action(acpi_desc->dev, acpi_nfit_remove_resource, res); + if (ret) { + remove_resource(res); + return ret; + } + + return 0; +} + static int acpi_nfit_find_poison(struct acpi_nfit_desc *acpi_desc, struct nd_region_desc *ndr_desc) { @@ -1781,6 +1823,12 @@ static int acpi_nfit_register_region(struct acpi_nfit_desc *acpi_desc, nvdimm_bus = acpi_desc->nvdimm_bus; if (nfit_spa_type(spa) == NFIT_SPA_PM) { + rc = acpi_nfit_insert_resource(acpi_desc, ndr_desc); + if (rc) + dev_warn(acpi_desc->dev, + "failed to insert pmem resource to iomem: %d\n", + rc); + rc = acpi_nfit_find_poison(acpi_desc, ndr_desc); if (rc) { dev_err(acpi_desc->dev, -- cgit v1.2.3 From 55155291b32d24371256adbcc67f9f53cf3f314f Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Wed, 9 Mar 2016 09:21:54 +1100 Subject: pmem: don't allocate unused major device number When alloc_disk(0) or alloc_disk-node(0, XX) is used, the ->major number is completely ignored: all devices are allocated with a major of BLOCK_EXT_MAJOR. So there is no point allocating pmem_major. Signed-off-by: NeilBrown Signed-off-by: Dan Williams --- drivers/nvdimm/pmem.c | 19 +------------------ 1 file changed, 1 insertion(+), 18 deletions(-) (limited to 'drivers') diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 74e2569910d8..ba8d5b6bfad0 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -50,8 +50,6 @@ struct pmem_device { struct badblocks bb; }; -static int pmem_major; - static bool is_bad_pmem(struct badblocks *bb, sector_t sector, unsigned int len) { if (bb->count) { @@ -231,8 +229,6 @@ static int pmem_attach_disk(struct device *dev, return -ENOMEM; } - disk->major = pmem_major; - disk->first_minor = 0; disk->fops = &pmem_fops; disk->private_data = pmem; disk->queue = pmem->pmem_queue; @@ -579,26 +575,13 @@ static struct nd_device_driver nd_pmem_driver = { static int __init pmem_init(void) { - int error; - - pmem_major = register_blkdev(0, "pmem"); - if (pmem_major < 0) - return pmem_major; - - error = nd_driver_register(&nd_pmem_driver); - if (error) { - unregister_blkdev(pmem_major, "pmem"); - return error; - } - - return 0; + return nd_driver_register(&nd_pmem_driver); } module_init(pmem_init); static void pmem_exit(void) { driver_unregister(&nd_pmem_driver.drv); - unregister_blkdev(pmem_major, "pmem"); } module_exit(pmem_exit); -- cgit v1.2.3 From ec56151d382c2140851b4f25203af9016ba84fea Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 10 Mar 2016 08:59:28 +1100 Subject: nvdimm/blk: don't allocate unused major device number When alloc_disk(0) is used ->major is completely ignored, all devices are allocated with a "major" of BLOCK_EXT_MAJOR. So don't allocate nd_blk_major Signed-off-by: NeilBrown Signed-off-by: Dan Williams --- drivers/nvdimm/blk.c | 18 +----------------- 1 file changed, 1 insertion(+), 17 deletions(-) (limited to 'drivers') diff --git a/drivers/nvdimm/blk.c b/drivers/nvdimm/blk.c index 91a336ea8c4f..e9ff9229d942 100644 --- a/drivers/nvdimm/blk.c +++ b/drivers/nvdimm/blk.c @@ -31,8 +31,6 @@ struct nd_blk_device { u32 internal_lbasize; }; -static int nd_blk_major; - static u32 nd_blk_meta_size(struct nd_blk_device *blk_dev) { return blk_dev->nsblk->lbasize - blk_dev->sector_size; @@ -264,7 +262,6 @@ static int nd_blk_attach_disk(struct nd_namespace_common *ndns, } disk->driverfs_dev = &ndns->dev; - disk->major = nd_blk_major; disk->first_minor = 0; disk->fops = &nd_blk_fops; disk->private_data = blk_dev; @@ -358,25 +355,12 @@ static struct nd_device_driver nd_blk_driver = { static int __init nd_blk_init(void) { - int rc; - - rc = register_blkdev(0, "nd_blk"); - if (rc < 0) - return rc; - - nd_blk_major = rc; - rc = nd_driver_register(&nd_blk_driver); - - if (rc < 0) - unregister_blkdev(nd_blk_major, "nd_blk"); - - return rc; + return nd_driver_register(&nd_blk_driver); } static void __exit nd_blk_exit(void) { driver_unregister(&nd_blk_driver.drv); - unregister_blkdev(nd_blk_major, "nd_blk"); } MODULE_AUTHOR("Ross Zwisler "); -- cgit v1.2.3 From ff8e92d5d94b99aab39f439d532cba435947dfc0 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Thu, 10 Mar 2016 08:59:28 +1100 Subject: nvdimm/btt: don't allocate unused major device number alloc_disk(0) does not require or use a ->major number, all devices are allocated with a major of BLOCK_EXT_MAJOR. So don't allocate btt_major. Signed-off-by: NeilBrown Signed-off-by: Dan Williams --- drivers/nvdimm/btt.c | 19 ++----------------- 1 file changed, 2 insertions(+), 17 deletions(-) (limited to 'drivers') diff --git a/drivers/nvdimm/btt.c b/drivers/nvdimm/btt.c index efb2c1ceef98..c32cbb593600 100644 --- a/drivers/nvdimm/btt.c +++ b/drivers/nvdimm/btt.c @@ -31,8 +31,6 @@ enum log_ent_request { LOG_OLD_ENT }; -static int btt_major; - static int arena_read_bytes(struct arena_info *arena, resource_size_t offset, void *buf, size_t n) { @@ -1246,7 +1244,6 @@ static int btt_blk_init(struct btt *btt) nvdimm_namespace_disk_name(ndns, btt->btt_disk->disk_name); btt->btt_disk->driverfs_dev = &btt->nd_btt->dev; - btt->btt_disk->major = btt_major; btt->btt_disk->first_minor = 0; btt->btt_disk->fops = &btt_fops; btt->btt_disk->private_data = btt; @@ -1423,22 +1420,11 @@ EXPORT_SYMBOL(nvdimm_namespace_detach_btt); static int __init nd_btt_init(void) { - int rc; - - btt_major = register_blkdev(0, "btt"); - if (btt_major < 0) - return btt_major; + int rc = 0; debugfs_root = debugfs_create_dir("btt", NULL); - if (IS_ERR_OR_NULL(debugfs_root)) { + if (IS_ERR_OR_NULL(debugfs_root)) rc = -ENXIO; - goto err_debugfs; - } - - return 0; - - err_debugfs: - unregister_blkdev(btt_major, "btt"); return rc; } @@ -1446,7 +1432,6 @@ static int __init nd_btt_init(void) static void __exit nd_btt_exit(void) { debugfs_remove_recursive(debugfs_root); - unregister_blkdev(btt_major, "btt"); } MODULE_ALIAS_ND_DEVICE(ND_DEVICE_BTT); -- cgit v1.2.3