April 2015 - Linux-nvdimm

[Linux-nvdimm] [PATCH v2 00/20] libnd: non-volatile memory device support

by Dan Williams

Changes since v1 [1]: Incorporates feedback received prior to April 24. 1/ Ingo said [2]: "So why on earth is this whole concept and the naming itself ('drivers/block/nd/' stands for 'NFIT Defined', apparently) revolving around a specific 'firmware' mindset and revolving around specific, weirdly named, overly complicated looking firmware interfaces that come with their own new weird glossary??" Indeed, we of course consulted the NFIT specification to determine the shape of the sub-system, but then let its terms and data structures permeate too deep into the implementation. That is fixed now with all NFIT specifics factored out into acpi.c. The NFIT is no longer required reading to review libnd. Only three concepts are needed: i/ PMEM - contiguous memory range where cpu stores are persistent once they are flushed through the memory controller. ii/ BLK - mmio apertures (sliding windows) that can be programmed to access an aperture's-worth of persistent media at a time. iii/ DPA - "dimm-physical-address", address space local to a dimm. A dimm may provide both PMEM-mode and BLK-mode access to a range of DPA. libnd manages allocation of DPA to either PMEM or BLK-namespaces to resolve this aliasing. The v1..v2 diffstat below shows the migration of nfit-specifics to acpi.c and the new state of libnd being nfit-free. "nd" now only refers to "non-volatile devices". Note, reworked documentation will return once the review has settled. Documentation/blockdev/nd.txt | 867 --------------------- MAINTAINERS | 34 +- arch/ia64/kernel/efi.c | 5 +- arch/x86/kernel/e820.c | 11 +- arch/x86/kernel/pmem.c | 2 +- drivers/block/Makefile | 2 +- drivers/block/nd/Kconfig | 135 ++-- drivers/block/nd/Makefile | 32 +- drivers/block/nd/acpi.c | 1506 +++++++++++++++++++++++++++++++------ drivers/block/nd/acpi_nfit.h | 321 ++++++++ drivers/block/nd/blk.c | 27 +- drivers/block/nd/btt.c | 6 +- drivers/block/nd/btt_devs.c | 8 +- drivers/block/nd/bus.c | 337 +++++---- drivers/block/nd/core.c | 574 +------------- drivers/block/nd/dimm.c | 11 - drivers/block/nd/dimm_devs.c | 292 ++----- drivers/block/nd/e820.c | 100 +++ drivers/block/nd/libnd.h | 122 +++ drivers/block/nd/namespace_devs.c | 10 +- drivers/block/nd/nd-private.h | 107 +-- drivers/block/nd/nd.h | 91 +-- drivers/block/nd/nfit.h | 238 ------ drivers/block/nd/pmem.c | 56 +- drivers/block/nd/region.c | 78 +- drivers/block/nd/region_devs.c | 783 +++---------------- drivers/block/nd/test/iomap.c | 86 +-- drivers/block/nd/test/nfit.c | 1115 +++++++++++++++------------ drivers/block/nd/test/nfit_test.h | 15 +- include/uapi/linux/ndctl.h | 130 ++-- 30 files changed, 3166 insertions(+), 3935 deletions(-) delete mode 100644 Documentation/blockdev/nd.txt create mode 100644 drivers/block/nd/acpi_nfit.h create mode 100644 drivers/block/nd/e820.c create mode 100644 drivers/block/nd/libnd.h delete mode 100644 drivers/block/nd/nfit.h [1]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000484.html [2]: https://lists.01.org/pipermail/linux-nvdimm/2015-April/000520.html 2/ Christoph asked the pmem ida conversion to be moved to its own patch (done), and to consider leaving the current pmem.c in drivers/block/. Instead, I converted the e820-type-12 enabling to be the first non-ACPI-NFIT based consumer of libnd. The new nd_e820 driver simply registers e820-type-12 ranges as libnd PMEM regions. Among other things this conversion enables BTT for these ranges. The alternative is to move drivers/block/nd/nd.h internals out to include/linux/ which I think is worse. 3/ Toshi reported that the NFIT parsing fails to handle the case of a PMEM range with a single-dimm (non-aliasing) interleave description. Support for this case was added and is tested by default by the nfit_test.1 configuration. 4/ Toshi reported that we should not be treating a missing _STA property as a "dimm disabled by firmware" case. (fixed). 5/ Christoph noted that ND_ARCH_HAS_IOREMAP_CACHE needs to be moved to arch code. It is gone for now and we'll revisit when adding cached mappings back to the PMEM driver. 6/ Toshi mentioned that the presence of two different nd_bus_probe() functions was confusing. (cleaned up). 7/ Robert asked for s/btt_checksum/nd_btt_checksum/ (done). 8/ Linda asked for nfit_test to honor dynamic cma reservations via the cma= command line (done). The cma requirements have also been reduced to 128M as only the simulated DAX regions need CMA. The rest can use vmalloc(). --- Available here: git://git.kernel.org/pub/scm/linux/kernel/git/djbw/nvdimm nd-v2 --- Dan Williams (18): e820, efi: add ACPI 6.0 persistent memory types libnd, nd_acpi: initial libnd infrastructure and NFIT support nd_acpi, nfit-test: manufactured NFITs for interface development libnd: ndctl class device, and nd bus attributes libnd, nd_acpi: dimm/memory-devices libnd: ndctl.h, the nd ioctl abi libnd, nd_dimm: dimm driver and base libnd device-driver infrastructure libnd, nd_acpi: regions (block-data-window, persistent memory, volatile memory) libnd: support for legacy (non-aliasing) nvdimms pmem: use ida libnd, nd_pmem: add libnd support to the pmem driver libnd, nd_acpi: add interleave-set state-tracking infrastructure libnd: namespace indices: read and validate libnd: pmem label sets and namespace instantiation. libnd: blk labels and namespace instantiation libnd: write pmem label set libnd: write blk label set libnd: infrastructure for btt devices Ross Zwisler (1): libnd, nd_acpi, nd_blk: driver for BLK-mode access persistent memory Vishal Verma (1): nd_btt: atomic sector updates Documentation/blockdev/btt.txt | 273 ++++++ arch/arm64/kernel/efi.c | 1 arch/ia64/kernel/efi.c | 4 arch/x86/boot/compressed/eboot.c | 4 arch/x86/include/uapi/asm/e820.h | 1 arch/x86/kernel/e820.c | 26 + arch/x86/kernel/pmem.c | 2 arch/x86/platform/efi/efi.c | 3 drivers/block/Kconfig | 13 drivers/block/Makefile | 2 drivers/block/nd/Kconfig | 129 +++ drivers/block/nd/Makefile | 41 + drivers/block/nd/acpi.c | 1505 +++++++++++++++++++++++++++++++++ drivers/block/nd/acpi_nfit.h | 321 +++++++ drivers/block/nd/blk.c | 264 ++++++ drivers/block/nd/btt.c | 1423 +++++++++++++++++++++++++++++++ drivers/block/nd/btt.h | 185 ++++ drivers/block/nd/btt_devs.c | 443 ++++++++++ drivers/block/nd/bus.c | 770 +++++++++++++++++ drivers/block/nd/core.c | 471 ++++++++++ drivers/block/nd/dimm.c | 115 +++ drivers/block/nd/dimm_devs.c | 507 +++++++++++ drivers/block/nd/e820.c | 100 ++ drivers/block/nd/label.c | 925 ++++++++++++++++++++ drivers/block/nd/label.h | 143 +++ drivers/block/nd/libnd.h | 122 +++ drivers/block/nd/namespace_devs.c | 1701 +++++++++++++++++++++++++++++++++++++ drivers/block/nd/nd-private.h | 114 ++ drivers/block/nd/nd.h | 261 ++++++ drivers/block/nd/pmem.c | 114 ++ drivers/block/nd/region.c | 159 +++ drivers/block/nd/region_devs.c | 637 ++++++++++++++ drivers/block/nd/test/Makefile | 5 drivers/block/nd/test/iomap.c | 151 +++ drivers/block/nd/test/nfit.c | 1131 +++++++++++++++++++++++++ drivers/block/nd/test/nfit_test.h | 26 + include/linux/efi.h | 3 include/linux/nd.h | 98 ++ include/uapi/linux/Kbuild | 1 include/uapi/linux/ndctl.h | 199 ++++ 40 files changed, 12345 insertions(+), 48 deletions(-) create mode 100644 Documentation/blockdev/btt.txt create mode 100644 drivers/block/nd/Kconfig create mode 100644 drivers/block/nd/Makefile create mode 100644 drivers/block/nd/acpi.c create mode 100644 drivers/block/nd/acpi_nfit.h create mode 100644 drivers/block/nd/blk.c create mode 100644 drivers/block/nd/btt.c create mode 100644 drivers/block/nd/btt.h create mode 100644 drivers/block/nd/btt_devs.c create mode 100644 drivers/block/nd/bus.c create mode 100644 drivers/block/nd/core.c create mode 100644 drivers/block/nd/dimm.c create mode 100644 drivers/block/nd/dimm_devs.c create mode 100644 drivers/block/nd/e820.c create mode 100644 drivers/block/nd/label.c create mode 100644 drivers/block/nd/label.h create mode 100644 drivers/block/nd/libnd.h create mode 100644 drivers/block/nd/namespace_devs.c create mode 100644 drivers/block/nd/nd-private.h create mode 100644 drivers/block/nd/nd.h rename drivers/block/{pmem.c => nd/pmem.c} (68%) create mode 100644 drivers/block/nd/region.c create mode 100644 drivers/block/nd/region_devs.c create mode 100644 drivers/block/nd/test/Makefile create mode 100644 drivers/block/nd/test/iomap.c create mode 100644 drivers/block/nd/test/nfit.c create mode 100644 drivers/block/nd/test/nfit_test.h create mode 100644 include/linux/nd.h create mode 100644 include/uapi/linux/ndctl.h

7 years, 4 months

12
91
0 / 0

[Linux-nvdimm] [GIT PULL] PMEM driver for v4.1

by Ingo Molnar

Linus, Please pull the latest x86-pmem-for-linus git tree from: git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git x86-pmem-for-linus # HEAD: 4c1eaa2344fb26bb5e936fb4d8ee307343ea0089 drivers/block/pmem: Fix 32-bit build warning in pmem_alloc() This is the initial support for the pmem block device driver: persistent non-volatile memory space mapped into the system's physical memory space as large physical memory regions. The driver is based on Intel code, written by Ross Zwisler, with fixes by Boaz Harrosh, integrated with x86 e820 memory resource management and tidied up by Christoph Hellwig. Note that there were two other separate pmem driver submissions to lkml: but apparently all parties (Ross Zwisler, Boaz Harrosh) are reasonably happy with this initial version. This version enables minimal support that enables persistent memory devices out in the wild to work as block devices, identified through a magic (non-standard) e820 flag and auto-discovered if CONFIG_X86_PMEM_LEGACY=y, or added explicitly through manipulating the memory maps via the "memmap=..." boot option with the new, special '!' modifier character. Limitations: this is a regular block device, and since the pmem areas are not struct page backed, they are invisible to the rest of the system (other than the block IO device), so direct IO to/from pmem areas, direct mmap() or XIP is not possible yet. The page cache will also shadow and double buffer pmem contents, etc. Initial support is for x86. Thanks, Ingo ------------------> Christoph Hellwig (1): x86/mm: Add support for the non-standard protected e820 type Ingo Molnar (1): drivers/block/pmem: Fix 32-bit build warning in pmem_alloc() Ross Zwisler (1): drivers/block/pmem: Add a driver for persistent memory Documentation/kernel-parameters.txt | 6 + MAINTAINERS | 6 + arch/x86/Kconfig | 10 ++ arch/x86/include/uapi/asm/e820.h | 10 ++ arch/x86/kernel/Makefile | 1 + arch/x86/kernel/e820.c | 26 +++- arch/x86/kernel/pmem.c | 53 ++++++++ drivers/block/Kconfig | 11 ++ drivers/block/Makefile | 1 + drivers/block/pmem.c | 262 ++++++++++++++++++++++++++++++++++++ 10 files changed, 380 insertions(+), 6 deletions(-) create mode 100644 arch/x86/kernel/pmem.c create mode 100644 drivers/block/pmem.c diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index bfcb1a62a7b4..c87122dd790f 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1965,6 +1965,12 @@ bytes respectively. Such letter suffixes can also be entirely omitted. or memmap=0x10000$0x18690000 + memmap=nn[KMG]!ss[KMG] + [KNL,X86] Mark specific memory as protected. + Region of memory to be used, from ss to ss+nn. + The memory region may be marked as e820 type 12 (0xc) + and is NVDIMM or ADR memory. + memory_corruption_check=0/1 [X86] Some BIOSes seem to corrupt the first 64k of memory when doing things like suspend/resume. diff --git a/MAINTAINERS b/MAINTAINERS index 1de6afa8ee51..4517613dc638 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8071,6 +8071,12 @@ S: Maintained F: Documentation/blockdev/ramdisk.txt F: drivers/block/brd.c +PERSISTENT MEMORY DRIVER +M: Ross Zwisler <ross.zwisler(a)linux.intel.com> +L: linux-nvdimm(a)lists.01.org +S: Supported +F: drivers/block/pmem.c + RANDOM NUMBER DRIVER M: "Theodore Ts'o" <tytso(a)mit.edu> S: Maintained diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index b7d31ca55187..9e3bcd6f4a48 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -1430,6 +1430,16 @@ config ILLEGAL_POINTER_VALUE source "mm/Kconfig" +config X86_PMEM_LEGACY + bool "Support non-standard NVDIMMs and ADR protected memory" + help + Treat memory marked using the non-standard e820 type of 12 as used + by the Intel Sandy Bridge-EP reference BIOS as protected memory. + The kernel will offer these regions to the 'pmem' driver so + they can be used for persistent storage. + + Say Y if unsure. + config HIGHPTE bool "Allocate 3rd-level pagetables from highmem" depends on HIGHMEM diff --git a/arch/x86/include/uapi/asm/e820.h b/arch/x86/include/uapi/asm/e820.h index d993e33f5236..960a8a9dc4ab 100644 --- a/arch/x86/include/uapi/asm/e820.h +++ b/arch/x86/include/uapi/asm/e820.h @@ -33,6 +33,16 @@ #define E820_NVS 4 #define E820_UNUSABLE 5 +/* + * This is a non-standardized way to represent ADR or NVDIMM regions that + * persist over a reboot. The kernel will ignore their special capabilities + * unless the CONFIG_X86_PMEM_LEGACY=y option is set. + * + * ( Note that older platforms also used 6 for the same type of memory, + * but newer versions switched to 12 as 6 was assigned differently. Some + * time they will learn... ) + */ +#define E820_PRAM 12 /* * reserved RAM used by kernel itself diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index cdb1b70ddad0..971f18cd9ca0 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -94,6 +94,7 @@ obj-$(CONFIG_KVM_GUEST) += kvm.o kvmclock.o obj-$(CONFIG_PARAVIRT) += paravirt.o paravirt_patch_$(BITS).o obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= paravirt-spinlocks.o obj-$(CONFIG_PARAVIRT_CLOCK) += pvclock.o +obj-$(CONFIG_X86_PMEM_LEGACY) += pmem.o obj-$(CONFIG_PCSPKR_PLATFORM) += pcspeaker.o diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 46201deee923..11cc7d54ec3f 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -149,6 +149,9 @@ static void __init e820_print_type(u32 type) case E820_UNUSABLE: printk(KERN_CONT "unusable"); break; + case E820_PRAM: + printk(KERN_CONT "persistent (type %u)", type); + break; default: printk(KERN_CONT "type %u", type); break; @@ -343,7 +346,7 @@ int __init sanitize_e820_map(struct e820entry *biosmap, int max_nr_map, * continue building up new bios map based on this * information */ - if (current_type != last_type) { + if (current_type != last_type || current_type == E820_PRAM) { if (last_type != 0) { new_bios[new_bios_entry].size = change_point[chgidx]->addr - last_addr; @@ -688,6 +691,7 @@ void __init e820_mark_nosave_regions(unsigned long limit_pfn) register_nosave_region(pfn, PFN_UP(ei->addr)); pfn = PFN_DOWN(ei->addr + ei->size); + if (ei->type != E820_RAM && ei->type != E820_RESERVED_KERN) register_nosave_region(PFN_UP(ei->addr), pfn); @@ -748,7 +752,7 @@ u64 __init early_reserve_e820(u64 size, u64 align) /* * Find the highest page frame number we have available */ -static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type) +static unsigned long __init e820_end_pfn(unsigned long limit_pfn) { int i; unsigned long last_pfn = 0; @@ -759,7 +763,11 @@ static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type) unsigned long start_pfn; unsigned long end_pfn; - if (ei->type != type) + /* + * Persistent memory is accounted as ram for purposes of + * establishing max_pfn and mem_map. + */ + if (ei->type != E820_RAM && ei->type != E820_PRAM) continue; start_pfn = ei->addr >> PAGE_SHIFT; @@ -784,12 +792,12 @@ static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type) } unsigned long __init e820_end_of_ram_pfn(void) { - return e820_end_pfn(MAX_ARCH_PFN, E820_RAM); + return e820_end_pfn(MAX_ARCH_PFN); } unsigned long __init e820_end_of_low_ram_pfn(void) { - return e820_end_pfn(1UL<<(32 - PAGE_SHIFT), E820_RAM); + return e820_end_pfn(1UL << (32-PAGE_SHIFT)); } static void early_panic(char *msg) @@ -866,6 +874,9 @@ static int __init parse_memmap_one(char *p) } else if (*p == '$') { start_at = memparse(p+1, &p); e820_add_region(start_at, mem_size, E820_RESERVED); + } else if (*p == '!') { + start_at = memparse(p+1, &p); + e820_add_region(start_at, mem_size, E820_PRAM); } else e820_remove_range(mem_size, ULLONG_MAX - mem_size, E820_RAM, 1); @@ -907,6 +918,7 @@ static inline const char *e820_type_to_string(int e820_type) case E820_ACPI: return "ACPI Tables"; case E820_NVS: return "ACPI Non-volatile Storage"; case E820_UNUSABLE: return "Unusable memory"; + case E820_PRAM: return "Persistent RAM"; default: return "reserved"; } } @@ -940,7 +952,9 @@ void __init e820_reserve_resources(void) * pci device BAR resource and insert them later in * pcibios_resource_survey() */ - if (e820.map[i].type != E820_RESERVED || res->start < (1ULL<<20)) { + if (((e820.map[i].type != E820_RESERVED) && + (e820.map[i].type != E820_PRAM)) || + res->start < (1ULL<<20)) { res->flags |= IORESOURCE_BUSY; insert_resource(&iomem_resource, res); } diff --git a/arch/x86/kernel/pmem.c b/arch/x86/kernel/pmem.c new file mode 100644 index 000000000000..3420c874ddc5 --- /dev/null +++ b/arch/x86/kernel/pmem.c @@ -0,0 +1,53 @@ +/* + * Copyright (c) 2015, Christoph Hellwig. + */ +#include <linux/memblock.h> +#include <linux/platform_device.h> +#include <linux/slab.h> +#include <asm/e820.h> +#include <asm/page_types.h> +#include <asm/setup.h> + +static __init void register_pmem_device(struct resource *res) +{ + struct platform_device *pdev; + int error; + + pdev = platform_device_alloc("pmem", PLATFORM_DEVID_AUTO); + if (!pdev) + return; + + error = platform_device_add_resources(pdev, res, 1); + if (error) + goto out_put_pdev; + + error = platform_device_add(pdev); + if (error) + goto out_put_pdev; + return; + +out_put_pdev: + dev_warn(&pdev->dev, "failed to add 'pmem' (persistent memory) device!\n"); + platform_device_put(pdev); +} + +static __init int register_pmem_devices(void) +{ + int i; + + for (i = 0; i < e820.nr_map; i++) { + struct e820entry *ei = &e820.map[i]; + + if (ei->type == E820_PRAM) { + struct resource res = { + .flags = IORESOURCE_MEM, + .start = ei->addr, + .end = ei->addr + ei->size - 1, + }; + register_pmem_device(&res); + } + } + + return 0; +} +device_initcall(register_pmem_devices); diff --git a/drivers/block/Kconfig b/drivers/block/Kconfig index 1b8094d4d7af..eb1fed5bd516 100644 --- a/drivers/block/Kconfig +++ b/drivers/block/Kconfig @@ -404,6 +404,17 @@ config BLK_DEV_RAM_DAX and will prevent RAM block device backing store memory from being allocated from highmem (only a problem for highmem systems). +config BLK_DEV_PMEM + tristate "Persistent memory block device support" + help + Saying Y here will allow you to use a contiguous range of reserved + memory as one or more persistent block devices. + + To compile this driver as a module, choose M here: the module will be + called 'pmem'. + + If unsure, say N. + config CDROM_PKTCDVD tristate "Packet writing on CD/DVD media" depends on !UML diff --git a/drivers/block/Makefile b/drivers/block/Makefile index 02b688d1438d..9cc6c18a1c7e 100644 --- a/drivers/block/Makefile +++ b/drivers/block/Makefile @@ -14,6 +14,7 @@ obj-$(CONFIG_PS3_VRAM) += ps3vram.o obj-$(CONFIG_ATARI_FLOPPY) += ataflop.o obj-$(CONFIG_AMIGA_Z2RAM) += z2ram.o obj-$(CONFIG_BLK_DEV_RAM) += brd.o +obj-$(CONFIG_BLK_DEV_PMEM) += pmem.o obj-$(CONFIG_BLK_DEV_LOOP) += loop.o obj-$(CONFIG_BLK_CPQ_DA) += cpqarray.o obj-$(CONFIG_BLK_CPQ_CISS_DA) += cciss.o diff --git a/drivers/block/pmem.c b/drivers/block/pmem.c new file mode 100644 index 000000000000..eabf4a8d0085 --- /dev/null +++ b/drivers/block/pmem.c @@ -0,0 +1,262 @@ +/* + * Persistent Memory Driver + * + * Copyright (c) 2014, Intel Corporation. + * Copyright (c) 2015, Christoph Hellwig <hch(a)lst.de>. + * Copyright (c) 2015, Boaz Harrosh <boaz(a)plexistor.com>. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + */ + +#include <asm/cacheflush.h> +#include <linux/blkdev.h> +#include <linux/hdreg.h> +#include <linux/init.h> +#include <linux/platform_device.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/slab.h> + +#define PMEM_MINORS 16 + +struct pmem_device { + struct request_queue *pmem_queue; + struct gendisk *pmem_disk; + + /* One contiguous memory region per device */ + phys_addr_t phys_addr; + void *virt_addr; + size_t size; +}; + +static int pmem_major; +static atomic_t pmem_index; + +static void pmem_do_bvec(struct pmem_device *pmem, struct page *page, + unsigned int len, unsigned int off, int rw, + sector_t sector) +{ + void *mem = kmap_atomic(page); + size_t pmem_off = sector << 9; + + if (rw == READ) { + memcpy(mem + off, pmem->virt_addr + pmem_off, len); + flush_dcache_page(page); + } else { + flush_dcache_page(page); + memcpy(pmem->virt_addr + pmem_off, mem + off, len); + } + + kunmap_atomic(mem); +} + +static void pmem_make_request(struct request_queue *q, struct bio *bio) +{ + struct block_device *bdev = bio->bi_bdev; + struct pmem_device *pmem = bdev->bd_disk->private_data; + int rw; + struct bio_vec bvec; + sector_t sector; + struct bvec_iter iter; + int err = 0; + + if (bio_end_sector(bio) > get_capacity(bdev->bd_disk)) { + err = -EIO; + goto out; + } + + BUG_ON(bio->bi_rw & REQ_DISCARD); + + rw = bio_data_dir(bio); + sector = bio->bi_iter.bi_sector; + bio_for_each_segment(bvec, bio, iter) { + pmem_do_bvec(pmem, bvec.bv_page, bvec.bv_len, bvec.bv_offset, + rw, sector); + sector += bvec.bv_len >> 9; + } + +out: + bio_endio(bio, err); +} + +static int pmem_rw_page(struct block_device *bdev, sector_t sector, + struct page *page, int rw) +{ + struct pmem_device *pmem = bdev->bd_disk->private_data; + + pmem_do_bvec(pmem, page, PAGE_CACHE_SIZE, 0, rw, sector); + page_endio(page, rw & WRITE, 0); + + return 0; +} + +static long pmem_direct_access(struct block_device *bdev, sector_t sector, + void **kaddr, unsigned long *pfn, long size) +{ + struct pmem_device *pmem = bdev->bd_disk->private_data; + size_t offset = sector << 9; + + if (!pmem) + return -ENODEV; + + *kaddr = pmem->virt_addr + offset; + *pfn = (pmem->phys_addr + offset) >> PAGE_SHIFT; + + return pmem->size - offset; +} + +static const struct block_device_operations pmem_fops = { + .owner = THIS_MODULE, + .rw_page = pmem_rw_page, + .direct_access = pmem_direct_access, +}; + +static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res) +{ + struct pmem_device *pmem; + struct gendisk *disk; + int idx, err; + + err = -ENOMEM; + pmem = kzalloc(sizeof(*pmem), GFP_KERNEL); + if (!pmem) + goto out; + + pmem->phys_addr = res->start; + pmem->size = resource_size(res); + + err = -EINVAL; + if (!request_mem_region(pmem->phys_addr, pmem->size, "pmem")) { + dev_warn(dev, "could not reserve region [0x%pa:0x%zx]\n", &pmem->phys_addr, pmem->size); + goto out_free_dev; + } + + /* + * Map the memory as non-cachable, as we can't write back the contents + * of the CPU caches in case of a crash. + */ + err = -ENOMEM; + pmem->virt_addr = ioremap_nocache(pmem->phys_addr, pmem->size); + if (!pmem->virt_addr) + goto out_release_region; + + pmem->pmem_queue = blk_alloc_queue(GFP_KERNEL); + if (!pmem->pmem_queue) + goto out_unmap; + + blk_queue_make_request(pmem->pmem_queue, pmem_make_request); + blk_queue_max_hw_sectors(pmem->pmem_queue, 1024); + blk_queue_bounce_limit(pmem->pmem_queue, BLK_BOUNCE_ANY); + + disk = alloc_disk(PMEM_MINORS); + if (!disk) + goto out_free_queue; + + idx = atomic_inc_return(&pmem_index) - 1; + + disk->major = pmem_major; + disk->first_minor = PMEM_MINORS * idx; + disk->fops = &pmem_fops; + disk->private_data = pmem; + disk->queue = pmem->pmem_queue; + disk->flags = GENHD_FL_EXT_DEVT; + sprintf(disk->disk_name, "pmem%d", idx); + disk->driverfs_dev = dev; + set_capacity(disk, pmem->size >> 9); + pmem->pmem_disk = disk; + + add_disk(disk); + + return pmem; + +out_free_queue: + blk_cleanup_queue(pmem->pmem_queue); +out_unmap: + iounmap(pmem->virt_addr); +out_release_region: + release_mem_region(pmem->phys_addr, pmem->size); +out_free_dev: + kfree(pmem); +out: + return ERR_PTR(err); +} + +static void pmem_free(struct pmem_device *pmem) +{ + del_gendisk(pmem->pmem_disk); + put_disk(pmem->pmem_disk); + blk_cleanup_queue(pmem->pmem_queue); + iounmap(pmem->virt_addr); + release_mem_region(pmem->phys_addr, pmem->size); + kfree(pmem); +} + +static int pmem_probe(struct platform_device *pdev) +{ + struct pmem_device *pmem; + struct resource *res; + + if (WARN_ON(pdev->num_resources > 1)) + return -ENXIO; + + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!res) + return -ENXIO; + + pmem = pmem_alloc(&pdev->dev, res); + if (IS_ERR(pmem)) + return PTR_ERR(pmem); + + platform_set_drvdata(pdev, pmem); + + return 0; +} + +static int pmem_remove(struct platform_device *pdev) +{ + struct pmem_device *pmem = platform_get_drvdata(pdev); + + pmem_free(pmem); + return 0; +} + +static struct platform_driver pmem_driver = { + .probe = pmem_probe, + .remove = pmem_remove, + .driver = { + .owner = THIS_MODULE, + .name = "pmem", + }, +}; + +static int __init pmem_init(void) +{ + int error; + + pmem_major = register_blkdev(0, "pmem"); + if (pmem_major < 0) + return pmem_major; + + error = platform_driver_register(&pmem_driver); + if (error) + unregister_blkdev(pmem_major, "pmem"); + return error; +} +module_init(pmem_init); + +static void pmem_exit(void) +{ + platform_driver_unregister(&pmem_driver); + unregister_blkdev(pmem_major, "pmem"); +} +module_exit(pmem_exit); + +MODULE_AUTHOR("Ross Zwisler <ross.zwisler(a)linux.intel.com>"); +MODULE_LICENSE("GPL v2");

7 years, 4 months

8
29
0 / 0

[Linux-nvdimm] another pmem variant

by Christoph Hellwig

Here is another version of the same trivial pmem driver, because two obviously aren't enough. The first patch is the same pmem driver that Ross posted a short time ago, just modified to use platform_devices to find the persistant memory region instead of hardconding it in the Kconfig. This allows to keep pmem.c separate from any discovery mechanism, but still allow auto-discovery. The other two patches are a heavily rewritten version of the code that Intel gave to various storage vendors to discover the type 12 (and earlier type 6) nvdimms, which I massaged into a form that is hopefully suitable for mainline. Note that pmem.c really is the minimal version as I think we need something included ASAP. We'll eventually need to be able to do other I/O from and to it, and as most people know everyone has their own preferre method to do it, which I'd like to discuss once we have the basic driver in. This has been tested both with a real NVDIMM on a system with a type 12 capable bios, as well as with "fake persistent" memory using the memmap= option.

7 years, 4 months

8
32
0 / 0

[Linux-nvdimm] [PATCH] libnd: stable(ish) block device names

by Dan Williams

>From Toshi: We have seen a case that the region#->pmem# binding becomes inconsistent across a reboot when there are 8 NVDIMM cards (reported by Robert Elliott). This leads user to access a wrong device. While it is a bug for userspace to depend on stable device names from boot-to-boot, we should try not to give random results in the case where the configuration has not changed. Now that PMEM only ever attaches to 'region' devices emitted by libnd, we can simply carry the region-id all the way down to the pmem device name. Since BLK-regions allow multiple namespaces per-region we need to carry the <region-id>.<namespace-index> in the block device name if we want stable(ish) device names. Cc: Christoph Hellwig <hch(a)lst.de> Cc: Ross Zwisler <ross.zwisler(a)linux.intel.com> Reported-by: Toshi Kani <toshi.kani(a)hp.com> Reported-by: Robert Elliott <Elliott(a)hp.com> Signed-off-by: Dan Williams <dan.j.williams(a)intel.com> --- Will fold this into the v3 posting. v3 will be a rebase on top of the upcoming ACPICA NFIT enabling. drivers/block/nd/blk.c | 14 ++------------ drivers/block/nd/pmem.c | 21 +++++---------------- 2 files changed, 7 insertions(+), 28 deletions(-) diff --git a/drivers/block/nd/blk.c b/drivers/block/nd/blk.c index 8536ee8b2009..9a2b93bdeccc 100644 --- a/drivers/block/nd/blk.c +++ b/drivers/block/nd/blk.c @@ -28,11 +28,9 @@ struct nd_blk_device { struct nd_blk_region *ndbr; struct nd_io ndio; size_t disk_size; - int id; }; static int nd_blk_major; -static DEFINE_IDA(nd_blk_ida); static resource_size_t to_dev_offset(struct nd_namespace_blk *nsblk, resource_size_t ns_offset, unsigned int len) @@ -139,6 +137,7 @@ static const struct block_device_operations nd_blk_fops = { static int nd_blk_probe(struct device *dev) { struct nd_namespace_blk *nsblk = to_nd_namespace_blk(dev); + struct nd_region *nd_region = to_nd_region(dev->parent); struct nd_blk_device *blk_dev; resource_size_t disk_size; struct gendisk *disk; @@ -152,12 +151,6 @@ static int nd_blk_probe(struct device *dev) if (!blk_dev) return -ENOMEM; - blk_dev->id = ida_simple_get(&nd_blk_ida, 0, 0, GFP_KERNEL); - if (blk_dev->id < 0) { - err = blk_dev->id; - goto err_ida; - } - blk_dev->disk_size = disk_size; blk_dev->queue = blk_alloc_queue(GFP_KERNEL); @@ -186,7 +179,7 @@ static int nd_blk_probe(struct device *dev) disk->private_data = blk_dev; disk->queue = blk_dev->queue; disk->flags = GENHD_FL_EXT_DEVT; - sprintf(disk->disk_name, "ndblk%d", blk_dev->id); + sprintf(disk->disk_name, "ndblk%d.%d", nd_region->id, nsblk->id); set_capacity(disk, disk_size >> SECTOR_SHIFT); nd_bus_lock(dev); @@ -202,8 +195,6 @@ static int nd_blk_probe(struct device *dev) err_alloc_disk: blk_cleanup_queue(blk_dev->queue); err_alloc_queue: - ida_simple_remove(&nd_blk_ida, blk_dev->id); - err_ida: kfree(blk_dev); return err; } @@ -219,7 +210,6 @@ static int nd_blk_remove(struct device *dev) del_gendisk(blk_dev->disk); put_disk(blk_dev->disk); blk_cleanup_queue(blk_dev->queue); - ida_simple_remove(&nd_blk_ida, blk_dev->id); kfree(blk_dev); return 0; diff --git a/drivers/block/nd/pmem.c b/drivers/block/nd/pmem.c index 900dad61a6b9..5814c8aed481 100644 --- a/drivers/block/nd/pmem.c +++ b/drivers/block/nd/pmem.c @@ -37,11 +37,9 @@ struct pmem_device { phys_addr_t phys_addr; void *virt_addr; size_t size; - int id; }; static int pmem_major; -static DEFINE_IDA(pmem_ida); static void pmem_do_bvec(struct pmem_device *pmem, struct page *page, unsigned int len, unsigned int off, int rw, @@ -142,7 +140,7 @@ static const struct block_device_operations pmem_fops = { .direct_access = pmem_direct_access, }; -static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res) +static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res, int id) { struct pmem_device *pmem; struct gendisk *disk; @@ -153,19 +151,13 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res) if (!pmem) goto out; - pmem->id = ida_simple_get(&pmem_ida, 0, 0, GFP_KERNEL); - if (pmem->id < 0) { - err = pmem->id; - goto out_free_dev; - } - pmem->phys_addr = res->start; pmem->size = resource_size(res); err = -EINVAL; if (!request_mem_region(pmem->phys_addr, pmem->size, "pmem")) { dev_warn(dev, "could not reserve region [0x%pa:0x%zx]\n", &pmem->phys_addr, pmem->size); - goto out_free_ida; + goto out_free_dev; } /* @@ -190,12 +182,12 @@ static struct pmem_device *pmem_alloc(struct device *dev, struct resource *res) goto out_free_queue; disk->major = pmem_major; - disk->first_minor = PMEM_MINORS * pmem->id; + disk->first_minor = PMEM_MINORS * id; disk->fops = &pmem_fops; disk->private_data = pmem; disk->queue = pmem->pmem_queue; disk->flags = GENHD_FL_EXT_DEVT; - sprintf(disk->disk_name, "pmem%d", pmem->id); + sprintf(disk->disk_name, "pmem%d", id); disk->driverfs_dev = dev; set_capacity(disk, pmem->size >> 9); pmem->pmem_disk = disk; @@ -208,8 +200,6 @@ out_unmap: iounmap(pmem->virt_addr); out_release_region: release_mem_region(pmem->phys_addr, pmem->size); -out_free_ida: - ida_simple_remove(&pmem_ida, pmem->id); out_free_dev: kfree(pmem); out: @@ -223,7 +213,6 @@ static void pmem_free(struct pmem_device *pmem) blk_cleanup_queue(pmem->pmem_queue); iounmap(pmem->virt_addr); release_mem_region(pmem->phys_addr, pmem->size); - ida_simple_remove(&pmem_ida, pmem->id); kfree(pmem); } @@ -250,7 +239,7 @@ static int nd_pmem_probe(struct device *dev) } } - pmem = pmem_alloc(dev, &nsio->res); + pmem = pmem_alloc(dev, &nsio->res, nd_region->id); if (IS_ERR(pmem)) return PTR_ERR(pmem);

7 years, 5 months

1
1
0 / 0

[Linux-nvdimm] [ndctl PATCH] ndctl: updates for libnd

by Dan Williams

The new 'libnd' organization of the kernel enabling means that nfit specifics are pushed out of the generic core. Update libndctl to find nfit-specific attributes in an 'nfit' attribute sub-directory of an nd device. Cc: Toshi Kani <toshi.kani(a)hp.com> Cc: Robert Elliott <Elliott(a)hp.com> Cc: Linda Knippers <linda.knippers(a)hp.com> Signed-off-by: Dan Williams <dan.j.williams(a)intel.com> --- lib/libndctl.c | 248 +++++++++++++++++++++++++++----------------------- lib/libndctl.sym | 1 lib/ndctl/libndctl.h | 1 lib/test-dpa-alloc.c | 2 lib/test-libndctl.c | 46 +++++---- ndctl.h | 136 ++++++++++++++------------- 6 files changed, 230 insertions(+), 204 deletions(-) diff --git a/lib/libndctl.c b/lib/libndctl.c index 73e241fcee80..3274c094b7f6 100644 --- a/lib/libndctl.c +++ b/lib/libndctl.c @@ -98,6 +98,7 @@ struct ndctl_bus { int dimms_init; int btts_init; int regions_init; + int has_nfit; char *bus_path; char *bus_buf; size_t buf_len; @@ -326,10 +327,10 @@ struct ndctl_cmd { } iter; struct ndctl_cmd *source; union { - struct nfit_cmd_get_config_size get_size[0]; - struct nfit_cmd_get_config_data_hdr get_data[0]; - struct nfit_cmd_set_config_hdr set_data[0]; - struct nfit_cmd_vendor_hdr vendor[0]; + struct nd_cmd_get_config_size get_size[0]; + struct nd_cmd_get_config_data_hdr get_data[0]; + struct nd_cmd_set_config_hdr set_data[0]; + struct nd_cmd_vendor_hdr vendor[0]; char cmd_buf[0]; }; }; @@ -742,11 +743,11 @@ static int to_dsm_index(const char *name, int dimm) int i, end_cmd; if (dimm) { - end_cmd = NFIT_CMD_VENDOR; - cmd_name_fn = nfit_dimm_cmd_name; + end_cmd = ND_CMD_VENDOR; + cmd_name_fn = nd_dimm_cmd_name; } else { - end_cmd = NFIT_CMD_ARS_QUERY; - cmd_name_fn = nfit_bus_cmd_name; + end_cmd = ND_CMD_ARS_QUERY; + cmd_name_fn = nd_bus_cmd_name; } for (i = 1; i <= end_cmd; i++) { @@ -784,7 +785,7 @@ static int add_bus(void *parent, int id, const char *ctl_base) struct ndctl_bus *bus; char buf[SYSFS_ATTR_SIZE]; struct ndctl_ctx *ctx = parent; - char *path = calloc(1, strlen(ctl_base) + 20); + char *path = calloc(1, strlen(ctl_base) + 100); if (!path) return -ENOMEM; @@ -809,10 +810,14 @@ static int add_bus(void *parent, int id, const char *ctl_base) goto err_read; bus->dsm_mask = parse_commands(buf, 0); - sprintf(path, "%s/device/revision", ctl_base); - if (sysfs_read_attr(ctx, path, buf) < 0) - goto err_read; - bus->revision = strtoul(buf, NULL, 0); + sprintf(path, "%s/device/nfit/revision", ctl_base); + if (sysfs_read_attr(ctx, path, buf) < 0) { + bus->has_nfit = 0; + bus->revision = -1; + } else { + bus->has_nfit = 1; + bus->revision = strtoul(buf, NULL, 0); + } sprintf(path, "%s/device/provider", ctl_base); if (sysfs_read_attr(ctx, path, buf) < 0) @@ -890,6 +895,11 @@ NDCTL_EXPORT struct ndctl_bus *ndctl_bus_get_next(struct ndctl_bus *bus) return list_next(&ctx->busses, bus, list); } +NDCTL_EXPORT int ndctl_bus_has_nfit(struct ndctl_bus *bus) +{ + return bus->has_nfit; +} + /** * ndctl_bus_get_major - nd bus character device major number * @bus: ndctl_bus instance returned from ndctl_bus_get_{first|next} @@ -950,7 +960,7 @@ NDCTL_EXPORT struct ndctl_btt *ndctl_bus_get_btt_seed(struct ndctl_bus *bus) NDCTL_EXPORT const char *ndctl_bus_get_cmd_name(struct ndctl_bus *bus, int cmd) { - return nfit_bus_cmd_name(cmd); + return nd_bus_cmd_name(cmd); } NDCTL_EXPORT int ndctl_bus_is_cmd_supported(struct ndctl_bus *bus, @@ -1020,7 +1030,7 @@ static int add_dimm(void *parent, int id, const char *dimm_base) char buf[SYSFS_ATTR_SIZE]; struct ndctl_bus *bus = parent; struct ndctl_ctx *ctx = bus->ctx; - char *path = calloc(1, strlen(dimm_base) + 20); + char *path = calloc(1, strlen(dimm_base) + 100); if (!path) return -ENOMEM; @@ -1039,60 +1049,62 @@ static int add_dimm(void *parent, int id, const char *dimm_base) if (sscanf(buf, "%d:%d", &dimm->major, &dimm->minor) != 2) goto err_read; - sprintf(path, "%s/handle", dimm_base); + sprintf(path, "%s/commands", dimm_base); if (sysfs_read_attr(ctx, path, buf) < 0) goto err_read; - dimm->handle = strtoul(buf, NULL, 0); + dimm->dsm_mask = parse_commands(buf, 1); - sprintf(path, "%s/phys_id", dimm_base); - if (sysfs_read_attr(ctx, path, buf) < 0) - goto err_read; - dimm->phys_id = strtoul(buf, NULL, 0); + dimm->dimm_buf = calloc(1, strlen(dimm_base) + 50); + if (!dimm->dimm_buf) + goto err_read; + dimm->buf_len = strlen(dimm_base) + 50; dimm->dimm_path = strdup(dimm_base); if (!dimm->dimm_path) goto err_read; - dimm->dimm_buf = calloc(1, strlen(dimm_base) + 50); - if (!dimm->dimm_buf) - goto err_read; - dimm->buf_len = strlen(dimm_base) + 50; + dimm->handle = -1; + dimm->phys_id = -1; + dimm->vendor_id = -1; + dimm->serial = -1; + dimm->device_id = -1; + dimm->revision_id = -1; + dimm->format_id = -1; + if (!ndctl_bus_has_nfit(bus)) + goto out; - sprintf(path, "%s/vendor", dimm_base); + sprintf(path, "%s/nfit/handle", dimm_base); if (sysfs_read_attr(ctx, path, buf) < 0) - dimm->vendor_id = -1; - else - dimm->vendor_id = strtoul(buf, NULL, 0); + goto err_read; + dimm->handle = strtoul(buf, NULL, 0); - sprintf(path, "%s/commands", dimm_base); + sprintf(path, "%s/nfit/phys_id", dimm_base); if (sysfs_read_attr(ctx, path, buf) < 0) goto err_read; - dimm->dsm_mask = parse_commands(buf, 1); + dimm->phys_id = strtoul(buf, NULL, 0); - sprintf(path, "%s/serial", dimm_base); + sprintf(path, "%s/nfit/vendor", dimm_base); if (sysfs_read_attr(ctx, path, buf) < 0) - dimm->serial = -1; + dimm->vendor_id = -1; else + dimm->vendor_id = strtoul(buf, NULL, 0); + + sprintf(path, "%s/nfit/serial", dimm_base); + if (sysfs_read_attr(ctx, path, buf) == 0) dimm->serial = strtoul(buf, NULL, 0); - sprintf(path, "%s/device", dimm_base); - if (sysfs_read_attr(ctx, path, buf) < 0) - dimm->device_id = -1; - else + sprintf(path, "%s/nfit/device", dimm_base); + if (sysfs_read_attr(ctx, path, buf) == 0) dimm->device_id = strtoul(buf, NULL, 0); - sprintf(path, "%s/revision", dimm_base); - if (sysfs_read_attr(ctx, path, buf) < 0) - dimm->revision_id = -1; - else + sprintf(path, "%s/nfit/rev_id", dimm_base); + if (sysfs_read_attr(ctx, path, buf) == 0) dimm->revision_id = strtoul(buf, NULL, 0); - sprintf(path, "%s/format", dimm_base); - if (sysfs_read_attr(ctx, path, buf) < 0) - dimm->format_id = -1; - else + sprintf(path, "%s/nfit/format", dimm_base); + if (sysfs_read_attr(ctx, path, buf) == 0) dimm->format_id = strtoul(buf, NULL, 0); - + out: list_add(&bus->dimms, &dimm->list); free(path); @@ -1185,7 +1197,7 @@ NDCTL_EXPORT const char *ndctl_dimm_get_devname(struct ndctl_dimm *dimm) NDCTL_EXPORT const char *ndctl_dimm_get_cmd_name(struct ndctl_dimm *dimm, int cmd) { - return nfit_dimm_cmd_name(cmd); + return nd_dimm_cmd_name(cmd); } NDCTL_EXPORT int ndctl_dimm_is_cmd_supported(struct ndctl_dimm *dimm, @@ -1241,6 +1253,16 @@ NDCTL_EXPORT struct ndctl_dimm *ndctl_dimm_get_by_handle(struct ndctl_bus *bus, return NULL; } +static struct ndctl_dimm *ndctl_dimm_get_by_id(struct ndctl_bus *bus, unsigned int id) +{ + struct ndctl_dimm *dimm; + + ndctl_dimm_foreach(bus, dimm) + if (ndctl_dimm_get_id(dimm) == id) + return dimm; + return NULL; +} + static int add_region(void *parent, int id, const char *region_base) { int rc = -ENOMEM; @@ -1248,7 +1270,7 @@ static int add_region(void *parent, int id, const char *region_base) struct ndctl_region *region; struct ndctl_bus *bus = parent; struct ndctl_ctx *ctx = bus->ctx; - char *path = calloc(1, strlen(region_base) + 20); + char *path = calloc(1, strlen(region_base) + 100); if (!path) return -ENOMEM; @@ -1278,8 +1300,8 @@ static int add_region(void *parent, int id, const char *region_base) goto err_read; region->nstype = strtoul(buf, NULL, 0); - sprintf(path, "%s/spa_index", region_base); - if (region->nstype != ND_DEVICE_NAMESPACE_BLOCK) { + sprintf(path, "%s/nfit/spa_index", region_base); + if (ndctl_bus_has_nfit(bus)) { if (sysfs_read_attr(ctx, path, buf) < 0) goto err_read; region->spa_index = strtoul(buf, NULL, 0); @@ -1379,7 +1401,7 @@ NDCTL_EXPORT unsigned long long ndctl_region_get_available_size( switch (nstype) { case ND_DEVICE_NAMESPACE_PMEM: - case ND_DEVICE_NAMESPACE_BLOCK: + case ND_DEVICE_NAMESPACE_BLK: break; default: return 0; @@ -1414,7 +1436,7 @@ NDCTL_EXPORT unsigned int ndctl_region_get_type(struct ndctl_region *region) case ND_DEVICE_NAMESPACE_PMEM: return ND_DEVICE_REGION_PMEM; default: - return ND_DEVICE_REGION_BLOCK; + return ND_DEVICE_REGION_BLK; } } @@ -1446,13 +1468,13 @@ NDCTL_EXPORT struct ndctl_namespace *ndctl_region_get_namespace_seed( static const char *ndctl_device_type_name(int type) { switch (type) { - case ND_DEVICE_DIMM: return "dimm"; - case ND_DEVICE_REGION_PMEM: return "pmem"; - case ND_DEVICE_REGION_BLOCK: return "block"; - case ND_DEVICE_NAMESPACE_IO: return "namespace_io"; - case ND_DEVICE_NAMESPACE_PMEM: return "namespace_pmem"; - case ND_DEVICE_NAMESPACE_BLOCK: return "namespace_block"; - default: return "unknown"; + case ND_DEVICE_DIMM: return "dimm"; + case ND_DEVICE_REGION_PMEM: return "pmem"; + case ND_DEVICE_REGION_BLK: return "blk"; + case ND_DEVICE_NAMESPACE_IO: return "namespace_io"; + case ND_DEVICE_NAMESPACE_PMEM: return "namespace_pmem"; + case ND_DEVICE_NAMESPACE_BLK: return "namespace_blk"; + default: return "unknown"; } } @@ -1501,10 +1523,10 @@ NDCTL_EXPORT struct ndctl_dimm *ndctl_region_get_next_dimm(struct ndctl_region * return NULL; } -static struct nfit_cmd_vendor_tail *to_vendor_tail(struct ndctl_cmd *cmd) +static struct nd_cmd_vendor_tail *to_vendor_tail(struct ndctl_cmd *cmd) { - struct nfit_cmd_vendor_tail *tail = (struct nfit_cmd_vendor_tail *) - (cmd->cmd_buf + sizeof(struct nfit_cmd_vendor_hdr) + struct nd_cmd_vendor_tail *tail = (struct nd_cmd_vendor_tail *) + (cmd->cmd_buf + sizeof(struct nd_cmd_vendor_hdr) + cmd->vendor->in_length); return tail; } @@ -1518,13 +1540,13 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_vendor_specific( struct ndctl_cmd *cmd; size_t size; - if (!ndctl_dimm_is_cmd_supported(dimm, NFIT_CMD_VENDOR)) { + if (!ndctl_dimm_is_cmd_supported(dimm, ND_CMD_VENDOR)) { dbg(ctx, "unsupported cmd\n"); return NULL; } - size = sizeof(*cmd) + sizeof(struct nfit_cmd_vendor_hdr) - + sizeof(struct nfit_cmd_vendor_tail) + input_size + size = sizeof(*cmd) + sizeof(struct nd_cmd_vendor_hdr) + + sizeof(struct nd_cmd_vendor_tail) + input_size + output_size; cmd = calloc(1, size); @@ -1533,7 +1555,7 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_vendor_specific( cmd->dimm = dimm; ndctl_cmd_ref(cmd); - cmd->type = NFIT_CMD_VENDOR; + cmd->type = ND_CMD_VENDOR; cmd->size = size; cmd->status = 1; cmd->vendor->opcode = opcode; @@ -1547,7 +1569,7 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_vendor_specific( NDCTL_EXPORT ssize_t ndctl_cmd_vendor_set_input(struct ndctl_cmd *cmd, void *buf, unsigned int len) { - if (cmd->type != NFIT_CMD_VENDOR) + if (cmd->type != ND_CMD_VENDOR) return -EINVAL; len = min(len, cmd->vendor->in_length); memcpy(cmd->vendor->in_buf, buf, len); @@ -1556,7 +1578,7 @@ NDCTL_EXPORT ssize_t ndctl_cmd_vendor_set_input(struct ndctl_cmd *cmd, NDCTL_EXPORT ssize_t ndctl_cmd_vendor_get_output_size(struct ndctl_cmd *cmd) { - if (cmd->type != NFIT_CMD_VENDOR || cmd->status > 0) + if (cmd->type != ND_CMD_VENDOR || cmd->status > 0) return -EINVAL; if (cmd->status < 0) return cmd->status; @@ -1584,19 +1606,19 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_cfg_size(struct ndctl_dimm *di struct ndctl_cmd *cmd; size_t size; - if (!ndctl_dimm_is_cmd_supported(dimm, NFIT_CMD_GET_CONFIG_SIZE)) { + if (!ndctl_dimm_is_cmd_supported(dimm, ND_CMD_GET_CONFIG_SIZE)) { dbg(ctx, "unsupported cmd\n"); return NULL; } - size = sizeof(*cmd) + sizeof(struct nfit_cmd_get_config_size); + size = sizeof(*cmd) + sizeof(struct nd_cmd_get_config_size); cmd = calloc(1, size); if (!cmd) return NULL; cmd->dimm = dimm; ndctl_cmd_ref(cmd); - cmd->type = NFIT_CMD_GET_CONFIG_SIZE; + cmd->type = ND_CMD_GET_CONFIG_SIZE; cmd->size = size; cmd->status = 1; cmd->firmware_status = &cmd->get_size->status; @@ -1618,12 +1640,12 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_cfg_read(struct ndctl_cmd *cfg struct ndctl_cmd *cmd; size_t size; - if (!ndctl_dimm_is_cmd_supported(dimm, NFIT_CMD_GET_CONFIG_DATA)) { + if (!ndctl_dimm_is_cmd_supported(dimm, ND_CMD_GET_CONFIG_DATA)) { dbg(ctx, "unsupported cmd\n"); return NULL; } - if (cfg_size->type != NFIT_CMD_GET_CONFIG_SIZE + if (cfg_size->type != ND_CMD_GET_CONFIG_SIZE || cfg_size->status != 0) { dbg(ctx, "expected sucessfully completed cfg_size command\n"); return NULL; @@ -1633,7 +1655,7 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_cfg_read(struct ndctl_cmd *cfg return NULL; } - size = sizeof(*cmd) + sizeof(struct nfit_cmd_get_config_data_hdr) + size = sizeof(*cmd) + sizeof(struct nd_cmd_get_config_data_hdr) + cfg_size->get_size->max_xfer; cmd = calloc(1, size); if (!cmd) @@ -1641,7 +1663,7 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_cfg_read(struct ndctl_cmd *cfg cmd->dimm = cfg_size->dimm; cmd->refcount = 1; - cmd->type = NFIT_CMD_GET_CONFIG_DATA; + cmd->type = ND_CMD_GET_CONFIG_DATA; cmd->size = size; cmd->status = 1; cmd->get_data->in_offset = 0; @@ -1668,13 +1690,13 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_cfg_write(struct ndctl_cmd *cf struct ndctl_cmd *cmd; size_t size; - if (!ndctl_dimm_is_cmd_supported(dimm, NFIT_CMD_SET_CONFIG_DATA)) { + if (!ndctl_dimm_is_cmd_supported(dimm, ND_CMD_SET_CONFIG_DATA)) { dbg(ctx, "unsupported cmd\n"); return NULL; } /* enforce rmw */ - if (cfg_read->type != NFIT_CMD_GET_CONFIG_DATA + if (cfg_read->type != ND_CMD_GET_CONFIG_DATA || cfg_read->status != 0) { dbg(ctx, "expected sucessfully completed cfg_read command\n"); return NULL; @@ -1685,7 +1707,7 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_cfg_write(struct ndctl_cmd *cf return NULL; } - size = sizeof(*cmd) + sizeof(struct nfit_cmd_set_config_hdr) + size = sizeof(*cmd) + sizeof(struct nd_cmd_set_config_hdr) + cfg_read->iter.max_xfer + 4; cmd = calloc(1, size); if (!cmd) @@ -1693,13 +1715,13 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_cfg_write(struct ndctl_cmd *cf cmd->dimm = cfg_read->dimm; ndctl_cmd_ref(cmd); - cmd->type = NFIT_CMD_SET_CONFIG_DATA; + cmd->type = ND_CMD_SET_CONFIG_DATA; cmd->size = size; cmd->status = 1; cmd->set_data->in_offset = 0; cmd->set_data->in_length = cfg_read->iter.max_xfer; cmd->firmware_status = (u32 *) (cmd->cmd_buf - + sizeof(struct nfit_cmd_set_config_hdr) + cfg_read->iter.max_xfer); + + sizeof(struct nd_cmd_set_config_hdr) + cfg_read->iter.max_xfer); cmd->iter.offset = &cmd->set_data->in_offset; cmd->iter.max_xfer = cfg_read->iter.max_xfer; cmd->iter.data = cmd->set_data->in_buf; @@ -1714,7 +1736,7 @@ NDCTL_EXPORT struct ndctl_cmd *ndctl_dimm_cmd_new_cfg_write(struct ndctl_cmd *cf NDCTL_EXPORT unsigned int ndctl_cmd_cfg_size_get_size(struct ndctl_cmd *cfg_size) { - if (cfg_size->type == NFIT_CMD_GET_CONFIG_SIZE + if (cfg_size->type == ND_CMD_GET_CONFIG_SIZE && cfg_size->status == 0) return cfg_size->get_size->config_size; return 0; @@ -1723,7 +1745,7 @@ NDCTL_EXPORT unsigned int ndctl_cmd_cfg_size_get_size(struct ndctl_cmd *cfg_size NDCTL_EXPORT ssize_t ndctl_cmd_cfg_read_get_data(struct ndctl_cmd *cfg_read, void *buf, unsigned int len, unsigned int offset) { - if (cfg_read->type != NFIT_CMD_GET_CONFIG_DATA || cfg_read->status > 0) + if (cfg_read->type != ND_CMD_GET_CONFIG_DATA || cfg_read->status > 0) return -EINVAL; if (cfg_read->status < 0) return cfg_read->status; @@ -1738,7 +1760,7 @@ NDCTL_EXPORT ssize_t ndctl_cmd_cfg_read_get_data(struct ndctl_cmd *cfg_read, NDCTL_EXPORT ssize_t ndctl_cmd_cfg_write_set_data(struct ndctl_cmd *cfg_write, void *buf, unsigned int len, unsigned int offset) { - if (cfg_write->type != NFIT_CMD_SET_CONFIG_DATA || cfg_write->status < 1) + if (cfg_write->type != ND_CMD_SET_CONFIG_DATA || cfg_write->status < 1) return -EINVAL; if (cfg_write->status < 0) return cfg_write->status; @@ -1752,7 +1774,7 @@ NDCTL_EXPORT ssize_t ndctl_cmd_cfg_write_set_data(struct ndctl_cmd *cfg_write, NDCTL_EXPORT ssize_t ndctl_cmd_cfg_write_zero_data(struct ndctl_cmd *cfg_write) { - if (cfg_write->type != NFIT_CMD_SET_CONFIG_DATA || cfg_write->status < 1) + if (cfg_write->type != ND_CMD_SET_CONFIG_DATA || cfg_write->status < 1) return -EINVAL; if (cfg_write->status < 0) return cfg_write->status; @@ -1838,24 +1860,24 @@ static int to_ioctl_cmd(int cmd, int dimm) { if (!dimm) { switch (cmd) { - case NFIT_CMD_ARS_CAP: return NFIT_IOCTL_ARS_CAP; - case NFIT_CMD_ARS_START: return NFIT_IOCTL_ARS_START; - case NFIT_CMD_ARS_QUERY: return NFIT_IOCTL_ARS_QUERY; + case ND_CMD_ARS_CAP: return ND_IOCTL_ARS_CAP; + case ND_CMD_ARS_START: return ND_IOCTL_ARS_START; + case ND_CMD_ARS_QUERY: return ND_IOCTL_ARS_QUERY; default: return 0; }; } switch (cmd) { - case NFIT_CMD_SMART: return NFIT_IOCTL_SMART; - case NFIT_CMD_SMART_THRESHOLD: return NFIT_IOCTL_SMART_THRESHOLD; - case NFIT_CMD_DIMM_FLAGS: return NFIT_IOCTL_DIMM_FLAGS; - case NFIT_CMD_GET_CONFIG_SIZE: return NFIT_IOCTL_GET_CONFIG_SIZE; - case NFIT_CMD_GET_CONFIG_DATA: return NFIT_IOCTL_GET_CONFIG_DATA; - case NFIT_CMD_SET_CONFIG_DATA: return NFIT_IOCTL_SET_CONFIG_DATA; - case NFIT_CMD_VENDOR: return NFIT_IOCTL_VENDOR; - case NFIT_CMD_VENDOR_EFFECT_LOG_SIZE: - case NFIT_CMD_VENDOR_EFFECT_LOG: + case ND_CMD_SMART: return ND_IOCTL_SMART; + case ND_CMD_SMART_THRESHOLD: return ND_IOCTL_SMART_THRESHOLD; + case ND_CMD_DIMM_FLAGS: return ND_IOCTL_DIMM_FLAGS; + case ND_CMD_GET_CONFIG_SIZE: return ND_IOCTL_GET_CONFIG_SIZE; + case ND_CMD_GET_CONFIG_DATA: return ND_IOCTL_GET_CONFIG_DATA; + case ND_CMD_SET_CONFIG_DATA: return ND_IOCTL_SET_CONFIG_DATA; + case ND_CMD_VENDOR: return ND_IOCTL_VENDOR; + case ND_CMD_VENDOR_EFFECT_LOG_SIZE: + case ND_CMD_VENDOR_EFFECT_LOG: default: return 0; } @@ -1877,10 +1899,11 @@ static int do_cmd(int fd, int ioctl_cmd, struct ndctl_cmd *cmd) if (iter->total_xfer == 0) { rc = ioctl(fd, ioctl_cmd, cmd->cmd_buf); - dbg(ctx, "bus: %d dimm: %#x cmd: %s status: %d fw: %d\n", + dbg(ctx, "bus: %d dimm: %#x cmd: %s status: %d fw: %d (%s)\n", bus->id, cmd->dimm ? ndctl_dimm_get_handle(cmd->dimm) : 0, - name, rc, *(cmd->firmware_status)); + name, rc, *(cmd->firmware_status), rc < 0 ? + strerror(errno) : "success"); return rc; } @@ -1891,10 +1914,11 @@ static int do_cmd(int fd, int ioctl_cmd, struct ndctl_cmd *cmd) iter->max_xfer); *(cmd->iter.offset) = offset; rc = ioctl(fd, ioctl_cmd, cmd->cmd_buf); - dbg(ctx, "bus: %d dimm: %#x cmd: %s offset: %d status: (%d:%d)\n", + dbg(ctx, "bus: %d dimm: %#x cmd: %s offset: %d status: %d fw: %d (%s)\n", bus->id, cmd->dimm ? ndctl_dimm_get_handle(cmd->dimm) : 0, - name, offset, rc, *(cmd->firmware_status)); + name, offset, rc, *(cmd->firmware_status), + rc < 0 ? strerror(errno) : "success"); if (rc) return rc; @@ -2215,7 +2239,7 @@ static void mappings_init(struct ndctl_region *region) return; region->mappings_init = 1; - mapping_path = calloc(1, strlen(region->region_path) + 20); + mapping_path = calloc(1, strlen(region->region_path) + 100); if (!mapping_path) { err(ctx, "bus%d region%d: allocation failure\n", bus->id, region->id); @@ -2226,7 +2250,7 @@ static void mappings_init(struct ndctl_region *region) struct ndctl_mapping *mapping; unsigned long long offset, length; struct ndctl_dimm *dimm; - unsigned int handle; + unsigned int dimm_id; sprintf(mapping_path, "%s/mapping%d", region->region_path, i); if (sysfs_read_attr(ctx, mapping_path, buf) < 0) { @@ -2235,17 +2259,17 @@ static void mappings_init(struct ndctl_region *region) continue; } - if (sscanf(buf, "%x,%llu,%llu", &handle, &offset, + if (sscanf(buf, "nmem%u,%llu,%llu", &dimm_id, &offset, &length) != 3) { err(ctx, "bus%d mapping parse failure\n", ndctl_bus_get_id(bus)); continue; } - dimm = ndctl_dimm_get_by_handle(bus, handle); + dimm = ndctl_dimm_get_by_id(bus, dimm_id); if (!dimm) { - err(ctx, "bus%d region%d mapping%d: dimm lookup failure\n", - bus->id, region->id, i); + err(ctx, "bus%d region%d mapping%d: nmem%d lookup failure\n", + bus->id, region->id, i, dimm_id); continue; } @@ -2350,7 +2374,7 @@ static int parse_lbasize_supported(struct ndctl_ctx *ctx, const char *buf, static int add_namespace(void *parent, int id, const char *ndns_base) { - char *path = calloc(1, strlen(ndns_base) + 20); + char *path = calloc(1, strlen(ndns_base) + 100); struct ndctl_namespace *ndns, *ndns_dup; struct ndctl_region *region = parent; struct ndctl_bus *bus = region->bus; @@ -2379,7 +2403,7 @@ static int add_namespace(void *parent, int id, const char *ndns_base) ndns->size = strtoull(buf, NULL, 0); switch (ndns->type) { - case ND_DEVICE_NAMESPACE_BLOCK: + case ND_DEVICE_NAMESPACE_BLK: sprintf(path, "%s/sector_size", ndns_base); if (sysfs_read_attr(ctx, path, buf) < 0) goto err_read; @@ -2708,7 +2732,7 @@ NDCTL_EXPORT int ndctl_namespace_is_configured(struct ndctl_namespace *ndns) return pmem_namespace_is_configured(ndns); case ND_DEVICE_NAMESPACE_IO: return 1; - case ND_DEVICE_NAMESPACE_BLOCK: + case ND_DEVICE_NAMESPACE_BLK: return blk_namespace_is_configured(ndns); default: dbg(ctx, "%s: nstype: %d is_configured() not implemented\n", @@ -2875,7 +2899,7 @@ NDCTL_EXPORT int ndctl_namespace_set_size(struct ndctl_namespace *ndns, switch (ndctl_namespace_get_type(ndns)) { case ND_DEVICE_NAMESPACE_PMEM: - case ND_DEVICE_NAMESPACE_BLOCK: + case ND_DEVICE_NAMESPACE_BLK: return namespace_set_size(ndns, size); default: dbg(ctx, "%s: nstype: %d set size failed\n", @@ -2901,7 +2925,7 @@ NDCTL_EXPORT int ndctl_namespace_delete(struct ndctl_namespace *ndns) switch (ndctl_namespace_get_type(ndns)) { case ND_DEVICE_NAMESPACE_PMEM: - case ND_DEVICE_NAMESPACE_BLOCK: + case ND_DEVICE_NAMESPACE_BLK: break; default: dbg(ctx, "%s: nstype: %d delete failed\n", @@ -2966,7 +2990,7 @@ static int parse_lbasize_supported(struct ndctl_ctx *ctx, const char *buf, static int add_btt(void *parent, int id, const char *btt_base) { - char *path = calloc(1, strlen(btt_base) + 20); + char *path = calloc(1, strlen(btt_base) + 100); struct ndctl_bus *bus = parent; struct ndctl_ctx *ctx = bus->ctx; struct ndctl_btt *btt, *btt_dup; diff --git a/lib/libndctl.sym b/lib/libndctl.sym index 31f2ffce4b1f..eb62e36d647b 100644 --- a/lib/libndctl.sym +++ b/lib/libndctl.sym @@ -27,6 +27,7 @@ global: ndctl_bus_get_btt_seed; ndctl_bus_get_cmd_name; ndctl_bus_is_cmd_supported; + ndctl_bus_has_nfit; ndctl_bus_get_revision; ndctl_bus_get_id; ndctl_bus_get_provider; diff --git a/lib/ndctl/libndctl.h b/lib/ndctl/libndctl.h index 5c27a435b8fb..126c511e289e 100644 --- a/lib/ndctl/libndctl.h +++ b/lib/ndctl/libndctl.h @@ -96,6 +96,7 @@ struct ndctl_bus *ndctl_bus_get_next(struct ndctl_bus *bus); bus != NULL; \ bus = ndctl_bus_get_next(bus)) struct ndctl_ctx *ndctl_bus_get_ctx(struct ndctl_bus *bus); +int ndctl_bus_has_nfit(struct ndctl_bus *bus); unsigned int ndctl_bus_get_major(struct ndctl_bus *bus); unsigned int ndctl_bus_get_minor(struct ndctl_bus *bus); const char *ndctl_bus_get_devname(struct ndctl_bus *bus); diff --git a/lib/test-dpa-alloc.c b/lib/test-dpa-alloc.c index f94838ac0b9b..9a881f301c7f 100644 --- a/lib/test-dpa-alloc.c +++ b/lib/test-dpa-alloc.c @@ -90,7 +90,7 @@ static int do_test(struct ndctl_ctx *ctx) * to lowest with no excursions into BLK only ranges. */ ndctl_region_foreach(bus, region) { - if (ndctl_region_get_type(region) != ND_DEVICE_REGION_BLOCK) + if (ndctl_region_get_type(region) != ND_DEVICE_REGION_BLK) continue; dimm = ndctl_region_get_first_dimm(region); if (!dimm) diff --git a/lib/test-libndctl.c b/lib/test-libndctl.c index 710aeb1f52b7..9e59861128bf 100644 --- a/lib/test-libndctl.c +++ b/lib/test-libndctl.c @@ -54,7 +54,7 @@ * +----------------------------+--------+--------+ * * *) In this layout we have four dimms and two memory controllers in one - * socket. Each unique interface ("block" or "pmem") to DPA space + * socket. Each unique interface ("blk" or "pmem") to DPA space * is identified by a region device with a dynamically assigned id. * * *) The first portion of dimm0 and dimm1 are interleaved as REGION0. @@ -185,7 +185,7 @@ static struct namespace namespace1_pmem0 = { }; static struct namespace namespace2_blk0 = { - 0, "namespace_block", NULL, SZ_7M, + 0, "namespace_blk", NULL, SZ_7M, { 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, @@ -193,7 +193,7 @@ static struct namespace namespace2_blk0 = { }; static struct namespace namespace2_blk1 = { - 1, "namespace_block", NULL, SZ_11M, + 1, "namespace_blk", NULL, SZ_11M, { 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, @@ -201,7 +201,7 @@ static struct namespace namespace2_blk1 = { }; static struct namespace namespace3_blk0 = { - 0, "namespace_block", NULL, SZ_7M, + 0, "namespace_blk", NULL, SZ_7M, { 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, @@ -209,7 +209,7 @@ static struct namespace namespace3_blk0 = { }; static struct namespace namespace3_blk1 = { - 1, "namespace_block", NULL, SZ_11M, + 1, "namespace_blk", NULL, SZ_11M, { 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, 6, @@ -217,7 +217,7 @@ static struct namespace namespace3_blk1 = { }; static struct namespace namespace4_blk0 = { - 0, "namespace_block", &btt_settings, SZ_27M, + 0, "namespace_blk", &btt_settings, SZ_27M, { 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, 7, @@ -225,7 +225,7 @@ static struct namespace namespace4_blk0 = { }; static struct namespace namespace5_blk0 = { - 0, "namespace_block", &btt_settings, SZ_27M, + 0, "namespace_blk", &btt_settings, SZ_27M, { 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, 8, @@ -237,13 +237,13 @@ static struct region regions0[] = { { &namespace0_pmem0, NULL, }, }, { { 2 }, 4, 1, "pmem", SZ_64M, SZ_64M, { 1 }, { &namespace1_pmem0, NULL, }, }, - { { DIMM_HANDLE(0, 0, 0, 0, 0) }, 1, 1, "block", SZ_18M, SZ_32M, { }, + { { DIMM_HANDLE(0, 0, 0, 0, 0) }, 1, 1, "blk", SZ_18M, SZ_32M, { }, { &namespace2_blk0, &namespace2_blk1, NULL, }, }, - { { DIMM_HANDLE(0, 0, 0, 0, 1) }, 1, 1, "block", SZ_18M, SZ_32M, { }, + { { DIMM_HANDLE(0, 0, 0, 0, 1) }, 1, 1, "blk", SZ_18M, SZ_32M, { }, { &namespace3_blk0, &namespace3_blk1, NULL, }, }, - { { DIMM_HANDLE(0, 0, 1, 0, 0) }, 1, 1, "block", SZ_27M, SZ_32M, { }, + { { DIMM_HANDLE(0, 0, 1, 0, 0) }, 1, 1, "blk", SZ_27M, SZ_32M, { }, { &namespace4_blk0, NULL, }, }, - { { DIMM_HANDLE(0, 0, 1, 0, 1) }, 1, 1, "block", SZ_27M, SZ_32M, { }, + { { DIMM_HANDLE(0, 0, 1, 0, 1) }, 1, 1, "blk", SZ_27M, SZ_32M, { }, { &namespace5_blk0, NULL, }, }, }; @@ -252,7 +252,7 @@ static struct namespace namespace1 = { }; static struct region regions1[] = { - { { 1 }, 0, 1, "pmem", 0, SZ_32M, + { { 1 }, 1, 1, "pmem", 0, SZ_32M, .namespaces = { [0] = &namespace1, }, @@ -267,9 +267,9 @@ static struct btt btts1[] = { { 0, { 0, }, 2, { 512, 4096, }, }, }; -static unsigned long commands0 = 1UL << NFIT_CMD_GET_CONFIG_SIZE - | 1UL << NFIT_CMD_GET_CONFIG_DATA - | 1UL << NFIT_CMD_SET_CONFIG_DATA; +static unsigned long commands0 = 1UL << ND_CMD_GET_CONFIG_SIZE + | 1UL << ND_CMD_GET_CONFIG_DATA + | 1UL << ND_CMD_SET_CONFIG_DATA; static struct ndctl_dimm *get_dimm_by_handle(struct ndctl_bus *bus, unsigned int handle) { @@ -339,7 +339,7 @@ static struct ndctl_region *get_blk_region_by_dimm_handle(struct ndctl_bus *bus, ndctl_region_foreach(bus, region) { struct ndctl_mapping *map; - if (ndctl_region_get_type(region) != ND_DEVICE_REGION_BLOCK) + if (ndctl_region_get_type(region) != ND_DEVICE_REGION_BLK) continue; ndctl_mapping_foreach(region, map) { struct ndctl_dimm *dimm = ndctl_mapping_get_dimm(map); @@ -904,7 +904,7 @@ static int check_get_config_size(struct ndctl_dimm *dimm, struct check_cmd *chec static int check_get_config_data(struct ndctl_dimm *dimm, struct check_cmd *check) { - struct ndctl_cmd *cmd_size = check_cmds[NFIT_CMD_GET_CONFIG_SIZE].cmd; + struct ndctl_cmd *cmd_size = check_cmds[ND_CMD_GET_CONFIG_SIZE].cmd; struct ndctl_cmd *cmd = ndctl_dimm_cmd_new_cfg_read(cmd_size); static char buf[SZ_128K]; ssize_t rc; @@ -937,7 +937,7 @@ static int check_get_config_data(struct ndctl_dimm *dimm, struct check_cmd *chec static int check_set_config_data(struct ndctl_dimm *dimm, struct check_cmd *check) { - struct ndctl_cmd *cmd_read = check_cmds[NFIT_CMD_GET_CONFIG_DATA].cmd; + struct ndctl_cmd *cmd_read = check_cmds[ND_CMD_GET_CONFIG_DATA].cmd; struct ndctl_cmd *cmd = ndctl_dimm_cmd_new_cfg_write(cmd_read); char buf[20], result[sizeof(buf)]; size_t rc; @@ -1010,15 +1010,15 @@ static int check_commands(struct ndctl_bus *bus, struct ndctl_dimm *dimm, * For now, by coincidence, these are indexed in test execution * order such that check_get_config_data can assume that * check_get_config_size has updated - * check_cmd[NFIT_CMD_GET_CONFIG_SIZE].cmd and + * check_cmd[ND_CMD_GET_CONFIG_SIZE].cmd and * check_set_config_data can assume that both * check_get_config_size and check_get_config_data have run */ static struct check_cmd __check_cmds[] = { - [NFIT_CMD_GET_CONFIG_SIZE] = { check_get_config_size }, - [NFIT_CMD_GET_CONFIG_DATA] = { check_get_config_data }, - [NFIT_CMD_SET_CONFIG_DATA] = { check_set_config_data }, - [NFIT_CMD_SMART_THRESHOLD] = { }, + [ND_CMD_GET_CONFIG_SIZE] = { check_get_config_size }, + [ND_CMD_GET_CONFIG_DATA] = { check_get_config_data }, + [ND_CMD_SET_CONFIG_DATA] = { check_set_config_data }, + [ND_CMD_SMART_THRESHOLD] = { }, }; unsigned int i, rc; diff --git a/ndctl.h b/ndctl.h index abddfdd2cfb4..baf1c506fee8 100644 --- a/ndctl.h +++ b/ndctl.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2014, Intel Corporation. + * Copyright (c) 2014-2015, Intel Corporation. * * This program is free software; you can redistribute it and/or modify it * under the terms and conditions of the GNU Lesser General Public License, @@ -15,60 +15,60 @@ #include <linux/types.h> -struct nfit_cmd_smart { +struct nd_cmd_smart { __u32 status; __u8 data[128]; } __attribute__((packed)); -struct nfit_cmd_smart_threshold { +struct nd_cmd_smart_threshold { __u32 status; __u8 data[8]; } __attribute__((packed)); -struct nfit_cmd_dimm_flags { +struct nd_cmd_dimm_flags { __u32 status; __u32 flags; } __attribute__((packed)); -struct nfit_cmd_get_config_size { +struct nd_cmd_get_config_size { __u32 status; __u32 config_size; __u32 max_xfer; } __attribute__((packed)); -struct nfit_cmd_get_config_data_hdr { +struct nd_cmd_get_config_data_hdr { __u32 in_offset; __u32 in_length; __u32 status; __u8 out_buf[0]; } __attribute__((packed)); -struct nfit_cmd_set_config_hdr { +struct nd_cmd_set_config_hdr { __u32 in_offset; __u32 in_length; __u8 in_buf[0]; } __attribute__((packed)); -struct nfit_cmd_vendor_hdr { +struct nd_cmd_vendor_hdr { __u32 opcode; __u32 in_length; __u8 in_buf[0]; } __attribute__((packed)); -struct nfit_cmd_vendor_tail { +struct nd_cmd_vendor_tail { __u32 status; __u32 out_length; __u8 out_buf[0]; } __attribute__((packed)); -struct nfit_cmd_ars_cap { +struct nd_cmd_ars_cap { __u64 address; __u64 length; __u32 status; __u32 max_ars_out; } __attribute__((packed)); -struct nfit_cmd_ars_start { +struct nd_cmd_ars_start { __u64 address; __u64 length; __u16 type; @@ -76,15 +76,15 @@ struct nfit_cmd_ars_start { __u32 status; } __attribute__((packed)); -struct nfit_cmd_ars_query { +struct nd_cmd_ars_query { __u32 status; - __u16 out_length; + __u32 out_length; __u64 address; __u64 length; __u16 type; __u32 num_records; - struct nfit_ars_record { - __u32 nfit_handle; + struct nd_ars_record { + __u32 handle; __u32 flags; __u64 err_address; __u64 mask; @@ -92,31 +92,31 @@ struct nfit_cmd_ars_query { } __attribute__((packed)); enum { - NFIT_CMD_IMPLEMENTED = 0, + ND_CMD_IMPLEMENTED = 0, /* bus commands */ - NFIT_CMD_ARS_CAP = 1, - NFIT_CMD_ARS_START = 2, - NFIT_CMD_ARS_QUERY = 3, + ND_CMD_ARS_CAP = 1, + ND_CMD_ARS_START = 2, + ND_CMD_ARS_QUERY = 3, /* per-dimm commands */ - NFIT_CMD_SMART = 1, - NFIT_CMD_SMART_THRESHOLD = 2, - NFIT_CMD_DIMM_FLAGS = 3, - NFIT_CMD_GET_CONFIG_SIZE = 4, - NFIT_CMD_GET_CONFIG_DATA = 5, - NFIT_CMD_SET_CONFIG_DATA = 6, - NFIT_CMD_VENDOR_EFFECT_LOG_SIZE = 7, - NFIT_CMD_VENDOR_EFFECT_LOG = 8, - NFIT_CMD_VENDOR = 9, + ND_CMD_SMART = 1, + ND_CMD_SMART_THRESHOLD = 2, + ND_CMD_DIMM_FLAGS = 3, + ND_CMD_GET_CONFIG_SIZE = 4, + ND_CMD_GET_CONFIG_DATA = 5, + ND_CMD_SET_CONFIG_DATA = 6, + ND_CMD_VENDOR_EFFECT_LOG_SIZE = 7, + ND_CMD_VENDOR_EFFECT_LOG = 8, + ND_CMD_VENDOR = 9, }; -static __inline__ const char *nfit_bus_cmd_name(unsigned cmd) +static __inline__ const char *nd_bus_cmd_name(unsigned cmd) { - static const char *names[] = { - [NFIT_CMD_ARS_CAP] = "ars_cap", - [NFIT_CMD_ARS_START] = "ars_start", - [NFIT_CMD_ARS_QUERY] = "ars_query", + static const char * const names[] = { + [ND_CMD_ARS_CAP] = "ars_cap", + [ND_CMD_ARS_START] = "ars_start", + [ND_CMD_ARS_QUERY] = "ars_query", }; if (cmd < ARRAY_SIZE(names) && names[cmd]) @@ -124,18 +124,18 @@ static __inline__ const char *nfit_bus_cmd_name(unsigned cmd) return "unknown"; } -static __inline__ const char *nfit_dimm_cmd_name(unsigned cmd) +static __inline__ const char *nd_dimm_cmd_name(unsigned cmd) { - static const char *names[] = { - [NFIT_CMD_SMART] = "smart", - [NFIT_CMD_SMART_THRESHOLD] = "smart_thresh", - [NFIT_CMD_DIMM_FLAGS] = "flags", - [NFIT_CMD_GET_CONFIG_SIZE] = "get_size", - [NFIT_CMD_GET_CONFIG_DATA] = "get_data", - [NFIT_CMD_SET_CONFIG_DATA] = "set_data", - [NFIT_CMD_VENDOR_EFFECT_LOG_SIZE] = "effect_size", - [NFIT_CMD_VENDOR_EFFECT_LOG] = "effect_log", - [NFIT_CMD_VENDOR] = "vendor", + static const char * const names[] = { + [ND_CMD_SMART] = "smart", + [ND_CMD_SMART_THRESHOLD] = "smart_thresh", + [ND_CMD_DIMM_FLAGS] = "flags", + [ND_CMD_GET_CONFIG_SIZE] = "get_size", + [ND_CMD_GET_CONFIG_DATA] = "get_data", + [ND_CMD_SET_CONFIG_DATA] = "set_data", + [ND_CMD_VENDOR_EFFECT_LOG_SIZE] = "effect_size", + [ND_CMD_VENDOR_EFFECT_LOG] = "effect_log", + [ND_CMD_VENDOR] = "vendor", }; if (cmd < ARRAY_SIZE(names) && names[cmd]) @@ -145,51 +145,51 @@ static __inline__ const char *nfit_dimm_cmd_name(unsigned cmd) #define ND_IOCTL 'N' -#define NFIT_IOCTL_SMART _IOWR(ND_IOCTL, NFIT_CMD_SMART,\ - struct nfit_cmd_smart) +#define ND_IOCTL_SMART _IOWR(ND_IOCTL, ND_CMD_SMART,\ + struct nd_cmd_smart) -#define NFIT_IOCTL_SMART_THRESHOLD _IOWR(ND_IOCTL, NFIT_CMD_SMART_THRESHOLD,\ - struct nfit_cmd_smart_threshold) +#define ND_IOCTL_SMART_THRESHOLD _IOWR(ND_IOCTL, ND_CMD_SMART_THRESHOLD,\ + struct nd_cmd_smart_threshold) -#define NFIT_IOCTL_DIMM_FLAGS _IOWR(ND_IOCTL, NFIT_CMD_DIMM_FLAGS,\ - struct nfit_cmd_dimm_flags) +#define ND_IOCTL_DIMM_FLAGS _IOWR(ND_IOCTL, ND_CMD_DIMM_FLAGS,\ + struct nd_cmd_dimm_flags) -#define NFIT_IOCTL_GET_CONFIG_SIZE _IOWR(ND_IOCTL, NFIT_CMD_GET_CONFIG_SIZE,\ - struct nfit_cmd_get_config_size) +#define ND_IOCTL_GET_CONFIG_SIZE _IOWR(ND_IOCTL, ND_CMD_GET_CONFIG_SIZE,\ + struct nd_cmd_get_config_size) -#define NFIT_IOCTL_GET_CONFIG_DATA _IOWR(ND_IOCTL, NFIT_CMD_GET_CONFIG_DATA,\ - struct nfit_cmd_get_config_data_hdr) +#define ND_IOCTL_GET_CONFIG_DATA _IOWR(ND_IOCTL, ND_CMD_GET_CONFIG_DATA,\ + struct nd_cmd_get_config_data_hdr) -#define NFIT_IOCTL_SET_CONFIG_DATA _IOWR(ND_IOCTL, NFIT_CMD_SET_CONFIG_DATA,\ - struct nfit_cmd_set_config_hdr) +#define ND_IOCTL_SET_CONFIG_DATA _IOWR(ND_IOCTL, ND_CMD_SET_CONFIG_DATA,\ + struct nd_cmd_set_config_hdr) -#define NFIT_IOCTL_VENDOR _IOWR(ND_IOCTL, NFIT_CMD_VENDOR,\ - struct nfit_cmd_vendor_hdr) +#define ND_IOCTL_VENDOR _IOWR(ND_IOCTL, ND_CMD_VENDOR,\ + struct nd_cmd_vendor_hdr) -#define NFIT_IOCTL_ARS_CAP _IOWR(ND_IOCTL, NFIT_CMD_ARS_CAP,\ - struct nfit_cmd_ars_cap) +#define ND_IOCTL_ARS_CAP _IOWR(ND_IOCTL, ND_CMD_ARS_CAP,\ + struct nd_cmd_ars_cap) -#define NFIT_IOCTL_ARS_START _IOWR(ND_IOCTL, NFIT_CMD_ARS_START,\ - struct nfit_cmd_ars_start) +#define ND_IOCTL_ARS_START _IOWR(ND_IOCTL, ND_CMD_ARS_START,\ + struct nd_cmd_ars_start) -#define NFIT_IOCTL_ARS_QUERY _IOWR(ND_IOCTL, NFIT_CMD_ARS_QUERY,\ - struct nfit_cmd_ars_query) +#define ND_IOCTL_ARS_QUERY _IOWR(ND_IOCTL, ND_CMD_ARS_QUERY,\ + struct nd_cmd_ars_query) #define ND_DEVICE_DIMM 1 /* nd_dimm: container for "config data" */ #define ND_DEVICE_REGION_PMEM 2 /* nd_region: (parent of pmem namespaces) */ -#define ND_DEVICE_REGION_BLOCK 3 /* nd_region: (parent of block namespaces) */ +#define ND_DEVICE_REGION_BLK 3 /* nd_region: (parent of blk namespaces) */ #define ND_DEVICE_NAMESPACE_IO 4 /* legacy persistent memory */ #define ND_DEVICE_NAMESPACE_PMEM 5 /* persistent memory namespace (may alias) */ -#define ND_DEVICE_NAMESPACE_BLOCK 6 /* block-data-window namespace (may alias) */ +#define ND_DEVICE_NAMESPACE_BLK 6 /* block-data-window namespace (may alias) */ #define ND_DEVICE_BTT 7 /* block-translation table device */ enum nd_driver_flags { ND_DRIVER_DIMM = 1 << ND_DEVICE_DIMM, ND_DRIVER_REGION_PMEM = 1 << ND_DEVICE_REGION_PMEM, - ND_DRIVER_REGION_BLOCK = 1 << ND_DEVICE_REGION_BLOCK, + ND_DRIVER_REGION_BLK = 1 << ND_DEVICE_REGION_BLK, ND_DRIVER_NAMESPACE_IO = 1 << ND_DEVICE_NAMESPACE_IO, ND_DRIVER_NAMESPACE_PMEM = 1 << ND_DEVICE_NAMESPACE_PMEM, - ND_DRIVER_NAMESPACE_BLOCK = 1 << ND_DEVICE_NAMESPACE_BLOCK, + ND_DRIVER_NAMESPACE_BLK = 1 << ND_DEVICE_NAMESPACE_BLK, ND_DRIVER_BTT = 1 << ND_DEVICE_BTT, };

7 years, 5 months

1
0
0 / 0

[Linux-nvdimm] [PATCH 00/21] ND: NFIT-Defined / NVDIMM Subsystem

by Dan Williams

Since 2010 Intel has included non-volatile memory support on a few storage-focused platforms with a feature named ADR (Asynchronous DRAM Refresh). These platforms were mostly targeted at custom applications and never enjoyed standard discovery mechanisms for platform firmware to advertise non-volatile memory capabilities. This now changes with the publication of version 6 of the ACPI specification [1] and its inclusion of a new table for describing platform memory capabilities. The NVDIMM Firmware Interface Table (NFIT), along with new EFI and E820 memory types, enumerates persistent memory ranges, memory-mapped-I/O apertures, physical memory devices (DIMMs), and their associated properties. The ND-subsystem wraps a Linux device driver model around the objects and address boundaries defined in the specification and introduces 3 new drivers. nd_pmem: NFIT enabled version of the existing 'pmem' driver [2] nd_blk: mmio aperture method for accessing persistent storage nd_btt: give persistent memory disk semantics (atomic sector update) See the documentation in patch2 for more details, and there is supplemental documentation on pmem.io [4]. Please review, and patches welcome... For kicking the tires, this release is accompanied by a userspace management library 'ndctl' that includes unit tests (make check) for all of the kernel ABIs. The nfit_test.ko module can be used to explore a sample NFIT topology. [1]: http://www.uefi.org/sites/default/files/resources/ACPI_6.0.pdf [2]: https://git.kernel.org/cgit/linux/kernel/git/tip/tip.git/log/?h=x86/pmem [3]: https://github.com/pmem/ndctl [4]: http://pmem.io/documents/ -- Dan for the NFIT driver development team Andy Rudoff, Matthew Wilcox, Ross Zwisler, and Vishal Verma --- Dan Williams (19): e820, efi: add ACPI 6.0 persistent memory types ND NFIT-Defined/NVIDIMM Subsystem nd_acpi: initial core implementation and nfit skeleton nd: create an 'nd_bus' from an 'nfit_desc' nfit-test: manufactured NFITs for interface development nd: ndctl class device, and nd bus attributes nd: dimm devices (nfit "memory-devices") nd: ndctl.h, the nd ioctl abi nd_dimm: dimm driver and base nd-bus device-driver infrastructure nd: regions (block-data-window, persistent memory, volatile memory) nd_region: support for legacy nvdimms nd_pmem: add NFIT support to the pmem driver nd: add interleave-set state-tracking infrastructure nd: namespace indices: read and validate nd: pmem label sets and namespace instantiation. nd: blk labels and namespace instantiation nd: write pmem label set nd: write blk label set nd: infrastructure for btt devices Ross Zwisler (1): nd_blk: nfit blk driver Vishal Verma (1): nd_btt: atomic sector updates Documentation/blockdev/btt.txt | 273 ++++++ Documentation/blockdev/nd.txt | 867 +++++++++++++++++++ MAINTAINERS | 34 + arch/arm64/kernel/efi.c | 1 arch/ia64/kernel/efi.c | 1 arch/x86/boot/compressed/eboot.c | 4 arch/x86/include/uapi/asm/e820.h | 1 arch/x86/kernel/e820.c | 25 - arch/x86/platform/efi/efi.c | 3 drivers/block/Kconfig | 13 drivers/block/Makefile | 2 drivers/block/nd/Kconfig | 130 +++ drivers/block/nd/Makefile | 39 + drivers/block/nd/acpi.c | 443 ++++++++++ drivers/block/nd/blk.c | 269 ++++++ drivers/block/nd/btt.c | 1423 +++++++++++++++++++++++++++++++ drivers/block/nd/btt.h | 185 ++++ drivers/block/nd/btt_devs.c | 443 ++++++++++ drivers/block/nd/bus.c | 703 +++++++++++++++ drivers/block/nd/core.c | 963 +++++++++++++++++++++ drivers/block/nd/dimm.c | 126 +++ drivers/block/nd/dimm_devs.c | 701 +++++++++++++++ drivers/block/nd/label.c | 925 ++++++++++++++++++++ drivers/block/nd/label.h | 143 +++ drivers/block/nd/namespace_devs.c | 1697 +++++++++++++++++++++++++++++++++++++ drivers/block/nd/nd-private.h | 203 ++++ drivers/block/nd/nd.h | 310 +++++++ drivers/block/nd/nfit.h | 238 +++++ drivers/block/nd/pmem.c | 122 ++- drivers/block/nd/region.c | 95 ++ drivers/block/nd/region_devs.c | 1196 ++++++++++++++++++++++++++ drivers/block/nd/test/Makefile | 5 drivers/block/nd/test/iomap.c | 199 ++++ drivers/block/nd/test/nfit.c | 1018 ++++++++++++++++++++++ drivers/block/nd/test/nfit_test.h | 37 + include/linux/efi.h | 3 include/linux/nd.h | 98 ++ include/uapi/linux/Kbuild | 1 include/uapi/linux/ndctl.h | 199 ++++ 39 files changed, 13102 insertions(+), 36 deletions(-) create mode 100644 Documentation/blockdev/btt.txt create mode 100644 Documentation/blockdev/nd.txt create mode 100644 drivers/block/nd/Kconfig create mode 100644 drivers/block/nd/Makefile create mode 100644 drivers/block/nd/acpi.c create mode 100644 drivers/block/nd/blk.c create mode 100644 drivers/block/nd/btt.c create mode 100644 drivers/block/nd/btt.h create mode 100644 drivers/block/nd/btt_devs.c create mode 100644 drivers/block/nd/bus.c create mode 100644 drivers/block/nd/core.c create mode 100644 drivers/block/nd/dimm.c create mode 100644 drivers/block/nd/dimm_devs.c create mode 100644 drivers/block/nd/label.c create mode 100644 drivers/block/nd/label.h create mode 100644 drivers/block/nd/namespace_devs.c create mode 100644 drivers/block/nd/nd-private.h create mode 100644 drivers/block/nd/nd.h create mode 100644 drivers/block/nd/nfit.h rename drivers/block/{pmem.c => nd/pmem.c} (68%) create mode 100644 drivers/block/nd/region.c create mode 100644 drivers/block/nd/region_devs.c create mode 100644 drivers/block/nd/test/Makefile create mode 100644 drivers/block/nd/test/iomap.c create mode 100644 drivers/block/nd/test/nfit.c create mode 100644 drivers/block/nd/test/nfit_test.h create mode 100644 include/linux/nd.h create mode 100644 include/uapi/linux/ndctl.h

7 years, 5 months

13
83
0 / 0

[Linux-nvdimm] [ndctl PATCH] ndctl: reorder header include for ARRAY_SIZE() definition

by Dan Williams

On a system without /usr/include/linux/ndctl.h (provided by the kernel "make INSTALL_HDR_PATH=/usr headers_install"). The ndctl build falls back to using the local copy of ndctl.h, triggering these warnings. CC lib/test-dpa-alloc.o In file included from lib/test-dpa-alloc.c:33:0: ./ndctl.h: In function ‘nfit_bus_cmd_name’: ./ndctl.h:122:2: warning: implicit declaration of function ‘ARRAY_SIZE’ [-Wimplicit-function-declaration] if (cmd < ARRAY_SIZE(names) && names[cmd]) Arrange for ARRAY_SIZE() to be picked up from <ccan/array_size/array_size.h> as a fallback. Reported-by: Robert Elliott <Elliott(a)hp.com> Signed-off-by: Dan Williams <dan.j.williams(a)intel.com> --- lib/test-dpa-alloc.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/lib/test-dpa-alloc.c b/lib/test-dpa-alloc.c index 5b7db7c2f542..f94838ac0b9b 100644 --- a/lib/test-dpa-alloc.c +++ b/lib/test-dpa-alloc.c @@ -27,13 +27,14 @@ #include <uuid/uuid.h> #include <ndctl/libndctl.h> +#include <test-dpa-alloc.h> +#include <ccan/array_size/array_size.h> + #ifdef HAVE_NDCTL_H #include <linux/ndctl.h> #else #include <ndctl.h> #endif -#include <test-dpa-alloc.h> -#include <ccan/array_size/array_size.h> static const char *NFIT_TEST_MODULE = "nfit_test"; static const char *NFIT_PROVIDER0 = "nfit_test.0";

7 years, 5 months

1
0
0 / 0

[Linux-nvdimm] [PATCH] nd: fix 'sh' arch is CONFIG_SUPERH

by Dan Williams

Paul says: "...noticed this patch contains the only Kconfig typo I know by heart. Because I think you meant to say: depends on (X86 || IA64 || ARM || ARM64 || SUPERH || XTENSA)" Reported-by: Paul Bolle <pebolle(a)tiscali.nl> Signed-off-by: Dan Williams <dan.j.williams(a)intel.com> --- Thanks Paul! drivers/block/nd/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/block/nd/Kconfig b/drivers/block/nd/Kconfig index 72580cb0e39c..ba12776b44e7 100644 --- a/drivers/block/nd/Kconfig +++ b/drivers/block/nd/Kconfig @@ -1,5 +1,5 @@ config ND_ARCH_HAS_IOREMAP_CACHE - depends on (X86 || IA64 || ARM || ARM64 || SH || XTENSA) + depends on (X86 || IA64 || ARM || ARM64 || SUPERH || XTENSA) def_bool y menuconfig NFIT_DEVICES

7 years, 5 months

2
2
0 / 0

[Linux-nvdimm] Another NVDIMM test system available

by Phil Pokorny

Hello, I am new to the list and looks like I've got lucky timing. Penguin Computing just got some NVDIMM's and a BIOS to support them in our Tundra Relion 1930g system. (Haswell, DDR4, memory + flash + supercap, 8GB) I was able to compile and boot 4.0.0 and get the device detected as persistent. Loaded pmem which took ownership of it. Made an XFS filesystem on it and was able to read/write to it. As mentioned in another thread, performance isn't anything to write home about just yet, but I'm sure it will improve. I've also got some BIOS bugs to fix which I'm working on separately. I'll keep an eye out for benchmarks to run and patches to try. My goal is to be able to use an NVDIMM as a ZFS ZIL or other filesystem journal to accelerate filesystem performance. NVDIMM's are architecturally identical to the ZeusRAM SCSI drive used by many to accelerate ZFS performance. One would hope that we can get performance to equal or exceed that device (a 6Gbps SAS drive). Let me know how I can help. Phil P. -- Philip Pokorny, RHCE Chief Technology Officer PENGUIN COMPUTING, Inc www.penguincomputing.com Changing the world through technical innovation

7 years, 5 months

1
0
0 / 0

[Linux-nvdimm] pmem and i_dio_count overhead

by Elliott, Robert (Server Storage)

Jens, one of your patches from October 2013 never made it to the kernel, but would be beneficial for pmem. It helps IOPS about 15%. Original patch: https://lkml.org/lkml/2013/10/24/130 > From Jens Axboe > Subject [PATCH 05/11] direct-io: only inc/dec inode->i_dio_count for file systems > Date Thu, 24 Oct 2013 10:25:58 +0100 > > We don't need truncate protection for block devices, so add a flag > bypassing this cache line dirtying twice for every IO. This easily > contributes to 5-10% of the CPU time on high IOPS O_DIRECT testing. Here are perf top results while running fio to pmem devices using memcpy with non-temporal load and store instructions: 20.54% [pmem] [k] pmem_do_bvec.isra.6 <the memcpy function> 10.13% [kernel] [k] do_blockdev_direct_IO 5.93% [kernel] [k] inode_dio_done 4.46% [kernel] [k] bio_endio 3.07% fio [.] get_io_u 2.08% fio [.] do_io Inside do_blockdev_direct_io (10%), 60% of the time is spent atomically incrementing i_dio_count: │ static inline void atomic_inc(atomic_t *v) │ { │ asm volatile(LOCK_PREFIX "incl %0" 0.06 │ 225: lock incl 0x134(%r14) │ atomic_inc(&inode->i_dio_count); │ │ retval = 0; │ sdio.blkbits = blkbits; │ sdio.blkfactor = i_blkbits - blkbits; │ sdio.block_in_file = offset >> blkbits; 60.31 │ mov -0x1d0(%rbp),%rdx 0.16 │ mov %r12d,%ecx │ */ │ atomic_inc(&inode->i_dio_count); │ │ retval = 0; │ sdio.blkbits = blkbits; │ sdio.blkfactor = i_blkbits - blkbits; 0.00 │ sub %r12d,%ebx │ * Will be decremented at I/O completion time. │ */ │ atomic_inc(&inode->i_dio_count); inode_dio_done is taking all of its 5.8% time doing the corresponding atomic_dec. So, they're combining for 11.8% of the overall CPU time. The problem is more atomic contention than cache line dirtying. Applying your patch (changing the bitmask from 0x04 to 0x08, since 0x04 is taken now) eliminates those instructions from perf top and improves the high IOPS results by 5 to 15%. Attr Copy Read IOPS Write IOPS ==== ==== ========= ========== UC NT rd,wr 513 K 326 K with the patch: 510 K 325 K WB NT rd,wr 3.3 M 3.5 M with the patch: 3.8 M 3.9 M WC NT rd,wr 3.0 M 3.9 M with the patch: 3.1 M 4.1 M WT NT rd,wr 3.3 M 2.1 M with the patch: 3.7 M 3.7 M (there is some other test environment inconsistency with WT writes - I don't think this change really helped by 76%) --- Robert Elliott, HP Server Storage

7 years, 5 months

2
1
0 / 0

2022

2021

2020

2019

2018

2017

2016

2015

2014

Linux-nvdimm April 2015