linux_dsm_epyc7002/drivers/md/dm-linear.c
Linus Torvalds b6ffe9ba46 libnvdimm for 4.13
* Introduce the _flushcache() family of memory copy helpers and use them
   for persistent memory write operations on x86. The _flushcache()
   semantic indicates that the cache is either bypassed for the copy
   operation (movnt) or any lines dirtied by the copy operation are
   written back (clwb, clflushopt, or clflush).
 
 * Extend dax_operations with ->copy_from_iter() and ->flush()
   operations. These operations and other infrastructure updates allow
   all persistent memory specific dax functionality to be pushed into
   libnvdimm and the pmem driver directly. It also allows dax-specific
   sysfs attributes to be linked to a host device, for example:
       /sys/block/pmem0/dax/write_cache
 
 * Add support for the new NVDIMM platform/firmware mechanisms introduced
   in ACPI 6.2 and UEFI 2.7. This support includes the v1.2 namespace
   label format, extensions to the address-range-scrub command set, new
   error injection commands, and a new BTT (block-translation-table)
   layout. These updates support inter-OS and pre-OS compatibility.
 
 * Fix a longstanding memory corruption bug in nfit_test.
 
 * Make the pmem and nvdimm-region 'badblocks' sysfs files poll(2)
   capable.
 
 * Miscellaneous fixes and small updates across libnvdimm and the nfit
   driver.
 
 Acknowledgements that came after the branch was pushed:
 
 commit 6aa734a2f3 "libnvdimm, region, pmem: fix 'badblocks'
   sysfs_get_dirent() reference lifetime"
 Reviewed-by: Toshi Kani <toshi.kani@hpe.com>
 -----BEGIN PGP SIGNATURE-----
 
 iQIcBAABAgAGBQJZXsUtAAoJEB7SkWpmfYgCOXcP/06bncqTEvtgrOF2b7O8w+8e
 mTySD51RUn6UpkFd37SMRch+rmbojuqj465TAE7XIXgyLgIOJixKaTlHYUoEnP3X
 rC4Q/g5mN0nittMDwL+vQaa1lQWd2kbjOlrqCgnLHVEEJpHmiQussunjvir4G1U7
 5ROooP8W+qMK5y5XPLJAg/gyGhYkjpRSlDg3Eo5meZZ0IdURbI7+WCLKrPcQUERT
 WmDc9gLhJdSQVxBV/0m2gdAER4ADmFjcrlm8kjXRBhdlUmEFjM0zpvlHJutHTkks
 rNZWCmCJs0Sas+DmRKszFmvVFHRHqUVA3dWK4P6PJEX+tl7BwlPcxpbfacHTG2EZ
 btArFc584DZ+EIrim1cXXRvLFlxnKOFBtBeteFs7l2kZjEcN6S4I5OZgTyeDpe/i
 2WDpHWLQWibkcIzH9y1EuMBkYnQjTJl1pecHzJoTaC+jAQ+opLiY7EecjLmCmQS6
 MBYUeQZNufLGfT5b8KXfpKeiXhpFkYrAGp+ErfoH/6RKy2zqTdagN1yVhos2y+a7
 JJu/Weetpn8qv+KTGUShO8TGyWv3wU46YkG2rKWl0FL1+C+6LMMw1/L0A97lwVlg
 BpypVVyaNu1D22ifZ8O5wbqPIYghoZ5akA0CiduhX19cpl5rTeTd8EvLjvcYhZEZ
 pMHuMAqIcIyLhIe2/sRF
 =xKQB
 -----END PGP SIGNATURE-----

Merge tag 'libnvdimm-for-4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm

Pull libnvdimm updates from Dan Williams:
 "libnvdimm updates for the latest ACPI and UEFI specifications. This
  pull request also includes new 'struct dax_operations' enabling to
  undo the abuse of copy_user_nocache() for copy operations to pmem.

  The dax work originally missed 4.12 to address concerns raised by Al.

  Summary:

   - Introduce the _flushcache() family of memory copy helpers and use
     them for persistent memory write operations on x86. The
     _flushcache() semantic indicates that the cache is either bypassed
     for the copy operation (movnt) or any lines dirtied by the copy
     operation are written back (clwb, clflushopt, or clflush).

   - Extend dax_operations with ->copy_from_iter() and ->flush()
     operations. These operations and other infrastructure updates allow
     all persistent memory specific dax functionality to be pushed into
     libnvdimm and the pmem driver directly. It also allows dax-specific
     sysfs attributes to be linked to a host device, for example:
     /sys/block/pmem0/dax/write_cache

   - Add support for the new NVDIMM platform/firmware mechanisms
     introduced in ACPI 6.2 and UEFI 2.7. This support includes the v1.2
     namespace label format, extensions to the address-range-scrub
     command set, new error injection commands, and a new BTT
     (block-translation-table) layout. These updates support inter-OS
     and pre-OS compatibility.

   - Fix a longstanding memory corruption bug in nfit_test.

   - Make the pmem and nvdimm-region 'badblocks' sysfs files poll(2)
     capable.

   - Miscellaneous fixes and small updates across libnvdimm and the nfit
     driver.

  Acknowledgements that came after the branch was pushed: commit
  6aa734a2f3 ("libnvdimm, region, pmem: fix 'badblocks'
  sysfs_get_dirent() reference lifetime") was reviewed by Toshi Kani
  <toshi.kani@hpe.com>"

* tag 'libnvdimm-for-4.13' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: (42 commits)
  libnvdimm, namespace: record 'lbasize' for pmem namespaces
  acpi/nfit: Issue Start ARS to retrieve existing records
  libnvdimm: New ACPI 6.2 DSM functions
  acpi, nfit: Show bus_dsm_mask in sysfs
  libnvdimm, acpi, nfit: Add bus level dsm mask for pass thru.
  acpi, nfit: Enable DSM pass thru for root functions.
  libnvdimm: passthru functions clear to send
  libnvdimm, btt: convert some info messages to warn/err
  libnvdimm, region, pmem: fix 'badblocks' sysfs_get_dirent() reference lifetime
  libnvdimm: fix the clear-error check in nsio_rw_bytes
  libnvdimm, btt: fix btt_rw_page not returning errors
  acpi, nfit: quiet invalid block-aperture-region warnings
  libnvdimm, btt: BTT updates for UEFI 2.7 format
  acpi, nfit: constify *_attribute_group
  libnvdimm, pmem: disable dax flushing when pmem is fronting a volatile region
  libnvdimm, pmem, dax: export a cache control attribute
  dax: convert to bitmask for flags
  dax: remove default copy_from_iter fallback
  libnvdimm, nfit: enable support for volatile ranges
  libnvdimm, pmem: fix persistence warning
  ...
2017-07-07 09:44:06 -07:00

232 lines
5.3 KiB
C

/*
* Copyright (C) 2001-2003 Sistina Software (UK) Limited.
*
* This file is released under the GPL.
*/
#include "dm.h"
#include <linux/module.h>
#include <linux/init.h>
#include <linux/blkdev.h>
#include <linux/bio.h>
#include <linux/dax.h>
#include <linux/slab.h>
#include <linux/device-mapper.h>
#define DM_MSG_PREFIX "linear"
/*
* Linear: maps a linear range of a device.
*/
struct linear_c {
struct dm_dev *dev;
sector_t start;
};
/*
* Construct a linear mapping: <dev_path> <offset>
*/
static int linear_ctr(struct dm_target *ti, unsigned int argc, char **argv)
{
struct linear_c *lc;
unsigned long long tmp;
char dummy;
int ret;
if (argc != 2) {
ti->error = "Invalid argument count";
return -EINVAL;
}
lc = kmalloc(sizeof(*lc), GFP_KERNEL);
if (lc == NULL) {
ti->error = "Cannot allocate linear context";
return -ENOMEM;
}
ret = -EINVAL;
if (sscanf(argv[1], "%llu%c", &tmp, &dummy) != 1) {
ti->error = "Invalid device sector";
goto bad;
}
lc->start = tmp;
ret = dm_get_device(ti, argv[0], dm_table_get_mode(ti->table), &lc->dev);
if (ret) {
ti->error = "Device lookup failed";
goto bad;
}
ti->num_flush_bios = 1;
ti->num_discard_bios = 1;
ti->num_write_same_bios = 1;
ti->num_write_zeroes_bios = 1;
ti->private = lc;
return 0;
bad:
kfree(lc);
return ret;
}
static void linear_dtr(struct dm_target *ti)
{
struct linear_c *lc = (struct linear_c *) ti->private;
dm_put_device(ti, lc->dev);
kfree(lc);
}
static sector_t linear_map_sector(struct dm_target *ti, sector_t bi_sector)
{
struct linear_c *lc = ti->private;
return lc->start + dm_target_offset(ti, bi_sector);
}
static void linear_map_bio(struct dm_target *ti, struct bio *bio)
{
struct linear_c *lc = ti->private;
bio->bi_bdev = lc->dev->bdev;
if (bio_sectors(bio) || bio_op(bio) == REQ_OP_ZONE_RESET)
bio->bi_iter.bi_sector =
linear_map_sector(ti, bio->bi_iter.bi_sector);
}
static int linear_map(struct dm_target *ti, struct bio *bio)
{
linear_map_bio(ti, bio);
return DM_MAPIO_REMAPPED;
}
static int linear_end_io(struct dm_target *ti, struct bio *bio,
blk_status_t *error)
{
struct linear_c *lc = ti->private;
if (!*error && bio_op(bio) == REQ_OP_ZONE_REPORT)
dm_remap_zone_report(ti, bio, lc->start);
return DM_ENDIO_DONE;
}
static void linear_status(struct dm_target *ti, status_type_t type,
unsigned status_flags, char *result, unsigned maxlen)
{
struct linear_c *lc = (struct linear_c *) ti->private;
switch (type) {
case STATUSTYPE_INFO:
result[0] = '\0';
break;
case STATUSTYPE_TABLE:
snprintf(result, maxlen, "%s %llu", lc->dev->name,
(unsigned long long)lc->start);
break;
}
}
static int linear_prepare_ioctl(struct dm_target *ti,
struct block_device **bdev, fmode_t *mode)
{
struct linear_c *lc = (struct linear_c *) ti->private;
struct dm_dev *dev = lc->dev;
*bdev = dev->bdev;
/*
* Only pass ioctls through if the device sizes match exactly.
*/
if (lc->start ||
ti->len != i_size_read(dev->bdev->bd_inode) >> SECTOR_SHIFT)
return 1;
return 0;
}
static int linear_iterate_devices(struct dm_target *ti,
iterate_devices_callout_fn fn, void *data)
{
struct linear_c *lc = ti->private;
return fn(ti, lc->dev, lc->start, ti->len, data);
}
static long linear_dax_direct_access(struct dm_target *ti, pgoff_t pgoff,
long nr_pages, void **kaddr, pfn_t *pfn)
{
long ret;
struct linear_c *lc = ti->private;
struct block_device *bdev = lc->dev->bdev;
struct dax_device *dax_dev = lc->dev->dax_dev;
sector_t dev_sector, sector = pgoff * PAGE_SECTORS;
dev_sector = linear_map_sector(ti, sector);
ret = bdev_dax_pgoff(bdev, dev_sector, nr_pages * PAGE_SIZE, &pgoff);
if (ret)
return ret;
return dax_direct_access(dax_dev, pgoff, nr_pages, kaddr, pfn);
}
static size_t linear_dax_copy_from_iter(struct dm_target *ti, pgoff_t pgoff,
void *addr, size_t bytes, struct iov_iter *i)
{
struct linear_c *lc = ti->private;
struct block_device *bdev = lc->dev->bdev;
struct dax_device *dax_dev = lc->dev->dax_dev;
sector_t dev_sector, sector = pgoff * PAGE_SECTORS;
dev_sector = linear_map_sector(ti, sector);
if (bdev_dax_pgoff(bdev, dev_sector, ALIGN(bytes, PAGE_SIZE), &pgoff))
return 0;
return dax_copy_from_iter(dax_dev, pgoff, addr, bytes, i);
}
static void linear_dax_flush(struct dm_target *ti, pgoff_t pgoff, void *addr,
size_t size)
{
struct linear_c *lc = ti->private;
struct block_device *bdev = lc->dev->bdev;
struct dax_device *dax_dev = lc->dev->dax_dev;
sector_t dev_sector, sector = pgoff * PAGE_SECTORS;
dev_sector = linear_map_sector(ti, sector);
if (bdev_dax_pgoff(bdev, dev_sector, ALIGN(size, PAGE_SIZE), &pgoff))
return;
dax_flush(dax_dev, pgoff, addr, size);
}
static struct target_type linear_target = {
.name = "linear",
.version = {1, 4, 0},
.features = DM_TARGET_PASSES_INTEGRITY | DM_TARGET_ZONED_HM,
.module = THIS_MODULE,
.ctr = linear_ctr,
.dtr = linear_dtr,
.map = linear_map,
.end_io = linear_end_io,
.status = linear_status,
.prepare_ioctl = linear_prepare_ioctl,
.iterate_devices = linear_iterate_devices,
.direct_access = linear_dax_direct_access,
.dax_copy_from_iter = linear_dax_copy_from_iter,
.dax_flush = linear_dax_flush,
};
int __init dm_linear_init(void)
{
int r = dm_register_target(&linear_target);
if (r < 0)
DMERR("register failed %d", r);
return r;
}
void dm_linear_exit(void)
{
dm_unregister_target(&linear_target);
}