From: Dan Williams <dan.j.williams@intel.com> To: linux-nvdimm@lists.01.org Cc: Jan Kara <jack@suse.cz>, Matthew Wilcox <mawilcox@microsoft.com>, x86@kernel.org, linux-kernel@vger.kernel.org, viro@zeniv.linux.org.uk, linux-fsdevel@vger.kernel.org, hch@lst.de Subject: [PATCH v4 15/16] libnvdimm, pmem, dax: export a cache control attribute Date: Thu, 29 Jun 2017 10:54:17 -0700 [thread overview] Message-ID: <149875885755.10031.2960001252640880575.stgit@dwillia2-desk3.amr.corp.intel.com> (raw) In-Reply-To: <149875877608.10031.17813337234536358002.stgit@dwillia2-desk3.amr.corp.intel.com> The dax_flush() operation can be turned into a nop on platforms where firmware arranges for cpu caches to be flushed on a power-fail event. The ACPI 6.2 specification defines a mechanism for the platform to indicate this capability so the kernel can select the proper default. However, for other platforms, the administrator must toggle this setting manually. Given this flush setting is a dax-specific mechanism we advertise it through a 'dax' attribute group hanging off a host device. For example, a 'pmem0' block-device gets a 'dax' sysfs-subdirectory with a 'write_cache' attribute to control response to dax cache flush requests. This is similar to the 'queue/write_cache' attribute that appears under block devices. Cc: Jan Kara <jack@suse.cz> Cc: Jeff Moyer <jmoyer@redhat.com> Cc: Matthew Wilcox <mawilcox@microsoft.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Dan Williams <dan.j.williams@intel.com> --- drivers/dax/super.c | 79 +++++++++++++++++++++++++++++++++++++++++++++++++ drivers/nvdimm/pmem.c | 10 ++++++ include/linux/dax.h | 3 ++ 3 files changed, 92 insertions(+) diff --git a/drivers/dax/super.c b/drivers/dax/super.c index 8bf71195921b..4827251782a1 100644 --- a/drivers/dax/super.c +++ b/drivers/dax/super.c @@ -119,6 +119,8 @@ EXPORT_SYMBOL_GPL(__bdev_dax_supported); enum dax_device_flags { /* !alive + rcu grace period == no new operations / mappings */ DAXDEV_ALIVE, + /* gate whether dax_flush() calls the low level flush routine */ + DAXDEV_WRITE_CACHE, }; /** @@ -139,6 +141,71 @@ struct dax_device { const struct dax_operations *ops; }; +static ssize_t write_cache_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct dax_device *dax_dev = dax_get_by_host(dev_name(dev)); + ssize_t rc; + + WARN_ON_ONCE(!dax_dev); + if (!dax_dev) + return -ENXIO; + + rc = sprintf(buf, "%d\n", !!test_bit(DAXDEV_WRITE_CACHE, + &dax_dev->flags)); + put_dax(dax_dev); + return rc; +} + +static ssize_t write_cache_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t len) +{ + bool write_cache; + int rc = strtobool(buf, &write_cache); + struct dax_device *dax_dev = dax_get_by_host(dev_name(dev)); + + WARN_ON_ONCE(!dax_dev); + if (!dax_dev) + return -ENXIO; + + if (rc) + len = rc; + else if (write_cache) + set_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags); + else + clear_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags); + + put_dax(dax_dev); + return len; +} +static DEVICE_ATTR_RW(write_cache); + +static umode_t dax_visible(struct kobject *kobj, struct attribute *a, int n) +{ + struct device *dev = container_of(kobj, typeof(*dev), kobj); + struct dax_device *dax_dev = dax_get_by_host(dev_name(dev)); + + WARN_ON_ONCE(!dax_dev); + if (!dax_dev) + return 0; + + if (a == &dev_attr_write_cache.attr && !dax_dev->ops->flush) + return 0; + return a->mode; +} + +static struct attribute *dax_attributes[] = { + &dev_attr_write_cache.attr, + NULL, +}; + +struct attribute_group dax_attribute_group = { + .name = "dax", + .attrs = dax_attributes, + .is_visible = dax_visible, +}; +EXPORT_SYMBOL_GPL(dax_attribute_group); + /** * dax_direct_access() - translate a device pgoff to an absolute pfn * @dax_dev: a dax_device instance representing the logical memory range @@ -194,11 +261,23 @@ void dax_flush(struct dax_device *dax_dev, pgoff_t pgoff, void *addr, if (!dax_alive(dax_dev)) return; + if (!test_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags)) + return; + if (dax_dev->ops->flush) dax_dev->ops->flush(dax_dev, pgoff, addr, size); } EXPORT_SYMBOL_GPL(dax_flush); +void dax_write_cache(struct dax_device *dax_dev, bool wc) +{ + if (wc) + set_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags); + else + clear_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags); +} +EXPORT_SYMBOL_GPL(dax_write_cache); + bool dax_alive(struct dax_device *dax_dev) { lockdep_assert_held(&dax_srcu); diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 06f6c27ec1e9..7339d184070e 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -253,6 +253,11 @@ static const struct dax_operations pmem_dax_ops = { .flush = pmem_dax_flush, }; +static const struct attribute_group *pmem_attribute_groups[] = { + &dax_attribute_group, + NULL, +}; + static void pmem_release_queue(void *q) { blk_cleanup_queue(q); @@ -287,6 +292,7 @@ static int pmem_attach_disk(struct device *dev, struct pmem_device *pmem; struct resource pfn_res; struct request_queue *q; + struct device *gendev; struct gendisk *disk; void *addr; @@ -384,8 +390,12 @@ static int pmem_attach_disk(struct device *dev, put_disk(disk); return -ENOMEM; } + dax_write_cache(dax_dev, true); pmem->dax_dev = dax_dev; + gendev = disk_to_dev(disk); + gendev->groups = pmem_attribute_groups; + device_add_disk(dev, disk); if (devm_add_action_or_reset(dev, pmem_release_disk, pmem)) return -ENOMEM; diff --git a/include/linux/dax.h b/include/linux/dax.h index 73fca1bebaf3..8f39db7439c3 100644 --- a/include/linux/dax.h +++ b/include/linux/dax.h @@ -23,6 +23,8 @@ struct dax_operations { void (*flush)(struct dax_device *, pgoff_t, void *, size_t); }; +extern struct attribute_group dax_attribute_group; + #if IS_ENABLED(CONFIG_DAX) struct dax_device *dax_get_by_host(const char *host); void put_dax(struct dax_device *dax_dev); @@ -84,6 +86,7 @@ size_t dax_copy_from_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr, size_t bytes, struct iov_iter *i); void dax_flush(struct dax_device *dax_dev, pgoff_t pgoff, void *addr, size_t size); +void dax_write_cache(struct dax_device *dax_dev, bool wc); /* * We use lowest available bit in exceptional entry for locking, one bit for _______________________________________________ Linux-nvdimm mailing list Linux-nvdimm@lists.01.org https://lists.01.org/mailman/listinfo/linux-nvdimm
WARNING: multiple messages have this Message-ID (diff)
From: Dan Williams <dan.j.williams@intel.com> To: linux-nvdimm@lists.01.org Cc: Jan Kara <jack@suse.cz>, Matthew Wilcox <mawilcox@microsoft.com>, x86@kernel.org, linux-kernel@vger.kernel.org, Jeff Moyer <jmoyer@redhat.com>, viro@zeniv.linux.org.uk, linux-fsdevel@vger.kernel.org, Ross Zwisler <ross.zwisler@linux.intel.com>, hch@lst.de Subject: [PATCH v4 15/16] libnvdimm, pmem, dax: export a cache control attribute Date: Thu, 29 Jun 2017 10:54:17 -0700 [thread overview] Message-ID: <149875885755.10031.2960001252640880575.stgit@dwillia2-desk3.amr.corp.intel.com> (raw) In-Reply-To: <149875877608.10031.17813337234536358002.stgit@dwillia2-desk3.amr.corp.intel.com> The dax_flush() operation can be turned into a nop on platforms where firmware arranges for cpu caches to be flushed on a power-fail event. The ACPI 6.2 specification defines a mechanism for the platform to indicate this capability so the kernel can select the proper default. However, for other platforms, the administrator must toggle this setting manually. Given this flush setting is a dax-specific mechanism we advertise it through a 'dax' attribute group hanging off a host device. For example, a 'pmem0' block-device gets a 'dax' sysfs-subdirectory with a 'write_cache' attribute to control response to dax cache flush requests. This is similar to the 'queue/write_cache' attribute that appears under block devices. Cc: Jan Kara <jack@suse.cz> Cc: Jeff Moyer <jmoyer@redhat.com> Cc: Matthew Wilcox <mawilcox@microsoft.com> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Suggested-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Dan Williams <dan.j.williams@intel.com> --- drivers/dax/super.c | 79 +++++++++++++++++++++++++++++++++++++++++++++++++ drivers/nvdimm/pmem.c | 10 ++++++ include/linux/dax.h | 3 ++ 3 files changed, 92 insertions(+) diff --git a/drivers/dax/super.c b/drivers/dax/super.c index 8bf71195921b..4827251782a1 100644 --- a/drivers/dax/super.c +++ b/drivers/dax/super.c @@ -119,6 +119,8 @@ EXPORT_SYMBOL_GPL(__bdev_dax_supported); enum dax_device_flags { /* !alive + rcu grace period == no new operations / mappings */ DAXDEV_ALIVE, + /* gate whether dax_flush() calls the low level flush routine */ + DAXDEV_WRITE_CACHE, }; /** @@ -139,6 +141,71 @@ struct dax_device { const struct dax_operations *ops; }; +static ssize_t write_cache_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct dax_device *dax_dev = dax_get_by_host(dev_name(dev)); + ssize_t rc; + + WARN_ON_ONCE(!dax_dev); + if (!dax_dev) + return -ENXIO; + + rc = sprintf(buf, "%d\n", !!test_bit(DAXDEV_WRITE_CACHE, + &dax_dev->flags)); + put_dax(dax_dev); + return rc; +} + +static ssize_t write_cache_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t len) +{ + bool write_cache; + int rc = strtobool(buf, &write_cache); + struct dax_device *dax_dev = dax_get_by_host(dev_name(dev)); + + WARN_ON_ONCE(!dax_dev); + if (!dax_dev) + return -ENXIO; + + if (rc) + len = rc; + else if (write_cache) + set_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags); + else + clear_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags); + + put_dax(dax_dev); + return len; +} +static DEVICE_ATTR_RW(write_cache); + +static umode_t dax_visible(struct kobject *kobj, struct attribute *a, int n) +{ + struct device *dev = container_of(kobj, typeof(*dev), kobj); + struct dax_device *dax_dev = dax_get_by_host(dev_name(dev)); + + WARN_ON_ONCE(!dax_dev); + if (!dax_dev) + return 0; + + if (a == &dev_attr_write_cache.attr && !dax_dev->ops->flush) + return 0; + return a->mode; +} + +static struct attribute *dax_attributes[] = { + &dev_attr_write_cache.attr, + NULL, +}; + +struct attribute_group dax_attribute_group = { + .name = "dax", + .attrs = dax_attributes, + .is_visible = dax_visible, +}; +EXPORT_SYMBOL_GPL(dax_attribute_group); + /** * dax_direct_access() - translate a device pgoff to an absolute pfn * @dax_dev: a dax_device instance representing the logical memory range @@ -194,11 +261,23 @@ void dax_flush(struct dax_device *dax_dev, pgoff_t pgoff, void *addr, if (!dax_alive(dax_dev)) return; + if (!test_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags)) + return; + if (dax_dev->ops->flush) dax_dev->ops->flush(dax_dev, pgoff, addr, size); } EXPORT_SYMBOL_GPL(dax_flush); +void dax_write_cache(struct dax_device *dax_dev, bool wc) +{ + if (wc) + set_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags); + else + clear_bit(DAXDEV_WRITE_CACHE, &dax_dev->flags); +} +EXPORT_SYMBOL_GPL(dax_write_cache); + bool dax_alive(struct dax_device *dax_dev) { lockdep_assert_held(&dax_srcu); diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c index 06f6c27ec1e9..7339d184070e 100644 --- a/drivers/nvdimm/pmem.c +++ b/drivers/nvdimm/pmem.c @@ -253,6 +253,11 @@ static const struct dax_operations pmem_dax_ops = { .flush = pmem_dax_flush, }; +static const struct attribute_group *pmem_attribute_groups[] = { + &dax_attribute_group, + NULL, +}; + static void pmem_release_queue(void *q) { blk_cleanup_queue(q); @@ -287,6 +292,7 @@ static int pmem_attach_disk(struct device *dev, struct pmem_device *pmem; struct resource pfn_res; struct request_queue *q; + struct device *gendev; struct gendisk *disk; void *addr; @@ -384,8 +390,12 @@ static int pmem_attach_disk(struct device *dev, put_disk(disk); return -ENOMEM; } + dax_write_cache(dax_dev, true); pmem->dax_dev = dax_dev; + gendev = disk_to_dev(disk); + gendev->groups = pmem_attribute_groups; + device_add_disk(dev, disk); if (devm_add_action_or_reset(dev, pmem_release_disk, pmem)) return -ENOMEM; diff --git a/include/linux/dax.h b/include/linux/dax.h index 73fca1bebaf3..8f39db7439c3 100644 --- a/include/linux/dax.h +++ b/include/linux/dax.h @@ -23,6 +23,8 @@ struct dax_operations { void (*flush)(struct dax_device *, pgoff_t, void *, size_t); }; +extern struct attribute_group dax_attribute_group; + #if IS_ENABLED(CONFIG_DAX) struct dax_device *dax_get_by_host(const char *host); void put_dax(struct dax_device *dax_dev); @@ -84,6 +86,7 @@ size_t dax_copy_from_iter(struct dax_device *dax_dev, pgoff_t pgoff, void *addr, size_t bytes, struct iov_iter *i); void dax_flush(struct dax_device *dax_dev, pgoff_t pgoff, void *addr, size_t size); +void dax_write_cache(struct dax_device *dax_dev, bool wc); /* * We use lowest available bit in exceptional entry for locking, one bit for
next prev parent reply other threads:[~2017-06-29 17:59 UTC|newest] Thread overview: 72+ messages / expand[flat|nested] mbox.gz Atom feed top 2017-06-29 17:52 [PATCH v4 00/16] pmem: stop abusing copy_user_nocache(), and other reworks Dan Williams 2017-06-29 17:52 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 01/16] x86, uaccess: introduce copy_from_iter_flushcache for pmem / cache-bypass operations Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 02/16] dm: add ->copy_from_iter() dax operation support Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 03/16] filesystem-dax: convert to dax_copy_from_iter() Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 04/16] dax, pmem: introduce an optional 'flush' dax_operation Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 05/16] dm: add ->flush() dax operation support Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 06/16] filesystem-dax: convert to dax_flush() Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 07/16] x86, dax: replace clear_pmem() with open coded memset + dax_ops->flush Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 08/16] x86, dax, libnvdimm: remove wb_cache_pmem() indirection Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 09/16] x86, libnvdimm, pmem: move arch_invalidate_pmem() to libnvdimm Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 10/16] x86, libnvdimm, pmem: remove global pmem api Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:53 ` [PATCH v4 11/16] libnvdimm, pmem: fix persistence warning Dan Williams 2017-06-29 17:53 ` Dan Williams 2017-06-29 17:54 ` [PATCH v4 12/16] libnvdimm, nfit: enable support for volatile ranges Dan Williams 2017-06-29 17:54 ` Dan Williams 2017-06-29 19:20 ` Linda Knippers 2017-06-29 19:20 ` Linda Knippers 2017-06-29 20:42 ` Dan Williams 2017-06-29 20:42 ` Dan Williams 2017-06-29 21:16 ` Linda Knippers 2017-06-29 21:16 ` Linda Knippers 2017-06-29 21:50 ` Dan Williams 2017-06-29 21:50 ` Dan Williams 2017-06-29 22:12 ` Linda Knippers 2017-06-29 22:12 ` Linda Knippers 2017-06-29 22:28 ` Dan Williams 2017-06-29 22:28 ` Dan Williams 2017-06-29 22:35 ` Linda Knippers 2017-06-29 22:35 ` Linda Knippers 2017-06-29 22:43 ` Dan Williams 2017-06-29 22:43 ` Dan Williams 2017-06-29 22:49 ` Linda Knippers 2017-06-29 22:49 ` Linda Knippers 2017-06-29 22:58 ` Dan Williams 2017-06-29 22:58 ` Dan Williams 2017-06-29 23:14 ` Linda Knippers 2017-06-29 23:14 ` Linda Knippers 2017-06-30 1:28 ` Dan Williams 2017-06-30 1:28 ` Dan Williams 2017-07-05 23:46 ` Kani, Toshimitsu 2017-07-05 23:46 ` Kani, Toshimitsu 2017-07-06 0:07 ` Dan Williams 2017-07-06 0:07 ` Dan Williams 2017-07-06 1:17 ` Kani, Toshimitsu 2017-07-06 1:17 ` Kani, Toshimitsu 2017-07-06 2:08 ` Dan Williams 2017-07-06 2:08 ` Dan Williams 2017-07-06 2:11 ` hch 2017-07-06 2:11 ` hch 2017-07-06 2:53 ` Oliver 2017-07-06 2:53 ` Oliver 2017-07-06 2:56 ` hch 2017-07-06 2:56 ` hch 2017-06-29 17:54 ` [PATCH v4 13/16] dax: remove default copy_from_iter fallback Dan Williams 2017-06-29 17:54 ` Dan Williams 2017-06-29 17:54 ` [PATCH v4 14/16] dax: convert to bitmask for flags Dan Williams 2017-06-29 17:54 ` Dan Williams 2017-06-29 17:54 ` Dan Williams [this message] 2017-06-29 17:54 ` [PATCH v4 15/16] libnvdimm, pmem, dax: export a cache control attribute Dan Williams 2017-06-29 17:54 ` [PATCH v4 16/16] libnvdimm, pmem: disable dax flushing when pmem is fronting a volatile region Dan Williams 2017-06-29 17:54 ` Dan Williams
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=149875885755.10031.2960001252640880575.stgit@dwillia2-desk3.amr.corp.intel.com \ --to=dan.j.williams@intel.com \ --cc=hch@lst.de \ --cc=jack@suse.cz \ --cc=linux-fsdevel@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-nvdimm@lists.01.org \ --cc=mawilcox@microsoft.com \ --cc=viro@zeniv.linux.org.uk \ --cc=x86@kernel.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.