[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1453498573-6328-4-git-send-email-ross.zwisler@linux.intel.com>
Date: Fri, 22 Jan 2016 14:36:11 -0700
From: Ross Zwisler <ross.zwisler@...ux.intel.com>
To: linux-kernel@...r.kernel.org
Cc: Ross Zwisler <ross.zwisler@...ux.intel.com>,
Alexander Viro <viro@...iv.linux.org.uk>,
Andrew Morton <akpm@...ux-foundation.org>,
Dan Williams <dan.j.williams@...el.com>,
Dave Chinner <david@...morbit.com>, Jan Kara <jack@...e.com>,
Matthew Wilcox <willy@...ux.intel.com>,
linux-fsdevel@...r.kernel.org, linux-nvdimm@...ts.01.org
Subject: [PATCH v3 3/5] dax: improve documentation for fsync/msync
Several of the subtleties and assumptions of the DAX fsync/msync
implementation are not immediately obvious, so document them with comments.
Signed-off-by: Ross Zwisler <ross.zwisler@...ux.intel.com>
Reported-by: Jan Kara <jack@...e.cz>
Reviewed-by: Jan Kara <jack@...e.cz>
---
fs/dax.c | 31 +++++++++++++++++++++++++++++++
1 file changed, 31 insertions(+)
diff --git a/fs/dax.c b/fs/dax.c
index d589113..ab2faa9 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -335,6 +335,7 @@ static int dax_radix_entry(struct address_space *mapping, pgoff_t index,
int type, error = 0;
void *entry;
+ WARN_ON_ONCE(pmd_entry && !dirty);
__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
spin_lock_irq(&mapping->tree_lock);
@@ -350,6 +351,13 @@ static int dax_radix_entry(struct address_space *mapping, pgoff_t index,
if (!pmd_entry || type == RADIX_DAX_PMD)
goto dirty;
+
+ /*
+ * We only insert dirty PMD entries into the radix tree. This
+ * means we don't need to worry about removing a dirty PTE
+ * entry and inserting a clean PMD entry, thus reducing the
+ * range we would flush with a follow-up fsync/msync call.
+ */
radix_tree_delete(&mapping->page_tree, index);
mapping->nrexceptional--;
}
@@ -912,6 +920,21 @@ int __dax_pmd_fault(struct vm_area_struct *vma, unsigned long address,
}
dax_unmap_atomic(bdev, &dax);
+ /*
+ * For PTE faults we insert a radix tree entry for reads, and
+ * leave it clean. Then on the first write we dirty the radix
+ * tree entry via the dax_pfn_mkwrite() path. This sequence
+ * allows the dax_pfn_mkwrite() call to be simpler and avoid a
+ * call into get_block() to translate the pgoff to a sector in
+ * order to be able to create a new radix tree entry.
+ *
+ * The PMD path doesn't have an equivalent to
+ * dax_pfn_mkwrite(), though, so for a read followed by a
+ * write we traverse all the way through __dax_pmd_fault()
+ * twice. This means we can just skip inserting a radix tree
+ * entry completely on the initial read and just wait until
+ * the write to insert a dirty entry.
+ */
if (write) {
error = dax_radix_entry(mapping, pgoff, dax.sector,
true, true);
@@ -985,6 +1008,14 @@ int dax_pfn_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
{
struct file *file = vma->vm_file;
+ /*
+ * We pass NO_SECTOR to dax_radix_entry() because we expect that a
+ * RADIX_DAX_PTE entry already exists in the radix tree from a
+ * previous call to __dax_fault(). We just want to look up that PTE
+ * entry using vmf->pgoff and make sure the dirty tag is set. This
+ * saves us from having to make a call to get_block() here to look
+ * up the sector.
+ */
dax_radix_entry(file->f_mapping, vmf->pgoff, NO_SECTOR, false, true);
return VM_FAULT_NOPAGE;
}
--
2.5.0
Powered by blists - more mailing lists