提交 9f4736fe 编写于 作者: D Dan Williams

block: revert runtime dax control of the raw block device

Dynamically enabling DAX requires that the page cache first be flushed
and invalidated.  This must occur atomically with the change of DAX mode
otherwise we confuse the fsync/msync tracking and violate data
durability guarantees.  Eliminate the possibilty of DAX-disabled to
DAX-enabled transitions for now and revisit this for the next cycle.

Cc: Jan Kara <jack@suse.com>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: NDan Williams <dan.j.williams@intel.com>
上级 65f87ee7
......@@ -434,42 +434,6 @@ bool blkdev_dax_capable(struct block_device *bdev)
return true;
}
static int blkdev_daxset(struct block_device *bdev, unsigned long argp)
{
unsigned long arg;
int rc = 0;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
if (get_user(arg, (int __user *)(argp)))
return -EFAULT;
arg = !!arg;
if (arg == !!(bdev->bd_inode->i_flags & S_DAX))
return 0;
if (arg)
arg = S_DAX;
if (arg && !blkdev_dax_capable(bdev))
return -ENOTTY;
inode_lock(bdev->bd_inode);
if (bdev->bd_map_count == 0)
inode_set_flags(bdev->bd_inode, arg, S_DAX);
else
rc = -EBUSY;
inode_unlock(bdev->bd_inode);
return rc;
}
#else
static int blkdev_daxset(struct block_device *bdev, int arg)
{
if (arg)
return -ENOTTY;
return 0;
}
#endif
static int blkdev_flushbuf(struct block_device *bdev, fmode_t mode,
......@@ -634,8 +598,6 @@ int blkdev_ioctl(struct block_device *bdev, fmode_t mode, unsigned cmd,
case BLKTRACESETUP:
case BLKTRACETEARDOWN:
return blk_trace_ioctl(bdev, cmd, argp);
case BLKDAXSET:
return blkdev_daxset(bdev, arg);
case BLKDAXGET:
return put_int(arg, !!(bdev->bd_inode->i_flags & S_DAX));
break;
......
......@@ -1736,37 +1736,13 @@ static int blkdev_dax_pmd_fault(struct vm_area_struct *vma, unsigned long addr,
return __dax_pmd_fault(vma, addr, pmd, flags, blkdev_get_block, NULL);
}
static void blkdev_vm_open(struct vm_area_struct *vma)
{
struct inode *bd_inode = bdev_file_inode(vma->vm_file);
struct block_device *bdev = I_BDEV(bd_inode);
inode_lock(bd_inode);
bdev->bd_map_count++;
inode_unlock(bd_inode);
}
static void blkdev_vm_close(struct vm_area_struct *vma)
{
struct inode *bd_inode = bdev_file_inode(vma->vm_file);
struct block_device *bdev = I_BDEV(bd_inode);
inode_lock(bd_inode);
bdev->bd_map_count--;
inode_unlock(bd_inode);
}
static const struct vm_operations_struct blkdev_dax_vm_ops = {
.open = blkdev_vm_open,
.close = blkdev_vm_close,
.fault = blkdev_dax_fault,
.pmd_fault = blkdev_dax_pmd_fault,
.pfn_mkwrite = blkdev_dax_fault,
};
static const struct vm_operations_struct blkdev_default_vm_ops = {
.open = blkdev_vm_open,
.close = blkdev_vm_close,
.fault = filemap_fault,
.map_pages = filemap_map_pages,
};
......@@ -1774,18 +1750,14 @@ static const struct vm_operations_struct blkdev_default_vm_ops = {
static int blkdev_mmap(struct file *file, struct vm_area_struct *vma)
{
struct inode *bd_inode = bdev_file_inode(file);
struct block_device *bdev = I_BDEV(bd_inode);
file_accessed(file);
inode_lock(bd_inode);
bdev->bd_map_count++;
if (IS_DAX(bd_inode)) {
vma->vm_ops = &blkdev_dax_vm_ops;
vma->vm_flags |= VM_MIXEDMAP | VM_HUGEPAGE;
} else {
vma->vm_ops = &blkdev_default_vm_ops;
}
inode_unlock(bd_inode);
return 0;
}
......
......@@ -484,9 +484,6 @@ struct block_device {
int bd_fsfreeze_count;
/* Mutex for freeze */
struct mutex bd_fsfreeze_mutex;
#ifdef CONFIG_FS_DAX
int bd_map_count;
#endif
};
/*
......
......@@ -222,7 +222,6 @@ struct fsxattr {
#define BLKSECDISCARD _IO(0x12,125)
#define BLKROTATIONAL _IO(0x12,126)
#define BLKZEROOUT _IO(0x12,127)
#define BLKDAXSET _IO(0x12,128)
#define BLKDAXGET _IO(0x12,129)
#define BMAP_IOCTL 1 /* obsolete - kept for compatibility */
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册