https://github.com/torvalds/linux
Revision 9230a0b65b47fe6856c4468ec0175c4987e5bede authored by Dave Chinner on 20 November 2018, 06:50:08 UTC, committed by Darrick J. Wong on 21 November 2018, 18:10:53 UTC
Long saga. There have been days spent following this through dead end
after dead end in multi-GB event traces. This morning, after writing
a trace-cmd wrapper that enabled me to be more selective about XFS
trace points, I discovered that I could get just enough essential
tracepoints enabled that there was a 50:50 chance the fsx config
would fail at ~115k ops. If it didn't fail at op 115547, I stopped
fsx at op 115548 anyway.

That gave me two traces - one where the problem manifested, and one
where it didn't. After refining the traces to have the necessary
information, I found that in the failing case there was a real
extent in the COW fork compared to an unwritten extent in the
working case.

Walking back through the two traces to the point where the CWO fork
extents actually diverged, I found that the bad case had an extra
unwritten extent in it. This is likely because the bug it led me to
had triggered multiple times in those 115k ops, leaving stray
COW extents around. What I saw was a COW delalloc conversion to an
unwritten extent (as they should always be through
xfs_iomap_write_allocate()) resulted in a /written extent/:

xfs_writepage:        dev 259:0 ino 0x83 pgoff 0x17000 size 0x79a00 offset 0 length 0
xfs_iext_remove:      dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/2 offset 32 block 152 count 20 flag 1 caller xfs_bmap_add_extent_delay_real
xfs_bmap_pre_update:  dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/1 offset 1 block 4503599627239429 count 31 flag 0 caller xfs_bmap_add_extent_delay_real
xfs_bmap_post_update: dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/1 offset 1 block 121 count 51 flag 0 caller xfs_bmap_add_ex

Basically, Cow fork before:

	0 1            32          52
	+H+DDDDDDDDDDDD+UUUUUUUUUUU+
	   PREV		RIGHT

COW delalloc conversion allocates:

	  1	       32
	  +uuuuuuuuuuuu+
	  NEW

And the result according to the xfs_bmap_post_update trace was:

	0 1            32          52
	+H+wwwwwwwwwwwwwwwwwwwwwwww+
	   PREV

Which is clearly wrong - it should be a merged unwritten extent,
not an unwritten extent.

That lead me to look at the LEFT_FILLING|RIGHT_FILLING|RIGHT_CONTIG
case in xfs_bmap_add_extent_delay_real(), and sure enough, there's
the bug.

It takes the old delalloc extent (PREV) and adds the length of the
RIGHT extent to it, takes the start block from NEW, removes the
RIGHT extent and then updates PREV with the new extent.

What it fails to do is update PREV.br_state. For delalloc, this is
always XFS_EXT_NORM, while in this case we are converting the
delayed allocation to unwritten, so it needs to be updated to
XFS_EXT_UNWRITTEN. This LF|RF|RC case does not do this, and so
the resultant extent is always written.

And that's the bug I've been chasing for a week - a bmap btree bug,
not a reflink/dedupe/copy_file_range bug, but a BMBT bug introduced
with the recent in core extent tree scalability enhancements.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
1 parent 2c30717
Raw File
Tip revision: 9230a0b65b47fe6856c4468ec0175c4987e5bede authored by Dave Chinner on 20 November 2018, 06:50:08 UTC
xfs: delalloc -> unwritten COW fork allocation can go wrong
Tip revision: 9230a0b
strncpy_from_user.c
// SPDX-License-Identifier: GPL-2.0
#include <linux/compiler.h>
#include <linux/export.h>
#include <linux/kasan-checks.h>
#include <linux/thread_info.h>
#include <linux/uaccess.h>
#include <linux/kernel.h>
#include <linux/errno.h>

#include <asm/byteorder.h>
#include <asm/word-at-a-time.h>

#ifdef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
#define IS_UNALIGNED(src, dst)	0
#else
#define IS_UNALIGNED(src, dst)	\
	(((long) dst | (long) src) & (sizeof(long) - 1))
#endif

/*
 * Do a strncpy, return length of string without final '\0'.
 * 'count' is the user-supplied count (return 'count' if we
 * hit it), 'max' is the address space maximum (and we return
 * -EFAULT if we hit it).
 */
static inline long do_strncpy_from_user(char *dst, const char __user *src, long count, unsigned long max)
{
	const struct word_at_a_time constants = WORD_AT_A_TIME_CONSTANTS;
	long res = 0;

	/*
	 * Truncate 'max' to the user-specified limit, so that
	 * we only have one limit we need to check in the loop
	 */
	if (max > count)
		max = count;

	if (IS_UNALIGNED(src, dst))
		goto byte_at_a_time;

	while (max >= sizeof(unsigned long)) {
		unsigned long c, data;

		/* Fall back to byte-at-a-time if we get a page fault */
		unsafe_get_user(c, (unsigned long __user *)(src+res), byte_at_a_time);

		*(unsigned long *)(dst+res) = c;
		if (has_zero(c, &data, &constants)) {
			data = prep_zero_mask(c, data, &constants);
			data = create_zero_mask(data);
			return res + find_zero(data);
		}
		res += sizeof(unsigned long);
		max -= sizeof(unsigned long);
	}

byte_at_a_time:
	while (max) {
		char c;

		unsafe_get_user(c,src+res, efault);
		dst[res] = c;
		if (!c)
			return res;
		res++;
		max--;
	}

	/*
	 * Uhhuh. We hit 'max'. But was that the user-specified maximum
	 * too? If so, that's ok - we got as much as the user asked for.
	 */
	if (res >= count)
		return res;

	/*
	 * Nope: we hit the address space limit, and we still had more
	 * characters the caller would have wanted. That's an EFAULT.
	 */
efault:
	return -EFAULT;
}

/**
 * strncpy_from_user: - Copy a NUL terminated string from userspace.
 * @dst:   Destination address, in kernel space.  This buffer must be at
 *         least @count bytes long.
 * @src:   Source address, in user space.
 * @count: Maximum number of bytes to copy, including the trailing NUL.
 *
 * Copies a NUL-terminated string from userspace to kernel space.
 *
 * On success, returns the length of the string (not including the trailing
 * NUL).
 *
 * If access to userspace fails, returns -EFAULT (some data may have been
 * copied).
 *
 * If @count is smaller than the length of the string, copies @count bytes
 * and returns @count.
 */
long strncpy_from_user(char *dst, const char __user *src, long count)
{
	unsigned long max_addr, src_addr;

	if (unlikely(count <= 0))
		return 0;

	max_addr = user_addr_max();
	src_addr = (unsigned long)src;
	if (likely(src_addr < max_addr)) {
		unsigned long max = max_addr - src_addr;
		long retval;

		kasan_check_write(dst, count);
		check_object_size(dst, count, false);
		user_access_begin();
		retval = do_strncpy_from_user(dst, src, count, max);
		user_access_end();
		return retval;
	}
	return -EFAULT;
}
EXPORT_SYMBOL(strncpy_from_user);
back to top