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
ncsi-aen.c
/*
* Copyright Gavin Shan, IBM Corporation 2016.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*/
#include <linux/module.h>
#include <linux/kernel.h>
#include <linux/init.h>
#include <linux/netdevice.h>
#include <linux/skbuff.h>
#include <net/ncsi.h>
#include <net/net_namespace.h>
#include <net/sock.h>
#include "internal.h"
#include "ncsi-pkt.h"
static int ncsi_validate_aen_pkt(struct ncsi_aen_pkt_hdr *h,
const unsigned short payload)
{
u32 checksum;
__be32 *pchecksum;
if (h->common.revision != NCSI_PKT_REVISION)
return -EINVAL;
if (ntohs(h->common.length) != payload)
return -EINVAL;
/* Validate checksum, which might be zeroes if the
* sender doesn't support checksum according to NCSI
* specification.
*/
pchecksum = (__be32 *)((void *)(h + 1) + payload - 4);
if (ntohl(*pchecksum) == 0)
return 0;
checksum = ncsi_calculate_checksum((unsigned char *)h,
sizeof(*h) + payload - 4);
if (*pchecksum != htonl(checksum))
return -EINVAL;
return 0;
}
static int ncsi_aen_handler_lsc(struct ncsi_dev_priv *ndp,
struct ncsi_aen_pkt_hdr *h)
{
struct ncsi_aen_lsc_pkt *lsc;
struct ncsi_channel *nc;
struct ncsi_channel_mode *ncm;
bool chained;
int state;
unsigned long old_data, data;
unsigned long flags;
/* Find the NCSI channel */
ncsi_find_package_and_channel(ndp, h->common.channel, NULL, &nc);
if (!nc)
return -ENODEV;
/* Update the link status */
lsc = (struct ncsi_aen_lsc_pkt *)h;
spin_lock_irqsave(&nc->lock, flags);
ncm = &nc->modes[NCSI_MODE_LINK];
old_data = ncm->data[2];
data = ntohl(lsc->status);
ncm->data[2] = data;
ncm->data[4] = ntohl(lsc->oem_status);
netdev_dbg(ndp->ndev.dev, "NCSI: LSC AEN - channel %u state %s\n",
nc->id, data & 0x1 ? "up" : "down");
chained = !list_empty(&nc->link);
state = nc->state;
spin_unlock_irqrestore(&nc->lock, flags);
if (!((old_data ^ data) & 0x1) || chained)
return 0;
if (!(state == NCSI_CHANNEL_INACTIVE && (data & 0x1)) &&
!(state == NCSI_CHANNEL_ACTIVE && !(data & 0x1)))
return 0;
if (!(ndp->flags & NCSI_DEV_HWA) &&
state == NCSI_CHANNEL_ACTIVE)
ndp->flags |= NCSI_DEV_RESHUFFLE;
ncsi_stop_channel_monitor(nc);
spin_lock_irqsave(&ndp->lock, flags);
list_add_tail_rcu(&nc->link, &ndp->channel_queue);
spin_unlock_irqrestore(&ndp->lock, flags);
return ncsi_process_next_channel(ndp);
}
static int ncsi_aen_handler_cr(struct ncsi_dev_priv *ndp,
struct ncsi_aen_pkt_hdr *h)
{
struct ncsi_channel *nc;
unsigned long flags;
/* Find the NCSI channel */
ncsi_find_package_and_channel(ndp, h->common.channel, NULL, &nc);
if (!nc)
return -ENODEV;
spin_lock_irqsave(&nc->lock, flags);
if (!list_empty(&nc->link) ||
nc->state != NCSI_CHANNEL_ACTIVE) {
spin_unlock_irqrestore(&nc->lock, flags);
return 0;
}
spin_unlock_irqrestore(&nc->lock, flags);
ncsi_stop_channel_monitor(nc);
spin_lock_irqsave(&nc->lock, flags);
nc->state = NCSI_CHANNEL_INVISIBLE;
spin_unlock_irqrestore(&nc->lock, flags);
spin_lock_irqsave(&ndp->lock, flags);
nc->state = NCSI_CHANNEL_INACTIVE;
list_add_tail_rcu(&nc->link, &ndp->channel_queue);
spin_unlock_irqrestore(&ndp->lock, flags);
return ncsi_process_next_channel(ndp);
}
static int ncsi_aen_handler_hncdsc(struct ncsi_dev_priv *ndp,
struct ncsi_aen_pkt_hdr *h)
{
struct ncsi_channel *nc;
struct ncsi_channel_mode *ncm;
struct ncsi_aen_hncdsc_pkt *hncdsc;
unsigned long flags;
/* Find the NCSI channel */
ncsi_find_package_and_channel(ndp, h->common.channel, NULL, &nc);
if (!nc)
return -ENODEV;
spin_lock_irqsave(&nc->lock, flags);
ncm = &nc->modes[NCSI_MODE_LINK];
hncdsc = (struct ncsi_aen_hncdsc_pkt *)h;
ncm->data[3] = ntohl(hncdsc->status);
spin_unlock_irqrestore(&nc->lock, flags);
netdev_dbg(ndp->ndev.dev,
"NCSI: host driver %srunning on channel %u\n",
ncm->data[3] & 0x1 ? "" : "not ", nc->id);
return 0;
}
static struct ncsi_aen_handler {
unsigned char type;
int payload;
int (*handler)(struct ncsi_dev_priv *ndp,
struct ncsi_aen_pkt_hdr *h);
} ncsi_aen_handlers[] = {
{ NCSI_PKT_AEN_LSC, 12, ncsi_aen_handler_lsc },
{ NCSI_PKT_AEN_CR, 4, ncsi_aen_handler_cr },
{ NCSI_PKT_AEN_HNCDSC, 8, ncsi_aen_handler_hncdsc }
};
int ncsi_aen_handler(struct ncsi_dev_priv *ndp, struct sk_buff *skb)
{
struct ncsi_aen_pkt_hdr *h;
struct ncsi_aen_handler *nah = NULL;
int i, ret;
/* Find the handler */
h = (struct ncsi_aen_pkt_hdr *)skb_network_header(skb);
for (i = 0; i < ARRAY_SIZE(ncsi_aen_handlers); i++) {
if (ncsi_aen_handlers[i].type == h->type) {
nah = &ncsi_aen_handlers[i];
break;
}
}
if (!nah) {
netdev_warn(ndp->ndev.dev, "Invalid AEN (0x%x) received\n",
h->type);
return -ENOENT;
}
ret = ncsi_validate_aen_pkt(h, nah->payload);
if (ret) {
netdev_warn(ndp->ndev.dev,
"NCSI: 'bad' packet ignored for AEN type 0x%x\n",
h->type);
goto out;
}
ret = nah->handler(ndp, h);
if (ret)
netdev_err(ndp->ndev.dev,
"NCSI: Handler for AEN type 0x%x returned %d\n",
h->type, ret);
out:
consume_skb(skb);
return ret;
}
![swh spinner](/static/img/swh-spinner.gif)
Computing file changes ...