* [Cluster-devel] [GFS2 Patch][TRY 2] GFS2: Add rgrp information to block_alloc trace point
[not found] <30ec7483-cf44-4105-9e7f-6b924f0c6ddb@zmail12.collab.prod.int.phx2.redhat.com>
@ 2012-05-09 16:11 ` Bob Peterson
2012-05-11 9:56 ` Steven Whitehouse
0 siblings, 1 reply; 2+ messages in thread
From: Bob Peterson @ 2012-05-09 16:11 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
This is a second attempt at a patch that adds rgrp information to the
block allocation trace point for GFS2. As suggested, the patch was
modified to list the rgrp information _after_ the fields that exist today.
Again, the reason for this patch is to allow us to trace and debug
problems with the block reservations patch, which is still in the works.
We can debug problems with reservations if we can see what block allocations
result from the block reservations. It may also be handy in figuring out
if there are problems in rgrp free space accounting. In other words,
we can use it to track the rgrp and its free space along side the allocations
that are taking place.
Regards,
Bob Peterson
Red Hat File Systems
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
---
Author: Bob Peterson <rpeterso@redhat.com>
Date: Thu Apr 12 08:32:45 2012 -0500
GFS2: Add rgrp information to block_alloc trace point
This patch adds resource group information to the block allocation
trace point for GFS2. This makes it easier to debug problems with
resource groups, such as management of the number of free blocks.
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 645c16f..f74fb9b 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -1556,7 +1556,7 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks,
ip->i_inode.i_gid);
rgd->rd_free_clone -= *nblocks;
- trace_gfs2_block_alloc(ip, block, *nblocks,
+ trace_gfs2_block_alloc(ip, rgd, block, *nblocks,
dinode ? GFS2_BLKST_DINODE : GFS2_BLKST_USED);
*bn = block;
return 0;
@@ -1583,7 +1583,7 @@ void __gfs2_free_blocks(struct gfs2_inode *ip, u64 bstart, u32 blen, int meta)
rgd = rgblk_free(sdp, bstart, blen, GFS2_BLKST_FREE);
if (!rgd)
return;
- trace_gfs2_block_alloc(ip, bstart, blen, GFS2_BLKST_FREE);
+ trace_gfs2_block_alloc(ip, rgd, bstart, blen, GFS2_BLKST_FREE);
rgd->rd_free += blen;
rgd->rd_flags &= ~GFS2_RGF_TRIMMED;
gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
@@ -1621,7 +1621,7 @@ void gfs2_unlink_di(struct inode *inode)
rgd = rgblk_free(sdp, blkno, 1, GFS2_BLKST_UNLINKED);
if (!rgd)
return;
- trace_gfs2_block_alloc(ip, blkno, 1, GFS2_BLKST_UNLINKED);
+ trace_gfs2_block_alloc(ip, rgd, blkno, 1, GFS2_BLKST_UNLINKED);
gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
}
@@ -1651,7 +1651,7 @@ static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno)
void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip)
{
gfs2_free_uninit_di(rgd, ip->i_no_addr);
- trace_gfs2_block_alloc(ip, ip->i_no_addr, 1, GFS2_BLKST_FREE);
+ trace_gfs2_block_alloc(ip, rgd, ip->i_no_addr, 1, GFS2_BLKST_FREE);
gfs2_quota_change(ip, -1, ip->i_inode.i_uid, ip->i_inode.i_gid);
gfs2_meta_wipe(ip, ip->i_no_addr, 1);
}
diff --git a/fs/gfs2/trace_gfs2.h b/fs/gfs2/trace_gfs2.h
index dfa89cd..1b8b815 100644
--- a/fs/gfs2/trace_gfs2.h
+++ b/fs/gfs2/trace_gfs2.h
@@ -457,10 +457,10 @@ TRACE_EVENT(gfs2_bmap,
/* Keep track of blocks as they are allocated/freed */
TRACE_EVENT(gfs2_block_alloc,
- TP_PROTO(const struct gfs2_inode *ip, u64 block, unsigned len,
- u8 block_state),
+ TP_PROTO(const struct gfs2_inode *ip, struct gfs2_rgrpd *rgd,
+ u64 block, unsigned len, u8 block_state),
- TP_ARGS(ip, block, len, block_state),
+ TP_ARGS(ip, rgd, block, len, block_state),
TP_STRUCT__entry(
__field( dev_t, dev )
@@ -468,6 +468,8 @@ TRACE_EVENT(gfs2_block_alloc,
__field( u64, inum )
__field( u32, len )
__field( u8, block_state )
+ __field( u64, rd_addr )
+ __field( u32, rd_free_clone )
),
TP_fast_assign(
@@ -476,14 +478,18 @@ TRACE_EVENT(gfs2_block_alloc,
__entry->inum = ip->i_no_addr;
__entry->len = len;
__entry->block_state = block_state;
+ __entry->rd_addr = rgd->rd_addr;
+ __entry->rd_free_clone = rgd->rd_free_clone;
),
- TP_printk("%u,%u bmap %llu alloc %llu/%lu %s",
+ TP_printk("%u,%u bmap %llu alloc %llu/%lu %s rg:%llu rf:%u",
MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long long)__entry->inum,
(unsigned long long)__entry->start,
(unsigned long)__entry->len,
- block_state_name(__entry->block_state))
+ block_state_name(__entry->block_state),
+ (unsigned long long)__entry->rd_addr,
+ __entry->rd_free_clone)
);
#endif /* _TRACE_GFS2_H */
^ permalink raw reply related [flat|nested] 2+ messages in thread
* [Cluster-devel] [GFS2 Patch][TRY 2] GFS2: Add rgrp information to block_alloc trace point
2012-05-09 16:11 ` [Cluster-devel] [GFS2 Patch][TRY 2] GFS2: Add rgrp information to block_alloc trace point Bob Peterson
@ 2012-05-11 9:56 ` Steven Whitehouse
0 siblings, 0 replies; 2+ messages in thread
From: Steven Whitehouse @ 2012-05-11 9:56 UTC (permalink / raw)
To: cluster-devel.redhat.com
Hi,
Now in the -nmw git tree. Thanks,
Steve.
On Wed, 2012-05-09 at 12:11 -0400, Bob Peterson wrote:
> Hi,
>
> This is a second attempt at a patch that adds rgrp information to the
> block allocation trace point for GFS2. As suggested, the patch was
> modified to list the rgrp information _after_ the fields that exist today.
>
> Again, the reason for this patch is to allow us to trace and debug
> problems with the block reservations patch, which is still in the works.
> We can debug problems with reservations if we can see what block allocations
> result from the block reservations. It may also be handy in figuring out
> if there are problems in rgrp free space accounting. In other words,
> we can use it to track the rgrp and its free space along side the allocations
> that are taking place.
>
> Regards,
>
> Bob Peterson
> Red Hat File Systems
>
> Signed-off-by: Bob Peterson <rpeterso@redhat.com>
> ---
> Author: Bob Peterson <rpeterso@redhat.com>
> Date: Thu Apr 12 08:32:45 2012 -0500
>
> GFS2: Add rgrp information to block_alloc trace point
>
> This patch adds resource group information to the block allocation
> trace point for GFS2. This makes it easier to debug problems with
> resource groups, such as management of the number of free blocks.
>
> diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
> index 645c16f..f74fb9b 100644
> --- a/fs/gfs2/rgrp.c
> +++ b/fs/gfs2/rgrp.c
> @@ -1556,7 +1556,7 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks,
> ip->i_inode.i_gid);
>
> rgd->rd_free_clone -= *nblocks;
> - trace_gfs2_block_alloc(ip, block, *nblocks,
> + trace_gfs2_block_alloc(ip, rgd, block, *nblocks,
> dinode ? GFS2_BLKST_DINODE : GFS2_BLKST_USED);
> *bn = block;
> return 0;
> @@ -1583,7 +1583,7 @@ void __gfs2_free_blocks(struct gfs2_inode *ip, u64 bstart, u32 blen, int meta)
> rgd = rgblk_free(sdp, bstart, blen, GFS2_BLKST_FREE);
> if (!rgd)
> return;
> - trace_gfs2_block_alloc(ip, bstart, blen, GFS2_BLKST_FREE);
> + trace_gfs2_block_alloc(ip, rgd, bstart, blen, GFS2_BLKST_FREE);
> rgd->rd_free += blen;
> rgd->rd_flags &= ~GFS2_RGF_TRIMMED;
> gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
> @@ -1621,7 +1621,7 @@ void gfs2_unlink_di(struct inode *inode)
> rgd = rgblk_free(sdp, blkno, 1, GFS2_BLKST_UNLINKED);
> if (!rgd)
> return;
> - trace_gfs2_block_alloc(ip, blkno, 1, GFS2_BLKST_UNLINKED);
> + trace_gfs2_block_alloc(ip, rgd, blkno, 1, GFS2_BLKST_UNLINKED);
> gfs2_trans_add_bh(rgd->rd_gl, rgd->rd_bits[0].bi_bh, 1);
> gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data);
> }
> @@ -1651,7 +1651,7 @@ static void gfs2_free_uninit_di(struct gfs2_rgrpd *rgd, u64 blkno)
> void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip)
> {
> gfs2_free_uninit_di(rgd, ip->i_no_addr);
> - trace_gfs2_block_alloc(ip, ip->i_no_addr, 1, GFS2_BLKST_FREE);
> + trace_gfs2_block_alloc(ip, rgd, ip->i_no_addr, 1, GFS2_BLKST_FREE);
> gfs2_quota_change(ip, -1, ip->i_inode.i_uid, ip->i_inode.i_gid);
> gfs2_meta_wipe(ip, ip->i_no_addr, 1);
> }
> diff --git a/fs/gfs2/trace_gfs2.h b/fs/gfs2/trace_gfs2.h
> index dfa89cd..1b8b815 100644
> --- a/fs/gfs2/trace_gfs2.h
> +++ b/fs/gfs2/trace_gfs2.h
> @@ -457,10 +457,10 @@ TRACE_EVENT(gfs2_bmap,
> /* Keep track of blocks as they are allocated/freed */
> TRACE_EVENT(gfs2_block_alloc,
>
> - TP_PROTO(const struct gfs2_inode *ip, u64 block, unsigned len,
> - u8 block_state),
> + TP_PROTO(const struct gfs2_inode *ip, struct gfs2_rgrpd *rgd,
> + u64 block, unsigned len, u8 block_state),
>
> - TP_ARGS(ip, block, len, block_state),
> + TP_ARGS(ip, rgd, block, len, block_state),
>
> TP_STRUCT__entry(
> __field( dev_t, dev )
> @@ -468,6 +468,8 @@ TRACE_EVENT(gfs2_block_alloc,
> __field( u64, inum )
> __field( u32, len )
> __field( u8, block_state )
> + __field( u64, rd_addr )
> + __field( u32, rd_free_clone )
> ),
>
> TP_fast_assign(
> @@ -476,14 +478,18 @@ TRACE_EVENT(gfs2_block_alloc,
> __entry->inum = ip->i_no_addr;
> __entry->len = len;
> __entry->block_state = block_state;
> + __entry->rd_addr = rgd->rd_addr;
> + __entry->rd_free_clone = rgd->rd_free_clone;
> ),
>
> - TP_printk("%u,%u bmap %llu alloc %llu/%lu %s",
> + TP_printk("%u,%u bmap %llu alloc %llu/%lu %s rg:%llu rf:%u",
> MAJOR(__entry->dev), MINOR(__entry->dev),
> (unsigned long long)__entry->inum,
> (unsigned long long)__entry->start,
> (unsigned long)__entry->len,
> - block_state_name(__entry->block_state))
> + block_state_name(__entry->block_state),
> + (unsigned long long)__entry->rd_addr,
> + __entry->rd_free_clone)
> );
>
> #endif /* _TRACE_GFS2_H */
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2012-05-11 9:56 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <30ec7483-cf44-4105-9e7f-6b924f0c6ddb@zmail12.collab.prod.int.phx2.redhat.com>
2012-05-09 16:11 ` [Cluster-devel] [GFS2 Patch][TRY 2] GFS2: Add rgrp information to block_alloc trace point Bob Peterson
2012-05-11 9:56 ` Steven Whitehouse
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).