On Tue, Nov 08, 2016 at 06:27:12AM -0500, Sanidhya Solanki wrote:
> On Tue, 8 Nov 2016 10:20:43 +0800
> Qu Wenruo <quwenruo@xxxxxxxxxxxxxx> wrote:
>
> > Introduce the following trace points:
> > qgroup_update_reserve
> > qgroup_meta_reserve
> >
> > These trace points are handy to trace qgroup reserve space related
> > problems.
> >
> > Signed-off-by: Qu Wenruo <quwenruo@xxxxxxxxxxxxxx>
> > ---
> > Although I hope these trace points will not be used again.
> >
> > v2:
> > None
> > v3:
> > Separate from trace point timing modification patch.
> > ---
> > fs/btrfs/qgroup.c | 15 +++++++++++++++
> > include/trace/events/btrfs.h | 43 +++++++++++++++++++++++++++++++++++++++++++
> > 2 files changed, 58 insertions(+)
> >
> > diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c
> > index fc0c64e..c00f962 100644
> > --- a/fs/btrfs/qgroup.c
> > +++ b/fs/btrfs/qgroup.c
> > @@ -1068,6 +1068,8 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info,
> > qgroup->excl += sign * num_bytes;
> > qgroup->excl_cmpr += sign * num_bytes;
> > if (sign > 0) {
> > + trace_qgroup_update_reserve(fs_info, qgroup->qgroupid,
> > + qgroup->reserved, (s64)-num_bytes);
> > if (WARN_ON(qgroup->reserved < num_bytes))
> > report_reserved_underflow(fs_info, qgroup,
> > num_bytes);
> > @@ -1094,6 +1096,9 @@ static int __qgroup_excl_accounting(struct btrfs_fs_info *fs_info,
> > WARN_ON(sign < 0 && qgroup->excl < num_bytes);
> > qgroup->excl += sign * num_bytes;
> > if (sign > 0) {
> > + trace_qgroup_update_reserve(fs_info, qgroup->qgroupid,
> > + qgroup->reserved,
> > + (s64)-num_bytes);
> > if (WARN_ON(qgroup->reserved < num_bytes))
> > report_reserved_underflow(fs_info, qgroup,
> > num_bytes);
> > @@ -2178,6 +2183,8 @@ static int qgroup_reserve(struct btrfs_root *root, u64 num_bytes)
> >
> > qg = u64_to_ptr(unode->aux);
> >
> > + trace_qgroup_update_reserve(fs_info, qg->qgroupid,
> > + qg->reserved, num_bytes);
> > qg->reserved += num_bytes;
> > }
> >
> > @@ -2223,6 +2230,8 @@ void btrfs_qgroup_free_refroot(struct btrfs_fs_info *fs_info,
> >
> > qg = u64_to_ptr(unode->aux);
> >
> > + trace_qgroup_update_reserve(fs_info, qg->qgroupid,
> > + qg->reserved, (s64)-num_bytes);
> > if (WARN_ON(qgroup->reserved < num_bytes))
> > report_reserved_underflow(fs_info, qgroup, num_bytes);
> > else
> > @@ -2706,6 +2715,8 @@ int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes)
> > return 0;
> >
> > BUG_ON(num_bytes != round_down(num_bytes, root->nodesize));
> > + trace_qgroup_meta_reserve(root->fs_info, root->objectid,
> > + (s64)num_bytes);
> > ret = qgroup_reserve(root, num_bytes);
> > if (ret < 0)
> > return ret;
> > @@ -2724,6 +2735,8 @@ void btrfs_qgroup_free_meta_all(struct btrfs_root *root)
> > reserved = atomic_xchg(&root->qgroup_meta_rsv, 0);
> > if (reserved == 0)
> > return;
> > + trace_qgroup_meta_reserve(root->fs_info, root->objectid,
> > + (s64)-reserved);
> > qgroup_free(root, reserved);
> > }
> >
> > @@ -2736,6 +2749,8 @@ void btrfs_qgroup_free_meta(struct btrfs_root *root, int num_bytes)
> > BUG_ON(num_bytes != round_down(num_bytes, root->nodesize));
> > WARN_ON(atomic_read(&root->qgroup_meta_rsv) < num_bytes);
> > atomic_sub(num_bytes, &root->qgroup_meta_rsv);
> > + trace_qgroup_meta_reserve(root->fs_info, root->objectid,
> > + (s64)-num_bytes);
> > qgroup_free(root, num_bytes);
> > }
>
> Why is there a hyphen in some places after "(s64)" and not in others? Is that supposed to be
> type casting or something else?
Right, this looks strange and not clear from the context, besides that
the typecast should precede negation: -(s64)num_bytes
There's also a type mismatch in the 2nd tracepoint, where argument is
s64, the tracepoint entry is u64 and printf format is %lld.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
