Re: [PATCHv2 00/10] ext4: Improve FC trace events

[Ritesh Harjani <riteshh@linux.ibm.com>]

On 22/03/10 09:33PM, Steven Rostedt wrote:
> On Fri, 11 Mar 2022 07:49:31 +0530
> Ritesh Harjani <riteshh@linux.ibm.com> wrote:
>
> > > # cat /sys/kernel/tracing/events/ext4/ext4_fc_commit_stop/format
> >
> > I think you meant ext4_fc_stats.
>
> Sure.
>
> >
> > >
> > > and show me what it outputs.
> >
> > root@qemu:/home/qemu# cat /sys/kernel/tracing/events/ext4/ext4_fc_stats/format
> > name: ext4_fc_stats
> > ID: 986
> > format:
> >         field:unsigned short common_type;       offset:0;       size:2; signed:0;
> >         field:unsigned char common_flags;       offset:2;       size:1; signed:0;
> >         field:unsigned char common_preempt_count;       offset:3;       size:1; signed:0;
> >         field:int common_pid;   offset:4;       size:4; signed:1;
> >
> >         field:dev_t dev;        offset:8;       size:4; signed:0;
> >         field:unsigned int fc_ineligible_rc[EXT4_FC_REASON_MAX];        offset:12;      size:36;        signed:0;
>
> Bah, the above tells us how many items, and the TRACE_DEFINE_ENUM() doesn't
> modify this part of the file.

Then shall I just define TRACE_DEFINE_ENUM(EXT4_FC_REASON_MAX) in this patch.
Would that be the correct approach? Like how we have defined other enums.
We haven't yet defined EXT4_FC_REASON_MAX in current patch.
(as I saw it doesn't affect TP_STRUCT__entry())


>
> I could update it to do so though.

Please let me know if you have any patch for me to try.

Thanks
-ritesh


>
> -- Steve
>
>
> >         field:unsigned long fc_commits; offset:48;      size:8; signed:0;
> >         field:unsigned long fc_ineligible_commits;      offset:56;      size:8; signed:0;
> >         field:unsigned long fc_numblks; offset:64;      size:8; signed:0;
>