From: Trond Myklebust <trondmy@hammerspace.com>
To: "bfields@fieldses.org" <bfields@fieldses.org>,
"dwysocha@redhat.com" <dwysocha@redhat.com>
Cc: "linux-nfs@vger.kernel.org" <linux-nfs@vger.kernel.org>,
"chuck.lever@oracle.com" <chuck.lever@oracle.com>
Subject: Re: [PATCH v2 1/1] nfsd4: Expose the callback address and state of each NFS4 client
Date: Wed, 26 May 2021 01:32:03 +0000 [thread overview]
Message-ID: <d0466617bdf9856ae6eb56416ea5f3637340f49b.camel@hammerspace.com> (raw)
In-Reply-To: <20210525205845.GB4162@fieldses.org>
On Tue, 2021-05-25 at 16:58 -0400, Bruce Fields wrote:
> When I run trace-cmd report I get output like:
>
> [nfsd:nfsd_cb_state] function cb_state2str not defined
> [nfsd:nfsd_cb_shutdown] function cb_state2str not defined
> [nfsd:nfsd_cb_probe] function cb_state2str not defined
> [nfsd:nfsd_cb_lost] function cb_state2str not defined
>
> I don't know how this is supposed to work. Is it OK for tracepoint
> definitions
> to reference kernel functions if they're defined in the right way
> somehow? If
> not, I don't know what the solution would be for sharing this--define
> a macro
> that expands to the array literal and use that in both places? Or
> maybe just
> live with the the redundancy.
>
You need to store the string in the TP_fast_assign() section if you
want to be able to display it. Otherwise 'perf' tries (and fails) to
find the cb_state2str function so it can invoke it.
> --b.
>
> On Mon, May 17, 2021 at 04:29:45PM -0400, Dave Wysochanski wrote:
> > In addition to the client's address, display the callback channel
> > state and address in the 'info' file. Define and use a common
> > function for this information that can be used by both callback
> > trace events and the NFS4 client 'info' file.
> >
> > Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
> > ---
> > fs/nfsd/nfs4state.c | 2 ++
> > fs/nfsd/trace.c | 15 +++++++++++++++
> > fs/nfsd/trace.h | 9 ++-------
> > 3 files changed, 19 insertions(+), 7 deletions(-)
> >
> > diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
> > index b2573d3ecd3c..f3b8221bb543 100644
> > --- a/fs/nfsd/nfs4state.c
> > +++ b/fs/nfsd/nfs4state.c
> > @@ -2385,6 +2385,8 @@ static int client_info_show(struct seq_file
> > *m, void *v)
> > seq_printf(m, "\nImplementation time: [%lld,
> > %ld]\n",
> > clp->cl_nii_time.tv_sec, clp-
> > >cl_nii_time.tv_nsec);
> > }
> > + seq_printf(m, "callback state: %s\n", cb_state2str(clp-
> > >cl_cb_state));
> > + seq_printf(m, "callback address: %pISpc\n", &clp-
> > >cl_cb_conn.cb_addr);
> > drop_client(clp);
> >
> > return 0;
> > diff --git a/fs/nfsd/trace.c b/fs/nfsd/trace.c
> > index f008b95ceec2..6291b5d10824 100644
> > --- a/fs/nfsd/trace.c
> > +++ b/fs/nfsd/trace.c
> > @@ -2,3 +2,18 @@
> >
> > #define CREATE_TRACE_POINTS
> > #include "trace.h"
> > +
> > +const char *cb_state2str(const int state)
> > +{
> > + switch (state) {
> > + case NFSD4_CB_UP:
> > + return "UP";
> > + case NFSD4_CB_UNKNOWN:
> > + return "UNKNOWN";
> > + case NFSD4_CB_DOWN:
> > + return "DOWN";
> > + case NFSD4_CB_FAULT:
> > + return "FAULT";
> > + }
> > + return "UNDEFINED";
> > +}
> > diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h
> > index 10cc3aaf1089..8908d48b2aa6 100644
> > --- a/fs/nfsd/trace.h
> > +++ b/fs/nfsd/trace.h
> > @@ -876,12 +876,7 @@
> > TP_printk("client %08x:%08x", __entry->cl_boot, __entry-
> > >cl_id)
> > )
> >
> > -#define
> > show_cb_state(val) \
> > -
> > __print_symbolic(val,
> > \
> > - { NFSD4_CB_UP, "UP"
> > }, \
> > - { NFSD4_CB_UNKNOWN, "UNKNOWN"
> > }, \
> > - { NFSD4_CB_DOWN, "DOWN"
> > }, \
> > - { NFSD4_CB_FAULT, "FAULT"})
> > +const char *cb_state2str(const int state);
> >
> > DECLARE_EVENT_CLASS(nfsd_cb_class,
> > TP_PROTO(const struct nfs4_client *clp),
> > @@ -901,7 +896,7 @@
> > ),
> > TP_printk("addr=%pISpc client %08x:%08x state=%s",
> > __entry->addr, __entry->cl_boot, __entry->cl_id,
> > - show_cb_state(__entry->state))
> > + cb_state2str(__entry->state))
> > );
> >
> > #define DEFINE_NFSD_CB_EVENT(name) \
> > --
> > 1.8.3.1
--
Trond Myklebust
Linux NFS client maintainer, Hammerspace
trond.myklebust@hammerspace.com
prev parent reply other threads:[~2021-05-26 1:32 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-17 20:29 [PATCH v2 0/1] Add callback address and state to nfsd4 client info Dave Wysochanski
2021-05-17 20:29 ` [PATCH v2 1/1] nfsd4: Expose the callback address and state of each NFS4 client Dave Wysochanski
2021-05-25 20:58 ` Bruce Fields
2021-05-25 21:48 ` Chuck Lever III
2021-05-26 18:30 ` Bruce Fields
2021-05-26 19:13 ` David Wysochanski
2021-05-26 1:32 ` Trond Myklebust [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=d0466617bdf9856ae6eb56416ea5f3637340f49b.camel@hammerspace.com \
--to=trondmy@hammerspace.com \
--cc=bfields@fieldses.org \
--cc=chuck.lever@oracle.com \
--cc=dwysocha@redhat.com \
--cc=linux-nfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox