Re: [PATCH 03/61] trace: update VFS-layer trace events for u64 i_ino

Jeff Layton via Linux-f2fs-devel <[email protected]> Thu, 26 Feb 2026 12:53:42 -0500
Newsgroups gmane.linux.file-systems.f2fs,gmane.comp.video.dri.devel,gmane.linux.hams,gmane.comp.file-systems.ceph.devel,gmane.linux.kernel,gmane.linux.kernel.perf.user,gmane.linux.drivers.video-input-infrastructure,gmane.linux.kernel.cifs,gmane.comp.file-systems.nilfs.user,gmane.comp.file-systems.coda.general,gmane.linux.drivers.mtd,gmane.comp.freedesktop.amd-gfx,gmane.linux.kernel.autofs,gmane.linux.nfs,gmane.comp.file-systems.ext4,gmane.linux.kernel.mm,gmane.network.samba.internals,gmane.linux.file-systems,gmane.comp.file-systems.ecryptfs.general,gmane.linux.file-systems.union,gmane.linux.x25,gmane.comp.security.apparmor,gmane.linux.network,gmane.linux.kernel.lsm
Message-ID <[email protected]> 
On Thu, 2026-02-26 at 18:11 +0100, Jan Kara wrote:
> On Thu 26-02-26 10:55:05, Jeff Layton wrote:
> > Update trace event definitions in VFS-layer trace headers to use u64
> > instead of ino_t/unsigned long for inode number fields, and change
> > format strings from %lu/%lx to %llu/%llx to match.
> > 
> > This is needed because i_ino is now u64. Changing trace event field
> > types changes the binary trace format, but the self-describing format
> > metadata handles this transparently for modern trace-cmd and perf.
> > 
> > Files updated:
> >   - cachefiles.h, filelock.h, filemap.h, fs_dax.h, fsverity.h,
> >     hugetlbfs.h, netfs.h, readahead.h, timestamp.h, writeback.h
> > 
> > Signed-off-by: Jeff Layton <[email protected]>
> 
> ...
> 
> > diff --git a/include/trace/events/writeback.h b/include/trace/events/writeback.h
> > index 4d3d8c8f3a1bc3e5ef10fc96e3c6dbbd0cf00c98..cc7651749eb3ce1123cb3ea9496f0803a0f4c1a0 100644
> > --- a/include/trace/events/writeback.h
> > +++ b/include/trace/events/writeback.h
> > @@ -67,7 +67,7 @@ DECLARE_EVENT_CLASS(writeback_folio_template,
> >  
> >  	TP_STRUCT__entry (
> >  		__array(char, name, 32)
> > -		__field(ino_t, ino)
> > +		__field(u64, ino)
> >  		__field(pgoff_t, index)
> >  	),
> >  
> > @@ -79,9 +79,9 @@ DECLARE_EVENT_CLASS(writeback_folio_template,
> >  		__entry->index = folio->index;
> >  	),
> >  
> > -	TP_printk("bdi %s: ino=%lu index=%lu",
> > +	TP_printk("bdi %s: ino=%llu index=%lu",
> >  		__entry->name,
> > -		(unsigned long)__entry->ino,
> > +		(unsigned long long)__entry->ino,
> 
> No need for explicit typing to ULL?
>
> >  		__entry->index
> >  	)
> >  );
> > @@ -108,7 +108,7 @@ DECLARE_EVENT_CLASS(writeback_dirty_inode_template,
> >  
> >  	TP_STRUCT__entry (
> >  		__array(char, name, 32)
> > -		__field(ino_t, ino)
> > +		__field(u64, ino)
> >  		__field(unsigned long, state)
> >  		__field(unsigned long, flags)
> >  	),
> > @@ -123,9 +123,9 @@ DECLARE_EVENT_CLASS(writeback_dirty_inode_template,
> >  		__entry->flags		= flags;
> >  	),
> >  
> > -	TP_printk("bdi %s: ino=%lu state=%s flags=%s",
> > +	TP_printk("bdi %s: ino=%llu state=%s flags=%s",
> >  		__entry->name,
> > -		(unsigned long)__entry->ino,
> > +		(unsigned long long)__entry->ino,
> 
> And here as well? And many times below as well...
> 
> 								Honza

Good catch. I'll clean those up.

Thanks,
-- 
Jeff Layton <[email protected]>