| Message ID | f5769e20-970a-53f5-4fc7-2afd11304ce2@sandeen.net (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
On Fri, Dec 01, 2017 at 04:12:34PM -0600, Eric Sandeen wrote: > Advertise this config option along with the others. > > Signed-off-by: Eric Sandeen <sandeen@redhat.com> > --- > > diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h > index fcc5dfc..8cee8e8 100644 > --- a/fs/xfs/xfs_super.h > +++ b/fs/xfs/xfs_super.h > @@ -44,6 +44,12 @@ > # define XFS_REALTIME_STRING > #endif > > +#ifdef CONFIG_XFS_ONLINE_SCRUB > +# define XFS_SCRUB_STRING "scrub, " > +#else > +# define XFS_SCRUB_STRING > +#endif I don't mind this addition to the build options, though I had pictured scrub becoming an integral part of xfs some day and not remaining a configurable option. That said I haven't evaluated how much scrub bloats up xfs nor do I know if the kernel tinyfication people would actually want to turn it off?? (For my part I prefer it some day not be Kconfig option so that it won't bitrot ala CONFIG_XFS_RT=y.) --D > + > #ifdef DEBUG > # define XFS_DBG_STRING "debug" > #else > @@ -54,6 +60,7 @@ > #define XFS_BUILD_OPTIONS XFS_ACL_STRING \ > XFS_SECURITY_STRING \ > XFS_REALTIME_STRING \ > + XFS_SCRUB_STRING \ > XFS_DBG_STRING /* DBG must be last */ > > struct xfs_inode; > > -- > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 12/1/17 4:29 PM, Darrick J. Wong wrote: > On Fri, Dec 01, 2017 at 04:12:34PM -0600, Eric Sandeen wrote: >> Advertise this config option along with the others. >> >> Signed-off-by: Eric Sandeen <sandeen@redhat.com> >> --- >> >> diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h >> index fcc5dfc..8cee8e8 100644 >> --- a/fs/xfs/xfs_super.h >> +++ b/fs/xfs/xfs_super.h >> @@ -44,6 +44,12 @@ >> # define XFS_REALTIME_STRING >> #endif >> >> +#ifdef CONFIG_XFS_ONLINE_SCRUB >> +# define XFS_SCRUB_STRING "scrub, " >> +#else >> +# define XFS_SCRUB_STRING >> +#endif > > I don't mind this addition to the build options, though I had pictured > scrub becoming an integral part of xfs some day and not remaining a > configurable option. It could be removed at that time, unless it's considered an API? I'm not hung up on this, if there's no reason for it that's OK. > That said I haven't evaluated how much scrub bloats up xfs nor do I know > if the kernel tinyfication people would actually want to turn it off?? > > (For my part I prefer it some day not be Kconfig option so that it won't > bitrot ala CONFIG_XFS_RT=y.) I'd imagine that it'd be turned on more often than RT but that's just a guess. -Eric -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, Dec 01, 2017 at 04:31:04PM -0600, Eric Sandeen wrote: > On 12/1/17 4:29 PM, Darrick J. Wong wrote: > > On Fri, Dec 01, 2017 at 04:12:34PM -0600, Eric Sandeen wrote: > >> Advertise this config option along with the others. > >> > >> Signed-off-by: Eric Sandeen <sandeen@redhat.com> > >> --- > >> > >> diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h > >> index fcc5dfc..8cee8e8 100644 > >> --- a/fs/xfs/xfs_super.h > >> +++ b/fs/xfs/xfs_super.h > >> @@ -44,6 +44,12 @@ > >> # define XFS_REALTIME_STRING > >> #endif > >> > >> +#ifdef CONFIG_XFS_ONLINE_SCRUB > >> +# define XFS_SCRUB_STRING "scrub, " > >> +#else > >> +# define XFS_SCRUB_STRING > >> +#endif > > > > I don't mind this addition to the build options, though I had pictured > > scrub becoming an integral part of xfs some day and not remaining a > > configurable option. > > It could be removed at that time, unless it's considered an API? > > I'm not hung up on this, if there's no reason for it that's OK. > > > That said I haven't evaluated how much scrub bloats up xfs nor do I know > > if the kernel tinyfication people would actually want to turn it off?? > > > > (For my part I prefer it some day not be Kconfig option so that it won't > > bitrot ala CONFIG_XFS_RT=y.) > > I'd imagine that it'd be turned on more often than RT but that's just a guess. Ok, now that we've merged all this stuff, it looks like disabling the scrub code alone can reduce the module size by ~8%. I can imagine the occasional system builder who wants xfs but doesn't care about online fsck, so I think I'll keep the kconfig option. Therefore, it ought to go in the build string. Looks ok, Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> --D > > -Eric > > -- > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h index fcc5dfc..8cee8e8 100644 --- a/fs/xfs/xfs_super.h +++ b/fs/xfs/xfs_super.h @@ -44,6 +44,12 @@ # define XFS_REALTIME_STRING #endif +#ifdef CONFIG_XFS_ONLINE_SCRUB +# define XFS_SCRUB_STRING "scrub, " +#else +# define XFS_SCRUB_STRING +#endif + #ifdef DEBUG # define XFS_DBG_STRING "debug" #else @@ -54,6 +60,7 @@ #define XFS_BUILD_OPTIONS XFS_ACL_STRING \ XFS_SECURITY_STRING \ XFS_REALTIME_STRING \ + XFS_SCRUB_STRING \ XFS_DBG_STRING /* DBG must be last */ struct xfs_inode;
Advertise this config option along with the others. Signed-off-by: Eric Sandeen <sandeen@redhat.com> --- -- To unsubscribe from this list: send the line "unsubscribe linux-xfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html