> -----Original Message-----
> From: netdev-ow...@vger.kernel.org <netdev-ow...@vger.kernel.org> On
> Behalf Of Andy Shevchenko
> Sent: Wednesday, August 21, 2019 2:56 PM
> To: Joseph Qi <joseph...@linux.alibaba.com>
> Cc: Mark Fasheh <m...@fasheh.com>; Joel Becker <jl...@evilplan.org>;
> ocfs2-de...@oss.oracle.com; Ariel Elior <ael...@marvell.com>; Sudarsana
> Reddy Kalluru <skall...@marvell.com>; GR-everest-linux-l2 <GR-everest-
> linux...@marvell.com>; David S. Miller <da...@davemloft.net>;
> netdev@vger.kernel.org; Colin Ian King <colin.k...@canonical.com>
> Subject: Re: [PATCH v1] ocfs2/dlm: Move BITS_TO_BYTES() to bitops.h for
> wider use
>
> On Wed, Aug 21, 2019 at 09:29:04AM +0800, Joseph Qi wrote:
> > On 19/8/21 00:31, Andy Shevchenko wrote:
> > > There are users already and will be more of BITS_TO_BYTES() macro.
> > > Move it to bitops.h for wider use.
> > >
> > > Signed-off-by: Andy Shevchenko <andriy.shevche...@linux.intel.com>
> > > ---
> > > drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h | 1 -
> > > fs/ocfs2/dlm/dlmcommon.h | 4 ----
> > > include/linux/bitops.h | 1 +
> > > 3 files changed, 1 insertion(+), 5 deletions(-)
> > >
> > > diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
> > > b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
> > > index 066765fbef06..0a59a09ef82f 100644
> > > --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
> > > +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_init.h
> > > @@ -296,7 +296,6 @@ static inline void bnx2x_dcb_config_qm(struct
> bnx2x *bp, enum cos_mode mode,
> > > * possible, the driver should only write the valid vnics into the
> > > internal
> > > * ram according to the appropriate port mode.
> > > */
> > > -#define BITS_TO_BYTES(x) ((x)/8)>
> > I don't think this is a equivalent replace, or it is in fact wrong
> > before?
>
> I was thinking about this one and there are two applications:
> - calculus of the amount of structures of certain type per PAGE
> (obviously off-by-one error in the original code IIUC purpose of
> STRUCT_SIZE)
> - calculus of some threshold based on line speed in bytes per second
> (I dunno it will have any difference on the Gbs / 100 MBs speeds)
>
I see that both the implementations (existing vs new) yield same value for
standard speeds 10G (i.e.,10000), 1G (1000) that device supports. Hence the
change look to be ok.
> > > /* CMNG constants, as derived from system spec calculations */
> > >
> > > diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h
> > > index aaf24548b02a..0463dce65bb2 100644
> > > --- a/fs/ocfs2/dlm/dlmcommon.h
> > > +++ b/fs/ocfs2/dlm/dlmcommon.h
> > > @@ -688,10 +688,6 @@ struct dlm_begin_reco
> > > __be32 pad2;
> > > };
> > >
> > > -
> > > -#define BITS_PER_BYTE 8
> > > -#define BITS_TO_BYTES(bits)
> > > (((bits)+BITS_PER_BYTE-1)/BITS_PER_BYTE)
> > > -
> > For ocfs2 part, it looks good to me.
> > Reviewed-by: Joseph Qi <joseph...@linux.alibaba.com>
>
> Thanks!
>
> >
> > > struct dlm_query_join_request
> > > {
> > > u8 node_idx;
> > > diff --git a/include/linux/bitops.h b/include/linux/bitops.h index
> > > cf074bce3eb3..79d80f5ddf7b 100644
> > > --- a/include/linux/bitops.h
> > > +++ b/include/linux/bitops.h
> > > @@ -5,6 +5,7 @@
> > > #include <linux/bits.h>
> > >
> > > #define BITS_PER_TYPE(type) (sizeof(type) * BITS_PER_BYTE)
> > > +#define BITS_TO_BYTES(nr) DIV_ROUND_UP(nr, BITS_PER_BYTE)
> > > #define BITS_TO_LONGS(nr) DIV_ROUND_UP(nr,
> BITS_PER_TYPE(long))
> > >
> > > extern unsigned int __sw_hweight8(unsigned int w);
> > >
>
> --
> With Best Regards,
> Andy Shevchenko
>
