On Thu, May 31, 2018 at 09:02:50PM -0400, Doug Ledford wrote:
> On Tue, 2018-05-29 at 14:56 +0300, Leon Romanovsky wrote:
> > From: Leon Romanovsky <leonro@xxxxxxxxxxxx>
> >
> > Let's perform checks in-place instead of BUG_ONs.
> >
> > Signed-off-by: Leon Romanovsky <leonro@xxxxxxxxxxxx>
> > ---
> > drivers/infiniband/core/mad.c | 11 +++++++----
> > 1 file changed, 7 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c
> > index 6a205224582d..c37079f86341 100644
> > --- a/drivers/infiniband/core/mad.c
> > +++ b/drivers/infiniband/core/mad.c
> > @@ -1556,7 +1556,8 @@ static int add_oui_reg_req(struct ib_mad_reg_req *mad_reg_req,
> > mad_reg_req->oui, 3)) {
> > method = &(*vendor_table)->vendor_class[
> > vclass]->method_table[i];
> > - BUG_ON(!*method);
> > + if (!*method)
> > + goto error3;
> > goto check_in_use;
> > }
> > }
> > @@ -1566,10 +1567,12 @@ static int add_oui_reg_req(struct ib_mad_reg_req *mad_reg_req,
> > vclass]->oui[i])) {
> > method = &(*vendor_table)->vendor_class[
> > vclass]->method_table[i];
> > - BUG_ON(*method);
> > /* Allocate method table for this OUI */
> > - if ((ret = allocate_method_table(method)))
> > - goto error3;
> > + if (*method) {
> ^
> Shouldn't this be (!*method)?
Doug,
I thought over night what I wanted in this patch and what went wrong.
My intention was:
if (*method)
goto error3;
ret = allocate_method_table(method);
if (ret)
oto error3;
I see that you didn't update your for-next yet, can you fix the patch
prior updating for-next?
Thanks
>
> > + ret = allocate_method_table(method);
> > + if (ret)
> > + goto error3;
> > + }
> > memcpy((*vendor_table)->vendor_class[vclass]->oui[i],
> > mad_reg_req->oui, 3);
> > goto check_in_use;
>
> --
> Doug Ledford <dledford@xxxxxxxxxx>
> GPG KeyID: B826A3330E572FDD
> Key fingerprint = AE6B 1BDA 122B 23B4 265B 1274 B826 A333 0E57 2FDD
Attachment:
signature.asc
Description: PGP signature
