On Thu, Jan 03, 2019 at 11:30:19AM +0530, Devesh Sharma wrote:
> On Thu, Jan 3, 2019 at 2:26 AM Jason Gunthorpe <jgg@xxxxxxxxxxxx> wrote:
> >
> > On Mon, Dec 31, 2018 at 01:10:01AM -0500, Devesh Sharma wrote:
> > > Adding setup and destroy routines for chip-context. The
> > > chip context would be used frequently in control and data
> > > path to take execution flow depending on the chip type.
> > > chip context structure pointer is added to the relevant
> > > data structures.
> > >
> > > Signed-off-by: Selvin Xavier <selvin.xavier@xxxxxxxxxxxx>
> > > Signed-off-by: Devesh Sharma <devesh.sharma@xxxxxxxxxxxx>
> > > drivers/infiniband/hw/bnxt_re/bnxt_re.h | 1 +
> > > drivers/infiniband/hw/bnxt_re/main.c | 43 ++++++++++++++++++++++++++++++
> > > drivers/infiniband/hw/bnxt_re/qplib_fp.h | 1 +
> > > drivers/infiniband/hw/bnxt_re/qplib_rcfw.h | 1 +
> > > drivers/infiniband/hw/bnxt_re/qplib_res.h | 15 ++++++++++-
> > > 5 files changed, 60 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/infiniband/hw/bnxt_re/bnxt_re.h b/drivers/infiniband/hw/bnxt_re/bnxt_re.h
> > > index 31baa893..6c16515 100644
> > > +++ b/drivers/infiniband/hw/bnxt_re/bnxt_re.h
> > > @@ -124,6 +124,7 @@ struct bnxt_re_dev {
> > > #define BNXT_RE_FLAG_ISSUE_ROCE_STATS 29
> > > struct net_device *netdev;
> > > unsigned int version, major, minor;
> > > + struct bnxt_qplib_chip_ctx *chip_ctx;
> > > struct bnxt_en_dev *en_dev;
> > > struct bnxt_msix_entry msix_entries[BNXT_RE_MAX_MSIX];
> > > int num_msix;
> > > diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c
> > > index e7a997f..e3e1944 100644
> > > +++ b/drivers/infiniband/hw/bnxt_re/main.c
> > > @@ -80,6 +80,38 @@
> > > static struct workqueue_struct *bnxt_re_wq;
> > > static void bnxt_re_ib_unreg(struct bnxt_re_dev *rdev);
> > >
> > > +static void bnxt_re_destroy_chip_ctx(struct bnxt_re_dev *rdev)
> > > +{
> > > + struct bnxt_qplib_chip_ctx *chip_ctx;
> > > +
> > > + chip_ctx = rdev->chip_ctx;
> > > + rdev->chip_ctx = NULL;
> > > + rdev->rcfw.res = NULL;
> > > + rdev->qplib_res.cctx = NULL;
> > > + kfree(chip_ctx);
> > > +}
> > > +
> > > +static int bnxt_re_setup_chip_ctx(struct bnxt_re_dev *rdev)
> > > +{
> > > + struct bnxt_qplib_chip_ctx *chip_ctx;
> > > + struct bnxt_en_dev *en_dev;
> > > + struct bnxt *bp;
> > > +
> > > + en_dev = rdev->en_dev;
> > > + bp = netdev_priv(en_dev->net);
> > > +
> > > + chip_ctx = kzalloc(sizeof(*chip_ctx), GFP_KERNEL);
> > > + if (!chip_ctx)
> > > + return -ENOMEM;
> >
> > It seems a bit weird to kzalloc this? Why not just inline it in the
> > rdev struct?
> I did not wanted to pile up another structure to the bnxt_re_device
> structure and increase its size by the sizeof (struct
> bnxt_qplib_chip_ctx).
> eventually this struct may grow bigger. I am not aware if inline'ing
> is preferred, if that is the case, I will change it.
If struct is not near PAGE_SIZE don't worry about it
Jason
