Re: [PATCH v16 12/16] IB, arm64: untag user pointers in ib_uverbs_(re)reg_mr()

[Andrey Konovalov]

On Mon, Jun 3, 2019 at 7:46 PM Jason Gunthorpe <jgg@xxxxxxxx> wrote:
>
> On Mon, Jun 03, 2019 at 06:55:14PM +0200, Andrey Konovalov wrote:
> > This patch is a part of a series that extends arm64 kernel ABI to allow to
> > pass tagged user pointers (with the top byte set to something else other
> > than 0x00) as syscall arguments.
> >
> > ib_uverbs_(re)reg_mr() use provided user pointers for vma lookups (through
> > e.g. mlx4_get_umem_mr()), which can only by done with untagged pointers.
> >
> > Untag user pointers in these functions.
> >
> > Signed-off-by: Andrey Konovalov <andreyknvl@xxxxxxxxxx>
> >  drivers/infiniband/core/uverbs_cmd.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> >
> > diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
> > index 5a3a1780ceea..f88ee733e617 100644
> > +++ b/drivers/infiniband/core/uverbs_cmd.c
> > @@ -709,6 +709,8 @@ static int ib_uverbs_reg_mr(struct uverbs_attr_bundle *attrs)
> >       if (ret)
> >               return ret;
> >
> > +     cmd.start = untagged_addr(cmd.start);
> > +
> >       if ((cmd.start & ~PAGE_MASK) != (cmd.hca_va & ~PAGE_MASK))
> >               return -EINVAL;
>
> I feel like we shouldn't thave to do this here, surely the cmd.start
> should flow unmodified to get_user_pages, and gup should untag it?
>
> ie, this sort of direction for the IB code (this would be a giant
> patch, so I didn't have time to write it all, but I think it is much
> saner):

Hi Jason,

ib_uverbs_reg_mr() passes cmd.start to mlx4_get_umem_mr(), which calls
find_vma(), which only accepts untagged addresses. Could you explain
how your patch helps?

Thanks!

>
> diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c
> index 54628ef879f0ce..7b3b736c87c253 100644
> --- a/drivers/infiniband/core/umem.c
> +++ b/drivers/infiniband/core/umem.c
> @@ -193,7 +193,7 @@ EXPORT_SYMBOL(ib_umem_find_best_pgsz);
>   * @access: IB_ACCESS_xxx flags for memory being pinned
>   * @dmasync: flush in-flight DMA when the memory region is written
>   */
> -struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
> +struct ib_umem *ib_umem_get(struct ib_udata *udata, void __user *addr,
>                             size_t size, int access, int dmasync)
>  {
>         struct ib_ucontext *context;
> @@ -201,7 +201,7 @@ struct ib_umem *ib_umem_get(struct ib_udata *udata, unsigned long addr,
>         struct page **page_list;
>         unsigned long lock_limit;
>         unsigned long new_pinned;
> -       unsigned long cur_base;
> +       void __user *cur_base;
>         struct mm_struct *mm;
>         unsigned long npages;
>         int ret;
> diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
> index 5a3a1780ceea4d..94389e7f12371f 100644
> --- a/drivers/infiniband/core/uverbs_cmd.c
> +++ b/drivers/infiniband/core/uverbs_cmd.c
> @@ -735,7 +735,8 @@ static int ib_uverbs_reg_mr(struct uverbs_attr_bundle *attrs)
>                 }
>         }
>
> -       mr = pd->device->ops.reg_user_mr(pd, cmd.start, cmd.length, cmd.hca_va,
> +       mr = pd->device->ops.reg_user_mr(pd, u64_to_user_ptr(cmd.start),
> +                                        cmd.length, cmd.hca_va,
>                                          cmd.access_flags,
>                                          &attrs->driver_udata);
>         if (IS_ERR(mr)) {
> diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c
> index 4d033796dcfcc2..bddbb952082fc5 100644
> --- a/drivers/infiniband/hw/mlx5/mr.c
> +++ b/drivers/infiniband/hw/mlx5/mr.c
> @@ -786,7 +786,7 @@ static int mr_cache_max_order(struct mlx5_ib_dev *dev)
>  }
>
>  static int mr_umem_get(struct mlx5_ib_dev *dev, struct ib_udata *udata,
> -                      u64 start, u64 length, int access_flags,
> +                      void __user *start, u64 length, int access_flags,
>                        struct ib_umem **umem, int *npages, int *page_shift,
>                        int *ncont, int *order)
>  {
> @@ -1262,8 +1262,8 @@ struct ib_mr *mlx5_ib_reg_dm_mr(struct ib_pd *pd, struct ib_dm *dm,
>                                  attr->access_flags, mode);
>  }
>
> -struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
> -                                 u64 virt_addr, int access_flags,
> +struct ib_mr *mlx5_ib_reg_user_mr(struct ib_pd *pd, void __user *start,
> +                                 u64 length, u64 virt_addr, int access_flags,
>                                   struct ib_udata *udata)
>  {
>         struct mlx5_ib_dev *dev = to_mdev(pd->device);
> diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
> index ec6446864b08e9..b3c8eaaa35c760 100644
> --- a/include/rdma/ib_verbs.h
> +++ b/include/rdma/ib_verbs.h
> @@ -2464,8 +2464,8 @@ struct ib_device_ops {
>         struct ib_mr *(*reg_user_mr)(struct ib_pd *pd, u64 start, u64 length,
>                                      u64 virt_addr, int mr_access_flags,
>                                      struct ib_udata *udata);
> -       int (*rereg_user_mr)(struct ib_mr *mr, int flags, u64 start, u64 length,
> -                            u64 virt_addr, int mr_access_flags,
> +       int (*rereg_user_mr)(struct ib_mr *mr, int flags, void __user *start,
> +                            u64 length, u64 virt_addr, int mr_access_flags,
>                              struct ib_pd *pd, struct ib_udata *udata);
>         int (*dereg_mr)(struct ib_mr *mr, struct ib_udata *udata);
>         struct ib_mr *(*alloc_mr)(struct ib_pd *pd, enum ib_mr_type mr_type,