On 08/07/2019 12:14, Michal Kalderon wrote:
> diff --git a/drivers/infiniband/core/device.c
> b/drivers/infiniband/core/device.c
> index 8a6ccb936dfe..a830c2c5d691 100644
> --- a/drivers/infiniband/core/device.c
> +++ b/drivers/infiniband/core/device.c
> @@ -2521,6 +2521,7 @@ void ib_set_device_ops(struct ib_device *dev, const
> struct ib_device_ops *ops)
> SET_DEVICE_OP(dev_ops, map_mr_sg_pi);
> SET_DEVICE_OP(dev_ops, map_phys_fmr);
> SET_DEVICE_OP(dev_ops, mmap);
> + SET_DEVICE_OP(dev_ops, mmap_free);
> SET_DEVICE_OP(dev_ops, modify_ah);
> SET_DEVICE_OP(dev_ops, modify_cq);
> SET_DEVICE_OP(dev_ops, modify_device);
> diff --git a/drivers/infiniband/core/rdma_core.c
> b/drivers/infiniband/core/rdma_core.c
> index ccf4d069c25c..7166741834c8 100644
> --- a/drivers/infiniband/core/rdma_core.c
> +++ b/drivers/infiniband/core/rdma_core.c
> @@ -817,6 +817,7 @@ static void ufile_destroy_ucontext(struct ib_uverbs_file
> *ufile,
> rdma_restrack_del(&ucontext->res);
>
> ib_dev->ops.dealloc_ucontext(ucontext);
> + rdma_user_mmap_entries_remove_free(ucontext);
This should happen before dealloc_ucontext.
> +struct rdma_user_mmap_entry *
> +rdma_user_mmap_entry_get(struct ib_ucontext *ucontext, u64 key, u64 len)
> +{
> + struct rdma_user_mmap_entry *entry;
> + u64 mmap_page;
> +
> + mmap_page = key >> PAGE_SHIFT;
> + if (mmap_page > U32_MAX)
> + return NULL;
> +
> + entry = xa_load(&ucontext->mmap_xa, mmap_page);
> + if (!entry || rdma_user_mmap_get_key(entry) != key ||
I wonder if the 'rdma_user_mmap_get_key(entry) != key' check is still needed.
> +/*
> + * This is only called when the ucontext is destroyed and there can be no
> + * concurrent query via mmap or allocate on the xarray, thus we can be sure
> no
> + * other thread is using the entry pointer. We also know that all the BAR
> + * pages have either been zap'd or munmaped at this point. Normal pages are
> + * refcounted and will be freed at the proper time.
> + */
> +void rdma_user_mmap_entries_remove_free(struct ib_ucontext *ucontext)
> +{
> + struct rdma_user_mmap_entry *entry;
> + unsigned long mmap_page;
> +
> + xa_for_each(&ucontext->mmap_xa, mmap_page, entry) {
> + xa_erase(&ucontext->mmap_xa, mmap_page);
> +
> + ibdev_dbg(ucontext->device,
> + "mmap: obj[0x%p] key[%#llx] addr[%#llx] len[%#llx]
> removed\n",
> + entry->obj, rdma_user_mmap_get_key(entry),
> + entry->address, entry->length);
> + if (ucontext->device->ops.mmap_free)
> + ucontext->device->ops.mmap_free(entry->address,
> + entry->length,
> + entry->mmap_flag);
Pass entry instead?
> + kfree(entry);
> + }
> +}
> +
> void uverbs_user_mmap_disassociate(struct ib_uverbs_file *ufile)
> {
> struct rdma_umap_priv *priv, *next_priv;
> diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
> index 26e9c2594913..54ce3fdae180 100644
> --- a/include/rdma/ib_verbs.h
> +++ b/include/rdma/ib_verbs.h
> @@ -1425,6 +1425,8 @@ struct ib_ucontext {
> * Implementation details of the RDMA core, don't use in drivers:
> */
> struct rdma_restrack_entry res;
> + struct xarray mmap_xa;
> + u32 mmap_xa_page;
> };
>
> struct ib_uobject {
> @@ -2311,6 +2313,7 @@ struct ib_device_ops {
> struct ib_udata *udata);
> void (*dealloc_ucontext)(struct ib_ucontext *context);
> int (*mmap)(struct ib_ucontext *context, struct vm_area_struct *vma);
> + void (*mmap_free)(u64 address, u64 length, u8 mmap_flag);
I feel like this callback needs some documentation.
> void (*disassociate_ucontext)(struct ib_ucontext *ibcontext);
> int (*alloc_pd)(struct ib_pd *pd, struct ib_udata *udata);
> void (*dealloc_pd)(struct ib_pd *pd, struct ib_udata *udata);
> @@ -2706,9 +2709,23 @@ void ib_set_client_data(struct ib_device *device,
> struct ib_client *client,
> void ib_set_device_ops(struct ib_device *device,
> const struct ib_device_ops *ops);
>
> +#define RDMA_USER_MMAP_INVALID U64_MAX
> +struct rdma_user_mmap_entry {
> + void *obj;
I know EFA is the culprit here, but please remove the extra space :).
> + u64 address;
> + u64 length;
> + u32 mmap_page;
> + u8 mmap_flag;
> +};
> +
