From: Oded Gabbay <ogabbay@kernel.org>
To: linux-kernel@vger.kernel.org, gregkh@linuxfoundation.org
Cc: linux-rdma@vger.kernel.org, daniel.vetter@ffwll.ch,
sleybo@amazon.com, galpress@amazon.com,
dri-devel@lists.freedesktop.org, christian.koenig@amd.com,
linaro-mm-sig@lists.linaro.org, jgg@ziepe.ca,
dledford@redhat.com, hch@lst.de, amd-gfx@lists.freedesktop.org,
alexander.deucher@amd.com, Tomer Tayar <ttayar@habana.ai>,
leonro@nvidia.com, linux-media@vger.kernel.org
Subject: [PATCH v4 1/2] habanalabs: define uAPI to export FD for DMA-BUF
Date: Mon, 5 Jul 2021 16:03:13 +0300 [thread overview]
Message-ID: <20210705130314.11519-2-ogabbay@kernel.org> (raw)
In-Reply-To: <20210705130314.11519-1-ogabbay@kernel.org>
User process might want to share the device memory with another
driver/device, and to allow it to access it over PCIe (P2P).
To enable this, we utilize the dma-buf mechanism and add a dma-buf
exporter support, so the other driver can import the device memory and
access it.
The device memory is allocated using our existing allocation uAPI,
where the user will get a handle that represents the allocation.
The user will then need to call the new
uAPI (HL_MEM_OP_EXPORT_DMABUF_FD) and give the handle as a parameter.
The driver will return a FD that represents the DMA-BUF object that
was created to match that allocation.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
---
include/uapi/misc/habanalabs.h | 28 +++++++++++++++++++++++++++-
1 file changed, 27 insertions(+), 1 deletion(-)
diff --git a/include/uapi/misc/habanalabs.h b/include/uapi/misc/habanalabs.h
index 18765eb75b65..c5cbd60696d7 100644
--- a/include/uapi/misc/habanalabs.h
+++ b/include/uapi/misc/habanalabs.h
@@ -808,6 +808,10 @@ union hl_wait_cs_args {
#define HL_MEM_OP_UNMAP 3
/* Opcode to map a hw block */
#define HL_MEM_OP_MAP_BLOCK 4
+/* Opcode to create DMA-BUF object for an existing device memory allocation
+ * and to export an FD of that DMA-BUF back to the caller
+ */
+#define HL_MEM_OP_EXPORT_DMABUF_FD 5
/* Memory flags */
#define HL_MEM_CONTIGUOUS 0x1
@@ -879,11 +883,26 @@ struct hl_mem_in {
/* Virtual address returned from HL_MEM_OP_MAP */
__u64 device_virt_addr;
} unmap;
+
+ /* HL_MEM_OP_EXPORT_DMABUF_FD */
+ struct {
+ /* Handle returned from HL_MEM_OP_ALLOC. In Gaudi,
+ * where we don't have MMU for the device memory, the
+ * driver expects a physical address (instead of
+ * a handle) in the device memory space.
+ */
+ __u64 handle;
+ /* Size of memory allocation. Relevant only for GAUDI */
+ __u64 mem_size;
+ } export_dmabuf_fd;
};
/* HL_MEM_OP_* */
__u32 op;
- /* HL_MEM_* flags */
+ /* HL_MEM_* flags.
+ * For the HL_MEM_OP_EXPORT_DMABUF_FD opcode, this field holds the
+ * DMA-BUF file/FD flags.
+ */
__u32 flags;
/* Context ID - Currently not in use */
__u32 ctx_id;
@@ -920,6 +939,13 @@ struct hl_mem_out {
__u32 pad;
};
+
+ /* Returned in HL_MEM_OP_EXPORT_DMABUF_FD. Represents the
+ * DMA-BUF object that was created to describe a memory
+ * allocation on the device's memory space. The FD should be
+ * passed to the importer driver
+ */
+ __u64 fd;
};
};
--
2.25.1
next prev parent reply other threads:[~2021-07-05 13:03 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-05 13:03 [PATCH v4 0/2] Add p2p via dmabuf to habanalabs Oded Gabbay
2021-07-05 13:03 ` Oded Gabbay [this message]
2021-07-05 13:03 ` [PATCH v4 2/2] habanalabs: add support for dma-buf exporter Oded Gabbay
2021-07-05 16:52 ` Jason Gunthorpe
2021-07-06 9:44 ` Oded Gabbay
2021-07-06 13:54 ` Jason Gunthorpe
2021-07-06 14:00 ` Oded Gabbay
2021-07-06 8:40 ` [PATCH v4 0/2] Add p2p via dmabuf to habanalabs Daniel Vetter
2021-07-06 10:03 ` Oded Gabbay
2021-07-06 10:36 ` Daniel Vetter
2021-07-06 10:47 ` Daniel Vetter
2021-07-06 12:07 ` Daniel Vetter
2021-07-06 13:44 ` Jason Gunthorpe
2021-07-06 14:09 ` Daniel Vetter
2021-07-06 14:56 ` Jason Gunthorpe
2021-07-06 15:52 ` Daniel Vetter
2021-07-06 14:23 ` Jason Gunthorpe
2021-07-06 14:39 ` Daniel Vetter
2021-07-06 15:25 ` Jason Gunthorpe
2021-07-06 15:49 ` Daniel Vetter
2021-07-06 16:07 ` Daniel Vetter
2021-07-06 17:28 ` Jason Gunthorpe
[not found] ` <20210706173137.GA7840@lst.de>
2021-07-06 17:59 ` Jason Gunthorpe
2021-07-09 14:47 ` Dennis Dalessandro
2021-07-06 16:29 ` Jason Gunthorpe
2021-07-06 17:35 ` Daniel Vetter
2021-07-06 18:03 ` Daniel Vetter
2021-07-06 18:31 ` Jason Gunthorpe
2021-07-06 19:06 ` Daniel Vetter
2021-07-06 19:09 ` Alex Deucher
[not found] ` <20210706122110.GA18273@lst.de>
2021-07-06 12:23 ` [Linaro-mm-sig] " Daniel Vetter
2021-07-06 12:45 ` Oded Gabbay
2021-07-06 13:17 ` Daniel Vetter
2021-07-06 13:45 ` Oded Gabbay
2021-07-07 12:17 ` Christian König
2021-07-07 12:54 ` Daniel Vetter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210705130314.11519-2-ogabbay@kernel.org \
--to=ogabbay@kernel.org \
--cc=alexander.deucher@amd.com \
--cc=amd-gfx@lists.freedesktop.org \
--cc=christian.koenig@amd.com \
--cc=daniel.vetter@ffwll.ch \
--cc=dledford@redhat.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=galpress@amazon.com \
--cc=gregkh@linuxfoundation.org \
--cc=hch@lst.de \
--cc=jgg@ziepe.ca \
--cc=leonro@nvidia.com \
--cc=linaro-mm-sig@lists.linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-media@vger.kernel.org \
--cc=linux-rdma@vger.kernel.org \
--cc=sleybo@amazon.com \
--cc=ttayar@habana.ai \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).