[PATCHv8 0/6] ublk zero copy support

[PATCHv8 0/6] ublk zero copy support

[Keith Busch]

From: Keith Busch <kbusch@kernel.org>

This one completed liburing 'make runtests' successfully.

Changes from v7:

  Prep patch, mostly from Jens, that removes the "do_import" parameter
  from the generic rw prep.

  Added check for kernel buffers in rw's loop submit. This file
  operation requires __user pointers, so can't use it there (Pavel)

  Added a bool, is_kbuf, so that we don't use the existence of the
  "release" callback to distinguish user vs kernel buffers. And had user
  buffers define its own release function so that we can remove a branch
  check on free.

  The io node and imu caching is moved from the table to ring ctx. This
  is the simplest solution to situations where the node outlives the
  table it came from.

  Fixed missing (parens) logical error checking for ublk flags.

  Minor cleanups to reduce diff churn.

Keith Busch (5):
  io_uring/rw: move buffer_select outside generic prep
  io_uring/rw: move fixed buffer import to issue path
  io_uring: add support for kernel registered bvecs
  ublk: zc register/unregister bvec
  io_uring: cache nodes and mapped buffers

Xinyu Zhang (1):
  nvme: map uring_cmd data even if address is 0

 drivers/block/ublk_drv.c       |  59 ++++++++--
 drivers/nvme/host/ioctl.c      |   2 +-
 include/linux/io_uring/cmd.h   |   7 ++
 include/linux/io_uring_types.h |   2 +
 include/uapi/linux/ublk_cmd.h  |   4 +
 io_uring/filetable.c           |   2 +-
 io_uring/io_uring.c            |   5 +
 io_uring/opdef.c               |   4 +-
 io_uring/rsrc.c                | 189 +++++++++++++++++++++++++++++----
 io_uring/rsrc.h                |  13 ++-
 io_uring/rw.c                  |  85 ++++++++++-----
 io_uring/rw.h                  |   2 +
 12 files changed, 318 insertions(+), 56 deletions(-)

-- 
2.43.5

[PATCHv8 1/6] io_uring/rw: move buffer_select outside generic prep

[Keith Busch]

From: Keith Busch <kbusch@kernel.org>

Cleans up the generic rw prep to not require the do_import flag. Use a
different prep function for callers that might need buffer select.

Based-on-a-patch-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Keith Busch <kbusch@kernel.org>
---
 io_uring/rw.c | 45 ++++++++++++++++++++++++++++-----------------
 1 file changed, 28 insertions(+), 17 deletions(-)

diff --git a/io_uring/rw.c b/io_uring/rw.c
index 788f06fbd7db1..b21b423b3cf8f 100644
--- a/io_uring/rw.c
+++ b/io_uring/rw.c
@@ -248,8 +248,8 @@ static int io_prep_rw_pi(struct io_kiocb *req, struct io_rw *rw, int ddir,
 	return ret;
 }
 
-static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe,
-		      int ddir, bool do_import)
+static int __io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+			int ddir)
 {
 	struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
 	unsigned ioprio;
@@ -285,14 +285,6 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 	rw->len = READ_ONCE(sqe->len);
 	rw->flags = READ_ONCE(sqe->rw_flags);
 
-	if (do_import && !io_do_buffer_select(req)) {
-		struct io_async_rw *io = req->async_data;
-
-		ret = io_import_rw_buffer(ddir, req, io, 0);
-		if (unlikely(ret))
-			return ret;
-	}
-
 	attr_type_mask = READ_ONCE(sqe->attr_type_mask);
 	if (attr_type_mask) {
 		u64 attr_ptr;
@@ -307,26 +299,45 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 	return 0;
 }
 
+static int io_rw_do_import(struct io_kiocb *req, int ddir)
+{
+	if (io_do_buffer_select(req))
+		return 0;
+
+	return io_import_rw_buffer(ddir, req, req->async_data, 0);
+}
+
+static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+		      int ddir)
+{
+	int ret;
+
+	ret = __io_prep_rw(req, sqe, ddir);
+	if (unlikely(ret))
+		return ret;
+
+	return io_rw_do_import(req, ddir);
+}
+
 int io_prep_read(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
-	return io_prep_rw(req, sqe, ITER_DEST, true);
+	return io_prep_rw(req, sqe, ITER_DEST);
 }
 
 int io_prep_write(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
-	return io_prep_rw(req, sqe, ITER_SOURCE, true);
+	return io_prep_rw(req, sqe, ITER_SOURCE);
 }
 
 static int io_prep_rwv(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 		       int ddir)
 {
-	const bool do_import = !(req->flags & REQ_F_BUFFER_SELECT);
 	int ret;
 
-	ret = io_prep_rw(req, sqe, ddir, do_import);
+	ret = io_prep_rw(req, sqe, ddir);
 	if (unlikely(ret))
 		return ret;
-	if (do_import)
+	if (!(req->flags & REQ_F_BUFFER_SELECT))
 		return 0;
 
 	/*
@@ -353,7 +364,7 @@ static int io_prep_rw_fixed(struct io_kiocb *req, const struct io_uring_sqe *sqe
 	struct io_async_rw *io;
 	int ret;
 
-	ret = io_prep_rw(req, sqe, ddir, false);
+	ret = __io_prep_rw(req, sqe, ddir);
 	if (unlikely(ret))
 		return ret;
 
@@ -386,7 +397,7 @@ int io_read_mshot_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	if (!(req->flags & REQ_F_BUFFER_SELECT))
 		return -EINVAL;
 
-	ret = io_prep_rw(req, sqe, ITER_DEST, false);
+	ret = __io_prep_rw(req, sqe, ITER_DEST);
 	if (unlikely(ret))
 		return ret;
 
-- 
2.43.5

Re: [PATCHv8 1/6] io_uring/rw: move buffer_select outside generic prep

[Ming Lei]

On Thu, Feb 27, 2025 at 02:39:11PM -0800, Keith Busch wrote:
> From: Keith Busch <kbusch@kernel.org>
> 
> Cleans up the generic rw prep to not require the do_import flag. Use a
> different prep function for callers that might need buffer select.
> 
> Based-on-a-patch-by: Jens Axboe <axboe@kernel.dk>
> Signed-off-by: Keith Busch <kbusch@kernel.org>
> ---

Reviewed-by: Ming Lei <ming.lei@redhat.com>

thanks,
Ming

[PATCHv8 2/6] io_uring/rw: move fixed buffer import to issue path

[Keith Busch]

From: Keith Busch <kbusch@kernel.org>

Registered buffers may depend on a linked command, which makes the prep
path too early to import. Move to the issue path when the node is
actually needed like all the other users of fixed buffers.

Signed-off-by: Keith Busch <kbusch@kernel.org>
---
 io_uring/opdef.c |  4 ++--
 io_uring/rw.c    | 39 ++++++++++++++++++++++++++++++---------
 io_uring/rw.h    |  2 ++
 3 files changed, 34 insertions(+), 11 deletions(-)

diff --git a/io_uring/opdef.c b/io_uring/opdef.c
index 89f50ecadeaf3..9511262c513e4 100644
--- a/io_uring/opdef.c
+++ b/io_uring/opdef.c
@@ -105,7 +105,7 @@ const struct io_issue_def io_issue_defs[] = {
 		.iopoll_queue		= 1,
 		.async_size		= sizeof(struct io_async_rw),
 		.prep			= io_prep_read_fixed,
-		.issue			= io_read,
+		.issue			= io_read_fixed,
 	},
 	[IORING_OP_WRITE_FIXED] = {
 		.needs_file		= 1,
@@ -119,7 +119,7 @@ const struct io_issue_def io_issue_defs[] = {
 		.iopoll_queue		= 1,
 		.async_size		= sizeof(struct io_async_rw),
 		.prep			= io_prep_write_fixed,
-		.issue			= io_write,
+		.issue			= io_write_fixed,
 	},
 	[IORING_OP_POLL_ADD] = {
 		.needs_file		= 1,
diff --git a/io_uring/rw.c b/io_uring/rw.c
index b21b423b3cf8f..7bc23802a388e 100644
--- a/io_uring/rw.c
+++ b/io_uring/rw.c
@@ -357,31 +357,30 @@ int io_prep_writev(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	return io_prep_rwv(req, sqe, ITER_SOURCE);
 }
 
-static int io_prep_rw_fixed(struct io_kiocb *req, const struct io_uring_sqe *sqe,
+static int io_init_rw_fixed(struct io_kiocb *req, unsigned int issue_flags,
 			    int ddir)
 {
 	struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
-	struct io_async_rw *io;
+	struct io_async_rw *io = req->async_data;
 	int ret;
 
-	ret = __io_prep_rw(req, sqe, ddir);
-	if (unlikely(ret))
-		return ret;
+	if (io->bytes_done)
+		return 0;
 
-	io = req->async_data;
-	ret = io_import_reg_buf(req, &io->iter, rw->addr, rw->len, ddir, 0);
+	ret = io_import_reg_buf(req, &io->iter, rw->addr, rw->len, ddir,
+				issue_flags);
 	iov_iter_save_state(&io->iter, &io->iter_state);
 	return ret;
 }
 
 int io_prep_read_fixed(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
-	return io_prep_rw_fixed(req, sqe, ITER_DEST);
+	return __io_prep_rw(req, sqe, ITER_DEST);
 }
 
 int io_prep_write_fixed(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 {
-	return io_prep_rw_fixed(req, sqe, ITER_SOURCE);
+	return __io_prep_rw(req, sqe, ITER_SOURCE);
 }
 
 /*
@@ -1147,6 +1146,28 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags)
 	}
 }
 
+int io_read_fixed(struct io_kiocb *req, unsigned int issue_flags)
+{
+	int ret;
+
+	ret = io_init_rw_fixed(req, issue_flags, ITER_DEST);
+	if (unlikely(ret))
+		return ret;
+
+	return io_read(req, issue_flags);
+}
+
+int io_write_fixed(struct io_kiocb *req, unsigned int issue_flags)
+{
+	int ret;
+
+	ret = io_init_rw_fixed(req, issue_flags, ITER_SOURCE);
+	if (unlikely(ret))
+		return ret;
+
+	return io_write(req, issue_flags);
+}
+
 void io_rw_fail(struct io_kiocb *req)
 {
 	int res;
diff --git a/io_uring/rw.h b/io_uring/rw.h
index a45e0c71b59d6..bf121b81ebe84 100644
--- a/io_uring/rw.h
+++ b/io_uring/rw.h
@@ -38,6 +38,8 @@ int io_prep_read(struct io_kiocb *req, const struct io_uring_sqe *sqe);
 int io_prep_write(struct io_kiocb *req, const struct io_uring_sqe *sqe);
 int io_read(struct io_kiocb *req, unsigned int issue_flags);
 int io_write(struct io_kiocb *req, unsigned int issue_flags);
+int io_read_fixed(struct io_kiocb *req, unsigned int issue_flags);
+int io_write_fixed(struct io_kiocb *req, unsigned int issue_flags);
 void io_readv_writev_cleanup(struct io_kiocb *req);
 void io_rw_fail(struct io_kiocb *req);
 void io_req_rw_complete(struct io_kiocb *req, io_tw_token_t tw);
-- 
2.43.5

Re: [PATCHv8 2/6] io_uring/rw: move fixed buffer import to issue path

[Ming Lei]

On Thu, Feb 27, 2025 at 02:39:12PM -0800, Keith Busch wrote:
> From: Keith Busch <kbusch@kernel.org>
> 
> Registered buffers may depend on a linked command, which makes the prep
> path too early to import. Move to the issue path when the node is
> actually needed like all the other users of fixed buffers.
> 
> Signed-off-by: Keith Busch <kbusch@kernel.org>

Reviewed-by: Ming Lei <ming.lei@redhat.com>

Thanks,
Ming

[PATCHv8 3/6] nvme: map uring_cmd data even if address is 0

[Keith Busch]

From: Xinyu Zhang <xizhang@purestorage.com>

When using kernel registered bvec fixed buffers, the "address" is
actually the offset into the bvec rather than userspace address.
Therefore it can be 0.

We can skip checking whether the address is NULL before mapping
uring_cmd data. Bad userspace address will be handled properly later when
the user buffer is imported.

With this patch, we will be able to use the kernel registered bvec fixed
buffers in io_uring NVMe passthru with ublk zero-copy support.

Reviewed-by: Caleb Sander Mateos <csander@purestorage.com>
Reviewed-by: Jens Axboe <axboe@kernel.dk>
Reviewed-by: Ming Lei <ming.lei@redhat.com>
Signed-off-by: Xinyu Zhang <xizhang@purestorage.com>
Signed-off-by: Keith Busch <kbusch@kernel.org>
---
 drivers/nvme/host/ioctl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c
index fb266cf1f8c66..98a0750c0cda5 100644
--- a/drivers/nvme/host/ioctl.c
+++ b/drivers/nvme/host/ioctl.c
@@ -512,7 +512,7 @@ static int nvme_uring_cmd_io(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 		return PTR_ERR(req);
 	req->timeout = d.timeout_ms ? msecs_to_jiffies(d.timeout_ms) : 0;
 
-	if (d.addr && d.data_len) {
+	if (d.data_len) {
 		ret = nvme_map_user_request(req, d.addr,
 			d.data_len, nvme_to_user_ptr(d.metadata),
 			d.metadata_len, ioucmd, vec, issue_flags);
-- 
2.43.5

[PATCHv8 4/6] io_uring: add support for kernel registered bvecs

[Keith Busch]

From: Keith Busch <kbusch@kernel.org>

Provide an interface for the kernel to leverage the existing
pre-registered buffers that io_uring provides. User space can reference
these later to achieve zero-copy IO.

User space must register an empty fixed buffer table with io_uring in
order for the kernel to make use of it.

Signed-off-by: Keith Busch <kbusch@kernel.org>
---
 include/linux/io_uring/cmd.h |   7 ++
 io_uring/io_uring.c          |   3 +
 io_uring/rsrc.c              | 123 +++++++++++++++++++++++++++++++++--
 io_uring/rsrc.h              |   9 +++
 io_uring/rw.c                |   3 +
 5 files changed, 138 insertions(+), 7 deletions(-)

diff --git a/include/linux/io_uring/cmd.h b/include/linux/io_uring/cmd.h
index 87150dc0a07cf..cf8d80d847344 100644
--- a/include/linux/io_uring/cmd.h
+++ b/include/linux/io_uring/cmd.h
@@ -4,6 +4,7 @@
 
 #include <uapi/linux/io_uring.h>
 #include <linux/io_uring_types.h>
+#include <linux/blk-mq.h>
 
 /* only top 8 bits of sqe->uring_cmd_flags for kernel internal use */
 #define IORING_URING_CMD_CANCELABLE	(1U << 30)
@@ -125,4 +126,10 @@ static inline struct io_uring_cmd_data *io_uring_cmd_get_async_data(struct io_ur
 	return cmd_to_io_kiocb(cmd)->async_data;
 }
 
+int io_buffer_register_bvec(struct io_uring_cmd *cmd, struct request *rq,
+			    void (*release)(void *), unsigned int index,
+			    unsigned int issue_flags);
+void io_buffer_unregister_bvec(struct io_uring_cmd *cmd, unsigned int index,
+			       unsigned int issue_flags);
+
 #endif /* _LINUX_IO_URING_CMD_H */
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index db1c0792def63..2f5dd47e7dbf5 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -3947,6 +3947,9 @@ static int __init io_uring_init(void)
 
 	io_uring_optable_init();
 
+	/* imu->dir is u8 */
+	BUILD_BUG_ON((IO_IMU_DEST | IO_IMU_SOURCE) > U8_MAX);
+
 	/*
 	 * Allow user copy in the per-command field, which starts after the
 	 * file in io_kiocb and until the opcode field. The openat2 handling
diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c
index f814526982c36..0eceaf2e03777 100644
--- a/io_uring/rsrc.c
+++ b/io_uring/rsrc.c
@@ -9,6 +9,7 @@
 #include <linux/hugetlb.h>
 #include <linux/compat.h>
 #include <linux/io_uring.h>
+#include <linux/io_uring/cmd.h>
 
 #include <uapi/linux/io_uring.h>
 
@@ -101,17 +102,23 @@ int io_buffer_validate(struct iovec *iov)
 	return 0;
 }
 
-static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_rsrc_node *node)
+static void io_release_ubuf(void *priv)
 {
-	struct io_mapped_ubuf *imu = node->buf;
+	struct io_mapped_ubuf *imu = priv;
 	unsigned int i;
 
-	if (!refcount_dec_and_test(&imu->refs))
-		return;
 	for (i = 0; i < imu->nr_bvecs; i++)
 		unpin_user_page(imu->bvec[i].bv_page);
+}
+
+static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_mapped_ubuf *imu)
+{
+	if (!refcount_dec_and_test(&imu->refs))
+		return;
+
 	if (imu->acct_pages)
 		io_unaccount_mem(ctx, imu->acct_pages);
+	imu->release(imu->priv);
 	kvfree(imu);
 }
 
@@ -451,7 +458,7 @@ void io_free_rsrc_node(struct io_ring_ctx *ctx, struct io_rsrc_node *node)
 		break;
 	case IORING_RSRC_BUFFER:
 		if (node->buf)
-			io_buffer_unmap(ctx, node);
+			io_buffer_unmap(ctx, node->buf);
 		break;
 	default:
 		WARN_ON_ONCE(1);
@@ -761,6 +768,10 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,
 	imu->len = iov->iov_len;
 	imu->nr_bvecs = nr_pages;
 	imu->folio_shift = PAGE_SHIFT;
+	imu->release = io_release_ubuf;
+	imu->priv = imu;
+	imu->is_kbuf = false;
+	imu->dir = IO_IMU_DEST | IO_IMU_SOURCE;
 	if (coalesced)
 		imu->folio_shift = data.folio_shift;
 	refcount_set(&imu->refs, 1);
@@ -857,6 +868,95 @@ int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg,
 	return ret;
 }
 
+int io_buffer_register_bvec(struct io_uring_cmd *cmd, struct request *rq,
+			    void (*release)(void *), unsigned int index,
+			    unsigned int issue_flags)
+{
+	struct io_ring_ctx *ctx = cmd_to_io_kiocb(cmd)->ctx;
+	struct io_rsrc_data *data = &ctx->buf_table;
+	struct req_iterator rq_iter;
+	struct io_mapped_ubuf *imu;
+	struct io_rsrc_node *node;
+	struct bio_vec bv, *bvec;
+	u16 nr_bvecs;
+	int ret = 0;
+
+	io_ring_submit_lock(ctx, issue_flags);
+	if (index >= data->nr) {
+		ret = -EINVAL;
+		goto unlock;
+	}
+	index = array_index_nospec(index, data->nr);
+
+	if (data->nodes[index]) {
+		ret = -EBUSY;
+		goto unlock;
+	}
+
+	node = io_rsrc_node_alloc(IORING_RSRC_BUFFER);
+	if (!node) {
+		ret = -ENOMEM;
+		goto unlock;
+	}
+
+	nr_bvecs = blk_rq_nr_phys_segments(rq);
+	imu = kvmalloc(struct_size(imu, bvec, nr_bvecs), GFP_KERNEL);
+	if (!imu) {
+		kfree(node);
+		ret = -ENOMEM;
+		goto unlock;
+	}
+
+	imu->ubuf = 0;
+	imu->len = blk_rq_bytes(rq);
+	imu->acct_pages = 0;
+	imu->folio_shift = PAGE_SHIFT;
+	imu->nr_bvecs = nr_bvecs;
+	refcount_set(&imu->refs, 1);
+	imu->release = release;
+	imu->priv = rq;
+	imu->is_kbuf = true;
+
+	if (op_is_write(req_op(rq)))
+		imu->dir = IO_IMU_SOURCE;
+	else
+		imu->dir = IO_IMU_DEST;
+
+	bvec = imu->bvec;
+	rq_for_each_bvec(bv, rq, rq_iter)
+		*bvec++ = bv;
+
+	node->buf = imu;
+	data->nodes[index] = node;
+unlock:
+	io_ring_submit_unlock(ctx, issue_flags);
+	return ret;
+}
+EXPORT_SYMBOL_GPL(io_buffer_register_bvec);
+
+void io_buffer_unregister_bvec(struct io_uring_cmd *cmd, unsigned int index,
+			       unsigned int issue_flags)
+{
+	struct io_ring_ctx *ctx = cmd_to_io_kiocb(cmd)->ctx;
+	struct io_rsrc_data *data = &ctx->buf_table;
+	struct io_rsrc_node *node;
+
+	io_ring_submit_lock(ctx, issue_flags);
+	if (index >= data->nr)
+		goto unlock;
+	index = array_index_nospec(index, data->nr);
+
+	node = data->nodes[index];
+	if (!node || !node->buf->is_kbuf)
+		goto unlock;
+
+	io_put_rsrc_node(ctx, node);
+	data->nodes[index] = NULL;
+unlock:
+	io_ring_submit_unlock(ctx, issue_flags);
+}
+EXPORT_SYMBOL_GPL(io_buffer_unregister_bvec);
+
 static int io_import_fixed(int ddir, struct iov_iter *iter,
 			   struct io_mapped_ubuf *imu,
 			   u64 buf_addr, size_t len)
@@ -871,6 +971,8 @@ static int io_import_fixed(int ddir, struct iov_iter *iter,
 	/* not inside the mapped region */
 	if (unlikely(buf_addr < imu->ubuf || buf_end > (imu->ubuf + imu->len)))
 		return -EFAULT;
+	if (!(imu->dir & (1 << ddir)))
+		return -EFAULT;
 
 	/*
 	 * Might not be a start of buffer, set size appropriately
@@ -883,8 +985,8 @@ static int io_import_fixed(int ddir, struct iov_iter *iter,
 		/*
 		 * Don't use iov_iter_advance() here, as it's really slow for
 		 * using the latter parts of a big fixed buffer - it iterates
-		 * over each segment manually. We can cheat a bit here, because
-		 * we know that:
+		 * over each segment manually. We can cheat a bit here for user
+		 * registered nodes, because we know that:
 		 *
 		 * 1) it's a BVEC iter, we set it up
 		 * 2) all bvecs are the same in size, except potentially the
@@ -898,8 +1000,15 @@ static int io_import_fixed(int ddir, struct iov_iter *iter,
 		 */
 		const struct bio_vec *bvec = imu->bvec;
 
+		/*
+		 * Kernel buffer bvecs, on the other hand, don't necessarily
+		 * have the size property of user registered ones, so we have
+		 * to use the slow iter advance.
+		 */
 		if (offset < bvec->bv_len) {
 			iter->iov_offset = offset;
+		} else if (imu->is_kbuf) {
+			iov_iter_advance(iter, offset);
 		} else {
 			unsigned long seg_skip;
 
diff --git a/io_uring/rsrc.h b/io_uring/rsrc.h
index f0e9080599646..7600e2736eeb3 100644
--- a/io_uring/rsrc.h
+++ b/io_uring/rsrc.h
@@ -20,6 +20,11 @@ struct io_rsrc_node {
 	};
 };
 
+enum {
+	IO_IMU_DEST	= 1 << ITER_DEST,
+	IO_IMU_SOURCE	= 1 << ITER_SOURCE,
+};
+
 struct io_mapped_ubuf {
 	u64		ubuf;
 	unsigned int	len;
@@ -27,6 +32,10 @@ struct io_mapped_ubuf {
 	unsigned int    folio_shift;
 	refcount_t	refs;
 	unsigned long	acct_pages;
+	void		(*release)(void *);
+	void		*priv;
+	bool		is_kbuf;
+	u8		dir;
 	struct bio_vec	bvec[] __counted_by(nr_bvecs);
 };
 
diff --git a/io_uring/rw.c b/io_uring/rw.c
index 7bc23802a388e..5ee9f8949e8ba 100644
--- a/io_uring/rw.c
+++ b/io_uring/rw.c
@@ -629,6 +629,7 @@ static inline loff_t *io_kiocb_ppos(struct kiocb *kiocb)
  */
 static ssize_t loop_rw_iter(int ddir, struct io_rw *rw, struct iov_iter *iter)
 {
+	struct io_kiocb *req = cmd_to_io_kiocb(rw);
 	struct kiocb *kiocb = &rw->kiocb;
 	struct file *file = kiocb->ki_filp;
 	ssize_t ret = 0;
@@ -644,6 +645,8 @@ static ssize_t loop_rw_iter(int ddir, struct io_rw *rw, struct iov_iter *iter)
 	if ((kiocb->ki_flags & IOCB_NOWAIT) &&
 	    !(kiocb->ki_filp->f_flags & O_NONBLOCK))
 		return -EAGAIN;
+	if ((req->flags & REQ_F_BUF_NODE) && req->buf_node->buf->is_kbuf)
+		return -EFAULT;
 
 	ppos = io_kiocb_ppos(kiocb);
 
-- 
2.43.5

Re: [PATCHv8 4/6] io_uring: add support for kernel registered bvecs

[Ming Lei]

On Thu, Feb 27, 2025 at 02:39:14PM -0800, Keith Busch wrote:
> From: Keith Busch <kbusch@kernel.org>
> 
> Provide an interface for the kernel to leverage the existing
> pre-registered buffers that io_uring provides. User space can reference
> these later to achieve zero-copy IO.
> 
> User space must register an empty fixed buffer table with io_uring in
> order for the kernel to make use of it.
> 
> Signed-off-by: Keith Busch <kbusch@kernel.org>

Reviewed-by: Ming Lei <ming.lei@redhat.com>

Thanks,
Ming

[PATCHv8 5/6] ublk: zc register/unregister bvec

[Keith Busch]

From: Keith Busch <kbusch@kernel.org>

Provide new operations for the user to request mapping an active request
to an io uring instance's buf_table. The user has to provide the index
it wants to install the buffer.

A reference count is taken on the request to ensure it can't be
completed while it is active in a ring's buf_table.

Signed-off-by: Keith Busch <kbusch@kernel.org>
---
 drivers/block/ublk_drv.c      | 59 ++++++++++++++++++++++++++++++-----
 include/uapi/linux/ublk_cmd.h |  4 +++
 2 files changed, 56 insertions(+), 7 deletions(-)

diff --git a/drivers/block/ublk_drv.c b/drivers/block/ublk_drv.c
index e8f52d8341fba..8d7d6862a80f5 100644
--- a/drivers/block/ublk_drv.c
+++ b/drivers/block/ublk_drv.c
@@ -51,6 +51,9 @@
 /* private ioctl command mirror */
 #define UBLK_CMD_DEL_DEV_ASYNC	_IOC_NR(UBLK_U_CMD_DEL_DEV_ASYNC)
 
+#define UBLK_IO_REGISTER_IO_BUF		_IOC_NR(UBLK_U_IO_REGISTER_IO_BUF)
+#define UBLK_IO_UNREGISTER_IO_BUF	_IOC_NR(UBLK_U_IO_UNREGISTER_IO_BUF)
+
 /* All UBLK_F_* have to be included into UBLK_F_ALL */
 #define UBLK_F_ALL (UBLK_F_SUPPORT_ZERO_COPY \
 		| UBLK_F_URING_CMD_COMP_IN_TASK \
@@ -197,12 +200,14 @@ struct ublk_params_header {
 
 static bool ublk_abort_requests(struct ublk_device *ub, struct ublk_queue *ubq);
 
+static inline struct request *__ublk_check_and_get_req(struct ublk_device *ub,
+		struct ublk_queue *ubq, int tag, size_t offset);
 static inline unsigned int ublk_req_build_flags(struct request *req);
 static inline struct ublksrv_io_desc *ublk_get_iod(struct ublk_queue *ubq,
 						   int tag);
 static inline bool ublk_dev_is_user_copy(const struct ublk_device *ub)
 {
-	return ub->dev_info.flags & UBLK_F_USER_COPY;
+	return ub->dev_info.flags & (UBLK_F_USER_COPY | UBLK_F_SUPPORT_ZERO_COPY);
 }
 
 static inline bool ublk_dev_is_zoned(const struct ublk_device *ub)
@@ -592,7 +597,7 @@ static void ublk_apply_params(struct ublk_device *ub)
 
 static inline bool ublk_support_user_copy(const struct ublk_queue *ubq)
 {
-	return ubq->flags & UBLK_F_USER_COPY;
+	return ubq->flags & (UBLK_F_USER_COPY | UBLK_F_SUPPORT_ZERO_COPY);
 }
 
 static inline bool ublk_need_req_ref(const struct ublk_queue *ubq)
@@ -1758,6 +1763,45 @@ static inline void ublk_prep_cancel(struct io_uring_cmd *cmd,
 	io_uring_cmd_mark_cancelable(cmd, issue_flags);
 }
 
+static void ublk_io_release(void *priv)
+{
+	struct request *rq = priv;
+	struct ublk_queue *ubq = rq->mq_hctx->driver_data;
+
+	ublk_put_req_ref(ubq, rq);
+}
+
+static int ublk_register_io_buf(struct io_uring_cmd *cmd,
+				struct ublk_queue *ubq, unsigned int tag,
+				const struct ublksrv_io_cmd *ub_cmd,
+				unsigned int issue_flags)
+{
+	struct ublk_device *ub = cmd->file->private_data;
+	int index = (int)ub_cmd->addr, ret;
+	struct request *req;
+
+	req = __ublk_check_and_get_req(ub, ubq, tag, 0);
+	if (!req)
+		return -EINVAL;
+
+	ret = io_buffer_register_bvec(cmd, req, ublk_io_release, index,
+				      issue_flags);
+	if (ret) {
+		ublk_put_req_ref(ubq, req);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int ublk_unregister_io_buf(struct io_uring_cmd *cmd,
+				  const struct ublksrv_io_cmd *ub_cmd,
+				  unsigned int issue_flags)
+{
+	io_buffer_unregister_bvec(cmd, ub_cmd->addr, issue_flags);
+	return 0;
+}
+
 static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd,
 			       unsigned int issue_flags,
 			       const struct ublksrv_io_cmd *ub_cmd)
@@ -1809,6 +1853,10 @@ static int __ublk_ch_uring_cmd(struct io_uring_cmd *cmd,
 
 	ret = -EINVAL;
 	switch (_IOC_NR(cmd_op)) {
+	case UBLK_IO_REGISTER_IO_BUF:
+		return ublk_register_io_buf(cmd, ubq, tag, ub_cmd, issue_flags);
+	case UBLK_IO_UNREGISTER_IO_BUF:
+		return ublk_unregister_io_buf(cmd, ub_cmd, issue_flags);
 	case UBLK_IO_FETCH_REQ:
 		/* UBLK_IO_FETCH_REQ is only allowed before queue is setup */
 		if (ublk_queue_ready(ubq)) {
@@ -2475,7 +2523,7 @@ static int ublk_ctrl_add_dev(struct io_uring_cmd *cmd)
 		 * buffer by pwrite() to ublk char device, which can't be
 		 * used for unprivileged device
 		 */
-		if (info.flags & UBLK_F_USER_COPY)
+		if (info.flags & (UBLK_F_USER_COPY | UBLK_F_SUPPORT_ZERO_COPY))
 			return -EINVAL;
 	}
 
@@ -2543,9 +2591,6 @@ static int ublk_ctrl_add_dev(struct io_uring_cmd *cmd)
 		goto out_free_dev_number;
 	}
 
-	/* We are not ready to support zero copy */
-	ub->dev_info.flags &= ~UBLK_F_SUPPORT_ZERO_COPY;
-
 	ub->dev_info.nr_hw_queues = min_t(unsigned int,
 			ub->dev_info.nr_hw_queues, nr_cpu_ids);
 	ublk_align_max_io_size(ub);
@@ -2876,7 +2921,7 @@ static int ublk_ctrl_get_features(struct io_uring_cmd *cmd)
 {
 	const struct ublksrv_ctrl_cmd *header = io_uring_sqe_cmd(cmd->sqe);
 	void __user *argp = (void __user *)(unsigned long)header->addr;
-	u64 features = UBLK_F_ALL & ~UBLK_F_SUPPORT_ZERO_COPY;
+	u64 features = UBLK_F_ALL;
 
 	if (header->len != UBLK_FEATURES_LEN || !header->addr)
 		return -EINVAL;
diff --git a/include/uapi/linux/ublk_cmd.h b/include/uapi/linux/ublk_cmd.h
index 8093acdeaa114..7255b36b5cf63 100644
--- a/include/uapi/linux/ublk_cmd.h
+++ b/include/uapi/linux/ublk_cmd.h
@@ -94,6 +94,10 @@
 	_IOWR('u', UBLK_IO_COMMIT_AND_FETCH_REQ, struct ublksrv_io_cmd)
 #define	UBLK_U_IO_NEED_GET_DATA		\
 	_IOWR('u', UBLK_IO_NEED_GET_DATA, struct ublksrv_io_cmd)
+#define	UBLK_U_IO_REGISTER_IO_BUF	\
+	_IOWR('u', 0x23, struct ublksrv_io_cmd)
+#define	UBLK_U_IO_UNREGISTER_IO_BUF	\
+	_IOWR('u', 0x24, struct ublksrv_io_cmd)
 
 /* only ABORT means that no re-fetch */
 #define UBLK_IO_RES_OK			0
-- 
2.43.5

Re: [PATCHv8 5/6] ublk: zc register/unregister bvec

[Ming Lei]

On Thu, Feb 27, 2025 at 02:39:15PM -0800, Keith Busch wrote:
> From: Keith Busch <kbusch@kernel.org>
> 
> Provide new operations for the user to request mapping an active request
> to an io uring instance's buf_table. The user has to provide the index
> it wants to install the buffer.
> 
> A reference count is taken on the request to ensure it can't be
> completed while it is active in a ring's buf_table.
> 
> Signed-off-by: Keith Busch <kbusch@kernel.org>

Reviewed-by: Ming Lei <ming.lei@redhat.com>

Thanks,
Ming

[PATCHv8 6/6] io_uring: cache nodes and mapped buffers

[Keith Busch]

From: Keith Busch <kbusch@kernel.org>

Frequent alloc/free cycles on these is pretty costly. Use an io cache to
more efficiently reuse these buffers.

Signed-off-by: Keith Busch <kbusch@kernel.org>
---
 include/linux/io_uring_types.h |  2 +
 io_uring/filetable.c           |  2 +-
 io_uring/io_uring.c            |  2 +
 io_uring/rsrc.c                | 70 +++++++++++++++++++++++++++-------
 io_uring/rsrc.h                |  4 +-
 5 files changed, 64 insertions(+), 16 deletions(-)

diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h
index c0fe8a00fe53a..3ce87dcd99eec 100644
--- a/include/linux/io_uring_types.h
+++ b/include/linux/io_uring_types.h
@@ -294,6 +294,8 @@ struct io_ring_ctx {
 
 		struct io_file_table	file_table;
 		struct io_rsrc_data	buf_table;
+		struct io_alloc_cache	node_cache;
+		struct io_alloc_cache	imu_cache;
 
 		struct io_submit_state	submit_state;
 
diff --git a/io_uring/filetable.c b/io_uring/filetable.c
index dd8eeec97acf6..a21660e3145ab 100644
--- a/io_uring/filetable.c
+++ b/io_uring/filetable.c
@@ -68,7 +68,7 @@ static int io_install_fixed_file(struct io_ring_ctx *ctx, struct file *file,
 	if (slot_index >= ctx->file_table.data.nr)
 		return -EINVAL;
 
-	node = io_rsrc_node_alloc(IORING_RSRC_FILE);
+	node = io_rsrc_node_alloc(ctx, IORING_RSRC_FILE);
 	if (!node)
 		return -ENOMEM;
 
diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 2f5dd47e7dbf5..8f542a5f20a60 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -292,6 +292,7 @@ static void io_free_alloc_caches(struct io_ring_ctx *ctx)
 	io_alloc_cache_free(&ctx->uring_cache, kfree);
 	io_alloc_cache_free(&ctx->msg_cache, kfree);
 	io_futex_cache_free(ctx);
+	io_rsrc_cache_free(ctx);
 }
 
 static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
@@ -339,6 +340,7 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
 	ret |= io_alloc_cache_init(&ctx->msg_cache, IO_ALLOC_CACHE_MAX,
 			    sizeof(struct io_kiocb), 0);
 	ret |= io_futex_cache_init(ctx);
+	ret |= io_rsrc_cache_init(ctx);
 	if (ret)
 		goto free_ref;
 	init_completion(&ctx->ref_comp);
diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c
index 0eceaf2e03777..450b4c039334d 100644
--- a/io_uring/rsrc.c
+++ b/io_uring/rsrc.c
@@ -33,6 +33,8 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,
 #define IORING_MAX_FIXED_FILES	(1U << 20)
 #define IORING_MAX_REG_BUFFERS	(1U << 14)
 
+#define IO_CACHED_BVECS_SEGS	32
+
 int __io_account_mem(struct user_struct *user, unsigned long nr_pages)
 {
 	unsigned long page_limit, cur_pages, new_pages;
@@ -111,6 +113,22 @@ static void io_release_ubuf(void *priv)
 		unpin_user_page(imu->bvec[i].bv_page);
 }
 
+static struct io_mapped_ubuf *io_alloc_imu(struct io_ring_ctx *ctx,
+					   int nr_bvecs)
+{
+	if (nr_bvecs <= IO_CACHED_BVECS_SEGS)
+		return io_cache_alloc(&ctx->imu_cache, GFP_KERNEL);
+	return kvmalloc(struct_size_t(struct io_mapped_ubuf, bvec, nr_bvecs),
+			GFP_KERNEL);
+}
+
+static void io_free_imu(struct io_ring_ctx *ctx, struct io_mapped_ubuf *imu)
+{
+	if (imu->nr_bvecs > IO_CACHED_BVECS_SEGS ||
+	    !io_alloc_cache_put(&ctx->imu_cache, imu))
+		kvfree(imu);
+}
+
 static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_mapped_ubuf *imu)
 {
 	if (!refcount_dec_and_test(&imu->refs))
@@ -119,22 +137,44 @@ static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_mapped_ubuf *imu)
 	if (imu->acct_pages)
 		io_unaccount_mem(ctx, imu->acct_pages);
 	imu->release(imu->priv);
-	kvfree(imu);
 }
 
-struct io_rsrc_node *io_rsrc_node_alloc(int type)
+struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx, int type)
 {
 	struct io_rsrc_node *node;
 
-	node = kzalloc(sizeof(*node), GFP_KERNEL);
+	node = io_cache_alloc(&ctx->node_cache, GFP_KERNEL);
 	if (node) {
 		node->type = type;
 		node->refs = 1;
+		node->tag = 0;
+		node->file_ptr = 0;
 	}
 	return node;
 }
 
-__cold void io_rsrc_data_free(struct io_ring_ctx *ctx, struct io_rsrc_data *data)
+bool io_rsrc_cache_init(struct io_ring_ctx *ctx)
+{
+	const int imu_cache_size = struct_size_t(struct io_mapped_ubuf, bvec,
+						 IO_CACHED_BVECS_SEGS);
+	const int node_size = sizeof(struct io_rsrc_node);
+	bool ret;
+
+	ret = io_alloc_cache_init(&ctx->node_cache, IO_ALLOC_CACHE_MAX,
+				  node_size, 0);
+	ret |= io_alloc_cache_init(&ctx->imu_cache, IO_ALLOC_CACHE_MAX,
+				   imu_cache_size, 0);
+	return ret;
+}
+
+void io_rsrc_cache_free(struct io_ring_ctx *ctx)
+{
+	io_alloc_cache_free(&ctx->node_cache, kfree);
+	io_alloc_cache_free(&ctx->imu_cache, kfree);
+}
+
+__cold void io_rsrc_data_free(struct io_ring_ctx *ctx,
+			      struct io_rsrc_data *data)
 {
 	if (!data->nr)
 		return;
@@ -207,7 +247,7 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx,
 				err = -EBADF;
 				break;
 			}
-			node = io_rsrc_node_alloc(IORING_RSRC_FILE);
+			node = io_rsrc_node_alloc(ctx, IORING_RSRC_FILE);
 			if (!node) {
 				err = -ENOMEM;
 				fput(file);
@@ -465,7 +505,8 @@ void io_free_rsrc_node(struct io_ring_ctx *ctx, struct io_rsrc_node *node)
 		break;
 	}
 
-	kfree(node);
+	if (!io_alloc_cache_put(&ctx->node_cache, node))
+		kvfree(node);
 }
 
 int io_sqe_files_unregister(struct io_ring_ctx *ctx)
@@ -527,7 +568,7 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg,
 			goto fail;
 		}
 		ret = -ENOMEM;
-		node = io_rsrc_node_alloc(IORING_RSRC_FILE);
+		node = io_rsrc_node_alloc(ctx, IORING_RSRC_FILE);
 		if (!node) {
 			fput(file);
 			goto fail;
@@ -732,7 +773,7 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,
 	if (!iov->iov_base)
 		return NULL;
 
-	node = io_rsrc_node_alloc(IORING_RSRC_BUFFER);
+	node = io_rsrc_node_alloc(ctx, IORING_RSRC_BUFFER);
 	if (!node)
 		return ERR_PTR(-ENOMEM);
 	node->buf = NULL;
@@ -752,10 +793,11 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,
 			coalesced = io_coalesce_buffer(&pages, &nr_pages, &data);
 	}
 
-	imu = kvmalloc(struct_size(imu, bvec, nr_pages), GFP_KERNEL);
+	imu = io_alloc_imu(ctx, nr_pages);
 	if (!imu)
 		goto done;
 
+	imu->nr_bvecs = nr_pages;
 	ret = io_buffer_account_pin(ctx, pages, nr_pages, imu, last_hpage);
 	if (ret) {
 		unpin_user_pages(pages, nr_pages);
@@ -766,7 +808,6 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,
 	/* store original address for later verification */
 	imu->ubuf = (unsigned long) iov->iov_base;
 	imu->len = iov->iov_len;
-	imu->nr_bvecs = nr_pages;
 	imu->folio_shift = PAGE_SHIFT;
 	imu->release = io_release_ubuf;
 	imu->priv = imu;
@@ -789,7 +830,8 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,
 	}
 done:
 	if (ret) {
-		kvfree(imu);
+		if (imu)
+			io_free_imu(ctx, imu);
 		if (node)
 			io_put_rsrc_node(ctx, node);
 		node = ERR_PTR(ret);
@@ -893,14 +935,14 @@ int io_buffer_register_bvec(struct io_uring_cmd *cmd, struct request *rq,
 		goto unlock;
 	}
 
-	node = io_rsrc_node_alloc(IORING_RSRC_BUFFER);
+	node = io_rsrc_node_alloc(ctx, IORING_RSRC_BUFFER);
 	if (!node) {
 		ret = -ENOMEM;
 		goto unlock;
 	}
 
 	nr_bvecs = blk_rq_nr_phys_segments(rq);
-	imu = kvmalloc(struct_size(imu, bvec, nr_bvecs), GFP_KERNEL);
+	imu = io_alloc_imu(ctx, nr_bvecs);
 	if (!imu) {
 		kfree(node);
 		ret = -ENOMEM;
@@ -1137,7 +1179,7 @@ static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx
 		if (!src_node) {
 			dst_node = NULL;
 		} else {
-			dst_node = io_rsrc_node_alloc(IORING_RSRC_BUFFER);
+			dst_node = io_rsrc_node_alloc(ctx, IORING_RSRC_BUFFER);
 			if (!dst_node) {
 				ret = -ENOMEM;
 				goto out_free;
diff --git a/io_uring/rsrc.h b/io_uring/rsrc.h
index 7600e2736eeb3..27e545694d01e 100644
--- a/io_uring/rsrc.h
+++ b/io_uring/rsrc.h
@@ -48,7 +48,9 @@ struct io_imu_folio_data {
 	unsigned int	nr_folios;
 };
 
-struct io_rsrc_node *io_rsrc_node_alloc(int type);
+bool io_rsrc_cache_init(struct io_ring_ctx *ctx);
+void io_rsrc_cache_free(struct io_ring_ctx *ctx);
+struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx, int type);
 void io_free_rsrc_node(struct io_ring_ctx *ctx, struct io_rsrc_node *node);
 void io_rsrc_data_free(struct io_ring_ctx *ctx, struct io_rsrc_data *data);
 int io_rsrc_data_alloc(struct io_rsrc_data *data, unsigned nr);
-- 
2.43.5

Re: [PATCHv8 6/6] io_uring: cache nodes and mapped buffers

[Jens Axboe]

On 2/27/25 3:39 PM, Keith Busch wrote:
> @@ -119,22 +137,44 @@ static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_mapped_ubuf *imu)
>  	if (imu->acct_pages)
>  		io_unaccount_mem(ctx, imu->acct_pages);
>  	imu->release(imu->priv);
> -	kvfree(imu);
>  }

io_free_imu(ctx, imu);

?

-- 
Jens Axboe

Re: [PATCHv8 0/6] ublk zero copy support

[Jens Axboe]

On Thu, 27 Feb 2025 14:39:10 -0800, Keith Busch wrote:
> This one completed liburing 'make runtests' successfully.
> 
> Changes from v7:
> 
>   Prep patch, mostly from Jens, that removes the "do_import" parameter
>   from the generic rw prep.
> 
> [...]

Applied, thanks!

[1/6] io_uring/rw: move buffer_select outside generic prep
      commit: 2a61e63891add7817e35a2323347ed8d354acf84
[2/6] io_uring/rw: move fixed buffer import to issue path
      commit: ff92d824d0b55e35ed2ee77021cbd2ed3e7ae7a2
[3/6] nvme: map uring_cmd data even if address is 0
      commit: 99fde895ff56ac2241e7b7b4566731d72f2fdaa7
[4/6] io_uring: add support for kernel registered bvecs
      commit: 27cb27b6d5ea401143ca3648983342bb820c4be9
[5/6] ublk: zc register/unregister bvec
      commit: 1f6540e2aabb7372e68223a3669019589c3e30ad
[6/6] io_uring: cache nodes and mapped buffers
      commit: ed9f3112a8a8f6e6919d3b9da2651fa302df7be3

Best regards,
-- 
Jens Axboe