From: Maxim Levitsky <mlevitsk@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>, qemu-devel@nongnu.org
Cc: kwolf@redhat.com, vsementsov@virtuozzo.com,
Max Reitz <mreitz@redhat.com>,
qemu-block@nongnu.org, Tom Yan <tom.ty89@gmail.com>
Subject: Re: [PATCH v4 3/7] block: add max_hw_transfer to BlockLimits
Date: Wed, 09 Jun 2021 19:12:55 +0300 [thread overview]
Message-ID: <9378247cb67d5b1afe7f4d83fab22a16ed610ae3.camel@redhat.com> (raw)
In-Reply-To: <20210608131634.423904-4-pbonzini@redhat.com>
On Tue, 2021-06-08 at 15:16 +0200, Paolo Bonzini wrote:
> For block host devices, I/O can happen through either the kernel file
> descriptor I/O system calls (preadv/pwritev, io_submit, io_uring)
> or the SCSI passthrough ioctl SG_IO.
>
> In the latter case, the size of each transfer can be limited by the
> HBA, while for file descriptor I/O the kernel is able to split and
> merge I/O in smaller pieces as needed. Applying the HBA limits to
> file descriptor I/O results in more system calls and suboptimal
> performance, so this patch splits the max_transfer limit in two:
> max_transfer remains valid and is used in general, while max_hw_transfer
> is limited to the maximum hardware size. max_hw_transfer can then be
> included by the scsi-generic driver in the block limits page, to ensure
> that the stricter hardware limit is used.
>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This is mostly the same as my patch
https://www.mail-archive.com/qemu-devel@nongnu.org/msg768264.html
I called this max_ioctl_transfer, since this limit is only relevant
to the .ioctl, but max_hw_transfer is fine as well.
So this patch looks OK, but I might have missed something
as I haven't touched this area for a long time.
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Best regards,
Maxim Levitsky
> ---
> block/block-backend.c | 12 ++++++++++++
> block/file-posix.c | 2 +-
> block/io.c | 1 +
> hw/scsi/scsi-generic.c | 2 +-
> include/block/block_int.h | 7 +++++++
> include/sysemu/block-backend.h | 1 +
> 6 files changed, 23 insertions(+), 2 deletions(-)
>
> diff --git a/block/block-backend.c b/block/block-backend.c
> index 15f1ea4288..2ea1412a54 100644
> --- a/block/block-backend.c
> +++ b/block/block-backend.c
> @@ -1953,6 +1953,18 @@ uint32_t blk_get_request_alignment(BlockBackend *blk)
> return bs ? bs->bl.request_alignment : BDRV_SECTOR_SIZE;
> }
>
> +/* Returns the maximum hardware transfer length, in bytes; guaranteed nonzero */
> +uint64_t blk_get_max_hw_transfer(BlockBackend *blk)
> +{
> + BlockDriverState *bs = blk_bs(blk);
> + uint64_t max = INT_MAX;
> +
> + if (bs) {
> + max = MIN_NON_ZERO(bs->bl.max_hw_transfer, bs->bl.max_transfer);
> + }
> + return max;
> +}
> +
> /* Returns the maximum transfer length, in bytes; guaranteed nonzero */
> uint32_t blk_get_max_transfer(BlockBackend *blk)
> {
> diff --git a/block/file-posix.c b/block/file-posix.c
> index 670c577bfe..c9746d3eb6 100644
> --- a/block/file-posix.c
> +++ b/block/file-posix.c
> @@ -1234,7 +1234,7 @@ static void raw_refresh_limits(BlockDriverState *bs, Error **errp)
> int ret = sg_get_max_transfer_length(s->fd);
>
> if (ret > 0 && ret <= BDRV_REQUEST_MAX_BYTES) {
> - bs->bl.max_transfer = pow2floor(ret);
> + bs->bl.max_hw_transfer = pow2floor(ret);
> }
>
> ret = sg_get_max_segments(s->fd);
> diff --git a/block/io.c b/block/io.c
> index 323854d063..089b99bb0c 100644
> --- a/block/io.c
> +++ b/block/io.c
> @@ -127,6 +127,7 @@ static void bdrv_merge_limits(BlockLimits *dst, const BlockLimits *src)
> {
> dst->opt_transfer = MAX(dst->opt_transfer, src->opt_transfer);
> dst->max_transfer = MIN_NON_ZERO(dst->max_transfer, src->max_transfer);
> + dst->max_hw_transfer = MIN_NON_ZERO(dst->max_hw_transfer, src->max_hw_transfer);
> dst->opt_mem_alignment = MAX(dst->opt_mem_alignment,
> src->opt_mem_alignment);
> dst->min_mem_alignment = MAX(dst->min_mem_alignment,
> diff --git a/hw/scsi/scsi-generic.c b/hw/scsi/scsi-generic.c
> index 82e1e2ee79..3762dce749 100644
> --- a/hw/scsi/scsi-generic.c
> +++ b/hw/scsi/scsi-generic.c
> @@ -179,7 +179,7 @@ static void scsi_handle_inquiry_reply(SCSIGenericReq *r, SCSIDevice *s)
> (r->req.cmd.buf[1] & 0x01)) {
> page = r->req.cmd.buf[2];
> if (page == 0xb0) {
> - uint32_t max_transfer = blk_get_max_transfer(s->conf.blk);
> + uint64_t max_transfer = blk_get_max_hw_transfer(s->conf.blk);
> uint32_t max_iov = blk_get_max_iov(s->conf.blk);
>
> assert(max_transfer);
> diff --git a/include/block/block_int.h b/include/block/block_int.h
> index 057d88b1fc..f1a54db0f8 100644
> --- a/include/block/block_int.h
> +++ b/include/block/block_int.h
> @@ -695,6 +695,13 @@ typedef struct BlockLimits {
> * clamped down. */
> uint32_t max_transfer;
>
> + /* Maximal hardware transfer length in bytes. Applies whenever
> + * transfers to the device bypass the kernel I/O scheduler, for
> + * example with SG_IO. If larger than max_transfer or if zero,
> + * blk_get_max_hw_transfer will fall back to max_transfer.
> + */
> + uint64_t max_hw_transfer;
> +
> /* memory alignment, in bytes so that no bounce buffer is needed */
> size_t min_mem_alignment;
>
> diff --git a/include/sysemu/block-backend.h b/include/sysemu/block-backend.h
> index 5423e3d9c6..9ac5f7bbd3 100644
> --- a/include/sysemu/block-backend.h
> +++ b/include/sysemu/block-backend.h
> @@ -208,6 +208,7 @@ void blk_eject(BlockBackend *blk, bool eject_flag);
> int blk_get_flags(BlockBackend *blk);
> uint32_t blk_get_request_alignment(BlockBackend *blk);
> uint32_t blk_get_max_transfer(BlockBackend *blk);
> +uint64_t blk_get_max_hw_transfer(BlockBackend *blk);
> int blk_get_max_iov(BlockBackend *blk);
> void blk_set_guest_block_size(BlockBackend *blk, int align);
> void *blk_try_blockalign(BlockBackend *blk, size_t size);
next prev parent reply other threads:[~2021-06-09 16:22 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-08 13:16 [PATCH v4 0/7] block: file-posix queue Paolo Bonzini
2021-06-08 13:16 ` [PATCH v4 1/7] file-posix: fix max_iov for /dev/sg devices Paolo Bonzini
2021-06-08 17:34 ` Eric Blake
2021-06-08 19:14 ` Vladimir Sementsov-Ogievskiy
2021-06-09 16:08 ` Maxim Levitsky
2021-06-23 15:42 ` Max Reitz
2021-06-24 7:33 ` Vladimir Sementsov-Ogievskiy
2021-06-08 13:16 ` [PATCH v4 2/7] scsi-generic: pass max_segments via max_iov field in BlockLimits Paolo Bonzini
2021-06-08 17:42 ` Eric Blake
2021-06-09 16:10 ` Maxim Levitsky
2021-06-08 13:16 ` [PATCH v4 3/7] block: add max_hw_transfer to BlockLimits Paolo Bonzini
2021-06-08 17:48 ` Eric Blake
2021-06-09 16:12 ` Maxim Levitsky [this message]
2021-06-08 13:16 ` [PATCH v4 4/7] file-posix: try BLKSECTGET on block devices too, do not round to power of 2 Paolo Bonzini
2021-06-08 17:53 ` Eric Blake
2021-06-09 16:15 ` Maxim Levitsky
2021-06-08 13:16 ` [PATCH v4 5/7] block: feature detection for host block support Paolo Bonzini
2021-06-23 15:44 ` Max Reitz
2021-06-08 13:16 ` [PATCH v4 6/7] block: check for sys/disk.h Paolo Bonzini
2021-06-08 13:16 ` [PATCH v4 7/7] block: detect DKIOCGETBLOCKCOUNT/SIZE before use Paolo Bonzini
2021-06-23 15:47 ` Max Reitz
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=9378247cb67d5b1afe7f4d83fab22a16ed610ae3.camel@redhat.com \
--to=mlevitsk@redhat.com \
--cc=kwolf@redhat.com \
--cc=mreitz@redhat.com \
--cc=pbonzini@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.org \
--cc=tom.ty89@gmail.com \
--cc=vsementsov@virtuozzo.com \
/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).