From: Stefan Reiter <s.reiter@proxmox.com>
To: pve-devel@lists.proxmox.com, pbs-devel@lists.proxmox.com
Subject: [pve-devel] [PATCH qemu 02/11] PVE: block/pbs: fast-path reads without allocation if possible
Date: Mon, 11 Jan 2021 12:14:00 +0100 [thread overview]
Message-ID: <20210111111409.32385-3-s.reiter@proxmox.com> (raw)
In-Reply-To: <20210111111409.32385-1-s.reiter@proxmox.com>
...and switch over to g_malloc/g_free while at it to align with other
QEMU code.
Tracing shows the fast-path is taken almost all the time, though not
100% so the slow one is still necessary.
Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
---
For that slight unnoticable performance boost :)
block/pbs.c | 17 ++++++++++++++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/block/pbs.c b/block/pbs.c
index 1481a2bfd1..fbf0d8d845 100644
--- a/block/pbs.c
+++ b/block/pbs.c
@@ -200,7 +200,16 @@ static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
BDRVPBSState *s = bs->opaque;
int ret;
char *pbs_error = NULL;
- uint8_t *buf = malloc(bytes);
+ uint8_t *buf;
+ bool inline_buf = true;
+
+ /* for single-buffer IO vectors we can fast-path the write directly to it */
+ if (qiov->niov == 1 && qiov->iov->iov_len >= bytes) {
+ buf = qiov->iov->iov_base;
+ } else {
+ inline_buf = false;
+ buf = g_malloc(bytes);
+ }
ReadCallbackData rcb = {
.co = qemu_coroutine_self(),
@@ -218,8 +227,10 @@ static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
return -EIO;
}
- qemu_iovec_from_buf(qiov, 0, buf, bytes);
- free(buf);
+ if (!inline_buf) {
+ qemu_iovec_from_buf(qiov, 0, buf, bytes);
+ g_free(buf);
+ }
return ret;
}
--
2.20.1
WARNING: multiple messages have this Message-ID
From: Stefan Reiter <s.reiter@proxmox.com>
To: pve-devel@lists.proxmox.com, pbs-devel@lists.proxmox.com
Subject: [pbs-devel] [PATCH qemu 02/11] PVE: block/pbs: fast-path reads without allocation if possible
Date: Mon, 11 Jan 2021 12:14:00 +0100 [thread overview]
Message-ID: <20210111111409.32385-3-s.reiter@proxmox.com> (raw)
In-Reply-To: <20210111111409.32385-1-s.reiter@proxmox.com>
...and switch over to g_malloc/g_free while at it to align with other
QEMU code.
Tracing shows the fast-path is taken almost all the time, though not
100% so the slow one is still necessary.
Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
---
For that slight unnoticable performance boost :)
block/pbs.c | 17 ++++++++++++++---
1 file changed, 14 insertions(+), 3 deletions(-)
diff --git a/block/pbs.c b/block/pbs.c
index 1481a2bfd1..fbf0d8d845 100644
--- a/block/pbs.c
+++ b/block/pbs.c
@@ -200,7 +200,16 @@ static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
BDRVPBSState *s = bs->opaque;
int ret;
char *pbs_error = NULL;
- uint8_t *buf = malloc(bytes);
+ uint8_t *buf;
+ bool inline_buf = true;
+
+ /* for single-buffer IO vectors we can fast-path the write directly to it */
+ if (qiov->niov == 1 && qiov->iov->iov_len >= bytes) {
+ buf = qiov->iov->iov_base;
+ } else {
+ inline_buf = false;
+ buf = g_malloc(bytes);
+ }
ReadCallbackData rcb = {
.co = qemu_coroutine_self(),
@@ -218,8 +227,10 @@ static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
return -EIO;
}
- qemu_iovec_from_buf(qiov, 0, buf, bytes);
- free(buf);
+ if (!inline_buf) {
+ qemu_iovec_from_buf(qiov, 0, buf, bytes);
+ g_free(buf);
+ }
return ret;
}
--
2.20.1
next prev parent reply other threads:[~2021-01-11 11:15 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-11 11:13 [pve-devel] [PATCH 00/11] live-restore for PBS snapshots Stefan Reiter
2021-01-11 11:13 ` [pbs-devel] " Stefan Reiter
2021-01-11 11:13 ` [pve-devel] [PATCH qemu 01/11] PVE: explicitly add libuuid as linking dependency Stefan Reiter
2021-01-11 11:13 ` [pbs-devel] " Stefan Reiter
2021-01-12 12:04 ` [pve-devel] " Thomas Lamprecht
2021-01-12 12:04 ` Thomas Lamprecht
2021-01-18 10:27 ` [pve-devel] [PATCH pve-qemu] explicitly specify " Stefan Reiter
2021-01-27 8:29 ` Stefan Reiter
2021-01-11 11:14 ` Stefan Reiter [this message]
2021-01-11 11:14 ` [pbs-devel] [PATCH qemu 02/11] PVE: block/pbs: fast-path reads without allocation if possible Stefan Reiter
2021-01-12 9:29 ` [pve-devel] " Wolfgang Bumiller
2021-01-12 9:29 ` [pbs-devel] " Wolfgang Bumiller
2021-01-11 11:14 ` [pve-devel] [PATCH qemu 03/11] block: add alloc-track driver Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-12 10:54 ` [pve-devel] " Wolfgang Bumiller
2021-01-12 10:54 ` [pbs-devel] " Wolfgang Bumiller
2021-01-12 11:29 ` Stefan Reiter
2021-01-12 11:29 ` [pbs-devel] " Stefan Reiter
2021-01-12 13:42 ` Wolfgang Bumiller
2021-01-12 13:42 ` [pbs-devel] " Wolfgang Bumiller
2021-01-11 11:14 ` [pve-devel] [PATCH proxmox-backup 04/11] RemoteChunkReader: add LRU cached variant Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-11 11:14 ` [pve-devel] [PATCH proxmox-backup-qemu 05/11] access: use bigger cache and LRU chunk reader Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-11 11:14 ` [pve-devel] [PATCH qemu-server 06/11] make qemu_drive_mirror_monitor more generic Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-12 13:19 ` [pve-devel] " Wolfgang Bumiller
2021-01-12 13:19 ` [pbs-devel] " Wolfgang Bumiller
2021-01-11 11:14 ` [pve-devel] [PATCH qemu-server 07/11] cfg2cmd: allow PBS snapshots as backing files for drives Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-28 16:25 ` [pve-devel] " Thomas Lamprecht
2021-01-28 16:25 ` [pbs-devel] " Thomas Lamprecht
2021-01-11 11:14 ` [pve-devel] [PATCH qemu-server 08/11] enable live-restore for PBS Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-11 11:14 ` [pve-devel] [PATCH qemu-server 09/11] extract register_qmeventd_handle to QemuServer.pm Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-11 11:14 ` [pve-devel] [PATCH qemu-server 10/11] live-restore: register qmeventd handle Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-11 11:14 ` [pve-devel] [PATCH manager 11/11] ui: restore: add live-restore checkbox Stefan Reiter
2021-01-11 11:14 ` [pbs-devel] " Stefan Reiter
2021-01-11 15:50 ` [pve-devel] [PATCH 00/11] live-restore for PBS snapshots aderumier
2021-01-11 15:50 ` [pbs-devel] " aderumier
2021-01-11 16:42 ` Stefan Reiter
2021-01-12 9:10 ` aderumier
2021-01-12 10:31 ` [pve-devel] [pbs-devel] " Thomas Lamprecht
2021-01-12 10:31 ` [pbs-devel] [pve-devel] " Thomas Lamprecht
2021-01-12 11:23 ` [pve-devel] [pbs-devel] " Thomas Lamprecht
2021-01-12 11:23 ` [pbs-devel] [pve-devel] " Thomas Lamprecht
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=20210111111409.32385-3-s.reiter@proxmox.com \
--to=s.reiter@proxmox.com \
--cc=pbs-devel@lists.proxmox.com \
--cc=pve-devel@lists.proxmox.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.