5b15e2ecaf
Notable changes: * The only big change is the switch to using a custom QIOChannel for savevm-async, because the previously used QEMUFileOps was dropped. Changes to the current implementation: * Switch to vector based methods as required for an IO channel. For short reads the passed-in IO vector is stuffed with zeroes at the end, just to be sure. * For reading: The documentation in include/io/channel.h states that at least one byte should be read, so also error out when whe are at the very end instead of returning 0. * For reading: Fix off-by-one error when request goes beyond end. The wrong code piece was: if ((pos + size) > maxlen) { size = maxlen - pos - 1; } Previously, the last byte would not be read. It's actually possible to get a snapshot .raw file that has content all the way up the final 512 byte (= BDRV_SECTOR_SIZE) boundary without any trailing zero bytes (I wrote a script to do it). Luckily, it didn't cause a real issue, because qemu_loadvm_state() is not interested in the final (i.e. QEMU_VM_VMDESCRIPTION) section. The buffer for reading it is simply freed up afterwards and the function will assume that it read the whole section, even if that's not the case. * For writing: Make use of the generated blk_pwritev() wrapper instead of manually wrapping the coroutine to simplify and save a few lines. * Adapt to changed interfaces for blk_{pread,pwrite}: * a9262f551e ("block: Change blk_{pread,pwrite}() param order") * 3b35d4542c ("block: Add a 'flags' param to blk_pread()") * bf5b16fa40 ("block: Make blk_{pread,pwrite}() return 0 on success") Those changes especially affected the qemu-img dd patches, because the context also changed, but also some of our block drivers used the functions. * Drop qemu-common.h include: it got renamed after essentially everything was moved to other headers. The only remaining user I could find for things dropped from the header between 7.0 and 7.1 was qemu_get_vm_name() in the iscsi-initiatorname patch, but it already includes the header to which the function was moved. Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
217 lines
7.7 KiB
Diff
217 lines
7.7 KiB
Diff
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
|
|
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
|
|
Date: Mon, 4 May 2020 11:05:08 +0200
|
|
Subject: [PATCH] PVE: add optional buffer size to QEMUFile
|
|
|
|
So we can use a 4M buffer for savevm-async which should
|
|
increase performance storing the state onto ceph.
|
|
|
|
Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
|
|
[increase max IOV count in QEMUFile to actually write more data]
|
|
Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
|
|
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
|
|
[FE: adapt to removal of QEMUFileOps]
|
|
Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
|
|
---
|
|
migration/qemu-file.c | 49 +++++++++++++++++++++++++++-------------
|
|
migration/qemu-file.h | 2 ++
|
|
migration/savevm-async.c | 5 ++--
|
|
3 files changed, 38 insertions(+), 18 deletions(-)
|
|
|
|
diff --git a/migration/qemu-file.c b/migration/qemu-file.c
|
|
index 4f400c2e52..21e8998867 100644
|
|
--- a/migration/qemu-file.c
|
|
+++ b/migration/qemu-file.c
|
|
@@ -31,8 +31,8 @@
|
|
#include "trace.h"
|
|
#include "qapi/error.h"
|
|
|
|
-#define IO_BUF_SIZE 32768
|
|
-#define MAX_IOV_SIZE MIN_CONST(IOV_MAX, 64)
|
|
+#define DEFAULT_IO_BUF_SIZE 32768
|
|
+#define MAX_IOV_SIZE MIN_CONST(IOV_MAX, 256)
|
|
|
|
struct QEMUFile {
|
|
const QEMUFileHooks *hooks;
|
|
@@ -55,7 +55,8 @@ struct QEMUFile {
|
|
|
|
int buf_index;
|
|
int buf_size; /* 0 when writing */
|
|
- uint8_t buf[IO_BUF_SIZE];
|
|
+ size_t buf_allocated_size;
|
|
+ uint8_t *buf;
|
|
|
|
DECLARE_BITMAP(may_free, MAX_IOV_SIZE);
|
|
struct iovec iov[MAX_IOV_SIZE];
|
|
@@ -106,7 +107,9 @@ bool qemu_file_mode_is_not_valid(const char *mode)
|
|
return false;
|
|
}
|
|
|
|
-static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
|
|
+static QEMUFile *qemu_file_new_impl(QIOChannel *ioc,
|
|
+ bool is_writable,
|
|
+ size_t buffer_size)
|
|
{
|
|
QEMUFile *f;
|
|
|
|
@@ -115,6 +118,8 @@ static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
|
|
object_ref(ioc);
|
|
f->ioc = ioc;
|
|
f->is_writable = is_writable;
|
|
+ f->buf_allocated_size = buffer_size;
|
|
+ f->buf = malloc(buffer_size);
|
|
|
|
return f;
|
|
}
|
|
@@ -125,17 +130,27 @@ static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
|
|
*/
|
|
QEMUFile *qemu_file_get_return_path(QEMUFile *f)
|
|
{
|
|
- return qemu_file_new_impl(f->ioc, !f->is_writable);
|
|
+ return qemu_file_new_impl(f->ioc, !f->is_writable, DEFAULT_IO_BUF_SIZE);
|
|
}
|
|
|
|
QEMUFile *qemu_file_new_output(QIOChannel *ioc)
|
|
{
|
|
- return qemu_file_new_impl(ioc, true);
|
|
+ return qemu_file_new_impl(ioc, true, DEFAULT_IO_BUF_SIZE);
|
|
+}
|
|
+
|
|
+QEMUFile *qemu_file_new_output_sized(QIOChannel *ioc, size_t buffer_size)
|
|
+{
|
|
+ return qemu_file_new_impl(ioc, true, buffer_size);
|
|
}
|
|
|
|
QEMUFile *qemu_file_new_input(QIOChannel *ioc)
|
|
{
|
|
- return qemu_file_new_impl(ioc, false);
|
|
+ return qemu_file_new_impl(ioc, false, DEFAULT_IO_BUF_SIZE);
|
|
+}
|
|
+
|
|
+QEMUFile *qemu_file_new_input_sized(QIOChannel *ioc, size_t buffer_size)
|
|
+{
|
|
+ return qemu_file_new_impl(ioc, false, buffer_size);
|
|
}
|
|
|
|
void qemu_file_set_hooks(QEMUFile *f, const QEMUFileHooks *hooks)
|
|
@@ -393,7 +408,7 @@ static ssize_t qemu_fill_buffer(QEMUFile *f)
|
|
do {
|
|
len = qio_channel_read(f->ioc,
|
|
(char *)f->buf + pending,
|
|
- IO_BUF_SIZE - pending,
|
|
+ f->buf_allocated_size - pending,
|
|
&local_error);
|
|
if (len == QIO_CHANNEL_ERR_BLOCK) {
|
|
if (qemu_in_coroutine()) {
|
|
@@ -443,6 +458,8 @@ int qemu_fclose(QEMUFile *f)
|
|
}
|
|
g_clear_pointer(&f->ioc, object_unref);
|
|
|
|
+ free(f->buf);
|
|
+
|
|
/* If any error was spotted before closing, we should report it
|
|
* instead of the close() return value.
|
|
*/
|
|
@@ -497,7 +514,7 @@ static void add_buf_to_iovec(QEMUFile *f, size_t len)
|
|
{
|
|
if (!add_to_iovec(f, f->buf + f->buf_index, len, false)) {
|
|
f->buf_index += len;
|
|
- if (f->buf_index == IO_BUF_SIZE) {
|
|
+ if (f->buf_index == f->buf_allocated_size) {
|
|
qemu_fflush(f);
|
|
}
|
|
}
|
|
@@ -523,7 +540,7 @@ void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, size_t size)
|
|
}
|
|
|
|
while (size > 0) {
|
|
- l = IO_BUF_SIZE - f->buf_index;
|
|
+ l = f->buf_allocated_size - f->buf_index;
|
|
if (l > size) {
|
|
l = size;
|
|
}
|
|
@@ -570,8 +587,8 @@ size_t qemu_peek_buffer(QEMUFile *f, uint8_t **buf, size_t size, size_t offset)
|
|
size_t index;
|
|
|
|
assert(!qemu_file_is_writable(f));
|
|
- assert(offset < IO_BUF_SIZE);
|
|
- assert(size <= IO_BUF_SIZE - offset);
|
|
+ assert(offset < f->buf_allocated_size);
|
|
+ assert(size <= f->buf_allocated_size - offset);
|
|
|
|
/* The 1st byte to read from */
|
|
index = f->buf_index + offset;
|
|
@@ -621,7 +638,7 @@ size_t qemu_get_buffer(QEMUFile *f, uint8_t *buf, size_t size)
|
|
size_t res;
|
|
uint8_t *src;
|
|
|
|
- res = qemu_peek_buffer(f, &src, MIN(pending, IO_BUF_SIZE), 0);
|
|
+ res = qemu_peek_buffer(f, &src, MIN(pending, f->buf_allocated_size), 0);
|
|
if (res == 0) {
|
|
return done;
|
|
}
|
|
@@ -655,7 +672,7 @@ size_t qemu_get_buffer(QEMUFile *f, uint8_t *buf, size_t size)
|
|
*/
|
|
size_t qemu_get_buffer_in_place(QEMUFile *f, uint8_t **buf, size_t size)
|
|
{
|
|
- if (size < IO_BUF_SIZE) {
|
|
+ if (size < f->buf_allocated_size) {
|
|
size_t res;
|
|
uint8_t *src = NULL;
|
|
|
|
@@ -680,7 +697,7 @@ int qemu_peek_byte(QEMUFile *f, int offset)
|
|
int index = f->buf_index + offset;
|
|
|
|
assert(!qemu_file_is_writable(f));
|
|
- assert(offset < IO_BUF_SIZE);
|
|
+ assert(offset < f->buf_allocated_size);
|
|
|
|
if (index >= f->buf_size) {
|
|
qemu_fill_buffer(f);
|
|
@@ -832,7 +849,7 @@ static int qemu_compress_data(z_stream *stream, uint8_t *dest, size_t dest_len,
|
|
ssize_t qemu_put_compression_data(QEMUFile *f, z_stream *stream,
|
|
const uint8_t *p, size_t size)
|
|
{
|
|
- ssize_t blen = IO_BUF_SIZE - f->buf_index - sizeof(int32_t);
|
|
+ ssize_t blen = f->buf_allocated_size - f->buf_index - sizeof(int32_t);
|
|
|
|
if (blen < compressBound(size)) {
|
|
return -1;
|
|
diff --git a/migration/qemu-file.h b/migration/qemu-file.h
|
|
index fa13d04d78..914f1a63a8 100644
|
|
--- a/migration/qemu-file.h
|
|
+++ b/migration/qemu-file.h
|
|
@@ -63,7 +63,9 @@ typedef struct QEMUFileHooks {
|
|
} QEMUFileHooks;
|
|
|
|
QEMUFile *qemu_file_new_input(QIOChannel *ioc);
|
|
+QEMUFile *qemu_file_new_input_sized(QIOChannel *ioc, size_t buffer_size);
|
|
QEMUFile *qemu_file_new_output(QIOChannel *ioc);
|
|
+QEMUFile *qemu_file_new_output_sized(QIOChannel *ioc, size_t buffer_size);
|
|
void qemu_file_set_hooks(QEMUFile *f, const QEMUFileHooks *hooks);
|
|
int qemu_fclose(QEMUFile *f);
|
|
|
|
diff --git a/migration/savevm-async.c b/migration/savevm-async.c
|
|
index b3692739a0..e65a5e3482 100644
|
|
--- a/migration/savevm-async.c
|
|
+++ b/migration/savevm-async.c
|
|
@@ -367,7 +367,7 @@ void qmp_savevm_start(bool has_statefile, const char *statefile, Error **errp)
|
|
|
|
QIOChannel *ioc = QIO_CHANNEL(qio_channel_savevm_async_new(snap_state.target,
|
|
&snap_state.bs_pos));
|
|
- snap_state.file = qemu_file_new_output(ioc);
|
|
+ snap_state.file = qemu_file_new_output_sized(ioc, 4 * 1024 * 1024);
|
|
|
|
if (!snap_state.file) {
|
|
error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
|
|
@@ -500,7 +500,8 @@ int load_snapshot_from_blockdev(const char *filename, Error **errp)
|
|
blk_op_block_all(be, blocker);
|
|
|
|
/* restore the VM state */
|
|
- f = qemu_file_new_input(QIO_CHANNEL(qio_channel_savevm_async_new(be, &bs_pos)));
|
|
+ f = qemu_file_new_input_sized(QIO_CHANNEL(qio_channel_savevm_async_new(be, &bs_pos)),
|
|
+ 4 * 1024 * 1024);
|
|
if (!f) {
|
|
error_setg(errp, "Could not open VM state file");
|
|
goto the_end;
|