From: Pierre Riteau <Pierre.Riteau@irisa.fr>
To: qemu-devel@nongnu.org
Cc: Pierre Riteau <Pierre.Riteau@irisa.fr>
Subject: [Qemu-devel] [PATCH 2/2] migration: Fix calculation of bytes_transferred
Date: Wed, 12 May 2010 15:12:44 +0200 [thread overview]
Message-ID: <1273669964-29783-1-git-send-email-Pierre.Riteau@irisa.fr> (raw)
When a page with all identical bytes is transferred, it is counted
as a full page (TARGET_PAGE_SIZE) although only one byte is actually
sent. Fix this by changing ram_save_block() to return the number of
bytes sent instead of a boolean value. This makes bandwidth
estimation, and consequently downtime estimation, more precise.
Signed-off-by: Pierre Riteau <Pierre.Riteau@irisa.fr>
---
arch_init.c | 21 ++++++++++++---------
1 files changed, 12 insertions(+), 9 deletions(-)
diff --git a/arch_init.c b/arch_init.c
index cf6b7b0..76317af 100644
--- a/arch_init.c
+++ b/arch_init.c
@@ -108,7 +108,7 @@ static int ram_save_block(QEMUFile *f)
static ram_addr_t current_addr = 0;
ram_addr_t saved_addr = current_addr;
ram_addr_t addr = 0;
- int found = 0;
+ int bytes_sent = 0;
while (addr < last_ram_offset) {
if (cpu_physical_memory_get_dirty(current_addr, MIGRATION_DIRTY_FLAG)) {
@@ -123,19 +123,20 @@ static int ram_save_block(QEMUFile *f)
if (is_dup_page(p, *p)) {
qemu_put_be64(f, current_addr | RAM_SAVE_FLAG_COMPRESS);
qemu_put_byte(f, *p);
+ bytes_sent = 1;
} else {
qemu_put_be64(f, current_addr | RAM_SAVE_FLAG_PAGE);
qemu_put_buffer(f, p, TARGET_PAGE_SIZE);
+ bytes_sent = TARGET_PAGE_SIZE;
}
- found = 1;
break;
}
addr += TARGET_PAGE_SIZE;
current_addr = (saved_addr + addr) % last_ram_offset;
}
- return found;
+ return bytes_sent;
}
static uint64_t bytes_transferred;
@@ -206,11 +207,11 @@ int ram_save_live(Monitor *mon, QEMUFile *f, int stage, void *opaque)
bwidth = qemu_get_clock_ns(rt_clock);
while (!qemu_file_rate_limit(f)) {
- int ret;
+ int bytes_sent;
- ret = ram_save_block(f);
- bytes_transferred += ret * TARGET_PAGE_SIZE;
- if (ret == 0) { /* no more blocks */
+ bytes_sent = ram_save_block(f);
+ bytes_transferred += bytes_sent;
+ if (bytes_sent == 0) { /* no more blocks */
break;
}
}
@@ -226,9 +227,11 @@ int ram_save_live(Monitor *mon, QEMUFile *f, int stage, void *opaque)
/* try transferring iterative blocks of memory */
if (stage == 3) {
+ int bytes_sent;
+
/* flush all remaining blocks regardless of rate limiting */
- while (ram_save_block(f) != 0) {
- bytes_transferred += TARGET_PAGE_SIZE;
+ while ((bytes_sent = ram_save_block(f)) != 0) {
+ bytes_transferred += bytes_sent;
}
cpu_physical_memory_set_dirty_tracking(0);
}
--
1.7.1
next reply other threads:[~2010-05-12 13:12 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-05-12 13:12 Pierre Riteau [this message]
2010-05-19 11:13 ` [Qemu-devel] [PATCH 2/2] migration: Fix calculation of bytes_transferred Pierre Riteau
2010-06-01 18:26 ` Anthony Liguori
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=1273669964-29783-1-git-send-email-Pierre.Riteau@irisa.fr \
--to=pierre.riteau@irisa.fr \
--cc=qemu-devel@nongnu.org \
/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).