From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
Subject: [PULL 06/12] icount: get rid of static variable
Date: Thu, 1 Apr 2021 13:22:17 +0200 [thread overview]
Message-ID: <20210401112223.55711-7-pbonzini@redhat.com> (raw)
In-Reply-To: <20210401112223.55711-1-pbonzini@redhat.com>
From: Pavel Dovgalyuk <pavel.dovgalyuk@ispras.ru>
This patch moves static last_delta variable into timers_state
structure to allow correct vmstate operations with icount shift=auto enabled.
Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
Message-Id: <161701335066.1180180.7104085247702343395.stgit@pasha-ThinkPad-X280>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
softmmu/cpu-timers.c | 5 +++--
softmmu/icount.c | 9 +++------
softmmu/timers-state.h | 2 ++
3 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/softmmu/cpu-timers.c b/softmmu/cpu-timers.c
index cd38595245..34ddfa02f1 100644
--- a/softmmu/cpu-timers.c
+++ b/softmmu/cpu-timers.c
@@ -188,11 +188,12 @@ static const VMStateDescription icount_vmstate_adjust_timers = {
static const VMStateDescription icount_vmstate_shift = {
.name = "timer/icount/shift",
- .version_id = 1,
- .minimum_version_id = 1,
+ .version_id = 2,
+ .minimum_version_id = 2,
.needed = icount_shift_state_needed,
.fields = (VMStateField[]) {
VMSTATE_INT16(icount_time_shift, TimersState),
+ VMSTATE_INT64(last_delta, TimersState),
VMSTATE_END_OF_LIST()
}
};
diff --git a/softmmu/icount.c b/softmmu/icount.c
index dbcd8c3594..21341a4ce4 100644
--- a/softmmu/icount.c
+++ b/softmmu/icount.c
@@ -176,9 +176,6 @@ static void icount_adjust(void)
int64_t cur_icount;
int64_t delta;
- /* Protected by TimersState mutex. */
- static int64_t last_delta;
-
/* If the VM is not running, then do nothing. */
if (!runstate_is_running()) {
return;
@@ -193,20 +190,20 @@ static void icount_adjust(void)
delta = cur_icount - cur_time;
/* FIXME: This is a very crude algorithm, somewhat prone to oscillation. */
if (delta > 0
- && last_delta + ICOUNT_WOBBLE < delta * 2
+ && timers_state.last_delta + ICOUNT_WOBBLE < delta * 2
&& timers_state.icount_time_shift > 0) {
/* The guest is getting too far ahead. Slow time down. */
qatomic_set(&timers_state.icount_time_shift,
timers_state.icount_time_shift - 1);
}
if (delta < 0
- && last_delta - ICOUNT_WOBBLE > delta * 2
+ && timers_state.last_delta - ICOUNT_WOBBLE > delta * 2
&& timers_state.icount_time_shift < MAX_ICOUNT_SHIFT) {
/* The guest is getting too far behind. Speed time up. */
qatomic_set(&timers_state.icount_time_shift,
timers_state.icount_time_shift + 1);
}
- last_delta = delta;
+ timers_state.last_delta = delta;
qatomic_set_i64(&timers_state.qemu_icount_bias,
cur_icount - (timers_state.qemu_icount
<< timers_state.icount_time_shift));
diff --git a/softmmu/timers-state.h b/softmmu/timers-state.h
index db4e60f18f..8c262ce139 100644
--- a/softmmu/timers-state.h
+++ b/softmmu/timers-state.h
@@ -43,6 +43,8 @@ typedef struct TimersState {
/* Conversion factor from emulated instructions to virtual clock ticks. */
int16_t icount_time_shift;
+ /* Icount delta used for shift auto adjust. */
+ int64_t last_delta;
/* Compensate for varying guest execution speed. */
int64_t qemu_icount_bias;
--
2.30.1
next prev parent reply other threads:[~2021-04-01 11:29 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-04-01 11:22 [PULL 00/12] Misc patches for QEMU 6.0-rc2 Paolo Bonzini
2021-04-01 11:22 ` [PULL 01/12] meson: Propagate gnutls dependency to migration Paolo Bonzini
2021-04-01 11:22 ` [PULL 02/12] target/i386: Verify memory operand for lcall and ljmp Paolo Bonzini
2021-04-01 11:22 ` [PULL 03/12] qapi: qom: do not use target-specific conditionals Paolo Bonzini
2021-04-01 11:22 ` [PULL 04/12] replay: fix recursive checkpoints Paolo Bonzini
2021-04-01 11:22 ` [PULL 05/12] Revert "qom: use qemu_printf to print help for user-creatable objects" Paolo Bonzini
2021-04-01 11:22 ` Paolo Bonzini [this message]
2021-04-01 11:22 ` [PULL 07/12] replay: notify CPU on event Paolo Bonzini
2021-04-01 11:22 ` [PULL 08/12] target/openrisc: fix icount handling for timer instructions Paolo Bonzini
2021-04-01 11:22 ` [PULL 09/12] configure: Do not use default_feature for EXESUF Paolo Bonzini
2021-04-01 11:22 ` [PULL 10/12] hexagon: do not specify executables as inputs Paolo Bonzini
2021-04-01 11:22 ` [PULL 11/12] hexagon: do not specify Python scripts " Paolo Bonzini
2021-04-01 11:22 ` [PULL 12/12] docs: Add a QEMU Code of Conduct and Conflict Resolution Policy document Paolo Bonzini
2021-04-01 16:08 ` [PULL 00/12] Misc patches for QEMU 6.0-rc2 Peter Maydell
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=20210401112223.55711-7-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=Pavel.Dovgalyuk@ispras.ru \
--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).