From: "Alex Bennée" <alex.bennee@linaro.org>
To: dovgaluk@ispras.ru, rth@twiddle.net, pbonzini@redhat.com
Cc: peter.maydell@linaro.org, qemu-devel@nongnu.org,
mttcg@listserver.greensocs.com, fred.konrad@greensocs.com,
a.rigo@virtualopensystems.com, cota@braap.org,
bobby.prani@gmail.com, nikunj@linux.vnet.ibm.com,
"Alex Bennée" <alex.bennee@linaro.org>,
"Peter Crosthwaite" <crosthwaite.peter@gmail.com>
Subject: [Qemu-devel] [PATCH v2 10/12] cpu-exec: update icount after each TB_EXIT
Date: Wed, 5 Apr 2017 14:25:01 +0100 [thread overview]
Message-ID: <20170405132503.32125-11-alex.bennee@linaro.org> (raw)
In-Reply-To: <20170405132503.32125-1-alex.bennee@linaro.org>
There is no particular reason we shouldn't update the global system
icount time as we exit each TranslationBlock run. This ensures the
main-loop doesn't have to wait until we exit to the outer loop for
executed instructions to be credited to timer_state.
The prepare_icount_for_run function is slightly tweaked to match the
logic we run in cpu_loop_exec_tb.
Based on Paolo's original suggestion.
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
cpu-exec.c | 14 +++++++-------
cpus.c | 18 +++++-------------
2 files changed, 12 insertions(+), 20 deletions(-)
diff --git a/cpu-exec.c b/cpu-exec.c
index 748cb66bca..63a56d0407 100644
--- a/cpu-exec.c
+++ b/cpu-exec.c
@@ -600,13 +600,13 @@ static inline void cpu_loop_exec_tb(CPUState *cpu, TranslationBlock *tb,
/* Instruction counter expired. */
assert(use_icount);
#ifndef CONFIG_USER_ONLY
- if (cpu->icount_extra) {
- /* Refill decrementer and continue execution. */
- cpu->icount_extra += insns_left;
- insns_left = MIN(0xffff, cpu->icount_extra);
- cpu->icount_extra -= insns_left;
- cpu->icount_decr.u16.low = insns_left;
- } else {
+ /* Ensure global icount has gone forward */
+ cpu_update_icount(cpu);
+ /* Refill decrementer and continue execution. */
+ insns_left = MIN(0xffff, cpu->icount_budget);
+ cpu->icount_decr.u16.low = insns_left;
+ cpu->icount_extra = cpu->icount_budget - insns_left;
+ if (!cpu->icount_extra) {
/* Execute any remaining instructions, then let the main loop
* handle the next event.
*/
diff --git a/cpus.c b/cpus.c
index 71c3baba87..ff75af449a 100644
--- a/cpus.c
+++ b/cpus.c
@@ -1204,8 +1204,7 @@ static void handle_icount_deadline(void)
static void prepare_icount_for_run(CPUState *cpu)
{
if (use_icount) {
- int64_t count;
- int decr;
+ int insns_left;
/* These should always be cleared by process_icount_data after
* each vCPU execution. However u16.high can be raised
@@ -1214,17 +1213,10 @@ static void prepare_icount_for_run(CPUState *cpu)
g_assert(cpu->icount_decr.u16.low == 0);
g_assert(cpu->icount_extra == 0);
-
- count = tcg_get_icount_limit();
-
- /* To calculate what we have executed so far we need to know
- * what we originally budgeted to run this cycle */
- cpu->icount_budget = count;
-
- decr = (count > 0xffff) ? 0xffff : count;
- count -= decr;
- cpu->icount_decr.u16.low = decr;
- cpu->icount_extra = count;
+ cpu->icount_budget = tcg_get_icount_limit();
+ insns_left = MIN(0xffff, cpu->icount_budget);
+ cpu->icount_decr.u16.low = insns_left;
+ cpu->icount_extra = cpu->icount_budget - insns_left;
}
}
--
2.11.0
next prev parent reply other threads:[~2017-04-05 13:25 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-04-05 13:24 [Qemu-devel] [PATCH v2 00/12] icount and misc MTTCG fixes for 2.9-rc4 Alex Bennée
2017-04-05 13:24 ` [Qemu-devel] [PATCH v2 01/12] scripts/qemugdb/mtree.py: fix up mtree dump Alex Bennée
2017-04-05 13:24 ` [Qemu-devel] [PATCH v2 02/12] scripts/qemu-gdb/timers.py: new helper to dump timer state Alex Bennée
2017-04-05 13:24 ` [Qemu-devel] [PATCH v2 03/12] scripts/replay-dump.py: replay log dumper Alex Bennée
2017-04-05 13:24 ` [Qemu-devel] [PATCH v2 04/12] target/i386/misc_helper: wrap BQL around another IRQ generator Alex Bennée
2017-04-05 13:24 ` [Qemu-devel] [PATCH v2 05/12] cpus: remove icount handling from qemu_tcg_cpu_thread_fn Alex Bennée
2017-04-05 13:24 ` [Qemu-devel] [PATCH v2 06/12] cpus: check cpu->running in cpu_get_icount_raw() Alex Bennée
2017-04-05 13:24 ` [Qemu-devel] [PATCH v2 07/12] cpus: move icount preparation out of tcg_exec_cpu Alex Bennée
2017-04-05 13:24 ` [Qemu-devel] [PATCH v2 08/12] cpus: don't credit executed instructions before they have run Alex Bennée
2017-04-05 13:25 ` [Qemu-devel] [PATCH v2 09/12] cpus: introduce cpu_update_icount helper Alex Bennée
2017-04-05 14:08 ` Paolo Bonzini
2017-04-05 14:34 ` Alex Bennée
2017-04-05 15:00 ` Paolo Bonzini
2017-04-05 13:25 ` Alex Bennée [this message]
2017-04-05 13:25 ` [Qemu-devel] [PATCH v2 11/12] cpus: call cpu_update_icount on read Alex Bennée
2017-04-05 14:07 ` Paolo Bonzini
2017-04-07 11:35 ` Alex Bennée
2017-04-07 12:19 ` Paolo Bonzini
2017-04-07 13:14 ` Alex Bennée
2017-04-07 18:42 ` Richard Henderson
2017-04-05 13:25 ` [Qemu-devel] [PATCH v2 12/12] replay: assert time only goes forward Alex Bennée
2017-04-05 13:33 ` Pavel Dovgalyuk
2017-04-05 13:49 ` Paolo Bonzini
2017-04-05 14:37 ` Alex Bennée
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=20170405132503.32125-11-alex.bennee@linaro.org \
--to=alex.bennee@linaro.org \
--cc=a.rigo@virtualopensystems.com \
--cc=bobby.prani@gmail.com \
--cc=cota@braap.org \
--cc=crosthwaite.peter@gmail.com \
--cc=dovgaluk@ispras.ru \
--cc=fred.konrad@greensocs.com \
--cc=mttcg@listserver.greensocs.com \
--cc=nikunj@linux.vnet.ibm.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=rth@twiddle.net \
/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).