From: Luiz Capitulino <lcapitulino@redhat.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Cc: "kwolf@redhat.com" <kwolf@redhat.com>,
"amit.shah@redhat.com" <amit.shah@redhat.com>,
"aliguori@us.ibm.com" <aliguori@us.ibm.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
"armbru@redhat.com" <armbru@redhat.com>
Subject: Re: [Qemu-devel] [PATCH 4/9] runstate_set(): Check for valid transitions
Date: Tue, 6 Sep 2011 13:47:18 -0300 [thread overview]
Message-ID: <20110906134718.4fb8a130@doriath> (raw)
In-Reply-To: <4E664260.3010303@siemens.com>
On Tue, 06 Sep 2011 17:55:12 +0200
Jan Kiszka <jan.kiszka@siemens.com> wrote:
> On 2011-09-06 15:14, Luiz Capitulino wrote:
> > This commit could have been folded with the previous one, however
> > doing it separately will allow for easy bisect and revert if needed.
> >
> > Checking and testing all valid transitions wasn't trivial, chances
> > are this will need broader testing to become more stable.
> >
> > Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
> > ---
> > vl.c | 149 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-
> > 1 files changed, 148 insertions(+), 1 deletions(-)
> >
> > diff --git a/vl.c b/vl.c
> > index 9926d2a..fe3628a 100644
> > --- a/vl.c
> > +++ b/vl.c
> > @@ -332,9 +332,156 @@ bool runstate_check(RunState state)
> > return current_run_state == state;
> > }
> >
> > +/* This function will abort() on invalid state transitions */
> > void runstate_set(RunState new_state)
> > {
> > - assert(new_state < RSTATE_MAX);
> > + switch (current_run_state) {
> > + case RSTATE_NO_STATE:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + case RSTATE_IN_MIGRATE:
> > + case RSTATE_PRE_LAUNCH:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_DEBUG:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_IN_MIGRATE:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + case RSTATE_PRE_LAUNCH:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_PANICKED:
> > + switch (new_state) {
> > + case RSTATE_PAUSED:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_IO_ERROR:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_PAUSED:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_POST_MIGRATE:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_PRE_LAUNCH:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + case RSTATE_POST_MIGRATE:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_PRE_MIGRATE:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + case RSTATE_POST_MIGRATE:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_RESTORE:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_RUNNING:
> > + switch (new_state) {
> > + case RSTATE_DEBUG:
> > + case RSTATE_PANICKED:
> > + case RSTATE_IO_ERROR:
> > + case RSTATE_PAUSED:
> > + case RSTATE_PRE_MIGRATE:
> > + case RSTATE_RESTORE:
> > + case RSTATE_SAVEVM:
> > + case RSTATE_SHUTDOWN:
> > + case RSTATE_WATCHDOG:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_SAVEVM:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_SHUTDOWN:
> > + switch (new_state) {
> > + case RSTATE_PAUSED:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + case RSTATE_WATCHDOG:
> > + switch (new_state) {
> > + case RSTATE_RUNNING:
> > + goto transition_ok;
> > + default:
> > + /* invalid transition */
> > + abort();
> > + }
> > + abort();
> > + default:
> > + fprintf(stderr, "current run state is invalid: %s\n",
> > + runstate_as_string());
> > + abort();
> > + }
> > +
> > +transition_ok:
> > current_run_state = new_state;
> > }
> >
>
> I haven't looked at the transitions yet, but just to make the function
> smaller: you could fold identical blocks together, e.g.
I thought about doing that but I fear it's error-prone: you extend
RSTATE_PAUSED and forget about RSTATE_IO_ERROR.
I think it's better to have different things separated, that's, each state
has its own switch statement.
>
> case RSTATE_IO_ERROR:
> case RSTATE_PAUSED:
> switch (new_state) {
> case RSTATE_RUNNING:
> goto transition_ok;
> default:
> /* invalid transition */
> abort();
> }
> abort();
>
> Jan
>
next prev parent reply other threads:[~2011-09-06 16:47 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-09-06 13:14 [Qemu-devel] [PATCH v4 0/9]: Introduce the RunState type Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 1/9] Move vm_state_notify() prototype from cpus.h to sysemu.h Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 2/9] Replace the VMSTOP macros with a proper state type Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 3/9] RunState: Add additional states Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 4/9] runstate_set(): Check for valid transitions Luiz Capitulino
2011-09-06 15:55 ` Jan Kiszka
2011-09-06 16:47 ` Luiz Capitulino [this message]
2011-09-06 17:42 ` Lluís Vilanova
2011-09-09 12:58 ` Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 5/9] Drop the incoming_expected global variable Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 6/9] Drop the vm_running " Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 7/9] Monitor/QMP: Don't allow cont on bad VM state Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 8/9] QMP: query-status: Introduce 'status' key Luiz Capitulino
2011-09-06 13:14 ` [Qemu-devel] [PATCH 9/9] HMP: info status: Print the VM state Luiz Capitulino
-- strict thread matches above, loose matches on Subject: below --
2011-09-09 20:25 [Qemu-devel] [PATCH v5 0/9]: Introduce the RunState type Luiz Capitulino
2011-09-09 20:25 ` [Qemu-devel] [PATCH 4/9] runstate_set(): Check for valid transitions Luiz Capitulino
2011-09-14 3:06 ` Luiz Capitulino
2011-09-14 19:55 ` Blue Swirl
2011-09-14 20:23 ` Luiz Capitulino
2011-09-14 20:49 [Qemu-devel] [PULL 0/9]: QMP queue Luiz Capitulino
2011-09-14 20:49 ` [Qemu-devel] [PATCH 4/9] runstate_set(): Check for valid transitions Luiz Capitulino
2011-09-15 20:05 [Qemu-devel] [PULL 0/9]: QMP queue Luiz Capitulino
2011-09-15 20:05 ` [Qemu-devel] [PATCH 4/9] runstate_set(): Check for valid transitions Luiz Capitulino
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=20110906134718.4fb8a130@doriath \
--to=lcapitulino@redhat.com \
--cc=aliguori@us.ibm.com \
--cc=amit.shah@redhat.com \
--cc=armbru@redhat.com \
--cc=jan.kiszka@siemens.com \
--cc=kwolf@redhat.com \
--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).