diff options
| author | Marc G. Fournier <scrappy@hub.org> | 1998-07-09 03:29:11 +0000 |
|---|---|---|
| committer | Marc G. Fournier <scrappy@hub.org> | 1998-07-09 03:29:11 +0000 |
| commit | a0659e3e2c79be49feb4aa527d823c71d7bcaf07 (patch) | |
| tree | 7dc58e872cdfce999dd0848000fcd3f2e4edbafa /src/backend/commands/async.c | |
| parent | 8bf61820f033d7df0e89433a0272d7b2567378b8 (diff) | |
| download | postgresql-a0659e3e2c79be49feb4aa527d823c71d7bcaf07.tar.gz postgresql-a0659e3e2c79be49feb4aa527d823c71d7bcaf07.zip | |
From: Tom Lane <tgl@sss.pgh.pa.us>
Making PQrequestCancel safe to call in a signal handler turned out to be
much easier than I feared. So here are the diffs.
Some notes:
* I modified the postmaster's packet "iodone" callback interface to allow
the callback routine to return a continue-or-drop-connection return
code; this was necessary to allow the connection to be closed after
receiving a Cancel, rather than proceeding to launch a new backend...
Being a neatnik, I also made the iodone proc have a typechecked
parameter list.
* I deleted all code I could find that had to do with OOB.
* I made some edits to ensure that all signals mentioned in the code
are referred to symbolically not by numbers ("SIGUSR2" not "2").
I think Bruce may have already done at least some of the same edits;
I hope that merging these patches is not too painful.
Diffstat (limited to 'src/backend/commands/async.c')
| -rw-r--r-- | src/backend/commands/async.c | 12 |
1 files changed, 6 insertions, 6 deletions
diff --git a/src/backend/commands/async.c b/src/backend/commands/async.c index b4b354cfc52..b80cbb8f345 100644 --- a/src/backend/commands/async.c +++ b/src/backend/commands/async.c @@ -7,7 +7,7 @@ * * * IDENTIFICATION - * $Header: /cvsroot/pgsql/src/backend/commands/async.c,v 1.34 1998/06/27 04:53:29 momjian Exp $ + * $Header: /cvsroot/pgsql/src/backend/commands/async.c,v 1.35 1998/07/09 03:28:44 scrappy Exp $ * *------------------------------------------------------------------------- */ @@ -21,11 +21,11 @@ * 2.a If the process is the same as the backend process that issued * notification (we are notifying something that we are listening), * signal the corresponding frontend over the comm channel. - * 2.b For all other listening processes, we send kill(2) to wake up + * 2.b For all other listening processes, we send kill(SIGUSR2) to wake up * the listening backend. - * 3. Upon receiving a kill(2) signal from another backend process notifying - * that one of the relation that we are listening is being notified, - * we can be in either of two following states: + * 3. Upon receiving a kill(SIGUSR2) signal from another backend process + * notifying that one of the relation that we are listening is being + * notified, we can be in either of two following states: * 3.a We are sleeping, wake up and signal our frontend. * 3.b We are in middle of another transaction, wait until the end of * of the current transaction and signal our frontend. @@ -46,7 +46,7 @@ * (which takes place after commit) to all listeners on this relation. * * 3. Async. notification results in all backends listening on relation - * to be woken up, by a process signal kill(2), with name of relation + * to be woken up, by a process signal kill(SIGUSR2), with name of relation * passed in shared memory. * * 4. Each backend notifies its respective frontend over the comm |