Re: WaitLatch - Mailing list pgsql-hackers

From Robert Haas
Subject Re: WaitLatch
Date
Msg-id AANLkTim1R0Yso+fgZrf4KXfEY70cw50fcKaTwE40sChP@mail.gmail.com
Whole thread Raw
In response to WaitLatch  (aaliya zarrin <aaliya.zarrin@gmail.com>)
List pgsql-hackers
On Mon, Dec 13, 2010 at 5:48 AM, aaliya zarrin <aaliya.zarrin@gmail.com> wrote:
> drainSelfPipe() function will read the data and remove it from the
> descriptor?

Yes...

> Coz the signal (SIGUSR1) is writing sata to selfPipe descriptor but unable
> to wake up latch on select system call.

The signal won't cause the select() to return, but the byte that the
signal handler writes to the self-pipe ought to make it return.

> Data is not avaialable at select in WaitLatch fun. Coz it is read by
> drailSelfPipe().

I'm not sure what you're talking about here.  drainSelfPipe() is
called before calling SELECT.  As soon as a self-pipe byte is written,
the SELECT will return.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company


pgsql-hackers by date:

Previous
From: Dimitri Fontaine
Date:
Subject: Re: CommitFest wrap-up
Next
From: Robert Haas
Date:
Subject: Re: pg_is_in_recovery=1