Re: Patch submission

From: Adam Haberlach <adam(at)newsnipple(dot)com>
To: Thomas Lockhart <lockhart(at)alumni(dot)caltech(dot)edu>
Cc: pgsql-hackers(at)postgresql(dot)org
Subject: Re: Patch submission
Date: 2000-05-02 15:45:11
Message-ID: 20000502084511.A18509@ricochet.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Tue, May 02, 2000 at 01:40:43PM +0000, Thomas Lockhart wrote:
> > How do I go about submitting a patch to the tree? I've
> > done the work to get libpq running on BeOS, although it looks
> > like psql has some curses problems (I can pass a query on the
> > command line and have it work). It isn't much of a patch--mostly
> > Makefiles and a hand-written getprotobyname and a few places
> > where I have twiddled with set/get sockopt.
>
> We had heard earlier that getting the server-side part of the code
> ported to BeOS was not likely to be possible, due to differences in
> Posix compliance and threading models. Is that still the case, and you
> are only planning the client-side libraries for now? I'd like to
> mention it as a possibility in the "Unsupported Platforms" list in the
> upcoming release...

The server-side requires mmap() support, which BeOS does not
currently have. Other then that, PostgreSQL only uses fork()/exec()
AFAIK, so it should work. There are also networking issues, since
sockets are not file descriptors in BeOS for at least another release.
The network stack is being rewritten and will soon be in Beta. I've
done quick&dirty ports of the client-side before, but this is the first
time I'm doing it in ways that don't break everything else.

I'll have to find some time to sort out the reason that psql
is exiting as soon as it starts and bundle my patches up. I should
be able to get this done within the week (although I said that last
week). You could mention BeOS in the Platforms list as "client support
pending" I guess. Server support will be somewhat pointless until the
new network stack is released and even then will require some work
to get the shared memory/IPC working.

--
Adam Haberlach |"You have to understand that the
adam(at)newsnipple(dot)com | entire 'Net is based on people with
http://www.newsnipple.com/ | too much free time on their hands."

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 2000-05-02 15:52:33 Re: shmem_seq may be a bad idea
Previous Message Tom Lane 2000-05-02 15:39:07 Re: psql \l error