| From: | Noah Misch <noah(at)leadboat(dot)com> |
|---|---|
| To: | David Rowley <dgrowleyml(at)gmail(dot)com> |
| Cc: | Robert Haas <robertmhaas(at)gmail(dot)com>, Andres Freund <andres(at)2ndquadrant(dot)com>, PostgreSQL-development <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: space reserved for WAL record does not match what was written: panic on windows |
| Date: | 2013-10-18 20:46:18 |
| Message-ID: | 20131018204618.GA368286@tornado.leadboat.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers pgsql-www |
On Fri, Oct 18, 2013 at 09:05:38PM +1300, David Rowley wrote:
> As for signed vs unsigned, I've not looked at all of the places where
> MAXALIGN is used, but I just assumed it was for memory addresses, if this
> is the case then I'm confused why we'd ever want a negative valued memory
> address?
The result will invariably be cast to a pointer type before use, at which
point it's no longer negative. (That's not to say we should keep using signed
math, but it doesn't cause active problems for memory addresses.)
> This might be an obvious one, but can anyone tell me why the casts are in
> the macro at all? Can a compiler not decide for itself which type it should
> be using?
The casts allow passing values of pointer type, which are not valid as
arguments to the bitwise AND operator.
--
Noah Misch
EnterpriseDB http://www.enterprisedb.com
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Josh Berkus | 2013-10-18 21:16:04 | Re: Turning recovery.conf into GUCs |
| Previous Message | Andres Freund | 2013-10-18 20:35:11 | Re: Turning recovery.conf into GUCs |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Noah Misch | 2013-10-21 01:10:13 | Re: signed vs. unsigned in TYPEALIGN (was Re: space reserved for WAL record does not match what was written: panic on windows) |
| Previous Message | David Rowley | 2013-10-18 08:05:38 | Re: space reserved for WAL record does not match what was written: panic on windows |