From: | Jan Kim <kim(at)mpiz-koeln(dot)mpg(dot)de> |
---|---|
To: | pgsql-admin(at)postgresql(dot)org |
Subject: | Compilation problems on Digital OSF1 4.0 |
Date: | 1998-12-08 13:52:11 |
Message-ID: | 19981208145211.A22990@mars.mpiz-koeln.mpg.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
Dear PostgreSQLers,
I try to build the new PostgreSQL v6.4 on our Alpha workstations running
OSF/1 4.0 with egcs-2.91.57 19980901 (egcs-1.1 release). The INSTALL
file says there are "minor patchable problems". However, I'm not sure
how to patch this one:
The sources seem to assume that slock_t can be typecast to int. But
OTOH, src/include/os.h says
#include <sys/mman.h> /* for msemaphore */
typedef msemaphore slock_t;
and our /usr/include/sys/mman.h has:
typedef struct {
int msem_state; /* The semaphore is locked is non-zero. */
int msem_wanted; /* Threads are waiting on the semaphore. */
}msemaphore;
Now, I patched the "offending" line in src/include/os.h to:
typedef int slock_t;
As a result, I have now managed to get everything through the compiler
(after some other patches which were indeed minor). However, my technical
knowledge about semaphores, test-and-set etc. is limited and at the same
time, it seems to me that fumbling in this area can be disastrous in a
database system, I would appreciate if some knowledgeable person could
comment on this.
Greetinx & thanx in advance, Jan
--
+- Jan T. Kim -------------------------------------------------------+
| email: kim(at)mpiz-koeln(dot)mpg(dot)de |
| WWW: http://www.mpiz-koeln.mpg.de/~kim/ |
*-----=< hierarchical systems are for files, not for humans >=-----*
From | Date | Subject | |
---|---|---|---|
Next Message | Murat Balkas | 1998-12-08 14:56:02 | Re: your mail |
Previous Message | Murat Balkas | 1998-12-08 12:45:58 | install problem under solaris |