> First things first, this isn't an oracle question but rather a series =
of questions based on a previous message on the oracle-l list. If you =
think I should post it on a different list please let me know which one.
>=20
> The message I am talking about can be found here.
> http://www.freelists.org/archives/oracle-l/02-2004/msg03007.html
>=20
> Also, I am new to the Linux world (just thought I'd let you know what =
kind of a guy your dealing with...newbie).
>=20
> Here is the following information about our Linux server.
>=20
> We are running SUSE LINUX Enterprise Server 9. From the files located =
in /usr/src, the version is 2.6.5-7.97. I'm not sure how else to get =
the version.
> There are 8 IA64 processors.
>=20
> I have been researching ways to get direct I/O to work on our Linux =
machine when I stumbled across the above message. Within the message =
was a small code example that tested direct I/O. I tried this code out =
to see if it would work. I could not get the code to compile as is. =
With some modifying (changed the headers), I was able to get it to =
compile but the direct I/O read kept giving me EINVAL (Invalid =
Argument).
>=20
> Here is the modified code that I used.
>=20
> #include <stdio.h>
> #include <stdlib.h>
> #include <sys/types.h>
> #include <sys/stat.h>
> #include <errno.h>
> #include <string.h>
> #include <fcntl.h>
> #include <unistd.h>
> #define BUFFSIZE 65536
> #define ALIGN 4096
>=20
> int main()=20
> {=20
> char *buff;
> int stat1=3D0,stat2=3D0,stat3=3D0;
> int fd1=3D0,fd2=3D0;
> if (stat3=3Dposix_memalign((void **)&buff,ALIGN,BUFFSIZE)) {
> fprintf(stderr,"ALIGN ERR:%s\n",strerror(stat3));
> exit(0);
> }
>=20
> fd1=3Dopen("/home/rrd/workspace/IORead.txt", =
O_RDONLY|O_DIRECT,S_IRWXU);
>=20
> =
fd2=3Dopen("/home/rrd/workspace/IOWrite.txt",O_CREAT|O_WRONLY|O_DIRECT,S_=
IRWXU);
> while(stat1=3Dread(fd1,buff,BUFFSIZE)) {
> if (errno) {
> fprintf(stderr,"READ ERR:%s\n",strerror(errno));
> exit(0);
> }
> stat2=3Dwrite(fd2,buff,(unsigned) stat1);
> if (errno) {
> fprintf(stderr,"WRITE ERR:%s\n",strerror(errno));
> exit(0);
> }
> }
> close(fd1);
> close(fd2);
> =20
> return 0;
> }
>=20
> Also, is there a programmatic way of retrieving the alignment and =
block size? I have seen the fcntl call for F_DIOINFO. However, I have =
read articles that Linux doesn't have this. Instead one must use ioctl =
with XFS_IOC_DIOINFO. When I do a search in /usr/, the only *DIOINFO =
that I find is in xfs_fs.h. When I try calling this function, I get an =
error and from what I gather it is because the file doesn't support this =
type of ioctl. Does this mean I am missing some feature?
>=20
> I have been searching Google about direct I/O and have come across =
countless articles that seem to contradict one another. Some articles =
say that Linux doesn't support O_DIRECT and that you need to convert =
your file system to XFS and use those libraries. Others say it does =
support it. From the example above it seems it doesn't support it. =
What must be done to the file system (if anything) to get direct I/O to =
work? Do I need to convert to XFS?
>=20
> Could it be that I'm using the wrong libraries? How can I find out =
what libraries it is using?
>=20
> Is there anything I should look at on the server to make sure I have =
the most up-to-date stuff? Any other ideas on how I should proceed?
>=20
> Any help would greatly be appreciated.
>=20
> James Kramer
>=20
>=20
> THIS COMMUNICATION MAY CONTAIN CONFIDENTIAL AND/OR OTHERWISE =
PROPRIETARY MATERIAL and is thus for use only by the intended recipient. =
If you received this in error, please contact the sender and delete the =
e-mail and its attachments from all computers.=20
>=20
>=20
--
http://www.freelists.org/webpage/oracle-l
Received on Wed Dec 08 2004 - 15:02:13 CST
