FreeBSD 5.x and Bad File Descriptor errors. Why?
Hello What follows is a description of a problem I used to have when running FreeBSD 5.0 and 5.1. I am not running FreeBSD right now, but I am considering going back to it but I need to figure out how to prevent this issue from happening again. My system has 2 harddrives, a 16 GB Seagate that I use for backups and mount it under /mnt/backup and a 40 GB Maxtor that I am using for everything else. After the installation of the OS (usually about 8-12 days of running non-stop) I start getting Bad File Descriptor errors on random files all around the Maxtor drive and I have to go to single-user mode in order to run a full fsck on the system. After that, the system works, until in 8-12 days time even more files get corrupted this way and the process has to be repeated. Eventually, so many files are damaged that a full OS reinstall is required. Now if not for a few things, I'd probably come to the conclusion that my Maxtor HD is dying on me as my Seagate drive isn't causing me any headaches. However this doesn't seem to be the case, as if Linux (EXT3) or Windows (NTFS) are used, no data loss ever occurs even if the system is left running for many weeks in a row. Now what exactly could be causing this bizarre behavior ? If this is of any help, the exact model number of the HD is MAXTOR 4K040H2 and I was using UFS2 on both drives. Thanks in advance. Sincerely, Dan Naumov ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 5.x and Bad File Descriptor errors. Why?
On Wed, Dec 03, 2003 at 01:43:57PM +0200, Dan Naumov wrote: Hello What follows is a description of a problem I used to have when running FreeBSD 5.0 and 5.1. I am not running FreeBSD right now, but I am considering going back to it but I need to figure out how to prevent this issue from happening again. My system has 2 harddrives, a 16 GB Seagate that I use for backups and mount it under /mnt/backup and a 40 GB Maxtor that I am using for everything else. After the installation of the OS (usually about 8-12 days of running non-stop) I start getting Bad File Descriptor errors on random files all around the Maxtor drive and I have to go to single-user mode in order to run a full fsck on the system. After that, the system works, until in 8-12 days time even more files get corrupted this way and the process has to be repeated. Eventually, so many files are damaged that a full OS reinstall is required. You get that error from what command(s)? Have you tried with 5.2, which has a new ATA driver? Kris pgp0.pgp Description: PGP signature
Re: FreeBSD 5.x and Bad File Descriptor errors. Why?
Kris Kennaway wrote You get that error from what command(s)? Have you tried with 5.2, which has a new ATA driver? Kris I usually discover that I am having data corruption when trying to update ports. A file that has a Bad File Descriptor refuses to let cvsup overwrite and/or remove itself. I've also had make installworld fail on me for the very same reason. A file in the base system would get corrupted and would prevent anything from overwriting itself. These problems would go away after a full system fsck, but the fsck would remove the damaged files completely, sometimes leaving the system in a severely broken state. And even if it did not, data corruption would happen again in 8-12 days. Sincerely, Dan Naumov ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 5.x and Bad File Descriptor errors. Why?
On Wed, Dec 03, 2003 at 05:55:29PM +0200, Dan Naumov wrote: Kris Kennaway wrote You get that error from what command(s)? Have you tried with 5.2, which has a new ATA driver? Kris I usually discover that I am having data corruption when trying to update ports. A file that has a Bad File Descriptor refuses to let cvsup overwrite and/or remove itself. I've also had make installworld fail on me for the very same reason. A file in the base system would get corrupted and would prevent anything from overwriting itself. These problems would go away after a full system fsck, but the fsck would remove the damaged files completely, sometimes leaving the system in a severely broken state. And even if it did not, data corruption would happen again in 8-12 days. Are there any other console or system messages logged? It sounds like there should be some other error reported by the kernel before the userland command receives the EBADF. Kris pgp0.pgp Description: PGP signature
Re: FreeBSD 5.x and Bad File Descriptor errors. Why?
Dan Naumov wrote: What follows is a description of a problem I used to have when running FreeBSD 5.0 and 5.1. [...] After the installation of the OS (usually about 8-12 days of running non-stop) I start getting Bad File Descriptor errors on random files all around the Maxtor drive and I have to go to single-user mode in order to run a full fsck on the system. After that, the system works, until in 8-12 days time even more files get corrupted this way and the process has to be repeated. Eventually, so many files are damaged that a full OS reinstall is required. I had exactly the same experiences while using the same FreeBSD versions. The filesystems slowly died without any direct sign pointing to hardware problems. But, after having exchanged the harddisk with a newer one, the effect never occured again. k.j. ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 5.x and Bad File Descriptor errors. Why?
On Wed, 3 Dec 2003, Dan Naumov wrote: causing this bizarre behavior ? If this is of any help, the exact model number of the HD is MAXTOR 4K040H2 and I was using UFS2 on both drives. Thanks in advance. Maxtor has a diag tool that goes onto a floppy that you could run. I have this exact disk in my sparc64 here and the first time I used it, it kept running over bad sectors and generally causing mayhem. Running the full scan with the tool, it prompted to fix some problems, then the bad sectors went away. I reinstalled the OS to clear out the broken files and I never had any problems afterward. The program is called PowerMax and is available from the downloads section under either diagnostics or utilities. -- Doug White| FreeBSD: The Power to Serve [EMAIL PROTECTED] | www.FreeBSD.org ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: FreeBSD 5.x and Bad File Descriptor errors. Why?
Doug White wrote: Maxtor has a diag tool that goes onto a floppy that you could run. I have this exact disk in my sparc64 here and the first time I used it, it kept running over bad sectors and generally causing mayhem. Running the full scan with the tool, it prompted to fix some problems, then the bad sectors went away. I reinstalled the OS to clear out the broken files and I never had any problems afterward. The program is called PowerMax and is available from the downloads section under either diagnostics or utilities. I've tried running PowerMax on the drive with a through scan but no errors were found. Sincerely, Dan Naumov ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to [EMAIL PROTECTED]
Port emulators/vmware3 + FreeBSD 5.x = kernel panic
Submitter-Id: current-users Originator:Vyacheslav I. Ivanchenko Confidential: no Synopsis: Port emulators/vmware3 + FreeBSD 5.x = kernel panic Severity: critical Priority: high Category: ports Class: sw-bug Release: FreeBSD 5.1-CURRENT i386 Environment: System: FreeBSD dhs.net.ru 5.1-CURRENT FreeBSD 5.1-CURRENT #0: Wed Oct 22 13:02:13 IRKST 2003 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/K20030324 i386 Description: Port emulators/vmware3 is very unstable on FreeBSD 5.x The first time I have traced this problem in the May of this year after I have upgraded FreeBSD from 5.0-RELEASE to 5.1-BETA version with the purpose to try emulators/vmware3 port. Actually this problem is evident on 5.1-CURRENT (20031022) version. # uname -a FreeBSD dhs.net.ru 5.1-CURRENT FreeBSD 5.1-CURRENT #0: Wed Oct 22 13:02:13 IRKST 2003 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/K20030324 i386 Recently I have reinstalled vmware3 from the ports in hope that the problem is corrected but it was evident that nothing has changed. I have installed the $FreeBSD port: ports/emulators/vmware3/Makefile,v 1.63 2003/10/22 18:22:07 knu Exp $ and began to use it in Host-only mode. On vmware3 I start Win98SE. I use Plain disk - fat32.dsk, the size 3 GB: DRIVETYPE ide CYLINDERS 6242 HEADS 16 SECTORS 63 ACCESS/usr/compat/linux/vmware/win98/fat32.mbr 0 63 ACCESS/usr/compat/linux/vmware/win98/fat32.dat 63 6291936 My configuration file for vmware3: #!/usr/local/lib/vmware/bin/vmware libdir = /usr/local/lib/vmware/lib config.version = 2 virtualHW.version = 1 displayName = Windows 98 # CD-ROM ide1:0.present = TRUE ide1:0.fileName = /dev/cdrom ide1:0.deviceType = atapi-cdrom ide1:0.startConnected = TRUE # Virtual hard disk on primary master ide0:0.present = TRUE ide0:0.fileName = /usr/compat/linux/vmware/win98/fat32.dsk ide0:0.deviceType = plainDisk ide0:0.mode = persistent ide0:0.writethrough = TRUE # Floppy floppy0.present = TRUE floppy0.fileName = /dev/fd0 floppy0.startConnected = TRUE # Networked to host only subnet vmnet1.HostOnlyAddress = 192.168.254.1 vmnet1.HostOnlyNetMask = 255.255.255.0 ethernet0.present = TRUE ethernet0.connectionType = hostOnly # Memory size memsize = 284 # Nvram nvram = win98.nvram # Log file log.fileName = win98.log # Hints guestOS = win98 mouse.fileName = /dev/sysmouse tools.remindInstall = FALSE sound.present = FALSE sound.device = /dev/dsp extension.converttonew = DONE ide0:0.renamedToNewExtension = TRUE gui.powerOnAtStartup = FALSE apmSuspend = FALSE suspendToDisk = TRUE gui.exitAtPowerOff = FALSE hard-disk.enableIBR = FALSE gui.fullScreenResize = TRUE resume.repeatable = FALSE disable_acceleration = FALSE gui.fullScreenAtPowerOn = false redoLogDir = logging = TRUE debug = TRUE Vmware3 itself works normally if FAR manager is not started in Windows. But the basic system sometimes hangs, after I terminate the vmware3 and I continue to work with FreeBSD. At any moment after work with vmware3 the system can independently disconnect or hang by herself. It happens rather occasionally but after work with vmware3 only. In the file /var/log/messages which I looked through after such cases, there are no messages. Also I have found out, that after work with vmware3, the system cannot normally execute a command reboot. I have started the computer then I have started the vmware3, then I have unloaded it and then I run the reboot command. Here is the output: ... [***skiped***] ... syncing disks, buffers remaining... 25 25 2 2 1 1 done Uptime: 7m37s WARNING: Driver mistake: destroy_dev on 202/0 panic: don't do that Uptime: 7m37s ... [***skiped***] ... WARNING: Driver mistake: destroy_dev on 202/0 panic: don't do that Uptime: 7m37s Fatal trap 12: page fault while in kernel mode fault virtual address = 0x83e58959 fault code = supervisor write, page not present instruction pointer = 0x8:0xc0783e60 stack pointer = 0x10:0xc512194c frame pointer = 0x10:0xc5121950 code segment= base 0x0, limit 0xf, type 0x1b = DPL 0, pres 1, def32 1, gran 1 processor eflags= resume, IOPL=0 current process = 873 (reboot) kernel: type 12 trap, code 0 If vmware3 is not started the system perfectly works for a few weeks not causing any censures. P. S. Working with FAR manager in Win98SE under vmware3 on 99% hangs up the Windows. I did not observe such problem with vmware2. Enormous quantity of messages VMX|NOT TESTED (warning) F(104):1916 fill in a win98.log file after FAR manager is started. Here an example of such work with vmware3: Oct 23 11:44:31: VMX|Log for VMware Workstation pid=1488 version=unreleased build=build-2242 option=Release.3.2.1 Oct 23 11:44:31: VMX
Re: New technologies in FreeBSD 5.x vs. 4.x
On Sat, Feb 08, 2003 at 01:23:58PM +0100, taxman wrote: Very detailed information for every commit can be found at: http://www.freebsd.org/cgi/cvsweb.cgi/ Adrian, For your FreeBSD news article, if you want to get in contact with developers who have added new things to FreeBSD, I recommend you look at the latest status reports in 2002: http://www.freebsd.org/news/status/status.html Those reports have nice summaries of new features, with contact e-mails for the developers who added those features. -- Craig Rodrigues http://home.attbi.com/~rodrigc [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
New technologies in FreeBSD 5.x vs. 4.x
Hi, I'm about to write an article on FreeBSD for PC Magazine Romania and I would like to concentrate on the new technologies introduced in FreeBSD 5.x. Where can I find a (preferrably detailed) list of the new technologies introduced with FreeBSD 5.x ? I would also like, if possible, to get in touch with a few of the main developers that took charge of coding them. Thank you very much, Adrian Penisoara Ady (@freebsd.ady.ro, @rofug.ro) ROFUG founder To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: New technologies in FreeBSD 5.x vs. 4.x
[ Bcc'd to -current following initial crosspost. Followups to -hackers or directly to Mr. Penisoara. ] On Thu, Feb 06, 2003 at 03:17:15PM +0200, Adrian Penisoara wrote: Where can I find a (preferrably detailed) list of the new technologies introduced with FreeBSD 5.x ? Greetings Adrian, The most significant new features are listed in the Release Announcement[0]. There is greater detail available in the Release Notes under What's New[1]. I would also like, if possible, to get in touch with a few of the main developers that took charge of coding them. You can contact the person listed as MAINTAINER in the Makefile for the part of the source tree you are concerned with. If there is no MAINTAINER listed, a quick trip through CVSWeb[2] will reveal who has been the most active committer to that section of the tree. I would, of course, recommend courtesy and patience when approaching FreeBSD develoeprs for an interview. Most of them are quite busy. [0] - http://www.freebsd.org/releases/5.0R/announce.html [1] - http://www.freebsd.org/releases/5.0R/relnotes-i386.html [2] - http://www.freebsd.org/cgi/cvsweb.cgi/ Good luck, Brandon D. Valentine -- [EMAIL PROTECTED] http://www.geekpunk.net We've been raised on replicas of fake and winding roads, and day after day up on this beautiful stage we've been playing tambourine for minimum wage, but we are real; I know we are real. -- David Berman To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
On 20-Jan-02 Robert Watson wrote: On Sat, 19 Jan 2002, Alp Atici wrote: Is gcc 3.x going to be the default compiler starting from FBSD 5.x series? Is the development on current branch compiled using gcc 3.0 (or up)? Is 5.x series going to be based on a preemptible kernel? Can't answer the gcc question, but yes, John Baldwin currently has support for preemption in his SMPng development tree. The kernel is already somewhat preemptive. The kernel in 5.0 will certainly be preemptible, as making a kernel SMP safe makes it laregly preemptible (i.e, safe for preemption) as well. Making the kernel fully preemptive (i.e., we can switch tasks on any setrunqueue() if the conditions favor that) is actually a fairly esay thing to do, I'm just not sure how well it works right now. :) I just recently fixed some bugs in the alpha pmap code that should help out with getting our kernel closer to that goal. -- John Baldwin [EMAIL PROTECTED]http://www.FreeBSD.org/~jhb/ Power Users Use the Power to Serve! - http://www.FreeBSD.org/ To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
On 23 Jan, Terry Lambert wrote: [gcc 3.0.x bug] Actually, that was against 3.0 at -O2. If that's been fixed, I guess we can cut over, as soom as the non-x86 code generation for our other supported platforms works again (tried compiling your RedHat for Alpha lately?). gcc 3.0.3 has problems with -O3 -funroll-loops, -O -funroll-loops or -O3 without -funroll-loops seems to work. Someone told me the CVS version of gcc has a fix for this. Someone told me that the only place you can get CVS versions of GCC are by installing RedHat Linux. 8-) 8-) 8-). :-) Why don't we wait until there is a GCC release that actually works? It wasn't my intention to suggest to use a CVS version of GCC. Bye, Alexander. -- It is easier to fix Unix than to live with NT. http://www.Leidinger.net Alexander @ Leidinger.net GPG fingerprint = C518 BC70 E67F 143F BE91 3365 79E2 9C60 B006 3FE7 To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
Terry Lambert wrote: Jason Andresen wrote: Odd, I can't reproduce that under RedHat: I guess you are saying there is a conspiracy to make GCC work only on RedHat? I could believe that... 8-) 8-) 8-). Actually, that was against 3.0 at -O2. If that's been fixed, I guess we can cut over, as soom as the non-x86 code generation for our other supported platforms works again (tried compiling your RedHat for Alpha lately?). Hmm: RedHat rogue/pts/0 (6 ~): gcc3 -O3 -funroll-loops -Wall -pedantic -o gcctest gcctest.c gcctest.c:2: warning: return type defaults to `int' gcctest.c: In function `main': gcctest.c:5: warning: implicit declaration of function `foo' gcctest.c:9: warning: implicit declaration of function `printf' gcctest.c:12: warning: control reaches end of non-void function RedHat rogue/pts/0 (7 ~): ./gcctest hello, stupid compiler! RedHat rogue/pts/0 24 (8 ~): gcc3 -O2 -funroll-loops -Wall -pedantic -o gcctest gcctest.c gcctest.c:2: warning: return type defaults to `int' gcctest.c: In function `main': gcctest.c:5: warning: implicit declaration of function `foo' gcctest.c:9: warning: implicit declaration of function `printf' gcctest.c:12: warning: control reaches end of non-void function RedHat rogue/pts/0 (9 ~): ./gcctest hello, stupid compiler! RedHat rogue/pts/0 24 (10 ~): gcc3 -O2 -Wall -pedantic -o gcctest gcctest.c gcctest.c:2: warning: return type defaults to `int' gcctest.c: In function `main': gcctest.c:5: warning: implicit declaration of function `foo' gcctest.c:9: warning: implicit declaration of function `printf' gcctest.c:12: warning: control reaches end of non-void function RedHat rogue/pts/0 (11 ~): uname -a Linux rogue 2.4.7-10 #1 Thu Sep 6 17:27:27 EDT 2001 i686 unknown RedHat rogue/pts/0 (12 ~): gcc3 -v Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/3.0.2/specs Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --disable-checking --host=i386-redhat-linux Thread model: posix gcc version 3.0.2 20010905 (Red Hat Linux 7.1 3.0.1-3) RedHat rogue/pts/0 (13 ~): Hmm, I can't reproduce the problem with any setting... Maybe it is time to look into upgrading GCC. -- \ |_ _|__ __|_ \ __| Jason Andresen[EMAIL PROTECTED] |\/ | ||/ _| Network and Distributed Systems Engineer _| _|___| _| _|_\___| Office: 703-883-7755 To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
Alexander Leidinger wrote: [gcc 3.0.x bug] Actually, that was against 3.0 at -O2. If that's been fixed, I guess we can cut over, as soom as the non-x86 code generation for our other supported platforms works again (tried compiling your RedHat for Alpha lately?). gcc 3.0.3 has problems with -O3 -funroll-loops, -O -funroll-loops or -O3 without -funroll-loops seems to work. Someone told me the CVS version of gcc has a fix for this. Someone told me that the only place you can get CVS versions of GCC are by installing RedHat Linux. 8-) 8-) 8-). Why don't we wait until there is a GCC release that actually works? -- Terry To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
Terry Lambert wrote:
Alp Atici wrote:
Is gcc 3.x going to be the default compiler starting from
FBSD 5.x series? Is the development on current branch
compiled using gcc 3.0 (or up)?
I think that the cut over will happen after the compiler
no longer core dumps on:
Odd, I can't reproduce that under RedHat:
RedHat gypsy/pts/0 (5 ~): cat gcctest.c
main()
{
int i;
i = foo();
switch( i) {
default:
printf( hello, stupid compiler!\n);
break;
}
}
int
foo()
{
return( 6);
}
RedHat gypsy/pts/0 (6 ~): gcc3 -v
Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/3.0.2/specs
Configured with: ../configure --prefix=/usr --mandir=/usr/share/man
--infodir=/usr/share/info --enable-shared --enable-threads=posix
--disable-checking --host=i386-redhat-linux
Thread model: posix
gcc version 3.0.2 20010905 (Red Hat Linux 7.1 3.0.1-3)
RedHat gypsy/pts/0 (7 ~): gcc3 -Wall -pedantic -o gcctest gcctest.c
gcctest.c:2: warning: return type defaults to `int'
gcctest.c: In function `main':
gcctest.c:5: warning: implicit declaration of function `foo'
gcctest.c:9: warning: implicit declaration of function `printf'
gcctest.c:12: warning: control reaches end of non-void function
RedHat gypsy/pts/0 (8 ~): ./gcctest
hello, stupid compiler!
--
\ |_ _|__ __|_ \ __| Jason Andresen[EMAIL PROTECTED]
|\/ | ||/ _| Network and Distributed Systems Engineer
_| _|___| _| _|_\___| Office: 703-883-7755
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
FreeBSD 5.x
Is gcc 3.x going to be the default compiler starting from FBSD 5.x series? Is the development on current branch compiled using gcc 3.0 (or up)? Is 5.x series going to be based on a preemptible kernel? Thanks, Alp To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
[-hackers removed from CC. One list is enough.] Alp Atici [EMAIL PROTECTED] writes: Is gcc 3.x going to be the default compiler starting from FBSD 5.x series? Is the development on current branch compiled using gcc 3.0 (or up)? Yes. Not yet. Is 5.x series going to be based on a preemptible kernel? That's the plan. Best regards, Mike Barcroft To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
On Sat, 19 Jan 2002, Alp Atici wrote: Is gcc 3.x going to be the default compiler starting from FBSD 5.x series? Is the development on current branch compiled using gcc 3.0 (or up)? Is 5.x series going to be based on a preemptible kernel? Can't answer the gcc question, but yes, John Baldwin currently has support for preemption in his SMPng development tree. Robert N M Watson FreeBSD Core Team, TrustedBSD Project [EMAIL PROTECTED] NAI Labs, Safeport Network Services To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
Alp Atici wrote:
Is gcc 3.x going to be the default compiler starting from
FBSD 5.x series? Is the development on current branch
compiled using gcc 3.0 (or up)?
I think that the cut over will happen after the compiler
no longer core dumps on:
main()
{
int i;
i = foo();
switch( i) {
default:
printf( hello, stupid compiler!\n);
break;
}
}
int
foo()
{
return( 6);
}
Is 5.x series going to be based on a preemptible kernel?
A multithreaded kernel. Do ISRs count?
-- Terry
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
In the last episode (Jan 19), Terry Lambert said:
Alp Atici wrote:
Is gcc 3.x going to be the default compiler starting from FBSD 5.x
series? Is the development on current branch compiled using gcc 3.0
(or up)?
I think that the cut over will happen after the compiler
no longer core dumps on:
main()
{
int i;
i = foo();
switch( i) {
default:
printf( hello, stupid compiler!\n);
break;
}
}
int
foo()
{
return( 6);
}
Doesn't core on me (gcc30+bounds-checking port, FreeBSD-current). In
fact, I've got USE_GCC30 in my make.conf and build all my ports with it
(at least the ports that aren't broken and hardcode cc or gcc).
--
Dan Nelson
[EMAIL PROTECTED]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
Re: FreeBSD 5.x
Dan Nelson [EMAIL PROTECTED] writes:
In the last episode (Jan 19), Terry Lambert said:
Alp Atici wrote:
Is gcc 3.x going to be the default compiler starting from FBSD 5.x
series? Is the development on current branch compiled using gcc 3.0
(or up)?
I think that the cut over will happen after the compiler
no longer core dumps on:
main()
{
int i;
i = foo();
switch( i) {
default:
printf( hello, stupid compiler!\n);
break;
}
}
int
foo()
{
return( 6);
}
Doesn't core on me (gcc30+bounds-checking port, FreeBSD-current). In
fact, I've got USE_GCC30 in my make.conf and build all my ports with it
(at least the ports that aren't broken and hardcode cc or gcc).
Interesting. The sparc64 toolchain suffers from this problem, so a
number of files on the sparc64 p4 branch have custom versions.
Anyway, I'm told this problem has been fixed in 3.1, which is the
planned version of GCC for 5.0-RELEASE.
Best regards,
Mike Barcroft
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-current in the body of the message
