DragonFlyBSD bugtracker: Issueshttps://bugs.dragonflybsd.org/https://bugs.dragonflybsd.org/favicon.ico?16293952082019-11-27T18:12:16ZDragonFlyBSD bugtracker
Redmine DragonFlyBSD - Bug #3217 (New): rescue tools: make install fails if rescue folder doesn't existhttps://bugs.dragonflybsd.org/issues/32172019-11-27T18:12:16ZAnonymous
<p>boojum# ls -l /rescue<br />ls: /rescue: No such file or directory</p>
<p>boojum# make install<br />--------------------------------------------------------------</p>
<blockquote><blockquote><blockquote>
<p>installing rescue tools to /rescue ...</p>
</blockquote></blockquote></blockquote>
--------------------------------------------------------------<br />(cd /usr/src/initrd/oinit; LC_ALL=C make -m /usr/src/initrd/../share/mk BINDIR=/rescue install)<br />install -s -o root -g wheel -m 555 oinit /rescue<br />(cd /usr/src/initrd/rescue; LC_ALL=C make -m /usr/src/initrd/../share/mk BINDIR=/rescue install)<br />install -s -o root -g wheel -m 555 rescue /rescue<br />install -o root -g wheel -m 555 /usr/src/initrd/rescue/dowpa /rescue/dowpa<br />install: /rescue/dowpa: Not a directory
<ul>
<li>Error code 71</li>
</ul>
Stop.<br />make<sup><a href="#fn1">1</a></sup>: stopped in /usr/src/initrd/rescue
<ul>
<li>Error code 1</li>
</ul>
<p>Stop.<br />make: stopped in /usr/src/initrd</p> DragonFlyBSD - Bug #3205 (Feedback): Go compiler net test failinghttps://bugs.dragonflybsd.org/issues/32052019-09-18T13:28:07ZAnonymous
<p>A recent commit appears to have broken the net test for the Go compiler:<br /><a class="external" href="https://build.golang.org/log/58be31cfd1a92ba9582fdf33e01f79e03184e59b">https://build.golang.org/log/58be31cfd1a92ba9582fdf33e01f79e03184e59b</a></p>
<p>This was working on commit be02f354 and started failing when I upgraded to b7d3e1.</p> DragonFlyBSD - Bug #3194 (New): Hammer kernel crash on mirror-stream of PFS after upgrade (assert...https://bugs.dragonflybsd.org/issues/31942019-06-29T19:53:41ZAnonymous
<p>I operate two HAMMER arrays, each with PFSs that are mirrored via ``hammer mirror-stream`` from an array master to an array slave.</p>
<p>I recently upgraded to DragonFly 5.6.</p>
<p>After 18 hours of activity, I encountered a kernel fault that reliably can be provoked by running ``hammer mirror-stream`` on one PFS.</p>
<p>I've disabled PFS mirroring until this can be corrected.</p>
<p>Here is the transcribed error from the kernel fault:</p>
<p>```<br />panic: assertion "cursor->flags & HAMMER_CURSOR_ITERATE_CHECK" failed in hammer_btree_iterate at /usr/src/sys/vfs/hammer/hammer_btree.c:263<br />cpuid = 2<br />Trace beginning at frame 0xfffff801e6875158<br />hammer_btree_iterate() at hammer_btree_iterate+0x839 0xffffffff80900db9<br />hammer_btree_iterate() at hammer_btree_iterate+0x839 0xffffffff80900db9<br />hammer_mirror_delete_to.irsra.2() at hammer_mirror_delete_to.isra.2+0x12 0xffffffff80914d92<br />hammer_ioc_mirror_write() at hammer_ioc_mirror_write+0x3b7 0xffffffff809159a7<br />hammer_ioctl() at hammer_ioctl+0xebe 0xffffffff809143de<br />hammer_vop_ioctl() at hammer_vop_ioctl+0x48 0xffffffff8092d458<br />Debugger("panic")</p>
<p>CPU2 stopping CPUs: 0x0000000b<br /> stopped<br />Stopped at Debugger+0x7c: movb $0,0xfc9be9(%rip)<br />db> <br />```</p>
<p>Unfortunately, nothing was saved to /var/crash for this fault except an empty ``kern.0`` file.</p> DragonFlyBSD - Submit #3041 (New): firmware: Remove embedding of multiple images in one module.https://bugs.dragonflybsd.org/issues/30412017-05-25T12:23:23ZAnonymous
<p>Hello,</p>
<p>this patch removes possibility of embedding more than one firmware image in one kernel module through the parent reference in the firmware_register() function.</p>
<p>This patch is a preparation for firmware subsystem modification for moving firmware images from kernel modules to userland.</p>
<p>The mechanism is not used and it can be functionally fully replaced by putting each firmware image in its own module.</p>
<p>Removing the functionality significantly simplifies handling of firmware images. If firmware images are moved to userland the logical grouping of modules could be expressed by putting the related firmware images into one directory if needed.</p>
<p>jan</p> DragonFlyBSD - Bug #2915 (New): Hammer mirror-copy problemhttps://bugs.dragonflybsd.org/issues/29152016-05-17T00:23:36ZAnonymous
<p>DragonFly v4.5.0.843.gfe3b7-DEVELOPMENT</p>
<p>When I mirror copy a master to a slave and then upgrade the slave, the new master pfs can't be mirror copied. This is reproducible but only between two distinct hammer filesystems. If you do it all on the same filesystem, the problem doesn't appear to occur.</p>
<p>boojum# hammer pfs-master /pfs/master<br />Creating PFS <a class="issue tracker-1 status-5 priority-4 priority-default closed" title="Bug: Buildworld error/panic (Closed)" href="https://bugs.dragonflybsd.org/issues/13">#13</a> succeeded!<br />/pfs/master<br /> sync-beg-tid=0x0000000000000001<br /> sync-end-tid=0x00000001b44c0b20<br /> shared-uuid=1191b9a4-1bc4-11e6-8e1d-418d5cb760e2<br /> unique-uuid=1191b9aa-1bc4-11e6-8e1d-418d5cb760e2<br /> label="" <br /> prune-min=00:00:00<br /> operating as a MASTER<br /> snapshots directory defaults to /var/hammer/<pfs><br />boojum# cp /COPYRIGHT /pfs/master/ <br />boojum# ls <del>l /pfs/master/<br />total 13<br />-r--r--r-</del> 1 root wheel 6686 16-May-2016 17:12 COPYRIGHT<br />boojum# hammer <del>y mirror-copy /pfs/master /volumes/BACKUP3/pfs/slave <br />PFS slave /volumes/BACKUP3/pfs/slave does not exist. Auto create new slave PFS!<br />Creating PFS <a class="issue tracker-1 status-5 priority-4 priority-default closed" title="Bug: panic (Closed)" href="https://bugs.dragonflybsd.org/issues/31">#31</a> succeeded!<br />/volumes/BACKUP3/pfs/slave<br /> sync-beg-tid=0x0000000000000001<br /> sync-end-tid=0x0000000000000001<br /> shared-uuid=1191b9a4-1bc4-11e6-8e1d-418d5cb760e2<br /> unique-uuid=2e551d04-1bc4-11e6-8e1d-418d5cb760e2<br /> label="" <br /> prune-min=00:00:00<br /> operating as a SLAVE<br /> snapshots directory defaults to /var/hammer/<pfs><br />Prescan to break up bulk transfer<br />Prescan 1 chunks, total 0 MBytes (7296)<br />Mirror-read /pfs/master succeeded<br />boojum# ls -l /volumes/BACKUP3/pfs/slave/<br />total 13<br />-r--r--r-</del> 1 root wheel 6686 16-May-2016 17:12 COPYRIGHT<br />boojum# hammer pfs-upgrade /volumes/BACKUP3/pfs/slave <br />pfs-upgrade of PFS#31 () succeeded<br />boojum# hammer -y mirror-copy /volumes/BACKUP3/pfs/slave /volumes/BACKUP3/pfs/slave2<br />PFS slave /volumes/BACKUP3/pfs/slave2 does not exist. Auto create new slave PFS!<br />Creating PFS <a class="issue tracker-1 status-5 priority-4 priority-default closed" title="Bug: Better news about that wi panic (Closed)" href="https://bugs.dragonflybsd.org/issues/32">#32</a> succeeded!<br />/volumes/BACKUP3/pfs/slave2<br /> sync-beg-tid=0x0000000000000001<br /> sync-end-tid=0x0000000000000001<br /> shared-uuid=1191b9a4-1bc4-11e6-8e1d-418d5cb760e2<br /> unique-uuid=5ef43f68-1bc4-11e6-8e1d-418d5cb760e2<br /> label="" <br /> prune-min=00:00:00<br /> operating as a SLAVE<br /> snapshots directory defaults to /var/hammer/<pfs><br />Prescan to break up bulk transfer<br />Prescan 1 chunks, total 0 MBytes (0)<br />Mirror-read /volumes/BACKUP3/pfs/slave succeeded<br />boojum# ls -l /volumes/BACKUP3/pfs/slave2 <br />lrwxr-xr-x 1 root wheel 10 16-May-2016 17:14 /volumes/BACKUP3/pfs/slave2 -> @@0x00000001000420d0:00032<br />boojum# ls -l /volumes/BACKUP3/pfs/slave2/<br />ls: /volumes/BACKUP3/pfs/slave2/: No such file or directory<br />boojum#</p> DragonFlyBSD - Bug #2870 (New): Broken text and icons when glamor acceleration is usedhttps://bugs.dragonflybsd.org/issues/28702015-12-19T15:09:16Z375gnu
<p>I'm using DF 4.4.1 with radeonkms, my videocard is radeon 7750. When I disable 2D acceleration in xorg.conf then output is correct, but when I enable it I see garbage, the more applications is run the more garbage I see, letters in gtk2/3 applications are broken, icons are black.</p>
<p>It depends on number of application started, i.e. when I use startx (twm + 3 xterm) and start 1 gtk application then it has only icons broken, qt applications are clean, but I use startxfce then everything is broken.</p>
<p>It does not matter what 3D (mesa) driver is used, radeonsi or llvmpipe.</p> DragonFlyBSD - Bug #2806 (New): failed to configure a link-local address on ath0 (errno = 22)https://bugs.dragonflybsd.org/issues/28062015-03-16T07:28:08ZChingyuanphil.cyc@gmail.com
<p>Hello,</p>
<p>My Laptop's wireless device is "Qualcomm Atheros AR9485 rev 01".</p>
<p>Athought DFB can detect it, it can not be initialized.</p>
<p>The error message as follows :</p>
<p>#ifconfig ath0 up</p>
<p>ath0: _ath_power_restore_power_state : refcnt=0?<br />kernel : in6_ifattach_linklocal : failed to configure a link-local address on ath0 (errno = 22)</p> DragonFlyBSD - Submit #2438 (Feedback): TRIM fixeshttps://bugs.dragonflybsd.org/issues/24382012-10-22T04:59:20ZAnonymous
<p>This patch is to fix bugs associated with TRIM.</p>
<p>If trim is on as a option, display that when typing "mount".</p>
<p>Change post-trim ffs_blkfree_cg() to use taskqueue_swi_mp and get mp token when modifying freemap.</p>
<p>Make sure TRIM works with softdep. Stash a copy of that vnode's mount point in the ufs inode so that if we are using softdep, we can get access to the mount point through the faked up inode (created in freeblocks). The original mount point path (ip->i_devvp->v_mount->mnt_flag) doesn't have the mount point options.</p>
<p>Tim</p> DragonFlyBSD - Bug #2297 (New): strange NFS (client) error messages / problemshttps://bugs.dragonflybsd.org/issues/22972012-02-02T14:28:09ZAnonymous
<p>Hi,</p>
<p>I have DragonFly mounting my home dir via NFS (server is running Fedora 16) and noticed the following warnings on the console:</p>
<p>dfly.my.lan kernel log messages:</p>
<blockquote>
<p>nfs send error 32 for server f16srv:/home/cosmicdj<br />receive error 54 from nfs server f16srv:/home/cosmicdj<br />nfs send error 32 for server f16srv:/home/cosmicdj<br />receive error 54 from nfs server f16srv:/home/cosmicdj</p>
</blockquote>
<p>+ tons more, always error 32 or 54</p>
<p>Today, I got this:</p>
<p>cosmicdj@dfly:/home/cosmicdj> ls<br />^T<br />load: 0.00 cmd: ls 3244 [getblk] 0.00u 0.00s 0% 2496k<br />^C<br />^T<br />load: 0.00 cmd: ls 3244 [getblk] 0.00u 0.00s 0% 2496k<br />^C<br />^Z<br />^T<br />load: 0.00 cmd: ls 3244 [getblk] 0.00u 0.00s 0% 2496k</p>
<p>Connection to dflyclosed. (<- killed my shell)</p>
<p>I tried to kill 'ls', even via -9. Didn't help. So I ended up killing the shell and tried to umount/remount my NFS homedir via console:</p>
<p>root@dfly # umount /home/cosmicdj<br />^T<br />load: 0.00 cmd: umount 3295 [clock] 0.00u 0.00s 0% 1444k<br />[diagnostic] cache_lock: blocked on 0xffffffe07d31e7f8 ".history_beastie" <br />^C<br />^Z<br />^T<br />load: 0.00 cmd: umount 3295 [clock] 0.00u 0.09s 0% 1444k<br />[diagnostic] cache_lock: blocked on 0xffffffe07d319158 "" <br />[diagnostic] cache_lock: blocked on 0xffffffe07d319158 "" <br />[diagnostic] cache_lock: blocked on 0xffffffe07d319158 ""</p>
<p>I get "[diagnostic] cache_lock: blocked on 0xffffffe07d319158 """ errors when I try to login via ssh again (I guess sshd tries to read my key...)</p>
<p>So it was working fine, but after, say 1 day uptime it doesn't work anymore :(</p> DragonFlyBSD - Bug #2292 (New): re interface with jumbo frames (mtu larger than 1500) hangs after...https://bugs.dragonflybsd.org/issues/22922012-01-27T09:04:09ZAnonymous
<p>How to reproduce:<br />ifconfig re0 mtu 4088<br />cd /usr; make pkgsrc-create<br />wait a bit<br />re0 completely hangs; no ping; nothing when tcpdumping; nothing dmesg either</p>
<p>doing a<br />ifconfig re0 mtu 1500<br />gets it back to life though...</p>
<p>uname:<br />DragonFly v3.1.0.47.ge2743-DEVELOPMENT #0: Tue Jan 24 09:47:50 UTC 2012<br /> <a class="email" href="mailto:root@pkgbox64.dragonflybsd.org">root@pkgbox64.dragonflybsd.org</a>:/usr/obj/usr/src/sys/X86_64_GENERIC</p>
<p>dmesg:<br />re0: <RealTek 8111/8168 PCIe Gigabit Ethernet> port 0xe800-0xe8ff mem 0xfbffc000-0xfbffffff,0xfbffb000-0xfbffbfff irq 36 at device 0.0 on pci5<br />re0: Hardware rev. 0x2c000000; MAC ver. 0x2e; PCI-E 125MHz<br />miibus0: <MII bus> on re0<br />rgephy0: <8211B/RTL8169S/8110S media interface> on miibus0<br />rgephy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-FDX, auto</p> DragonFlyBSD - Bug #2052 (New): Kernel panic: CPU APIC ID out of rangehttps://bugs.dragonflybsd.org/issues/20522011-04-21T19:38:13ZAnonymous
<p>Hello,</p>
<p>booting DragonFly BSD from a CD ends with a kernel panic. A screenshot of the console is attached. The machine is a Dell PowerEdge M610 blade. The same problem occurs on both of our M610s, one with older 45nm Xeons and one with newer 32nm Xeons. I'd like to run some performance benchmarks of the Hammer file system on the machines, so a fully working SMP system with ACPI support is probably needed. If you need more debugging information, please let me know how to get it. A piece of advice on how to circumvent this issue would be very helpful. :-)</p>
<p>Andrej</p> DragonFlyBSD - Bug #1695 (New): NFS-related system breakdownhttps://bugs.dragonflybsd.org/issues/16952010-03-14T06:22:19ZAnonymous
<p>Hi all.</p>
<p>I do a cp(1) from /mnt/nfs (which is a zfs fileystem NFS-exported by an<br />opensolaris installation) to my /home/beket directory (HAMMER fs).</p>
<p>After a few MBs having being copied, cp(1) process stalls in 'getblk' state.<br />Also it is unkillable by ^C. From that point there's downhill.</p>
<p>Many commands will block if I issue them, such as 'mount' or 'df'. I can still<br />view top(1) update its contents but the system is on the edge. dmesg shows<br />sparse messages 'nfs server 10.0.0.1:/export/nfs: not responding' or<br />'[diagnostic]: $address block on cache_something ""'. If I break into the<br />debugger, I get nothing unusual. scgetc < sckbdevent < kbd_xxxx < taskqueue_yyyy<br />etc.</p>
<p>I tried to kill X and the system hang to the point that a cold reset was<br />necessary. I once managed to resume after typing 'c' in db> prompt, but that<br />doesn't always succeed.</p>
<p>This situation is, I think, 100% reproducible. Also, I don't have problems<br />copying stuff from|to a linux NFS client.</p>
<p>Cheers</p> DragonFlyBSD - Submit #1398 (In Progress): hdestroy(3) restricts hash key to point to malloc'ed s...https://bugs.dragonflybsd.org/issues/13982009-06-11T02:08:49ZAnonymous
<p>Salute.</p>
<p>hdestroy(3) frees the memory pointed to by the hash key. In other words it expects the user to always have malloc()'ed rather than used static allocation for the hash key. This doesn't apply to the data associated with the key.</p>
<p>Although POSIX standard doesn't say much on this particular topic:</p>
<ol>
<li>This is unnecessarily restrictive. If the user wants static allocation, we should allow this. If she wants dynamic then let <strong>her</strong> free the memory she malloc()'ed.</li>
<li>It is in conflict with the example code in the POSIX page. The code segfaults if you add an hdestroy() call in the end of it.</li>
<li>Programs that target other implementations may segfault in DragonFly (that'show I discovered it). AFAIK sunOS 5.10 and a recent glibc work fine, whereas {Net, Free, DragonFly}BSD all are affected because they share the same code. (One could argue that all programs written with the *BSD version in mind would<br />result in a memory leak. But still I think these programs (if any) should be fixed.)</li>
</ol>
<p>Any thoughts ?</p>
<p>Cheers,<br />Stathis</p> DragonFlyBSD - Bug #1397 (Feedback): jobs -l output inconsistency when called from scripthttps://bugs.dragonflybsd.org/issues/13972009-06-08T00:49:49ZAnonymous
<p>Salute.</p>
<p>The jobs(1) utility gives different output when called from a script and when<br />from an interactive shell.</p>
<pre>
[beket@voyager ~] cat testjobs.sh
#!/bin/sh
sleep 30 &
jobs -l
[beket@voyager ~] sh testjobs.sh
[1] + 10005 Running
[beket@voyager ~] sleep 30 &
[1] 10006
[beket@voyager ~] jobs -l
[1]+ 10006 Running sleep 30 &
[beket@voyager ~]
</pre>
<p>It is not clear whether the jobs(1) should work at all inside a script. POSIX<br />says that since it doesn't fall into the 'special' built-in category a new<br />environment (subshell?) would be created upon its invocation. Even this is true,<br />the jobs aren't specific to the shell environment, so they should be visible to<br />jobs(1). And in any case, the command should either print nothing or print all<br />the fields.</p>
<p>NetBSD 5.0:<br /><pre>
$ sh testjobs.sh
[1] + 27159 Running sleep 30
</pre></p>
<p>SunOS 5.10:<br /><pre>
tuxillo@solaris$ /usr/xpg4/bin/sh testjobs.sh
[1] + 11754 Running <command unknown>
</pre></p>
<p>FreeBSD: same as us. (kindly reported by vstemen at #dragonflybsd).</p>
<p>Any thoughts ?</p>
<p>Best regards,<br />Stathis</p> DragonFlyBSD - Bug #1390 (In Progress): Use id_t type for {get,set}priority()https://bugs.dragonflybsd.org/issues/13902009-05-27T00:25:04ZAnonymous
<p>Salute!</p>
<p>Both get- and set-priority() functions take a `who' argument that may refer to <br />process ID, group ID or a user ID depending on the situation. The id_t type,<br />which is already available in our src tree, guarantees that it's large enough to<br />hold pid, gid, etc.</p>
<p>The attached patch replaces `int' with `id_t' wherever appropriate. I have done<br />a build{world,kernel} and install{world,kernel} and I don't broke anything. Plus<br />some test cases I have, continue to pass.</p>
<p>If anyone objects to this patch speak now or forever hold your peace!</p>
<p>Cheers,<br />Stathis</p>