Bug#966716: initramfs-tools: update-initramfs fails with no conf.d/resume and veracrypt volume mounted
Package: initramfs-tools Version: 0.133+deb10u1 Severity: important Dear Maintainer, update-initramfs failed during a system upgrade with the error: >cryptsetup: WARNING: target "hdd500_crypt" not found in /etc/crypttab where "hdd500_crypt" is my hdd, it contains /home and a small swap partition(less than my ram). The system is installed in a ssd "ssd_crypt". In my experience this would make the system unbootable, it would loop on the message "vg1 not found", "vg1" is a lvm volume group whose physical volumes are just "hdd500_crypt". Fortunately I didn't reboot to find out. fstab and crypttab attached. * What led up to the situation? I deleted /etc/initramfs-tools/conf.d/resume and had a veracrypt(truecrypt fork) volume mounted while calling update-initramfs -u -k all. I did so because: I don't use hibernation (my swap 4gb is smaller than my ram 16gb) and; My ssd and hdd are encrypted with the same password, if I delete the resume file I'm asked only once for the password at boot. * What exactly did you do (or not do) that was effective (or ineffective)? I unmounted the truecrypt volume and/or re-created /etc/iniramfs- tools/conf.d/resume (even with an invalid swap path it works). * What was the outcome of this action? I wouldn't expect update-initramfs -u -k all to fail just because there is no resume partition set AND there is a veracrypt volume mounted. A user could easily oversee this and accidentally make the system unbootable during an upgrade just by having a veracrypt volume mounted. * What outcome did you expect instead? To always get an error(not an warning) when there is no resume device set or to always get an error(not an warning) when there is a veracrypt volume mouted. -- Package-specific info: -- initramfs sizes -rw-r--r-- 1 root root 39M Aug 2 09:06 /boot/initrd.img-4.19.0-10-amd64 -rw-r--r-- 1 root root 39M Aug 2 09:07 /boot/initrd.img-4.19.0-8-amd64 -rw-r--r-- 1 root root 39M Aug 2 09:07 /boot/initrd.img-4.19.0-9-amd64 -- /proc/cmdline BOOT_IMAGE=/vmlinuz-4.19.0-9-amd64 root=/dev/mapper/vg0-root ro quiet -- resume RESUME=/dev/mapper/vg1-swap-- /proc/filesystems ext3 ext2 ext4 vfat xfs fuseblk jfs msdos minix hfs hfsplus qnx4 ufs btrfs -- lsmod Module Size Used by btrfs1400832 0 zstd_compress 172032 1 btrfs zstd_decompress81920 1 btrfs xxhash 16384 2 zstd_compress,zstd_decompress xor24576 1 btrfs raid6_pq 122880 1 btrfs ufs86016 0 qnx4 16384 0 hfsplus 118784 0 hfs69632 0 minix 40960 0 msdos 20480 0 jfs 208896 0 uas28672 0 usb_storage73728 3 uas ctr16384 6 ccm20480 9 nft_chain_route_ipv416384 1 xt_CHECKSUM16384 1 nft_chain_nat_ipv4 16384 4 ipt_MASQUERADE 16384 3 nf_nat_ipv416384 2 ipt_MASQUERADE,nft_chain_nat_ipv4 tun49152 1 bridge188416 0 stp16384 1 bridge llc16384 2 bridge,stp devlink77824 0 fuse 122880 7 vboxnetadp 28672 0 vboxnetflt 32768 0 vboxdrv 491520 2 vboxnetadp,vboxnetflt binfmt_misc20480 1 xfs 1458176 1 arc4 16384 2 snd_hda_codec_hdmi 57344 1 rtl8192cu 81920 0 rtl_usb24576 1 rtl8192cu rtl8192c_common61440 1 rtl8192cu nls_ascii 16384 1 nls_cp437 20480 1 rtlwifi98304 3 rtl8192c_common,rtl_usb,rtl8192cu vfat 20480 1 fat86016 2 msdos,vfat mac80211 835584 3 rtl_usb,rtl8192cu,rtlwifi nf_log_ipv616384 0 cfg80211 774144 2 rtlwifi,mac80211 ip6t_REJECT16384 0 nf_reject_ipv6 16384 1 ip6t_REJECT xt_hl 16384 0 ip6_tables 32768 0 ip6t_rt16384 0 eeepc_wmi 16384 0 asus_wmi 32768 1 eeepc_wmi sparse_keymap 16384 1 asus_wmi rfkill 28672 4 asus_wmi,cfg80211 video 45056 1 asus_wmi wmi_bmof 16384 0 nvidia_drm 49152 5 drm_kms_helper208896 1 nvidia_drm serio_raw 16384 0 edac_mce_amd 28672 0 drm 495616 8 drm_kms_helper,nvidia_drm nf_log_ipv416384 0 nf_log_common 16384 2 nf_log_ipv4,nf_log_ipv6 ipt_REJECT 16384 2 nf_reject_ipv4 16384 1 ipt_REJECT snd_hda_codec_realtek 122880 1 kvm_amd 106496 0 snd_hda_codec_generic86016 1 snd_hda_codec_realtek xt_LOG 1638
Bug#594288: kfreebsd-8: non-standard gcc/g++ used for build (gcc-4.3)
Btw FreeBSD handbook (http://www.freebsd.org/doc/en/books/developers-handbook/kerneldebug.html#KERNELDEBUG-OBTAIN) says dumpon and savecore are needed to save and restore crash dumps so I ported them and added them to freebsd-utils in case you need them. On 9/13/10, Tuco wrote: > On 9/13/10, Petr Salinger wrote: >> But when compiled with "-O1" it boots (and runs) fine on kfreebsd-amd64. > > From > http://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html#Optimize-Options > > -O2 turns on all optimization flags specified by -O. It also turns on > the following optimization flags: > > -fthread-jumps > -falign-functions -falign-jumps > -falign-loops -falign-labels > -fcaller-saves > -fcrossjumping > -fcse-follow-jumps -fcse-skip-blocks > -fdelete-null-pointer-checks > -fexpensive-optimizations > -fgcse -fgcse-lm > -finline-small-functions > -findirect-inlining > -fipa-sra > -foptimize-sibling-calls > -fpartial-inlining > -fpeephole2 > -fregmove > -freorder-blocks -freorder-functions > -frerun-cse-after-loop > -fsched-interblock -fsched-spec > -fschedule-insns -fschedule-insns2 > -fstrict-aliasing -fstrict-overflow > -ftree-switch-conversion > -ftree-pre > -ftree-vrp > > so we could find the faulty one by recombining them? It is just 5 > builds, I can do this if you like. > -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#594288: kfreebsd-8: non-standard gcc/g++ used for build (gcc-4.3)
On 9/13/10, Petr Salinger wrote: > But when compiled with "-O1" it boots (and runs) fine on kfreebsd-amd64. >From http://gcc.gnu.org/onlinedocs/gcc/Optimize-Options.html#Optimize-Options -O2 turns on all optimization flags specified by -O. It also turns on the following optimization flags: -fthread-jumps -falign-functions -falign-jumps -falign-loops -falign-labels -fcaller-saves -fcrossjumping -fcse-follow-jumps -fcse-skip-blocks -fdelete-null-pointer-checks -fexpensive-optimizations -fgcse -fgcse-lm -finline-small-functions -findirect-inlining -fipa-sra -foptimize-sibling-calls -fpartial-inlining -fpeephole2 -fregmove -freorder-blocks -freorder-functions -frerun-cse-after-loop -fsched-interblock -fsched-spec -fschedule-insns -fschedule-insns2 -fstrict-aliasing -fstrict-overflow -ftree-switch-conversion -ftree-pre -ftree-vrp so we could find the faulty one by recombining them? It is just 5 builds, I can do this if you like. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#567992: status of openjdk on bsd in debian?
On 9/9/10, Torsten Werner wrote: > On Wed, Sep 8, 2010 at 12:49 PM, Matthias Klose wrote: >> Any news about this? It looks like there exists a port for openjdk-6 too, >> see >> >> http://www.freebsd.org/cgi/cvsweb.cgi/ports/java/openjdk6/ >> http://www.freebsd.org/cgi/cvsweb.cgi/ports/java/openjdk7/ > > Interesting. Does someone know how to get the real code from those links? The tarball URL is in Makefile, the patches are in files. If you wonder about the inexistant bsd directory in the patches, it's just a rename of the linux one. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#567992: status of openjdk on bsd in debian?
On 9/8/10, Matthias Klose wrote:
> Any news about this? It looks like there exists a port for openjdk-6 too,
I started porting openjdk-6 but it was too much work and gave up. I'm
attaching my incomplete patch in case someone wants to continue.
> see
> http://www.freebsd.org/cgi/cvsweb.cgi/ports/java/openjdk6/
> http://www.freebsd.org/cgi/cvsweb.cgi/ports/java/openjdk7/
The FreeBSD port might have some useful bits, but is not very useful
in general. Most of the changes have to do with system detection
(uname), which is different on Debian GNU/kFreeBSD, and with userland
utilities which we don't need.
--- openjdk/jdk/make/common/shared/Platform.gmk~ 2010-06-29 22:07:19.0 -0400
+++ openjdk/jdk/make/common/shared/Platform.gmk 2010-06-30 07:34:10.0 -0400
@@ -190,7 +190,7 @@
endif
# Platform settings specific to Linux
-ifeq ($(SYSTEM_UNAME), Linux)
+ifeq ($(SYSTEM_UNAME), GNU/kFreeBSD)
PLATFORM = linux
OS_NAME = linux
OS_VERSION := $(shell uname -r)
@@ -320,10 +320,12 @@
REQUIRED_OS_VERSION = 2.4.19-SMP
endif
endif
+ ifeq ($(SYSTEM_UNAME), Linux)
ifneq ($(ARCH), ia64)
# ALSA 0.9.1 and above
REQUIRED_ALSA_VERSION = ^((0[.]9[.][1-9])|(1[.]0[.][0-9]))[0-9]*
endif
+ endif
# How much RAM does this machine have:
MB_OF_MEMORY := $(shell m=`free -m | fgrep Mem: | sed -e 's...@\ \ *@ @g' | cut -d' ' -f2`; if [ "$$m" -lt 384 ]; then echo 384; else echo $$m; fi)
endif
--- openjdk/jdk/make/common/shared/Sanity.gmk~ 2010-06-29 21:47:17.0 -0400
+++ openjdk/jdk/make/common/shared/Sanity.gmk 2010-06-30 07:37:40.0 -0400
@@ -74,6 +74,7 @@
$(ECHO) "Unknown linux"; \
fi )
endif
+ ifeq ($(SYSTEM_UNAME), Linux)
ifneq ($(ARCH), ia64)
ifdef CROSS_COMPILATION
# assume alsa version is correct and do not check it
@@ -84,6 +85,7 @@
ALSA_VERSION = `if [ -f "$(ALSA_VERSION_CHECK)" ] ; then $(ALSA_VERSION_CHECK) ; fi`
endif
endif
+ endif
endif
ifeq ($(PLATFORM), windows)
--- openjdk/jdk/make/common/shared/Sanity-Settings.gmk~ 2010-02-16 22:14:12.0 -0500
+++ openjdk/jdk/make/common/shared/Sanity-Settings.gmk 2010-06-30 07:39:08.0 -0400
@@ -183,9 +183,11 @@
endif
ifeq ($(PLATFORM),linux)
ALL_SETTINGS+=$(call addRequiredSetting,LINUX_VERSION)
+ ifeq ($(SYSTEM_UNAME), Linux)
ifneq ($(ARCH), ia64)
ALL_SETTINGS+=$(call addRequiredSetting,ALSA_VERSION)
endif
+ endif
endif
ALL_SETTINGS+=$(call addRequiredVersionSetting,OS_VERSION)
ALL_SETTINGS+=$(call addRequiredSetting,OS_NAME)
--- openjdk/corba/make/common/shared/Platform.gmk~ 2010-08-04 07:33:11.0 -0400
+++ openjdk/corba/make/common/shared/Platform.gmk 2010-08-04 08:52:08.590184486 -0400
@@ -175,7 +175,7 @@
endif
# Platform settings specific to Linux
-ifeq ($(SYSTEM_UNAME), Linux)
+ifeq ($(SYSTEM_UNAME), GNU/kFreeBSD)
PLATFORM = linux
OS_NAME = linux
OS_VERSION := $(shell uname -r)
--- openjdk/hotspot/make/defs.make~ 2010-08-11 22:08:56.0 -0400
+++ openjdk/hotspot/make/defs.make 2010-08-12 10:38:20.0 -0400
@@ -123,14 +123,14 @@
endif
# If not SunOS and not Linux, assume Windows
-ifneq ($(OS), Linux)
- ifneq ($(OS), SunOS)
-OSNAME=windows
- else
-OSNAME=solaris
- endif
-else
+ifeq ($(OS), Linux)
+ OSNAME=linux
+else ifeq ($(OS), SunOS)
+ OSNAME=solaris
+else ifeq ($(OS), GNU/kFreeBSD)
OSNAME=linux
+else
+ OSNAME=windows
endif
# Determinations of default make arguments and platform specific settings
--- openjdk/hotspot/make/linux/Makefile~ 2010-08-11 22:09:03.0 -0400
+++ openjdk/hotspot/make/linux/Makefile 2010-08-12 10:50:49.0 -0400
@@ -233,6 +233,9 @@
SUPPORTED_OS_VERSION = 2.4% 2.5% 2.6% 2.7%
OS_VERSION := $(shell uname -r)
EMPTY_IF_NOT_SUPPORTED = $(filter $(SUPPORTED_OS_VERSION),$(OS_VERSION))
+ifeq ($(shell uname -s), GNU/kFreeBSD)
+EMPTY_IF_NOT_SUPPORTED = something
+endif
check_os_version:
ifeq ($(DISABLE_HOTSPOT_OS_VERSION_CHECK)$(EMPTY_IF_NOT_SUPPORTED),)
--- openjdk/hotspot/src/os/linux/vm/attachListener_linux.cpp.old 2010-08-12 11:35:39.0 -0400
+++ openjdk/hotspot/src/os/linux/vm/attachListener_linux.cpp 2010-08-12 11:51:12.0 -0400
@@ -31,6 +31,7 @@
#include
#include
#include
+#include
// The attach mechanism on Linux uses a UNIX domain socket. An attach listener
// thread is created at startup or is created on-demand via a signal from
@@ -330,6 +331,9 @@
// get the credentials of the peer and check the effective uid/guid
// - check with jeff on this.
+uid_t peer_uid;
+gid_t peer_gid;
+#if defined(SO_PEERCRED)
struct ucred cred_info;
socklen_t optlen = sizeof(cred_info);
if (::getsockopt(s, SOL_SOCKET, SO_PEERCRED, (void*)&cred_info, &optlen) == -1) {
@@ -337,10 +341,25 @@
RESTARTABLE(::close(s), res);
continue;
}
+peer_uid = cred_info.uid;
+peer_gid = cred_info.gid;
+#elif defined(LOCAL_PEERCRED)
+struct xucred cred_info;
+soc
Bug#593658: default compression=on?
Package: zfsutils Version: 8.1-3 Tags: patch Compression in ZFS has a triple benefit in IO performance, disk space and energy consumption, this article explains (with benchmarks): http://blogs.sun.com/partnertech/entry/zfs_compression_perf_disk_space IO improvement isn't obvious: reading and writing takes less time when compressing or decompressing because CPU is not the bottleneck. The linked article explains in more detail. I suggest enabling compression by default when creating new ZFS filesystems. --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio.h 2010-05-24 06:09:36.0 -0400 +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio.h 2010-08-19 18:52:33.995673472 -0400 @@ -98,7 +98,7 @@ }; #define ZIO_COMPRESS_ON_VALUE ZIO_COMPRESS_LZJB -#define ZIO_COMPRESS_DEFAULT ZIO_COMPRESS_OFF +#define ZIO_COMPRESS_DEFAULT ZIO_COMPRESS_ON #define ZIO_FAILURE_MODE_WAIT 0 #define ZIO_FAILURE_MODE_CONTINUE 1
Bug#592676: can't import pool after exporting it
On 8/13/10, Tuco wrote: > On 8/12/10, Petr Salinger wrote: >>> Attached patch fixes the first problem: zpool searches for disks in >>> /dev/dsk instead of /dev, but then importing still fails. >> >> Do you have access to our SVN, don't you ? ;-) > > Yes but I'm thinking I'm not sure this would be the right fix, as the > problem persists. Also, FreeBSD zpool is not affected, and they build > from the same source. I'm confused by this. I found the real problem, and also commited a few other fixes. With zfsutils from SVN I can 'zpool import' now :-) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#591837: Wine 1.2 packages with wine-gecko
For kfreebsd, wine 1.2 has the same problem, except in a different
place. Here's a new patch. Again I verified it works by running a
simple windows program.
For the mingw32 stuff, I got this error when trying to install your package:
Unpacking mingw-w32 (from mingw-w32_0~20100726-0.1_all.deb) ...
dpkg: error processing mingw-w32_0~20100726-0.1_all.deb (--install):
trying to overwrite '/usr/i586-mingw32msvc/include/mbstring.h', which
is also in package mingw32-runtime 3.13-1
On 8/12/10, Stephen Kitt wrote:
> Hi everyone,
>
> I've prepared packages of gcc-mingw32 4.5.0, a snapshot of mingw-w64,
> wine-gecko 1.0.0 and wine 1.2; they are available on
> http://www.sk2.org/wine/
> along with everything needed to rebuild them (build gcc-mingw32 first, then
> mingw-w64, wine-gecko and wine).
>
> I've already attempted to contact Ove and the Wine packaging team but have
> not received any response so far. Given the time of year it's possible
> everyone involved is on holiday! Since there is some unrest around having
> newer versions of Wine in Debian, and since I'd done most of the work
> already, I thought I'd send this announcement and avoid unnecessary
> duplicate
> work.
>
> gcc-mingw32 and ming-w64 are just the minimum required to build wine-gecko;
> I'm not sure they're appropriate for replacing the existing Mingw32-based
> packages in Debian. I haven't contacted the people involved yet.
>
> wine-gecko itself works on i386, and as far as I can determine on amd64 (at
> least the Steam browser works). It should probably end up with the version
> in
> the package name, to allow multiple versions to be installed in parallel
> (for
> instance if future versions of wine-unstable use a newer version of
> wine-gecko, once wine-unstable becomes installable alongside wine).
>
> The wine package includes everything in Ove's git tree, and newer programs
> shipped in Wine 1.2. The changelog is complete, but doesn't list upstream's
> change summaries as Ove usually does.
>
> I'd appreciate your feedback! Regarding #591837, the provided patch doesn't
> work but it may be possible that Wine 1.2 builds as-is on kfreebsd-i386.
>
> Regards,
>
> Stephen
>
--- loader/main.c~ 2010-07-16 15:05:45.0 +
+++ loader/main.c 2010-08-13 10:41:58.0 +
@@ -164,7 +164,7 @@
return 1;
}
-#elif defined(__FreeBSD__) && defined(__i386__)
+#elif (defined(__FreeBSD__) || defined(__FreeBSD_kernel__)) && defined(__i386__)
static int pre_exec(void)
{
Bug#592785: please default to OSS on kfreebsd
Package: libao Version: 1.0.0-4 Severity: normal User: debian-...@lists.debian.org Usertags: kfreebsd X-Debbugs-CC: debian-...@lists.debian.org /etc/libao.conf has default_driver=alsa. kFreeBSD provides ALSA using an emulation layer but it is better to use OSS when possible (Aurelien Jarno explains in http://lists.debian.org/debian-bsd/2010/08/msg00033.html). Could you set default driver to OSS on GNU/kFreeBSD? -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#592676: can't import pool after exporting it
Here's the patch.
On 8/11/10, Tuco wrote:
> Package: zfsutils
> Version: 8.1-3
> Severity: important
>
> t...@debian:~$ sudo zpool create test /dev/ad6s1
> t...@debian:~$ sudo zpool export test
> t...@debian:~$ sudo zpool import test
> cannot open '/dev/dsk': must be an absolute path
> cannot import 'test': no such pool available
>
> Attached patch fixes the first problem: zpool searches for disks in
> /dev/dsk instead of /dev, but then importing still fails.
>
>
>
> --
> To UNSUBSCRIBE, email to debian-bsd-requ...@lists.debian.org
> with a subject of "unsubscribe". Trouble? Contact
> listmas...@lists.debian.org
> Archive:
> http://lists.debian.org/aanlkti=ryop75wyeqku4opz-pejwgwz=r4xtilhdh...@mail.gmail.com
>
>
--- a/cddl/contrib/opensolaris/cmd/zpool/zpool_main.c
+++ b/cddl/contrib/opensolaris/cmd/zpool/zpool_main.c
@@ -1468,7 +1468,11 @@
if (searchdirs == NULL) {
searchdirs = safe_malloc(sizeof (char *));
+#ifdef __sun
searchdirs[0] = "/dev/dsk";
+#else
+ searchdirs[0] = "/dev";
+#endif
nsearch = 1;
}
--- a/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_import.c
+++ b/cddl/contrib/opensolaris/lib/libzfs/common/libzfs_import.c
@@ -848,7 +848,11 @@
size_t pathleft;
struct stat64 statbuf;
nvlist_t *ret = NULL, *config;
+#ifdef __sun
static char *default_dir = "/dev/dsk";
+#else
+ static char *default_dir = "/dev";
+#endif
int fd;
pool_list_t pools = { 0 };
pool_entry_t *pe, *penext;
@@ -894,9 +898,11 @@
* reading the labels skips a bunch of slow operations during
* close(2) processing, so we replace /dev/dsk with /dev/rdsk.
*/
+#ifdef __sun
if (strcmp(path, "/dev/dsk/") == 0)
rdsk = "/dev/rdsk/";
else
+#endif
rdsk = path;
if ((dirp = opendir(rdsk)) == NULL) {
Bug#592676: can't import pool after exporting it
Package: zfsutils Version: 8.1-3 Severity: important t...@debian:~$ sudo zpool create test /dev/ad6s1 t...@debian:~$ sudo zpool export test t...@debian:~$ sudo zpool import test cannot open '/dev/dsk': must be an absolute path cannot import 'test': no such pool available Attached patch fixes the first problem: zpool searches for disks in /dev/dsk instead of /dev, but then importing still fails. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#592292: report correct disk size on kfreebsd
Package: util-linux
Version: 2.17.2-3.1
Severity: important
Tags: patch
User: debian-...@lists.debian.org
Usertags: kfreebsd
X-Debbugs-CC: debian-...@lists.debian.org
Hi,
This fix makes fdisk report the correct disk size on Debian GNU/kFreeBSD.
diff -ur util-linux-2.17.2/configure.ac util-linux-2.17.2.fix/configure.ac
--- util-linux-2.17.2/configure.ac 2010-08-07 08:33:22.0 -0400
+++ util-linux-2.17.2.fix/configure.ac 2010-08-07 07:47:36.752135651 -0400
@@ -94,7 +94,7 @@
AC_SYS_LARGEFILE
-AC_CHECK_HEADERS([linux/compiler.h linux/blkpg.h linux/major.h], [], [], [
+AC_CHECK_HEADERS([linux/compiler.h linux/blkpg.h linux/major.h sys/disk.h], [], [], [
#ifdef HAVE_LINUX_COMPILER_H
#include
#endif
diff -ur util-linux-2.17.2/lib/blkdev.c util-linux-2.17.2.fix/lib/blkdev.c
--- util-linux-2.17.2/lib/blkdev.c 2010-03-22 04:05:42.0 -0400
+++ util-linux-2.17.2.fix/lib/blkdev.c 2010-08-07 08:09:42.223428053 -0400
@@ -3,6 +3,10 @@
#include
#include
+#ifdef HAVE_SYS_DISK_H
+#include
+#endif
+
#include "blkdev.h"
#include "linux_version.h"
@@ -67,6 +71,17 @@
return -1;
#endif /* BLKGETSIZE */
+#ifdef DIOCGMEDIASIZE
+ {
+ off_t size;
+
+ if (ioctl(fd, DIOCGMEDIASIZE, &size) >= 0) {
+ *bytes = size;
+ return 0;
+ }
+ }
+#endif /* DIOCGMEDIASIZE */
+
*bytes = blkdev_find_size(fd);
return 0;
}
Bug#591816: lib32ncurses on kfreebsd-amd64
On 8/5/10, Sven Joachim wrote: > …for wine you would also need ia32-libs, wouldn't you? I don't know if > anybody wants to maintain that mess on yet another architecture. I made ia32-libs work, but yes it's a mess :-(. I just sent a mail to debian-bsd asking about that. > Besides, wine has not even been built on kfreebsd-i386 yet. I made it build, my patch is in http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=591837 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#591837: wine on kfreebsd-i386
Package: wine Version: 1.0.1-3 Severity: important Tags: patch User: debian-...@lists.debian.org Usertags: kfreebsd X-Debbugs-CC: debian-...@lists.debian.org Hi With this patch, wine builds on kfreebsd-i386. I also tested it with a simple Windows program. --- configure.ac 2010-08-05 12:21:31.0 -0400 +++ configure.ac.fixed 2010-08-05 16:08:36.856317532 -0400 @@ -1308,10 +1308,10 @@ case $host_cpu in *i[[3456789]]86*) case $host_os in - linux* | k*bsd*-gnu) + linux*) AC_SUBST(MAIN_BINARY,"wine-glibc") AC_SUBST(EXTRA_BINARIES,"wine-kthread wine-pthread wine-preloader") ;; - freebsd*) + freebsd* | kfreebsd*-gnu) AC_SUBST(MAIN_BINARY,"wine-freebsd") AC_SUBST(EXTRA_BINARIES,"wine-pthread") ;; esac
Bug#591823: please include init.d script
On 8/5/10, Henrik Riomar wrote:
> Package: zfsutils
> Version: 8.1-1
> Severity: wishlist
>
> When rebooting the zfs volumes that where active at reboot are not
> available until you manually perform;
> zfs mount -a
/etc/init.d/mountall.sh runs mount -a, maybe mount can be fixed to
also do ZFS? Like adding a system("zfs mount -a") call
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#591816: lib32ncurses on kfreebsd-amd64
Package: ncurses
Version: 5.7+20100313-2
Severity: wishlist
Tags: patch
User: debian-...@lists.debian.org
Usertags: kfreebsd
X-Debbugs-CC: debian-...@lists.debian.org
Hi
This patch adds lib32ncurses on kfreebsd-amd64. It is needed for wine.
diff -ur ncurses-5.7+20100313/debian/control ncurses-5.7+20100313.fix/debian/control
--- ncurses-5.7+20100313/debian/control 2010-08-05 13:04:06.0 -0400
+++ ncurses-5.7+20100313.fix/debian/control 2010-08-05 11:40:55.888154343 -0400
@@ -101,7 +101,7 @@
architecure.
Package: lib32ncurses5
-Architecture: amd64 ppc64
+Architecture: amd64 ppc64 kfreebsd-amd64
Priority: optional
Depends: ${shlibs:Depends}, ${misc:Depends}
Conflicts: libc6-i386 (<= 2.9-18)
@@ -114,7 +114,7 @@
architecure.
Package: lib32ncurses5-dev
-Architecture: amd64 ppc64
+Architecture: amd64 ppc64 kfreebsd-amd64
Section: libdevel
Priority: optional
Depends: lib32ncurses5 (= ${binary:Version}), libncurses5-dev, lib32c-dev, ${misc:Depends}
@@ -127,7 +127,7 @@
architecure.
Package: lib32ncursesw5
-Architecture: amd64 ppc64
+Architecture: amd64 ppc64 kfreebsd-amd64
Priority: optional
Depends: ${shlibs:Depends}, ${misc:Depends}
Conflicts: libc6-i386 (<= 2.9-18)
@@ -139,7 +139,7 @@
architecure.
Package: lib32ncursesw5-dev
-Architecture: amd64 ppc64
+Architecture: amd64 ppc64 kfreebsd-amd64
Section: libdevel
Priority: optional
Depends: lib32ncursesw5 (= ${binary:Version}), libncursesw5-dev, lib32c-dev, ${misc:Depends}
diff -ur ncurses-5.7+20100313/debian/rules ncurses-5.7+20100313.fix/debian/rules
--- ncurses-5.7+20100313/debian/rules 2010-08-05 13:04:06.0 -0400
+++ ncurses-5.7+20100313.fix/debian/rules 2010-08-05 10:45:48.297694583 -0400
@@ -66,7 +66,7 @@
build_64 = build-64
endif
-ifeq ($(DEB_HOST_ARCH),amd64)
+ifneq ($(findstring amd64,$(DEB_HOST_ARCH)),)
build_32_target = i486-$(DEB_HOST_GNU_SYSTEM)
build_32 = build-32 build-wide-32
lib32 = /lib32
Bug#591721: patch for kFreeBSD
Package: svgalib
Version: 1:1.4.3-29
Severity: important
Tags: patch
User: debian-...@lists.debian.org
Usertags: kfreebsd
X-Debbugs-CC: debian-...@lists.debian.org
Hi,
I made svgalib build on GNU/kFreeBSD, here is a patch.
diff -ur svgalib-1.4.3.orig/debian/control svgalib-1.4.3/debian/control
--- svgalib-1.4.3.orig/debian/control 2009-11-14 00:50:00.0 -0500
+++ svgalib-1.4.3/debian/control 2010-08-04 19:56:00.117398564 -0400
@@ -7,7 +7,7 @@
Vcs-Browser: http://git.hadrons.org/?p=debian/pkgs/svgalib.git
Vcs-Git: git://git.hadrons.org/git/debian/pkgs/svgalib.git
Standards-Version: 3.8.3
-Build-Depends: debhelper (>= 7), libx86-dev [amd64 i386]
+Build-Depends: debhelper (>= 7), libx86-dev [amd64 i386 kfreebsd-amd64 kfreebsd-i386]
Package: svgalib-bin
Architecture: any
diff -ur svgalib-1.4.3.orig/src/keyboard/keyboard.c svgalib-1.4.3/src/keyboard/keyboard.c
--- svgalib-1.4.3.orig/src/keyboard/keyboard.c 2010-08-04 13:43:15.0 -0400
+++ svgalib-1.4.3/src/keyboard/keyboard.c 2010-08-04 19:54:40.714634914 -0400
@@ -33,7 +33,11 @@
/* should also be useful for svgalib programs using the keyboard. It misses */
/* a few KERNEL ifdefs around kernel data structures though. */
#include "keyboard_lnx.h"
+#if defined(__linux__)
#include
+#elif defined(__FreeBSD_kernel__)
+#include
+#endif
/* Needed to check uid of keymap files */
#include
#include
@@ -385,6 +389,22 @@
__svgalib_kbd_fd = -1;
}
+static inline int vt_getactive(int fd)
+{
+int v;
+#ifdef VT_GETSTATE
+struct vt_stat vts;
+if (ioctl(fd, VT_GETSTATE, &vts) == 0)
+v = vts.v_active;
+else
+v = -1;
+#else
+if (ioctl(fd, VT_GETACTIVE, &v) != 0)
+v = -1;
+#endif
+return v;
+}
+
/* For now, we assume there's no console switching. */
/* (Actually, there won't be any unless we catch the console switching */
/* keys). */
@@ -578,7 +598,6 @@
/* VT switch. */
/* *** what about F11 & F12? */
int j, vt = 0;
- struct vt_stat vts;
for (j = 0; j < 12; j++)
if (functionkey_state & (1 << j)) {
vt = j + 1;
@@ -587,9 +606,7 @@
}
/* Do not switch vt's if need not to */
- ioctl(__svgalib_tty_fd, VT_GETSTATE, &vts);
-
- if(vt != vts.v_active) {
+ if(vt != vt_getactive(__svgalib_tty_fd)) {
/* if switching vt's, need to clear keystates */
keyboard_clearstate();
/*
diff -ur svgalib-1.4.3.orig/src/mouse/ms.c svgalib-1.4.3/src/mouse/ms.c
--- svgalib-1.4.3.orig/src/mouse/ms.c 2010-08-04 13:43:15.0 -0400
+++ svgalib-1.4.3/src/mouse/ms.c 2010-08-04 19:49:23.495541626 -0400
@@ -117,7 +117,9 @@
tty.c_iflag = IGNBRK | IGNPAR;
tty.c_oflag = 0;
tty.c_lflag = 0;
+#ifdef __linux__
tty.c_line = 0;
+#endif
tty.c_cc[VTIME] = 0;
tty.c_cc[VMIN] = 1;
@@ -527,7 +529,9 @@
tty.c_iflag = IGNBRK | IGNPAR;
tty.c_oflag = 0;
tty.c_lflag = 0;
+#ifdef __linux__
tty.c_line = 0;
+#endif
tty.c_cc[VTIME] = 0;
tty.c_cc[VMIN] = 1;
tty.c_cflag = cflag[m_type] | B1200;
diff -ur svgalib-1.4.3.orig/src/vga.c svgalib-1.4.3/src/vga.c
--- svgalib-1.4.3.orig/src/vga.c 2010-08-04 13:43:15.0 -0400
+++ svgalib-1.4.3/src/vga.c 2010-08-04 19:53:16.757293455 -0400
@@ -25,7 +25,11 @@
#include
#include
#include
+#if defined(__linux__)
#include
+#elif defined(__FreeBSD_kernel__)
+#include
+#endif
#include
#include
#include
@@ -781,7 +785,7 @@
/* Leave keyboard alone when rawkeyboard is enabled! */
if (__svgalib_kbd_fd < 0) {
/* set graphics mode termio parameters */
- ioctl(0, TCSETSW, &graph_termio);
+ tcsetattr(0, TCSADRAIN, &graph_termio);
}
}
@@ -791,7 +795,7 @@
/* Leave keyboard alone when rawkeyboard is enabled! */
if (__svgalib_kbd_fd < 0) {
/* restore text mode termio parameters */
- ioctl(0, TCSETSW, &text_termio);
+ tcsetattr(0, TCSADRAIN, &text_termio);
}
}
@@ -802,9 +806,9 @@
/* Well, one could argue that sigint is not enabled at all when in __svgalib_nosigint
but sometimes they *still* are enabled b4 graph_termio is set.. */
-ioctl(0, TCGETS, &cur_termio);
+tcgetattr(0, &cur_termio);
cur_termio.c_lflag &= ~ISIG;
-ioctl(0, TCSETSW, &cur_termio);
+tcsetattr(0, TCSADRAIN, &cur_termio);
}
@@ -814,9 +818,9 @@
if (__svgalib_nosigint) /* do not reenable, they are often reenabled by text_termio */
return;
-ioctl(0, TCGETS, &cur_termio);
+tcgetattr(0, &cur_termio);
cur_termio.c_lflag |= ISIG;
-ioctl(0, TCSETSW, &cur_termio);
+tcsetattr(0, TCSADRAIN, &cur_termio);
}
/* The following is rather messy and inelegant. The only solution I can */
@@ -938,10 +942,26 @@
return 0;
}
+static inline int vt_getactive(int fd)
+{
+int v;
+#ifdef VT_GETSTATE
+struct vt_stat vts;
+if (ioctl(fd, VT_GETSTATE, &vts) == 0)
+v = vts.v_active;
+else
+v = -1;
+#else
+if (ioctl(fd, VT_GETACTIVE, &v) != 0)
+v = -1;
+#endif
+return
Bug#591529: does not compile
Package: kfreebsd-kernel-headers Version: 0.47 Severity: normal When trying to build getnetconfig.c (obtained from src/sys/rpc/ in FreeBSD): In file included from getnetconfig.c:35: /usr/include/sys/systm.h:46:62: error: sys/stdint.h: No such file or directory In file included from getnetconfig.c:35: /usr/include/sys/systm.h:137: warning: no previous prototype for ‘longjmp’ /usr/include/sys/systm.h: In function ‘longjmp’: /usr/include/sys/systm.h:137: error: expected declaration specifiers before ‘__dead2’ /usr/include/sys/systm.h:138: error: expected ‘)’ before ‘addr’ lots of errors follow, this header is unusable :-( -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#559364: [RFC]: Bug#559364: regression: console text is red
On 8/2/10, Aurelien Jarno wrote: > I confirm the issue here, so it seems we have to revert the options in > the kernel. That's a pitty because the goal was to use the UTF-8 > terminal to be able to have non-english language in debian-installer. Maybe this sounds crazy, but what if you put this in /sbin/init: #!/bin/sh exec xinit /usr/bin/xterm -e /sbin/oldinit and then you have UTF-8? -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#590730: missing
Package: kfreebsd-kernel-headers Version: 0.50 Severity: wishlist Hi is needed for ZFS too. Sorry I ddn't mention before but it wan't needed with FreeBSD 8.0 source. With 8.1 it is. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#590639: add libzfs
On 7/28/10, Petr Salinger wrote: > Thanks for your work, but IMO, it does not belong into > source package freebsd-libs. Not all kfreebsd specific libs/utils > should come from freebsd-libs/freebsd-utils source package, > see i.e. ufsutils for UFS filesystem. > > Please instead try to prepare new source package zfsutils, > which would produce needed binary packages with both libs and utils. > Please base your work on 8.1 upstream sources, the kernel have been > already uploaded and libs/utils will follow soon, they are prepared > in our SVN. OK. I'm not very used to Debian packaging but I think I can adjust freebsd-libs or freebsd-utils to provide zfs libs and utils only. > You also might "Request to join" at [1] and get access > into our SVN [2] and put your work there. Thanks, I'll have a look at that. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#590639: add libzfs
Package: freebsd-libs
Severity: wishlist
Tags: patch
This patch adds libzfs and related libraries to freebsd-libs:
libzfs
libavl
libnvpair
libuutil
These are all the libraries necessary to build the ZFS tools (zpool
and zfs). I will provide a patch for freebsd-utils too.
diff -Nur freebsd-libs-8.0.old/debian/control freebsd-libs-8.0/debian/control
--- freebsd-libs-8.0.old/debian/control 2010-07-27 19:54:52.0 -0400
+++ freebsd-libs-8.0/debian/control 2010-07-27 19:59:31.942892493 -0400
@@ -218,3 +218,17 @@
Description: Development files for libkiconv
This package contains the header files and static library needed to
compile applications that use libkiconv.
+
+Package: libzfs0
+Architecture: kfreebsd-any
+Depends: ${shlibs:Depends}, ${misc:Depends}
+Description: FreeBSD ZFS and related libraries
+ The ZFS libraries provide support for managing ZFS filesystems.
+
+Package: libzfs-dev
+Section: libdevel
+Architecture: kfreebsd-any
+Depends: libzfs0 (= ${binary:Version}), ${misc:Depends}, libc0.1-dev | libc-dev
+Description: Development files for libzfs
+ This package contains the header files and static library needed to
+ compile applications that use libzfs.
diff -Nur freebsd-libs-8.0.old/debian/libzfs0.install freebsd-libs-8.0/debian/libzfs0.install
--- freebsd-libs-8.0.old/debian/libzfs0.install 1969-12-31 19:00:00.0 -0500
+++ freebsd-libs-8.0/debian/libzfs0.install 2010-07-27 20:18:10.902701936 -0400
@@ -0,0 +1,4 @@
+usr/lib/libavl*.so.* lib
+usr/lib/libzfs*.so.* lib
+usr/lib/libuutil*.so.* lib
+usr/lib/libnvpair*.so.* lib
diff -Nur freebsd-libs-8.0.old/debian/libzfs-dev.install freebsd-libs-8.0/debian/libzfs-dev.install
--- freebsd-libs-8.0.old/debian/libzfs-dev.install 1969-12-31 19:00:00.0 -0500
+++ freebsd-libs-8.0/debian/libzfs-dev.install 2010-07-27 20:03:31.842799801 -0400
@@ -0,0 +1,8 @@
+usr/lib/libavl*.a
+usr/lib/libavl.so
+usr/lib/libzfs*.a
+usr/lib/libzfs.so
+usr/lib/libuutil*.a
+usr/lib/libuutil.so
+usr/lib/libnvpair*.a
+usr/lib/libnvpair.so
diff -Nur freebsd-libs-8.0.old/debian/patches/14_zfs_glibc_kludge.diff freebsd-libs-8.0/debian/patches/14_zfs_glibc_kludge.diff
--- freebsd-libs-8.0.old/debian/patches/14_zfs_glibc_kludge.diff 1969-12-31 19:00:00.0 -0500
+++ freebsd-libs-8.0/debian/patches/14_zfs_glibc_kludge.diff 2010-07-27 20:08:27.215970169 -0400
@@ -0,0 +1,96 @@
+diff -ur -x debian freebsd-libs-8.0.old/cddl/compat/opensolaris/misc/mnttab.c freebsd-libs-8.0/cddl/compat/opensolaris/misc/mnttab.c
+--- freebsd-libs-8.0.old/cddl/compat/opensolaris/misc/mnttab.c 2008-11-17 15:49:29.0 -0500
freebsd-libs-8.0/cddl/compat/opensolaris/misc/mnttab.c 2010-07-27 20:07:02.656479850 -0400
+@@ -89,7 +89,7 @@
+ }
+
+ int
+-getmntany(FILE *fd __unused, struct mnttab *mgetp, struct mnttab *mrefp)
++getmntany(FILE *fd, struct mnttab *mgetp, struct mnttab *mrefp)
+ {
+ static struct statfs *sfs = NULL;
+ static char mntopts[MNTMAXSTR];
+diff -ur -x debian freebsd-libs-8.0.old/cddl/compat/opensolaris/misc/zmount.c freebsd-libs-8.0/cddl/compat/opensolaris/misc/zmount.c
+--- freebsd-libs-8.0.old/cddl/compat/opensolaris/misc/zmount.c 2008-11-17 15:49:29.0 -0500
freebsd-libs-8.0/cddl/compat/opensolaris/misc/zmount.c 2010-07-27 20:07:02.659058030 -0400
+@@ -92,6 +92,9 @@
+ iov = NULL;
+ iovlen = 0;
+ build_iovec(&iov, &iovlen, "fstype", fstype, (size_t)-1);
++#ifndef __DECONST
++#define __DECONST(type, var) ((type)(uintptr_t)(const void *)(var))
++#endif
+ build_iovec(&iov, &iovlen, "fspath", __DECONST(char *, dir),
+ (size_t)-1);
+ build_iovec(&iov, &iovlen, "from", __DECONST(char *, spec), (size_t)-1);
+diff -ur -x debian freebsd-libs-8.0.old/cddl/contrib/opensolaris/lib/libuutil/common/uu_pname.c freebsd-libs-8.0/cddl/contrib/opensolaris/lib/libuutil/common/uu_pname.c
+--- freebsd-libs-8.0.old/cddl/contrib/opensolaris/lib/libuutil/common/uu_pname.c 2008-03-28 18:16:18.0 -0400
freebsd-libs-8.0/cddl/contrib/opensolaris/lib/libuutil/common/uu_pname.c 2010-07-27 20:07:02.660028620 -0400
+@@ -31,7 +31,7 @@
+ #include
+ #include
+ #include
+-#include
++#include
+ #include
+ #include
+ #include
+diff -ur -x debian freebsd-libs-8.0.old/sys/cddl/compat/opensolaris/rpc/xdr.h freebsd-libs-8.0/sys/cddl/compat/opensolaris/rpc/xdr.h
+--- freebsd-libs-8.0.old/sys/cddl/compat/opensolaris/rpc/xdr.h 2009-08-19 20:08:58.0 -0400
freebsd-libs-8.0/sys/cddl/compat/opensolaris/rpc/xdr.h 2010-07-27 20:07:02.661041790 -0400
+@@ -32,7 +32,31 @@
+
+ #include_next
+
+-#ifndef _KERNEL
++/* Copied from the FreeBSD version of */
++#ifdef __GLIBC__
++/*
++ * These are XDR control operators
++ */
++
++#define XDR_GET_BYTES_AVAIL 1
++#define XDR_PEEK 2
++#define XDR_SKIPBYTES 3
++
++struct xdr_bytesrec {
++ bool_t xc_is_last_record;
++ size_t xc_num_avail;
++};
++
++typedef struct xdr_bytesrec xdr_bytesrec;
++#endif
++
++#if defined(__GLIBC__)
++
++/* GNU libc doesn't implement XDR control method */
++#define XDR_CONTROL
Bug#590637: add libzfs
Package: freebsd-libs Severity: wishlist Tags: patch This patch adds libzfs and related libraries to freebsd-libs: libzfs libavl libnvpair libuutil These are all the libraries necessary to build the ZFS tools (zpool and zfs). I will provide a patch for freebsd-utils too. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#590628: please add
Package: kfreebsd-kernel-headers Version: 0.47 Severity: wishlist Tags: patch Hi, could you please add ? It is needed by libzfs. diff -ur kfreebsd-kernel-headers-0.47/debian/install kfreebsd-kernel-headers-0.47.sdt/debian/install --- kfreebsd-kernel-headers-0.47/debian/install 2009-11-29 09:34:55.0 -0500 +++ kfreebsd-kernel-headers-0.47.sdt/debian/install 2010-07-26 19:52:59.183650383 -0400 @@ -262,6 +262,7 @@ sys/sys/rtprio.husr/include/sys sys/sys/runq.h usr/include/sys sys/sys/sbuf.h usr/include/sys +sys/sys/sdt.h usr/include/sys sys/sys/selinfo.h usr/include/sys sys/sys/sigio.h usr/include/sys sys/sys/signalvar.h usr/include/sys
Bug#587630: kernel panic when using ZFS
On Mon, Jul 19, 2010 at 6:33 PM, Aurelien Jarno wrote: > This has been caused by the switch from freebsd-buildutils 7.2 to 8.0 > and is also likely the cause of bug#587629. > > I will try to narrow down the problem even more. Just tried your fixed package and it is working fine, thank you! -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#587630: kernel panic when using ZFS
On 7/9/10, Tuco wrote: > This doesn't happen when using a native FreeBSD kernel on my Debian > GNU/kFreeBSD. I noticed a few patches related to ZFS, but disabling > them has no effect. I'm not a kernel developer, anything I can do to > help solve this? To be a bit more specific, for this workaround to work one doesn't need to copy the whole kernel. If I copy opensolaris.ko and zfs.ko over from FreeBSD (only those files), then ZFS works perfectly. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#587630: kernel panic when using ZFS
On 6/30/10, Tuco wrote: > $ zpool create mypool da1 > panic: stack overflow detected; backtrace may be corrupted > cpuid = 0 > Uptime: 1m10s > Cannot dump. Device not defined or unavailable. > Automatic reboot in 15 seconds - press a key on the console to abort This doesn't happen when using a native FreeBSD kernel on my Debian GNU/kFreeBSD. I noticed a few patches related to ZFS, but disabling them has no effect. I'm not a kernel developer, anything I can do to help solve this? -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#588538: fix kern.module_path for native kernel
On 7/9/10, Petr Salinger wrote: >> Using a native FreeBSD kernel on Debian GNU/kFreeBSD brings trouble >> because of kern.module_path not being set early enough (it is needed >> when /etc/init.d/freebsd-utils mounts /proc). Here is a patch. > > What do you mean by native FreeBSD kernel ? > The stock FreeBSD kernel without GNU/kFreeBSD patches ? Yes > At least 007_clone_signals.diff is essential for threaded programs. :-( -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#588538: fix kern.module_path for native kernel
Package: freebsd-utils Version: 8.0-11 Severity: wishlist Tags: patch Hi! Using a native FreeBSD kernel on Debian GNU/kFreeBSD brings trouble because of kern.module_path not being set early enough (it is needed when /etc/init.d/freebsd-utils mounts /proc). Here is a patch. --- /etc/init.d/freebsd-utils~ 2010-07-02 22:02:19.0 -0400 +++ /etc/init.d/freebsd-utils 2010-07-09 09:16:37.0 -0400 @@ -28,6 +28,8 @@ case "$1" in start) +sysctl "kern.module_path=/lib/modules/`uname -r`;/boot/kernel" + if [ "$(readlink /etc/mtab)" != "/proc/mounts" ] ; then log_warning_msg "Warning: replacing /etc/mtab by a symlink to /proc/mounts." rm -f /etc/mtab
Bug#587630: kernel panic when using ZFS
Package: kfreebsd-image-8.0-1-amd64 Severity: important Version: 8.0-9 I installed ZFS utilities by copying them from FreeBSD. When running zpool, kernel panics: $ zpool create mypool da1 panic: stack overflow detected; backtrace may be corrupted cpuid = 0 Uptime: 1m10s Cannot dump. Device not defined or unavailable. Automatic reboot in 15 seconds - press a key on the console to abort -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#587629: can't load any module
Package: kfreebsd-image-7.3-1-amd64 Severity: important Version: 7.3-4 I can't load any module: $ kldload pf link_elf_obj: symbol __stack_chk_fail undefined kldload: /lib/modules/7.3-1-amd64/pf.ko: Unsupported file type kldload: can't load pf: Exec format error -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#587588: build on kfreebsd
On 6/30/10, Cyril Brulebois wrote: > Thanks for that. > > Some comments: > - it'd be nice to tag the bug as having a patch. > - it'd also be nice to usertag the bug when submitting. > - it's not a serious bug, since pwlib never built on kfreebsd-*, >that's only important. Ok. Thanks to both of you for the advice, I will follow these guidelines. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#587588: build on kfreebsd
Package: pwlib
Version: 1.10.10-3
Severity: serious
This patch fixes build problems of pwlib on kfreebsd.
Only in pwlib-1.10.10: config.guess
Only in pwlib-1.10.10: config.sub
diff -ur pwlib-1.10.10/configure pwlib-1.10.10.fix/configure
--- pwlib-1.10.10/configure 2007-09-03 16:57:56.0 -0400
+++ pwlib-1.10.10.fix/configure 2010-06-29 20:09:53.0 -0400
@@ -2639,7 +2639,7 @@
NEEDS_PTHREADS=yes
case "$target_os" in
- linux*) OSTYPE=linux ;
+ linux*|kfreebsd*) OSTYPE=linux ;
OSRELEASE="\"`uname -r`\"";
OS_TAG="P_LINUX" ;
need_pragma=yes ;
diff -ur pwlib-1.10.10/configure.ac pwlib-1.10.10.fix/configure.ac
--- pwlib-1.10.10/configure.ac 2007-09-03 16:57:57.0 -0400
+++ pwlib-1.10.10.fix/configure.ac 2010-06-29 20:09:38.0 -0400
@@ -62,7 +62,7 @@
NEEDS_PTHREADS=yes
case "$target_os" in
- linux*) OSTYPE=linux ;
+ linux*|kfreebsd*) OSTYPE=linux ;
OSRELEASE="\"`uname -r`\"";
OS_TAG="P_LINUX" ;
need_pragma=yes ;
diff -ur pwlib-1.10.10/debian/control pwlib-1.10.10.fix/debian/control
--- pwlib-1.10.10/debian/control 2010-06-29 20:38:29.0 -0400
+++ pwlib-1.10.10.fix/debian/control 2010-06-29 20:32:17.0 -0400
@@ -62,7 +62,7 @@
Package: libpt-1.10.10-plugins-v4l
Section: libs
-Architecture: any
+Architecture: a
Depends: ${shlibs:Depends}
Conflicts: libpt-plugins-v4l
Replaces: libpt-plugins-v4l
@@ -74,7 +74,7 @@
Package: libpt-1.10.10-plugins-v4l2
Section: libs
-Architecture: any
+Architecture: a
Depends: ${shlibs:Depends}
Conflicts: libpt-plugins-v4l2
Replaces: libpt-plugins-v4l2
@@ -96,7 +96,7 @@
interface..
Package: libpt-1.10.10-plugins-alsa
-Architecture: any
+Architecture: a
Depends: ${shlibs:Depends}
Recommends: alsa-base
Conflicts: libpt-plugins-alsa
Only in pwlib-1.10.10.fix/debian: libpt-1.10.10-dbg.dirs
Only in pwlib-1.10.10.fix/debian: libpt-1.10.10-plugins-alsa.debhelper.log
Only in pwlib-1.10.10.fix/debian: libpt-1.10.10-plugins-v4l2.debhelper.log
Only in pwlib-1.10.10.fix/debian: libpt-1.10.10-plugins-v4l.debhelper.log
diff -ur pwlib-1.10.10/debian/rules pwlib-1.10.10.fix/debian/rules
--- pwlib-1.10.10/debian/rules 2010-06-29 20:38:29.0 -0400
+++ pwlib-1.10.10.fix/debian/rules 2010-06-29 20:31:54.0 -0400
@@ -207,14 +207,14 @@
# plugins
#$(PACKAGE)-plugins-v4l
- install -d -m 755 debian/$(PACKAGE)-plugins-v4l/usr/lib/pwlib/$(SHLIBSVER)/device/videoinput/
- cp plugins/pwlib/device/videoinput/v4l_pwplugin.so \
- debian/$(PACKAGE)-plugins-v4l/usr/lib/pwlib/$(SHLIBSVER)/device/videoinput/
+# install -d -m 755 debian/$(PACKAGE)-plugins-v4l/usr/lib/pwlib/$(SHLIBSVER)/device/videoinput/
+# cp plugins/pwlib/device/videoinput/v4l_pwplugin.so \
+# debian/$(PACKAGE)-plugins-v4l/usr/lib/pwlib/$(SHLIBSVER)/device/videoinput/
#$(PACKAGE)-plugins-v4l2
- install -d -m 755 debian/$(PACKAGE)-plugins-v4l2/usr/lib/pwlib/$(SHLIBSVER)/device/videoinput/
- cp plugins/pwlib/device/videoinput/v4l2_pwplugin.so \
- debian/$(PACKAGE)-plugins-v4l2/usr/lib/pwlib/$(SHLIBSVER)/device/videoinput/
+# install -d -m 755 debian/$(PACKAGE)-plugins-v4l2/usr/lib/pwlib/$(SHLIBSVER)/device/videoinput/
+# cp plugins/pwlib/device/videoinput/v4l2_pwplugin.so \
+# debian/$(PACKAGE)-plugins-v4l2/usr/lib/pwlib/$(SHLIBSVER)/device/videoinput/
#$(PACKAGE)-plugins-oss
install -d -m 755 debian/$(PACKAGE)-plugins-oss/usr/lib/pwlib/$(SHLIBSVER)/device/sound/
@@ -222,9 +222,9 @@
debian/$(PACKAGE)-plugins-oss/usr/lib/pwlib/$(SHLIBSVER)/device/sound/
#$(PACKAGE)-plugins-alsa
- install -d -m 755 debian/$(PACKAGE)-plugins-alsa/usr/lib/pwlib/$(SHLIBSVER)/device/sound/
- cp plugins/pwlib/device/sound/alsa_pwplugin.so \
- debian/$(PACKAGE)-plugins-alsa/usr/lib/pwlib/$(SHLIBSVER)/device/sound/
+# install -d -m 755 debian/$(PACKAGE)-plugins-alsa/usr/lib/pwlib/$(SHLIBSVER)/device/sound/
+# cp plugins/pwlib/device/sound/alsa_pwplugin.so \
+# debian/$(PACKAGE)-plugins-alsa/usr/lib/pwlib/$(SHLIBSVER)/device/sound/
# Must not depend on anything. This is to be called by
# binary-arch/binary-multi
Only in pwlib-1.10.10.fix/plugins/vidinput_dc: Makefile
diff -ur pwlib-1.10.10/src/ptlib/unix/pipechan.cxx pwlib-1.10.10.fix/src/ptlib/unix/pipechan.cxx
--- pwlib-1.10.10/src/ptlib/unix/pipechan.cxx 2005-11-30 07:47:42.0 -0500
+++ pwlib-1.10.10.fix/src/ptlib/unix/pipechan.cxx 2010-06-29 20:21:34.0 -0400
@@ -149,7 +149,7 @@
#include
#if defined(P_LINUX) || defined(P_SOLARIS)
-#include
+#include
#endif
#if defined(P_MACOSX)
diff -ur pwlib-1.10.10/src/ptlib/unix/serchan.cxx pwlib-1.10.10.fix/src/ptlib/unix/serchan.cxx
--- pwlib-1.10.10/src/ptlib/unix/serchan.cxx 2005-11-30 07:47:42.0 -0500
+++ pwlib-1.10.10.fix/src/ptlib/unix/serchan.cxx 2010-06-29 20:22:47.0 -0400
@@ -435,7 +435,7 @@
// save new baud rate
baudRate = newBaudRate;
-#if defined(P_FREEBSD) || defined(P_OPENBSD) || defin
Bug#586539: gdm problem fixed
Hi, This patch solved the problem for me: --- gdm-2.20.10/daemon/getvt.h 2009-03-19 00:59:27.0 -0400 +++ gdm-2.20.10.fixed/daemon/getvt.h2010-06-26 07:57:32.0 -0400 @@ -27,7 +27,7 @@ #endif #endif -#if defined (__FreeBSD__) || defined (__DragonFly__) +#if defined (__FreeBSD__) || defined (__FreeBSD_kernel__) || defined (__DragonFly__) #define GDM_USE_CONSIO_VT #endif -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#469528: hal install fails on kfreebsd
Hi This happened to me, not when upgrading but when installing hal for first time on a kFreeBSD system. After the first time, if I start dbus manually, then hal works. This is the error message: Setting up hal (0.5.14-2) ... Reloading system message bus config...Failed to open connection to "system" message bus: Failed to connect to socket /var/run/dbus/system_bus_socket: Connection refused invoke-rc.d: initscript dbus, action "force-reload" failed. Starting Hardware abstraction layer: haldinvoke-rc.d: initscript hal, action "start" failed. dpkg: error processing hal (--configure): subprocess installed post-installation script returned error exit status 1 Errors were encountered while processing: hal E: Sub-process /usr/bin/dpkg returned an error code (1) -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#587162: ttyd0 -> cuau0
Package: sysvinit Version: 2.88dsf-9 With kFreeBSD 8.0, serial devices are named cuau*, not ttyd* anymore. --- /tmp/inittab 2010-06-25 18:31:11.0 +0200 +++ /etc/inittab 2010-06-25 18:32:04.0 +0200 @@ -60,8 +60,8 @@ # Example how to put a getty on a serial line (for a terminal) # -#T0:23:respawn:/sbin/getty -L ttyd0 9600 vt100 -#T1:23:respawn:/sbin/getty -L ttyd1 9600 vt100 +#T0:23:respawn:/sbin/getty -L cuau0 9600 vt100 +#T1:23:respawn:/sbin/getty -L cuau1 9600 vt100 # Example how to put a getty on a modem line. #
Bug#586203: postinst fails because of unrecognized filesystem
Package: linux-base Version: 2.6.32-15 Severity: grave My fstab had the wrong entry for / (hda1 is part of a ZFS pool, my real / as per grub config is hde1), then postinst fails. I think it could report gracefully that my fstab was fubar. Note: wrong entry in fstab is quite common, ironically, because it didn't migrate to UUID yet. I think this problem will happen lots of times as your script migrates systems to UUID. Setting up linux-base (2.6.32-15) ... tune2fs 1.41.12 (17-May-2010) tune2fs: Bad magic number in super-block while trying to open /dev/hda1 Couldn't find valid filesystem superblock. tune2fs failed: 256 at /var/lib/dpkg/info/linux-base.postinst line 1045, line 10. dpkg: error processing linux-base (--configure): subprocess installed post-installation script returned error exit status 9 -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#586150: alsa-lib: confmisc.c:768:(parse_card) cannot find card '0'
Package: libasound2 Version: 1.0.23-1 libasound has a funny way of saying that my user doesn't have the necessary permissions (audio group) to access /dev/snd/ googling for this error returns many results for other kind of problems, which makes it hard to figure out what's going on. $ aplay audiodump.wav Playing WAVE 'audiodump.wav' : Signed 16 bit Little Endian, Rate 44100 Hz, Stereo ^C $ sudo sudo -u nobody aplay audiodump.wav ALSA lib confmisc.c:768:(parse_card) cannot find card '0' ALSA lib conf.c:4154:(_snd_config_evaluate) function snd_func_card_driver returned error: No such file or directory ALSA lib confmisc.c:392:(snd_func_concat) error evaluating strings ALSA lib conf.c:4154:(_snd_config_evaluate) function snd_func_concat returned error: No such file or directory ALSA lib confmisc.c:1251:(snd_func_refer) error evaluating name ALSA lib conf.c:4154:(_snd_config_evaluate) function snd_func_refer returned error: No such file or directory ALSA lib conf.c:4633:(snd_config_expand) Evaluate error: No such file or directory ALSA lib pcm.c:2211:(snd_pcm_open_noupdate) Unknown PCM default aplay: main:608: audio open error: No such file or directory -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#577524: Same problem
> So it'd be great if somebody who has the same problem can run: > apt-get remove iceowl calendar-google-provider calendar-timezones > and send the output of: > find /usr/lib/iceowl /usr/share/iceowl I had the same problem, run this command, then neither /usr/lib/iceowl or /usr/share/iceowl was there. Then I reinstalled those 3 packages, and the problem was gone. HTH -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#579742: closed by Didier Raboud (Bug#579742: fixed in win32-loader 0.6.16)
Cool, thanks. On Wed, May 5, 2010 at 7:03 PM, Debian Bug Tracking System < ow...@bugs.debian.org> wrote: > This is an automatic notification regarding your Bug report > which was filed against the win32-loader package: > > #579742: kfreebsd > > It has been closed by Didier Raboud . > > Their explanation is attached below along with your original report. > If this explanation is unsatisfactory and you have not received a > better one in a separate message then please contact Didier Raboud < > did...@raboud.com> by > replying to this email. > > > -- > 579742: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=579742 > Debian Bug Tracking System > Contact ow...@bugs.debian.org with problems > > > -- Forwarded message -- > From: Didier Raboud > To: 579742-cl...@bugs.debian.org > Date: Wed, 05 May 2010 17:02:36 + > Subject: Bug#579742: fixed in win32-loader 0.6.16 > Source: win32-loader > Source-Version: 0.6.16 > > We believe that the bug you reported is fixed in the latest version of > win32-loader, which is due to be installed in the Debian FTP archive: > > win32-loader_0.6.16.dsc > to main/w/win32-loader/win32-loader_0.6.16.dsc > win32-loader_0.6.16.tar.gz > to main/w/win32-loader/win32-loader_0.6.16.tar.gz > win32-loader_0.6.16_all.deb > to main/w/win32-loader/win32-loader_0.6.16_all.deb > > > > A summary of the changes between this version and the previous one is > attached. > > Thank you for reporting the bug, which will now be closed. If you > have further comments please address them to 579...@bugs.debian.org, > and the maintainer will reopen the bug report if appropriate. > > Debian distribution maintenance software > pp. > Didier Raboud (supplier of updated win32-loader > package) > > (This message was generated automatically at their request; if you > believe that there is a problem with it please contact the archive > administrators by mailing ftpmas...@debian.org) > > > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > Format: 1.8 > Date: Wed, 05 May 2010 18:28:39 +0200 > Source: win32-loader > Binary: win32-loader > Architecture: source all > Version: 0.6.16 > Distribution: unstable > Urgency: low > Maintainer: Debian Install System Team > Changed-By: Didier Raboud > Description: > win32-loader - Debian-Installer loader for win32 > Closes: 579742 > Changes: > win32-loader (0.6.16) unstable; urgency=low > . > * Add basic kFreeBSD support (Closes: #579742) > + What works: fetching and booting a kFreeBSD d-i from standalone > version > - What doesn't: preseeding, stable install, g-i, media install > . > * Move files to /usr/share/win32/ to be consistent with other win32 > packages. > * Sync linux bootparam VIDEO_MODE definition with rest of d-i > - Revert the grub.cfg generation code update (aka s/vga=/set > gfxpayload/g) > Checksums-Sha1: > df5ead877a0a88bec004d082f08aa83449386866 1288 win32-loader_0.6.16.dsc > 61735eb22f5e6a6f373830884d052fc80dd9dac2 170716 win32-loader_0.6.16.tar.gz > 38e02ec7401e45128e044ed0e55d0d22ce112d0c 733428 > win32-loader_0.6.16_all.deb > Checksums-Sha256: > 5cb83af48823dd997e19993c66dcff522b1a22ca743efc9aa33293fb2c5fa9d1 1288 > win32-loader_0.6.16.dsc > fd0e6536e041c975b460f303a48962390a6e948acb153462cc22ce5575d99001 170716 > win32-loader_0.6.16.tar.gz > e4d638eedb686a1252abfcf2fd5e6e78c6686cc83b0b5f99ad0a131f697f99b4 733428 > win32-loader_0.6.16_all.deb > Files: > 232b4d51fdd425d569c853b5e31e232c 1288 utils extra win32-loader_0.6.16.dsc > d1ba9c0e743d09c1406aaa9c175dec5e 170716 utils extra > win32-loader_0.6.16.tar.gz > 5fbcd8a7393ee65e819da124a1998b0b 733428 utils extra > win32-loader_0.6.16_all.deb > > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.4.10 (GNU/Linux) > > iJwEAQECAAYFAkvhoi4ACgkQ884eR6Y9JhRszwP+ND65/i4HocZ7NInCqN+pHKEA > pK6gjsIUMeEXYc39QMJjjA4jtORfj8mVnH0f0k33aDdpf1b7KK0kfnZb3CAVoG1l > tTRAPqNvi7jVptdkaziUq8f96c/7/EDHtJOg3u/vqq3Lx79reVjRIKDfev9NeaXf > /cSMRZWyC2JwJaNjG3U= > =0URh > -END PGP SIGNATURE- > > > > > -- Forwarded message -- > From: Tuco > To: sub...@bugs.debian.org > Date: Fri, 30 Apr 2010 13:02:54 +0200 > Subject: kfreebsd > Package: win32-loader > > Hi. Loading kfreebsd from win32-loader is simple, I made this patch based > on grub.cfg from the D-I images. > > Sorry I don't know about NSIS part, but if the files are in > C:/win32-loader/ it should boot. > > >
Bug#579742: kfreebsd
Package: win32-loader
Hi. Loading kfreebsd from win32-loader is simple, I made this patch based
on grub.cfg from the D-I images.
Sorry I don't know about NSIS part, but if the files are in C:/win32-loader/
it should boot.
Index: main.nsi
===
--- main.nsi (revision 63034)
+++ main.nsi (working copy)
@@ -653,8 +653,10 @@
$\n'
${Endif}
FileWrite $0 "\
-linux /win32-loader/linux $preseed_cmdline$\n\
-initrd /win32-loader/initrd.gz$\n\
+kfreebsd /win32-loader/kfreebsd.gz$\n\
+kfreebsd_module_elf /win32-loader/acpi.ko$\n\
+kfreebsd_module /win32-loader/mfsroot.gz type=mfs_root$\n\
+set kFreeBSD.vfs.root.mountfrom=ufs:/dev/md0$\n\
boot$\n"
FileClose $0
Bug#576085: silently dies when syslog isn't present
Package: lirc
Severity: important
daemon dies silently when syslog isn't present. it's very confusing
as I had to use strace to figure out:
connect(5, {sa_family=AF_FILE, path="/dev/log"...}, 110) = -1
ECONNREFUSED (Connection refused)
package should ensure there's a working syslog, or at least print a
useful error when there isn't.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#572791: please loosen libfuse-dev version requirement
Package: zfs-fuse Version: 2.7.4-1.1+lenny1 Severity: minor libfuse-dev in Build-Depends seems too strict. INSTALL file says it can work with FUSE 2.5.x or later. I tried with libfuse-dev 2.7.4-1.1+lenny1 and this produced a working package for lenny. -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
