Re: libc6 Compatibility
>>>>> On Wed, 15 Jun 2022 20:26:13 +0200 (CEST), Sysadmin Lists said:
>
> > --------
> > From: Martin Simmons
> > Sent: Wed Jun 15 19:04:48 CEST 2022
> > To: Sysadmin Lists
> > Cc: ,
> > Subject: Re: libc6 Compatibility
> >
> >
> > >>>>> On Tue, 14 Jun 2022 21:55:58 +0200 (CEST), Sysadmin Lists said:
> > >
> > > >
> > > > From: Dmitry Chagin
> > > > Sent: Tue Jun 14 09:24:37 CEST 2022
> > > > To: Sysadmin Lists
> > > > Cc: Freebsd Emulation
> > > > Subject: Re: libc6 Compatibility
> > > >
> > > >
> > > > On Tue, Jun 14, 2022 at 03:26:03AM +0200, Sysadmin Lists wrote:
> > > > > Does Linux compat not work with Ubuntu's newest libc6 update?
> > > > > I used to run Brave browser from it, but now I get this error message:
> > > > >
> > > > > $ /compat/ubuntu/opt/brave.com/brave/brave
> > > > > ELF interpreter /lib64/ld-linux-x86-64.so.2 not found, error 2
> > > > > Abort trap
> > > > >
> > > > > The shared object is loaded in memory:
> > > > > $ ldd /opt/brave.com/brave/brave | grep ld-
> > > > > /lib64/ld-linux-x86-64.so.2 (0x01021000)
> > > > >
> > > > > And it exists on the filesystem:
> > > > > $ find /compat/ubuntu/lib**/ -name ld-\* -exec ls -lh '{}' +
> > > > > -rwxr-xr-x 1 root wheel 187K Dec 16 2020
> > > > > /compat/ubuntu/lib/x86_64-linux-gnu/ld-2.31.so
> > > > > lrwxr-xr-x 1 root wheel 10B Dec 16 2020
> > > > > /compat/ubuntu/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2 -> ld-2.31.so
> > > > > lrwxr-xr-x 1 root root 34B Mar 23 2021
> > > > > /compat/ubuntu/lib64/ld-linux-x86-64.so.2 ->
> > > > > ../lib/x86_64-linux-gnu/ld-2.31.so
> > > > >
> > > > > This is the update that broke it:
> > > > > $ apt-get install libc6
> > > > > Get:1 http://archive.ubuntu.com/ubuntu focal-updates/main amd64 libc6
> > > > > amd64 2.31-0ubuntu9.9 [2722 kB]
> > > > >
> > > >
> > > > try ktrace it, aka ktrace -di /compat/ubuntu/opt/brave.com/brave/brave
> > > > then kdump -HAR -m 128 > brave.log
> > > >
> > > >
> > >
> > > Thanks, Dmitry. That helped me find the problem. Here's the full log:
> > >
> > > $ cat brave.log
> > > 8184 101222 ktrace 0.00 RET F64 ktrace 0
> > > 8184 101222 ktrace 0.14 CALL F64
> > > execve(0x7fffeda3,0x7fffeac0,0x7fffead0)
> > > 8184 101222 ktrace 0.03 NAMI F64
> > > "/compat/ubuntu/opt/brave.com/brave/brave"
> > > 8184 101222 ktrace 0.055726 NAMI F64
> > > "/compat/ubuntu/lib64/ld-linux-x86-64.so.2"
> > > 8184 101222 ktrace 0.32 NAMI F64 "/lib64/ld-linux-x86-64.so.2"
> > >
> > > And here's an updated `ls' printout after upgrading libc6:
> > > $ find /compat/ubuntu/lib**/ -name ld-\* -exec ls -lh '{}' +
> > > -rwxr-xr-x 1 root wheel 187K Apr 6 18:24
> > > /compat/ubuntu/lib/x86_64-linux-gnu/ld-2.31.so
> > > lrwxr-xr-x 1 root wheel 10B Apr 6 18:24
> > > /compat/ubuntu/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2 -> ld-2.31.so
> > > lrwxr-xr-x 1 root wheel 32B Apr 6 18:24
> > > /compat/ubuntu/lib64/ld-linux-x86-64.so.2 ->
> > > /lib/x86_64-linux-gnu/ld-2.31.so
> > >
> > > The new .so is missing 'previous directory' dots in the symlink path.
> >
> > It looks like the link with the dots was made by hand in 2021, presumably to
> > fix this.
>
> That might be the case. I remember having to make changes to the shared
> object
> before. The question then becomes, why only this update to libc6 and not
> previous updates?
Based on the file dates, I suspect there were no previous updates to libc6
installed since Dec 16 2020.
> > > I thought Linux compat was designed to search compat.linux.emul_path
> > > for shared libs first, but apparently it works differently than that.
> >
> > What does linux compat do in general with absolute symlinks? Does it check
> > the target with compat.linux.emul_path prepended?
>
> That's what the documentation say
Re: libc6 Compatibility
> On Tue, 14 Jun 2022 21:55:58 +0200 (CEST), Sysadmin Lists said:
>
> >
> > From: Dmitry Chagin
> > Sent: Tue Jun 14 09:24:37 CEST 2022
> > To: Sysadmin Lists
> > Cc: Freebsd Emulation
> > Subject: Re: libc6 Compatibility
> >
> >
> > On Tue, Jun 14, 2022 at 03:26:03AM +0200, Sysadmin Lists wrote:
> > > Does Linux compat not work with Ubuntu's newest libc6 update?
> > > I used to run Brave browser from it, but now I get this error message:
> > >
> > > $ /compat/ubuntu/opt/brave.com/brave/brave
> > > ELF interpreter /lib64/ld-linux-x86-64.so.2 not found, error 2
> > > Abort trap
> > >
> > > The shared object is loaded in memory:
> > > $ ldd /opt/brave.com/brave/brave | grep ld-
> > > /lib64/ld-linux-x86-64.so.2 (0x01021000)
> > >
> > > And it exists on the filesystem:
> > > $ find /compat/ubuntu/lib**/ -name ld-\* -exec ls -lh '{}' +
> > > -rwxr-xr-x 1 root wheel 187K Dec 16 2020
> > > /compat/ubuntu/lib/x86_64-linux-gnu/ld-2.31.so
> > > lrwxr-xr-x 1 root wheel 10B Dec 16 2020
> > > /compat/ubuntu/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2 -> ld-2.31.so
> > > lrwxr-xr-x 1 root root 34B Mar 23 2021
> > > /compat/ubuntu/lib64/ld-linux-x86-64.so.2 ->
> > > ../lib/x86_64-linux-gnu/ld-2.31.so
> > >
> > > This is the update that broke it:
> > > $ apt-get install libc6
> > > Get:1 http://archive.ubuntu.com/ubuntu focal-updates/main amd64 libc6
> > > amd64 2.31-0ubuntu9.9 [2722 kB]
> > >
> >
> > try ktrace it, aka ktrace -di /compat/ubuntu/opt/brave.com/brave/brave
> > then kdump -HAR -m 128 > brave.log
> >
> >
>
> Thanks, Dmitry. That helped me find the problem. Here's the full log:
>
> $ cat brave.log
> 8184 101222 ktrace 0.00 RET F64 ktrace 0
> 8184 101222 ktrace 0.14 CALL F64
> execve(0x7fffeda3,0x7fffeac0,0x7fffead0)
> 8184 101222 ktrace 0.03 NAMI F64
> "/compat/ubuntu/opt/brave.com/brave/brave"
> 8184 101222 ktrace 0.055726 NAMI F64
> "/compat/ubuntu/lib64/ld-linux-x86-64.so.2"
> 8184 101222 ktrace 0.32 NAMI F64 "/lib64/ld-linux-x86-64.so.2"
>
> And here's an updated `ls' printout after upgrading libc6:
> $ find /compat/ubuntu/lib**/ -name ld-\* -exec ls -lh '{}' +
> -rwxr-xr-x 1 root wheel 187K Apr 6 18:24
> /compat/ubuntu/lib/x86_64-linux-gnu/ld-2.31.so
> lrwxr-xr-x 1 root wheel 10B Apr 6 18:24
> /compat/ubuntu/lib/x86_64-linux-gnu/ld-linux-x86-64.so.2 -> ld-2.31.so
> lrwxr-xr-x 1 root wheel 32B Apr 6 18:24
> /compat/ubuntu/lib64/ld-linux-x86-64.so.2 -> /lib/x86_64-linux-gnu/ld-2.31.so
>
> The new .so is missing 'previous directory' dots in the symlink path.
It looks like the link with the dots was made by hand in 2021, presumably to
fix this.
> I thought Linux compat was designed to search compat.linux.emul_path
> for shared libs first, but apparently it works differently than that.
What does linux compat do in general with absolute symlinks? Does it check
the target with compat.linux.emul_path prepended?
__Martin
Re: Old bug in patch-src_VBox_Additions_freebsd_vboxvfs_vboxvfs__vfsops.c
Is the memcpy wrong too? It looks like fsinfo.serial is a single uint32_t (i.e. 32 bits), but mp->mnt_stat.f_fsid is a pair of int32_t (i.e. 64 bits), so copying this pair from fsinfo.serial is wrong. __Martin > On Mon, 22 Jun 2020 14:22:05 -0400, Mikhail T said: > > Gentlemen! > > An old bug in the patch is causing compiler-warnings, and leads to > erroneous behavior where pointers are bigger than 32-bit. > > Moreover, given the memcpy right after it, the bzero is simply not > needed at all. Instead of removing the bogus ampersand, the entire line > should be deleted. (I would've replaced the memcpy with an assignment > too, but that's not as pressing.) > > Can I commit this? > > Index: files/patch-src_VBox_Additions_freebsd_vboxvfs_vboxvfs__vfsops.c > === > --- files/patch-src_VBox_Additions_freebsd_vboxvfs_vboxvfs__vfsops.c > (revision 539883) > +++ files/patch-src_VBox_Additions_freebsd_vboxvfs_vboxvfs__vfsops.c > (working copy) > @@ -11,7 +11,7 @@ > * > * This file is part of VirtualBox Open Source Edition (OSE), as > * available from http://www.virtualbox.org. This file is free > software; > -@@ -14,245 +9,479 @@ > +@@ -14,245 +9,478 @@ > * VirtualBox OSE distribution. VirtualBox OSE is distributed in the > * hope that it will be useful, but WITHOUT ANY WARRANTY of any kind. > */ > @@ -466,7 +466,6 @@ > + > + MNT_ILOCK(mp); > + mp->mnt_data = vboxfsmp; > *-+ bzero(&mp->mnt_stat.f_fsid, sizeof(&mp->mnt_stat.f_fsid));* > + /* f_fsid is int32_t but serial is uint32_t, convert */ > + memcpy(&mp->mnt_stat.f_fsid, &fsinfo.serial, > sizeof(mp->mnt_stat.f_fsid)); > + mp->mnt_flag |= MNT_LOCAL; > > Thanks. Yours, > > -mi > > ___ > freebsd-emulation@freebsd.org mailing list > https://lists.freebsd.org/mailman/listinfo/freebsd-emulation > To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org" > ___ freebsd-emulation@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: virtualbox-ose-additions - build fails due to env mixup
> On Mon, 12 Nov 2018 09:53:17 +0100, Domagoj Smolčić said: > > virtualbox-ose-additions-5.2.22 and below > > If user which su-ed to root, has in it's .shrc file, inclusion of i.e; > . $HOME/_some_file.sh, during build, that file is included from /root/ > instead and build fails with: > .: Can't open /root/_some_file.sh > As /usr/local/bin/kmk reports (kBuild-0.1.9998_11) > > WORKAROUND: > Uncomment inclusion line > OR > Log in directly as a root I suggest using "su - root" to become root. __Martin ___ freebsd-emulation@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: FreeBSD Port: emulators/virtualbox-ose-nox11
You are probably getting packages from the quarterly snapshot rather than
latest.
You can switch to the latest packages by creating a file
/usr/local/etc/pkg/repos/FreeBSD.conf containing these three lines:
FreeBSD: {
url: "pkg+http://pkg.FreeBSD.org/${ABI}/latest";
}
Note that this affects all packages, not just vbox. See
https://forums.freebsd.org/threads/52843/ for a discussion.
__Martin
> On Tue, 27 Dec 2016 02:03:30 +0100, Pascal Thiel said:
>
> Hi,
>
> I just did a clean installation of FreeBSD 11 and wanted to install
> *virtualbox-ose-nox11* on it in order to run *phpvirtualbox*. However pkg
> only found *virtualbox-ose-nox11* under its old name *virtualbox-ose-lite*
> (which was built without the webservice enabled).
>
> Is there something you can do about that?
>
> Thanks in advance and have a nice day :)
>
>
> best regards
>
> Pascal
>
> PS: I found above mentioned ports here
>
Re: Call for Testing: CentOS c6_64 ports now with 32bit libraries.
>>>>> On Mon, 8 Feb 2016 21:26:55 +, Miguel C said: > > --089e011770cf92b957052b48ded9 > Content-Type: text/plain; charset=UTF-8 > > On Mon, Feb 8, 2016 at 6:19 PM, Martin Simmons wrote: > > > >>>>> On Mon, 8 Feb 2016 01:17:45 +, Miguel C said: > > > > > > On Sun, Jan 31, 2016 at 3:05 AM, Miguel C > > wrote: > > > > > > > > > > > > > > > On Sun, Jan 31, 2016 at 2:44 AM, Johannes Jost Meixner < > > > > johan...@meixner.dk> wrote: > > > > > > > >> -BEGIN PGP SIGNED MESSAGE- > > > >> Hash: SHA256 > > > >> > > > >> Hi Miguel, > > > >> > > > >> you'll have to svn up to the latest portstree, apply the patch, > > > >> update/reinstall the relevant packages (in this case, > > > >> devel/linux-c6-devtools), and try again. > > > >> > > > >> I'm actually using git, but I'm assuming master branch is in sync > > with svn > > > > > > > > FYI linux-skype_oss_wrapper still fails to build after the patch with: > > > > > > Stop. > > > make: stopped in > > > /usr/home/miguelc/gitstuff/freebsd-ports/audio/linux-skype_oss_wrapper > > > [miguelc@r2d2:~/gitstuff/freebsd-ports/audio/linux-skype_oss_wrapper ]% > > > sudo make MAKE_JOBS_UNSAFE=yes > > > ===> Building for linux-skype_oss_wrapper-0.1.1 > > > /compat/linux/usr/bin/gcc -m32 libpulse.o -o libpulse.so.0 > > > -fstack-protector -shared -lpthread > > > /usr/bin/ld: cannot find /lib/libpthread.so.0 > > > collect2: ld returned 1 exit status > > > *** Error code 1 > > > > > > I can see libpthread.so in /compat/linux/usr/lib > > > > > > find /compat/linux/ -name 'libpthread*' > > > /compat/linux/lib64/libpthread-2.12.so > > > /compat/linux/lib64/libpthread.so.0 > > > /compat/linux/usr/lib/libpthread.so.0 > > > /compat/linux/usr/lib/libpthread_nonshared.a > > > /compat/linux/usr/lib/libpthread.so > > > /compat/linux/usr/lib64/libpthread_nonshared.a > > > /compat/linux/usr/lib64/libpthread.so > > > > Just an idea: have you installed the 32-bit glibc? That should provide > > /compat/linux/lib/libpthread.so.0 etc. > > > > > Hum, I was under the impression that 32-bit libs would still be installed > alongside... wasn't that the purpose of the PATCH? Sorry, you are right, it should be installed by linux_base-c6. Nevertheless, you are somehow missing at least these files: /compat/linux/lib/libpthread-2.12.so /compat/linux/lib/libpthread.so.0 Do you have the two libc files below? /compat/linux/lib/libc-2.12.so /compat/linux/lib/libc.so.6 If not, then I suspect your linux_base-c6 doesn't contain the patch. __Martin ___ freebsd-emulation@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: Call for Testing: CentOS c6_64 ports now with 32bit libraries.
> On Mon, 8 Feb 2016 01:17:45 +, Miguel C said: > > On Sun, Jan 31, 2016 at 3:05 AM, Miguel C wrote: > > > > > > > On Sun, Jan 31, 2016 at 2:44 AM, Johannes Jost Meixner < > > johan...@meixner.dk> wrote: > > > >> -BEGIN PGP SIGNED MESSAGE- > >> Hash: SHA256 > >> > >> Hi Miguel, > >> > >> you'll have to svn up to the latest portstree, apply the patch, > >> update/reinstall the relevant packages (in this case, > >> devel/linux-c6-devtools), and try again. > >> > >> I'm actually using git, but I'm assuming master branch is in sync with svn > > > > FYI linux-skype_oss_wrapper still fails to build after the patch with: > > Stop. > make: stopped in > /usr/home/miguelc/gitstuff/freebsd-ports/audio/linux-skype_oss_wrapper > [miguelc@r2d2:~/gitstuff/freebsd-ports/audio/linux-skype_oss_wrapper ]% > sudo make MAKE_JOBS_UNSAFE=yes > ===> Building for linux-skype_oss_wrapper-0.1.1 > /compat/linux/usr/bin/gcc -m32 libpulse.o -o libpulse.so.0 > -fstack-protector -shared -lpthread > /usr/bin/ld: cannot find /lib/libpthread.so.0 > collect2: ld returned 1 exit status > *** Error code 1 > > I can see libpthread.so in /compat/linux/usr/lib > > find /compat/linux/ -name 'libpthread*' > /compat/linux/lib64/libpthread-2.12.so > /compat/linux/lib64/libpthread.so.0 > /compat/linux/usr/lib/libpthread.so.0 > /compat/linux/usr/lib/libpthread_nonshared.a > /compat/linux/usr/lib/libpthread.so > /compat/linux/usr/lib64/libpthread_nonshared.a > /compat/linux/usr/lib64/libpthread.so Just an idea: have you installed the 32-bit glibc? That should provide /compat/linux/lib/libpthread.so.0 etc. __Martin ___ freebsd-emulation@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: /compat/linux and /usr/compat/linux
> On Tue, 10 Feb 2015 18:12:43 -0800, Kevin Oberman said: > > On Tue, Feb 10, 2015 at 2:54 PM, Chris H wrote: > > > On Tue, 10 Feb 2015 12:34:19 -0800 Patrick Powell > > wrote > > > > > On 02/08/15 02:10, Gary Jennejohn wrote: > > > > On Fri, 06 Feb 2015 08:43:15 -0800 > > > > Patrick Powell wrote: > > > > > > > >> The latest linux-base appears to install into /compat, rather than > > > >> /usr/compat > > > >> > > > >> However, some FreeBSD FAQ and other documents appear to refer to > > > >> /usr/compat. > > > >> > > > >> Question: when did the move from using /usr/compat to /compat take > > > >> place (just out of curiosity)? > > > >> > > > >> Question: can linproc be in /compat/linux/proc or > > /usr/compat/linux/proc? > > > >> Question: if you have a /usr/compat/linux/lib directory will this be > > > >> added to the > > > >> libraries for linux emulation? > > > >> > > > > /compat has historically always been a symbolic link to /usr/compat: > > > > > > > > lrwxr-xr-x 1 root wheel 10 Nov 25 2011 /compat -> usr/compat > > > > > > > > linprocfs on /usr/compat/linux/proc (linprocfs, local) > > > > but in my /etc/fstab I have > > > > linproc /compat/linux/proc linprocfs rw0 > > 0 > > > > > > > > Since /compat is a symbolic link it resolves to /usr/compat when it's > > > > used. > > > > > > > # uname -a > > > FreeBSD astart2.astart.com 9.3-RELEASE-p9 FreeBSD 9.3-RELEASE-p9 #0: Tue > > > Jan 27 10:43:40 UTC 2015 > > > r...@amd64-builder.daemonology.net:/usr/obj/usr/src/sys/GENERIC amd64 > > > > > > # ls -l / |grep compat > > > > > > drwxr-xr-x 3 root wheel512 Dec 3 14:41 compat > > > > > > # ls -l /usr |grep compat > > > drwxr-xr-x 3 root wheel512 Nov 21 21:06 compat > > > > > > Note that neither /compat or /usr/compat are symbolic links. > > Interesting. On a RELENG_9 (9.3-STABLE); I only return > > /compat > > no linux, or compat available in /usr > > So what's the *real* story behind all this? > > > > > ls -ld /compat > lrwxr-xr-x 1 root wheel 11 Feb 8 09:18 /compat@ -> /usr/compat > > It is NOT created by the OS or any port, as far as I know. FWIW, the pre-9.0 installer (sysinstall) creates /usr/compat with /compat linked to it, but it bsdinstall doesn't. See http://lists.freebsd.org/pipermail/freebsd-emulation/2011-December/009339.html __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: FreeBSD/i386 guests using excessive host CPU
> On Wed, 12 Dec 2012 12:43:36 +1100, Peter Jeremy said: > > [Please copy me in any replies] > > My VPS provider will not offer 32-bit FreeBSD guests because they use > excessive (ESXi) host CPU time. I have done some experiments using > VirtualBox on my FreeBSD/amd64 8-stable host and see the same. > > An idle 32-bit FreeBSD 9 or 10 guest uses very roughly 25% more host > CPU than an equivalent 64-bit guest. An idle 32-bit FreeBSD 8 guest > uses very roughly 50% more host CPU than an equivalent 64-bit guest. > > This problem does not show up with Linux or Windows7 but occurs with > both VirtualBox on AMD and ESXi on Intel CPUs. > > Since the problem shows up with different virtualisation environments > and on different CPUs, this suggests this is a FreeBSD guest issue. > My suspicion is that the cost of handling an interrupt on FreeBSD/i386 > is higher than the cost on amd64 - and this is more obvious on 8.x > because it doesn't use dynamic ticks. > > Has anyone else seen this behaviour and does anyone have any suggestions > for improving the FreeBSD/i386 performance. I can't repeat that with idle FreeBSD 8.0 guests running in VirtualBox 4.1.18 on Linux kernel 2.6.35.14-106.fc14.x86_64. In fact, the 64-bit one uses 4.8% of the CPU whereas the 32-bit one uses 4.2% (according to ps "bsdtime" measured over 1000 seconds). What absolute percentages are you seeing? __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: /usr/ports/print/acroread9 & ldd shared lib errors (fwd)
> On Thu, 23 Aug 2012 14:29:49 +0200, Julian H Stacey said: > > Hi, > Hiroki Sato wrote: > > "Julian H. Stacey" wrote > > > > jh> Hi h...@freebsd.org, > > jh> Sorry, just realised I should have sent you a copy of this > > jh> > > jh> Forwarded from: "Julian Stacey" > > http://www.berklix.com/~jhs/ > > jh> --- Forwarded Message > > jh> > > jh> To: freebsd-emulation@freebsd.org > > jh> Subject: /usr/ports/print/acroread9 & ldd shared lib errors > > jh> From: "Julian H. Stacey" > > jh> Organization: http://berklix.com BSD Unix Linux Consultancy, Munich > > Germany > > jh> User-agent: EXMH on FreeBSD http://berklix.com/free/ > > jh> X-URL: http://www.berklix.com > > jh> In-reply-to: Your message "Wed, 22 Aug 2012 18:41:06 +0200." > > jh> <20120822164106.GE4301@localhost> > > jh> Date: Thu, 23 Aug 2012 00:35:43 +0200 > > jh> Sender: j...@fire.js.berklix.net > > jh> > > jh> Hi freebsd-emulation@ > > jh> I am trying to get current version of /usr/ports/print/acroread9 > > jh> to work on FreeBSD8.2-RELEASE src/ based systems, both i386 & amd64 > > jh> available here (also available: 8.3-REL & 9.0-REL amd64 if necessary). > > jh> > > jh> I am seeing errors from ldd, shown on my > > jh> > > http://berklix.com/~jhs/src/bsd/fixes/FreeBSD/doc/en_US.ISO8859-1/books/handbook/linuxemu-adobe.html > > jh> > > jh> & request help/ suggestions. > > jh> Once I have this working, I'd happily smarten the page & then send-p > > jh> to help others. > > jh> > > jh> I have eg libBIB.so but not sure where to toss it ? > > jh> I currently have 2 copies even, in compat & local. > > > > I cannot understand what your problem is. acroread9 did not run on > > your system? > > Right, It starts, but various errors, both to invoking command line > & also to pop up windows, please see > > > http://berklix.com/~jhs/src/bsd/fixes/FreeBSD/doc/en_US.ISO8859-1/books/handbook/linuxemu-adobe.html > > where I documented checking everything to see what was wrong, & found > ldd shows some libs are still unresolved per URL above: > i386 > libBIB.so => not found > amd64 > libBIB.so: cannot open shared object file > libs remained missing after > (cd /usr/local/Adobe/Reader9;tar cf - . )|(cd > /compat/linux/Adobe/Reader9 && tar xf - ) > ldconfig -R > /compat/linux/Adobe/Reader9/ENU/Adobe/Reader9/Reader/intellinux/lib > ldd > /usr/local/Adobe/Reader9/ENU/Adobe/Reader9/Reader/intellinux/bin/acroread > > Please see my web page at text lines: > because per quote from `man ldconfig`: > Filenames must conform to the lib > > I or you in port could cludge those numeric failures, (rename, relink) > but is it necessary to ? > or should the Linux ABI/emulator work round it ? > I presume it's common for Linux to have .so.[0-9].[0.9] & FreeBSD to use just > .so.[0-9] > So I guess FreeBSD might already have some standard mapping mechanism for > Linux lib names ? > > I dont know if acroread libs should be in /usr/local or /compat ? > > I don't know if I should be setting more ldconfig vars in rc.conf ? > > Currently in rc.conf I have > ldconfig_paths="/usr/lib/compat /usr/local/lib > /usr/local/lib/compat/pkg \ > /usr/local/lib/browser_plugins/gnash /usr/local/lib/gio/modules \ > /usr/local/Adobe/Reader9/ENU/Adobe/Reader9/Reader/intellinux/lib" > I have no other ldconfig vars set, so I'm inheriting /etc/defaults/rc.conf > Should I set other ldconfig vars ? > eg any of ldconfig_insecure ldconfig32_paths ldconfig_paths_aout > ldconfig_local_dirs ldconfig_local32_dirs > Should I be copying those libs somewhere else ? to a 32 directory maybe ? The libs in Reader9/Reader/intellinux/lib are private to acroread, so I'm not convinced that they should be put in the ldconfig cache. On native Linux, these libs are located without any special ldconfig guff because the acroread front-end script (Reader9/bin/acroread) points LD_LIBRARY_PATH to them. Maybe that script is malfunctioning under the emulator? You could try running it with sh -x. > After > ldconfig -R > /compat/linux/Adobe/Reader9/ENU/Adobe/Reader9/Reader/intellinux/lib > on amd64 > acroread > click File Open (& dont get as far as selecting any file) > I see > libfam.so.0: cannot open shared object file: No such file or directory > Failed to load module: /usr/lib/gio/modules/libgiofam.so > (acroread:92034): GLib-GIO-WARNING **: Could not initialize inotify > (acroread:92034): GLib-GIO-WARNING **: Could not initialize inotify > (acroread:92034): Gtk-WARNING **: Unable to find default local > directory monitor type > (acroread:92034): Gtk-WARNING **: Unable to find default local > directory monitor type > > On i386 just after clicking File > libfam.so.0: cannot open shared object file: No such file or directory > Failed to load module: /usr/lib/gio/modules/libgiofam.so > (acroread:
Re: VirtualBox-4.1.10 doesn't forward ports into the guests
> On Mon, 02 Apr 2012 12:00:58 -0700, Yuri said: > > I use these commands: > VBoxManage setextradata MyGuest > "VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/Protocol" TCP > VBoxManage setextradata MyGuest > "VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/GuestPort" 22 > VBoxManage setextradata MyGuest > "VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/HostPort" > > They worked the first time. Then I deleted that guest, created a new one > with he same name. Now they don't work on this guest: > $ ssh -p 127.0.0.1 > ssh: connect to host 127.0.0.1 port : Connection refused > > Even though: > $ VBoxManage getextradata MyGuest enumerate > Key: GUI/LastCloseAction, Value: shutdown > Key: GUI/LastGuestSizeHint, Value: 720,400 > Key: GUI/LastNormalWindowPosition, Value: 573,23,720,443 > Key: VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/GuestPort, Value: 22 > Key: VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/HostPort, Value: > Key: VBoxInternal/Devices/pcnet/0/LUN#0/Config/ssh/Protocol, Value: TCP How did you delete the guest? The extra data are stored in the MyGuest.vbox file, but possible also cached in the vbox daemons. __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: FreeBSD 9.0 on VirtualBox 4.1.8
> On Mon, 30 Jan 2012 19:07:25 -0500, Robert Simmons said: > > I encountered a problem installing FreeBSD 9.0-RELEASE on VirtualBox > 4.1.8, and I wanted to share my fix for it, so it can be added to the > FreeBSD wiki and Handbook if appropriate. > > Using the default settings of VirtualBox, the following error occurs > during boot from the iso image: > http://i.imgur.com/kO93X.png > > I'm using Kubuntu 11.10, and I installed VirtualBox from the official PPA: > deb http://download.virtualbox.org/virtualbox/debian oneiric contrib > > The FreeBSD ISO that I used was "FreeBSD-9.0-RELEASE-amd64-dvd1.iso" > > The following checkbox needs to be checked in settings to be able to > boot FreeBSD: > Settings > System > Motherboard > Enable IO APIC How did you create the VM? If you choose "FreeBSD (64-bit)" as the OS type (to match the FreeBSD platform "amd64") then it should check Enable IO APIC autoamatically. __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: Wine-fbsd64 updated to 1.3.36 (32bit Wine for 64bit FreeBSD)
> On Thu, 19 Jan 2012 20:45:13 +0100, =?ISO-8859-15?Q?Andr=E9 Hentschel?= > said: > > >> >>> > >> >>> To answer your question directly, I have put no thought or effort into > >> >>> wine64. > >> >> > >> >> There is still a need for the 32bit wine, as the wow64 build will ONLY > >> >> run 64bit windows apps, so for the majority of applications and games, > >> >> we'll still need the 32bit wine. > >> > > >> > Is that because wow64 is unstable in wine64? > >> > > >> > Wow64's reason for existence is to run 32-bit Windows binaries on 64-bit > >> > Windows/Wine and it works pretty well for most applications. > >> > > >> > __Martin > >> > ___ > >> > freebsd-emulation at freebsd.org mailing list > >> > http://lists.freebsd.org/mailman/listinfo/freebsd-emulation > >> > To unsubscribe, send any mail to "freebsd-emulation-unsubscribe at > >> > freebsd.org" > >> wow64 doesnt -exist- in wine64 unless you compile the 32bit wine > >> alongside it. > >> > >> You have to build it twice, to get two binaries "wine" and "wine64" and > >> if you build them "properly" what happens when you attempt to execute a > >> 64bit app with "wine" it's automatically relaunched with "wine64" But > >> the reverse is NOT true. > > > > OK, that's what I would call unstable. Eventually though, I would expect a > > packaged wine64 to include wow64, just like Windows does. > > That's not unstable, that's the way it needs to be done. Yes, sorry, I didn't mean to sound critical of Wine. > To clarify this confusing thread: > wine64 as standalone can run 64-bit windows apps (on a 64-bit processor of > course) > wine(32) can run 32-bit windows apps > a wow64 setup means wine64 and wine(32) side by side with some > exchange mechanisms to run 32-bit apps in the same wineprefix as 64-bit ones > > Linux distros handle packaging that differently: > fedora uses the i386 package as a dependency, arch builds both together I'm glad that it is just a packaging issue (the original topic of this thread). __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: Wine-fbsd64 updated to 1.3.36 (32bit Wine for 64bit FreeBSD)
>>>>> On Wed, 11 Jan 2012 11:47:40 -0600, Chuck Burns said: > > On 01/11/12 10:24, Martin Simmons wrote: > >>>>>> On Tue, 10 Jan 2012 17:20:04 -0600, Chuck Burns said: > >> > >> On 01/10/12 10:49, David Naylor wrote: > >>> I wasn't aware of WoW64 (and how it will make my packages redundent :->). > >>> I > >>> see there are two wiki pages: > >>>+ http://wiki.winehq.org/Wine64 > >>>+ http://wiki.winehq.org/Wine64ForPackagers > >>> > >>> The latter makes this statement "Wine 64bit works at the moment only on > >>> Linux." which would indicate that, at this stage, FreeBSD is precluded. > >>> That, > >>> however, is just my speculation. > >>> > >>> To answer your question directly, I have put no thought or effort into > >>> wine64. > >> > >> There is still a need for the 32bit wine, as the wow64 build will ONLY > >> run 64bit windows apps, so for the majority of applications and games, > >> we'll still need the 32bit wine. > > > > Is that because wow64 is unstable in wine64? > > > > Wow64's reason for existence is to run 32-bit Windows binaries on 64-bit > > Windows/Wine and it works pretty well for most applications. > > > > __Martin > > ___ > > freebsd-emulation@freebsd.org mailing list > > http://lists.freebsd.org/mailman/listinfo/freebsd-emulation > > To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org" > wow64 doesnt -exist- in wine64 unless you compile the 32bit wine > alongside it. > > You have to build it twice, to get two binaries "wine" and "wine64" and > if you build them "properly" what happens when you attempt to execute a > 64bit app with "wine" it's automatically relaunched with "wine64" But > the reverse is NOT true. OK, that's what I would call unstable. Eventually though, I would expect a packaged wine64 to include wow64, just like Windows does. __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: Wine-fbsd64 updated to 1.3.36 (32bit Wine for 64bit FreeBSD)
> On Tue, 10 Jan 2012 17:20:04 -0600, Chuck Burns said: > > On 01/10/12 10:49, David Naylor wrote: > > I wasn't aware of WoW64 (and how it will make my packages redundent :->). I > > see there are two wiki pages: > > + http://wiki.winehq.org/Wine64 > > + http://wiki.winehq.org/Wine64ForPackagers > > > > The latter makes this statement "Wine 64bit works at the moment only on > > Linux." which would indicate that, at this stage, FreeBSD is precluded. > > That, > > however, is just my speculation. > > > > To answer your question directly, I have put no thought or effort into > > wine64. > > There is still a need for the 32bit wine, as the wow64 build will ONLY > run 64bit windows apps, so for the majority of applications and games, > we'll still need the 32bit wine. Is that because wow64 is unstable in wine64? Wow64's reason for existence is to run 32-bit Windows binaries on 64-bit Windows/Wine and it works pretty well for most applications. __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: "headed" virtualbox + vnc
> On Thu, 01 Dec 2011 14:11:55 +0200, Andriy Gapon said: > > Is it possible to start a guess VM with GUI and also be able to access it via > VNC? > All google hits for freebsd+virtualbox+vnc are about VBoxHeadless. Probably not simultaneously, but you can close the GUI with the "Save the machine state" option and then bring it back to life as a headless VM (presumably with vnc). Use VBoxManage controlvm ... savestate to stop the headless VM and then start it again in GUI mode. __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: [ports-i...@freebsd.org: linux-f10-flashplugin-10.3r181.34 failed on i386 9]
> On Wed, 10 Aug 2011 23:20:04 +0300, Volodymyr Kostyrko said: > > PS: As far as I understand the main problem is about mirroring correct > player versions. Anyone (even me) can apply for a redistribution rights > and make those files available for download the way we want (the way > each version will have unique url). Wouldn't that be dangerous, because it would allow easy use of old versions that contain critical vulnerabilities? __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: wine64
> On Tue, 21 Jun 2011 14:09:41 -1000, Bruce Meier said: > > I downloaded wine-fbsd64-1.3.21,1.tbz from > http://www.mediafire.com/wine_fbsd64 and it > installed without a problem. However when I run it I get the following > error: > > "ELF interpreter /libexec/ld-elf.so.1 not found Abort". > > I installed on an amd64 using freebsd-8.2. > Can anyone help? You need to install FreeBSD's 32-bit runtime compatibility libraries (the lib32 dist). __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
Re: AW: AW: rc.d script for running multiple VirtualBox VMs from /etc/rc.conf
> On Tue, 12 Apr 2011 14:09:24 -0700, Tobias Oberstein said: > > In the meantime, I've implemented stop_cmd as synchronous (it does not > return until VM is stopped) and also restart_cmd. > > I somehow works .. but there seems to be a watchdog in place: I get > > "watchdog timeout 30s expired ... shutodown NOW" > > Is it possible to override the 30s watchdog? Maybe set rcshutdown_timeout (see man rc.conf)? __Martin ___ freebsd-emulation@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-emulation To unsubscribe, send any mail to "freebsd-emulation-unsubscr...@freebsd.org"
