From gary.jennejohn at freenet.de Sun Mar 1 09:57:14 2009 From: gary.jennejohn at freenet.de (Gary Jennejohn) Date: Sun Mar 1 09:57:20 2009 Subject: testing qemu svn r6636 on FreeBSD; future of qemu on FreeBSD... In-Reply-To: <20090228213620.GA64626@saturn.kn-bremen.de> References: <20090225201550.3d3032b5@ernst.jennejohn.org> <20090228171520.GA56888@saturn.kn-bremen.de> <20090228201650.064c0ee4@ernst.jennejohn.org> <20090228213620.GA64626@saturn.kn-bremen.de> Message-ID: <20090301185709.60194bce@ernst.jennejohn.org> On Sat, 28 Feb 2009 22:36:20 +0100 Juergen Lock wrote: > On Sat, Feb 28, 2009 at 08:16:50PM +0100, Gary Jennejohn wrote: {snip old stuff] > > > It is possible that you got hit by the kqemu tsc vs smp problem, i.e. > > > passing `notsc' to the guest kernel or forcing qemu onto one cpu > > > (cpuset -l 0 qemu ...) may have helped there, sorry I should have thought > > > of that earlier... > > > > > > > I just tried these suggestions (with the standard qemu-devel) and they > > don't seem to do any harm :-P > > > ..but they didn't help either? (i.e. w/o -no-acpi) > Yes it did. It boots without -no-apci and with notpc just fine. > > > > Now I'm thinking about giving the patched qemu-devel another try, since > > I've found some workarounds for booting. I'll see whether I can at least > > boot into the installed openSUSE using it. > OK, the patched qemu-devel can boot the already installed openSUSE. I tried it with and without kernel-kqemu and both worked as long as I booted with notpc. Can't explain why the installation failed, though. --- Gary Jennejohn From nox at jelal.kn-bremen.de Sun Mar 1 15:24:50 2009 From: nox at jelal.kn-bremen.de (Juergen Lock) Date: Sun Mar 1 15:25:03 2009 Subject: testing qemu svn r6636 on FreeBSD; future of qemu on FreeBSD... In-Reply-To: <20090301185709.60194bce@ernst.jennejohn.org> References: <20090225201550.3d3032b5@ernst.jennejohn.org> <20090228171520.GA56888@saturn.kn-bremen.de> <20090228201650.064c0ee4@ernst.jennejohn.org> <20090228213620.GA64626@saturn.kn-bremen.de> <20090301185709.60194bce@ernst.jennejohn.org> Message-ID: <20090301231316.GA95918@saturn.kn-bremen.de> On Sun, Mar 01, 2009 at 06:57:09PM +0100, Gary Jennejohn wrote: > On Sat, 28 Feb 2009 22:36:20 +0100 > Juergen Lock wrote: > > > On Sat, Feb 28, 2009 at 08:16:50PM +0100, Gary Jennejohn wrote: > {snip old stuff] > > > > It is possible that you got hit by the kqemu tsc vs smp problem, i.e. > > > > passing `notsc' to the guest kernel or forcing qemu onto one cpu > > > > (cpuset -l 0 qemu ...) may have helped there, sorry I should have thought > > > > of that earlier... > > > > > > > > > > I just tried these suggestions (with the standard qemu-devel) and they > > > don't seem to do any harm :-P > > > > > ..but they didn't help either? (i.e. w/o -no-acpi) > > > > Yes it did. It boots without -no-apci and with notpc just fine. > You mean notsc, right...? > > > > > > Now I'm thinking about giving the patched qemu-devel another try, since > > > I've found some workarounds for booting. I'll see whether I can at least > > > boot into the installed openSUSE using it. > > > > OK, the patched qemu-devel can boot the already installed openSUSE. > > I tried it with and without kernel-kqemu and both worked as long as > I booted with notpc. > ..also notsc here? :) > Can't explain why the installation failed, though. OK, thanx, Juergen From gary.jennejohn at freenet.de Sun Mar 1 16:32:19 2009 From: gary.jennejohn at freenet.de (Gary Jennejohn) Date: Sun Mar 1 16:32:31 2009 Subject: testing qemu svn r6636 on FreeBSD; future of qemu on FreeBSD... In-Reply-To: <20090301231316.GA95918@saturn.kn-bremen.de> References: <20090225201550.3d3032b5@ernst.jennejohn.org> <20090228171520.GA56888@saturn.kn-bremen.de> <20090228201650.064c0ee4@ernst.jennejohn.org> <20090228213620.GA64626@saturn.kn-bremen.de> <20090301185709.60194bce@ernst.jennejohn.org> <20090301231316.GA95918@saturn.kn-bremen.de> Message-ID: <20090302013214.51d73098@ernst.jennejohn.org> On Mon, 2 Mar 2009 00:13:16 +0100 Juergen Lock wrote: > On Sun, Mar 01, 2009 at 06:57:09PM +0100, Gary Jennejohn wrote: > > Yes it did. It boots without -no-apci and with notpc just fine. > > > You mean notsc, right...? > > > > > > > > > Now I'm thinking about giving the patched qemu-devel another try, since > > > > I've found some workarounds for booting. I'll see whether I can at least > > > > boot into the installed openSUSE using it. > > > > > > > OK, the patched qemu-devel can boot the already installed openSUSE. > > > > I tried it with and without kernel-kqemu and both worked as long as > > I booted with notpc. > > > ..also notsc here? :) > Yes, I meant notsc everywhere :-P --- Gary Jennejohn From bsam at ipt.ru Sun Mar 1 23:17:26 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Sun Mar 1 23:17:33 2009 Subject: Linux gtk2.10+ In-Reply-To: <49A838D3.2010408@gmail.com> (Eitan Adler's message of "Fri\, 27 Feb 2009 14\:02\:43 -0500") References: <49A838D3.2010408@gmail.com> Message-ID: <68523380@bb.ipt.ru> Eitan Adler writes: > This is a follow up to a previous message I sent to this list about > updating linux-gtk to a more recent version of GTK. > At the time the response was "Boris has a patch, wait for the ports to > melt (unfreeze)". Any updates? Sorry for a delay. I'm back from a very hard business trip but should settle down my time in two days. Then I'll go further with linux ports. WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From bugmaster at FreeBSD.org Mon Mar 2 03:07:31 2009 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Mar 2 03:10:53 2009 Subject: Current problem reports assigned to freebsd-emulation@FreeBSD.org Message-ID: <200903021106.n22B6oYT057259@freefall.freebsd.org> Note: to view an individual PR, use: http://www.freebsd.org/cgi/query-pr.cgi?pr=(number). The following is a listing of current problems submitted by FreeBSD users. These represent problem reports covering all versions including experimental development code and obsolete releases. S Tracker Resp. Description -------------------------------------------------------------------------------- p kern/131099 emulation [linux] [patch] readdir broken on Linux emulation. o kern/130724 emulation [linprocfs] [patch] cpuinfo in linprocfs is dated, cau o kern/129169 emulation [linux] [patch] Linux Emulation ENOTCONN error using n f ports/127018 emulation Linuxulator incapable of using FreeBSD's LDAP environm o kern/126232 emulation [linux] Linux ioctl TCGETS (0x5401) always fails o ports/121800 emulation x11-toolkits/linux-openmotif - OpenMotif upgrade to 2. o kern/97326 emulation [linux] file descriptor leakage in linux emulation o ports/91318 emulation [fix] graphics/linux_dri: works on amd64 too o kern/91293 emulation [svr4] [patch] *Experimental* Update to the SVR4 emula o kern/73777 emulation [linux] [patch] linux emulation: root dir special hand a kern/72920 emulation [linux]: path "prefixing" is not done on unix domain s o kern/41543 emulation [patch] [request] easier wine/w23 support o kern/39201 emulation [linux] [patch] ptrace(2) and rfork(RFLINUXTHPN) confu o kern/29698 emulation [linux] [patch] linux ipcs doesn'work o kern/21463 emulation [linux] Linux compatability mode should not allow setu o kern/11165 emulation [ibcs2] IBCS2 doesn't work correctly with PID_MAX 9999 16 problems total. From novembre at gmail.com Mon Mar 2 14:17:50 2009 From: novembre at gmail.com (Novembre) Date: Mon Mar 2 14:17:58 2009 Subject: Flash 9 crash problems Message-ID: <3b47caa90903021351g4ceb832ap866cccd940cfb68b@mail.gmail.com> I have upgraded my 7.0-RELEASE system to RELENG_7 as of 02/19/2009. I had Firefox 2.0.0.12 installed before the upgrade. Then, I installed Adobe Flash 9 following the instructions here http://docs.freebsd.org/cgi/getmsg.cgi?fetch=51075+0+archive/2008/freebsd-emulation/20081102.freebsd-emulation So I have linux_enable="YES" in /etc/rc.conf, compat.linux.osrelease=2.6.16 in /etc/sysctl.conf, and OVERRIDE_LINUX_BASE_PORT=f8 in /etc/make.conf, and the following line in /etc/fstab: linproc /compat/linux/proc linprocfs rw 0 0 I then installed www/nspluginwrapper and www/linux-flashplugin9. The plugins showed up in Firefox when I type about:pugins as ---------- Shockwave Flash File name: npwrapper.libflashplayer.so Shockwave Flash 9.0 r152 MIME Type Description Suffixes Enabled application/x-shockwave-flash Shockwave Flash swf Yes application/futuresplash FutureSplash Player spl Yes ---------- After that, I can see flash videos on youtube and other websites. The problem is that when I close the tabs containing a flash video (most of the times when it is playing, but sometimes even when it is paused), npviewer.bin crashes with a core dump (npviewer.bin.core) in my home directory and it leaves a lot of zombie npviewer.bin processes running. Note that Firefox does not crash and is stable. At this stage, if I want to view another video, a gray box appears and nothing plays. If I kill all the zombie processes, the videos load and play just fine. Another minor problem is that if I view the videos in full screen mode, it appears that some frames are lost, and the video is not smooth. This is the result of 'uname -a' FreeBSD rsx4.physics.uiuc.edu 7.1-STABLE FreeBSD 7.1-STABLE #0: Thu Feb 19 18:46:55 CST 2009 soheil@rsx4.physics.uiuc.edu:/usr/obj/usr/src/sys/RSX4CUSTOM i386 Does anybody know what I should do to fix this? Thanks a lot :) From Alexander at Leidinger.net Mon Mar 2 23:34:42 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Mon Mar 2 23:34:49 2009 Subject: Flash 9 crash problems In-Reply-To: <3b47caa90903021351g4ceb832ap866cccd940cfb68b@mail.gmail.com> References: <3b47caa90903021351g4ceb832ap866cccd940cfb68b@mail.gmail.com> Message-ID: <20090303083430.64927hdcfxdcmjok@webmail.leidinger.net> Quoting Novembre (from Mon, 2 Mar 2009 15:51:59 -0600): > Does anybody know what I should do to fix this? Yes, wait (or apply some patch if you feel confident enough). The person which has a fix for this got commit access recently (not only for this fix). He is in the process of going to commit his changes to -current. There are patches floating around for the problem you have (look in the archive of this mailinglist and search for "futex"), but I don't know if they apply to RELENG_7. FYI: 2.6.16 is not officially supported on RELENG_7, as it does not contain everything what is in -current and makes up 2.6.16 emulation. The problem you see with Flash can be solved in RELENG_7 too (AFAIK), but some other things can not as they involve changes which we can not do in RELENG_7. Feel free to report problems, but don't expect that everything can be solved. Bye, Alexander. -- Toilet Toupee, n: Any shag carpet that causes the lid to become top-heavy, thus creating endless annoyance to male users. -- Rich Hall, "Sniglets" http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From bsam at ipt.ru Tue Mar 3 07:54:55 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Tue Mar 3 07:55:03 2009 Subject: [patch] Mk/bsd.linux-rpm.mk: introduce some linux-f8 infrastructure Message-ID: <14905754@bb.ipt.ru> Hello List, I'm going to commit an attached patch. This patch is a noop for current ports structure (no need to ask an approval from portmgr) but is needed for new linux-f8 ports. There are two points (both taken from emulators/linux_base-f8): . define LINUX_OSRELEASE variable (linux-f8 packages should be build with compat.linux.osrelease=2.6.16); . define MASTER_SITE_SUBDIR and MASTER_SITE_SRC_SUBDIR when LINUX_DIST_VER=8 (no such ports ATM). Opinions are welcome. Thanks. -------------- next part -------------- A non-text attachment was scrubbed... Name: Mk.diff Type: text/x-patch Size: 1872 bytes Desc: a patch for bsd.linux-rpm.mk Url : http://lists.freebsd.org/pipermail/freebsd-emulation/attachments/20090303/bab4e3de/Mk.bin -------------- next part -------------- WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From bsam at ipt.ru Tue Mar 3 08:43:56 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Tue Mar 3 08:44:04 2009 Subject: [patch] Mk/bsd.port.mk introduce bsd.linux-apps.mk Message-ID: <93462813@bb.ipt.ru> Hello List, here is a patch i'm going to ask an approval of portmgr to commit. Imho there is no need to do an experimental run with because this patch only introduces a new bsd.linux-apps.mk to bsd.port.mk. Currently no ports are using this structure and nothing should changed. The new bsd.linux-apps.mk is much like bsd.gnome.mk and give an infrastructure for non linux distribution ports to use different linux distributions. Example: --- acroread8 Makefile now --- RUN_DEPENDS= acroreadwrapper>=0.0.20080906:${PORTSDIR}/print/acroreadwrapper \ ${LINUXBASE}/usr/lib/libgtk-x11-2.0.so.0:${PORTSDIR}/x11-toolkits/linux-gtk2 \ ${LINUXBASE}/usr/lib/libpango-1.0.so.0:${PORTSDIR}/x11-toolkits/linux-pango \ ${LOCALBASE}/lib/linux-nvu/libgtkembedmoz.so:${PORTSDIR}/www/linux-nvu \ ${LINUXBASE}/usr/lib/gtk-2.0/immodules/im-scim.so:${PORTSDIR}/textproc/linux-scim-gtk ----- --- acroread8 Makefile in the future --- RUN_DEPENDS= acroreadwrapper>=0.0.20080906:${PORTSDIR}/print/acroreadwrapper USE_LINUX_APPS= gtk2 pango scimlibs webauth ----- Then acroread8 may be used both with linux_base-fc4 linux_base-f8 and their infrastructure ports. Opinions are welcome. Thanks: -------------- next part -------------- A non-text attachment was scrubbed... Name: Mk.diff.apps Type: text/x-patch Size: 14897 bytes Desc: introduce bsd.linux-apps.mk Url : http://lists.freebsd.org/pipermail/freebsd-emulation/attachments/20090303/cb9b268f/Mk.diff.bin -------------- next part -------------- WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From novembre at gmail.com Tue Mar 3 18:41:04 2009 From: novembre at gmail.com (Novembre) Date: Tue Mar 3 18:41:10 2009 Subject: Flash 9 crash problems In-Reply-To: <20090303083430.64927hdcfxdcmjok@webmail.leidinger.net> References: <3b47caa90903021351g4ceb832ap866cccd940cfb68b@mail.gmail.com> <20090303083430.64927hdcfxdcmjok@webmail.leidinger.net> Message-ID: <3b47caa90903031841m75eb3553jdd38cc47972990cc@mail.gmail.com> On Tue, Mar 3, 2009 at 1:34 AM, Alexander Leidinger wrote: > Quoting Novembre (from Mon, 2 Mar 2009 15:51:59 > -0600): > > Does anybody know what I should do to fix this? >> > > Yes, wait (or apply some patch if you feel confident enough). > > The person which has a fix for this got commit access recently (not only > for this fix). He is in the process of going to commit his changes to > -current. There are patches floating around for the problem you have (look > in the archive of this mailinglist and search for "futex"), but I don't know > if they apply to RELENG_7. > > FYI: 2.6.16 is not officially supported on RELENG_7, as it does not contain > everything what is in -current and makes up 2.6.16 emulation. The problem > you see with Flash can be solved in RELENG_7 too (AFAIK), but some other > things can not as they involve changes which we can not do in RELENG_7. Feel > free to report problems, but don't expect that everything can be solved. > > Bye, > Alexander. > > -- > Toilet Toupee, n: > Any shag carpet that causes the lid to become top-heavy, thus > creating endless annoyance to male users. > -- Rich Hall, "Sniglets" > > http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 > http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 > I found two posts containing patches regarding this issue: 1) http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2008/freebsd-emulation/20081228.freebsd-emulation 2) http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2009/freebsd-emulation/20090215.freebsd-emulation The first post contains two patches, of which one is marked specifically for 7-STABLE (as of Dec 21, 2008 when the patches were made available). The second post also has a patch, but I don't know whether it is for 8-CURRENT or 7-STABLE. Which one should I apply? Thanks :) From dchagin at freebsd.org Tue Mar 3 20:57:27 2009 From: dchagin at freebsd.org (Chagin Dmitry) Date: Tue Mar 3 20:57:58 2009 Subject: Flash 9 crash problems In-Reply-To: <3b47caa90903031841m75eb3553jdd38cc47972990cc@mail.gmail.com> References: <3b47caa90903021351g4ceb832ap866cccd940cfb68b@mail.gmail.com> <20090303083430.64927hdcfxdcmjok@webmail.leidinger.net> <3b47caa90903031841m75eb3553jdd38cc47972990cc@mail.gmail.com> Message-ID: <20090304045718.GA2274@dchagin.static.corbina.ru> On Tue, Mar 03, 2009 at 08:41:03PM -0600, Novembre wrote: > On Tue, Mar 3, 2009 at 1:34 AM, Alexander Leidinger > wrote: > > > Quoting Novembre (from Mon, 2 Mar 2009 15:51:59 > > -0600): > > > > Does anybody know what I should do to fix this? > >> > > > > Yes, wait (or apply some patch if you feel confident enough). > > > > The person which has a fix for this got commit access recently (not only > > for this fix). He is in the process of going to commit his changes to > > -current. There are patches floating around for the problem you have (look > > in the archive of this mailinglist and search for "futex"), but I don't know > > if they apply to RELENG_7. > > > > FYI: 2.6.16 is not officially supported on RELENG_7, as it does not contain > > everything what is in -current and makes up 2.6.16 emulation. The problem > > you see with Flash can be solved in RELENG_7 too (AFAIK), but some other > > things can not as they involve changes which we can not do in RELENG_7. Feel > > free to report problems, but don't expect that everything can be solved. > > > > Bye, > > Alexander. > > > > -- > > Toilet Toupee, n: > > Any shag carpet that causes the lid to become top-heavy, thus > > creating endless annoyance to male users. > > -- Rich Hall, "Sniglets" > > > > http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 > > http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 > > > > > I found two posts containing patches regarding this issue: > 1) > http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2008/freebsd-emulation/20081228.freebsd-emulation > 2) > http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2009/freebsd-emulation/20090215.freebsd-emulation > > The first post contains two patches, of which one is marked specifically for > 7-STABLE (as of Dec 21, 2008 when the patches were made available). The > second post also has a patch, but I don't know whether it is for 8-CURRENT > or 7-STABLE. Which one should I apply? > hi, we start commiting, please wait a couple of weeks... -- Have fun! chd -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 196 bytes Desc: not available Url : http://lists.freebsd.org/pipermail/freebsd-emulation/attachments/20090304/f47b839c/attachment.pgp From Alexander at Leidinger.net Tue Mar 3 23:43:24 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Tue Mar 3 23:43:32 2009 Subject: Flash 9 crash problems In-Reply-To: <3b47caa90903031841m75eb3553jdd38cc47972990cc@mail.gmail.com> References: <3b47caa90903021351g4ceb832ap866cccd940cfb68b@mail.gmail.com> <20090303083430.64927hdcfxdcmjok@webmail.leidinger.net> <3b47caa90903031841m75eb3553jdd38cc47972990cc@mail.gmail.com> Message-ID: <20090304084315.587314cev8vm3f34@webmail.leidinger.net> Quoting Novembre (from Tue, 3 Mar 2009 20:41:03 -0600): > I found two posts containing patches regarding this issue: > 1) > http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2008/freebsd-emulation/20081228.freebsd-emulation > 2) > http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2009/freebsd-emulation/20090215.freebsd-emulation > > The first post contains two patches, of which one is marked specifically for > 7-STABLE (as of Dec 21, 2008 when the patches were made available). The > second post also has a patch, but I don't know whether it is for 8-CURRENT > or 7-STABLE. Which one should I apply? If you do not want to wait: the patch utility has the option "-C" which just tries to apply the patch but does not really apply it. This way you can see if a patch is applicable to what you have or not. This way you can test if one of all those patches applies at all to what you have. In the worst case it applies, but you get a kernel panic upon use of some linux stuff (you may lose data because of this). In the best case at least one applies and it works perfectly. Bye, Alexander. -- If we see the light at the end of the tunnel It's the light of an oncoming train. -- Robert Lowell http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From Alexander at Leidinger.net Wed Mar 4 00:31:49 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Wed Mar 4 00:31:55 2009 Subject: [patch] Mk/bsd.linux-rpm.mk: introduce some linux-f8 infrastructure In-Reply-To: <14905754@bb.ipt.ru> References: <14905754@bb.ipt.ru> Message-ID: <20090304093137.10733oaty524f0is@webmail.leidinger.net> Quoting Boris Samorodov (from Tue, 03 Mar 2009 18:55:01 +0300): > I'm going to commit an attached patch. This patch is a noop for > current ports structure (no need to ask an approval from portmgr) > but is needed for new linux-f8 ports. While I agree with you... does portmgr agree with you too? Sometimes they have different opinions than committers about what should be tested in an exp-build. > There are two points (both taken from emulators/linux_base-f8): > . define LINUX_OSRELEASE variable (linux-f8 packages should be > build with compat.linux.osrelease=2.6.16); Why do you use ECHO_CMD? Is there some whitespace which needs to be deleted? > . define MASTER_SITE_SUBDIR and MASTER_SITE_SRC_SUBDIR when > LINUX_DIST_VER=8 (no such ports ATM). Why is there ifndef MASTER_SITES (only in the context of your patch, so not really related)? Is there a difference for "?=" between an empty MASTER_SITES and an undefined one? Bye, Alexander. -- And this is good old Boston, The home of the bean and the cod, Where the Lowells talk only to Cabots, And the Cabots talk only to God. http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From Alexander at Leidinger.net Wed Mar 4 00:40:19 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Wed Mar 4 00:40:30 2009 Subject: [patch] Mk/bsd.port.mk introduce bsd.linux-apps.mk In-Reply-To: <93462813@bb.ipt.ru> References: <93462813@bb.ipt.ru> Message-ID: <20090304094008.16846rstcvof4ngo@webmail.leidinger.net> Quoting Boris Samorodov (from Tue, 03 Mar 2009 19:44:02 +0300): > The new bsd.linux-apps.mk is much like bsd.gnome.mk and give an > infrastructure for non linux distribution ports to use different > linux distributions. Example: Yeah! Great work! Bye, Alexander. -- << WAIT >> http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From bsam at ipt.ru Wed Mar 4 01:23:37 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Wed Mar 4 01:23:45 2009 Subject: [patch] Mk/bsd.linux-rpm.mk: introduce some linux-f8 infrastructure In-Reply-To: <20090304093137.10733oaty524f0is@webmail.leidinger.net> (Alexander Leidinger's message of "Wed\, 04 Mar 2009 09\:31\:37 +0100") References: <14905754@bb.ipt.ru> <20090304093137.10733oaty524f0is@webmail.leidinger.net> Message-ID: <11370520@bb.ipt.ru> Alexander Leidinger writes: > Quoting Boris Samorodov (from Tue, 03 Mar 2009 18:55:01 +0300): > >> I'm going to commit an attached patch. This patch is a noop for >> current ports structure (no need to ask an approval from portmgr) >> but is needed for new linux-f8 ports. > > While I agree with you... does portmgr agree with you too? Sometimes > they have different opinions than committers about what should be > tested in an exp-build. Agreed, I'll contact portmgr with this patch too. >> There are two points (both taken from emulators/linux_base-f8): >> . define LINUX_OSRELEASE variable (linux-f8 packages should be >> build with compat.linux.osrelease=2.6.16); > > Why do you use ECHO_CMD? Is there some whitespace which needs to be deleted? Can't recall for sure but I've got some unwanted results while doing some experiments (was it an output to stderr if no linuxulator kldloaded?). >> . define MASTER_SITE_SUBDIR and MASTER_SITE_SRC_SUBDIR when >> LINUX_DIST_VER=8 (no such ports ATM). > > Why is there ifndef MASTER_SITES (only in the context of your patch, This ifndef was introduced by sat@ at Rev 1.9. I'm not aware of the real need. > so not really related)? Is there a difference for "?=" between an > empty MASTER_SITES and an undefined one? Yea, after ifndef introduction "?=" lost it's purpose. Alexander, I fully agree with your concerns, I'm aware of them. But for now I try to change as little as it is possible so to face less problems. Since new infrastructure and linux-f8 port are introduced work should go further. ;-) Thanks for your feedback. WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From Alexander at Leidinger.net Wed Mar 4 01:32:57 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Wed Mar 4 01:33:04 2009 Subject: [patch] Mk/bsd.linux-rpm.mk: introduce some linux-f8 infrastructure In-Reply-To: <11370520@bb.ipt.ru> References: <14905754@bb.ipt.ru> <20090304093137.10733oaty524f0is@webmail.leidinger.net> <11370520@bb.ipt.ru> Message-ID: <20090304103247.417765zrsapwwow0@webmail.leidinger.net> Quoting Boris Samorodov (from Wed, 04 Mar 2009 12:23:35 +0300): > Alexander Leidinger writes: >> Quoting Boris Samorodov (from Tue, 03 Mar 2009 >>> . define MASTER_SITE_SUBDIR and MASTER_SITE_SRC_SUBDIR when >>> LINUX_DIST_VER=8 (no such ports ATM). >> >> Why is there ifndef MASTER_SITES (only in the context of your patch, > > This ifndef was introduced by sat@ at Rev 1.9. I'm not aware of the > real need. I had a look at it. It is about protecting MASTER_SITE_* to be set, e.g. imagine if someone sets MASTER_SITE=file:///..., and at this location there's a flat file structure. In this case you do not want to have MASTER_SITE_* to be set. So I think you need to change your patch to keep this functionality. Bye, Alexander. -- An elephant is a mouse with an operating system. http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From bsam at ipt.ru Wed Mar 4 02:01:30 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Wed Mar 4 02:01:37 2009 Subject: [patch] Mk/bsd.linux-rpm.mk: introduce some linux-f8 infrastructure In-Reply-To: <20090304103247.417765zrsapwwow0@webmail.leidinger.net> (Alexander Leidinger's message of "Wed\, 04 Mar 2009 10\:32\:47 +0100") References: <14905754@bb.ipt.ru> <20090304093137.10733oaty524f0is@webmail.leidinger.net> <11370520@bb.ipt.ru> <20090304103247.417765zrsapwwow0@webmail.leidinger.net> Message-ID: <79218247@bb.ipt.ru> Alexander Leidinger writes: > Quoting Boris Samorodov (from Wed, 04 Mar 2009 12:23:35 +0300): >> Alexander Leidinger writes: >>> Quoting Boris Samorodov (from Tue, 03 Mar 2009 > >>>> . define MASTER_SITE_SUBDIR and MASTER_SITE_SRC_SUBDIR when >>>> LINUX_DIST_VER=8 (no such ports ATM). >>> >>> Why is there ifndef MASTER_SITES (only in the context of your patch, >> >> This ifndef was introduced by sat@ at Rev 1.9. I'm not aware of the >> real need. > > I had a look at it. It is about protecting MASTER_SITE_* to be set, > e.g. imagine if someone sets MASTER_SITE=file:///..., and at this > location there's a flat file structure. In this case you do not want > to have MASTER_SITE_* to be set. > > So I think you need to change your patch to keep this functionality. Yes, good catch, thanks! The first .endif was misplaced. Seems that the following patch is what you meant: -------------- next part -------------- A non-text attachment was scrubbed... Name: Mk.diff Type: text/x-patch Size: 1811 bytes Desc: a patch for bsd.linux-rpm.mk Url : http://lists.freebsd.org/pipermail/freebsd-emulation/attachments/20090304/8d5f24aa/Mk.bin -------------- next part -------------- WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From Alexander at Leidinger.net Wed Mar 4 02:25:35 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Wed Mar 4 02:25:41 2009 Subject: [patch] Mk/bsd.linux-rpm.mk: introduce some linux-f8 infrastructure In-Reply-To: <79218247@bb.ipt.ru> References: <14905754@bb.ipt.ru> <20090304093137.10733oaty524f0is@webmail.leidinger.net> <11370520@bb.ipt.ru> <20090304103247.417765zrsapwwow0@webmail.leidinger.net> <79218247@bb.ipt.ru> Message-ID: <20090304112524.45016b7uvodhvvok@webmail.leidinger.net> Quoting Boris Samorodov (from Wed, 04 Mar 2009 13:01:28 +0300): > Alexander Leidinger writes: >> Quoting Boris Samorodov (from Wed, 04 Mar 2009 >> 12:23:35 +0300): >>> Alexander Leidinger writes: >>>> Quoting Boris Samorodov (from Tue, 03 Mar 2009 >> >>>>> . define MASTER_SITE_SUBDIR and MASTER_SITE_SRC_SUBDIR when >>>>> LINUX_DIST_VER=8 (no such ports ATM). >>>> >>>> Why is there ifndef MASTER_SITES (only in the context of your patch, >>> >>> This ifndef was introduced by sat@ at Rev 1.9. I'm not aware of the >>> real need. >> >> I had a look at it. It is about protecting MASTER_SITE_* to be set, >> e.g. imagine if someone sets MASTER_SITE=file:///..., and at this >> location there's a flat file structure. In this case you do not want >> to have MASTER_SITE_* to be set. >> >> So I think you need to change your patch to keep this functionality. > > Yes, good catch, thanks! The first .endif was misplaced. Seems that > the following patch is what you meant: Yes, looks commit ready. Personally I would remove the '?' from MASTER_SITE?=, as it is handled by the ifndef. This way it does not confuse readers. MAybe even putting a comment there what this is all about. Bye, Alexander. -- Grelb's Reminder: Eighty percent of all people consider themselves to be above average drivers. http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From bsam at ipt.ru Wed Mar 4 02:56:01 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Wed Mar 4 02:56:08 2009 Subject: [patch] Mk/bsd.linux-rpm.mk: introduce some linux-f8 infrastructure In-Reply-To: <20090304112524.45016b7uvodhvvok@webmail.leidinger.net> (Alexander Leidinger's message of "Wed\, 04 Mar 2009 11\:25\:24 +0100") References: <14905754@bb.ipt.ru> <20090304093137.10733oaty524f0is@webmail.leidinger.net> <11370520@bb.ipt.ru> <20090304103247.417765zrsapwwow0@webmail.leidinger.net> <79218247@bb.ipt.ru> <20090304112524.45016b7uvodhvvok@webmail.leidinger.net> Message-ID: <13134976@bb.ipt.ru> Alexander Leidinger writes: > Quoting Boris Samorodov (from Wed, 04 Mar 2009 13:01:28 +0300): >> Alexander Leidinger writes: >>> Quoting Boris Samorodov (from Wed, 04 Mar 2009 >>> 12:23:35 +0300): >>>> Alexander Leidinger writes: >>>>> Quoting Boris Samorodov (from Tue, 03 Mar 2009 >>> >>>>>> . define MASTER_SITE_SUBDIR and MASTER_SITE_SRC_SUBDIR when >>>>>> LINUX_DIST_VER=8 (no such ports ATM). >>>>> >>>>> Why is there ifndef MASTER_SITES (only in the context of your patch, >>>> >>>> This ifndef was introduced by sat@ at Rev 1.9. I'm not aware of the >>>> real need. >>> >>> I had a look at it. It is about protecting MASTER_SITE_* to be set, >>> e.g. imagine if someone sets MASTER_SITE=file:///..., and at this >>> location there's a flat file structure. In this case you do not want >>> to have MASTER_SITE_* to be set. >>> >>> So I think you need to change your patch to keep this functionality. >> >> Yes, good catch, thanks! The first .endif was misplaced. Seems that >> the following patch is what you meant: > > Yes, looks commit ready. > > Personally I would remove the '?' from MASTER_SITE?=, as it is handled > by the ifndef. This way it does not confuse readers. MAybe even > putting a comment there what this is all about. Fixed: -------------- next part -------------- A non-text attachment was scrubbed... Name: Mk.diff Type: text/x-patch Size: 2050 bytes Desc: a patch for bsd.linux-rpm.mk Url : http://lists.freebsd.org/pipermail/freebsd-emulation/attachments/20090304/3449727c/Mk.bin -------------- next part -------------- WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From netchild at FreeBSD.org Wed Mar 4 12:27:57 2009 From: netchild at FreeBSD.org (netchild@FreeBSD.org) Date: Wed Mar 4 12:28:03 2009 Subject: kern/131099: [linux] [patch] readdir broken on Linux emulation. Message-ID: <200903042027.n24KRujU020379@freefall.freebsd.org> Synopsis: [linux] [patch] readdir broken on Linux emulation. State-Changed-From-To: patched->closed State-Changed-By: netchild State-Changed-When: Wed Mar 4 20:27:45 UTC 2009 State-Changed-Why: MFCed to RELENG_7. http://www.freebsd.org/cgi/query-pr.cgi?pr=131099 From dfilter at FreeBSD.ORG Wed Mar 4 12:30:07 2009 From: dfilter at FreeBSD.ORG (dfilter service) Date: Wed Mar 4 12:30:13 2009 Subject: kern/131099: commit references a PR Message-ID: <200903042030.n24KU7u4020536@freefall.freebsd.org> The following reply was made to PR kern/131099; it has been noted by GNATS. From: dfilter@FreeBSD.ORG (dfilter service) To: bug-followup@FreeBSD.org Cc: Subject: Re: kern/131099: commit references a PR Date: Wed, 4 Mar 2009 20:26:51 +0000 (UTC) Author: netchild Date: Wed Mar 4 20:26:39 2009 New Revision: 189370 URL: http://svn.freebsd.org/changeset/base/189370 Log: MFC r188572: Fix an edge-case of the linux readdir: We need the size of a linux dirent structure, not the size of a pointer to it. PR: 131099 Submitted by: Andreas Kies Modified: stable/7/sys/ (props changed) stable/7/sys/compat/linux/linux_file.c stable/7/sys/contrib/pf/ (props changed) stable/7/sys/dev/ath/ath_hal/ (props changed) stable/7/sys/dev/cxgb/ (props changed) Modified: stable/7/sys/compat/linux/linux_file.c ============================================================================== --- stable/7/sys/compat/linux/linux_file.c Wed Mar 4 18:36:48 2009 (r189369) +++ stable/7/sys/compat/linux/linux_file.c Wed Mar 4 20:26:39 2009 (r189370) @@ -438,7 +438,7 @@ getdents_common(struct thread *td, struc /* readdir(2) case. Always struct dirent. */ if (is64bit) return (EINVAL); - nbytes = sizeof(linux_dirent); + nbytes = sizeof(*linux_dirent); justone = 1; } else justone = 0; _______________________________________________ svn-src-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-all To unsubscribe, send any mail to "svn-src-all-unsubscribe@freebsd.org" From kalinoj1 at iem.pw.edu.pl Wed Mar 4 17:23:32 2009 From: kalinoj1 at iem.pw.edu.pl (Jedrzej Kalinowski) Date: Wed Mar 4 17:23:38 2009 Subject: kqemu and devfs - any luck? Message-ID: <20090305010525.GA69474@volt.iem.pw.edu.pl> Hello, I'm trying to set up kqemu (kqemu-devel from -current) to be accessible to usual non-wheel users. I've created a ruleset that looks like this: # devfs rule show 100 path kqemu mode 666 200 path kqemu unhide than I do # devfs rule applyset but sadly my /dev/kqemu are still owned by root:wheel and have 660 permissions. I'm proceeding exactly in the way that's described in man 8 devfs. Maybe someone else succeded with this and could give me a tip on how to cope with this issue? Or maybe someone can at least confirm that it does not work, and the PR should be posted? I'm working on a -CURRENT system.. Regards, Jedrzej Kalinowski From nox at jelal.kn-bremen.de Thu Mar 5 09:46:55 2009 From: nox at jelal.kn-bremen.de (Juergen Lock) Date: Thu Mar 5 09:47:01 2009 Subject: kqemu and devfs - any luck? In-Reply-To: <20090305010525.GA69474@volt.iem.pw.edu.pl> Message-ID: <200903051745.n25HjMvK004963@saturn.kn-bremen.de> In article <20090305010525.GA69474@volt.iem.pw.edu.pl> you write: >Hello, > >I'm trying to set up kqemu (kqemu-devel from -current) to be accessible to usual >non-wheel users. I've created a ruleset that looks like this: > ># devfs rule show >100 path kqemu mode 666 Try: path 'kqemu*' ... (kqemu uses device cloning.) >200 path kqemu unhide > Not sure that works with cloning as expected, at least it doesn't do anything here... >than I do ># devfs rule applyset > That also didn't work here, I did: /etc/rc.d/devfs restart >but sadly my /dev/kqemu are still owned by root:wheel and have 660 >permissions. I'm proceeding exactly in the way that's described in man 8 devfs. > >Maybe someone else succeded with this and could give me a tip on how to >cope with this issue? Or maybe someone can at least confirm that it does >not work, and the PR should be posted? I'm working on a -CURRENT >system.. 7-stable here... HTH, Juergen From bakul at bitblocks.com Thu Mar 5 10:23:37 2009 From: bakul at bitblocks.com (Bakul Shah) Date: Thu Mar 5 10:23:43 2009 Subject: kqemu and devfs - any luck? In-Reply-To: Your message of "Thu, 05 Mar 2009 02:05:26 +0100." <20090305010525.GA69474@volt.iem.pw.edu.pl> References: <20090305010525.GA69474@volt.iem.pw.edu.pl> Message-ID: <20090305181331.A0F875B5E@mail.bitblocks.com> On Thu, 05 Mar 2009 02:05:26 +0100 Jedrzej Kalinowski wrote: > Hello, > > I'm trying to set up kqemu (kqemu-devel from -current) to be accessible to us > ual > non-wheel users. I've created a ruleset that looks like this: > > # devfs rule show > 100 path kqemu mode 666 > 200 path kqemu unhide > > than I do > # devfs rule applyset > > but sadly my /dev/kqemu are still owned by root:wheel and have 660 > permissions. I'm proceeding exactly in the way that's described in man 8 devfs > > Maybe someone else succeded with this and could give me a tip on how to > cope with this issue? Or maybe someone can at least confirm that it does > not work, and the PR should be posted? I'm working on a -CURRENT > system.. > For loadable modules you can use devfs.conf so try something like the following as root. # cat >> /etc/devfs.conf < References: <20090305010525.GA69474@volt.iem.pw.edu.pl> <200903051745.n25HjMvK004963@saturn.kn-bremen.de> Message-ID: <20090305182825.GA24816@volt.iem.pw.edu.pl> Thu, Mar 05, 2009 at 06:45:22PM +0100, Juergen Lock napisa?(a): > In article <20090305010525.GA69474@volt.iem.pw.edu.pl> you write: > >Hello, > > > >I'm trying to set up kqemu (kqemu-devel from -current) to be accessible to usual > >non-wheel users. I've created a ruleset that looks like this: > > > ># devfs rule show > >100 path kqemu mode 666 > > Try: path 'kqemu*' ... (kqemu uses device cloning.) > > >200 path kqemu unhide > > > Not sure that works with cloning as expected, at least it doesn't > do anything here... > > >than I do > ># devfs rule applyset > > > That also didn't work here, I did: > /etc/rc.d/devfs restart > You were right, I was missing the proper wildcard, I can confirm it works fine: # devfs rule add path 'kqemu*' mode 666 group wheel # devfs rule show 100 path kqemu* group wheel mode 666 # devfs rule apply 100 > > 7-stable here... > > HTH, > Juergen Thanks for your help, Jedrzej Kalinowski From phusion2k at gmail.com Fri Mar 6 09:04:30 2009 From: phusion2k at gmail.com (Phusion) Date: Fri Mar 6 09:04:37 2009 Subject: Help configuring qemu networking tap mode Message-ID: I have qemu installed and have a guest os running Windows XP Professional. The host os is FreeBSD 7.1-RELEASE. I have the guest os working with the default networking. qemu -hda /data/qemu/winxp.img -m 256 -localtime -std-vga I would like to get networking tap mode working. I have set the following already. # sudo cat /boot/loader.conf if_bridge_load="YES" if_tap_load="YES" # sudo cat /etc/rc.conf ifconfig_bfe0="DHCP" cloned_interfaces="bridge0" ifconfig_bridge0="addm bfe0 up" For an example network that uses DHCP for 192.168.1.x, I would like to have the guest os have an IP address (static or dynamic) in the same range. Where do I make the necessary changes? Phusion From swhetzel at gmail.com Sat Mar 7 05:02:59 2009 From: swhetzel at gmail.com (Scot Hetzel) Date: Sat Mar 7 05:03:06 2009 Subject: Help configuring qemu networking tap mode Message-ID: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> >I would like to get networking tap mode working. I have set the >following already. : >For an example network that uses DHCP for 192.168.1.x, I would like to >have the guest os have an IP address (static or dynamic) in the same >range. Where do I make the necessary changes? Once you have setup up tap networking, you just have to configure the guest for DHCP, and it will automatically pull the DHCP address from your networks DHCP server. The below Qemu Tap Network Setup is how I setup my host system to allow the guest systems to connect to my network. This setup doesn't use any qemu startup/shutdown scripts to add/remove the tap* interfaces from bridge0. Scot ============================================= Qemu Tap Network Setup: 1. Setup tap* networking in rc.conf: cloned_interfaces="bridge0 tap0 tap1" ifconfig_bridge0="addm rl10 addm tap0 addm tap1" ifconfig_tap0="up" ifconfig_tap1="up" 2. modify /etc/sysctl.conf net.link.tap.up_on_open=1 net.link.tap.user_open=1 3. create a tap group in /etc/group and add the users who can access the /dev/tap* to the tap group. 4. add entry to /etc/devfs.conf for tap* # /dev/tap* perm tap0 0660 own tap0 root:tap perm tap1 0660 own tap1 root:tap 5. reboot system Installing Windows XP on QEMU: 1. Create disk image: qemu-img create -f raw Windows_XP.img 8G 2. start qemu to install Windows XP: Place the Windows XP CD into the cdrom drive, and execute qemu: qemu -name windows_xp -localtime -m 512 -net nic,macaddr=52:54:DE:AD:BE:EF,model=rtl8139 -net tap,ifname=tap0,script=no -hda Windows_XP.img -cdrom /dev/acd0 -boot d NOTE: 1. /dev/acd0 can be replaced with path to a cdrom iso image 2. to use user mode networking just drop -net tap* and/or -net nic*. From phusion2k at gmail.com Sat Mar 7 13:01:41 2009 From: phusion2k at gmail.com (Phusion) Date: Sat Mar 7 13:01:47 2009 Subject: Help configuring qemu networking tap mode In-Reply-To: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> References: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> Message-ID: On Sat, Mar 7, 2009 at 6:42 AM, Scot Hetzel wrote: >>I would like to get networking tap mode working. I have set the >>following already. > : >>For an example network that uses DHCP for 192.168.1.x, I would like to >>have the guest os have an IP address (static or dynamic) in the same >>range. Where do I make the necessary changes? > > Once you have setup up tap networking, you just have to configure the > guest for DHCP, and it will automatically pull the DHCP address from > your networks DHCP server. > > The below Qemu Tap Network Setup is how I setup my host system to > allow the guest systems to connect to my network. ?This setup doesn't > use any qemu startup/shutdown scripts to add/remove the tap* > interfaces from bridge0. > > Scot > ============================================= > Qemu Tap Network Setup: > > 1. Setup tap* networking in rc.conf: > > cloned_interfaces="bridge0 tap0 tap1" > ifconfig_bridge0="addm rl10 addm tap0 addm tap1" > ifconfig_tap0="up" > ifconfig_tap1="up" > > 2. modify /etc/sysctl.conf > > net.link.tap.up_on_open=1 > net.link.tap.user_open=1 > > 3. create a tap group in /etc/group and add the users who can access > the /dev/tap* to the tap group. > > 4. add entry to /etc/devfs.conf for tap* > > # /dev/tap* > perm tap0 0660 > own tap0 root:tap > perm tap1 0660 > own tap1 root:tap > > 5. reboot system > > Installing Windows XP on QEMU: > > 1. Create disk image: > > qemu-img create -f raw Windows_XP.img 8G > > 2. start qemu to install Windows XP: > > Place the Windows XP CD into the cdrom drive, and execute qemu: > > qemu -name windows_xp -localtime -m 512 -net > nic,macaddr=52:54:DE:AD:BE:EF,model=rtl8139 -net > tap,ifname=tap0,script=no -hda Windows_XP.img -cdrom /dev/acd0 -boot d > > NOTE: > 1. /dev/acd0 can be replaced with path to a cdrom iso image > 2. to use user mode networking just drop -net tap* and/or -net nic*. > With your help, I got this guest OS working. It now gets a DHCP address. There is still a problem related to the pf firewall on the host OS running FreeBSD 7.1. When I have the firewall disabled from running the guest OS works meaning it gets a DHCP address and can access the Internet. When the firewall is enabled, it doesn't get a DHCP address and can't access the Internet. Here is a copy of the /etc/pf.conf file. ext_if="bfe0" lo_if="lo0" tcp_services="{ 22 }" icmp_types="{ echoreq }" set block-policy return set loginterface $ext_if set skip on $lo_if scrub in nat-anchor "ftp-proxy/*" rdr-anchor "ftp-proxy/*" nat on $ext_if from !($ext_if) -> ($ext_if:0) rdr pass on $ext_if proto tcp to port 21 -> 127.0.0.1 port 8021 block in block drop in quick inet6 all block drop out quick inet6 all pass out keep state anchor "ftp-proxy/*" antispoof quick for { $lo_if } pass in on $ext_if proto tcp from any to ($ext_if) port $tcp_services flags S/SA keep state pass in inet proto icmp all icmp-type $icmp_types keep state I would like to get it working as well as have it set where there is port-forwarding for the host OS and the guest OS. Please advise and thanks for the help so far. Phusion From dougb at FreeBSD.org Sat Mar 7 15:08:27 2009 From: dougb at FreeBSD.org (Doug Barton) Date: Sat Mar 7 15:08:34 2009 Subject: Help configuring qemu networking tap mode In-Reply-To: <49B2F950.9010803@FreeBSD.org> References: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> <49B2F950.9010803@FreeBSD.org> Message-ID: <49B2FE65.6000009@FreeBSD.org> Doug Barton wrote: > Scot Hetzel wrote: > >> ============================================= >> Qemu Tap Network Setup: >> >> 1. Setup tap* networking in rc.conf: >> >> cloned_interfaces="bridge0 tap0 tap1" >> ifconfig_bridge0="addm rl10 addm tap0 addm tap1" > > I changed rl10 here to bge0 since that's the nic I have. Am I missing > something? And the answer is .... yes! I tried actually following the instructions and it worked. I realized that since in your command line you're telling qemu to emulate a realtek nic that's where the rl10 comes from. So I'm past the Network Install phase and we're now finalizing the install. :) BTW, I forgot to mention that I'm using qemu-devel, and so far I haven't installed kqemu since I've read some messages where people have had problems with it recently on -current. Couple more quick questions, is there some middle ground between size of the default window that pops up when I run the command line you supplied and full screen mode? This window is fine for the installation since it's just a little bigger than my typical terminal window, but I'd like something larger to do actual work on. Glancing through the man page it seems like perhaps -std-vga might be the ticket? Finally it looks to me like I'll need to set up samba on the host if I want the XP guest to have access to files that are not in the image, correct? The man page mentions the -smb option for the user mode network stack, will that work for tap too? Doug -- This .signature sanitized for your protection From dougb at FreeBSD.org Sat Mar 7 15:13:23 2009 From: dougb at FreeBSD.org (Doug Barton) Date: Sat Mar 7 15:13:29 2009 Subject: Help configuring qemu networking tap mode In-Reply-To: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> References: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> Message-ID: <49B2F950.9010803@FreeBSD.org> Scot, Thanks again for showing us all the way with this. I am trying to set mine up now but I'm running into a few "issues." I'll add some notes below. The problem I'm having right now is that the windows installer seems to be permanently stuck on the "Installing Network" phase. The indicator in the bottom right of the window is doing it's thing, and the message in the main window is changing, but the progress bar for "Installing Network" has not budged. Scot Hetzel wrote: > ============================================= > Qemu Tap Network Setup: > > 1. Setup tap* networking in rc.conf: > > cloned_interfaces="bridge0 tap0 tap1" > ifconfig_bridge0="addm rl10 addm tap0 addm tap1" I changed rl10 here to bge0 since that's the nic I have. Am I missing something? > ifconfig_tap0="up" > ifconfig_tap1="up" > > 2. modify /etc/sysctl.conf > > net.link.tap.up_on_open=1 > net.link.tap.user_open=1 > > 3. create a tap group in /etc/group and add the users who can access > the /dev/tap* to the tap group. > > 4. add entry to /etc/devfs.conf for tap* > > # /dev/tap* > perm tap0 0660 > own tap0 root:tap > perm tap1 0660 > own tap1 root:tap > > 5. reboot system FWIW, you may be able to get away without rebooting here. You can run each of those sysctl commands on the command line, then do '/etc/rc.d/netif start' and it should pick up your new settings. Also, if you're using a stripped down kernel you'll need the if_tap, if_bridge and bridgestp modules. > Installing Windows XP on QEMU: > > 1. Create disk image: > > qemu-img create -f raw Windows_XP.img 8G > > 2. start qemu to install Windows XP: > > Place the Windows XP CD into the cdrom drive, and execute qemu: > > qemu -name windows_xp -localtime -m 512 -net > nic,macaddr=52:54:DE:AD:BE:EF,model=rtl8139 -net > tap,ifname=tap0,script=no -hda Windows_XP.img -cdrom /dev/acd0 -boot d I assumed that the above should have been one long line, so that's how I ran it. Thanks again, Doug -- This .signature sanitized for your protection From dougb at FreeBSD.org Sat Mar 7 16:11:13 2009 From: dougb at FreeBSD.org (Doug Barton) Date: Sat Mar 7 16:11:18 2009 Subject: Help configuring qemu networking tap mode In-Reply-To: <49B2FE65.6000009@FreeBSD.org> References: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> <49B2F950.9010803@FreeBSD.org> <49B2FE65.6000009@FreeBSD.org> Message-ID: <49B30D19.2090203@FreeBSD.org> Doug Barton wrote: > So I'm past the Network Install phase and we're now finalizing the > install. :) Well, bad news ... the install is finished, but networking is not working. If I do user mode networking it gets an IP address but can't see the outside world. If I follow your instructions from the previous e-mail it never gets an IP address at all, the network status thing says "limited or no connectivity." Any ideas where to look? > BTW, I forgot to mention that I'm using qemu-devel, and so far I > haven't installed kqemu since I've read some messages where people > have had problems with it recently on -current. > > Couple more quick questions, is there some middle ground between size > of the default window that pops up when I run the command line you > supplied and full screen mode? I found the answer to this one, use the controls inside windows to bump up the resolution. > Finally it looks to me like I'll need to set up samba on the host if I > want the XP guest to have access to files that are not in the image, > correct? The man page mentions the -smb option for the user mode > network stack, will that work for tap too? Doug -- This .signature sanitized for your protection From dougb at FreeBSD.org Sat Mar 7 18:45:50 2009 From: dougb at FreeBSD.org (Doug Barton) Date: Sat Mar 7 18:45:56 2009 Subject: Help configuring qemu networking tap mode In-Reply-To: <49B30D19.2090203@FreeBSD.org> References: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> <49B2F950.9010803@FreeBSD.org> <49B2FE65.6000009@FreeBSD.org> <49B30D19.2090203@FreeBSD.org> Message-ID: <49B33157.9040106@FreeBSD.org> Doug Barton wrote: > Doug Barton wrote: >> So I'm past the Network Install phase and we're now finalizing the >> install. :) > > Well, bad news ... the install is finished, but networking is not > working. If I do user mode networking it gets an IP address but can't > see the outside world. If I follow your instructions from the previous > e-mail it never gets an IP address at all, the network status thing > says "limited or no connectivity." Turns out that there is a page on the wiki that has some good suggestions, although it could use a bit of tweaking. I got networking up by doing this in the host: myif=bge0 ifconfig tap create ifconfig bridge create ifconfig bridge0 addm tap0 addm $myif up Then I was able to start qemu with the tap command line and now networking works. I also installed kqemu, but I'm noticing that even with that working it's still pretty slow. I have a pretty fast core 2 duo and qemu is often pegging one of the cpus. Still, it's interesting enough to keep trying to make work ... Doug -- This .signature sanitized for your protection From swhetzel at gmail.com Sun Mar 8 00:40:05 2009 From: swhetzel at gmail.com (Scot Hetzel) Date: Sun Mar 8 00:40:12 2009 Subject: Help configuring qemu networking tap mode In-Reply-To: <49B33157.9040106@FreeBSD.org> References: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> <49B2F950.9010803@FreeBSD.org> <49B2FE65.6000009@FreeBSD.org> <49B30D19.2090203@FreeBSD.org> <49B33157.9040106@FreeBSD.org> Message-ID: <790a9fff0903080040j7f8548e0ue1d382afcdb74cbe@mail.gmail.com> On 3/7/09, Doug Barton wrote: > Doug Barton wrote: > > Doug Barton wrote: > >> So I'm past the Network Install phase and we're now finalizing the > >> install. :) > > > > Well, bad news ... the install is finished, but networking is not > > working. If I do user mode networking it gets an IP address but can't > > see the outside world. If I follow your instructions from the previous > > e-mail it never gets an IP address at all, the network status thing > > says "limited or no connectivity." > > > Turns out that there is a page on the wiki that has some good > suggestions, although it could use a bit of tweaking. I got networking > up by doing this in the host: > The How To was based on that wiki page and several others. Plus my own creation of not depending on the qemu startup/shutdown scripts to add/remove the tap device to/from the bridge. > myif=bge0 > > ifconfig tap create > ifconfig bridge create > ifconfig bridge0 addm tap0 addm $myif up > As you found out you need to add your hosts interface (bge0, rl0) to the bridge. > Then I was able to start qemu with the tap command line and now > networking works. > I was at work and couldn't answer your early questions, but it is good to hear that you were able to make it work. On 3/7/09, Doug Barton wrote: > Finally it looks to me like I'll need to set up samba on the host if I > want the XP guest to have access to files that are not in the image, > correct? The man page mentions the -smb option for the user mode > network stack, will that work for tap too? > You'll need to setup samba to allow the XP guest access to the files on the host system. I haven't tried the -smb option under a tap setup, it might work. Scot From dougb at FreeBSD.org Sun Mar 8 18:00:30 2009 From: dougb at FreeBSD.org (Doug Barton) Date: Sun Mar 8 18:00:37 2009 Subject: qemu 0.10.0 - cd/dvd drive access problems Message-ID: <49B46A29.7020500@FreeBSD.org> One of my main reasons for wanting to install qemu was to be able to use Windows tools for things like archiving DVDs. Now that I have qemu working pretty well for most things, I tried to start using it for DVD/CD access and it's not working well at all. I installed XP via a typical installation CD, so I thought I had it made in this area. :( I'm using '-cdrom /dev/acd0' the same way I did for the install. If there is a disc in the drive when I start qemu I can "see" the disc, and copy some small files from it. However trying to copy/read large files (say, 1G) doesn't work at all. I get an error saying that sector of the disc cannot be read (this is with known-good discs). I tried with and without the cd-rom DMA compile-time OPTION, no luck. If I remove the disc from the drive, or if there is no disc in the drive when I start qemu, inserting one after it starts produces no results. Every time I try to access the drive I get the "Please insert a disc into drive D:" error message. I'm using a fairly up to date 8-current as the host, btw. Any suggestions? Doug -- This .signature sanitized for your protection From daichi at ongs.co.jp Sun Mar 8 20:40:52 2009 From: daichi at ongs.co.jp (Daichi GOTO) Date: Sun Mar 8 20:40:59 2009 Subject: How to config timezone of linux_base-f8 Message-ID: <49B48FC3.70105@ongs.co.jp> Title says everything :) I want to use JST as default timezone for linux_base-f8. Anyone has any ideas? -- Daichi GOTO, http://people.freebsd.org/~daichi From gary.jennejohn at freenet.de Mon Mar 9 02:37:08 2009 From: gary.jennejohn at freenet.de (Gary Jennejohn) Date: Mon Mar 9 02:37:14 2009 Subject: qemu 0.10.0 - cd/dvd drive access problems In-Reply-To: <49B46A29.7020500@FreeBSD.org> References: <49B46A29.7020500@FreeBSD.org> Message-ID: <20090309103706.0a44fb26@ernst.jennejohn.org> On Sun, 08 Mar 2009 18:00:25 -0700 Doug Barton wrote: > One of my main reasons for wanting to install qemu was to be able to > use Windows tools for things like archiving DVDs. Now that I have qemu > working pretty well for most things, I tried to start using it for > DVD/CD access and it's not working well at all. I installed XP via a > typical installation CD, so I thought I had it made in this area. :( > > I'm using '-cdrom /dev/acd0' the same way I did for the install. If > there is a disc in the drive when I start qemu I can "see" the disc, > and copy some small files from it. However trying to copy/read large > files (say, 1G) doesn't work at all. I get an error saying that sector > of the disc cannot be read (this is with known-good discs). I tried > with and without the cd-rom DMA compile-time OPTION, no luck. If I > remove the disc from the drive, or if there is no disc in the drive > when I start qemu, inserting one after it starts produces no results. > Every time I try to access the drive I get the "Please insert a disc > into drive D:" error message. > > I'm using a fairly up to date 8-current as the host, btw. > > > Any suggestions? > Do you have vfs_aio in your kernel or as a module? I've run openSUSE in qemu and used the DVD drive with no problems at all; I have vfs_aio in my kernel. However, I don't know how much data were transfered by openSUSE. Since I was merely loading packages from the DVD it may have been only a few MB. One thing I've noticed is that the DVD drive is locked and I can't open the drawer once qemu starts, even if there is no disc in the drive. Apparently openSUSE acceses the drive (i.e. locks it) even when it's empty. --- Gary Jennejohn From phusion2k at gmail.com Mon Mar 9 06:46:09 2009 From: phusion2k at gmail.com (Phusion) Date: Mon Mar 9 06:46:19 2009 Subject: Help configuring qemu networking tap mode In-Reply-To: <790a9fff0903080040j7f8548e0ue1d382afcdb74cbe@mail.gmail.com> References: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> <49B2F950.9010803@FreeBSD.org> <49B2FE65.6000009@FreeBSD.org> <49B30D19.2090203@FreeBSD.org> <49B33157.9040106@FreeBSD.org> <790a9fff0903080040j7f8548e0ue1d382afcdb74cbe@mail.gmail.com> Message-ID: On Sun, Mar 8, 2009 at 4:40 AM, Scot Hetzel wrote: > On 3/7/09, Doug Barton wrote: >> Doug Barton wrote: >> ?> Doug Barton wrote: >> ?>> So I'm past the Network Install phase and we're now finalizing the >> ?>> install. :) >> ?> >> ?> Well, bad news ... the install is finished, but networking is not >> ?> working. If I do user mode networking it gets an IP address but can't >> ?> see the outside world. If I follow your instructions from the previous >> ?> e-mail it never gets an IP address at all, the network status thing >> ?> says "limited or no connectivity." >> >> >> Turns out that there is a page on the wiki that has some good >> ?suggestions, although it could use a bit of tweaking. I got networking >> ?up by doing this in the host: >> > The How To was based on ?that wiki page and several others. ?Plus my > own creation of not depending on the qemu startup/shutdown scripts to > add/remove the tap device to/from the bridge. > >> ?myif=bge0 >> >> ?ifconfig tap create >> ?ifconfig bridge create >> ?ifconfig bridge0 addm tap0 addm $myif up >> > As you found out you need to add your hosts interface (bge0, rl0) to the bridge. > >> ?Then I was able to start qemu with the tap command line and now >> ?networking works. >> > I was at work and couldn't answer your early questions, but it is good > to hear that ?you were able to make it work. > > On 3/7/09, Doug Barton wrote: >> ?Finally it looks to me like I'll need to set up samba on the host if I >> ?want the XP guest to have access to files that are not in the image, >> ?correct? The man page mentions the -smb option for the user mode >> ?network stack, will that work for tap too? >> > You'll need to setup samba to allow the XP guest access to the files > on the host system. ?I haven't tried the -smb option under a tap > setup, it might work. > > Scot > _______________________________________________ > freebsd-emulation@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-emulation > To unsubscribe, send any mail to "freebsd-emulation-unsubscribe@freebsd.org" > I am curious how your local firewall configuration looks on the host operating system. I am using pf firewall. Let me know. Phusion From bsam at ipt.ru Mon Mar 9 06:50:44 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Mon Mar 9 06:50:50 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: <49B48FC3.70105@ongs.co.jp> (Daichi GOTO's message of "Mon\, 09 Mar 2009 12\:40\:51 +0900") References: <49B48FC3.70105@ongs.co.jp> Message-ID: <80900349@bb.ipt.ru> On Mon, 09 Mar 2009 12:40:51 +0900 Daichi GOTO wrote: > Title says everything :) > I want to use JST as default timezone for linux_base-f8. > Anyone has any ideas? There were a thread called "Linux compat 2.6.16 reports time incorrect" back in October 2008 and a workaround. The main problem seems to be an incompatibility between current native and linux timezone versions. WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From bugmaster at FreeBSD.org Mon Mar 9 10:15:03 2009 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Mar 9 10:15:48 2009 Subject: Current problem reports assigned to freebsd-emulation@FreeBSD.org Message-ID: <200903091715.n29HF2O3045204@freefall.freebsd.org> Note: to view an individual PR, use: http://www.freebsd.org/cgi/query-pr.cgi?pr=(number). The following is a listing of current problems submitted by FreeBSD users. These represent problem reports covering all versions including experimental development code and obsolete releases. S Tracker Resp. Description -------------------------------------------------------------------------------- o kern/130724 emulation [linprocfs] [patch] cpuinfo in linprocfs is dated, cau o kern/129169 emulation [linux] [patch] Linux Emulation ENOTCONN error using n f ports/127018 emulation Linuxulator incapable of using FreeBSD's LDAP environm o kern/126232 emulation [linux] Linux ioctl TCGETS (0x5401) always fails o ports/121800 emulation x11-toolkits/linux-openmotif - OpenMotif upgrade to 2. o kern/97326 emulation [linux] file descriptor leakage in linux emulation o ports/91318 emulation [fix] graphics/linux_dri: works on amd64 too o kern/91293 emulation [svr4] [patch] *Experimental* Update to the SVR4 emula o kern/73777 emulation [linux] [patch] linux emulation: root dir special hand a kern/72920 emulation [linux]: path "prefixing" is not done on unix domain s o kern/41543 emulation [patch] [request] easier wine/w23 support o kern/39201 emulation [linux] [patch] ptrace(2) and rfork(RFLINUXTHPN) confu o kern/29698 emulation [linux] [patch] linux ipcs doesn'work o kern/21463 emulation [linux] Linux compatability mode should not allow setu o kern/11165 emulation [ibcs2] IBCS2 doesn't work correctly with PID_MAX 9999 15 problems total. From nox at jelal.kn-bremen.de Mon Mar 9 11:05:05 2009 From: nox at jelal.kn-bremen.de (Juergen Lock) Date: Mon Mar 9 11:05:24 2009 Subject: qemu 0.10.0 - cd/dvd drive access problems In-Reply-To: <20090309103706.0a44fb26@ernst.jennejohn.org> References: <49B46A29.7020500@FreeBSD.org> Message-ID: <200903091802.n29I2nYK004571@saturn.kn-bremen.de> In article <20090309103706.0a44fb26@ernst.jennejohn.org> you write: >On Sun, 08 Mar 2009 18:00:25 -0700 >Doug Barton wrote: > >> One of my main reasons for wanting to install qemu was to be able to >> use Windows tools for things like archiving DVDs. Now that I have qemu >> working pretty well for most things, I tried to start using it for >> DVD/CD access and it's not working well at all. I installed XP via a >> typical installation CD, so I thought I had it made in this area. :( >> >> I'm using '-cdrom /dev/acd0' the same way I did for the install. If >> there is a disc in the drive when I start qemu I can "see" the disc, >> and copy some small files from it. However trying to copy/read large >> files (say, 1G) doesn't work at all. I get an error saying that sector >> of the disc cannot be read (this is with known-good discs). Hmm. I wonder if this bug is related to the opensuse installer segfault, have you tried with the (now older, still at the 20080620 snapshot) qemu-devel port thats in the tree? You could also try the new "werror" option for -drive (possible values are: report - report errors to a guest as IO errors ignore - continue as if nothing happened stop - stop VM on any error and retry last command on resume enospc - stop vm on ENOSPC error and retry last command on resume all other errors are reported to a guest. I think the default is "enospc" in 0.10.0 . Yeah -cdrom is readonly but who knows...) And finally you could try disabling aio (untested, I hope the following still works:) Index: Makefile @@ -47,7 +47,7 @@ WITHOUT_CPU_CFLAGS=yes #to avoid problems with register allocation CFLAGS:= ${CFLAGS:C/-fno-tree-vrp//} -CONFIGURE_ARGS+= --prefix=${PREFIX} --cc=${CC} +CONFIGURE_ARGS+= --prefix=${PREFIX} --cc=${CC} --disable-aio .if defined(WITHOUT_SDL) CONFIGURE_ARGS+= --disable-sdl --disable-gfx-check >> I tried >> with and without the cd-rom DMA compile-time OPTION, no luck. If I >> remove the disc from the drive, or if there is no disc in the drive >> when I start qemu, inserting one after it starts produces no results. >> Every time I try to access the drive I get the "Please insert a disc >> into drive D:" error message. >> Try using the `change' monitor command (`change ide1-cd0 /dev/acd0' in this case.) Also with Windows guests istr people having to click `eject' in the guest too before it would recognize a new disc. >> I'm using a fairly up to date 8-current as the host, btw. >> >> >> Any suggestions? >> > >Do you have vfs_aio in your kernel or as a module? Actually qemu 0.10.0 (and also the last few snapshots I posted updates for) now does its own aio emulation using threads, so the aio(4) kld or kernel option is no longer needed. > I've run openSUSE >in qemu and used the DVD drive with no problems at all; I have vfs_aio >in my kernel. However, I don't know how much data were transfered by >openSUSE. Since I was merely loading packages from the DVD it may have >been only a few MB. > >One thing I've noticed is that the DVD drive is locked and I can't open >the drawer once qemu starts, even if there is no disc in the drive. >Apparently openSUSE acceses the drive (i.e. locks it) even when it's >empty. > Hmm I haven't seen that here... Maybe -current locks the door when the /dev/acd0 device node is open now too, instead of only when its mounted like its here. Good luck, Juergen From dougb at FreeBSD.org Mon Mar 9 14:08:02 2009 From: dougb at FreeBSD.org (Doug Barton) Date: Mon Mar 9 14:08:08 2009 Subject: qemu 0.10.0 - cd/dvd drive access problems In-Reply-To: <20090309103706.0a44fb26@ernst.jennejohn.org> References: <49B46A29.7020500@FreeBSD.org> <20090309103706.0a44fb26@ernst.jennejohn.org> Message-ID: On Mon, 9 Mar 2009, Gary Jennejohn wrote: > On Sun, 08 Mar 2009 18:00:25 -0700 > Doug Barton wrote: > >> One of my main reasons for wanting to install qemu was to be able to >> use Windows tools for things like archiving DVDs. Now that I have qemu >> working pretty well for most things, I tried to start using it for >> DVD/CD access and it's not working well at all. I installed XP via a >> typical installation CD, so I thought I had it made in this area. :( >> >> I'm using '-cdrom /dev/acd0' the same way I did for the install. If >> there is a disc in the drive when I start qemu I can "see" the disc, >> and copy some small files from it. However trying to copy/read large >> files (say, 1G) doesn't work at all. I get an error saying that sector >> of the disc cannot be read (this is with known-good discs). I tried >> with and without the cd-rom DMA compile-time OPTION, no luck. If I >> remove the disc from the drive, or if there is no disc in the drive >> when I start qemu, inserting one after it starts produces no results. >> Every time I try to access the drive I get the "Please insert a disc >> into drive D:" error message. >> >> I'm using a fairly up to date 8-current as the host, btw. >> >> >> Any suggestions? >> > > Do you have vfs_aio in your kernel or as a module? I'm running the aio module, which seems to incorporate the vfs_aio stuff, yes. > I've run openSUSE in qemu and used the DVD drive with no problems at > all; I have vfs_aio in my kernel. Are you talking about 'options VFS_AIO'? > One thing I've noticed is that the DVD drive is locked and I can't open > the drawer once qemu starts, even if there is no disc in the drive. > Apparently openSUSE acceses the drive (i.e. locks it) even when it's > empty. I forgot to mention this bit. If I click the eject menu option on the CD/DVD in windows it doesn't actually eject it like native windows does, but it does allow me to remove the disc manually. After that even if I do reinsert a disc the windows guest doesn't "see" it. Thanks for the response, Doug -- This .signature sanitized for your protection From dougb at FreeBSD.org Mon Mar 9 14:17:03 2009 From: dougb at FreeBSD.org (Doug Barton) Date: Mon Mar 9 14:17:09 2009 Subject: qemu 0.10.0 - cd/dvd drive access problems In-Reply-To: <200903091802.n29I2nYK004571@saturn.kn-bremen.de> References: <49B46A29.7020500@FreeBSD.org> <200903091802.n29I2nYK004571@saturn.kn-bremen.de> Message-ID: On Mon, 9 Mar 2009, Juergen Lock wrote: > In article <20090309103706.0a44fb26@ernst.jennejohn.org> you write: >> On Sun, 08 Mar 2009 18:00:25 -0700 >> Doug Barton wrote: >> >>> One of my main reasons for wanting to install qemu was to be able to >>> use Windows tools for things like archiving DVDs. Now that I have qemu >>> working pretty well for most things, I tried to start using it for >>> DVD/CD access and it's not working well at all. I installed XP via a >>> typical installation CD, so I thought I had it made in this area. :( >>> >>> I'm using '-cdrom /dev/acd0' the same way I did for the install. If >>> there is a disc in the drive when I start qemu I can "see" the disc, >>> and copy some small files from it. However trying to copy/read large >>> files (say, 1G) doesn't work at all. I get an error saying that sector >>> of the disc cannot be read (this is with known-good discs). > > Hmm. I wonder if this bug is related to the opensuse installer segfault, Not sure I follow you there, sorry. > have you tried with the (now older, still at the 20080620 snapshot) > qemu-devel port thats in the tree? That's actually what I started with (i.e., I compiled that port and used it to install my windows guest) but I switched to the 0.10.0 before I started goofing around with the CD/DVD stuff. I'll try going back to the older version, but it's not likely I'll be able to get to it today. > You could also try the new "werror" > option for -drive (possible values are: > report - report errors to a guest as IO errors > ignore - continue as if nothing happened > stop - stop VM on any error and retry last command on resume > enospc - stop vm on ENOSPC error and retry last command on resume > all other errors are reported to a guest. > I think the default is "enospc" in 0.10.0 . Yeah -cdrom is readonly > but who knows...) Errrr, -cdrom being read-only is a problem, I need to be able to write CDs and DVDs. Is there way to do that with the -drive option? I looked at the man page but couldn't quite grasp it well enough to get a command line to try. BTW, any hope that we'll have working USB in qemu on -current any time soon? I can't help thinking that some of these issues might be avoided if the windows guest could "see" my external drive directly. It would also allow me to give syncing my iPhone a try, which is another big motivator for me to be trying this at all. > And finally you could try disabling aio (untested, I hope the following > still works:) If I don't have any luck with the other options I'll give this a try as well, thanks. > Actually qemu 0.10.0 (and also the last few snapshots I posted updates > for) now does its own aio emulation using threads, so the aio(4) kld or > kernel option is no longer needed. Ah, interesting. I wonder if it's possible that the kernel module is interfering? Doug -- This .signature sanitized for your protection From nox at jelal.kn-bremen.de Mon Mar 9 15:21:09 2009 From: nox at jelal.kn-bremen.de (Juergen Lock) Date: Mon Mar 9 15:21:16 2009 Subject: qemu 0.10.0 - cd/dvd drive access problems In-Reply-To: References: <49B46A29.7020500@FreeBSD.org> <200903091802.n29I2nYK004571@saturn.kn-bremen.de> Message-ID: <20090309221828.GA13046@saturn.kn-bremen.de> On Mon, Mar 09, 2009 at 02:16:54PM -0700, Doug Barton wrote: > On Mon, 9 Mar 2009, Juergen Lock wrote: > >> In article <20090309103706.0a44fb26@ernst.jennejohn.org> you write: >>> On Sun, 08 Mar 2009 18:00:25 -0700 >>> Doug Barton wrote: >>> >>>> One of my main reasons for wanting to install qemu was to be able to >>>> use Windows tools for things like archiving DVDs. Now that I have qemu >>>> working pretty well for most things, I tried to start using it for >>>> DVD/CD access and it's not working well at all. I installed XP via a >>>> typical installation CD, so I thought I had it made in this area. :( >>>> >>>> I'm using '-cdrom /dev/acd0' the same way I did for the install. If >>>> there is a disc in the drive when I start qemu I can "see" the disc, >>>> and copy some small files from it. However trying to copy/read large >>>> files (say, 1G) doesn't work at all. I get an error saying that sector >>>> of the disc cannot be read (this is with known-good discs). >> >> Hmm. I wonder if this bug is related to the opensuse installer segfault, > > Not sure I follow you there, sorry. > Sorry I'm talking about a failure experienced by Gary... >> have you tried with the (now older, still at the 20080620 snapshot) >> qemu-devel port thats in the tree? > ...that didn't happen anymore for him after going back to the 20080620 qemu-devel port. > That's actually what I started with (i.e., I compiled that port and used it > to install my windows guest) but I switched to the 0.10.0 before I started > goofing around with the CD/DVD stuff. I'll try going back to the older > version, but it's not likely I'll be able to get to it today. > >> You could also try the new "werror" >> option for -drive (possible values are: >> report - report errors to a guest as IO errors >> ignore - continue as if nothing happened >> stop - stop VM on any error and retry last command on resume >> enospc - stop vm on ENOSPC error and retry last command on resume >> all other errors are reported to a guest. >> I think the default is "enospc" in 0.10.0 . Yeah -cdrom is readonly >> but who knows...) > > Errrr, -cdrom being read-only is a problem, I need to be able to write CDs > and DVDs. Is there way to do that with the -drive option? Not on FreeBSD atm. (this would require the scsi passthru support to be ported at least.) > I looked at the > man page but couldn't quite grasp it well enough to get a command line to > try. > > BTW, any hope that we'll have working USB in qemu on -current any time > soon? Well, the original author of the FreeBSD usb host code is too busy with other stuff now, so it seems we need another volunteer... > I can't help thinking that some of these issues might be avoided if > the windows guest could "see" my external drive directly. Oh its an usb cd drive? Then the read problems you saw could have something to do with usb as well... Anyway, scsi passthru would most likely be faster than usb passthru anyway, if the latter even works at all for burning. > It would also > allow me to give syncing my iPhone a try, which is another big motivator > for me to be trying this at all. > Btw I got a report of a bricked iphone in connection with the old qemu ush host code, although iirc that guy was trying to upgrade the firmware... (which might have made the device drop from the bus for a bit which the current usb host code doesn't really handle, at least thats my theory...) >> And finally you could try disabling aio (untested, I hope the following >> still works:) > > If I don't have any luck with the other options I'll give this a try as > well, thanks. > >> Actually qemu 0.10.0 (and also the last few snapshots I posted updates >> for) now does its own aio emulation using threads, so the aio(4) kld or >> kernel option is no longer needed. > > Ah, interesting. I wonder if it's possible that the kernel module is > interfering? I would very much doubt that, it shouldn't do anything until the aio syscalls are actually used. Cheers, Juergen From nox at jelal.kn-bremen.de Mon Mar 9 15:38:53 2009 From: nox at jelal.kn-bremen.de (Juergen Lock) Date: Mon Mar 9 15:38:59 2009 Subject: qemu 0.10.0 - cd/dvd drive access problems In-Reply-To: References: <49B46A29.7020500@FreeBSD.org> <20090309103706.0a44fb26@ernst.jennejohn.org> Message-ID: <200903092235.n29MZCFF014159@saturn.kn-bremen.de> In article you write: >On Mon, 9 Mar 2009, Gary Jennejohn wrote: > >> On Sun, 08 Mar 2009 18:00:25 -0700 >> Doug Barton wrote: >> >>> One of my main reasons for wanting to install qemu was to be able to >>> use Windows tools for things like archiving DVDs. Now that I have qemu >>> working pretty well for most things, I tried to start using it for >>> DVD/CD access and it's not working well at all. I installed XP via a >>> typical installation CD, so I thought I had it made in this area. :( >>> >>> I'm using '-cdrom /dev/acd0' the same way I did for the install. If >>> there is a disc in the drive when I start qemu I can "see" the disc, >>> and copy some small files from it. However trying to copy/read large >>> files (say, 1G) doesn't work at all. I get an error saying that sector >>> of the disc cannot be read (this is with known-good discs). I tried >>> with and without the cd-rom DMA compile-time OPTION, no luck. If I >>> remove the disc from the drive, or if there is no disc in the drive >>> when I start qemu, inserting one after it starts produces no results. >>> Every time I try to access the drive I get the "Please insert a disc >>> into drive D:" error message. >>> >>> I'm using a fairly up to date 8-current as the host, btw. >>> >>> >>> Any suggestions? >>> >> >> Do you have vfs_aio in your kernel or as a module? > >I'm running the aio module, which seems to incorporate the vfs_aio stuff, >yes. > (Not needed with qemu 0.10.0 anymore as already mentioned.) >> I've run openSUSE in qemu and used the DVD drive with no problems at >> all; I have vfs_aio in my kernel. > >Are you talking about 'options VFS_AIO'? > >> One thing I've noticed is that the DVD drive is locked and I can't open >> the drawer once qemu starts, even if there is no disc in the drive. >> Apparently openSUSE acceses the drive (i.e. locks it) even when it's >> empty. > >I forgot to mention this bit. If I click the eject menu option on the >CD/DVD in windows it doesn't actually eject it like native windows does, >but it does allow me to remove the disc manually. After that even if I do >reinsert a disc the windows guest doesn't "see" it. > Have you tried the mentioned `change ide1-cd0' in qemu's monitor? (This probably would benefit from scsi passthru as well...) HTH, Juergen From daichi at ongs.co.jp Mon Mar 9 20:28:49 2009 From: daichi at ongs.co.jp (Daichi GOTO) Date: Mon Mar 9 20:28:56 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: <80900349@bb.ipt.ru> References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> Message-ID: <49B5DE6E.6030905@ongs.co.jp> Boris Samorodov wrote: > On Mon, 09 Mar 2009 12:40:51 +0900 Daichi GOTO wrote: > >> Title says everything :) > >> I want to use JST as default timezone for linux_base-f8. >> Anyone has any ideas? > > There were a thread called "Linux compat 2.6.16 reports time incorrect" > back in October 2008 and a workaround. > > The main problem seems to be an incompatibility between current native > and linux timezone versions. > > > WBR Thanks. I have checked above mails. I understand. TO: bsam, linux_base-f8 maintainer I know :) How about to add tzdata-2008h-1.fc8.noarch.rpm to Ports Collections and dependency of linux_base-f8? If f8 has tzdata there, user can use their own tz just copying it to /compat/linux/etc/localtime. It looks good idea. -- Daichi GOTO, http://people.freebsd.org/~daichi From daichi at ongs.co.jp Mon Mar 9 21:04:29 2009 From: daichi at ongs.co.jp (Daichi GOTO) Date: Mon Mar 9 21:04:35 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: <49B5DE6E.6030905@ongs.co.jp> References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> <49B5DE6E.6030905@ongs.co.jp> Message-ID: <49B5E6CB.1010608@ongs.co.jp> Daichi GOTO wrote: > Boris Samorodov wrote: >> On Mon, 09 Mar 2009 12:40:51 +0900 Daichi GOTO wrote: >> >>> Title says everything :) >> >>> I want to use JST as default timezone for linux_base-f8. >>> Anyone has any ideas? >> >> There were a thread called "Linux compat 2.6.16 reports time incorrect" >> back in October 2008 and a workaround. >> >> The main problem seems to be an incompatibility between current native >> and linux timezone versions. >> >> >> WBR > > Thanks. I have checked above mails. I understand. > > > TO: bsam, linux_base-f8 maintainer I know :) > > How about to add tzdata-2008h-1.fc8.noarch.rpm to Ports Collections > and dependency of linux_base-f8? If f8 has tzdata there, user can > use their own tz just copying it to /compat/linux/etc/localtime. It > looks good idea. > Uh? The tzdata-2008i-1.fc8.noarch.rpm has 3 JST tzdata files. % rpm2cpio tzdata-2008i-1.fc8.noarch.rpm| cpio -idv | grep Tokyo ./usr/share/zoneinfo/right/Asia/Tokyo ./usr/share/zoneinfo/posix/Asia/Tokyo ./usr/share/zoneinfo/Asia/Tokyo 4067 blocks % % md5 ./usr/share/zoneinfo/right/Asia/Tokyo ./usr/share/zoneinfo/posix/Asia/Tokyo ./usr/share/zoneinfo/Asia/Tokyo MD5 (./usr/share/zoneinfo/right/Asia/Tokyo) = 0fce60d78288efb3685c1e2a19fb8e2b MD5 (./usr/share/zoneinfo/posix/Asia/Tokyo) = f17769e8eb35e7a08cfedfe9b2f6b5ef MD5 (./usr/share/zoneinfo/Asia/Tokyo) = f17769e8eb35e7a08cfedfe9b2f6b5ef % Which file should I use? Anyone has any ideas? -- Daichi GOTO, http://people.freebsd.org/~daichi From swhetzel at gmail.com Mon Mar 9 22:19:41 2009 From: swhetzel at gmail.com (Scot Hetzel) Date: Mon Mar 9 22:19:47 2009 Subject: Help configuring qemu networking tap mode In-Reply-To: References: <790a9fff0903070442v66de9f4ar140cf6e2a6f716ac@mail.gmail.com> <49B2F950.9010803@FreeBSD.org> <49B2FE65.6000009@FreeBSD.org> <49B30D19.2090203@FreeBSD.org> <49B33157.9040106@FreeBSD.org> <790a9fff0903080040j7f8548e0ue1d382afcdb74cbe@mail.gmail.com> Message-ID: <790a9fff0903092219n335e6901s9cde51f06a5e4cec@mail.gmail.com> On 3/9/09, Phusion wrote: > I am curious how your local firewall configuration looks on the host > operating system. I am using pf firewall. Let me know. > Currently, I haven't tried using QEMU behind the host firewall. Scot From bsam at ipt.ru Mon Mar 9 22:40:47 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Mon Mar 9 22:40:54 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: <49B5DE6E.6030905@ongs.co.jp> (Daichi GOTO's message of "Tue\, 10 Mar 2009 12\:28\:46 +0900") References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> <49B5DE6E.6030905@ongs.co.jp> Message-ID: <11308704@bb.ipt.ru> On Tue, 10 Mar 2009 12:28:46 +0900 Daichi GOTO wrote: > How about to add tzdata-2008h-1.fc8.noarch.rpm to Ports Collections > and dependency of linux_base-f8? If f8 has tzdata there, user can > use their own tz just copying it to /compat/linux/etc/localtime. It > looks good idea. It's on my TODO list but not the first one. ATM I'm working on new linux infrastructure ports. And as always patches are welcome. ;-) WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From daichi at ongs.co.jp Mon Mar 9 22:44:30 2009 From: daichi at ongs.co.jp (Daichi GOTO) Date: Mon Mar 9 22:44:36 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: <11308704@bb.ipt.ru> References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> <49B5DE6E.6030905@ongs.co.jp> <11308704@bb.ipt.ru> Message-ID: <49B5FE3C.4080104@ongs.co.jp> Boris Samorodov wrote: > On Tue, 10 Mar 2009 12:28:46 +0900 Daichi GOTO wrote: > >> How about to add tzdata-2008h-1.fc8.noarch.rpm to Ports Collections >> and dependency of linux_base-f8? If f8 has tzdata there, user can >> use their own tz just copying it to /compat/linux/etc/localtime. It >> looks good idea. > > It's on my TODO list but not the first one. ATM I'm working on new > linux infrastructure ports. And as always patches are welcome. ;-) > > > WBR Sure I understand. Thanks ;-) -- Daichi GOTO, http://people.freebsd.org/~daichi From Alexander at Leidinger.net Tue Mar 10 00:55:31 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Tue Mar 10 00:55:39 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: <11308704@bb.ipt.ru> References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> <49B5DE6E.6030905@ongs.co.jp> <11308704@bb.ipt.ru> Message-ID: <20090310085518.19936d06a3qyy2w4@webmail.leidinger.net> Quoting Boris Samorodov (from Tue, 10 Mar 2009 08:40:47 +0300): > On Tue, 10 Mar 2009 12:28:46 +0900 Daichi GOTO wrote: > >> How about to add tzdata-2008h-1.fc8.noarch.rpm to Ports Collections >> and dependency of linux_base-f8? If f8 has tzdata there, user can >> use their own tz just copying it to /compat/linux/etc/localtime. It >> looks good idea. > > It's on my TODO list but not the first one. ATM I'm working on new > linux infrastructure ports. And as always patches are welcome. ;-) I want to add that there's still hope that the timezone stuff in -current gets updated soonish (which means we don't need tzdata in f8, as 7.1 will not switch to f8 by default as it would 1. violate POLA and 2. does not have the necessary emulation stuff to be compatible with 2.6.16). Bye, Alexander. -- "Eureka!" -Professor "Did you build the Smell-o-scope?" -Fry "No. I remembered that I built one last year." -Professor http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From gary.jennejohn at freenet.de Tue Mar 10 03:01:20 2009 From: gary.jennejohn at freenet.de (Gary Jennejohn) Date: Tue Mar 10 03:01:26 2009 Subject: qemu 0.10.0 - cd/dvd drive access problems In-Reply-To: References: <49B46A29.7020500@FreeBSD.org> <20090309103706.0a44fb26@ernst.jennejohn.org> Message-ID: <20090310110117.00fcdf6b@ernst.jennejohn.org> On Mon, 9 Mar 2009 14:07:57 -0700 (PDT) Doug Barton wrote: > On Mon, 9 Mar 2009, Gary Jennejohn wrote: > > I've run openSUSE in qemu and used the DVD drive with no problems at > > all; I have vfs_aio in my kernel. > > Are you talking about 'options VFS_AIO'? > Yes. I wasn't aware that qemu had incorporated this functionality. As Juergen wrote, it doesn't seem to interfere. --- Gary Jennejohn From scf at FreeBSD.org Tue Mar 10 11:34:47 2009 From: scf at FreeBSD.org (Sean C. Farley) Date: Tue Mar 10 11:34:54 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: <20090310085518.19936d06a3qyy2w4@webmail.leidinger.net> References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> <49B5DE6E.6030905@ongs.co.jp> <11308704@bb.ipt.ru> <20090310085518.19936d06a3qyy2w4@webmail.leidinger.net> Message-ID: On Tue, 10 Mar 2009, Alexander Leidinger wrote: *snip* >> It's on my TODO list but not the first one. ATM I'm working on new >> linux infrastructure ports. And as always patches are welcome. ;-) > > I want to add that there's still hope that the timezone stuff in > -current gets updated soonish (which means we don't need tzdata in f8, > as 7.1 will not switch to f8 by default as it would 1. violate POLA > and 2. does not have the necessary emulation stuff to be compatible > with 2.6.16). Out of curiosity, when you say necessary emulation stuff to be compatible with 2.6.16, how compatible is it? I use it on 7-STABLE quite well with Skype and Acroread8, somewhat well with Flash plugin 9 and not well at all with linux-ut (at least with the nvidia driver). Will all that is necessary to use full 2.6.16 emulation be eventually MFC'd, or will only some of it? Of course, keeping the default at 2.4.2 emulation is perfectly fine. On a related note, would it be possible to be able to run one application under 2.4.2 emulation while another is under 2.6.16 in the future? I thought it may help in some situations. There could be a /compat/linux.f8 (with /compat/linux linking to it) and /compat/linux.fc4 where the application can be told via an environment variable to use a specific install. Sean -- scf@FreeBSD.org From rdivacky at FreeBSD.org Tue Mar 10 11:47:09 2009 From: rdivacky at FreeBSD.org (Roman Divacky) Date: Tue Mar 10 11:47:16 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> <49B5DE6E.6030905@ongs.co.jp> <11308704@bb.ipt.ru> <20090310085518.19936d06a3qyy2w4@webmail.leidinger.net> Message-ID: <20090310184655.GA96302@freebsd.org> > On a related note, would it be possible to be able to run one > application under 2.4.2 emulation while another is under 2.6.16 in the > future? I thought it may help in some situations. There could be a > /compat/linux.f8 (with /compat/linux linking to it) and > /compat/linux.fc4 where the application can be told via an environment > variable to use a specific install. speaking from top of my head... yes, this could be done, theoretically. the 2.6 sysctl would have to be made per-proc (as netbsd does iirc), and then it should work. noone has done it yet From Alexander at Leidinger.net Tue Mar 10 13:14:36 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Tue Mar 10 13:14:43 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: <20090310184655.GA96302@freebsd.org> References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> <49B5DE6E.6030905@ongs.co.jp> <11308704@bb.ipt.ru> <20090310085518.19936d06a3qyy2w4@webmail.leidinger.net> <20090310184655.GA96302@freebsd.org> Message-ID: <20090310211421.00005a18@unknown> On Tue, 10 Mar 2009 19:46:55 +0100 Roman Divacky wrote: > > On a related note, would it be possible to be able to run one > > application under 2.4.2 emulation while another is under 2.6.16 in > > the future? I thought it may help in some situations. There could > > be a /compat/linux.f8 (with /compat/linux linking to it) and > > /compat/linux.fc4 where the application can be told via an > > environment variable to use a specific install. > > speaking from top of my head... yes, this could be done, > theoretically. > > the 2.6 sysctl would have to be made per-proc (as netbsd does iirc), > and then it should work. How to set it before starting the corresponding linux program? I see a chicken and egg problem here. Bye, Alexander. From Alexander at Leidinger.net Tue Mar 10 13:26:03 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Tue Mar 10 13:26:10 2009 Subject: How to config timezone of linux_base-f8 In-Reply-To: References: <49B48FC3.70105@ongs.co.jp> <80900349@bb.ipt.ru> <49B5DE6E.6030905@ongs.co.jp> <11308704@bb.ipt.ru> <20090310085518.19936d06a3qyy2w4@webmail.leidinger.net> Message-ID: <20090310212547.00004fcd@unknown> On Tue, 10 Mar 2009 13:34:45 -0500 (CDT) "Sean C. Farley" wrote: > On Tue, 10 Mar 2009, Alexander Leidinger wrote: > > *snip* > > >> It's on my TODO list but not the first one. ATM I'm working on new > >> linux infrastructure ports. And as always patches are welcome. ;-) > > > > I want to add that there's still hope that the timezone stuff in > > -current gets updated soonish (which means we don't need tzdata in > > f8, as 7.1 will not switch to f8 by default as it would 1. violate > > POLA and 2. does not have the necessary emulation stuff to be > > compatible with 2.6.16). > > Out of curiosity, when you say necessary emulation stuff to be > compatible with 2.6.16, how compatible is it? I use it on 7-STABLE > quite well with Skype and Acroread8, somewhat well with Flash plugin > 9 and not well at all with linux-ut (at least with the nvidia driver). I would say it is compatible enough to run Skype Acroread8 and a little bit of Flash... ;-) What is missing is for example the *at() funktions. Soon there will be a different futex implementation in -current. There may be more which escapes my memory ATM. > Will all that is necessary to use full 2.6.16 emulation be eventually > MFC'd, or will only some of it? Of course, keeping the default at > 2.4.2 emulation is perfectly fine. The *at() stuff can not be MFCed, as it depends upon VFS stuff which will not be MFCed due to API/ABI concerns. The futex stuff can maybe MFCed, but I don't really know. > On a related note, would it be possible to be able to run one > application under 2.4.2 emulation while another is under 2.6.16 in > the future? I thought it may help in some situations. There could > be a /compat/linux.f8 (with /compat/linux linking to it) and > /compat/linux.fc4 where the application can be told via an > environment variable to use a specific install. Currently it's either 2.4 or 2.6 (don't switch while a linux process is (still) running), and with the current implementation I don't really see an easy way to do this. Someone could make a copy of the current linuxulator and use a different ELF-brand for the copy. The copy would then have to be modified to use a different path for the linux base, and every non-static function/symbol would need to be renamed. Praktically I would say too much work, not enough return of investment. Would be cool to have from a geeky point, but if you really _need_ this (business stuff instead of "I would like to"), take an older version of FreeBSD and let it run on another machine. Bye, Alexander. From novembre at gmail.com Wed Mar 11 17:14:12 2009 From: novembre at gmail.com (Novembre) Date: Wed Mar 11 17:14:18 2009 Subject: Flash 9 crash problems In-Reply-To: <20090304084315.587314cev8vm3f34@webmail.leidinger.net> References: <3b47caa90903021351g4ceb832ap866cccd940cfb68b@mail.gmail.com> <20090303083430.64927hdcfxdcmjok@webmail.leidinger.net> <3b47caa90903031841m75eb3553jdd38cc47972990cc@mail.gmail.com> <20090304084315.587314cev8vm3f34@webmail.leidinger.net> Message-ID: <3b47caa90903111714p4b82f672t6597b4fa84cc2ba6@mail.gmail.com> On Wed, Mar 4, 2009 at 2:43 AM, Alexander Leidinger wrote: > Quoting Novembre (from Tue, 3 Mar 2009 20:41:03 > -0600): > > I found two posts containing patches regarding this issue: >> 1) >> >> http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2008/freebsd-emulation/20081228.freebsd-emulation >> 2) >> >> http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2009/freebsd-emulation/20090215.freebsd-emulation >> >> The first post contains two patches, of which one is marked specifically >> for >> 7-STABLE (as of Dec 21, 2008 when the patches were made available). The >> second post also has a patch, but I don't know whether it is for 8-CURRENT >> or 7-STABLE. Which one should I apply? >> > > If you do not want to wait: the patch utility has the option "-C" which > just tries to apply the patch but does not really apply it. This way you can > see if a patch is applicable to what you have or not. This way you can test > if one of all those patches applies at all to what you have. In the worst > case it applies, but you get a kernel panic upon use of some linux stuff > (you may lose data because of this). In the best case at least one applies > and it works perfectly. > > Bye, > Alexander. > > -- > If we see the light at the end of the tunnel > It's the light of an oncoming train. > -- Robert Lowell > > > http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 > http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 > I copied both files to /usr/src/sys/ and tried patching with -C option. It asks me a question about which file to patch! Here's the result: #pwd /usr/src/sys #patch -C < futexes_partial_II.patch Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |diff --git a/sys/amd64/amd64/elf_machdep.c b/sys/amd64/amd64/elf_machdep.c |index 4f6d178..0aea61d 100644 |--- a/sys/amd64/amd64/elf_machdep.c |+++ b/sys/amd64/amd64/elf_machdep.c -------------------------- File to patch: # patch -C < mega-st7.linux.patch Hmm... Looks like a unified diff to me... The text leading up to this was: -------------------------- |diff --git a/sys/amd64/include/elf.h b/sys/amd64/include/elf.h |index 0a386be..3c2cd20 100644 |--- a/sys/amd64/include/elf.h |+++ b/sys/amd64/include/elf.h -------------------------- File to patch: I have no idea what to tell it, so I just press Ctrl-C to cancel the command. Any ideas? Thanks From bsam at ipt.ru Wed Mar 11 23:14:58 2009 From: bsam at ipt.ru (Boris Samorodov) Date: Wed Mar 11 23:15:04 2009 Subject: Flash 9 crash problems In-Reply-To: <3b47caa90903111714p4b82f672t6597b4fa84cc2ba6@mail.gmail.com> (novembre@gmail.com's message of "Wed\, 11 Mar 2009 19\:14\:11 -0500") References: <3b47caa90903021351g4ceb832ap866cccd940cfb68b@mail.gmail.com> <20090303083430.64927hdcfxdcmjok@webmail.leidinger.net> <3b47caa90903031841m75eb3553jdd38cc47972990cc@mail.gmail.com> <20090304084315.587314cev8vm3f34@webmail.leidinger.net> <3b47caa90903111714p4b82f672t6597b4fa84cc2ba6@mail.gmail.com> Message-ID: <02319391@bb.ipt.ru> On Wed, 11 Mar 2009 19:14:11 -0500 Novembre wrote: > On Wed, Mar 4, 2009 at 2:43 AM, Alexander Leidinger > wrote: > > Quoting Novembre (from Tue, 3 Mar 2009 20:41:03 > > -0600): > > > > I found two posts containing patches regarding this issue: > >> 1) > >> > >> http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2008/freebsd-emulation/20081228.freebsd-emulation > >> 2) > >> > >> http://docs.freebsd.org/cgi/getmsg.cgi?fetch=0+0+archive/2009/freebsd-emulation/20090215.freebsd-emulation > >> > >> The first post contains two patches, of which one is marked specifically > >> for > >> 7-STABLE (as of Dec 21, 2008 when the patches were made available). The > >> second post also has a patch, but I don't know whether it is for 8-CURRENT > >> or 7-STABLE. Which one should I apply? > >> > > > > If you do not want to wait: the patch utility has the option "-C" which > > just tries to apply the patch but does not really apply it. This way you can > > see if a patch is applicable to what you have or not. This way you can test > > if one of all those patches applies at all to what you have. In the worst > > case it applies, but you get a kernel panic upon use of some linux stuff > > (you may lose data because of this). In the best case at least one applies > > and it works perfectly. > > > I copied both files to /usr/src/sys/ and tried patching with -C option. It > asks me a question about > which file to patch! Here's the result: > #pwd > /usr/src/sys > #patch -C < futexes_partial_II.patch # patch -Cp2 ... > Hmm... Looks like a unified diff to me... > The text leading up to this was: > -------------------------- > |diff --git a/sys/amd64/amd64/elf_machdep.c b/sys/amd64/amd64/elf_machdep.c > |index 4f6d178..0aea61d 100644 > |--- a/sys/amd64/amd64/elf_machdep.c > |+++ b/sys/amd64/amd64/elf_machdep.c > -------------------------- > File to patch: > # patch -C < mega-st7.linux.patch The same here. > Hmm... Looks like a unified diff to me... > The text leading up to this was: > -------------------------- > |diff --git a/sys/amd64/include/elf.h b/sys/amd64/include/elf.h > |index 0a386be..3c2cd20 100644 > |--- a/sys/amd64/include/elf.h > |+++ b/sys/amd64/include/elf.h > -------------------------- > File to patch: > I have no idea what to tell it, so I just press Ctrl-C to cancel the > command. Any ideas? man patch, please WBR -- Boris Samorodov (bsam) Research Engineer, http://www.ipt.ru Telephone & Internet SP FreeBSD Committer, http://www.FreeBSD.org The Power To Serve From c.r.n.a at wanadoo.fr Thu Mar 12 02:18:44 2009 From: c.r.n.a at wanadoo.fr (Nicolas) Date: Thu Mar 12 02:18:50 2009 Subject: Flash 9 crash problems References: 20090304084315.587314cev8vm3f34@webmail.leidinger.net Message-ID: <49B8DDCE.5000904@wanadoo.fr> Hi Novembre, If you want to apply this patch, you need to replace all "--- a/sys/" by "--- /sys/", and "+++ b/sys/" by "+++ /sys/" copy the patch on "/usr/src" and apply with command "patch -p0 < nameof patch.diff" This patch does not build on 8-Current ! Niko. From rbgarga at gmail.com Thu Mar 12 04:43:09 2009 From: rbgarga at gmail.com (Renato Botelho) Date: Thu Mar 12 04:43:15 2009 Subject: Flash 9 crash problems In-Reply-To: <49B8DDCE.5000904@wanadoo.fr> References: <49B8DDCE.5000904@wanadoo.fr> Message-ID: <747dc8f30903120443m19e96f62nde5feb6e6fc23870@mail.gmail.com> On Thu, Mar 12, 2009 at 7:02 AM, Nicolas wrote: > Hi Novembre, > > If you want to apply this patch, you need to replace all "--- a/sys/" by > "--- /sys/", and "+++ b/sys/" by "+++ /sys/" copy the patch on "/usr/src" > and apply with command "patch -p0 < nameof patch.diff" You can use patch -p1 instead of replace path inside patch. -- Renato Botelho From bugmaster at FreeBSD.org Mon Mar 16 04:06:53 2009 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Mar 16 04:07:38 2009 Subject: Current problem reports assigned to freebsd-emulation@FreeBSD.org Message-ID: <200903161106.n2GB6qwp043205@freefall.freebsd.org> Note: to view an individual PR, use: http://www.freebsd.org/cgi/query-pr.cgi?pr=(number). The following is a listing of current problems submitted by FreeBSD users. These represent problem reports covering all versions including experimental development code and obsolete releases. S Tracker Resp. Description -------------------------------------------------------------------------------- o kern/130724 emulation [linprocfs] [patch] cpuinfo in linprocfs is dated, cau o kern/129169 emulation [linux] [patch] Linux Emulation ENOTCONN error using n f ports/127018 emulation Linuxulator incapable of using FreeBSD's LDAP environm o kern/126232 emulation [linux] Linux ioctl TCGETS (0x5401) always fails o ports/121800 emulation x11-toolkits/linux-openmotif - OpenMotif upgrade to 2. o kern/97326 emulation [linux] file descriptor leakage in linux emulation o ports/91318 emulation [fix] graphics/linux_dri: works on amd64 too o kern/91293 emulation [svr4] [patch] *Experimental* Update to the SVR4 emula o kern/73777 emulation [linux] [patch] linux emulation: root dir special hand a kern/72920 emulation [linux]: path "prefixing" is not done on unix domain s o kern/41543 emulation [patch] [request] easier wine/w23 support o kern/39201 emulation [linux] [patch] ptrace(2) and rfork(RFLINUXTHPN) confu o kern/29698 emulation [linux] [patch] linux ipcs doesn'work o kern/21463 emulation [linux] Linux compatability mode should not allow setu o kern/11165 emulation [ibcs2] IBCS2 doesn't work correctly with PID_MAX 9999 15 problems total. From bds at waywood.co.uk Mon Mar 16 10:06:47 2009 From: bds at waywood.co.uk (Barnaby Scott) Date: Mon Mar 16 10:06:53 2009 Subject: wine without X Message-ID: <33eaa170b754b20e442a5b99acac280a.squirrel@webmail.gradwell.net> I hope this is the right place to be asking... I have been trying to figure out a way to install wine, but without X. I tried installing from ports, but I couldn't see an obvious way to to avoid all the mainly X-related dependencies being built. Searching around I see there is some talk of being able to do this with Linux, but I am not nearly knowledgeable enough to be confident mucking about with Makefiles etc to try and reproduce the necessary steps. Surely someone else (who knows what they're doing!) has trodden this path before me? In case you're wondering why I want to do this, it is for a render slave for the renderer Vray. It consists of a single Windows executable which has absolutely no visible output at all for the user - it is not strictly a Windows service, but is more or less daemon-like. I know it works with wine under Linux, so assume it will work under FreeBSD, but I *really* want to avoid installing X on all the boxes in question, as it is so clearly unnecessary in this case. Any pointers gratefully received. Barnaby Scott From nox at jelal.kn-bremen.de Sat Mar 21 13:20:38 2009 From: nox at jelal.kn-bremen.de (Juergen Lock) Date: Sat Mar 21 13:20:45 2009 Subject: want to test a patch? (was: Re: qemu 0.10.0 - cd/dvd drive access problems) In-Reply-To: <200903092235.n29MZCFF014159@saturn.kn-bremen.de> References: <49B46A29.7020500@FreeBSD.org> <20090309103706.0a44fb26@ernst.jennejohn.org> Message-ID: <200903212017.n2LKHCuq070764@saturn.kn-bremen.de> In article <200903092235.n29MZCFF014159@saturn.kn-bremen.de> I write: >In article you write: >>On Mon, 9 Mar 2009, Gary Jennejohn wrote: >> >>> On Sun, 08 Mar 2009 18:00:25 -0700 >>> Doug Barton wrote: >[...] >>> One thing I've noticed is that the DVD drive is locked and I can't open >>> the drawer once qemu starts, even if there is no disc in the drive. >>> Apparently openSUSE acceses the drive (i.e. locks it) even when it's >>> empty. >> I'm not sure but this seems to be a bug in our acd driver, it seems to lock the door when you open the device node regardless if there is a disc in the drive or not... >>I forgot to mention this bit. If I click the eject menu option on the >>CD/DVD in windows it doesn't actually eject it like native windows does, >>but it does allow me to remove the disc manually. After that even if I do >>reinsert a disc the windows guest doesn't "see" it. >> > Have you tried the mentioned `change ide1-cd0' in qemu's monitor? >(This probably would benefit from scsi passthru as well...) After finding out that there actually is code in qemu now that attempts to support these things on Linux I played with that a bit and came up with the patch below that adds some FreeBSD support. You will still have to tell the guest to eject the disc when you want to take it out/change it (or do the mentioned change command in the monitor if the guest doesnt have an eject command), but other than that a passed cd drive should now work like it now does on Linux hosts. I also did some fixes to the scsi cdrom emulation while I was at it (eject/load command support was missing too), so passing a cd drive as -drive if=scsi,media=cdrom,file=/dev/acd0 (or also /dev/cd0) should now work as well. (It is still readonly tho, scsi passthru support would be the next thing to be ported i.e. so you could do file=/dev/pass0...) Enjoy, Juergen Index: qemu/block-raw-posix.c @@ -55,6 +55,7 @@ #ifdef __FreeBSD__ #include #include +#include #endif #ifdef __OpenBSD__ @@ -105,6 +106,9 @@ int fd_got_error; int fd_media_changed; #endif +#if defined(__FreeBSD__) + int cd_open_flags; +#endif uint8_t* aligned_buf; } BDRVRawState; @@ -112,6 +116,12 @@ static int fd_open(BlockDriverState *bs); +#if defined(__FreeBSD__) +static int cd_open(BlockDriverState *bs); +#endif + +static int raw_is_inserted(BlockDriverState *bs); + static int raw_open(BlockDriverState *bs, const char *filename, int flags) { BDRVRawState *s = bs->opaque; @@ -747,6 +757,9 @@ int64_t size; #ifdef _BSD struct stat sb; +#ifdef __FreeBSD__ + int reopened = 0; +#endif #endif #ifdef __sun__ struct dk_minfo minfo; @@ -759,6 +772,9 @@ return ret; #ifdef _BSD +#ifdef __FreeBSD__ +again: +#endif if (!fstat(fd, &sb) && (S_IFCHR & sb.st_mode)) { #ifdef DIOCGMEDIASIZE if (ioctl(fd, DIOCGMEDIASIZE, (off_t *)&size)) @@ -768,6 +784,19 @@ #else size = lseek(fd, 0LL, SEEK_END); #endif +#ifdef __FreeBSD__ + switch(s->type) { + case FTYPE_CD: + /* XXX FreeBSD acd returns UINT_MAX sectors for an empty drive */ + if (size == 2048LL * (unsigned)-1) + size = 0; + /* XXX no disc? maybe we need to reopen... */ + if (size <= 0 && !reopened && cd_open(bs) >= 0) { + reopened = 1; + goto again; + } + } +#endif } else #endif #ifdef __sun__ @@ -958,6 +987,14 @@ bs->sg = 1; } #endif +#if defined(__FreeBSD__) + if (strstart(filename, "/dev/cd", NULL) || + strstart(filename, "/dev/acd", NULL)) { + s->type = FTYPE_CD; + s->cd_open_flags = open_flags; + } +#endif + s->fd = -1; fd = open(filename, open_flags, 0644); if (fd < 0) { ret = -errno; @@ -966,6 +1003,11 @@ return ret; } s->fd = fd; +#if defined(__FreeBSD__) + /* make sure the door isnt locked at this time */ + if (s->type == FTYPE_CD) + ioctl (s->fd, CDIOCALLOW); +#endif #if defined(__linux__) /* close fd so that we can reopen it as needed */ if (s->type == FTYPE_FD) { @@ -1132,7 +1174,116 @@ return ioctl(s->fd, req, buf); } -#else +#elif defined(__FreeBSD__) + +static int fd_open(BlockDriverState *bs) +{ + BDRVRawState *s = bs->opaque; + + /* this is just to ensure s->fd is sane (its called by io ops) */ + if (s->fd >= 0) + return 0; + return -EIO; +} + +static int cd_open(BlockDriverState *bs) +{ +#if defined(__FreeBSD__) + BDRVRawState *s = bs->opaque; + int fd; + + switch(s->type) { + case FTYPE_CD: + /* XXX force reread of possibly changed/newly loaded disc, + * FreeBSD seems to not notice sometimes... */ + if (s->fd >= 0) + close (s->fd); + fd = open(bs->filename, s->cd_open_flags, 0644); + if (fd < 0) { + s->fd = -1; + return -EIO; + } + s->fd = fd; + /* make sure the door isnt locked at this time */ + ioctl (s->fd, CDIOCALLOW); + } +#endif + return 0; +} + +static int raw_is_inserted(BlockDriverState *bs) +{ + BDRVRawState *s = bs->opaque; + + switch(s->type) { + case FTYPE_CD: + return (raw_getlength(bs) > 0); + case FTYPE_FD: + /* XXX handle this */ + /* FALLTHRU */ + default: + return 1; + } +} + +static int raw_media_changed(BlockDriverState *bs) +{ + return -ENOTSUP; +} + +static int raw_eject(BlockDriverState *bs, int eject_flag) +{ + BDRVRawState *s = bs->opaque; + + switch(s->type) { + case FTYPE_CD: + if (s->fd < 0) + return -ENOTSUP; + (void) ioctl (s->fd, CDIOCALLOW); + if (eject_flag) { + if (ioctl (s->fd, CDIOCEJECT) < 0) + perror("CDIOCEJECT"); + } else { + if (ioctl (s->fd, CDIOCCLOSE) < 0) + perror("CDIOCCLOSE"); + } + if (cd_open(bs) < 0) + return -ENOTSUP; + break; + case FTYPE_FD: + /* XXX handle this */ + /* FALLTHRU */ + default: + return -ENOTSUP; + } + return 0; +} + +static int raw_set_locked(BlockDriverState *bs, int locked) +{ + BDRVRawState *s = bs->opaque; + + switch(s->type) { + case FTYPE_CD: + if (s->fd < 0) + return -ENOTSUP; + if (ioctl (s->fd, (locked ? CDIOCPREVENT : CDIOCALLOW)) < 0) { + /* Note: an error can happen if the distribution automatically + mounts the CD-ROM */ + // perror("CDROM_LOCKDOOR"); + } + break; + default: + return -ENOTSUP; + } + return 0; +} + +static int raw_ioctl(BlockDriverState *bs, unsigned long int req, void *buf) +{ + return -ENOTSUP; +} +#else /* !linux && !FreeBSD */ static int fd_open(BlockDriverState *bs) { @@ -1163,7 +1314,7 @@ { return -ENOTSUP; } -#endif /* !linux */ +#endif /* !linux && !FreeBSD */ BlockDriver bdrv_host_device = { "host_device", Index: qemu/hw/scsi-disk.c @@ -417,16 +417,26 @@ switch (command) { case 0x0: DPRINTF("Test Unit Ready\n"); + if (!bdrv_is_inserted(s->bdrv)) + goto notready; break; case 0x03: DPRINTF("Request Sense (len %d)\n", len); if (len < 4) goto fail; memset(outbuf, 0, 4); + r->buf_len = 4; + if (s->sense == SENSE_NOT_READY && len >= 18) { + memset(outbuf, 0, 18); + r->buf_len = 18; + outbuf[7] = 10; + /* asc 0x3a, ascq 0: Medium not present */ + outbuf[12] = 0x3a; + outbuf[13] = 0; + } outbuf[0] = 0xf0; outbuf[1] = 0; outbuf[2] = s->sense; - r->buf_len = 4; break; case 0x12: DPRINTF("Inquiry (len %d)\n", len); @@ -725,6 +735,10 @@ break; case 0x1b: DPRINTF("Start Stop Unit\n"); + if (bdrv_get_type_hint(s->bdrv) == BDRV_TYPE_CDROM && + (buf[4] & 2)) + /* load/eject medium */ + bdrv_eject(s->bdrv, !(buf[4] & 1)); break; case 0x1e: DPRINTF("Prevent Allow Medium Removal (prevent = %d)\n", buf[4] & 3); @@ -754,6 +768,7 @@ outbuf[7] = 0; r->buf_len = 8; } else { + notready: scsi_command_complete(r, STATUS_CHECK_CONDITION, SENSE_NOT_READY); return 0; } @@ -790,6 +805,7 @@ start_track = buf[6]; bdrv_get_geometry(s->bdrv, &nb_sectors); DPRINTF("Read TOC (track %d format %d msf %d)\n", start_track, format, msf >> 1); + nb_sectors /= s->cluster_size; switch(format) { case 0: toclen = cdrom_read_toc(nb_sectors, outbuf, msf, start_track); From bugmaster at FreeBSD.org Mon Mar 23 04:06:55 2009 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Mar 23 04:07:43 2009 Subject: Current problem reports assigned to freebsd-emulation@FreeBSD.org Message-ID: <200903231106.n2NB6rrH003959@freefall.freebsd.org> Note: to view an individual PR, use: http://www.freebsd.org/cgi/query-pr.cgi?pr=(number). The following is a listing of current problems submitted by FreeBSD users. These represent problem reports covering all versions including experimental development code and obsolete releases. S Tracker Resp. Description -------------------------------------------------------------------------------- o kern/130724 emulation [linprocfs] [patch] cpuinfo in linprocfs is dated, cau o kern/129169 emulation [linux] [patch] Linux Emulation ENOTCONN error using n f ports/127018 emulation Linuxulator incapable of using FreeBSD's LDAP environm o kern/126232 emulation [linux] Linux ioctl TCGETS (0x5401) always fails o ports/121800 emulation x11-toolkits/linux-openmotif - OpenMotif upgrade to 2. o kern/97326 emulation [linux] file descriptor leakage in linux emulation o ports/91318 emulation [fix] graphics/linux_dri: works on amd64 too o kern/91293 emulation [svr4] [patch] *Experimental* Update to the SVR4 emula o kern/73777 emulation [linux] [patch] linux emulation: root dir special hand a kern/72920 emulation [linux]: path "prefixing" is not done on unix domain s o kern/41543 emulation [patch] [request] easier wine/w23 support o kern/39201 emulation [linux] [patch] ptrace(2) and rfork(RFLINUXTHPN) confu o kern/29698 emulation [linux] [patch] linux ipcs doesn'work o kern/21463 emulation [linux] Linux compatability mode should not allow setu o kern/11165 emulation [ibcs2] IBCS2 doesn't work correctly with PID_MAX 9999 15 problems total. From gary.jennejohn at freenet.de Tue Mar 24 03:24:53 2009 From: gary.jennejohn at freenet.de (Gary Jennejohn) Date: Tue Mar 24 03:24:59 2009 Subject: want to test a patch? (was: Re: qemu 0.10.0 - cd/dvd drive access problems) In-Reply-To: <200903212017.n2LKHCuq070764@saturn.kn-bremen.de> References: <49B46A29.7020500@FreeBSD.org> <20090309103706.0a44fb26@ernst.jennejohn.org> <200903212017.n2LKHCuq070764@saturn.kn-bremen.de> Message-ID: <20090324112448.6d3bab52@ernst.jennejohn.org> On Sat, 21 Mar 2009 21:17:12 +0100 (CET) Juergen Lock wrote: [snip] > After finding out that there actually is code in qemu now that attempts > to support these things on Linux I played with that a bit and came up with > the patch below that adds some FreeBSD support. You will still have to > tell the guest to eject the disc when you want to take it out/change it > (or do the mentioned change command in the monitor if the guest doesnt > have an eject command), but other than that a passed cd drive should > now work like it now does on Linux hosts. I also did some fixes > to the scsi cdrom emulation while I was at it (eject/load command > support was missing too), so passing a cd drive as > -drive if=scsi,media=cdrom,file=/dev/acd0 (or also /dev/cd0) should now > work as well. (It is still readonly tho, scsi passthru support would be > the next thing to be ported i.e. so you could do file=/dev/pass0...) > [snip patch] It's a little late now, since you've already updated the port, but... I can use eject in my openSUSE client to open the drawer of the DVD drive. I also noticed that I don't need to use notsc to boot openSUSE with the latest version of the qemu port :) --- Gary Jennejohn From bsdgroup.md at gmail.com Wed Mar 25 18:10:01 2009 From: bsdgroup.md at gmail.com (Rusu Silviu) Date: Wed Mar 25 18:10:07 2009 Subject: linuxnero does not see SATA recorder Message-ID: On a 7.1 RELEASE have an Optiark SATA DVD recorder using it as /dev/cd0 with growisofs have linux_base-fc-4_14 installed nero is linux-nero-3.5.0.1 have proc mounted by linproc /compat/linux/proc linprocfs rw 0 0 have linux enabled in rc.conf and it is kld loaded i can start nero, introducing SN, create projects etc but it does not see my recorder running it under root(not sudo) at the start it is complaining the /proc is not mounted, when in fact it is /compat/linux/proc exists and have util info as cpuinfo, meminfo etc tried to ln -s /compat/linux/proc / same complaints at startup the /compat/linux/proc/scsi/scsi file is empty tried to ln -s /dev/cd0 /dev/sg0 also tried ln -s /dev/cd0 /compat/linux/dev/sg0 nothing ... any ideas? thank you From Alexander at Leidinger.net Thu Mar 26 00:03:25 2009 From: Alexander at Leidinger.net (Alexander Leidinger) Date: Thu Mar 26 00:03:31 2009 Subject: linuxnero does not see SATA recorder In-Reply-To: References: Message-ID: <20090326080314.10727gsz1jw5szs4@webmail.leidinger.net> Quoting Rusu Silviu (from Thu, 26 Mar 2009 02:36:53 +0200): > On a 7.1 RELEASE have an Optiark SATA DVD recorder > using it as /dev/cd0 with growisofs > > have linux_base-fc-4_14 installed > nero is linux-nero-3.5.0.1 > > have proc mounted by > linproc /compat/linux/proc linprocfs rw 0 0 > > have linux enabled in rc.conf and it is kld loaded > > i can start nero, introducing SN, create projects etc > but it does not see my recorder > running it under root(not sudo) > > at the start it is complaining the /proc is not mounted, when in fact it is > /compat/linux/proc exists and have util info as cpuinfo, meminfo etc Plese run nero within ktrace and provide the output of linux_kdump. MAybe it is trying to access some /proc entries, which are not emulated. > tried to ln -s /compat/linux/proc / > same complaints at startup > > the /compat/linux/proc/scsi/scsi file is empty Can you provide an example what this file should contain in your case? I think I remember that there's some linux scsi compat stuff somewhere, but it needs to be enabled by hand. I can not look it up ATM, but have a look at the NOTES files if there's something related. Bye, Alexander. -- NOTICE: Anyone seen smoking will be assumed to be on fire and will be summarily put out. http://www.Leidinger.net Alexander @ Leidinger.net: PGP ID = B0063FE7 http://www.FreeBSD.org netchild @ FreeBSD.org : PGP ID = 72077137 From kalinoj1 at iem.pw.edu.pl Thu Mar 26 02:12:09 2009 From: kalinoj1 at iem.pw.edu.pl (=?UTF-8?Q?J=C4=99drzej_Kalinowski?=) Date: Thu Mar 26 02:12:15 2009 Subject: qemu 0.10.1 PCAP networking on CURRENT Message-ID: <962f2e5dc273dac6f2a2aaf017763bb2@iem.pw.edu.pl> Hi, I'm working with latest qemu port (0.10.1) submitted by Juergen Lock. Pcap networking didn't work ok for me - it was terribly slow. I'm on a 8.0-CURRENT system, so I don't know if it is also the case in older releases. Nevertheless, if someone experiences similiar problems, you can try this patch - it helped me: --- /usr/ports/emulators/qemu/files/pcap-patch.orig 2009-03-08 18:09:43.000000000 +0100 +++ /usr/local/emulators/qemu/files/pcap-patch 2009-03-26 09:16:43.000000000 +0100 @@ -87,7 +87,7 @@ echo "CONFIG_SLIRP=yes" >> $config_mak echo "#define CONFIG_SLIRP 1" >> $config_h Index: net.c -@@ -105,6 +105,13 @@ +@@ -105,6 +105,15 @@ #include "qemu_socket.h" @@ -95,7 +95,9 @@ +#if defined(_WIN32) +#define WPCAP 1 +#endif ++#define PCAP_DONT_INCLUDE_PCAP_BPF_H +#include ++#include +#endif + #if defined(CONFIG_SLIRP) -- JK From personrp at UPMC.EDU Thu Mar 26 05:49:28 2009 From: personrp at UPMC.EDU (Person, Roderick) Date: Thu Mar 26 05:49:35 2009 Subject: linuxnero does not see SATA recorder In-Reply-To: References: Message-ID: <1AE59099C6D80E41BEB64A1768AFB4EA0A75C060@msxmbxnsprd18.acct.upmchs.net> -----Original Message----- > On a 7.1 RELEASE have an Optiark SATA DVD recorder using it as /dev/cd0 with growisofs > > have linux_base-fc-4_14 installed > nero is linux-nero-3.5.0.1 >any ideas? I don't think this has anything to do with being a SATA drive. I have tried linuxnero with SCSI and IDE drives and I have never been able to make it see any drive. I gave up on it over a year ago. Rod Person Sr. Programmer (412)454-2616 http://www.ccbh.com "And of all the passions the most mischievous is pleasure. Why so? Because all things are the slave of pleasure; and because the life of the wicked is governed by pleasure as by a master." - Philo The Second Book of the Treatise on The Allegories of the Sacred Laws, after the Work of the Six Days of Creation. From vinnix.bsd at gmail.com Thu Mar 26 07:20:14 2009 From: vinnix.bsd at gmail.com (=?ISO-8859-1?Q?Vin=EDcius_Abrah=E3o?=) Date: Thu Mar 26 07:20:21 2009 Subject: Problem when compiling kernel Message-ID: <1e31c7980903260656w3b8a93f5xf4e6a24255dbf98d@mail.gmail.com> Hi guys, I have a problem when I trying to compile my kernel: (...) /usr/src/sys/amd64/linux32/linux32_sysvec.c: In function 'exec_linux_setregs': /usr/src/sys/amd64/linux32/linux32_sysvec.c:835: error: '__LINUX_NPXCW__' undeclared (first use in this function) /usr/src/sys/amd64/linux32/linux32_sysvec.c:835: error: (Each undeclared identifier is reported only once /usr/src/sys/amd64/linux32/linux32_sysvec.c:835: error: for each function it appears in.) *** Error code 1 Stop in /usr/obj/usr/src/sys/VINNIX. *** Error code 1 Stop in /usr/src. *** Error code 1 My FreeBSD is: FreeBSD vinnix.corp.triarius.com.br 7.1-STABLE FreeBSD 7.1-STABLE #54: Thu Mar 5 19:50:49 BRT 2009 root@vinnix.corp.triarius.com.br:/usr/obj/usr/src/sys/VINNIX amd64 My kernel config file: http://pastebin.com/m2c74fbf1 And I run cvsup yesterday! Thanks!, Vinnix From jkim at FreeBSD.org Thu Mar 26 09:49:09 2009 From: jkim at FreeBSD.org (Jung-uk Kim) Date: Thu Mar 26 09:49:15 2009 Subject: qemu 0.10.1 PCAP networking on CURRENT In-Reply-To: <962f2e5dc273dac6f2a2aaf017763bb2@iem.pw.edu.pl> References: <962f2e5dc273dac6f2a2aaf017763bb2@iem.pw.edu.pl> Message-ID: <200903261249.00104.jkim@FreeBSD.org> [CC added] On Thursday 26 March 2009 04:55 am, J?drzej Kalinowski wrote: > Hi, > > I'm working with latest qemu port (0.10.1) submitted by Juergen > Lock. Pcap networking didn't work ok for me - it was terribly slow. > > I'm on a 8.0-CURRENT system, so I don't know if it is also the case > in older releases. Nevertheless, if someone experiences similiar > problems, you can try this patch - it helped me: --- (PATCH removed) --- libpcap 1.0.0 was imported few days ago but it didn't merge the following local hack: http://www.freebsd.org/cgi/cvsweb.cgi/src/contrib/libpcap/pcap.h.diff?r1=1.13;r2=1.14 Your patch is harmless but it should be fixed in pcap.h and/or net/bpf.h properly not to step on each other unless the plan is to add PCAP_DONT_INCLUDE_PCAP_BPF_H everywhere in our ports tree to get FreeBSD-specific definitions. For example, the QEMU patch needs at least two ioctl definitions to get it really working, i.e., BIOCIMMEDIATE and BIOCFEEDBACK, which are only present in net/bpf.h. In fact, pcap-bpf.c itself has a hack for AIX like this: #ifdef _AIX /* * Make "pcap.h" not include "pcap/bpf.h"; we are going to include the * native OS version, as we need "struct bpf_config" from it. */ #define PCAP_DONT_INCLUDE_PCAP_BPF_H #include /* * Prevent bpf.h from redefining the DLT_ values to their * IFT_ values, as we're going to return the standard libpcap * values, not IBM's non-standard IFT_ values. */ #undef _AIX #include #define _AIX ... #else /* _AIX */ #include #endif /* _AIX */ This is absolutely disgusting. IMHO, I don't think it is a good idea to pollute ports tree with local patches like that. Jung-uk Kim From rdivacky at freebsd.org Thu Mar 26 13:38:29 2009 From: rdivacky at freebsd.org (Roman Divacky) Date: Thu Mar 26 13:38:36 2009 Subject: Problem when compiling kernel In-Reply-To: <1e31c7980903260656w3b8a93f5xf4e6a24255dbf98d@mail.gmail.com> References: <1e31c7980903260656w3b8a93f5xf4e6a24255dbf98d@mail.gmail.com> Message-ID: <20090326203750.GA15634@freebsd.org> On Thu, Mar 26, 2009 at 10:56:28AM -0300, Vin?cius Abrah?o wrote: > Hi guys, I have a problem when I trying to compile my kernel: > > (...) > /usr/src/sys/amd64/linux32/linux32_sysvec.c: In function > 'exec_linux_setregs': > /usr/src/sys/amd64/linux32/linux32_sysvec.c:835: error: '__LINUX_NPXCW__' > undeclared (first use in this function) > /usr/src/sys/amd64/linux32/linux32_sysvec.c:835: error: (Each undeclared > identifier is reported only once > /usr/src/sys/amd64/linux32/linux32_sysvec.c:835: error: for each function it > appears in.) > *** Error code 1 > > Stop in /usr/obj/usr/src/sys/VINNIX. > *** Error code 1 > > Stop in /usr/src. > *** Error code 1 > > My FreeBSD is: > FreeBSD vinnix.corp.triarius.com.br 7.1-STABLE FreeBSD 7.1-STABLE #54: Thu > Mar 5 19:50:49 BRT 2009 > root@vinnix.corp.triarius.com.br:/usr/obj/usr/src/sys/VINNIX > amd64 > > My kernel config file: > http://pastebin.com/m2c74fbf1 > > And I run cvsup yesterday! this was already fixed, please recvsup and it should just work From avg at icyb.net.ua Fri Mar 27 05:10:22 2009 From: avg at icyb.net.ua (Andriy Gapon) Date: Fri Mar 27 05:10:29 2009 Subject: qemu+kqemu: hard lockup with windows guest (32-bit on 64-bit) Message-ID: <49CCBE43.5080309@icyb.net.ua> System: stable/7 amd64 r190321 qemu-0.10.0_1 kqemu-kmod-devel-1.4.0.p1_1 I experience rare complete system lockups while running windows (98, win2k) guests with kqemu enabled. The guests are obviously 32-bit. I can't provide any diagnostics because it's a total freeze and there is no useful info after rebooting. -- Andriy Gapon From dchagin at freebsd.org Fri Mar 27 05:59:01 2009 From: dchagin at freebsd.org (Chagin Dmitry) Date: Fri Mar 27 05:59:07 2009 Subject: qemu+kqemu: hard lockup with windows guest (32-bit on 64-bit) In-Reply-To: <49CCBE43.5080309@icyb.net.ua> References: <49CCBE43.5080309@icyb.net.ua> Message-ID: <20090327125852.GA6042@dchagin.static.corbina.ru> On Fri, Mar 27, 2009 at 01:53:39PM +0200, Andriy Gapon wrote: > > System: > stable/7 amd64 r190321 > qemu-0.10.0_1 > kqemu-kmod-devel-1.4.0.p1_1 > > I experience rare complete system lockups while running windows (98, win2k) guests > with kqemu enabled. The guests are obviously 32-bit. > I can't provide any diagnostics because it's a total freeze and there is no useful > info after rebooting. > SW_WATCHDOG can help you with debugging of similar bugs. -- Have fun! chd -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 196 bytes Desc: not available Url : http://lists.freebsd.org/pipermail/freebsd-emulation/attachments/20090327/12092d7e/attachment.pgp From avg at icyb.net.ua Fri Mar 27 06:17:33 2009 From: avg at icyb.net.ua (Andriy Gapon) Date: Fri Mar 27 06:17:40 2009 Subject: qemu+kqemu: hard lockup with windows guest (32-bit on 64-bit) In-Reply-To: <20090327125852.GA6042@dchagin.static.corbina.ru> References: <49CCBE43.5080309@icyb.net.ua> <20090327125852.GA6042@dchagin.static.corbina.ru> Message-ID: <49CCD1E9.2060307@icyb.net.ua> on 27/03/2009 14:58 Chagin Dmitry said the following: > On Fri, Mar 27, 2009 at 01:53:39PM +0200, Andriy Gapon wrote: >> System: >> stable/7 amd64 r190321 >> qemu-0.10.0_1 >> kqemu-kmod-devel-1.4.0.p1_1 >> >> I experience rare complete system lockups while running windows (98, win2k) guests >> with kqemu enabled. The guests are obviously 32-bit. >> I can't provide any diagnostics because it's a total freeze and there is no useful >> info after rebooting. >> > > SW_WATCHDOG can help you with debugging of similar bugs. It's an interesting idea. Now that you mentioned it I realized something else - I have a hardware watchdog and it is tested to work, but this time it didn't fire. So either watchdogd process was able to keep running or even chipset for so screwed so badly that it could not reset system. I have hard time believing in the second case, but the first one is also under suspicion - system didn't respond to any input (even power button press), it couldn't be pinged, etc. -- Andriy Gapon From ktullavik at gmail.com Fri Mar 27 13:33:54 2009 From: ktullavik at gmail.com (Kjell Tore Ullavik) Date: Fri Mar 27 13:34:00 2009 Subject: wine, mixer pcm volume reset to 0:0 Message-ID: <49CD3288.7090102@gmail.com> Hello! I played around with wine-1.1.17. I don't know why it's marked ignore, it works for me on CURRENT (with windows firefox and flash 10 among other things) Anyway, I dicovered a regression since wine-1.1.0. When using music players like Winamp and Spotify, the mixer pcm volume keeps beeing reset to 0:0 between each song beeing played. If I do 'mixer pcm 100:100' I can hear the rest of the song, before volume is set to 0:0 again. Affected systems: FreeBSD 8 CURRENT wine-1.1.17 FreeBSD 7 STABLE from july. wine-1.1.16 Not affected: FreeBSD 7 STABLE from july wine-1.1.0 Applications tested: Winamp 5 Lite 5.551 Spotify 0.3.11 From ktullavik at gmail.com Fri Mar 27 18:39:16 2009 From: ktullavik at gmail.com (Kjell Tore Ullavik) Date: Fri Mar 27 18:39:22 2009 Subject: wine, mixer pcm volume reset to 0:0 In-Reply-To: <49CD3288.7090102@gmail.com> References: <49CD3288.7090102@gmail.com> Message-ID: <49CD7FB0.4000807@gmail.com> Kjell Tore Ullavik wrote: > Hello! > > I played around with wine-1.1.17. I don't know why it's marked ignore, > it works for me on CURRENT (with windows firefox > and flash 10 among other things) > > Anyway, I dicovered a regression since wine-1.1.0. When using music > players like Winamp and Spotify, the mixer pcm > volume keeps beeing reset to 0:0 between each song beeing played. If I > do 'mixer pcm 100:100' > I can hear the rest of the song, before volume is set to 0:0 again. > > Affected systems: > FreeBSD 8 CURRENT > wine-1.1.17 > > FreeBSD 7 STABLE from july. > wine-1.1.16 > > Not affected: > FreeBSD 7 STABLE from july > wine-1.1.0 > > Applications tested: > Winamp 5 Lite 5.551 > Spotify 0.3.11 > > Further testing shows this regression was introduced with wine-1.1.6, and is still present in wine-1.1.18 (released yesterday). From tijl at ulyssis.org Sat Mar 28 03:02:12 2009 From: tijl at ulyssis.org (Tijl Coosemans) Date: Sat Mar 28 03:02:19 2009 Subject: wine, mixer pcm volume reset to 0:0 In-Reply-To: <49CD3288.7090102@gmail.com> References: <49CD3288.7090102@gmail.com> Message-ID: <200903281051.49122.tijl@ulyssis.org> On Friday 27 March 2009 21:09:44 Kjell Tore Ullavik wrote: > I played around with wine-1.1.17. I don't know why it's marked > ignore, it works for me on CURRENT (with windows firefox and flash > 10 among other things) > > Anyway, I dicovered a regression since wine-1.1.0. When using music > players like Winamp and Spotify, the mixer pcm volume keeps beeing > reset to 0:0 between each song beeing played. If I do 'mixer pcm > 100:100' I can hear the rest of the song, before volume is set to > 0:0 again. What if you set it to 90:90? From gelraen.ua at gmail.com Sat Mar 28 05:17:01 2009 From: gelraen.ua at gmail.com (Maxim Ignatenko) Date: Sat Mar 28 05:17:07 2009 Subject: wine, mixer pcm volume reset to 0:0 In-Reply-To: <200903281051.49122.tijl@ulyssis.org> References: <49CD3288.7090102@gmail.com> <200903281051.49122.tijl@ulyssis.org> Message-ID: 2009/3/28 Tijl Coosemans : > On Friday 27 March 2009 21:09:44 Kjell Tore Ullavik wrote: >> I played around with wine-1.1.17. I don't know why it's marked >> ignore, it works for me on CURRENT (with windows firefox and flash >> 10 among other things) >> >> Anyway, I dicovered a regression since wine-1.1.0. When using music >> players like Winamp and Spotify, the mixer pcm volume keeps beeing >> reset to 0:0 between each song beeing played. If I do 'mixer pcm >> 100:100' I can hear the rest of the song, before volume is set to >> 0:0 again. > > What if you set it to 90:90? > _______________________________________________ > freebsd-emulation@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-emulation > To unsubscribe, send any mail to "freebsd-emulation-unsubscribe@freebsd.org" > It doesn't important what exact value is, it's reset to 0 when running application do something except just playing single stream (it is my impression). I have the same issue on -CUREENT, and it seems to appear only with OSS sound driver in wine. With JACK all works normally (except crashes with "err:ntdll:RtlpWaitForCriticalSection section 0x146c44 "dsound.c: DirectSoundDevice.mixlock" wait timed out in thread 002e, blocked by 0000, retrying (60 sec)"), but I failed to get any sound with NAS, while native NAS applications (i.e. games/qnetwalk) works fine. From sven.schellack at gmx.de Sat Mar 28 05:28:59 2009 From: sven.schellack at gmx.de (Sven Schellack) Date: Sat Mar 28 05:29:06 2009 Subject: wine, mixer pcm volume reset to 0:0 In-Reply-To: <49CD7FB0.4000807@gmail.com> References: <49CD3288.7090102@gmail.com> <49CD7FB0.4000807@gmail.com> Message-ID: <20090328120217.319080@gmx.net> -------- Original-Nachricht -------- > Datum: Sat, 28 Mar 2009 02:38:56 +0100 > Von: Kjell Tore Ullavik > An: emulation@freebsd.org, gerald@FreeBSD.org > Betreff: Re: wine, mixer pcm volume reset to 0:0 > Kjell Tore Ullavik wrote: > > Hello! > > > > I played around with wine-1.1.17. I don't know why it's marked ignore, > > it works for me on CURRENT (with windows firefox > > and flash 10 among other things) > > > > Anyway, I dicovered a regression since wine-1.1.0. When using music > > players like Winamp and Spotify, the mixer pcm > > volume keeps beeing reset to 0:0 between each song beeing played. If I > > do 'mixer pcm 100:100' > > I can hear the rest of the song, before volume is set to 0:0 again. > > > > Affected systems: > > FreeBSD 8 CURRENT > > wine-1.1.17 > > > > FreeBSD 7 STABLE from july. > > wine-1.1.16 > > > > Not affected: > > FreeBSD 7 STABLE from july > > wine-1.1.0 > > > > Applications tested: > > Winamp 5 Lite 5.551 > > Spotify 0.3.11 > > > > > Further testing shows this regression was introduced with wine-1.1.6, > and is still present in wine-1.1.18 (released yesterday). Hi, sounds a bit like http://bugs.winehq.org/show_bug.cgi?id=15710 - I could reproduce this behaviour with several games. World of Warcraft for example starts muted and gets muted again each time you'll change to another zone or re-log with a different character. -- Psssst! Schon vom neuen GMX MultiMessenger geh?rt? Der kann`s mit allen: http://www.gmx.net/de/go/multimessenger01 From nox at jelal.kn-bremen.de Sat Mar 28 11:30:51 2009 From: nox at jelal.kn-bremen.de (Juergen Lock) Date: Sat Mar 28 11:30:58 2009 Subject: qemu+kqemu: hard lockup with windows guest (32-bit on 64-bit) In-Reply-To: <49CCD1E9.2060307@icyb.net.ua> References: <49CCBE43.5080309@icyb.net.ua> <20090327125852.GA6042@dchagin.static.corbina.ru> <49CCD1E9.2060307@icyb.net.ua> Message-ID: <20090328182909.GA90582@saturn.kn-bremen.de> On Fri, Mar 27, 2009 at 03:17:29PM +0200, Andriy Gapon wrote: > on 27/03/2009 14:58 Chagin Dmitry said the following: > > On Fri, Mar 27, 2009 at 01:53:39PM +0200, Andriy Gapon wrote: > >> System: > >> stable/7 amd64 r190321 > >> qemu-0.10.0_1 > >> kqemu-kmod-devel-1.4.0.p1_1 > >> > >> I experience rare complete system lockups while running windows (98, win2k) guests > >> with kqemu enabled. The guests are obviously 32-bit. > >> I can't provide any diagnostics because it's a total freeze and there is no useful > >> info after rebooting. > >> > > > > SW_WATCHDOG can help you with debugging of similar bugs. > > It's an interesting idea. > Now that you mentioned it I realized something else - I have a hardware watchdog > and it is tested to work, but this time it didn't fire. So either watchdogd > process was able to keep running or even chipset for so screwed so badly that it > could not reset system. I have hard time believing in the second case, but the > first one is also under suspicion - system didn't respond to any input (even power > button press), it couldn't be pinged, etc. Is that an smp box and do the hangs happen soon after starting up qemu? If yes there's a remote chance your box doesn't like the ltr that the remnants of the kqemu amd64 tss fixup code still do once on cpu 0 - I've just updated the kqemu ports to only add that code on OSVERSIONs where its actually needed, so you could try updating your kqemu. (I wanted to do that soon after 7.1 went out already but somehow forgot about it again until I saw this thread...) HTH, Juergen From bugmaster at FreeBSD.org Mon Mar 30 04:06:51 2009 From: bugmaster at FreeBSD.org (FreeBSD bugmaster) Date: Mon Mar 30 04:07:36 2009 Subject: Current problem reports assigned to freebsd-emulation@FreeBSD.org Message-ID: <200903301106.n2UB6oth054699@freefall.freebsd.org> Note: to view an individual PR, use: http://www.freebsd.org/cgi/query-pr.cgi?pr=(number). The following is a listing of current problems submitted by FreeBSD users. These represent problem reports covering all versions including experimental development code and obsolete releases. S Tracker Resp. Description -------------------------------------------------------------------------------- o kern/130724 emulation [linprocfs] [patch] cpuinfo in linprocfs is dated, cau o kern/129169 emulation [linux] [patch] Linux Emulation ENOTCONN error using n f ports/127018 emulation Linuxulator incapable of using FreeBSD's LDAP environm o kern/126232 emulation [linux] Linux ioctl TCGETS (0x5401) always fails o ports/121800 emulation x11-toolkits/linux-openmotif - OpenMotif upgrade to 2. o kern/97326 emulation [linux] file descriptor leakage in linux emulation o ports/91318 emulation [fix] graphics/linux_dri: works on amd64 too o kern/91293 emulation [svr4] [patch] *Experimental* Update to the SVR4 emula o kern/73777 emulation [linux] [patch] linux emulation: root dir special hand a kern/72920 emulation [linux]: path "prefixing" is not done on unix domain s o kern/41543 emulation [patch] [request] easier wine/w23 support o kern/39201 emulation [linux] [patch] ptrace(2) and rfork(RFLINUXTHPN) confu o kern/29698 emulation [linux] [patch] linux ipcs doesn'work o kern/21463 emulation [linux] Linux compatability mode should not allow setu o kern/11165 emulation [ibcs2] IBCS2 doesn't work correctly with PID_MAX 9999 15 problems total. From avg at icyb.net.ua Mon Mar 30 07:57:14 2009 From: avg at icyb.net.ua (Andriy Gapon) Date: Mon Mar 30 07:57:55 2009 Subject: qemu+kqemu: hard lockup with windows guest (32-bit on 64-bit) In-Reply-To: <20090328182909.GA90582@saturn.kn-bremen.de> References: <49CCBE43.5080309@icyb.net.ua> <20090327125852.GA6042@dchagin.static.corbina.ru> <49CCD1E9.2060307@icyb.net.ua> <20090328182909.GA90582@saturn.kn-bremen.de> Message-ID: <49D0DDC7.7070600@icyb.net.ua> on 28/03/2009 20:29 Juergen Lock said the following: > > Is that an smp box and do the hangs happen soon after starting up qemu? Yes and almost yes. Several times it happened soon after qemu start and one other time it happened after a while after qemu start, but I was installing OS using qemu then. > If yes there's a remote chance your box doesn't like the ltr that the > remnants of the kqemu amd64 tss fixup code still do once on cpu 0 - I've > just updated the kqemu ports to only add that code on OSVERSIONs where its > actually needed, so you could try updating your kqemu. (I wanted to do > that soon after 7.1 went out already but somehow forgot about it again > until I saw this thread...) OK, I am now running with kqemu-kmod-devel-1.4.0.p1_2 and qemu-0.10.1_1. Let's see what happens. -- Andriy Gapon From me at janh.de Tue Mar 31 04:20:17 2009 From: me at janh.de (Jan Henrik Sylvester) Date: Tue Mar 31 04:20:24 2009 Subject: acroread9 misses libxml2.so.2 Message-ID: <49D1F976.7000708@janh.de> I just installed acroread9 in parallel to acroread8, but I cannot use it: /usr/local/Adobe/Reader9/ENU/Adobe/Reader9/Reader/intellinux/bin/acroread: error while loading shared libraries: libxml2.so.2: cannot open shared object file: No such file or directory Starting acroread8 still works fine. I am on 7.1-RELEASE-p4 using compat.linux.osrelease=2.4.2 with the following linux_base dependent packages installed: acroread8-8.1.4_1 acroread9-9.1.0_1 google-earth-5.0.11337.1968 linux-alsa-lib-1.0.10.3 linux-atk-1.9.1_1 linux-cairo-1.0.2 linux-expat-1.95.8 linux-flashplugin-9.0r159 linux-fontconfig-2.2.3_7 linux-gtk2-2.6.10_1 linux-hicolor-icon-theme-0.5_1 linux-jpeg-6b.34 linux-libsigc-2.0.17 linux-nvu-1.0_1 linux-openssl-0.9.7f linux-pango-1.10.2_1 linux-png-1.2.8_2 linux-scim-gtk-fc4-1.4.4 linux-scim-libs-fc4-1.4.4_1 linux-tiff-3.7.1 linux-xorg-libs-6.8.2_5 linux_base-fc-4_14 linux_dri-7.0 nspluginwrapper-1.2.2_2 skype-2.0.0.72,1 `find /usr/compat/linux/ -name "*libxml*"` returns nothing. Does acroread9 need a newer linux_base? The port did not complain. Cheers, Jan Henrik From ktullavik at gmail.com Tue Mar 31 12:45:24 2009 From: ktullavik at gmail.com (Kjell Tore Ullavik) Date: Tue Mar 31 12:45:31 2009 Subject: acroread9 misses libxml2.so.2 In-Reply-To: <49D1F976.7000708@janh.de> References: <49D1F976.7000708@janh.de> Message-ID: <49D26B4E.6070508@gmail.com> You need to install textproc/linux-libxml, probably also graphics/linux-gdk-pixbuf, it still charshes for me with linux-base8 though. Bug report here: http://www.freebsd.org/cgi/query-pr.cgi?pr=133194 Jan Henrik Sylvester wrote: > I just installed acroread9 in parallel to acroread8, but I cannot use it: > > /usr/local/Adobe/Reader9/ENU/Adobe/Reader9/Reader/intellinux/bin/acroread: > error while loading shared libraries: libxml2.so.2: cannot open shared > object file: No such file or directory > > Starting acroread8 still works fine. > > I am on 7.1-RELEASE-p4 using compat.linux.osrelease=2.4.2 with the > following linux_base dependent packages installed: > > acroread8-8.1.4_1 > acroread9-9.1.0_1 > google-earth-5.0.11337.1968 > linux-alsa-lib-1.0.10.3 > linux-atk-1.9.1_1 > linux-cairo-1.0.2 > linux-expat-1.95.8 > linux-flashplugin-9.0r159 > linux-fontconfig-2.2.3_7 > linux-gtk2-2.6.10_1 > linux-hicolor-icon-theme-0.5_1 > linux-jpeg-6b.34 > linux-libsigc-2.0.17 > linux-nvu-1.0_1 > linux-openssl-0.9.7f > linux-pango-1.10.2_1 > linux-png-1.2.8_2 > linux-scim-gtk-fc4-1.4.4 > linux-scim-libs-fc4-1.4.4_1 > linux-tiff-3.7.1 > linux-xorg-libs-6.8.2_5 > linux_base-fc-4_14 > linux_dri-7.0 > nspluginwrapper-1.2.2_2 > skype-2.0.0.72,1 > > `find /usr/compat/linux/ -name "*libxml*"` returns nothing. > > Does acroread9 need a newer linux_base? The port did not complain. > > Cheers, > Jan Henrik > _______________________________________________ > freebsd-emulation@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-emulation > To unsubscribe, send any mail to > "freebsd-emulation-unsubscribe@freebsd.org" > From dchagin at freebsd.org Tue Mar 31 13:15:41 2009 From: dchagin at freebsd.org (Chagin Dmitry) Date: Tue Mar 31 13:15:47 2009 Subject: acroread9 misses libxml2.so.2 In-Reply-To: <49D26B4E.6070508@gmail.com> References: <49D1F976.7000708@janh.de> <49D26B4E.6070508@gmail.com> Message-ID: <20090331201533.GA10698@dchagin.static.corbina.ru> On Tue, Mar 31, 2009 at 09:13:18PM +0200, Kjell Tore Ullavik wrote: > > > You need to install textproc/linux-libxml, probably also > graphics/linux-gdk-pixbuf, > it still charshes for me with linux-base8 though. Bug report here: > > http://www.freebsd.org/cgi/query-pr.cgi?pr=133194 > > looks like new acroread depends on linux inotify API, which is not implemented in our linuxulator... thnx, let's try to implement. -- Have fun! chd -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 196 bytes Desc: not available Url : http://lists.freebsd.org/pipermail/freebsd-emulation/attachments/20090331/6014ee24/attachment.pgp From rdivacky at freebsd.org Tue Mar 31 13:57:01 2009 From: rdivacky at freebsd.org (Roman Divacky) Date: Tue Mar 31 13:57:08 2009 Subject: acroread9 misses libxml2.so.2 In-Reply-To: <20090331201533.GA10698@dchagin.static.corbina.ru> References: <49D1F976.7000708@janh.de> <49D26B4E.6070508@gmail.com> <20090331201533.GA10698@dchagin.static.corbina.ru> Message-ID: <20090331205615.GA32160@freebsd.org> On Wed, Apr 01, 2009 at 12:15:33AM +0400, Chagin Dmitry wrote: > On Tue, Mar 31, 2009 at 09:13:18PM +0200, Kjell Tore Ullavik wrote: > > > > > > You need to install textproc/linux-libxml, probably also > > graphics/linux-gdk-pixbuf, > > it still charshes for me with linux-base8 though. Bug report here: > > > > http://www.freebsd.org/cgi/query-pr.cgi?pr=133194 > > > > > > looks like new acroread depends on linux inotify API, which is not > implemented in our linuxulator... thnx, let's try to implement. if you are interseted you can finish my epoll() work... From sfourman at gmail.com Tue Mar 31 15:41:14 2009 From: sfourman at gmail.com (Sam Fourman Jr.) Date: Tue Mar 31 15:41:21 2009 Subject: [PATCH] futexes, flash9 related In-Reply-To: <11167f520902090511i52adaa7dm9c44025578aecf2c@mail.gmail.com> References: <20081221174939.GA33531@dchagin.dialup.corbina.ru> <747dc8f30902050220y7b1d726bj9f6f83afa843b520@mail.gmail.com> <20090206063550.GA2123@dchagin.static.corbina.ru> <1233919189.1866.13.camel@localhost> <20090206122354.GA5670@dchagin.static.corbina.ru> <1233923734.1866.33.camel@localhost> <20090208200541.GA59299@dchagin.static.corbina.ru> <747dc8f30902090501l58979723l8e0de51482131991@mail.gmail.com> <11167f520902090511i52adaa7dm9c44025578aecf2c@mail.gmail.com> Message-ID: <11167f520903311510x72e61c21sd4c639582b96c52@mail.gmail.com> On Mon, Feb 9, 2009 at 8:11 AM, Sam Fourman Jr. wrote: >>> please, try http://lnxx64.googlecode.com/files/futexes_partial_II.patch >>> first patch fail at i386... > > What is the syntax to apply this patch to -CURRENT, also can this > patch be applied to i386 as well as amd64? > > Sam > Does anyone know if this patch was ever committed to CURRENT? Sam Fourman Jr.