Re: [gentoo-user] Kernel upgrading and linux symlink
Am Samstag 31 Oktober 2009 17:52:52 schrieb Harry Putnam: > And in fact does it really matter if its pointing at the newly > installed or actual running kernel, when kernel compiling operations > take place? No, it doesn't really matter at all. AFAIK, kernel devs even recommend against the symlink. And you can do just fine w/o it. I need to compile one external module, nvidia. This can be done by providing some ENV variables which point to the correct kernel source: KBUILD_OUTPUT=/lib/modules/`uname -r`/build KERNEL_DIR=/lib/modules/`uname - r`/source paludis -i1 nvidia-drivers This points to the currently running kernel. You can easily replace "`uname - r`" with any version you want. Why do I do this? Because I always use the sources from kernel.org directly, and get them via git. I have cloned the kernel tree in /usr/src/linux-2.6, so that I can easily switch between different version by means of a simple "git checkout". The build output is stored in /usr/src/build-, by using "make O=../build-2.6.31 menuconfig", for example. HTH... Dirk signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] emerge ignores /etcportage/package.mask-RESOLVED
Maxim Wexler wrote: >> Do you have a space in front of the line or something else that may >> interfere with package.mask ? >> >> You sure you have the file as package.mask ? You know, spelled right >> and all? I mention because it sounds like something I would do. lol >> >> Dale >> > > I guess I shouldn't feel so embarrassed then. What a stupid question, > I thought. Of course I called it package.mask. What else would I call > it? So I checked. I called portage.mask :( > > mw > > > I couldn't think of any other reason for it not to work after someone else mentioned it being in package.unmask. At least I am good for something. ;-) Dale :-) :-)
Re: [gentoo-user] kernel 2.6.31 + nvidia
Confirmed On Sat, Oct 31, 2009 at 8:19 PM, Philip Webb wrote: > Just in case anyone else runs into this problem, which seems solved: > Nvidia-drivers 180.60 won't compile with Gentoo Kernel 2.6.31-r4 . > The problem is fixed by moving up to N-d 185.18.36-r1 (also "testing"). > > I've just rebooted & apps seem to start noticeably faster on Fluxbox. > > -- > ,, > SUPPORT ___//___, Philip Webb > ELECTRIC /] [] [] [] [] []| Cities Centre, University of Toronto > TRANSIT`-O--O---' purslowatchassdotutorontodotca > > >
Re: [gentoo-user] emerge ignores /etcportage/package.mask-RESOLVED
> Do you have a space in front of the line or something else that may > interfere with package.mask ? > > You sure you have the file as package.mask ? You know, spelled right > and all? I mention because it sounds like something I would do. lol > > Dale I guess I shouldn't feel so embarrassed then. What a stupid question, I thought. Of course I called it package.mask. What else would I call it? So I checked. I called portage.mask :( mw
Re: [gentoo-user] Re: depclean / db & php loop
On Sat, Oct 31, 2009 at 7:26 PM, walt wrote: > On 10/31/2009 04:42 PM, Mark Knecht wrote: >> ... >> If it matters this is a PPC Mac Mini used as a MythTV backend >> server... > > Oops, I forgot to ask: > > It may be very important -- but I don't have a clue what a MythTV > server does or what packages it depends on. > > Does your server use something like Apache/php to do the serving, > and do you really need php to do it? > Walt, Thanks for the info on the USE flag. I'll try that tomorrow. As for Myth and Apache, there is a MythWeb app that allows me to look at TV listings, what's recorded, what's coming up, etc. I don't even know what php is so I cannot say whether I need it or not. I suspect this is the sort of thing where a developer or whoever sets of basic profiles set something and maybe I could get rid of it if I knew what it did. It seems to be required by MythWeb and I don't seem to have any real USE flag control in MythWeb. To me Apache is just a tool, not an app, so I don't try to optimize how I use it but rather hope the Gentoo devs are making the right choices. Again, thanks! - Mark MacMini ~ # equery depends php [ Searching for packages depending on php... ] www-apps/mythweb-0.21_p17573 (=virtual/httpd-php-5*) MacMini ~ # MacMini ~ # emerge -pv mythweb apache These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] www-servers/apache-2.2.11-r2 USE="ldap ssl -debug -doc (-selinux) -sni -static -suexec -threads" APACHE2_MODULES="actions alias auth_basic authn_alias authn_anon authn_dbm authn_default authn_file authz_dbm authz_default authz_groupfile authz_host authz_owner authz_user autoindex cache dav dav_fs dav_lock deflate dir disk_cache env expires ext_filter file_cache filter headers include info log_config logio mem_cache mime mime_magic negotiation rewrite setenvif speling status unique_id userdir usertrack vhost_alias -asis -auth_digest -authn_dbd -cern_meta -charset_lite -dbd -dumpio -ident -imagemap -log_forensic -proxy -proxy_ajp -proxy_balancer -proxy_connect -proxy_ftp -proxy_http -substitute -version" APACHE2_MPMS="-event -itk -peruser -prefork -worker" 0 kB [ebuild R ] www-apps/mythweb-0.21_p17573 USE="-vhosts" 0 kB Total: 2 packages (2 reinstalls), Size of downloads: 0 kB MacMini ~ #
[gentoo-user] kernel 2.6.31 + nvidia
Just in case anyone else runs into this problem, which seems solved: Nvidia-drivers 180.60 won't compile with Gentoo Kernel 2.6.31-r4 . The problem is fixed by moving up to N-d 185.18.36-r1 (also "testing"). I've just rebooted & apps seem to start noticeably faster on Fluxbox. -- ,, SUPPORT ___//___, Philip Webb ELECTRIC /] [] [] [] [] []| Cities Centre, University of Toronto TRANSIT`-O--O---' purslowatchassdotutorontodotca
Re: [gentoo-user] Re: libsxlt and aclocal/autoconf failing during emerge
walt wrote: > On 10/31/2009 07:05 PM, Harry Putnam wrote: > >> ... >> /var/tmp/portage/dev-libs/libxslt-1.1.26/temp/autoconf.out >> * autoconf * >> * PWD: /var/tmp/portage/dev-libs/libxslt-1.1.26/work/libxslt-1.1.26 >> * autoconf >> >> configure.in:120: error: possibly undefined macro: AC_LIBTOOL_WIN32_DLL >> If this token and others are legitimate, please use m4_pattern_allow. >> See the Autoconf documentation. >> configure.in:121: error: possibly undefined macro: AM_PROG_LIBTOOL >> > > If I'm interpreting things correctly (unlikely) those two macros are defined > in /usr/share/libtool/libtld/aclocal.m4, which belongs to the libtool package. > (On my machine, sys-devel/libtool-2.2.6a). > > Do you have such a file/package on your machine? > > I was reluctant to post but I did a google search and although it appeared they were talking about Macs and such, they were talking about libtool and (e)make's friends. Could it be that there is a version mismatch between those three? Maybe one or two are newer than the others? I also noticed there were a couple other packages affected, which lead me to narrow my search a little more. Seems to be the same thoughts all the way around tho. Dale :-) :-)
[gentoo-user] Re: libsxlt and aclocal/autoconf failing during emerge
On 10/31/2009 07:05 PM, Harry Putnam wrote: > ... > /var/tmp/portage/dev-libs/libxslt-1.1.26/temp/autoconf.out > * autoconf * > * PWD: /var/tmp/portage/dev-libs/libxslt-1.1.26/work/libxslt-1.1.26 > * autoconf > > configure.in:120: error: possibly undefined macro: AC_LIBTOOL_WIN32_DLL > If this token and others are legitimate, please use m4_pattern_allow. > See the Autoconf documentation. > configure.in:121: error: possibly undefined macro: AM_PROG_LIBTOOL If I'm interpreting things correctly (unlikely) those two macros are defined in /usr/share/libtool/libtld/aclocal.m4, which belongs to the libtool package. (On my machine, sys-devel/libtool-2.2.6a). Do you have such a file/package on your machine?
[gentoo-user] Re: depclean / db & php loop
On 10/31/2009 04:42 PM, Mark Knecht wrote: > ... >If it matters this is a PPC Mac Mini used as a MythTV backend > server... Oops, I forgot to ask: It may be very important -- but I don't have a clue what a MythTV server does or what packages it depends on. Does your server use something like Apache/php to do the serving, and do you really need php to do it?
[gentoo-user] Re: depclean / db & php loop
On 10/31/2009 04:42 PM, Mark Knecht wrote: > ... > * sys-libs/db-4.6.21_p4 pulled in by: > * dev-lang/php-5.2.11 needs libdb-4.6.so > Calculating dependencies... done! > [ebuild R ] dev-lang/php-5.2.11 USE="apache2 berkdb bzip2 cgi cli ^^ Ah, the nomenclature is a bit confusing. libdb is supplied by the "Berkely" database package (the historical name) but portage calls the package "db". This "Berkely" distinguishes db from db2 (which is IBM, not Berkely). Anyway to get rid of the 'db' dependency you need to unset your berkdb USE flag and rebuild php. Very confusing!
[gentoo-user] Re: libsxlt and aclocal/autoconf failing during emerge
walt writes: > On 10/31/2009 06:03 PM, Harry Putnam wrote: >> Can anyone tell from the tail of emerge -vuD system >> what the trouble is with aclocal/autoconf and libxslt. >> ... >> * Failed Running autoconf ! >> * >> * Include in your bugreport the contents of: >> * /var/tmp/portage/dev-libs/libxslt-1.1.26/temp/autoconf.out > > Chances are that the file cited above will give a hint. > Does that file contain any error messages? Sorry, I hoped someone would have had the same problem and just recognize the tail messages. /var/tmp/portage/dev-libs/libxslt-1.1.26/temp/autoconf.out * autoconf * * PWD: /var/tmp/portage/dev-libs/libxslt-1.1.26/work/libxslt-1.1.26 * autoconf configure.in:120: error: possibly undefined macro: AC_LIBTOOL_WIN32_DLL If this token and others are legitimate, please use m4_pattern_allow. See the Autoconf documentation. configure.in:121: error: possibly undefined macro: AM_PROG_LIBTOOL They don't mean anything to me... do that to you?
[gentoo-user] Re: libsxlt and aclocal/autoconf failing during emerge
On 10/31/2009 06:03 PM, Harry Putnam wrote: > Can anyone tell from the tail of emerge -vuD system > what the trouble is with aclocal/autoconf and libxslt. > ... > * Failed Running autoconf ! > * > * Include in your bugreport the contents of: > * /var/tmp/portage/dev-libs/libxslt-1.1.26/temp/autoconf.out Chances are that the file cited above will give a hint. Does that file contain any error messages?
[gentoo-user] libsxlt and aclocal/autoconf failing during emerge
Can anyone tell from the tail of emerge -vuD system what the trouble is with aclocal/autoconf and libxslt. First it failed on aclocal... I re-emerged emerge -v sys-devel/aclocal-wrapper Tried again and it failed on autoconf emerge -v sys-devel/autoconf-wrapper But that didn't help it ... still fails at autoconf. [...] >>> Preparing source in >>> /var/tmp/portage/dev-libs/libxslt-1.1.26/work/libxslt-1.1.26 ... * Applying libxslt.m4-libxslt-1.1.26.patch ... [ ok ] * Applying libxslt-1.1.23-parallel-install.patch ... [ ok ] * Applying libxslt-1.1.26-undefined.patch ...[ ok ] * Running eautoreconf in '/var/tmp/portage/dev-libs/libxslt-1.1.26/work/libxslt-1.1.26' ... * Running aclocal ...[ ok ] * Running autoconf ... [ !! ] * Failed Running autoconf ! * * Include in your bugreport the contents of: * * /var/tmp/portage/dev-libs/libxslt-1.1.26/temp/autoconf.out * ERROR: dev-libs/libxslt-1.1.26 failed: * Failed Running autoconf ! * * Call stack: * ebuild.sh, line 49: Called src_prepare * environment, line 2797: Called eautoreconf * environment, line 944: Called eautoconf * environment, line 886: Called autotools_run_tool 'autoconf' * environment, line 425: Called die * The specific snippet of code: * die "Failed Running $1 !";
Re: [gentoo-user] emerge ignores /etcportage/package.mask
Maxim Wexler wrote: >> check that you don't have the ebuild listed in package.unmask >> >> > package.unmask is empty, haven't needed it yet. Only just started > stocking a fresh install. > > > Do you have a space in front of the line or something else that may interfere with package.mask ? You sure you have the file as package.mask ? You know, spelled right and all? I mention because it sounds like something I would do. lol Dale :-) :-)
Re: [gentoo-user] emerge ignores /etcportage/package.mask
> check that you don't have the ebuild listed in package.unmask > package.unmask is empty, haven't needed it yet. Only just started stocking a fresh install.
Re: [gentoo-user] Kernel upgrading and linux symlink
On Sat, Oct 31, 2009 at 4:41 PM, Alan McKinnon wrote: > On Sunday 01 November 2009 01:29:27 Mark Knecht wrote: >> So I think this thread addresses a question I've had about the kernel >> installation process over the years. I only copy bzImage to /boot with >> a rename to whatever this kernel is. I don't do anything with the >> other files - System.map and something else - which I don't even have >> on most of my systems anymore. They don't seem to be needed. Are they >> just things used in the old days but now too outdated or replaced by >> other stuff? (Like config.gz in the kernel, etc.) >> > > just run > > make && make modules_install && make install > > and everything gets moved to /boot and renamed. Don't worry about extra files, > they're small and not worthy of concern > > > -- > alan dot mckinnon at gmail dot com > > I don't run make install as it requires (or used to anyway) /boot to be mounted which mine never is unless I'm copying the kernel over. That's why I do the copy by hand. I could put it all in a script like Neil and others seem to do but I'll forget the script is there and not use it. I've done that before! Thanks, Mark
Re: [gentoo-user] Kernel upgrading and linux symlink
On 10/31/2009 7:29 PM, Mark Knecht wrote: So I think this thread addresses a question I've had about the kernel installation process over the years. I only copy bzImage to /boot with a rename to whatever this kernel is. I don't do anything with the other files - System.map and something else - which I don't even have on most of my systems anymore. They don't seem to be needed. Are they just things used in the old days but now too outdated or replaced by other stuff? (Like config.gz in the kernel, etc.) Having a copy of the config file is useful, for me anyway, when I have multiple types of kernel (one without PaX, one without SELinux, whatever) and a new version comes out, I can copy its config from /boot to .config and run make oldconfig. The System.map file is probably the least useful of the three for the average user. It's main use is for address resolution in oops messages. I think ps uses it for something as well. --Mike
Re: [gentoo-user] Re: Kernel upgrading and linux symlink
On Sat, Oct 31, 2009 at 4:40 PM, walt wrote: > On 10/31/2009 04:29 PM, Mark Knecht wrote: > >> So I think this thread addresses a question I've had about the kernel >> installation process over the years. I only copy bzImage to /boot with >> a rename to whatever this kernel is. I don't do anything with the >> other files - System.map and something else - which I don't even have >> on most of my systems anymore. They don't seem to be needed... > > They're needed if you are trying to debug kernel crashes, but AFAIK > not needed for anything else. > Thanks. - Mark
[gentoo-user] depclean / db & php loop
Hi, Earlier today emerge --depclean wouldn't remove db because 3 programs needed it so I rebuilt the 3 programs but it's still complaining about php. Sometimes I'll just remove the problem package - like db in this case - and then let an emerge -DuN @world / revdep-rebuild fix it, but I wanted to be careful with db as it might be important to portage or eix. What's the right way to fix this? If it matters this is a PPC Mac Mini used as a MythTV backend server. revdep-rebuild -ip says everything is clean and emerge -DuN @system @world is finished. Thanks in advance, Mark >>> Assigning files to packages... * In order to avoid breakage of link level dependencies, one or more * packages will not be removed. This can be solved by rebuilding the * packages that pulled them in. * * sys-libs/db-4.6.21_p4 pulled in by: * dev-lang/php-5.2.11 needs libdb-4.6.so * >>> Adding lib providers to graph... | Calculating dependencies... done! >>> No packages selected for removal by depclean >>> To see reverse dependencies, use --verbose Packages installed: 353 Packages in world:35 Packages in system: 52 Required packages:353 Number to remove: 0 MacMini ~ # emerge -pv php These are the packages that would be merged, in order: Calculating dependencies... done! [ebuild R ] dev-lang/php-5.2.11 USE="apache2 berkdb bzip2 cgi cli crypt gd gdbm iconv ldap mysql mysqli ncurses nls pcre posix readline reflection session spell spl ssl truetype unicode xml zlib (-adabas) -bcmath (-birdstep) -calendar -cdb -cjk -concurrentmodphp -ctype -curl -curlwrappers (-db2) -dbase (-dbmaker) -debug -discard-path -doc (-empress) (-empress-bcs) (-esoob) -exif -fastbuild (-fdftk) -filter (-firebird) -flatfile -force-cgi-redirect (-frontbase) -ftp -gd-external -gmp -hash -imap -inifile (-interbase) -iodbc -ipv6 (-java-external) -json -kerberos -kolab -ldap-sasl -libedit -mcve -mhash -msql -mssql (-oci8) (-oci8-instant-client) -odbc -pcntl -pdo -pic -postgres -qdbm -recode -sapdb -sharedext -sharedmem -simplexml -snmp -soap -sockets (-solid) -sqlite -suhosin (-sybase) (-sybase-ct) -sysvipc -threads -tidy -tokenizer -wddx -xmlreader -xmlrpc -xmlwriter -xpm -xsl -yaz -zip" 0 kB Total: 1 package (1 reinstall), Size of downloads: 0 kB MacMini ~ #
Re: [gentoo-user] Kernel upgrading and linux symlink
On Sunday 01 November 2009 01:29:27 Mark Knecht wrote: > So I think this thread addresses a question I've had about the kernel > installation process over the years. I only copy bzImage to /boot with > a rename to whatever this kernel is. I don't do anything with the > other files - System.map and something else - which I don't even have > on most of my systems anymore. They don't seem to be needed. Are they > just things used in the old days but now too outdated or replaced by > other stuff? (Like config.gz in the kernel, etc.) > just run make && make modules_install && make install and everything gets moved to /boot and renamed. Don't worry about extra files, they're small and not worthy of concern -- alan dot mckinnon at gmail dot com
[gentoo-user] Re: Kernel upgrading and linux symlink
On 10/31/2009 04:29 PM, Mark Knecht wrote: > So I think this thread addresses a question I've had about the kernel > installation process over the years. I only copy bzImage to /boot with > a rename to whatever this kernel is. I don't do anything with the > other files - System.map and something else - which I don't even have > on most of my systems anymore. They don't seem to be needed... They're needed if you are trying to debug kernel crashes, but AFAIK not needed for anything else.
Re: [gentoo-user] Kernel upgrading and linux symlink
On Sat, Oct 31, 2009 at 3:20 PM, Alan McKinnon wrote: > On Saturday 31 October 2009 23:26:33 Dale wrote: >> Denis wrote: >> > On Sat, Oct 31, 2009 at 12:52 PM, Harry Putnam wrote: >> >> And in fact does it really matter if its pointing at the newly >> >> installed or actual running kernel, when kernel compiling operations >> >> take place? >> > >> > When I upgrade a kernel, I first change the symlink using eselect to >> > point to the source I'm about to install. Then, after I configure and >> > compile the kernel, I use the "module-rebuild rebuild" to rebuild any >> > kernel modules against the new source. Move the bzImage to /boot, >> > reboot, and that's it. I don't know how genkernel changes any of this >> > - I use manual menuconfig. >> >> I'm about the same. I update the symlink, build the kernel, update >> nvidia-drivers against the new kernel before I forget, copy bzImage to >> /boot and edit grub. I reboot when I get the chance. > > I'm a forgetful old git. In my world it usually goes like this: > > emerge, build, install new kernel > carry on with work > boot into new kernel at some later point > observe xdm doing nothing on-screen > curse and swear mightily > Ctrl-Alt-F1 > login as root > check symlink > module-rebuild rebuild > modprobe -r nvidia && modprobe nvidia > /etc/init.d/xdm restart && logout > > takes about 6 minutes total, 6 minutes that I'll never get back :-) > > -- > alan dot mckinnon at gmail dot com So I think this thread addresses a question I've had about the kernel installation process over the years. I only copy bzImage to /boot with a rename to whatever this kernel is. I don't do anything with the other files - System.map and something else - which I don't even have on most of my systems anymore. They don't seem to be needed. Are they just things used in the old days but now too outdated or replaced by other stuff? (Like config.gz in the kernel, etc.) Thanks, Mark
Re: [gentoo-user] Kernel upgrading and linux symlink
On Sun, 1 Nov 2009 00:20:57 +0200, Alan McKinnon wrote: > I'm a forgetful old git. In my world it usually goes like this: > > emerge, build, install new kernel > carry on with work > boot into new kernel at some later point > observe xdm doing nothing on-screen > curse and swear mightily > Ctrl-Alt-F1 > login as root > check symlink > module-rebuild rebuild > modprobe -r nvidia && modprobe nvidia > /etc/init.d/xdm restart && logout > > takes about 6 minutes total, 6 minutes that I'll never get back :-) That's why I use a script to build and install the kernel and then install any external modules. With video drivers it's merely an inconvenience if you forget, but with wireless drivers, especially with DISTDIR on NFS, it means I would have to reboot back to the old kernel again. -- Neil Bothwick If you think that there is good in everybody, you haven't met everybody. signature.asc Description: PGP signature
[gentoo-user] firefox-3.5.4 and xulrunner
Someone posted recently about an oversight between latest version of xulrunner in the tree and the version required by firefox. A fixed ebuild is in the tree, but if you sync and update daily, firefox might fail to start and give you this on the console: $ Could not find compatible GRE between version 1.9.1.3 and 1.9.1.3. ^C [1]+ Exit 1 firefox just rebuild forefox to use the latest xulrunner -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Kernel upgrading and linux symlink
On Sunday 01 November 2009 00:52:42 Dale wrote: > > emerge, build, install new kernel > > carry on with work > > boot into new kernel at some later point > > observe xdm doing nothing on-screen > > curse and swear mightily > > Ctrl-Alt-F1 > > login as root > > check symlink > > module-rebuild rebuild > > modprobe -r nvidia && modprobe nvidia > > /etc/init.d/xdm restart && logout > > > > takes about 6 minutes total, 6 minutes that I'll never get back :-) > > > > > > Well, I'm a bit of a old git too. I just sort of build habits when I am > doing something which helps me to not forget. I'm dreading the new > baselayout upgrade. I don't have a "habit" for that yet. lol Go on, do it. You know you want to :-) Besides, what else do you have to do with your time except read new stuff in /etc/conf.d/ ? -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Kernel upgrading and linux symlink
Alan McKinnon wrote: > On Saturday 31 October 2009 23:26:33 Dale wrote: > >> Denis wrote: >> >>> On Sat, Oct 31, 2009 at 12:52 PM, Harry Putnam wrote: >>> And in fact does it really matter if its pointing at the newly installed or actual running kernel, when kernel compiling operations take place? >>> When I upgrade a kernel, I first change the symlink using eselect to >>> point to the source I'm about to install. Then, after I configure and >>> compile the kernel, I use the "module-rebuild rebuild" to rebuild any >>> kernel modules against the new source. Move the bzImage to /boot, >>> reboot, and that's it. I don't know how genkernel changes any of this >>> - I use manual menuconfig. >>> >> I'm about the same. I update the symlink, build the kernel, update >> nvidia-drivers against the new kernel before I forget, copy bzImage to >> /boot and edit grub. I reboot when I get the chance. >> > > I'm a forgetful old git. In my world it usually goes like this: > > emerge, build, install new kernel > carry on with work > boot into new kernel at some later point > observe xdm doing nothing on-screen > curse and swear mightily > Ctrl-Alt-F1 > login as root > check symlink > module-rebuild rebuild > modprobe -r nvidia && modprobe nvidia > /etc/init.d/xdm restart && logout > > takes about 6 minutes total, 6 minutes that I'll never get back :-) > > Well, I'm a bit of a old git too. I just sort of build habits when I am doing something which helps me to not forget. I'm dreading the new baselayout upgrade. I don't have a "habit" for that yet. lol Dale :-) :-)
Re: [gentoo-user] hal vs. devicekit
Mike Edenfield wrote:
> On 10/31/2009 4:59 PM, Alan McKinnon wrote:
>
>> Even it's author knows this (but apparently many distros do not)
>> which is why
>> he deprecated hal and started over with devicekit.
>
> Speaking of which...
>
> Has the switchover to devicekit officially started and I missed it?
> And if so, is there some migration guide of some sort I can peruse for
> assistance?
>
> I just upgraded Xorg and Gnome yesterday and my previously
> more-or-less-working with hal system went belly up. It no longer
> recognizes my synaptics touchpad, and in fact refused to start X until
> I re-merged the kdb/mouse drivers, and gnome-power-manager is totally
> useless.
>
> For the record, I'm all about moving to devicekit from hal if that's
> where the dev(s) are moving, but I'd rather do so while continuing to
> use my laptop :)
>
> --Mike
>
>
I noticed devicekit going in the tree the other day. I think I posted
it on this list.
r...@smoker / # eix devicekit
* sys-apps/devicekit
Available versions: ~003 {doc}
Homepage:http://www.freedesktop.org/wiki/Software/DeviceKit
Description: D-Bus abstraction for enumerating devices and
listening for device events using udev
* sys-apps/devicekit-disks
Available versions: ~008 {bash-completion debug doc}
Homepage:http://www.freedesktop.org/wiki/Software/DeviceKit
Description: Daemon providing interfaces to work with
storage devices
* sys-apps/devicekit-power
Available versions: ~009 ~011 {debug doc test}
Homepage:http://www.freedesktop.org/wiki/Software/DeviceKit
Description: D-Bus abstraction for enumerating power
devices and querying history and statistics
Found 3 matches.
r...@smoker / #
I'm not sure when it FIRST hit the tree but at least it is there. I
hope the config stuff is easier. Of course, if it just works, that
would be nice.
Dale
:-) :-)
Re: [gentoo-user] hal vs. devicekit (was: more about hal)
On Sunday 01 November 2009 00:25:55 Mike Edenfield wrote: > On 10/31/2009 4:59 PM, Alan McKinnon wrote: > > Even it's author knows this (but apparently many distros do not) which is > > why he deprecated hal and started over with devicekit. > > Speaking of which... > > Has the switchover to devicekit officially started and I > missed it? And if so, is there some migration guide of some > sort I can peruse for assistance? > > I just upgraded Xorg and Gnome yesterday and my previously > more-or-less-working with hal system went belly up. It no > longer recognizes my synaptics touchpad, and in fact refused > to start X until I re-merged the kdb/mouse drivers, and > gnome-power-manager is totally useless. > > For the record, I'm all about moving to devicekit from hal > if that's where the dev(s) are moving, but I'd rather do so > while continuing to use my laptop :) devicekit was only recently added to portage, and last time I looked the hardware it supported was nowhere near hal yet. So the answer to your question is unfortunately no -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] hal vs. devicekit (was: more about hal)
On 10/31/2009 4:59 PM, Alan McKinnon wrote: Even it's author knows this (but apparently many distros do not) which is why he deprecated hal and started over with devicekit. Speaking of which... Has the switchover to devicekit officially started and I missed it? And if so, is there some migration guide of some sort I can peruse for assistance? I just upgraded Xorg and Gnome yesterday and my previously more-or-less-working with hal system went belly up. It no longer recognizes my synaptics touchpad, and in fact refused to start X until I re-merged the kdb/mouse drivers, and gnome-power-manager is totally useless. For the record, I'm all about moving to devicekit from hal if that's where the dev(s) are moving, but I'd rather do so while continuing to use my laptop :) --Mike
Re: [gentoo-user] Kernel upgrading and linux symlink
On Saturday 31 October 2009 23:26:33 Dale wrote: > Denis wrote: > > On Sat, Oct 31, 2009 at 12:52 PM, Harry Putnam wrote: > >> And in fact does it really matter if its pointing at the newly > >> installed or actual running kernel, when kernel compiling operations > >> take place? > > > > When I upgrade a kernel, I first change the symlink using eselect to > > point to the source I'm about to install. Then, after I configure and > > compile the kernel, I use the "module-rebuild rebuild" to rebuild any > > kernel modules against the new source. Move the bzImage to /boot, > > reboot, and that's it. I don't know how genkernel changes any of this > > - I use manual menuconfig. > > I'm about the same. I update the symlink, build the kernel, update > nvidia-drivers against the new kernel before I forget, copy bzImage to > /boot and edit grub. I reboot when I get the chance. I'm a forgetful old git. In my world it usually goes like this: emerge, build, install new kernel carry on with work boot into new kernel at some later point observe xdm doing nothing on-screen curse and swear mightily Ctrl-Alt-F1 login as root check symlink module-rebuild rebuild modprobe -r nvidia && modprobe nvidia /etc/init.d/xdm restart && logout takes about 6 minutes total, 6 minutes that I'll never get back :-) -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Re: Kernel upgrading and linux symlink
On Saturday 31 October 2009 23:43:21 Mark Knecht wrote: > On Sat, Oct 31, 2009 at 2:27 PM, Nikos Chantziaras wrote: > > On 10/31/2009 11:07 PM, Alan McKinnon wrote: > >> On Saturday 31 October 2009 22:03:04 Nikos Chantziaras wrote: > For instance, you might be running 2.6.31-r4 and also have 2.6.31-r3 > installed. To install nvidia-drivers, you must build it twice - > against each kernel you want to use it with (nvidia-drivers builds and > installs a > kernel driver into /lib/modules/) > >>> > >>> It's a bit more obfuscated than that. Maybe nvidia-drivers work > >>> different, but ati-drivers will build against /usr/src/linux but > >>> install the actual modules in /lib/modules/running_kernel. If > >>> /usr/src/linux doesn't point to the running kernel, the modules will be > >>> installed in the wrong place. > >> > >> That is just so mind-bogglingly absurdly stupid I doubt if ATI should > >> even be > >> allowed near a computer > >> > >> Compiling code never depends on something running, it only depends on > >> things > >> being present that can be linked against. > >> > >> Thanks for reminding me why I insist on NVidia GPUs, I'd forgotten. > > > > This isn't ATI's installer. It's the ebuild that does this. > > And from deep memory it seems like there were other packages that > operated this way 8-10 years ago. I know in 1999 I had to be very > careful about where the linux link pointed, and while it's not as > necessary today to do so i'm still quite careful. > > I use the ATI drivers on my AMD64 machine. I think I've always found > that I needed to emerge fglrx after the new kernel had been booted but > never understood why. This email is helpful. It seems to me that if it > is the ebuild that's doing this is needs to be fixed. If I understand > correctly I could be building for 2.6.31 but installing in 2.6.29? > That's not right... I agree, the ebuild should be fixed. I can't think of any valid reason for that behaviour. -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] emerge ignores /etcportage/package.mask
On Saturday 31 October 2009 23:44:00 Maxim Wexler wrote: > Hi group, > > From the gentoo docs: > > "Older cards such as the GeForce FX 5 series should use the 173.x > drivers, such as nvidia-drivers-173.14.15. For these cards, you should > mask >=x11-drivers/nvidia-drivers-174.00 in your > /etc/portage/package.mask file. This will prevent newer versions of > the driver which are incompatible with your card from being > installed." > > But emerge ignores package.mask and insists on installing the newer > pkg. Searching reveals an old bug marked 'resolved' but nothing to say > what the resolution is. check that you don't have the ebuild listed in package.unmask That takes precedence over package.mask -- alan dot mckinnon at gmail dot com
[gentoo-user] emerge ignores /etcportage/package.mask
Hi group, >From the gentoo docs: "Older cards such as the GeForce FX 5 series should use the 173.x drivers, such as nvidia-drivers-173.14.15. For these cards, you should mask >=x11-drivers/nvidia-drivers-174.00 in your /etc/portage/package.mask file. This will prevent newer versions of the driver which are incompatible with your card from being installed." But emerge ignores package.mask and insists on installing the newer pkg. Searching reveals an old bug marked 'resolved' but nothing to say what the resolution is. Maxim
Re: [gentoo-user] Re: Kernel upgrading and linux symlink
On Sat, Oct 31, 2009 at 2:27 PM, Nikos Chantziaras wrote: > On 10/31/2009 11:07 PM, Alan McKinnon wrote: >> >> On Saturday 31 October 2009 22:03:04 Nikos Chantziaras wrote: For instance, you might be running 2.6.31-r4 and also have 2.6.31-r3 installed. To install nvidia-drivers, you must build it twice - against each kernel you want to use it with (nvidia-drivers builds and installs a kernel driver into /lib/modules/) >>> >>> It's a bit more obfuscated than that. Maybe nvidia-drivers work >>> different, but ati-drivers will build against /usr/src/linux but install >>> the actual modules in /lib/modules/running_kernel. If /usr/src/linux >>> doesn't point to the running kernel, the modules will be installed in >>> the wrong place. >> >> >> That is just so mind-bogglingly absurdly stupid I doubt if ATI should even >> be >> allowed near a computer >> >> Compiling code never depends on something running, it only depends on >> things >> being present that can be linked against. >> >> Thanks for reminding me why I insist on NVidia GPUs, I'd forgotten. > > This isn't ATI's installer. It's the ebuild that does this. And from deep memory it seems like there were other packages that operated this way 8-10 years ago. I know in 1999 I had to be very careful about where the linux link pointed, and while it's not as necessary today to do so i'm still quite careful. I use the ATI drivers on my AMD64 machine. I think I've always found that I needed to emerge fglrx after the new kernel had been booted but never understood why. This email is helpful. It seems to me that if it is the ebuild that's doing this is needs to be fixed. If I understand correctly I could be building for 2.6.31 but installing in 2.6.29? That's not right... Thanks to all for the info. Cheers, Mark
Re: [gentoo-user] Kmplayer, video and audio not syncing.
Alan McKinnon wrote: > On Saturday 31 October 2009 05:34:57 Dale wrote: > >> Jesús Guerrero wrote: >> >>> kmplayer can get in the middle, can you -please- test regular mplayer >>> from command line? >>> >>> I've had a similar issue a couple of weeks ago, and we were able to track >>> it down on the mplayer mailing lists. If it's the same bug, I was able to >>> consistently reproduce it in streams with ac3 (5.1) audio (stereo worked >>> fine), and only when using ALSA. So, check that and see if you can see a >>> pattern there. If you see that same pattern, then it might be the same >>> bug, it's been fixed in the development branch, and the following ebuilds >>> should work fine: >>> >>> 1.0_rc4_p20091026, 1.0_rc4_p20091026-r1, >>> >>> If not, then it's probably something else. But try with mplayer alone >>> when debugging, since kmplayer just adds another level to worry about. >>> For your reference, here's the bug I opened, there you can also find >>> links to the relevant mails in the mplayer ML. >>> >>> http://bugs.gentoo.org/show_bug.cgi?id=286020 >>> >> I can give it a shot at least. Do I just run mplayer path/to/file or do >> I need to add some options so that it will provide more info? >> > > If you run mplayer it will try to autodetect what to do. Or you can > look at the kmpayer config and find the corresponding options in mplayer's > man > page and use those to see if it makes a difference. > > Beware the mplayer man page - it's huge, complex and a horrendous read. > That's > unavoidable - happens with any complex software that covers 100s of possible > combinations. > > I noticed it was huge. That was why I asked. I figured a reply might be faster than me trying to read all that then understand it as well. That could take a looong while. LOL Dale :-) :-)
[gentoo-user] Re: Kernel upgrading and linux symlink
On 10/31/2009 11:07 PM, Alan McKinnon wrote: On Saturday 31 October 2009 22:03:04 Nikos Chantziaras wrote: For instance, you might be running 2.6.31-r4 and also have 2.6.31-r3 installed. To install nvidia-drivers, you must build it twice - against each kernel you want to use it with (nvidia-drivers builds and installs a kernel driver into /lib/modules/) It's a bit more obfuscated than that. Maybe nvidia-drivers work different, but ati-drivers will build against /usr/src/linux but install the actual modules in /lib/modules/running_kernel. If /usr/src/linux doesn't point to the running kernel, the modules will be installed in the wrong place. That is just so mind-bogglingly absurdly stupid I doubt if ATI should even be allowed near a computer Compiling code never depends on something running, it only depends on things being present that can be linked against. Thanks for reminding me why I insist on NVidia GPUs, I'd forgotten. This isn't ATI's installer. It's the ebuild that does this.
Re: [gentoo-user] Kernel upgrading and linux symlink
Denis wrote: > On Sat, Oct 31, 2009 at 12:52 PM, Harry Putnam wrote: > >> And in fact does it really matter if its pointing at the newly >> installed or actual running kernel, when kernel compiling operations >> take place? >> > > When I upgrade a kernel, I first change the symlink using eselect to > point to the source I'm about to install. Then, after I configure and > compile the kernel, I use the "module-rebuild rebuild" to rebuild any > kernel modules against the new source. Move the bzImage to /boot, > reboot, and that's it. I don't know how genkernel changes any of this > - I use manual menuconfig. > > > I'm about the same. I update the symlink, build the kernel, update nvidia-drivers against the new kernel before I forget, copy bzImage to /boot and edit grub. I reboot when I get the chance. If the kernel is borked, I change the symlink back and reboot to my old kernel. I try to keep the symlink pointing to the kernel I am running even if it is not absolutely necessary. Dale :-) :-)
Re: [gentoo-user] Re: Kernel upgrading and linux symlink
On Saturday 31 October 2009 22:03:04 Nikos Chantziaras wrote: > > For instance, you might be running 2.6.31-r4 and also have 2.6.31-r3 > > installed. To install nvidia-drivers, you must build it twice - against > > each kernel you want to use it with (nvidia-drivers builds and installs a > > kernel driver into /lib/modules/) > > It's a bit more obfuscated than that. Maybe nvidia-drivers work > different, but ati-drivers will build against /usr/src/linux but install > the actual modules in /lib/modules/running_kernel. If /usr/src/linux > doesn't point to the running kernel, the modules will be installed in > the wrong place. That is just so mind-bogglingly absurdly stupid I doubt if ATI should even be allowed near a computer Compiling code never depends on something running, it only depends on things being present that can be linked against. Thanks for reminding me why I insist on NVidia GPUs, I'd forgotten. -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Re: Kernel upgrading and linux symlink
On Saturday 31 October 2009 22:06:35 Harry Putnam wrote: > Alan McKinnon writes: > > Nikos is being kind to the document writers :-) > > Thanks for saving me from the Dunce cap...hehe. > > But I might yet acquire full rights to it.. > > So, is the symlink not really necessary? Doe something look at > /usr/src/linux for files? Yes, almost every app that installs kernel drivers will go to /usr/src/linux/ looking for kernel sources > For example, if you cd 'ed into the sources top dir. And started > `make' (after the makeconfig step), would it matter if there was a > symlink or not? It doesn't matter at all. There's a Makefile in that directory and it will do whatever it's supposed to do. The symlink to it is irrelevant and is only there as a shortcut with a known name -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] More about hal
On Saturday 31 October 2009 22:36:34 Dale wrote: > Harry Putnam wrote: > > When I noticed the thread here about `hal' that started a while back > > it caused me enough curiousity that i ran eix -Ic ^hal$ but found I > > have no `hal' installed. > > > > I have keep up with updates somewhat better than usual the last few > > months but don't remember when hal went away... I do remember having > > some trouble with keyboard/mouse, and hal not starting, but that was > > quite a while back. > > > > I don't remember making a conscious effort to get rid of hal either. > > > > Evey thing seems to work ok here. I do notice a problem on bootup > > where the keyboard/mouse I have attached to a KVM does not become > > usable until AFTER the grub prompt. Somewhere between there and the > > appearance of the login prompt it becomes usable. But that has gone > > on at least a yr if not longer. > > > > So is there something wrong if I have no hal installed. Is it just > > not necessary or has it been replaced? > > I think it goes to show that a computer can work fine without hal. > Somebody nudge the KDE folks. Doesn't KDE require hal? There's no valid technical reason why an app *must* use hal. An app can try and do all hardware detection by itself, but it just makes more sense to use one common layer for that rather than every app doing essentially the same action. KDE-4 itself does not directly require hal. Solid does though: $ equery depends hal * Searching for hal ... app-emulation/wine-1.1.32 (hal ? sys-apps/hal) app-laptop/laptop-mode-tools-1.52 (hal ? sys-apps/hal) app-misc/hal-cups-utils-0.6.19 (>=sys-apps/hal-0.5.10) app-misc/hal-info-20090716 (>=sys-apps/hal-0.5.10) kde-base/solid-4.3.2 (>=sys-apps/hal-0.5.9) media-gfx/gimp-2.6.7 (hal ? sys-apps/hal) media-libs/libgphoto2-2.4.7 (hal ? >=sys-apps/hal-0.5) media-libs/libgpod-0.7.2 (hal ? =sys-apps/hal-0.5*) media-video/gxine-0.5.903 (hal ? sys-apps/hal) media-video/vlc-1.0.3_rc (hal ? sys-apps/hal) net-misc/networkmanager-0.7.1_p20090824 (>=sys-apps/hal-0.5.10) sys-apps/pcsc-lite-1.5.5 (hal ? sys-apps/hal) sys-power/pm-utils-1.2.5 (>=sys-apps/hal-0.5.10) x11-base/xorg-server-1.7.1 (hal ? sys-apps/hal) x11-drivers/xf86-input-synaptics-1.2.0 (hal ? sys-apps/hal) x11-libs/e_dbus- (sys-apps/hal) There's nothing wrong with the *idea* of hal; it's the implementation of that app that makes it an utter piece of shit. It's called "feature-creep". Even it's author knows this (but apparently many distros do not) which is why he deprecated hal and started over with devicekit. -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Re: Kernel upgrading and linux symlink
On Sat, 31 Oct 2009 20:00:35 +0200, Nikos Chantziaras wrote: > So, to sum it up, when you update to a new kernel, first update the > symlink (or let the ebuild take care of it), build the kernel, boot it, > *then* rebuild external modules (like ati-drivers). If you know what > you're doing, you don't need to reboot before rebuilding external > modules, for example I do: > > ebuild /usr/portage/x11-drivers/ati-drivers/ati-drivers-N.ebuild compile > mkdir /lib/modules/KERNEL_VERSION/video > cp > /var/tmp/portage/x11-drivers/ati-drivers-N/work/common/lib/modules/fglrx/build_mod/2.6.x/fglrx.ko > > /lib/modules/KERNEL_VERSION/video/ I don't use ati, but with other modules, this is not necessary. As long as /usr/srx/linux contains a valid .config, you can build modules against that kernel without running or even compiling it. -- Neil Bothwick Thesaurus: ancient reptile with an excellent vocabulary signature.asc Description: PGP signature
Re: [gentoo-user] video capture from min-dv... kernel params
On Sat, 31 Oct 2009 10:05:42 -0500, Harry Putnam wrote: > I wondered if anyone here has used gentoo for mini-dv capture and if > so what do I need to do regarding the kernel? It's been a few years (my current camcorder has a hard disk and uses USB for transfer) but I don't think I built anything special into the kernel beyond the 1394 stuff. > I see apps called dvgrab and kino show up in many of the hits too. > > Is it just a matter of building specific modules or turning on certain > parameters in the kernel? I think the only one that caught me out was missing raw1384. > Are the tools dvgrab or kino self contained enough that I could do the > the capture with one or the other alone? Yes. Kino is a grabber and editor, dvgrab is the grabbing code from Kino released as a separate commandline program. If you only want to grab, dvgrab is enough. -- Neil Bothwick Our bikinis are exciting. They are simply the tops. signature.asc Description: PGP signature
Re: [gentoo-user] More about hal
Harry Putnam wrote: > When I noticed the thread here about `hal' that started a while back > it caused me enough curiousity that i ran eix -Ic ^hal$ but found I > have no `hal' installed. > > I have keep up with updates somewhat better than usual the last few > months but don't remember when hal went away... I do remember having > some trouble with keyboard/mouse, and hal not starting, but that was > quite a while back. > > I don't remember making a conscious effort to get rid of hal either. > > Evey thing seems to work ok here. I do notice a problem on bootup > where the keyboard/mouse I have attached to a KVM does not become > usable until AFTER the grub prompt. Somewhere between there and the > appearance of the login prompt it becomes usable. But that has gone > on at least a yr if not longer. > > So is there something wrong if I have no hal installed. Is it just > not necessary or has it been replaced? > > > > I think it goes to show that a computer can work fine without hal. Somebody nudge the KDE folks. Doesn't KDE require hal? Dale :-) :-)
Re: [gentoo-user] [OT[ How do I unautomate an OOO slideshow?
On Friday 30 October 2009, fe...@crowfix.com wrote: > People send me slideshows, some of which are interesting enough to > watch. With OOO-3.0, I had to click or use NextPage for each picture > and I got no sound, and that was fine with me. Now OOO 3.1 automates > these slideshows and they really aggravate me -- I use Linux because > it's my computer and I want to set my own pace, not be bored by auto > numbness. > > Is there some way to put a slideshow in manual mode, either on the > command line right from the start or during it? I know ESC quts it, > but that's not what I want. All the docco I find seems to think this > automation is just dandy and teaches how to enable it, but nothing on > how to turn it off. On my system, no slideshow advances automatically (at least none of the 7 I tried before writing this).
[gentoo-user] More about hal
When I noticed the thread here about `hal' that started a while back it caused me enough curiousity that i ran eix -Ic ^hal$ but found I have no `hal' installed. I have keep up with updates somewhat better than usual the last few months but don't remember when hal went away... I do remember having some trouble with keyboard/mouse, and hal not starting, but that was quite a while back. I don't remember making a conscious effort to get rid of hal either. Evey thing seems to work ok here. I do notice a problem on bootup where the keyboard/mouse I have attached to a KVM does not become usable until AFTER the grub prompt. Somewhere between there and the appearance of the login prompt it becomes usable. But that has gone on at least a yr if not longer. So is there something wrong if I have no hal installed. Is it just not necessary or has it been replaced?
[gentoo-user] Re: Kernel upgrading and linux symlink
Alan McKinnon writes: > Nikos is being kind to the document writers :-) Thanks for saving me from the Dunce cap...hehe. But I might yet acquire full rights to it.. So, is the symlink not really necessary? Doe something look at /usr/src/linux for files? For example, if you cd 'ed into the sources top dir. And started `make' (after the makeconfig step), would it matter if there was a symlink or not? (In the instant case I did create the symlink looking at newest sources, so all is well I hope... have yet to boot the creation. Waiting on some emerging left in emerge -vuD system)
[gentoo-user] Re: Kernel upgrading and linux symlink
On 10/31/2009 09:18 PM, Alan McKinnon wrote: On Saturday 31 October 2009 20:09:37 Harry Putnam wrote: Nikos Chantziaras writes: The link is created only if you have the "symlink" USE flag enabled. Also, "Gentoo requires that the [...] symbolic link points to the sources of the kernel you are running" is not entirely correct. It is required only when you want to build something against that kernel. . . . . Obviously, you need to create the symlink if you want to build the newly installed kernel, even though the system is still running an older one. Why is that obvious? That's what seemed confusing to me. Nothing about creating it with USE=symlin, eselect, or by hand is a problem. Or hard to follow, and I've always just done it by hand. Nikos is being kind to the document writers :-) In fact, the documentation is flat out wrong - there is no requirement for the symlink to point to the currently running kernel. It must point to the kernel sources you want to *configure* or use for an emerge that installs a kernel driver. For instance, you might be running 2.6.31-r4 and also have 2.6.31-r3 installed. To install nvidia-drivers, you must build it *twice* - against each kernel you want to use it with (nvidia-drivers builds and installs a kernel driver into /lib/modules/) It's a bit more obfuscated than that. Maybe nvidia-drivers work different, but ati-drivers will build against /usr/src/linux but install the actual modules in /lib/modules/running_kernel. If /usr/src/linux doesn't point to the running kernel, the modules will be installed in the wrong place.
Re: [gentoo-user] Re: removing kde4(sets)
On Saturday 31 October 2009 02:46:34 Neil Bothwick wrote: > On Fri, 30 Oct 2009 22:33:15 +0200, Alan McKinnon wrote: > > tomorrow I shall take myself off to see my special good friend Peter > > who is giving me a fantastic deal on a brand new mother of a > > motorcycle. And in the afternoon I shall watch the Blue Bulls thrash > > the living daylights out of the FreeState Cheetahs in the Currie Cup > > rugby final. > > > > Oh wait, that's all ZA stuff. You probably have no idea what I'm on > > about :-) > > Actually, we have motorcycles in this part of the world too :P REALLY But, but, but, but ... it pisses with rain all the time in the UK And the English are whingers of not. Now the Scots - they are different. They'll ride their bikes come rain, snow, hail, or anything else :-) And the Bulls won the rugby. They didn't have it easy, the Cheetahs put up a heck of a fight. -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Kmplayer, video and audio not syncing.
On Saturday 31 October 2009 05:34:57 Dale wrote: > Jesús Guerrero wrote: > > kmplayer can get in the middle, can you -please- test regular mplayer > > from command line? > > > > I've had a similar issue a couple of weeks ago, and we were able to track > > it down on the mplayer mailing lists. If it's the same bug, I was able to > > consistently reproduce it in streams with ac3 (5.1) audio (stereo worked > > fine), and only when using ALSA. So, check that and see if you can see a > > pattern there. If you see that same pattern, then it might be the same > > bug, it's been fixed in the development branch, and the following ebuilds > > should work fine: > > > > 1.0_rc4_p20091026, 1.0_rc4_p20091026-r1, > > > > If not, then it's probably something else. But try with mplayer alone > > when debugging, since kmplayer just adds another level to worry about. > > For your reference, here's the bug I opened, there you can also find > > links to the relevant mails in the mplayer ML. > > > > http://bugs.gentoo.org/show_bug.cgi?id=286020 > > I can give it a shot at least. Do I just run mplayer path/to/file or do > I need to add some options so that it will provide more info? If you run mplayer it will try to autodetect what to do. Or you can look at the kmpayer config and find the corresponding options in mplayer's man page and use those to see if it makes a difference. Beware the mplayer man page - it's huge, complex and a horrendous read. That's unavoidable - happens with any complex software that covers 100s of possible combinations. -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Xorg slow after upgrade
On Saturday 31 October 2009 17:34:30 Valmor de Almeida wrote: > Krzysztof Poc wrote: > > Hello I've recently upgraded my system and I found that Xorg redraws the > > screen > > very slowly. "top" shows me that "X" process takes up around 95% of the > > CPU while I have Intel Core 2 Duo. > > > > What's wrong with my system ? > > > > I have the following installed: > > kernel 2.6.30.7 #3 SMP (without PAE) > > xorg-server-1.6.3.901-r2 > > xorg-x11-7.4-r1 > > xf86-video-intel-2.8.1 > > > > I've also successfully migrated to X.org Server 1.6 and libxcb 1.4 as > > written > > in official gentoo documentation. > > > > great thanks for any indications > > I am using > > kernel 2.6.30.9 i686 > xorg-server 1.6.3.901-r2 > xorg-x11 7.4-r1 > xf86-video-intel 2.8.1 > hal 0.5.12_rc1-r8 > > Also have some unresolved issues with windowmaker-0.92.0-r8 not > accepting background images on workspaces, and glxgears results lower > (by half) than I had with xorg-server-1.5: > > -> glxgears > 2095 frames in 5.0 seconds = 418.739 FPS > 1800 frames in 5.0 seconds = 359.967 FPS > 1881 frames in 5.0 seconds = 376.091 FPS > 2097 frames in 5.0 seconds = 419.294 FPS > 2097 frames in 5.0 seconds = 419.286 FPS > 1887 frames in 5.0 seconds = 377.349 FPS > 1891 frames in 5.0 seconds = 378.111 FPS > 1911 frames in 5.0 seconds = 382.009 FPS Are either of you using KDE-4? I get this too, it's caused by plasma-desktop (top says it's 35% cpu) which makes X block on some IO (100% of one cpu). Killing and starting plasma-desktop fixes it for me. -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Re: Kernel upgrading and linux symlink
On Saturday 31 October 2009 20:09:37 Harry Putnam wrote: > Nikos Chantziaras writes: > > The link is created only if you have the "symlink" USE flag enabled. > > > > Also, "Gentoo requires that the [...] symbolic link points to the > > sources of the kernel you are running" is not entirely correct. It is > > required only when you want to build something against that > > kernel. > > > > . . . . Obviously, you need to create the symlink if you want to build > > the newly installed kernel, even though the system is still running an > > older one. > > Why is that obvious? That's what seemed confusing to me. > > Nothing about creating it with USE=symlin, eselect, or by hand is a > problem. Or hard to follow, and I've always just done it by hand. > Nikos is being kind to the document writers :-) In fact, the documentation is flat out wrong - there is no requirement for the symlink to point to the currently running kernel. It must point to the kernel sources you want to *configure* or use for an emerge that installs a kernel driver. For instance, you might be running 2.6.31-r4 and also have 2.6.31-r3 installed. To install nvidia-drivers, you must build it *twice* - against each kernel you want to use it with (nvidia-drivers builds and installs a kernel driver into /lib/modules/) USE="symlink" just runs ln -sfn /usr/src/ linux at the end of the merge , no further magic. It's purely a convenience thing, you can just as easily do that step yourself -- alan dot mckinnon at gmail dot com
[gentoo-user] Re: Kernel upgrading and linux symlink
On 10/31/2009 08:09 PM, Harry Putnam wrote: Nikos Chantziaras writes: The link is created only if you have the "symlink" USE flag enabled. Also, "Gentoo requires that the [...] symbolic link points to the sources of the kernel you are running" is not entirely correct. It is required only when you want to build something against that kernel. . . . . Obviously, you need to create the symlink if you want to build the newly installed kernel, even though the system is still running an older one. Why is that obvious? That's what seemed confusing to me. How obvious it is probably depends in prior knowledge here. There's absolutely nothing in a running system that needs anything from /usr/src/linux. If fact, you could, completely uninstall all kernel sources and totally wipe out /usr/src. It's only needed when you build something that needs kernel sources. Of course if you didn't know that, then yes, it's not obvious :)
Re: [gentoo-user] Kernel upgrading and linux symlink
On Sat, Oct 31, 2009 at 12:52 PM, Harry Putnam wrote: > And in fact does it really matter if its pointing at the newly > installed or actual running kernel, when kernel compiling operations > take place? When I upgrade a kernel, I first change the symlink using eselect to point to the source I'm about to install. Then, after I configure and compile the kernel, I use the "module-rebuild rebuild" to rebuild any kernel modules against the new source. Move the bzImage to /boot, reboot, and that's it. I don't know how genkernel changes any of this - I use manual menuconfig.
[gentoo-user] Re: Kernel upgrading and linux symlink
Nikos Chantziaras writes: > The link is created only if you have the "symlink" USE flag enabled. > > Also, "Gentoo requires that the [...] symbolic link points to the > sources of the kernel you are running" is not entirely correct. It is > required only when you want to build something against that > kernel. > . . . . Obviously, you need to create the symlink if you want to build > the newly installed kernel, even though the system is still running an > older one. Why is that obvious? That's what seemed confusing to me. Nothing about creating it with USE=symlin, eselect, or by hand is a problem. Or hard to follow, and I've always just done it by hand.
[gentoo-user] Re: Kernel upgrading and linux symlink
On 10/31/2009 06:52 PM, Harry Putnam wrote: Looking at the kernel upgrade pages at http://www.gentoo.org/doc/en/kernel-upgrade.xml Its a bit confusing about the symlink creation. I've wondered about it a few times. At the top, you're told how to get the sources and then a discussion of the symlink follows. It appears you are expected to change the symlink to the newly installed sources. But then it says: 4. Updating the /usr/src/linux symbolic link "Gentoo requires that the /usr/src/linux symbolic link points to the sources of the kernel you are running." So if gentoo `requires' the symlink to point to the running kernel why are we changing it to the newly installed but not yet compiled or started kernel? And in fact does it really matter if its pointing at the newly installed or actual running kernel, when kernel compiling operations take place? The link is created only if you have the "symlink" USE flag enabled. Also, "Gentoo requires that the [...] symbolic link points to the sources of the kernel you are running" is not entirely correct. It is required only when you want to build something against that kernel. Obviously, you need to create the symlink if you want to build the newly installed kernel, even though the system is still running an older one. And, btw, the symlink is changed easily with eselect: eselect kernel list to get a list and: eselect kernel set N to point the symlink to the Nth kernel. So, to sum it up, when you update to a new kernel, first update the symlink (or let the ebuild take care of it), build the kernel, boot it, *then* rebuild external modules (like ati-drivers). If you know what you're doing, you don't need to reboot before rebuilding external modules, for example I do: ebuild /usr/portage/x11-drivers/ati-drivers/ati-drivers-N.ebuild compile mkdir /lib/modules/KERNEL_VERSION/video cp /var/tmp/portage/x11-drivers/ati-drivers-N/work/common/lib/modules/fglrx/build_mod/2.6.x/fglrx.ko /lib/modules/KERNEL_VERSION/video/ But it general, just boot the new kernel and emerge external modules again; safer and more straight forward.
[gentoo-user] Kernel upgrading and linux symlink
Looking at the kernel upgrade pages at http://www.gentoo.org/doc/en/kernel-upgrade.xml Its a bit confusing about the symlink creation. I've wondered about it a few times. At the top, you're told how to get the sources and then a discussion of the symlink follows. It appears you are expected to change the symlink to the newly installed sources. But then it says: 4. Updating the /usr/src/linux symbolic link "Gentoo requires that the /usr/src/linux symbolic link points to the sources of the kernel you are running." So if gentoo `requires' the symlink to point to the running kernel why are we changing it to the newly installed but not yet compiled or started kernel? And in fact does it really matter if its pointing at the newly installed or actual running kernel, when kernel compiling operations take place?
Re: [gentoo-user] Re: glxgears with no ... gears?!
Mick wrote: > Am I the only one with dodgy glxgears? Any ideas at all? > > On Tuesday 06 October 2009, Mick wrote: >> I've updated xorg following all relevant instructions by gentoo devs on an >> old laptop and when I run glxgears I get no graphic and pedestrian speeds: >> >> $ glxgears >> 629 frames in 5.0 seconds = 125.622 FPS >> 629 frames in 5.0 seconds = 125.718 FPS >> 627 frames in 5.0 seconds = 125.377 FPS >> >> If I maximise the window, still no gears just a black terminal window and I >> get: >> >> 93 frames in 5.0 seconds = 18.474 FPS >> >> It's never been that bad before. I've noticed that emerge -uDv world >> pulled in xorg-drivers and it also emerged xf86-video-ati. I attach the >> list of the packages that were emerged today and my Xorg.0.log, just in >> case you see something amiss in there. >> >> Please ask if you need more info. >> >> PS. I have no xorg.conf, but can post the *.fdi files if needed. > > > No. See recent post: [gentoo-user] Xorg slow after upgrade -- Valmor
Re: [gentoo-user] Xorg slow after upgrade
Krzysztof Poc wrote: > Hello I've recently upgraded my system and I found that Xorg redraws the > screen > very slowly. "top" shows me that "X" process takes up around 95% of the CPU > while I have Intel Core 2 Duo. > > What's wrong with my system ? > > I have the following installed: > kernel 2.6.30.7 #3 SMP (without PAE) > xorg-server-1.6.3.901-r2 > xorg-x11-7.4-r1 > xf86-video-intel-2.8.1 > > I've also successfully migrated to X.org Server 1.6 and libxcb 1.4 as > written > in official gentoo documentation. > > great thanks for any indications > I am using kernel 2.6.30.9 i686 xorg-server 1.6.3.901-r2 xorg-x11 7.4-r1 xf86-video-intel 2.8.1 hal 0.5.12_rc1-r8 Also have some unresolved issues with windowmaker-0.92.0-r8 not accepting background images on workspaces, and glxgears results lower (by half) than I had with xorg-server-1.5: -> glxgears 2095 frames in 5.0 seconds = 418.739 FPS 1800 frames in 5.0 seconds = 359.967 FPS 1881 frames in 5.0 seconds = 376.091 FPS 2097 frames in 5.0 seconds = 419.294 FPS 2097 frames in 5.0 seconds = 419.286 FPS 1887 frames in 5.0 seconds = 377.349 FPS 1891 frames in 5.0 seconds = 378.111 FPS 1911 frames in 5.0 seconds = 382.009 FPS -- Valmor PS: I have fully recompiled, relinked, etc. all that is related to graphics.
[gentoo-user] video capture from min-dv... kernel params
My windowsXP video editing machine where I use various adobe tools to capture and edit video, has lost use of the ohci ports, and seemingly the usb ports as well. I've been seeing problems with those inputs for a while and today, finally the machine simply is not `seeing' the video cam attached to either of two ohci (1394) inputs. Unistall/re-install drivers etc... no soap. I'm going to need a pci card and will change over to 1394b when I get it. Hoping to use gentoo platform to capture video. Googling around this morning, I hoped to find a howto that at least listed what has to be compiled kernel wise. There is a vast amount of hits with strings like `gentoo howto capture video' `linux capture video' but an awful lot of it is anecdotal stuff from forums, seems to be lots of stuff about mythtv, and webcams and of course lots of what turns up is ancient (well, 3-5 yrs old). I wondered if anyone here has used gentoo for mini-dv capture and if so what do I need to do regarding the kernel? I see apps called dvgrab and kino show up in many of the hits too. Is it just a matter of building specific modules or turning on certain parameters in the kernel? Are the tools dvgrab or kino self contained enough that I could do the the capture with one or the other alone? I'm not really concerned with the editing, and will transfer the captured clips to a winxp machine where I can use Adobe premiere to edit them.
Re: [gentoo-user] executing commands on lots of servers at once
On Fri, Oct 30, 2009 at 2:36 AM, James wrote: > I have a cluster of Gentoo boxes I'd like to update all at once, along > with emerging specific packages to the servers simultaneously. > > Does anyone have any experience(s) with a good utility for doing this? > I've seen a few scripts online for accomplishing this but I'm not sure > how these scripts would handle the large amount of output that emerge > generates from all the servers simultaneously. > > Thoughts / ideas appreciated. > > -j > > you can use screen, introduced in last GMN. http://www.gentoo.org/news/en/gmn/20081130-newsletter.xml
Re: [gentoo-user] EeePC network problem
On Friday 30 October 2009 16:48:23 Philip Webb wrote: > How have others got Gentoo installed on their 1005HA's ? > -- did they all use wireless access to the Internet ? You could try the atl1c module, which works on my wife's 1005HA, although lspci shows an Attansic 1062 chip. Long shot maybe, but no harm in trying. Doesn't net-setup eth0 load the right module by itself? -- Rgds Peter
Re: [gentoo-user] Installing an old glibc to run a proprietary commercial tool (would that even help?)
I was in a similar position some years ago - grab a copy of the needed libs from somewhere and use "ldpreload" to load them into memory before running the application. Google will help. In some cases, you can symlink the needed lib names to existing later libs and run ldconfig before trying to run the app. This does work sometimes, but success varies ... BillK On Fri, 2009-10-30 at 19:27 -0400, Albert Hopkins wrote: > On Fri, 2009-10-30 at 14:52 -0700, Kyle Bader wrote: > > I dunno how > > flexible the vendor is but its worth asking :) > > They only support RHEL4. RHEL4 was released nearly 5 years ago and uses > the 2.6.9 kernel. I think that shows how flexible they are. :) > > > -- William Kenworthy Home in Perth!
