Re: [gentoo-user] Sipix Pocket Printer A6
>> I can get the Sipix Pocket Printer A6 installed in CUPS and it will >> get about 20% through a CUPS test page, but it always stops in the >> same spot and the LED just blinks. > > Does it do the same with standard printing jobs? I ask because I have an > Epson printer that failed on the CUPS test page, no matter what I tried, > but then I tried printing a photo from KDE and it worked perfectly. It works! The thing just needed new batteries. For anyone else reading this, flashing LED = replace batteries. It has a few problems: 1. The printer doesn't show up in Print dialogs. It shows up in localhost:631 and prints the test page and via lpr. Why wouldn't it show up in the Print dialogs? Is it because it's a serial printer? 2. I'm using a USB->serial converter and I get a permissions error when trying to print unless I chmod /dev/ttyUSB0. The file shows up before chmod as: crw-rw 1 root uucp The problem with chmod is it resets after unplug/plug. My user is in the uucp group. How can I enable printing after unplug/plug without chmod? 3. lpr doesn't work unless I set: # export CUPS_SERVER=localhost which resets after reboot. This wouldn't really be a problem if I could get #1 fixed above because then I could use the Print dialogs instead of lpr. 4. The printer feeds a lot of paper before it starts to print. Should that be fixed in the .upp or .ppd? 5. Quality isn't great, but there's probably not much to do about that. - Grant
Re: [gentoo-user] Sipix Pocket Printer A6
>>> I can get the Sipix Pocket Printer A6 installed in CUPS and it will >>> get about 20% through a CUPS test page, but it always stops in the >>> same spot and the LED just blinks. >> >> Does it do the same with standard printing jobs? I ask because I have an >> Epson printer that failed on the CUPS test page, no matter what I tried, >> but then I tried printing a photo from KDE and it worked perfectly. > > It works! The thing just needed new batteries. For anyone else > reading this, flashing LED = replace batteries. > > It has a few problems: > > 1. The printer doesn't show up in Print dialogs. It shows up in > localhost:631 and prints the test page and via lpr. Why wouldn't it > show up in the Print dialogs? Is it because it's a serial printer? > > 2. I'm using a USB->serial converter and I get a permissions error > when trying to print unless I chmod /dev/ttyUSB0. The file shows up > before chmod as: > > crw-rw 1 root uucp > > The problem with chmod is it resets after unplug/plug. My user is in > the uucp group. How can I enable printing after unplug/plug without > chmod? > > 3. lpr doesn't work unless I set: > > # export CUPS_SERVER=localhost > > which resets after reboot. This wouldn't really be a problem if I > could get #1 fixed above because then I could use the Print dialogs > instead of lpr. > > 4. The printer feeds a lot of paper before it starts to print. Should > that be fixed in the .upp or .ppd? > > 5. Quality isn't great, but there's probably not much to do about that. > > - Grant I should include info about how I got it to work. Just follow the Gentoo Printing Guide and additionally emerge foomatic-filters-ppds (if it is not already pulled in by cups) and psutils. After adding the printer at localhost:631, the device URI should look like this: serial:/dev/ttyUSB0?baud=115200+bits=8+parity=none+flow=hard Although your serial port device file would be different if you're using a real serial port instead of a USB->serial adapter. The device URI can be verified and edited in /etc/cups/printers.conf. Restart cups after editing that file. The sipixa6.upp file is required and is included with ghostscript-gpl which should be pulled in by foomatic-filters which should be pulled in by foomatic-filters-ppds. - Grant
[gentoo-user] /etc/conf.d/net tries to do too much
I configured my laptop as a wireless AP with hostapd. It works great as long as I 'ifconfig wlan0 192.168.1.1'. I can have /etc/init.d/net.wlan0 started automatically with hostapd, but I can't come up with an /etc/conf.d/net config that will work. It always fails by saying it can't set master mode or other parameters. I think hostapd is supposed to handle all of that except the IP configuration, but /etc/conf.d/net wants to do more than just the IP configuration. Does anyone know the right way to handle this? - Grant
[gentoo-user] Which packages did I unmerge?
Is there any way to find out which packages I unmerged today with depclean? I thought they would show up in /var/log/portage but apparently not. I'm getting a wireless card DMA error since unmerging them. - Grant
Re: [gentoo-user] Gentoo on SSD
> I'm thinking of re-installing Gentoo on an Intel 40 Megs SSD -- excluding the > most often writen dirs like /var, /tmp, /home --. What do you think ? I'll be > glad to hear about previous experiences. What about swap ? Is it safe to have > it > on the SSD ? Why exclude the most written dirs? I thought SSDs were projected to last longer than HDs? Also, from what I've read, SLC should last much longer than MLC. - Grant
Re: [gentoo-user] Which packages did I unmerge?
>> Is there any way to find out which packages I unmerged today with >> depclean? I thought they would show up in /var/log/portage but >> apparently not. I'm getting a wireless card DMA error since unmerging >> them. >> >> - Grant > > > app-portage/genlop Thank you, but I need to specify a package name with genlop, right? I'm trying to figure out which packages were unmerged yesterday by depclean. - Grant
Re: [gentoo-user] Gentoo on SSD
>> I thought SSDs were projected to >> last longer than HDs? Also, from what I've read, SLC should last much >> longer than MLC. > > It's the other way round: HD's last longer dan SSD's. [1] > > [1] http://en.wikipedia.org/wiki/Solid-state_drive#Disadvantages Thanks for the link. I did some Googleing too and I'm really surprised at what I found. It sounds like SSDs don't have the projected longevity they did when I researched this a year or so ago. I'm troubled by the ever-lurking possibility of an HD failure and I thought SSDs would be my way out. Is an HD the best choice for reliability? - Grant
Re: [gentoo-user] Re: Which packages did I unmerge?
>>>> Is there any way to find out which packages I unmerged today with >>>> depclean? I thought they would show up in /var/log/portage but >>>> apparently not. I'm getting a wireless card DMA error since unmerging >>>> them. >>>> >>>> - Grant >>> >>> >>> app-portage/genlop >> >> Thank you, but I need to specify a package name with genlop, right? >> I'm trying to figure out which packages were unmerged yesterday by >> depclean. >> >> - Grant >> >> >> > > genlop -u --date 1 days ago "*" > > or something similar, iirc Perfect, thank you. - Grant
Re: [gentoo-user] {OT} Flashing BIOS with Windows utility
>> XP on a USB stick It sounds like a Windows machine is necessary >> to build it, but once it's built it would be really handy. > > I have found my attempts at building this or a CD-based PE very frustrating > indeed. It seems like an awful lot of aggro when you could just download > exactly the same thing that someone else has already built, if only the > license permitted it. > > Perhaps it's just seemed like a lot of work because it's never been a goal > of mine to make a bootable Windows CD or USB stick - I've always been trying > to build them in order to help me achieve some other goal, and it's been far > more faffing about that I've been prepared to tolerate. > > There used to be plenty of PE builds available by the usual pirate sites, > but I have no idea whether that's still the case. I seem to be getting more > efficient at not needing them. > > Stroller. Has anyone tried winbuilder: http://winbuilder.net/ It's supposed to be great. - Grant
Re: [gentoo-user] rsync backup system
>> How is BackupPC to set up? Is it a whole new world to explore, or can >> it be set up quickly and easily? > > It takes a little while to get the hang of how the config files work, but > once you get it it takes no work at all. Restoring is as simple as > selecting the files you want in a browser and pressing a button. > > > -- > Neil Bothwick Has anyone tried backupninja? There is a new ebuild for it. https://labs.riseup.net/code/projects/show/backupninja/ Is BackupPC too excellent to consider an alternative? I'm going to set up one of these backup systems in the next few weeks. - Grant
Re: [gentoo-user] rsync backup system
>> A much better way is to run a dedicated agent on the client. If the server >> needs to schedule backups, it can ask the agent to do so using regular tcp >> traffic. The client can then do it's backup and rsync it over to the server >> when it's done, and that push can be done as a regular user on both ends. The >> actual backing up on the client must be done by root of course, no other user >> has the necessary access. > > Sounds great. Is there any software that works this way? > > Ward I'd like to know too. :) - Grant
Re: [gentoo-user] xfce4 tray icon problems
>>> I'm having a problem with xfce4 tray icons all of a sudden. The >>> wicd-client icon won't show up, even though the binary executes in a >>> terminal without error. I've tried restarting wicd and rebooting. >>> The twinkle icon shows up in the center of the screen instead of the >>> tray, and it has its own panel button which is separate from the main >>> app's panel button. Does anyone know what could be wrong? >>> >> >> I didn't understand very well what happen with you, possible because my >> bad english vocabulary, but have you tried run wicd-client with --no-tray >> parameter ? >> > Another guess, when I upgraded to xfce 4.6 it was necessarily emerge > x11-themes/tango-icon-theme and set xfce to use it. This package fixed every > icon problemthat I had in xfce menu. I'm sorry, I just saw your post now for some reason. I don't think the problem is with tango-icon-theme because I have 2 identical laptops with very similar configurations and neither of them have that package installed. Only 1 of the laptops is having this problem. The problem is with system tray icons that are not panel plugins. For example: wicd, twinkle, blueman, orage. The icons for those do not appear at all except for the twinkle icon which appears right in the center of the screen. All other icons work fine, and system tray icons that are panel plugins such as xfce4-mixer and clipman work fine. Any ideas? - Grant
Re: [gentoo-user] xfce4 tray icon problems
>>>>> I'm having a problem with xfce4 tray icons all of a sudden. The >>>>> wicd-client icon won't show up, even though the binary executes in a >>>>> terminal without error. I've tried restarting wicd and rebooting. >>>>> The twinkle icon shows up in the center of the screen instead of the >>>>> tray, and it has its own panel button which is separate from the main >>>>> app's panel button. Does anyone know what could be wrong? >>>>> >>>> >>>> I didn't understand very well what happen with you, possible because my >>>> bad english vocabulary, but have you tried run wicd-client with --no-tray >>>> parameter ? >>>> >>> Another guess, when I upgraded to xfce 4.6 it was necessarily emerge >>> x11-themes/tango-icon-theme and set xfce to use it. This package fixed every >>> icon problemthat I had in xfce menu. >> >> I'm sorry, I just saw your post now for some reason. I don't think >> the problem is with tango-icon-theme because I have 2 identical >> laptops with very similar configurations and neither of them have that >> package installed. Only 1 of the laptops is having this problem. >> >> The problem is with system tray icons that are not panel plugins. For >> example: wicd, twinkle, blueman, orage. The icons for those do not >> appear at all except for the twinkle icon which appears right in the >> center of the screen. All other icons work fine, and system tray >> icons that are panel plugins such as xfce4-mixer and clipman work >> fine. >> >> Any ideas? >> >> - Grant >> >> > > Please feel to contribute data and ideas here if appropriate: > > http://bugs.gentoo.org/show_bug.cgi?id=307685 > > It doesn't sound exactly like what I've seen but it's very similar. > > - Mark Thanks Mark. I really think this is a different issue. I emerged tango-icon-theme with no change. I have all icons except those that should be in the system tray, and even those missing system tray icons do appear in the xfce4 menu. There are no blank placeholders or anything like that. They just don't show up in the system tray at all. I did notice this from Twinkle: X Error: BadWindow (invalid Window parameter) 3 I researched it on Google but got nowhere. - Grant
[gentoo-user] Updates = slow firefox
I just updated a lot of packages on my laptop including xorg stuff, the intel-drivers, and firefox. Firefox is running really slowly now, with kind of a lag to everything. Does anyone know of anything to try in order to fix it? Do I need to disable or enable DRI? - Grant
Re: [gentoo-user] Updates = slow firefox
> did etc-update over write xorg.conf ? I actually don't use an xorg.conf at all. - Grant >> I just updated a lot of packages on my laptop including xorg stuff, >> the intel-drivers, and firefox. Firefox is running really slowly now, >> with kind of a lag to everything. Does anyone know of anything to try >> in order to fix it? Do I need to disable or enable DRI? >> >> - Grant
Re: [gentoo-user] Updates = slow firefox
>> ummm .. ok ? so your using the default xorg.conf .. could be why youve got >> bad performance. >> >> http://www.gentoo.org/doc/en/xorg-config.xml >> >> - >> deface > But the new version of xorg (from circa 1.7) dont really need xorg.conf, > moreover its discouraged to use one. (or just i think so) Exactly. > Grant, try to reemerge firefox (and if You haven't done it already the > x11-drivers/*) I re-emerged them with no change. I do think it has to do with x11-drivers/xf86-video-intel though. I've had this problem in the past, and the solution was to mask >x11-drivers/xf86-video-intel-2.7.1. Unfortunately, those drivers don't work with the latest xorg updates and now I'm on xf86-video-intel-2.9.1. My wife has an identical laptop with the same issue. - Grant
Re: [gentoo-user] Updates = slow firefox
>>> ummm .. ok ? so your using the default xorg.conf .. could be why youve got >>> bad performance. >>> >>> http://www.gentoo.org/doc/en/xorg-config.xml >>> >>> - >>> deface >> But the new version of xorg (from circa 1.7) dont really need xorg.conf, >> moreover its discouraged to use one. (or just i think so) > > You only dont need xorg.conf IF it works without itand even if it > does work without it, its unlikely to be the best config. > > grep ^\(EE /var/log/Xorg.0.log and post the result. Could this be the problem? # grep ^\(EE /var/log/Xorg.0.log (EE) Failed to load module "vesa" (module does not exist, 0) (EE) Failed to load module "fbdev" (module does not exist, 0) (EE) intel(0): [drm] Failed to open DRM device for : No such file or directory (EE) intel(0): Failed to become DRM master. (EE) intel(0): Failed to initialize kernel memory manager - Grant
Re: [gentoo-user] Updates = slow firefox
>> Could this be the problem? >> >> # grep ^\(EE /var/log/Xorg.0.log >> (EE) Failed to load module "vesa" (module does not exist, 0) >> (EE) Failed to load module "fbdev" (module does not exist, 0) >> (EE) intel(0): [drm] Failed to open DRM device for : No such file or >> directory >> (EE) intel(0): Failed to become DRM master. >> (EE) intel(0): Failed to initialize kernel memory manager > > The first two errors are fine; Xorg defaults to trying vesa and fbdev as > display drivers and you just don't have them. > > The last three are your problem. The intel video driver is unable to > properly access the DRM subsystem, which will definitely cause X to slow > to a crawl. > > The most likely cause of your errors is that the intel AGP driver (i810 > or i915, depending on your hardware) isn't getting loaded. If that's > the case, you should see an error such as: > > [drm] failed to load kernel module "i915" > > in Xorg.0.log just before the ones from intel. If the modules are being > loaded, you'll likely see some other errors around that same area. The > aren't tagged with (EE), unfortunately; try: > > # grep -5 'Failed to open DRM' Xorg.0.log > > You can also check your dmesg output to see if the devices are being > initialized properly: > > platypus log # dmesg | grep agp > Linux agpgart interface v0.103 > agpgart-intel :00:00.0: Intel 965GM Chipset > agpgart-intel :00:00.0: detected 7676K stolen memory > agpgart-intel :00:00.0: AGP aperture is 256M @ 0xe000 > > platypus log # dmesg | grep drm > [drm] Initialized drm 1.1.0 20060810 > [drm] set up 7M of stolen space > [drm] initialized overlay support > fb0: inteldrmfb frame buffer device > [drm] Initialized i915 1.6.0 20080730 for :00:02.0 on minor 0 > > If everything's working, you should have the following devices that the > Xorg driver needs: > > platypus log # ls -l /dev/dri > total 0 > crw-rw 1 root video 226, 0 Apr 20 13:11 card0 > crw-rw 1 root video 226, 64 Apr 20 13:11 controlD64 Ah, thank you so much. I needed to enable CONFIG_DRM_I915 in the kernel. - Grant
Re: [gentoo-user] Updates = slow firefox
>>> Could this be the problem? >>> >>> # grep ^\(EE /var/log/Xorg.0.log >>> (EE) Failed to load module "vesa" (module does not exist, 0) >>> (EE) Failed to load module "fbdev" (module does not exist, 0) >>> (EE) intel(0): [drm] Failed to open DRM device for : No such file or >>> directory >>> (EE) intel(0): Failed to become DRM master. >>> (EE) intel(0): Failed to initialize kernel memory manager >> >> The first two errors are fine; Xorg defaults to trying vesa and fbdev as >> display drivers and you just don't have them. >> >> The last three are your problem. The intel video driver is unable to >> properly access the DRM subsystem, which will definitely cause X to slow >> to a crawl. >> >> The most likely cause of your errors is that the intel AGP driver (i810 >> or i915, depending on your hardware) isn't getting loaded. If that's >> the case, you should see an error such as: >> >> [drm] failed to load kernel module "i915" >> >> in Xorg.0.log just before the ones from intel. If the modules are being >> loaded, you'll likely see some other errors around that same area. The >> aren't tagged with (EE), unfortunately; try: >> >> # grep -5 'Failed to open DRM' Xorg.0.log >> >> You can also check your dmesg output to see if the devices are being >> initialized properly: >> >> platypus log # dmesg | grep agp >> Linux agpgart interface v0.103 >> agpgart-intel :00:00.0: Intel 965GM Chipset >> agpgart-intel :00:00.0: detected 7676K stolen memory >> agpgart-intel :00:00.0: AGP aperture is 256M @ 0xe000 >> >> platypus log # dmesg | grep drm >> [drm] Initialized drm 1.1.0 20060810 >> [drm] set up 7M of stolen space >> [drm] initialized overlay support >> fb0: inteldrmfb frame buffer device >> [drm] Initialized i915 1.6.0 20080730 for :00:02.0 on minor 0 >> >> If everything's working, you should have the following devices that the >> Xorg driver needs: >> >> platypus log # ls -l /dev/dri >> total 0 >> crw-rw 1 root video 226, 0 Apr 20 13:11 card0 >> crw-rw 1 root video 226, 64 Apr 20 13:11 controlD64 > > Ah, thank you so much. I needed to enable CONFIG_DRM_I915 in the kernel. > > - Grant Strangely, now my laptop's brightness adjustment doesn't work via the keyboard shortcuts. Any ideas on that? - Grant
Re: [gentoo-user] Updates = slow firefox
>> Strangely, now my laptop's brightness adjustment doesn't work via the >> keyboard shortcuts. Any ideas on that? >> >> - Grant >> > Please share the beast model :P (or maybe ive missed it). > in kernel config You have multiple option for backlight eg. for thinkpad > there is extra one in "thinkpad specific acpi" maybe You have something > similar for Yours stuff. It's a Dell Vostro 1320. The keyboard shortcuts to change brightness were working great until I enabled DRM in the kernel. Can you tell me where in the kernel those options can be found, or part of the variable name that defines them? > And (its only my private opinion - could base on wrong facts :P) dont be > used to hal because the 1.8 xorg-server dont like it any more, > preferring udev, and future versions wouldn't probably support hal at > all. > > Lately i delete hal USE and now iam using udev - excepting auto mounting > usb stick etc. Can anyone confirm that as users we should be moving away from hal? - Grant
Re: [gentoo-user] Updates = slow firefox
>> Can anyone confirm that as users we should be moving away from hal? > > http://www.gentoo.org/proj/en/desktop/x/x11/xorg-server-1.8-upgrade-guide.xml > http://xorg.freedesktop.org/wiki/XorgHAL OK, and since xorg-server-1.7 doesn't have a udev USE flag, I should probably stick with hal until 1.8. Please let me know if that isn't the case. I'm on udev-149. - Grant
Re: [gentoo-user] Updates = slow firefox
>>>> Can anyone confirm that as users we should be moving away from hal? >>> >>> http://www.gentoo.org/proj/en/desktop/x/x11/xorg-server-1.8-upgrade-guide.xml >>> http://xorg.freedesktop.org/wiki/XorgHAL >> >> OK, and since xorg-server-1.7 doesn't have a udev USE flag, I should >> probably stick with hal until 1.8. Please let me know if that isn't >> the case. I'm on udev-149. > > If HAL is working for you, stick with it. If not, turn it off. Xorg > 1.7 works equally well with or without HAL. The main difference is how > much manually configuration you need to do. > > The relative stability of using/not using udev with Xorg 1.8 have yet to > determined. Could switching to udev from hal fix my brightness adjustment keys? - Grant
[gentoo-user] AT&T DSL + Westell modem/router = Gentoo woes
I recently switched from cable to DSL on a Gentoo router and I'm having trouble keeping a stable connection. Periodically I need to run '/etc/init.d/net.eth0 restart' before it will connect to the Westell modem (which is also a router although AT&T won't admit it). The weird thing is that rebooting the system doesn't reacquire the connection. I have to actually issue '/etc/init.d/net.eth0 restart'. Does anyone know why rebooting doesn't reacquire the connection and restarting the interface does? - Grant
Re: [gentoo-user] AT&T DSL + Westell modem/router = Gentoo woes
>> I recently switched from cable to DSL on a Gentoo router and I'm >> having trouble keeping a stable connection. Periodically I need to >> run '/etc/init.d/net.eth0 restart' before it will connect to the >> Westell modem (which is also a router although AT&T won't admit it). >> The weird thing is that rebooting the system doesn't reacquire the >> connection. I have to actually issue '/etc/init.d/net.eth0 restart'. >> Does anyone know why rebooting doesn't reacquire the connection and >> restarting the interface does? >> >> - Grant >> > I just use the modem to login and then enable ip pass thru to my router > which is a linksys running DD-WRT; > http://www.dslreports.com/faq/7073 > > This set-up has worked well for a few years without me having to do > anything. Could I get you to tell me a couple of your settings? You're doing something right if you haven't had to touch your's after setting it up. I'm trying to figure out the right connection type (or something like that) for which the options are something like Smart KeepAlive, Always Reconnect, and Connect On Demand. The other one is the IP lease timeout. It defaults to 10 minutes and I just set it to 99 days, but I don't want it to expire even after 99 days. I have a static IP but I think this is the timeout between the modem/router and the Gentoo router. - Grant
Re: [gentoo-user] AT&T DSL + Westell modem/router = Gentoo woes
>> >> I recently switched from cable to DSL on a Gentoo router and I'm >> >> having trouble keeping a stable connection. Periodically I need to >> >> run '/etc/init.d/net.eth0 restart' before it will connect to the >> >> Westell modem (which is also a router although AT&T won't admit it). >> >> The weird thing is that rebooting the system doesn't reacquire the >> >> connection. I have to actually issue '/etc/init.d/net.eth0 restart'. >> >> Does anyone know why rebooting doesn't reacquire the connection and >> >> restarting the interface does? >> >> >> >> - Grant >> > >> > I just use the modem to login and then enable ip pass thru to my router >> > which is a linksys running DD-WRT; >> > http://www.dslreports.com/faq/7073 >> > >> > This set-up has worked well for a few years without me having to do >> > anything. >> >> Could I get you to tell me a couple of your settings? You're doing >> something right if you haven't had to touch your's after setting it >> up. >> >> I'm trying to figure out the right connection type (or something like >> that) for which the options are something like Smart KeepAlive, Always >> Reconnect, and Connect On Demand. >> >> The other one is the IP lease timeout. It defaults to 10 minutes and >> I just set it to 99 days, but I don't want it to expire even after 99 >> days. I have a static IP but I think this is the timeout between the >> modem/router and the Gentoo router. > > I am not familiar with the modem in question, but if you are using your own > router the modem should be set up in fully bridged mode and the PPPoE > authentication will be managed by your Gentoo router. > > Your Gentoo router will also use dhcpcd (or similar) to obtain an IP address > from your ISP after it authenticates on their RADIUS server. > > In a nut-shell: You want to set up your modem to not do NAT, or dhcp, or > authentication at all, but transparently encapsulate your PPP into ATM packets > and send them off to your local DSLAM. All the communication with your ISP > will be managed by the router and your router will have the static IP address > given to you by your ISP. The modem will only have a LAN address which you > will need to set up manually on it, using its control panel. > > PS. I think that the modem connection type you refer to above should be Smart > KeepAlive - although as I said I'm not familiar with the particular hardware. > > HTH. > -- > Regards, > Mick Thanks Mick. The Westell does have an option to take PPPoE off of the device and I'd like to set that up soon. - Grant
Re: [gentoo-user] AT&T DSL + Westell modem/router = Gentoo woes
>> >> I recently switched from cable to DSL on a Gentoo router and I'm >> >> having trouble keeping a stable connection. Periodically I need to >> >> run '/etc/init.d/net.eth0 restart' before it will connect to the >> >> Westell modem (which is also a router although AT&T won't admit it). >> >> The weird thing is that rebooting the system doesn't reacquire the >> >> connection. I have to actually issue '/etc/init.d/net.eth0 restart'. >> >> Does anyone know why rebooting doesn't reacquire the connection and >> >> restarting the interface does? >> >> >> >> - Grant >> >> >> > I just use the modem to login and then enable ip pass thru to my router >> > which is a linksys running DD-WRT; >> > http://www.dslreports.com/faq/7073 >> > >> > This set-up has worked well for a few years without me having to do >> > anything. >> >> Could I get you to tell me a couple of your settings? You're doing >> something right if you haven't had to touch your's after setting it >> up. >> >> I'm trying to figure out the right connection type (or something like >> that) for which the options are something like Smart KeepAlive, Always >> Reconnect, and Connect On Demand. > I use Always On Are you sure that's what it's called? I don't have that option. >> The other one is the IP lease timeout. It defaults to 10 minutes and >> I just set it to 99 days, but I don't want it to expire even after 99 >> days. I have a static IP but I think this is the timeout between the >> modem/router and the Gentoo router. > I have DHCP lease time once per day And you don't have to restart your network interface once per day? I seem to need to restart the nic as often as that timeout setting. Any idea why that would happen? - Grant
Re: [gentoo-user] Updates = slow firefox
>> >> Strangely, now my laptop's brightness adjustment doesn't work via the >> >> keyboard shortcuts. Any ideas on that? >> >> >> >> - Grant >> >> >> > Please share the beast model :P (or maybe ive missed it). >> > in kernel config You have multiple option for backlight eg. for thinkpad >> > there is extra one in "thinkpad specific acpi" maybe You have something >> > similar for Yours stuff. >> >> It's a Dell Vostro 1320. The keyboard shortcuts to change brightness >> were working great until I enabled DRM in the kernel. Can you tell me >> where in the kernel those options can be found, or part of the >> variable name that defines them? >> > Try running xev and punching brightness keys, if you would see > "effects" (some text in terminal) then its OK :P You should change the > Acpi configs (etc/acpi/) or Gnome/KDE/Xfce/... bindings. I do see text in xev when pressing the brightness keys. > (if You dont know it already) > For acpi config You'll need "event id" try running acpi_listen. > eg. /etc/acpi/events/sleep: > event=ibm/hotkey HKEY 0080 00001004 > action=/etc/acpi/actions/sleep.sh > > and into actions you put scripts, try using xbacklight. You think I should use xbacklight or similar even though it was working on its own before? - Grant > If You wouldn't have any reaction in xev and acpi_listen i check the > option in kernel.
[gentoo-user] {OT} hardening SSL without rejecting users
I've been advised to harden my SSL in the following ways: 1. disable SSL 2.0 2. disable use of SSL ciphers which offer either weak or no encryption 3. disable anonymous SSL ciphers Will some website users not be able to use https if I do this? - Grant
Re: [gentoo-user] Updates = slow firefox
>> >> >> Strangely, now my laptop's brightness adjustment doesn't work via the >> >> >> keyboard shortcuts. Any ideas on that? >> >> >> >> >> >> - Grant >> >> >> >> >> > Please share the beast model :P (or maybe ive missed it). >> >> > in kernel config You have multiple option for backlight eg. for thinkpad >> >> > there is extra one in "thinkpad specific acpi" maybe You have something >> >> > similar for Yours stuff. >> >> >> >> It's a Dell Vostro 1320. The keyboard shortcuts to change brightness >> >> were working great until I enabled DRM in the kernel. Can you tell me >> >> where in the kernel those options can be found, or part of the >> >> variable name that defines them? >> >> >> > Try running xev and punching brightness keys, if you would see >> > "effects" (some text in terminal) then its OK :P You should change the >> > Acpi configs (etc/acpi/) or Gnome/KDE/Xfce/... bindings. >> >> I do see text in xev when pressing the brightness keys. >> >> > (if You dont know it already) >> > For acpi config You'll need "event id" try running acpi_listen. >> > eg. /etc/acpi/events/sleep: >> > event=ibm/hotkey HKEY 0080 1004 >> > action=/etc/acpi/actions/sleep.sh >> > >> > and into actions you put scripts, try using xbacklight. >> >> You think I should use xbacklight or similar even though it was >> working on its own before? >> >> - Grant >> >> >> > If You wouldn't have any reaction in xev and acpi_listen i check the >> > option in kernel. >> > I think that its better to have things done even if would be around then > dont done it at all. :) Yes but I think I should find the built-in mechanism which was allowing it to work before instead of writing my own script to make it work. Don't you think so? - Grant
Re: [gentoo-user] Updates = slow firefox
>> >> >> >> Strangely, now my laptop's brightness adjustment doesn't work via >> >> >> >> the >> >> >> >> keyboard shortcuts. Any ideas on that? >> >> >> >> >> >> >> >> - Grant >> >> >> >> >> >> >> > Please share the beast model :P (or maybe ive missed it). >> >> >> > in kernel config You have multiple option for backlight eg. for >> >> >> > thinkpad >> >> >> > there is extra one in "thinkpad specific acpi" maybe You have >> >> >> > something >> >> >> > similar for Yours stuff. >> >> >> >> >> >> It's a Dell Vostro 1320. The keyboard shortcuts to change brightness >> >> >> were working great until I enabled DRM in the kernel. Can you tell me >> >> >> where in the kernel those options can be found, or part of the >> >> >> variable name that defines them? >> >> >> >> >> > Try running xev and punching brightness keys, if you would see >> >> > "effects" (some text in terminal) then its OK :P You should change the >> >> > Acpi configs (etc/acpi/) or Gnome/KDE/Xfce/... bindings. >> >> >> >> I do see text in xev when pressing the brightness keys. >> >> >> >> > (if You dont know it already) >> >> > For acpi config You'll need "event id" try running acpi_listen. >> >> > eg. /etc/acpi/events/sleep: >> >> > event=ibm/hotkey HKEY 0080 1004 >> >> > action=/etc/acpi/actions/sleep.sh >> >> > >> >> > and into actions you put scripts, try using xbacklight. >> >> >> >> You think I should use xbacklight or similar even though it was >> >> working on its own before? >> >> >> >> - Grant >> >> >> >> >> >> > If You wouldn't have any reaction in xev and acpi_listen i check the >> >> > option in kernel. >> >> >> > I think that its better to have things done even if would be around then >> > dont done it at all. :) >> >> Yes but I think I should find the built-in mechanism which was >> allowing it to work before instead of writing my own script to make it >> work. Don't you think so? >> >> - Grant >> > Try built in Gnome\Kde\Xfce(etc) bindings i had some troubles (in xfce) > - keys with names XF86* starts to randomly changes names or disappear > from configs ... maybe its Your case too. > -- > Bartosz Szatkowski Got it, thank you for your help with this. I used xbacklight along with the xfce4 keyboard shortcut GUI settings. My backlight adjustment keystrokes are displayed as XF86MonBrightnessUp and *Down in those settings, so there must have been a mechanism adjusting the backlight based on that before I updated Xorg. Here are my xbacklight commands: xbacklight -inc 15 -steps 1 -time 0 xbacklight -dec 10 -steps 1 -time 0 Thanks again, Grant
Re: [gentoo-user] AT&T DSL + Westell modem/router = Gentoo woes
>> > I am not familiar with the modem in question, but if you are using your >> > own router the modem should be set up in fully bridged mode and the PPPoE >> > authentication will be managed by your Gentoo router. > >> Thanks Mick. The Westell does have an option to take PPPoE off of the >> device and I'd like to set that up soon. > > Right, without the PPPoE disabled the modem operates in a half-bridged mode > (essentially it is a router running dhcp and 1:1 NAT). I would disable NAT, > dhcp and PPPoE on the modem so that it is in fully-bridged mode and then see > if the problem is resolved. Your symptoms are typical of a half-bridged router > and a dynamic IP address from the ISP. Usually, the client on the LAN does > not know when the ISP's WAN side IP address has changed and will not pick up > the new address until the dhcp lease on the client has expired. To overcome > this botched implementation the half-bridged modem has a short lease. This > doesn't always work, as I suspect is the case here. > > HTH. > -- > Regards, > Mick Thank you Mick, I'm going to set that up ASAP. - Grant
[gentoo-user] Print server with hplip
I've been using a print and scan server with an Epson printer/scanner for a long time. I'm trying to set up the same thing with an HP and I've got everything working except remote printing. Local printing, local scanning and remote scanning work, but not remote printing. I have the client's IP in /etc/cups/cupsd.conf and the server's IP in /etc/cups/client.conf but the client won't pick up the server's HP for printing. Could there be any extra configuration required for remote printing when switching from gutenprint (I think) to hplip? I had to add saned to the lp group on the server before I could get remote scanning to work (not required with the Epson), and I'm wondering if there could be a similar detail I'm overlooking with remote printing. - Grant
[gentoo-user] Re: Print server with hplip
> I've been using a print and scan server with an Epson printer/scanner > for a long time. I'm trying to set up the same thing with an HP and > I've got everything working except remote printing. Local printing, > local scanning and remote scanning work, but not remote printing. I > have the client's IP in /etc/cups/cupsd.conf and the server's IP in > /etc/cups/client.conf but the client won't pick up the server's HP for > printing. > > Could there be any extra configuration required for remote printing > when switching from gutenprint (I think) to hplip? I had to add saned > to the lp group on the server before I could get remote scanning to > work (not required with the Epson), and I'm wondering if there could > be a similar detail I'm overlooking with remote printing. > > - Grant I guess hplip uses something called "JetDirect" for network printing. hplip in general seems to want to do things its own way. I don't think I'll bother with it. I'm going to get rid of this HP and get an Epson anyway since scan quality is sub-par. - Grant
[gentoo-user] new user can't run X apps
I created a new user with useradd but I get X errors like "cannot open display" and "cannot connect to X server" when I try to run X apps as the new user. I've tried restarting with the same results. Does anyone know why this is happening? - Grant
Re: [gentoo-user] new user can't run X apps
>> I created a new user with useradd but I get X errors like "cannot open >> display" and "cannot connect to X server" when I try to run X apps as >> the new user. I've tried restarting with the same results. Does >> anyone know why this is happening? > > Is the user in the video group? Yes, I should have said that. - Grant
[gentoo-user] I've been hacked.
I nmap'ed one of my remote Gentoo servers today and besides the expected open ports were these: 1080/tcp open socks 3128/tcp open squid-http 8080/tcp open http-proxy I'm not running any sort of proxy software that I know of and I should be the only person whatsoever with access to the machine. 'netstat -l' doesn't show any info on those ports at all so I suppose it's been hacked as well? I installed and ran 'rkhunter --check' (what happened to the chrootkit ebuild?) but it doesn't seem to be much use since I hadn't established a "file of stored file properties". What do you guys think is going on? What should I do from here? - Grant
Re: [gentoo-user] I've been hacked.
>> I nmap'ed one of my remote Gentoo servers today and besides the >> expected open ports were these: >> >> 1080/tcp open socks >> 3128/tcp open squid-http >> 8080/tcp open http-proxy >> >> I'm not running any sort of proxy software that I know of and I should >> be the only person whatsoever with access to the machine. 'netstat >> -l' doesn't show any info on those ports at all so I suppose it's been >> hacked as well? I installed and ran 'rkhunter --check' (what happened >> to the chrootkit ebuild?) but it doesn't seem to be much use since I >> hadn't established a "file of stored file properties". >> >> What do you guys think is going on? What should I do from here? > > What does lsof (I'd reinstall it afresh) show with regards to strange users? > What users the above services run under. If indeed they are not legitimate > and you confirm that they are not being run as packages that you installed, > then I'm afraid the only sane option is to reinstall. Wow. I'm actually seeing the same thing from other domains I nmap. Could my ISP have some kind of a weird environment set up that makes it look like there are ports such as these open on remote systems? Right now I'm on some kind of a shared connection where everyone has their own modem or router or whatever it is, but I think everyone's IP is the same. - Grant
Re: [gentoo-user] I've been hacked.
>>>>> I nmap'ed one of my remote Gentoo servers today and besides the >>>>> expected open ports were these: >>>>> >>>>> 1080/tcp open socks >>>>> 3128/tcp open squid-http >>>>> 8080/tcp open http-proxy >>>>> >>>>> I'm not running any sort of proxy software that I know of and I should >>>>> be the only person whatsoever with access to the machine. 'netstat >>>>> -l' doesn't show any info on those ports at all so I suppose it's been >>>>> hacked as well? I installed and ran 'rkhunter --check' (what happened >>>>> to the chrootkit ebuild?) but it doesn't seem to be much use since I >>>>> hadn't established a "file of stored file properties". >>>>> >>>>> What do you guys think is going on? What should I do from here? >>>>> >>>> >>>> What does lsof (I'd reinstall it afresh) show with regards to strange >>>> users? >>>> What users the above services run under. If indeed they are not >>>> legitimate >>>> and you confirm that they are not being run as packages that you >>>> installed, >>>> then I'm afraid the only sane option is to reinstall. >>>> >>> >>> Wow. I'm actually seeing the same thing from other domains I nmap. >>> Could my ISP have some kind of a weird environment set up that makes >>> it look like there are ports such as these open on remote systems? >>> Right now I'm on some kind of a shared connection where everyone has >>> their own modem or router or whatever it is, but I think everyone's IP >>> is the same. >>> >>> - Grant >>> >>> >> >> Hello, >> >> looks like, your ISP has a Transparent Proxy Setup running. Should I be worried about that? > Ports being shown as open does not mean that your machine is > listening, more like the firewall has some holes in it. If the Really? I thought a service had to be listening for the port to be open. So from nmap, there is no way to tell the difference between a port that isn't blocked by a firewall and one that is listening? > firewall is not configured/running on your server itself, then you may > be alright. Can you actually connect to your server using those > ports? If I enter the server's IP appended with one of the port numbers listed above into a web browser, I get: "tinyproxy 1.6.0 The page you requested was unavailable. The error code is listed below. In addition, the HTML file which has been configured as the page to be displayed when an error of this type was unavailable, with the error code 14 (Bad address). Please contact your administrator. Bad Request" The thing is, I get the same thing from any domain I enter appended with one of those ports. > Have you tried telnet, or nc -v -z to see if > they are open? Can you tell me what package nc is included in? - Grant
[gentoo-user] SIP client with bluetooth headset support?
My bluetooth headset was working great with twinkle, but now it's out of the tree and I can't find another SIP client that works with a bluetooth headset. Does anybody know of one? linphone is said to work, but I can't make it happen even after an exhaustive effort. - Grant
[gentoo-user] Re: SIP client with bluetooth headset support?
> My bluetooth headset was working great with twinkle, but now it's out > of the tree and I can't find another SIP client that works with a > bluetooth headset. Does anybody know of one? linphone is said to > work, but I can't make it happen even after an exhaustive effort. > > - Grant It works with skype. Please let me know if you have any ideas on getting it to work with linphone. I've added the following to ~/.linphonerc: alsadev=pcm.bluetooth and it appears in the linphone audio devices, but it doesn't function at all. - Grant
[gentoo-user] glx & dri fail to load for nv
I'm seeing tearing from the nv xorg driver. I think it's because the glx and dri modules are failing to load, and I think that's because I'm missing a kernel option or two. Can anyone tell me what I might be missing in the kernel for an Nvidia card? BTW, the proprietary nvidia driver works great but I'd like to get nv working too. - Grant
Re: [gentoo-user] Re: glx & dri fail to load for nv
>> I'm seeing tearing from the nv xorg driver. I think it's because the >> glx and dri modules are failing to load, and I think that's because >> I'm missing a kernel option or two. Can anyone tell me what I might >> be missing in the kernel for an Nvidia card? >> >> BTW, the proprietary nvidia driver works great but I'd like to get nv >> working too. > > A bit off-topic, but nv is dead upstream (NVidia dropped it). You might > want to look into the Nouveau open source driver if you have hardware > supported by it (it even supports 3D): > > http://nouveau.freedesktop.org > > It's better than the crappy "nv" driver by orders of magnitude. That's great! I'll ask the Nouveau list about VDPAU support. - Grant
[gentoo-user] nouveau-drm compile failure
I'm trying to compile nouveau-drm for my Nvidia 8400GS video card, but compilation fails: /var/tmp/portage/x11-base/nouveau-drm-20100316/work/master/drivers/gpu/drm/nouveau/nouveau_drv.c: In function 'nouveau_pci_suspend': /var/tmp/portage/x11-base/nouveau-drm-20100316/work/master/drivers/gpu/drm/nouveau/nouveau_drv.c:232: error: implicit declaration of function 'acquire_console_sem' /var/tmp/portage/x11-base/nouveau-drm-20100316/work/master/drivers/gpu/drm/nouveau/nouveau_drv.c:234: error: implicit declaration of function 'release_console_sem' I get kernel config warnings: * CONFIG_FB_CFB_FILLRECT: is not set when it should be. * CONFIG_FB_CFB_COPYAREA: is not set when it should be. * CONFIG_FB_CFB_IMAGEBLIT:is not set when it should be. But those kernel options are required to be automatically selected by a list of dependant options which is very long and very ridiculous. Are those options the reason for the compile failure? - Grant
Re: [gentoo-user] Re: nouveau-drm compile failure
>> I'm trying to compile nouveau-drm for my Nvidia 8400GS video card, but >> compilation fails: > > I don't think you should be using nouveau-drm in the first place. This > driver is now in the kernel itself. nouveau-drm was used before that driver > moved into the Linux kernel together with the other DRM drivers. Well, the xf86-video-nouveau-0.0.16_pre20100510 ebuild says: Nouveau DRM not detected. If you want any kind of acceleration with nouveau, emerge x11-base/nouveau-drm or enable CONFIG_DRM_NOUVEAU in the kernel. Nouveau DRM in kernel 2.6.33 is API incompatible to >=x11-libs/libdrm-2.4.18, please use x11-base/nouveau-drm instead. Unfortunately, the same ebuild depends on >=x11-libs/libdrm-2.4.19. I tried to ignore the warning but X fails with: (EE) Failed to load module "dri" (module does not exist, 0) (EE) Failed to load module "dri2" (module does not exist, 0) (EE) [drm] No DRICreatePCIBusID symbol (EE) No devices detected. The xf86-video-nouveau-0.0.15_pre20100213 ebuild only depends on >=x11-libs/libdrm-2.4.18_pre so I installed those but I still get the errors above when trying to start X. Any ideas? Do you have nouveau working? - Grant
Re: [gentoo-user] Re: nouveau-drm compile failure
>>> I'm trying to compile nouveau-drm for my Nvidia 8400GS video card, but >>> compilation fails: >> >> I don't think you should be using nouveau-drm in the first place. This >> driver is now in the kernel itself. nouveau-drm was used before that driver >> moved into the Linux kernel together with the other DRM drivers. > > Well, the xf86-video-nouveau-0.0.16_pre20100510 ebuild says: > > Nouveau DRM not detected. If you want any kind of > acceleration with nouveau, emerge x11-base/nouveau-drm or > enable CONFIG_DRM_NOUVEAU in the kernel. > Nouveau DRM in kernel 2.6.33 is API incompatible to >>=x11-libs/libdrm-2.4.18, please use x11-base/nouveau-drm > instead. > > Unfortunately, the same ebuild depends on >=x11-libs/libdrm-2.4.19. I > tried to ignore the warning but X fails with: > > (EE) Failed to load module "dri" (module does not exist, 0) > (EE) Failed to load module "dri2" (module does not exist, 0) > (EE) [drm] No DRICreatePCIBusID symbol > (EE) No devices detected. > > The xf86-video-nouveau-0.0.15_pre20100213 ebuild only depends on >>=x11-libs/libdrm-2.4.18_pre so I installed those but I still get the > errors above when trying to start X. > > Any ideas? Do you have nouveau working? > > - Grant I forgot to mention that I have CONFIG_DRM_NOUVEAU enabled in my 2.6.33 kernel and nouveau-firmware is loading successfully. - Grant
Re: [gentoo-user] Re: nouveau-drm compile failure
>> > I'm trying to compile nouveau-drm for my Nvidia 8400GS video card, but >> >> > compilation fails: >> I don't think you should be using nouveau-drm in the first place. This >> driver is now in the kernel itself. nouveau-drm was used before that >> driver moved into the Linux kernel together with the other DRM drivers. >> >> Also, as a consequence, trying to use nouveau-drm means you're getting >> an outdated driver, since AFAIK the updates happen in-kernel now. > > There's an API break between the nouveau driver and the drm in 2.6.3[23] (not > completely sure about the versions) > > Hence the workaround of nouveau-drm > > Hopefully it will be *very* temporary Have you installed nouveau-drm? If so, do you get kernel config warnings during the emerge? - Grant
Re: [gentoo-user] Re: nouveau-drm compile failure
>> There's an API break between the nouveau driver and the drm in 2.6.3[23] (not >> completely sure about the versions) >> >> Hence the workaround of nouveau-drm >> >> Hopefully it will be *very* temporary > > It is. 2.6.34 (which is in ~arch) fixes the API break with the Nouveau > driver. Supposedly it's an API break between the kernel and libdrm: "Nouveau DRM in kernel 2.6.33 is API incompatible to >=x11-libs/libdrm-2.4.18, please use x11-base/nouveau-drm instead." According to ebuild dependencies and messages, my combination of xf86-video-nouveau-0.0.15_pre20100213 and x11-libs/libdrm-2.4.18_pre should work. I get this same error no matter what combination of libdrm, xf86-video-nouveau, and nouveau-firmware I try, and whether or not I have CONFIG_DRM_NOUVEAU=y. This makes me think it's not related to the API break: (EE) Failed to load module "dri" (module does not exist, 0) (EE) Failed to load module "dri2" (module does not exist, 0) (EE) [drm] No DRICreatePCIBusID symbol (EE) No devices detected. Google says the error can be due to disabling dri in xorg.conf, but I don't have anything like that. Should CONFIG_DRM_NOUVEAU=y include the dri or dri2 module? - Grant
Re: [gentoo-user] Re: nouveau-drm compile failure
>> Google says the error can be due to disabling dri in xorg.conf, but I >> don't have anything like that. Should CONFIG_DRM_NOUVEAU=y include >> the dri or dri2 module? > > The dri and dri2 modules should be installed by x11-base/xorg-server. I've got all of these: /usr/lib64/opengl/xorg-x11/extensions/libdri.so /usr/lib64/opengl/xorg-x11/extensions/libdri2.so /usr/lib/opengl/xorg-x11/extensions/libdri2.so /usr/lib/opengl/xorg-x11/extensions/libdri.so but I get the dri and dri2 "module does not exist" errors. Can I specify a path in xorg.conf, or does anyone know what the problem might be? - Grant
[gentoo-user] Preventing a shared interrupt
Does anyone know how to prevent my "nouveau" video card and "ath" wifi card from sharing interrupt 19? I can't move the video card slot and I'd rather not move the wifi card slot if possible. BTW, does anyone see any other opportunity for interrupt optimization here? I'm trying to optimize for video and especially audio, and I'm using a USB 1.0 sound card. # cat /proc/interrupts CPU0 0: 473139 IO-APIC-edge timer 1: 2 IO-APIC-edge i8042 7: 1 IO-APIC-edge 8:127 IO-APIC-edge rtc0 9: 0 IO-APIC-fasteoi acpi 12: 3 IO-APIC-edge i8042 14: 0 IO-APIC-edge ide0 19: 135597 IO-APIC-fasteoi nouveau, ath 20: 2 IO-APIC-fasteoi ehci_hcd:usb2 21: 13860 IO-APIC-fasteoi ehci_hcd:usb1 22: 971970 IO-APIC-fasteoi ohci_hcd:usb4 23: 3446 IO-APIC-fasteoi ohci_hcd:usb3 28: 21524 PCI-MSI-edge ahci NMI: 0 Non-maskable interrupts LOC:2234210 Local timer interrupts SPU: 0 Spurious interrupts PMI: 0 Performance monitoring interrupts PND: 0 Performance pending work THR: 0 Threshold APIC interrupts MCE: 0 Machine check exceptions MCP: 6 Machine check polls ERR: 1 MIS: 0 - Grant
Re: [gentoo-user] Preventing a shared interrupt
> if your mobo does not supporting re-assigning interrupts: you can't. Would that be done in the BIOS? There is a list of interrupts in the BIOS and I can select either "Unassigned" or "Reserved" I think. What affect will reserving an interrupt have? - Grant
Re: [gentoo-user] Preventing a shared interrupt
>> > if your mobo does not supporting re-assigning interrupts: you can't. >> >> Would that be done in the BIOS? There is a list of interrupts in the >> BIOS and I can select either "Unassigned" or "Reserved" I think. What >> affect will reserving an interrupt have? >> >> - Grant > > I really don't know. Ask the support of your vendor ;) Fair enough. Can you tell me why a motherboard would assign 2 devices to the same IRQ when there are obviously many free ones? Is it just bad implementation? - Grant
Re: [gentoo-user] Re: nouveau-drm compile failure
>>>> Google says the error can be due to disabling dri in xorg.conf, but I >>>> don't have anything like that. Should CONFIG_DRM_NOUVEAU=y include >>>> the dri or dri2 module? >>> >>> The dri and dri2 modules should be installed by x11-base/xorg-server. >> >> I've got all of these: >> >> /usr/lib64/opengl/xorg-x11/extensions/libdri.so >> /usr/lib64/opengl/xorg-x11/extensions/libdri2.so >> /usr/lib/opengl/xorg-x11/extensions/libdri2.so >> /usr/lib/opengl/xorg-x11/extensions/libdri.so >> >> but I get the dri and dri2 "module does not exist" errors. Can I >> specify a path in xorg.conf, or does anyone know what the problem >> might be? > > I have this on my x86: > > (II) LoadModule: "dri" > (II) Loading /usr/lib/xorg/modules/extensions/libdri.so > (II) Module dri: vendor="X.Org Foundation" > compiled for 1.7.6, module version = 1.0.0 > ABI class: X.Org Server Extension, version 2.0 > (II) Loading extension XFree86-DRI > (II) LoadModule: "dri2" > (II) Loading /usr/lib/xorg/modules/extensions/libdri2.so > (II) Module dri2: vendor="X.Org Foundation" > compiled for 1.7.6, module version = 1.1.0 > ABI class: X.Org Server Extension, version 2.0 > (II) Loading extension DRI2 > (II) LoadModule: "nouveau" > > If you have the proprietary nvidia drivers installed, that package messes > with > the simlinks to the dri libraries when you use "eselect opengl set foo" so > maybe you have that set for the nvidia opengl version? Check to see what > those > simlinks are actually pointing to. Bingo! Walt, you aced it, thank you. I can now start X via the nouveau driver. The only problem I see in Xorg.0.log is: (EE) AIGLX error: dlopen of /usr/lib64/dri/nouveau_dri.so failed (/usr/lib64/dri/nouveau_dri.so: cannot open shared object file: No such file or directory) (EE) AIGLX: reverting to software rendering I see that xf86-video-nouveau-0.0.15_pre20100213 doesn't install that file. It does install the following file, but they don't look interchangeable to me: /usr/lib64/xorg/modules/drivers/nouveau_drv.so Should /usr/lib64/dri/nouveau_dri.so come from CONFIG_DRM_NOUVEAU=y in the kernel (which I have defined)? Does anyone have this file? - Grant
Re: [gentoo-user] Preventing a shared interrupt
>>>> > if your mobo does not supporting re-assigning interrupts: you can't. >>>> >>>> Would that be done in the BIOS? There is a list of interrupts in the >>>> BIOS and I can select either "Unassigned" or "Reserved" I think. What >>>> affect will reserving an interrupt have? >>>> >>>> - Grant >>> >>> I really don't know. Ask the support of your vendor ;) >> >> Fair enough. Can you tell me why a motherboard would assign 2 devices >> to the same IRQ when there are obviously many free ones? Is it just >> bad implementation? > > Looks like you have MSI support, so probably your hardware isn't > ancient enough for interrupt sharing to be a problem in the first > place. OK. Do higher numbered IRQ's have priority over lower numbered ones? I'm starting a new project where I'll build a super-minimal Gentoo system and I'm trying to learn as much as I can about optimizing for performance. - Grant
Re: [gentoo-user] Re: nouveau-drm compile failure
>> I can now start X via the >> nouveau driver. The only problem I see in Xorg.0.log is: >> >> (EE) AIGLX error: dlopen of /usr/lib64/dri/nouveau_dri.so failed >> (/usr/lib64/dri/nouveau_dri.so: cannot open shared object file: No >> such file or directory) >> (EE) AIGLX: reverting to software rendering > > According to the freedesktop.org documentation, AIGLX has to do with > 3D acceleration and is still under development, not supported for use > by the general public. They recommend for users of 2D acceleration > to put this in xorg.conf to get rid of the error message: > > Section "ServerLayout" > Identifier "X.org Configured" > Screen 0 "Screen0" 0 0 > Option "AIGLX" "false" < > > If you want to try 3D accel at your own risk, they say, you need to run > their latest development libraries from git. I don't do that, so I don't > know if the gentoo packages include that stuff. Maybe someone else knows? Thanks Walt. I don't need 3D so if that's what the error refers to I won't worry about it. The system is running great on Nouveau now. - Grant
Re: [gentoo-user] Preventing a shared interrupt
>> Does anyone know how to prevent my "nouveau" video card and "ath" wifi >> card from sharing interrupt 19? I can't move the video card slot and >> I'd rather not move the wifi card slot if possible. >> >> BTW, does anyone see any other opportunity for interrupt optimization >> here? I'm trying to optimize for video and especially audio, and I'm >> using a USB 1.0 sound card. >> >> # cat /proc/interrupts >> CPU0 >> 0: 473139 IO-APIC-edge timer >> 1: 2 IO-APIC-edge i8042 >> 7: 1 IO-APIC-edge >> 8: 127 IO-APIC-edge rtc0 >> 9: 0 IO-APIC-fasteoi acpi >> 12: 3 IO-APIC-edge i8042 >> 14: 0 IO-APIC-edge ide0 >> 19: 135597 IO-APIC-fasteoi nouveau, ath >> 20: 2 IO-APIC-fasteoi ehci_hcd:usb2 >> 21: 13860 IO-APIC-fasteoi ehci_hcd:usb1 >> 22: 971970 IO-APIC-fasteoi ohci_hcd:usb4 >> 23: 3446 IO-APIC-fasteoi ohci_hcd:usb3 >> 28: 21524 PCI-MSI-edge ahci > > FWIW a grep of the nouveau source doesnt bring up any hits on MSI, so it > looks like the driver doesnt support it yet. Likewise no hits for ath5k, > however ath9k does support it. Nice, thanks Adam. Sounds like ath9k is the way out. - Grant
Re: [gentoo-user] Preventing a shared interrupt
>>> Does anyone know how to prevent my "nouveau" video card and "ath" wifi >>> card from sharing interrupt 19? I can't move the video card slot and >>> I'd rather not move the wifi card slot if possible. >>> >>> BTW, does anyone see any other opportunity for interrupt optimization >>> here? I'm trying to optimize for video and especially audio, and I'm >>> using a USB 1.0 sound card. >>> >>> # cat /proc/interrupts >>> CPU0 >>> 0: 473139 IO-APIC-edge timer >>> 1: 2 IO-APIC-edge i8042 >>> 7: 1 IO-APIC-edge >>> 8: 127 IO-APIC-edge rtc0 >>> 9: 0 IO-APIC-fasteoi acpi >>> 12: 3 IO-APIC-edge i8042 >>> 14: 0 IO-APIC-edge ide0 >>> 19: 135597 IO-APIC-fasteoi nouveau, ath >>> 20: 2 IO-APIC-fasteoi ehci_hcd:usb2 >>> 21: 13860 IO-APIC-fasteoi ehci_hcd:usb1 >>> 22: 971970 IO-APIC-fasteoi ohci_hcd:usb4 >>> 23: 3446 IO-APIC-fasteoi ohci_hcd:usb3 >>> 28: 21524 PCI-MSI-edge ahci >> >> FWIW a grep of the nouveau source doesnt bring up any hits on MSI, so it >> looks like the driver doesnt support it yet. Likewise no hits for ath5k, >> however ath9k does support it. > > Nice, thanks Adam. Sounds like ath9k is the way out. > > - Grant Ah nevermind, I guess ath9k is 802.11n only. Maybe they'll add MSI support to ath5k. - Grant
[gentoo-user] {OT} Basic device for a Gentoo router/firewall?
Does anyone know of a basic device that would function well as a Gentoo router/firewall? Using typical hardware seems like overkill. I should be able to offload package compilation duties to another local machine on the network. It would also be nice if it were small, cheap, and power-efficient. - Grant
[gentoo-user] {OT} "dumb" ethernet mover
My cable internet outlet is across the room from my TV, and my Gentoo desktop attaches to my TV. I'm using a small wireless router to send the signal from the cable modem to my Gentoo system across the room. I don't like using a non-Gentoo decision-making device in my network, but I also don't want to build and maintain another Gentoo system for only firewall/router duties. Am I overlooking another option? I want a "dumb" device to move the ethernet connection from one side of the room to the other. There may not be anything like that. I just thought I'd ask. - Grant
Re: [gentoo-user] {OT} "dumb" ethernet mover
>> My cable internet outlet is across the room from my TV, and my Gentoo >> desktop attaches to my TV. I'm using a small wireless router to send >> the signal from the cable modem to my Gentoo system across the room. >> I don't like using a non-Gentoo decision-making device in my network, >> but I also don't want to build and maintain another Gentoo system for >> only firewall/router duties. Am I overlooking another option? I want >> a "dumb" device to move the ethernet connection from one side of the >> room to the other. > > You could use a pair of powerline adaptors to route the ethernet across > the mains wiring. I use them for my MythTV front end, because SWMBO would > go apeshit if I tried running ethernet cables through the living room. I thought about that, and it would be perfect except that I'm really into audio and I don't want to increase the interference in my apartment's electricity. - Grant
Re: [gentoo-user] Re: {OT} "dumb" ethernet mover
>> My cable internet outlet is across the room from my TV, and my Gentoo >> desktop attaches to my TV. I'm using a small wireless router to send >> the signal from the cable modem to my Gentoo system across the room. >> I don't like using a non-Gentoo decision-making device in my network, >> but I also don't want to build and maintain another Gentoo system for >> only firewall/router duties. Am I overlooking another option? I want >> a "dumb" device to move the ethernet connection from one side of the >> room to the other. > > Usually that's called a "cable". ;) > > Many wireless bridges have a "virtual cable" mode point-to-point > bridging mode that will let you pair them together so that they won't > talk to anything else and are just transparent layer 2 bridges. I've > got some DLink bridges that have a mode like that. You just set them > up next to each other and powered them both up while holding a button > down, and they'd find each other and pair-up. I hadn't heard of a wireless bridge before. That sounds about right. DD-WRT running as a wireless bridge wouldn't be so bad right? - Grant
Re: [gentoo-user] Re: {OT} "dumb" ethernet mover
>>> My cable internet outlet is across the room from my TV, and my Gentoo >>> desktop attaches to my TV. I'm using a small wireless router to send >>> the signal from the cable modem to my Gentoo system across the room. >>> I don't like using a non-Gentoo decision-making device in my network, > [...] > >> You could use a pair of powerline adaptors to route the ethernet across >> the mains wiring. I use them for my MythTV front end, because SWMBO would >> go apeshit if I tried running ethernet cables through the living room. > > Do those not count as "non-Gentoo decision-making devices"? When I said I didn't like to use non-Gentoo decision-making devices, I was just trying to let everyone in on my mentality. It would be ideal, but of course it means building and maintaining another Gentoo system and I always seem to be short on time as it is. - Grant
Re: [gentoo-user] Re: {OT} "dumb" ethernet mover
>>>> My cable internet outlet is across the room from my TV, and my Gentoo >>>> desktop attaches to my TV. ?I'm using a small wireless router to send >>>> the signal from the cable modem to my Gentoo system across the room. >>>> I don't like using a non-Gentoo decision-making device in my network, >>>> but I also don't want to build and maintain another Gentoo system for >>>> only firewall/router duties. ?Am I overlooking another option? ?I want >>>> a "dumb" device to move the ethernet connection from one side of the >>>> room to the other. >>> >>> Usually that's called a "cable". ?;) >>> >>> Many wireless bridges have a "virtual cable" mode point-to-point >>> bridging mode that will let you pair them together so that they won't >>> talk to anything else and are just transparent layer 2 bridges. ?I've >>> got some DLink bridges that have a mode like that. ?You just set them >>> up next to each other and powered them both up while holding a button >>> down, and they'd find each other and pair-up. >> >> I hadn't heard of a wireless bridge before. That sounds about right. >> DD-WRT running as a wireless bridge wouldn't be so bad right? > > I don't have any experience with DD-WRT. I use OpenWrt, but have > only used it in "normal" bridges and WAPs. Sorry to be getting even more OT, but why did you choose OpenWRT over DD-WRT? I didn't know there was more than one choice available. - Grant > By a "normal" bridge, I mean one that's running in infrastructure mode > talking to a WAP using the normal WEP and WAP authentication (the WAP > also happened to be running OpenWrt). > > For a simple point-to-point link you may want to try ad-hoc mode > instead of infrastrucure mode.
[gentoo-user] change memory speed?
I have my DDR800 RAM specifically set to DDR800 in the BIOS, but I get: # lshw -short -C memory H/W path Device Class Description = /0/0 memory 64KiB BIOS /0/3/5 memory 256KiB L1 cache /0/3/6 memory 1MiB L2 cache /0/24 memory 4GiB System Memory /0/24/0memory 2GiB DIMM DDR2 Synchronous 400 MHz (2.5 ns) /0/24/1memory 2GiB DIMM DDR2 Synchronous 400 MHz (2.5 ns) /0/24/2memory DIMM [empty] /0/24/3memory DIMM [empty] /0/4 memory RAM memory /0/1.2 memory RAM memory /0/1.4 memory RAM memory Does anyone know how to set the memory speed to 800 Mhz? - Grant
Re: [gentoo-user] Re: change memory speed?
>> I have my DDR800 RAM specifically set to DDR800 in the BIOS, but I get: >> >> # lshw -short -C memory >> H/W path Device Class Description >> = >> /0/0 memory 64KiB BIOS >> /0/3/5 memory 256KiB L1 cache >> /0/3/6 memory 1MiB L2 cache >> /0/24 memory 4GiB System Memory >> /0/24/0 memory 2GiB DIMM DDR2 Synchronous 400 MHz (2.5 >> ns) >> /0/24/1 memory 2GiB DIMM DDR2 Synchronous 400 MHz (2.5 >> ns) >> /0/24/2 memory DIMM [empty] >> /0/24/3 memory DIMM [empty] >> /0/4 memory RAM memory >> /0/1.2 memory RAM memory >> /0/1.4 memory RAM memory >> >> Does anyone know how to set the memory speed to 800 Mhz? > > DDR800 runs at 400Mhz, so everything is fine. > > For more details: > > http://en.wikipedia.org/wiki/DDR2_SDRAM#Chips_and_modules Thank you Nikos. - Grant
[gentoo-user] Patch via perl script in an ebuild?
I need to install the google-api-adwords-perl library, and it requires that SOAP-WSDL is patched with the soap_wsdl_patches.pl perl script. Can I have SOAP-WSDL patched via the perl script in an ebuild? - Grant
[gentoo-user] Re: Patch via perl script in an ebuild?
> I need to install the google-api-adwords-perl library, and it requires > that SOAP-WSDL is patched with the soap_wsdl_patches.pl perl script. > Can I have SOAP-WSDL patched via the perl script in an ebuild? > > - Grant Here is the perl script: http://pastebin.com/YM3G5sKn Can anyone with perl and ebuild knowledge determine if the script could be incorporated into an ebuild? David Abbott and I are working on getting google-api-adwords-perl into portage and this is a dependency. - Grant
Re: [gentoo-user] Re: Flash stopped working after Firefox 3.6.3->3.6.4 update.
>> Firefox got upgraded from 3.6.3 to 3.6.4 yesterday, and now the flash >> plugin won't work. I tried both "x86" and "~x86" versions, and both >> crash 100% of the time. > > The problem was caused by an old 9.x copy of the plugin that was left > in ~/.mozialla/plugins. After deleting the flashplayer plugin from > that directory, FF then found the "officially" installed 10.x version > and is now happy. I'm having the exact same problem, except I don't have a ~/.mozilla/plugins directory at all. Any ideas? - Grant (a different Grant)
Re: [gentoo-user] Re: Patch via perl script in an ebuild?
>>> I need to install the google-api-adwords-perl library, and it requires >>> that SOAP-WSDL is patched with the soap_wsdl_patches.pl perl script. >>> Can I have SOAP-WSDL patched via the perl script in an ebuild? >>> >>> - Grant >>> >> Here is the perl script: >> >> http://pastebin.com/YM3G5sKn >> >> Can anyone with perl and ebuild knowledge determine if the script >> could be incorporated into an ebuild? David Abbott and I are working >> on getting google-api-adwords-perl into portage and this is a >> dependency. > > (Some preliminary setup work, like cpan -i Text::Patch SOAP::WSDL.) > > Running the patch after fixing its paths ... will result invariably in: > Trying to patch > /usr/lib64/perl5/site_perl/5.10.1/SOAP/WSDL/XSD/Typelib/Builtin/anySimpleType.pm...Hunk > #1 failed at line 13. > > I'm no authority on perl, but AFAICT that patch does not match against > either SOAP::WSDL latest stable sources nor latest SOAP::WSDL dev release > code from CPAN. Actually, I couldn't find tarballs from CPAN which would > match with the failing lines of that patch. There might've been some earlier > 2.00-series releases, but they're ... long gone? > > Any idea of the version of SOAP::WSDL they are using for this at Google? Thank you Arttu. Here is the link to the SOAP::WSDL: http://soap-wsdl.svn.sourceforge.net/viewvc/soap-wsdl/SOAP-WSDL/branches/Typemap.tar.gz?view=tar&pathrev=846 from the README: http://code.google.com/p/google-api-adwords-perl/source/browse/trunk/README - Grant
Re: [gentoo-user] Re: Patch via perl script in an ebuild?
[snip] > Am I making any sense? I think all of that is right on. I need to find out why the patch isn't working though. > Theoretically (if you insist), you could still use the perl's > Text::Patch route as well, but (if I'm not entirely wrong, see the > excerpted attempted patch run above) the patch would still need to be > touched up to match properly with the _3 dev release code. And it > would add a dependency to Text::Patch, and make an odd call to perl in > the middle of the ebuild. (I assume it must be made explicitly as I > don't know if perl-module.eclass has any automation for this. Probably > not since AFAICT Text::Patch isn't even installed by default). Do you think it would be better to create a real patch than to use the perl patch (after we figure out why it isn't working)? I would think it would be easier to use the perl patch in case a different version is released so we don't have to re-create the patch each time. A Text::Patch dep wouldn't be so bad. What do you think? - Grant
Re: [gentoo-user] Re: Patch via perl script in an ebuild?
[snip]
> # Copyright 1999-2010 Gentoo Foundation
> # Distributed under the terms of the GNU General Public License v2
> # $Header: $
>
> EAPI="3"
>
> inherit perl-module
>
> DESCRIPTION="SOAP-WSDL provides a SOAP client with WSDL support."
> HOMEPAGE="http://soap-wsdl.svn.sourceforge.net";
> SRC_URI="http://soap-wsdl.svn.sourceforge.net/viewvc/soap-wsdl/SOAP-WSDL/branches/Typemap.tar.gz
> -> ${P}.tar.gz"
> LICENSE="Apache-2.0"
> SLOT="0"
> KEYWORDS="~amd64 ~x86"
> IUSE="google"
>
> DEPEND="${RDEPEND}
> virtual/perl-Module-Build"
> RDEPEND="dev-perl/SOAP-Lite
> dev-perl/Class-Std-Fast"
>
> src_prepare() {
> if use google ; then
> epatch "${FILESDIR}/${PN}.patch" | die "google patch failed"
> fi
> }
David, you mentioned in the bug that you couldn't get the patch to
work clean. I thought you got it working before when you posted the
patched Typemap for me to download. Did it work then but not now?
http://bugs.gentoo.org/show_bug.cgi?id=305621
- Grant
Re: [gentoo-user] Re: Flash stopped working after Firefox 3.6.3->3.6.4 update.
>>>> Firefox got upgraded from 3.6.3 to 3.6.4 yesterday, and now the flash >>>> plugin won't work. ?I tried both "x86" and "~x86" versions, and both >>>> crash 100% of the time. >>> >>> The problem was caused by an old 9.x copy of the plugin that was left >>> in ~/.mozialla/plugins. ?After deleting the flashplayer plugin from >>> that directory, FF then found the "officially" installed 10.x version >>> and is now happy. >> >> I'm having the exact same problem, except I don't have a >> ~/.mozilla/plugins directory at all. Any ideas? > > Search for libflashplayer.so files. > > Another of my machines had an old one in /usr/lib/nsbrowser/plugins. > That should have been a symlink to /opt/netscape/plugins/libflashplayer.so > which is where the ebuild puts the file. I did have an extra libflashplayer.so, I deleted it, I deleted pluginreg.dat, I re-emerged adobe-flash, and I restarted firefox, but flash still isn't working. It works in opera. Any ideas? - Grant
Re: [gentoo-user] Re: Patch via perl script in an ebuild?
>> David, you mentioned in the bug that you couldn't get the patch to >> work clean. I thought you got it working before when you posted the >> patched Typemap for me to download. Did it work then but not now? >> >> http://bugs.gentoo.org/show_bug.cgi?id=305621 >> >> - GrantSOAP-WSDL >> >> > The patched Typemap created by hand works fine, but when you attempt > to use a downloaded SOAP-WSDL and patch it failed for me, may need to > manually create a patch for SOAP-WSDL. > > -- > David Abbott (dabbott) I'm sure I'm confused because of my weak understanding of this. I thought Typemap was a component of the downloaded SOAP-WSDL. If not, what is Typemap and how did you patch it by hand? - Grant
Re: [gentoo-user] Re: Patch via perl script in an ebuild?
>> David, you mentioned in the bug that you couldn't get the patch to >> work clean. I thought you got it working before when you posted the >> patched Typemap for me to download. Did it work then but not now? >> >> http://bugs.gentoo.org/show_bug.cgi?id=305621 > > Ah, I didn't know you had this already mostly figured out in a bug. > Full disclosure requested -- from the very beginning! :D My mistake, I apologize. > After taking a quick peek: do those ebuilds succeed in installing > something? Since they don't follow CPAN conventions (esp. wrt naming) > I'm not sure what perl-module.eclass actually does for their > src_compile, src_install and other steps. Also, the "->" used in SRC > might have interesting side-effects to paths, as the tarball name no > longer matches its content directory's name. > > -- > Arttu V. -- Running Gentoo is like running with scissors The google-api-adwords-perl ebuild in the bug works great for me so far. I've only done basic testing with it, but it appears to be working. I will be able to do more extensive testing hopefully over the next few days. - Grant
Re: [gentoo-user] Re: Patch via perl script in an ebuild?
>>> David, you mentioned in the bug that you couldn't get the patch to >>> work clean. I thought you got it working before when you posted the >>> patched Typemap for me to download. Did it work then but not now? >>> >>> http://bugs.gentoo.org/show_bug.cgi?id=305621 >> >> Ah, I didn't know you had this already mostly figured out in a bug. >> Full disclosure requested -- from the very beginning! :D >> >> After taking a quick peek: do those ebuilds succeed in installing >> something? Since they don't follow CPAN conventions (esp. wrt naming) >> I'm not sure what perl-module.eclass actually does for their >> src_compile, src_install and other steps. Also, the "->" used in SRC >> might have interesting side-effects to paths, as the tarball name no >> longer matches its content directory's name. >> >> -- >> Arttu V. -- Running Gentoo is like running with scissors >> >> > I don't think I will use the software, I was just helping to put the > pieces together, Grant has been testing it as we move forward. Once we > get the pieces together and if it will work as intended I think we > could contact upstream to help with a more sensible approach to > packaging the patch for the stock SOAP-WSDL, so updating will be > easier to maintain. > Just my 2c :) > > > -- > David Abbott (dabbott) You think Google will change the way it packages the SOAP-WSDL patch if we request it? - Grant
Re: [gentoo-user] Re: Flash stopped working after Firefox 3.6.3->3.6.4 update.
>> > > I did have an extra libflashplayer.so, I deleted it, I deleted >> > > pluginreg.dat, I re-emerged adobe-flash, and I restarted firefox, but >> > > flash still isn't working. It works in opera. Any ideas? >> > >> > Try with another new profile : >> > $ firefox -ProfileManager >> >> Here it stopped working in the new Opera-10.6 and Konqueror, but still >> works in FF-3.6.4. o_O >> >> I'm starting a new thread for the Opera problem ... > > Latest flash does not have a 64 bit version. On amd64 you MUST use > nspluginwrapper. > > Safest is to just remerge the thing and let it convert all your plugins - the > command to do that is somewhat arcane and tedious but the ebuild does it all > automatically. Fixed by emerging nspluginwrapper. Thanks everyone! - Grant
[gentoo-user] 2.6.34 kernel compile error: DRM problem?
I tried to update my laptop and desktop to 2.6.34 but compilation fails with this error: drivers/gpu/drm/i915/i915_dma.c: In function 'i915_switcheroo_can_switch': drivers/gpu/drm/i915/i915_dma.c:1419: error: invalid operands to binary == (have 'atomic_t' and 'int') or: drivers/gpu/drm/nouveau/nouveau_state.c: In function 'nouveau_switcheroo_can_switch': drivers/gpu/drm/nouveau/nouveau_state.c:394: error: invalid operands to binary == (have 'atomic_t' and 'int') They compile if I disable DRM. I've tried with and without CONFIG_VGA_SWITCHEROO with the same result. Does anyone know how to fix this? I'm trying to use hardened-sources-2.6.34 from the anarchy overlay on both systems. - Grant
[gentoo-user] Why does high-res video drop frames at 60% CPU?
I've been using VDPAU acceleration to play back Blu-Ray rips for a while, but the extra layer is getting to be quite a hassle so I'm trying to get decent performance via software decoding. It has actually come a long way since the last time I tried and playing Blu-Ray rips via mplayer is nearly watchable. I'm using a dual-core 3.1Ghz CPU and one of the cores is only taxed up to 60% during playback, but frames are still being dropped constantly. Does anyone know where the bottleneck might be? - Grant
Re: [gentoo-user] Re: Why does high-res video drop frames at 60% CPU?
>> I've been using VDPAU acceleration to play back Blu-Ray rips for a >> while, but the extra layer is getting to be quite a hassle so I'm >> trying to get decent performance via software decoding. It has >> actually come a long way since the last time I tried and playing >> Blu-Ray rips via mplayer is nearly watchable. I'm using a dual-core >> 3.1Ghz CPU and one of the cores is only taxed up to 60% during >> playback, but frames are still being dropped constantly. Does anyone >> know where the bottleneck might be? > > Not sure. Could be wrong CPU load display; which tool do you use to get the > CPU load? I use top. On the mplayer list, people were saying they too get 60% CPU load but no playback problems. > Anyway, if you're not already doing so, you might want to try the > multithreaded version of mplayer so both CPU cores can do decoding. It's in > the "multimedia" overlay. More details here: > > http://forums.gentoo.org/viewtopic-t-789673.html I really don't think it's a CPU issue. What other factors could be at play? Could it be my nouveau video drivers? - Grant
Re: [gentoo-user] Re: Why does high-res video drop frames at 60% CPU?
> Regarding mplayer: There is another ebuild in the multimedia-overlay > that I prefer now to mplayer(-mt) and that also uses ffmpeg-mt : > media-video/mplayer-uau. > > It fetches the mplayer-version from one of the mplayer-devs and creates > a binary called "mplayer-uau" which can be installed at the same time as > the official mplayer package. I gave that a try and it smoothed everything right out, so I guess it was a CPU issue after all. Thank you for your help. How is mplayer-uau different from mplayer-mt? Maybe mplayer-mt can't be installed alongside mplayer? > Regarding your frame drops: it is highly likely that sound is the > problem. Please try playing the video with "-ao null" to see if that's > the case. I assume you use pulseaudio? Check if it has real time > capabilities (kill it, start it with verbose/debug in foreground, read > log). Also try "-ao alsa" and "-ao oss". No change with "-ao null". I actually don't use pulseaudio. > Your data-source (gard disk, network?) is fast enough? Copy 1GB into RAM > to be sure by ether using RAM-disk or cache-settings. It's running from an internal HD. > Try nvidia-binary. You'll get VDPAU in that case, which will result in > about 5% CPU usage when decoding h264! I'm trying to move away from VDPAU. It fixes the performance issue but the extra layer creates a new set of problems. - Grant
Re: [gentoo-user] Why does high-res video drop frames at 60% CPU?
> You have dual core so 60% means: > 50% (full one core) is for decoding, > and the rest 10% is for audio, resizing etc. > You can't play the video correctly because your "decoder" is not > multithreaded and uses just the one CPU at its fullest. > Try using multithreaded version of mplayer "mplayer-mt" (in some overlay > probably) with "lavdopts=threads=2" in mplayer config. Does anyone know if mplayer-uau uses "-lavdopts threads=2" by default? I tried with and without and there might have been a performance increase with, but I'm not sure. - Grant >> I've been using VDPAU acceleration to play back Blu-Ray rips for a >> while, but the extra layer is getting to be quite a hassle so I'm >> trying to get decent performance via software decoding. It has >> actually come a long way since the last time I tried and playing >> Blu-Ray rips via mplayer is nearly watchable. I'm using a dual-core >> 3.1Ghz CPU and one of the cores is only taxed up to 60% during >> playback, but frames are still being dropped constantly. Does anyone >> know where the bottleneck might be? >> >> - Grant
Re: [gentoo-user] Re: Why does high-res video drop frames at 60% CPU?
>>> Regarding mplayer: There is another ebuild in the multimedia-overlay >>> that I prefer now to mplayer(-mt) and that also uses ffmpeg-mt : >>> media-video/mplayer-uau. >>> >>> It fetches the mplayer-version from one of the mplayer-devs and creates >>> a binary called "mplayer-uau" which can be installed at the same time as >>> the official mplayer package. >> >> I gave that a try and it smoothed everything right out, so I guess it >> was a CPU issue after all. Thank you for your help. How is >> mplayer-uau different from mplayer-mt? Maybe mplayer-mt can't be > > mplayer from multimedia overlay is a replacement for the regular, official > mplayer with the only change being ffmpeg-mt. It can't be installed > alongside regular mplayer because it's a replacement and multithreading can > be enabled/disabled at runtime. > > mplayer-uau is a fork of MPlayer, previously known as "mplayer-git", before > the developer was kicked from the mplayer project for unknown reasons. > While mplayer from multimedia only adds the ffmpeg-mt patch, mplayer-uau > adds more patches. Does mplayer from multimedia overlay include mplayer-mt and ffmpeg-mt patches, or just ffmpeg-mt? Do you know what other significant patches mplayer-uau adds? - Grant
Re: [gentoo-user] Re: Why does high-res video drop frames at 60% CPU?
>> Does mplayer from multimedia overlay include mplayer-mt and ffmpeg-mt >> patches, or just ffmpeg-mt? > > "mplayer-mt" is mplayer with ffmpeg-mt. Sorry for my misunderstanding. I'd like to follow the latest mplayer with ffmpeg-mt support. Is there a way to either do that specifically, or to follow the latest mplayer version in the multimedia overlay? The situation is a little tricky since there is a version in portage that is newer than the newest version in the multimedia overlay. Also thanks to Stroller for clearing up the mplayer/mplayer-uau situation. - Grant
Re: [gentoo-user] Re: Why does high-res video drop frames at 60% CPU?
>>>> Does mplayer from multimedia overlay include mplayer-mt and ffmpeg-mt >>>> patches, or just ffmpeg-mt? >>> >>> "mplayer-mt" is mplayer with ffmpeg-mt. >> >> Sorry for my misunderstanding. I'd like to follow the latest mplayer >> with ffmpeg-mt support. > > There is no latest "mplayer with ffmpeg-mt" support. There is mplayer, > ffmpeg and ffmpeg-mt. To get what most people refer to as "mplayer-mt", you > have to replace the bundled ffmeg that is included in mplayer with > ffmpeg-mt. > > >> Is there a way to either do that >> specifically, or to follow the latest mplayer version in the >> multimedia overlay? The situation is a little tricky since there is a >> version in portage that is newer than the newest version in the >> multimedia overlay. > > That's because the developer who updates the multimedia overlay didn't yet > put the ebuild there. The latest version I released (which is based on the > latest available ffmpeg-mt sources) is "1.0_rc4_p20100626": > > http://forums.gentoo.org/viewtopic-t-789673.html Got it, thanks. Hopefully that shows up in the multimedia overlay soon. It's interesting that the mplayer-uau ebuild includes a threads USE flag in addition to the ffmpeg-mt one. - Grant
[gentoo-user] Re: 2.6.34 kernel compile error: DRM problem?
> I tried to update my laptop and desktop to 2.6.34 but compilation > fails with this error: > > drivers/gpu/drm/i915/i915_dma.c: In function 'i915_switcheroo_can_switch': > drivers/gpu/drm/i915/i915_dma.c:1419: error: invalid operands to > binary == (have 'atomic_t' and 'int') > > or: > > drivers/gpu/drm/nouveau/nouveau_state.c: In function > 'nouveau_switcheroo_can_switch': > drivers/gpu/drm/nouveau/nouveau_state.c:394: error: invalid operands > to binary == (have 'atomic_t' and 'int') > > They compile if I disable DRM. I've tried with and without > CONFIG_VGA_SWITCHEROO with the same result. Does anyone know how to > fix this? I'm trying to use hardened-sources-2.6.34 from the anarchy > overlay on both systems. > > - Grant Any idea on this guys? - Grant
[gentoo-user] Re: 2.6.34 kernel compile error: DRM problem?
> I tried to update my laptop and desktop to 2.6.34 but compilation > fails with this error: > > drivers/gpu/drm/i915/i915_dma.c: In function 'i915_switcheroo_can_switch': > drivers/gpu/drm/i915/i915_dma.c:1419: error: invalid operands to > binary == (have 'atomic_t' and 'int') > > or: > > drivers/gpu/drm/nouveau/nouveau_state.c: In function > 'nouveau_switcheroo_can_switch': > drivers/gpu/drm/nouveau/nouveau_state.c:394: error: invalid operands > to binary == (have 'atomic_t' and 'int') > > They compile if I disable DRM. I've tried with and without > CONFIG_VGA_SWITCHEROO with the same result. Does anyone know how to > fix this? I'm trying to use hardened-sources-2.6.34 from the anarchy > overlay on both systems. > > - Grant This is fixed with hardened-2.6.34-r2 in the anarchy overlay. - Grant
Re: [gentoo-user] Re: Patch via perl script in an ebuild?
>>>> David, you mentioned in the bug that you couldn't get the patch to >>>> work clean. I thought you got it working before when you posted the >>>> patched Typemap for me to download. Did it work then but not now? >>>> >>>> http://bugs.gentoo.org/show_bug.cgi?id=305621 >>>> >>>> - GrantSOAP-WSDL >>>> >>>> >>> The patched Typemap created by hand works fine, but when you attempt >>> to use a downloaded SOAP-WSDL and patch it failed for me, may need to >>> manually create a patch for SOAP-WSDL. >>> >>> -- >>> David Abbott (dabbott) >> >> I'm sure I'm confused because of my weak understanding of this. I >> thought Typemap was a component of the downloaded SOAP-WSDL. If not, >> what is Typemap and how did you patch it by hand? >> >> - Grant >> >> > Typemap is a branch of soap-wsdl; > http://soap-wsdl.svn.sourceforge.net/viewvc/soap-wsdl/SOAP-WSDL/branches/ > > SOAP::WSDL I don't think can even be used for this; > http://search.cpan.org/~mkutter/SOAP-WSDL-2.00.10/lib/SOAP/WSDL.pm > > I am with you Grant, it is confusing, that is why I said google needs > to make a patched version of Typemap to use with this and call it > something line google-typemap-0.1 and keep it updated to use with > their google-api-adwords-perl. This may happen as it appears the > project is very active. > > http://code.google.com/p/google-api-adwords-perl/source/browse/trunk/README I've updated the bug with the following good news: "The ebuild works perfectly. I've done extensive testing with the library now and all is well." - Grant
[gentoo-user] Can I do this remotely?
I recently had the problem described here: http://bbs.archlinux.org/viewtopic.php?id=70884 The solution worked. Now I'm being told the same system is having the same problem, but I'm away from there. Is there a way to fix it remotely? - Grant
Re: [gentoo-user] Can I do this remotely?
>> I recently had the problem described here: >> >> http://bbs.archlinux.org/viewtopic.php?id=70884 >> >> The solution worked. Now I'm being told the same system is having the >> same problem, but I'm away from there. Is there a way to fix it >> remotely? > > ssh in remotely and have a user log in to gdm locally. > > ps will show if xfce4-panel is running twice, so have the local user log out, > you edit the session startup files and have them try again. Do you know which file that is? I have: # cat /home/grant/.config/xfce4-session/xfce4-session.rc [General] SessionName=Default SaveOnExit=true [Splash Screen] Engine= but that doesn't seem to be it. - Grant
[gentoo-user] Does Firefox call Google?
I have a remote system on which shorewall blocks all outgoing 80/443 traffic except for 1 destination IP. I noticed that whenever someone logs in to an xfce4 session on that system, I see a bunch of rejected 80/443 requests from that system to various Google IPs from throughout their session. Does Firefox periodically make Google requests for some reason? The person logging in says they aren't attempting to access Google, and the home page is not set to go there. Does anyone know why this might be happening? - Grant
Re: [gentoo-user] Does Firefox call Google?
>> Does Firefox periodically make Google requests for >> some reason? The person logging in says they aren't attempting to >> access Google, and the home page is not set to go there. Does anyone >> know why this might be happening? > > If you haven’t disabled it, Firefox periodically downloads a current list of > malware addresses for its phising filter. Thank you. Is this the checkbox: "Tell me if the site I'm visiting is a suspected attack site." - Grant
Re: [gentoo-user] Re: Does Firefox call Google?
>>>> Does Firefox periodically make Google requests for >>>> some reason? The person logging in says they aren't attempting to >>>> access Google, and the home page is not set to go there. Does anyone >>>> know why this might be happening? >>> >>> If you haven’t disabled it, Firefox periodically downloads a current list >>> of >>> malware addresses for its phising filter. >> >> Thank you. Is this the checkbox: "Tell me if the site I'm visiting is >> a suspected attack site." > > Two options, in the "Security" tab: > > "Block reported attack sites" and "Block reported web forgeries." Thanks, you must be on 3.5. - Grant
Re: [gentoo-user] Re: Does Firefox call Google?
>>>> Does Firefox periodically make Google requests for >>>> some reason? The person logging in says they aren't attempting to >>>> access Google, and the home page is not set to go there. Does anyone >>>> know why this might be happening? >>> >>> If you haven’t disabled it, Firefox periodically downloads a current list >>> of >>> malware addresses for its phising filter. >> >> Thank you. Is this the checkbox: "Tell me if the site I'm visiting is >> a suspected attack site." > > Two options, in the "Security" tab: > > "Block reported attack sites" and "Block reported web forgeries." BTW, Firefox asks Google for the lists right? - Grant
Re: [gentoo-user] Does Firefox call Google?
>> [snip] and the home page is not set to go there.[snip] > > Hi, > > does this also include stuff like google analytics? Like are there adds on > the homepage? Good question, I will test for that. Does anyone know if Firefox downloads its website lists from Google? - Grant
[gentoo-user] HDMI-out working?
Does HDMI-out on laptops work on Gentoo? - Grant
Re: [gentoo-user] HDMI-out working?
>> Does HDMI-out on laptops work on Gentoo? > > Here with a Dell XPS Laptop M1330 and a Nvidia GeForce 9400M GS, it works fine > with nvidia-drivers just plug it, and all its ok :) Thank you. I'm planning on getting a laptop with the Intel GMA 4500MHD. I thought HDMI had some type of "copy protection" built in in some way, so I was thinking it might not work with Linux. - Grant
Re: [gentoo-user] HDMI-out working?
>>>> Does HDMI-out on laptops work on Gentoo? >>> >>> Here with a Dell XPS Laptop M1330 and a Nvidia GeForce 9400M GS, it works >>> fine >>> with nvidia-drivers just plug it, and all its ok :) >> >> Thank you. I'm planning on getting a laptop with the Intel GMA >> 4500MHD. I thought HDMI had some type of "copy protection" built in >> in some way, so I was thinking it might not work with Linux. > > You're probably thinking of HDCP which is copy-protection that > requires both ends (video card and monitor/TV) support HDCP (=paid > Intel for a license). HDMI is one of several connection formats that > supports HDCP. Exactly, that was it. - Grant
[gentoo-user] {OT} External video card with RCA output?
Does anyone know of a self-powered external video card (USB, Firewire, ExpressCard) that has composite/RCA output and works in Gentoo? I'd like to use it to connect my laptop (VGA output only) to TVs while I'm travelling. - Grant
Re: [gentoo-user] Re: {OT} External video card with RCA output?
>>> Does anyone know of a self-powered external video card (USB, Firewire, >>> ExpressCard) that has composite/RCA output and works in Gentoo? I'd >>> like to use it to connect my laptop (VGA output only) to TVs while I'm >>> travelling. >> >> Something like this? >> >> http://www.maplin.co.uk/Module.aspx?ModuleNo=223833 >> >> OS compatibility isn't an issue because it only deals with >> video signals. There is a USB connection to the computer, but >> only to supply power. > > I've used various widgets similar to that, and they work fine. > but remember than you're not going to get much more than > 640x480 resolution on a TV via composite video. Setting your > desktop to 800x600 usually works OK, but things are going to be > a bit on the fuzzy side. > > -- > Grant That looks pretty slick and it has good reviews: http://www.newegg.com/Product/Product.aspx?Item=N82E16815260011 Ideally, I'd like to get something with the option to use DVI or HDMI as well as composite. Then it would need to be a sound card too unless I was using separate audio cables. Does anyone know of anything? It might be even better to have a real external video and sound card connected via a real USB interface. That way, you could use only an HDMI cable, or you could run audio and video cables from the device to the TV, instead of audio cables going from the laptop to the TV. However, I'm not sure something like that exists. Here is something that looks close: http://www.amazon.com/gp/product/B001PSVBFY - Grant
[gentoo-user] New laptop is slow.
I just finished installing Gentoo on a Dell Vostro 1320 laptop. It has a 2.2Ghz Core Duo CPU, 3GB RAM, and a 7200RPM hard drive. Navigating within firefox is pretty slow. It's the response time of the application, not the network. It's much slower than my previous laptop which has much weaker specs. I noticed the HD light comes on when the system is pausing in firefox sometimes. I don't have any swap at all, I'm using the CFQ, and /tmp is mounted on tmpfs. Can anyone suggest where to look? - Grant
Re: [gentoo-user] New laptop is slow.
>> I just finished installing Gentoo on a Dell Vostro 1320 laptop. It >> has a 2.2Ghz Core Duo CPU, 3GB RAM, and a 7200RPM hard drive. >> Navigating within firefox is pretty slow. It's the response time of >> the application, not the network. It's much slower than my previous >> laptop which has much weaker specs. >> >> I noticed the HD light comes on when the system is pausing in firefox >> sometimes. I don't have any swap at all, I'm using the CFQ, and /tmp >> is mounted on tmpfs. Can anyone suggest where to look? > > When you're not using X does it still seem slow? You don't mention > what kind of video or arch vs ~arch, but I would try different > AccelMethod options in your xorg.conf. On my laptop everything in X > was very slow until I set > Option "AccelMethod" "EXA" Thank you James and Paul. I read this: http://www.gossamer-threads.com/lists/gentoo/user/195836 and downgraded to xf86-video-intel-2.7.1 and all is well. - Grant
