Re: cpufreqd as standard install?
On Thu, Mar 08, 2012 at 05:29:47PM -0500, Phillip Susi wrote: > You also need to remember that 3.4x the clock speed does not mean you > actually end up finishing your work 3.4x faster. Intel recommends > using the ondemand governor, so if you are claiming they are wrong, > and you save more power using performance, that's going to take a > little convincing. I checked on my desktop sandybridge core i5 > system, and I found that running stress -c 4 with powersave draws 150 > watts, and with ondemand, is nearly 200 watts. Idle, the system draws > 125 watts. Factoring out that baseline gives a cpu load of 25 vs 75 > watts depending on whether you run at 1600 MHz vs 3300 MHz, or 3x the > power for twice the speed. Do a SpecPOWER run with performance and with ondemand and you'll see that there's very little difference on modern hardware. > > But this is a very strange workload to be optimising for. First, it's > > entirely CPU-bound. If it involves IO then you're going to be keeping IO > > devices in a higher power state for longer, which wipes out the > > advantage. Second, it makes the assumption that the user doesn't care > > how much time it takes. That's basically never true. > > We're not talking about a 100% cpu bound workload of course; we're > talking about typical loads where the cpu is mostly idle, and the > question is whether it is better to spend a bit more time idle, and be > less efficient when actually executing instructions, or be more > efficient at the cost of a little less idle time. If you ignore the > time it takes to enter/exit the deep C states, you can model the > different execution speeds and how much energy they consume as a > simple 100% busy period followed by a fully idle period, with a total > duration being the same in both tests. You're ignoring far too many factors for this to be terribly relevant. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: cpufreqd as standard install?
On Thu, Mar 08, 2012 at 11:22:04AM -0500, Phillip Susi wrote: > On 3/8/2012 11:10 AM, Matthew Garrett wrote: > >Yes, if those are the actual power figures. But they're typically not > >going to be. > > Can you be a little less vague and hand wavy? My i7 draws about 7W when fully loaded at 800MHz, and about 27W when fully loaded at 2.7GHz. That's a 3.4x performance improvement at a 3.9x power increase. So, naively, that does result in a fixed amount of work being carried out in a smaller amount of energy, although not anywhere near the extent that you're describing. But this is a very strange workload to be optimising for. First, it's entirely CPU-bound. If it involves IO then you're going to be keeping IO devices in a higher power state for longer, which wipes out the advantage. Second, it makes the assumption that the user doesn't care how much time it takes. That's basically never true. The only reason not to use race-to-idle is because you have an amazingly specific workload, one that's CPU bound and not user-interactive. That discounts pretty much every desktop, mobile and server use case. It's really not worth worrying about. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: cpufreqd as standard install?
On Thu, Mar 08, 2012 at 11:03:42AM -0500, Phillip Susi wrote: > On 3/8/2012 9:47 AM, Matthew Garrett wrote: > >This is (broadly speaking) untrue. There's a bunch of fixed costs that a > >naive P=IV² doesn't take into account. Assuming a fixed amount of work, > >race to idle is almost always the most power efficient strategy. > > What fixed costs? If you spend 5 seconds working at full throttle > and consuming 100 watts, and then the next 25 seconds in deep C6 > consuming 0 watts, you've spent 500 joules of energy. If you > instead spend 10 seconds working at half frequency, consuming only > 30 watts, then the next 20 seconds in deep C6, you've only spent 300 > joules for the same work. When you factor in the typical increased > execution efficiency you get at the lower frequencies, you probably > could finish that work in only 9 seconds, cutting the energy > expenditure down to 270 joules. Yes, if those are the actual power figures. But they're typically not going to be. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: cpufreqd as standard install?
On Mon, Mar 05, 2012 at 09:39:34PM -0500, Phillip Susi wrote: > With correct frequency management, the lower power per instruction of > the lower frequencies outweighs the reduced time in the lower C > states. This is (broadly speaking) untrue. There's a bunch of fixed costs that a naive P=IV² doesn't take into account. Assuming a fixed amount of work, race to idle is almost always the most power efficient strategy. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: GNOME Panel dropped in 11.10
On Thu, May 05, 2011 at 11:46:24PM +0300, pec...@gmail.com wrote: > No, it is not the plan. Current GNOME Panel maitaner claimed that he > will support it as long as there will be neccessity. And bear in mind > that even if he drops towel, someone can take his place (not everyone > of course, but if there is enough need for that, someone will step > up). Mesa will shortly be sufficient for running gnome-shell even on systems without hardware 3D. At that point there won't be any necessity for gnome panel, although it's possible that someone will want to maintain it anyway. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: CPU scaling vs Temperature
On Sun, Mar 06, 2011 at 11:24:35AM -0500, John Moser wrote: > Why not have a user-configurable critical temperature? 'sensors' > reports my CPU's critical is 95C; but my system shuts down if I maintain > above 80C long enough. How about at 75C, CPU frequency scales down to > minimum? This seems sensible. The value can be user-adjustable. echo 75000 >/sys/class/thermal/whatever/passive -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Ubuntu for laptops
On Tue, Sep 22, 2009 at 10:40:05PM +0530, shirish शिरीष wrote: > a. Lots of background services which are started by default. Background services are either doing something (in which case the user expects them to be doing so), are idle (and therefore not consuming any energy) or are buggy (in which case they should just be fixed). > b. Also the kenel as well as packages should behave in a manner which > make battery life longer. This is not the case at the present . > > For e.g. > http://forums.fedoraforum.org/showthread.php?s=84bfb8cd28895385bee083cc30eed476&p=1267038#post1267038 No. If there are kernel bugs that increase power consumption, they need fixing for desktops and servers as well as laptops. > The use-cases are unique to laptops and laptops in battery-mode. Any > benefits on this are surely going to overlap with UNR as well. (as it > suffers from similar constraints) The use-cases are not unique to laptops. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Ubuntu for laptops
On Tue, Sep 22, 2009 at 11:40:27AM -0500, Patrick Goetz wrote: > If you google for ext4 and battery life you can find quite a bit of > discussion about how the default filesystem configuration prevents the > disk from being able to go to sleep. The bottom line is optimizing for > performance is more or less inversely proportional to optimizing for low > energy consumption. So yes, it would be useful to have a > laptop-optimized version. As far as filesystems go, pretty much everything that improves battery life does so by reducing the number of accesses - which also improves performance. I don't think that's the tradeoff you're thinking about. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Ubuntu for laptops
On Tue, Sep 22, 2009 at 09:37:01PM +0530, shirish शिरीष wrote: > Has there been any thought of having a distribution specifically for > laptop users (similar to the initiative taken for netbooks - Ubuntu > Netbook remix) otherwise laptop owners have to go through quite few > hoops to make it less battery intensive. What would the differences be? -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Call for Testers: Karmic kernel with sound controller powerdown fixes
On Sat, Jul 25, 2009 at 01:56:46PM -0400, Daniel Chen wrote: > [0] > https://lists.ubuntu.com/archives/ubuntu-devel-discuss/2009-May/008239.html > [1] http://kernel.ubuntu.com/~dtchen/test-kernels/powerdown/ I know that this is a very pedantic point, but when distributing kernel builds (or, indeed, any other GPLed code) it's helpful to include the source (or a pointer to the source) alongside it - otherwise you're personally responsible for providing the source to that precise binary for the next three years in accordance with the written offer you need to ship with or in the binary. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: problems with ubuntu in general
On Thu, May 28, 2009 at 04:21:28AM -0400, Mackenzie Morgan wrote: > On Monday 25 May 2009 11:15:21 am solaris manzur wrote: > > b) besides that OMG "Hibernating and suspending" This is a big trouble, > > > > and second, it is not working in all pcs why? > > Because many PCs have broken hardware (ie, the BIOS loaded onto the > motherboard reports incorrect information and violates the ACPI standard). > This is the cause of the majority of weird hardware issues. Most suspend/resume issues at this point are due to Linux deficiencies (such as our inability to reprogram nvidia hardware from scratch) rather than spec violations by the manufacturer. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Fake login screens
On Sat, Feb 14, 2009 at 06:54:03PM +0100, Vincenzo Ciancia wrote: > However, it seems to me that nobody is getting the point about fake > login screens: if I am an *user* of somebody else's network, how can I > protect myself from another *user* faking a login screen, used as the > only running X application, and stealing my password? ctrl+alt+backspace never protected you from that. It's a mappable keystroke, in the same way that ctrl+alt+fwhatever are. A malicious client could remap it away to something else, grab ctrl+alt+backspace, fake an X server restart by changing DPMS mode a few times and then give you a fake login screen. Arguing that something's a security feature without checking that it's actually a security feature isn't a good plan. -- Matthew Garrett | mj...@srcf.ucam.org -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: external monitor defaults in 8.10
On Tue, Nov 18, 2008 at 10:31:28PM +, Ian Lynch wrote: > On my IBM R40e Fn F7 does nothing - in fact none of the function keys > work at all! After a lot of messing around I have managed to get output > to a data projector to work but the local screen is very dim and Power > Manager Brightness Applet 2.24.0 has no effect and neither do the Fn > Home or end buttons. Turns out the R40e has an amusingly broken BIOS that results in the kernel listening for events in the wrong place. Should be fixed in the upstream kernel before too long. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Midnight Commander in 8.10
On Wed, Oct 29, 2008 at 05:01:33PM -0400, Felix Miata wrote: > On 2008/10/29 16:28 (GMT-0400) Phillip Susi composed: > > > How exactly is mc vital to fixing a broken X? > > It's vital to fixing broken __, __, __, ___, etc., etc. in > generic fashion (Citation needed) -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Joysticks/joypads/etc information needed for Ubuntu 8.10 Intrepid Ibex and later
Something like this:
diff --git a/hald/linux/device.c b/hald/linux/device.c
index 11f79cd..eff4ee5 100644
--- a/hald/linux/device.c
+++ b/hald/linux/device.c
@@ -1060,16 +1060,27 @@ input_test_abs (HalDevice *d, const char *sysfs_path)
hal_device_add_capability (d, "input.touchpad");
goto out;
} else {
- /*
-* This path is taken by VMware's USB mouse, which has
-* absolute axes, but no touch/pressure button.
-*/
- hal_device_add_capability (d, "input.mouse");
- goto out;
+ long bitmask_button[NBITS(KEY_MAX)];
+
+ s = hal_util_get_string_from_file (sysfs_path,
"capabilities/key");
+ if (s == NULL) {
+ hal_device_add_capability (d, "input.joystick");
+ goto out;
+ }
+
+ if (test_bit(BTN_LEFT, bitmask_button)) {
+ /*
+* This path is taken by VMware's USB mouse,
which has
+* absolute axes, but no touch/pressure button.
+*/
+ hal_device_add_capability (d, "input.mouse");
+ goto out;
+ }
+
+ hal_device_add_capability (d, "input.joystick");
}
}
-/* TODO: Hmm; this code looks sketchy... why do we do !test_bit on the
Y axis ?? */
if (test_bit(ABS_X, bitmask) && !test_bit(ABS_Y, bitmask)) {
long bitmask_touch[NBITS(KEY_MAX)];
@@ -1084,6 +1095,9 @@ input_test_abs (HalDevice *d, const char *sysfs_path)
hal_device_add_capability (d, "input.tablet");
}
}
+
+ if (test_bit(ABS_WHEEL, bitmask) || test_bit(ABS_THROTTLE, bitmask))
+ hal_device_add_capability (d, "input.joystick");
out:
;
}
--
Matthew Garrett | [EMAIL PROTECTED]
--
Ubuntu-devel-discuss mailing list
Ubuntu-devel-discuss@lists.ubuntu.com
Modify settings or unsubscribe at:
https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Joysticks/joypads/etc information needed for Ubuntu 8.10 Intrepid Ibex and later
Surely the correct solution is to fix hal so it flags these devices as input.joystick and not input.mouse? input_test_rel in hald/linux/device.c looks pretty dumb. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Bugs for NM 0.7
On Mon, Sep 08, 2008 at 11:51:59AM +0200, Alexander Sack wrote: > On Mon, Sep 08, 2008 at 10:44:49AM +0100, Matthew Garrett wrote: > > The thread was discussing the removal of network-admin - doesn't that > > modify /etc/network/interfaces? > > Yes it does that atm. But if network-admin is still wanted in the long > run - it could also write keyfile system configurations. Right, but I believe the issue is that without network-admin there's no graphical means of configuring the legacy networking interface. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Bugs for NM 0.7
On Mon, Sep 08, 2008 at 11:31:07AM +0200, Alexander Sack wrote: > On Mon, Sep 08, 2008 at 12:26:50AM +0100, Matthew Garrett wrote: > > To be fair to NM, this is a Debian/Ubuntu integration issue. System-wide > > configuration is present but requires a system-specific backend. > > > > NetworkManager has a distro independent system-wise backend called > "keyfile". And thats enabled by default in ubuntu. What isnt enabled > by default yet is the legacy backend for /etc/network/interfaces, but > that isnt the problem this is about afaict. The thread was discussing the removal of network-admin - doesn't that modify /etc/network/interfaces? -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Bugs for NM 0.7
On Sun, Sep 07, 2008 at 10:27:10PM +0300, Peteris Krisjanis wrote: > Btw, I haven't seen that system wide configuration on OpenSUSE and > Fedora. I would like to see it in action. So far I am very nervous > about ditching network-admin, because no matter how it was stuck in > development, or it lacked features, it worked, it had "over the > distros" feel and so far Network Manager has been "let's repeat > PulseAudio" all over the place. There should be very good > network-admin and NM integration or at least NM should heavily improve > their configuration dialogs and menus. Otherwise I still suggest to > leave network-admin and work on NM to improve it to get it finally > worthy to ditch good old g-s-t tool for good. To be fair to NM, this is a Debian/Ubuntu integration issue. System-wide configuration is present but requires a system-specific backend. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Intrepid compatibility with C3 CPUs
On Mon, Aug 18, 2008 at 07:31:48PM +0100, Sam Tygier wrote: > It appears that the kernel now requires the 'cmov' instruction. If i > understand right this is a i686 instruction. It's an optional i686 instruction. CPUs are allowed to claim i686 support without implementing it. Unfortunately, it's also about the only generally useful i686 instruction, so gcc generates it if you ask it to build for i686. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: usplash and alternate resolutions
On Tue, Jul 22, 2008 at 01:27:12PM -0400, Bill Filler wrote: > Hello, > Does anyone know if there is work underway in usplash to support > resolutions such as 1024x600 and 1280x800 (without stretching the > image), which are proving to be quite common in the netbook space? If > not, any hints as to the efforts of adding this support would be > appreciated. The most practical way of doing so would be to add code to parse the vesa mode list in an attempt to find a mode that matches the actual screen resolution, but a lot of hardware won't have this. Beyond that, wait for kernel modesetting support. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: pm-utils vs acpi-support
On Mon, Jul 14, 2008 at 01:26:31AM +0200, Przemysław Kulczycki wrote: > What's the status of pm-utils in Ubuntu? As far as I can see in Ubuntu > Hardy, pm-utils is installed but it's not used by default because > acpi-support is still present. It's used by default. > Maybe this could bring us closer to fixing the unfamous bug 59695. I don't see how. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: acpi-support: Lock screen with dbus during resume
I haven't actually tested this, but are you sure it works? You need the environment variable providing the DBus socket address, and that's only available from the user's running session. Also, I'm no longer involved in acpi-support development - patches should probably be sent to either Ubuntu or Debian's bug tracking system. Thanks, -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Wacom config hacks (was: An example of how things should *not* be done)
On Wed, Apr 23, 2008 at 03:41:52AM -0700, Bryce Harrington wrote: > No. That bug was mostly cosmetic IMHO, aside from the .xsession-errors > file growth in Kubuntu. Note there were dozens of other similar bug > reports that derived from this issue. In fact, a lot of users confused > the wacom errors in their Xorg logs for the cause of their (unrelated) X > problems. The bug that actually motivated this change was #83860 from > Henrik Omma. I talked with him at length about the pros/cons of this > change, but given that the "fix" for tablets was an ugly hack, and that > it was only needed for a small subset of tablet users anyway (who could > still uncomment the entries in xorg.conf to enable it), and that it > resulted in a completely unusuable Ubuntu for users needing accessible > login, we decided it was better to go with the current approach. The problem in this case is that there's no record of the discussion in question. Upstream asked for more information about the issue over a year ago, and it's clear that it's a gdm bug (I've changed the state of #83860, since there *wasn't* a fix released in gdm). Instead of fixing the problem, it was worked around in a way that reduces the functionality of the distribution. There's no record of why this decision was made, especially since upstream seem to be on record as being willing to fix it. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Please keep the old IDE chipset drivers so that external CD writers can work
No, something else is going on here. usb-storage is entirely unrelated to the IDE subsystem, so doesn't care whether you're using libata or not. Something else is triggering this bug. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Data retention reminder on live CD
On Mon, Mar 03, 2008 at 10:41:40PM +, (``-_-´´) -- Fernando wrote: > That might confuse an user who is trying to install (Ed/K/xpto)ubuntu to > disk, wouldnt it? > Also, with all the bugs with framebuffer, many users wont even be able to see > anything until they reach the Desktop. Just for the avoidance of confusion, we don't use a kernel framebuffer on x86 or x86_64. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Packaging kernel modules.
On Thu, Mar 06, 2008 at 12:17:15AM +0900, John McCabe-Dansted wrote: > What process do I need to follow to get this approved? > > (I have the source package and > http://www.debian.org/doc/maint-guide/ch-update.en.html > I can create a linux-ubuntu-modules source package that includes > compcache, but will that actually help anyone?) If you can create a diff to linux-ubuntu-modules and put it in launchpad, and then contact [EMAIL PROTECTED], that ought to do. I wouldn't necessarily expect this to happen until after hardy's released, though. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Packaging kernel modules.
On Wed, Mar 05, 2008 at 11:08:00PM +0900, John McCabe-Dansted wrote: > I am trying to package compcache. A much better solution is to get it included in linux-ubuntu-modules. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: acpi-support patches for Ubuntu hardy that need your attention
A number of these bugs are irrelevant for hardy, since the move to pm-utils. The lack of hotkeys_over_acpi is a kernel bug that's happened in almost every release I can remember, and I'm bored of asking the kernel team to ensure that patches don't get dropped between releases and being told it won't happen again. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: acpi-support patches for Ubuntu hardy that need your attention
On Fri, Feb 29, 2008 at 12:52:46AM -0600, Jerone Young wrote: > There is a round up of acpi-support patches that are waiting to go > into hardy. You are listed as the maintainer of acpi-support. The > launchpad bugzilla is here and it is a milestone for hardy Alpha-6 .. > > https://bugs.launchpad.net/ubuntu/+source/acpi-support/+bug/194609 The bug is incorrect. That's not an eject button, it's a dock eject request button. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: madwifi-source
On Fri, Feb 22, 2008 at 07:11:08AM +0900, Onno Benschop wrote: > I understand that, however, if you have a machine that has a card that > is not supported by the linux-restricted-modules, you would use > module-assistant to create a module to match your kernel. > > If you had the madwifi-source package, you could patch it and compile a > module in such a way that it would continue to be maintainable, rather > than get the source from madwifi.org, unpack it, make and make install > it and have unknown files scattered all over your file-system. Or you could apply the patch against linux-restricted-modules. > Finally, if the madwifi-source isn't available, then I suspect there's a > bug in module-assistant, seeing that it still has madwifi as an option. Yes, that's true. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: madwifi-source
The madwifi code is already in linux-restricted-modules, so there's no benefit in providing a separate source package as well. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: HAL fix for multiple battaries
On Wed, Feb 20, 2008 at 09:19:04AM -0800, Ted Gould wrote: > Yes, but that HAL patch ignores all battery entries in /proc if it finds > one in /sys. Which seems rather risky to me... but I was curious if > that seems logical to those who know the kernel better. Yes, the only way a battery can end up in /proc/acpi is if it's supported by the acpi battery driver, which has now been ported to sysfs. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
Though, having said that, I have found various issues - I've just uploaded a new pm-utils. It'd be nice to know if it worked for you. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
I can't see how this can happen - the same environment variables are set in both cases independent of the ordering. Is this repeatable? -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
On Tue, Jan 15, 2008 at 09:30:13AM -0500, Paul S wrote: > This is not good. I had a kernel update so had to reboot anyway. Now > I find that even Fn-Esc does not work. The only way to suspend is by > konsole with "sudo pm-suspend" and it still works ok. The > kde-power-manager menu "suspend" does not work. either nor the keyboard > "sleep" button. Ok. Sounds like KDE is failing to respond to the sleep button signal. I'm afraid I've no expertise beyond this point. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
On Tue, Jan 15, 2008 at 08:25:38AM -0500, Paul S wrote: > but same result .. /var/log/pm-suspend.log does not change, suspend / > resume work only from laptop Fn-Esc, not kde-power-manager menu or usb > keyboard sleep button and no "logger" message in /var/log/messages. > > Do I need to reboot after changing that file? If so, I did not. acpid needs restarting, so rebooting is the easiest way to do that. If Fn+Esc no longer works, there's some sort of KDE issue. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
On Mon, Jan 14, 2008 at 09:17:29PM -0500, Paul S wrote: > Since kde-power-manager won't work and my usb keyboard sleep button > doesn't work, I tried the Fn-Esc key combo on the laptop, which is the > suspend key combo. The laptop suspended and resumed fine. Now, when I > look at /var/log/messages, there is no line from the logger that I > inserted above. Also, when I look at /var/log/pm-suspend.log, it never > changed from the previous suspend. So, it appears something else is > functioning to suspend, other than pm-utils via hal. Yeah, something's clearly catching that. Can you try changing /etc/acpi/events/sleepbtn to call sleepbtn.sh and not sleep.sh? -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Proposal: include Brasero by default
On Tue, Jan 15, 2008 at 12:08:06AM +0100, Wouter Stomp wrote: > I would not consider features like inhibiting Gnome Power Manager from > suspending while burning niche, that is something that should just > work. Well, that's clearly a bug that needs fixing in Nautilus in any case. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
On Mon, Jan 14, 2008 at 11:10:49AM -0800, Matt Price wrote: > sudo pm-suspend Right, so it sounds more like the issue here is that we're passing quirks that are breaking your system. For reference, the quirks that are used by default on Ubuntu are currently: --quirk-dpms-on --quirk-vbestate-restore --quirk-vbemode-restore --quirk-vga-mode3 --quirk-vbe-post --quirk-reset-brightness If you pass those, does it fail? If so, can you try to figure out which one is causing it? -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
On Mon, Jan 14, 2008 at 10:30:33AM -0800, Matt Price wrote: > ... i suppose if the resume worked properly the reverse steps would be > present too? You'd hope so :) How exactly is the resume failing, and which graphics drivers are you using? > ah, this is my bad -- it seems an old script i had in > in /etc/acpi/lid-hibernate.sh was being called on lid closure; i thought > i'd verified that my old scripts had been replaced by the standard ones > on upgrade, but that one escaped my attention. that script calls > hibernate, which suspends & resumes successfully, while pm-utils appears > not to. Ah, interesting. Yes, that would explain the difference. I'll take a look at hibernate and see what it's doing by default (I haven't checked it out in some time). We are talking about suspend to RAM here, right? -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
On Mon, Jan 14, 2008 at 10:57:13AM -0500, Paul S wrote: > Are you the contact for these kubuntu issues as well as the ubuntu lead? No, I'm afraid I've got no experience of KDE at all. > Is it too early to start feedback on this stuff now, or should we > continue? Do you want us to feedback on this list or start bugs? Now is a fine time to provide feedback - there shouldn't be any major changes. Filing bugs is ok, but if you think an issue is generic then feel free to bring it up here. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Proposal: cdrkit vs. cdrtools
On Mon, Jan 14, 2008 at 05:23:11PM +0100, Stephan Hermann wrote: > So, regarding the removal of cdrtools form the archive, should we change > cdrkit binary packages to Provide: the old cdrtools binary package > names, and adding symlinks for the new binary tools to the old cdrtools > names, like gentoo is doing so? Yes. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: tracking down suspend/hibernate bugs?
On Sun, Jan 13, 2008 at 10:51:48PM -0800, Matt Price wrote: > in hardy it appears that acpi-support has been cut out entirely in favor > of pm-utils -- see the attached hal-system-power-suspend-linux script. > i've flipped through the pm-utils code & can't see exactly how it's > determined what is used to make the actual suspend call -- for instance, > i'm not sure what I would do if i wanted to use a custom kernel with > tuxonice for suspend, which i think still works best if it's called by > the hibernate script. If it's triggered by writing disk to /sys/power/state (which I believe it is, nowadays) then pm-hibernate will trigger it happily. Please don't use the hibernate script - any cases where it works and pm-utils doesn't are bugs that need fixing. > i alsoseem to have an odd problem -- if i choose 'suspend' from the > gnome 'quit' menu, then resume fails. if, however, i suspend just by > closing the lid on my laptop, suspend and resume both succeed. this > suggests to me that two different methods are being used, probably acpi > in the lid case and pm-utils in the gnome case. seems to me this > shouldn't happen; again, i'm not quite sure what's going on. Both cases should be the same. I'll check to make sure that the logout dialog is triggering pm-utils, but lid closure should certainly be going through hal. > apt cache gives this list as the ubuntu maintainer of pm-utils; the > changelog suggests that the package has just been ported over unchanged > from debian. if it's going to be the main power management tool, it > seems like an important piece of infrastructure; is anyone at ubuntu > watching over it at all? if so, please let me know what i can do to > help with debugging and stuff. Yes, I'm looking after it. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Proposal: cdrkit vs. cdrtools
On Mon, Jan 14, 2008 at 08:51:12AM -0500, Forest Bond wrote: > Didn't we just move back to cdrtools from cdrkit? Weren't these issues > resolved, or something? > > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=270060 That's from 2005. > Debian seems to be happy. Let's leave it be, hmm? No - cdrtools still links GPLed code into a CDDLed binary. It's undistributable in its current form. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: fsck on boot is major usability issue
On Fri, Dec 21, 2007 at 09:37:35AM +0100, Martin Pitt wrote: > We quickly discussed this at the last UDS. Most people were not in > favor of dropping the check completely, since occasionally, things > just go wrong, and you never notice until you actually run a check. Any severe corruption is likely to trigger the kernel's own sanity checks, at which point a check will be forced anyway. Otherwise we seem to be optimising for a massively uncommon case at the expense of everyone else. If ext3 had a habit of introducing corruption, we'd know about it by now. We should just skip the time and count based checks. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Changing dpkg-deb default compression from gzip to lzma for Hardy
On Tue, Dec 18, 2007 at 11:09:50AM +0100, Thilo Six wrote: > Only with OOo core moved from gz to 7z we save: 39084 - 27358 = 11726kB on > the CDs!! The LiveCDs don't include the binary debs - the compression is handled by squashfs. I /suspect/ that the limiting factor in the liveCDs is the speed of the CD drive rather than the compression mechanism, so investigating lzma there might be helpful. There appears to be an lzma-based fork of squashfs at http://www.squashfs-lzma.org/ -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Hibernate and Restriced Drivers (Was: 4 More days...)
On Tue, Oct 16, 2007 at 09:33:22PM +0200, Paulus Esterhazy wrote: > This bug, or this group of bugs, will be a source of annoyance to many > users. Basically, when you use restricted drivers (both NVidia and ATI), > your system will fail to resume from hibernation most of the time. As > restricted drivers are enabled by default, this should be considered a > regression from feisty. They are? > I hope this doesn't sound ungrateful. Ubuntu developers are doing a very > good job overall, and dealing with binary blobs isn't an easy task. It's > alright to know that something is broken right now, but it's worrying to > have the impression that no solution is in the offing at all. I'd love > to see some sort of Hibernation team created that tries to tackle the > problem in a systematic way. I've tried. We can't. The lack of source means we have absolutely no idea what these drivers do over suspend/resume, and there's no way to figure out what we should be doing in response. The long-term solution involves moving modesetting and video initialisation into the kernel, but I strongly suspect that that will end up breaking the binary drivers for a significant period of time until they adjust. I don't see any way we can make this work properly within the next 18 months. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Archive frozen for Gutsy release
On Wed, Oct 17, 2007 at 12:32:43AM +0100, Matt Hoy wrote: > Steve, > > Pretty major bug, yet seemingly simple fix, affects a fair number of people. If the fix were simple, it would have been fixed. We've no idea what's causing this, and it doesn't help that it's been filed against the wrong package. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: 4 More days...
With the exception of the Thunderbird and timidity issues (and I can't reproduce the timidity one here at all), every bug you've responded to in gutsy appears to be down to your issue with Hal. And, judging by the log there, your system was entirely broken: 03:13:09.373 [W] ids.c:294: Couldn't stat pci.ids file '/usr/share/misc/pci.ids', errno=13: Permission denied 03:13:09.373 [W] ids.c:515: Couldn't stat usb.ids file '/usr/share/misc/usb.ids', errno=13: Permission denied 03:13:09.373 [E] osspec.c:310: Unable to inotify_add_watch() for '/usr/share/hal/fdi/preprobe': Permission denied indicates either filesystem corruption or that something has heavily screwed with the permissions. It certainly doesn't look anything like a hal bug. What are you actually complaining about? -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Restricted tab-completion is annoying
On Thu, Oct 11, 2007 at 06:43:15PM -0700, Aaron C. de Bruyn wrote: > Agreed. I would love to have a system-wide disable option and/or a > per-account option. > For now I'll settle for what Gavin said in his message to the list. Toss > 'shopt -u progcomp' into your .bashrc Or stick it in /etc/bash.bashrc and then allow individual users to turn it on if they want to. The changes won't be automatically overwritten on upgrade. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Restricted tab-completion is annoying
On Fri, Oct 12, 2007 at 09:02:05AM -0400, Forest Bond wrote: > Enough with the madness. Please keep the Ubuntu code of conduct in consideration when posting to lists. Inflammatory language doesn't help reach any sort of solution. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: GL-screensavers acting up in Gutsy
On Sat, Oct 06, 2007 at 02:55:17AM +0300, Matthias Andersson wrote: > I noticed when I tried to reconfigure my screensaver that the GL > screensavers began to act strangely, in the preview window they work > fine until I drag a window on top of the preview, the window starts to > blink and become distorted. This is due to technological limitations in the current DRI implementation. Disable compiz to avoid these issues. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: That need to close bugs?
On Thu, Sep 20, 2007 at 12:58:25PM +0100, Colin Watson wrote: > I agree entirely. What drives *me* batty in turn is when people take a > confirmed, complete report, ask why it hasn't been fixed yet, and close > it as invalid because obviously something that's been around for a > couple of releases can't be relevant any more. Or when people reject > perfectly valid and complete wishlist bugs (note that it usually takes a > lot less for a wishlist bug to be complete) because they should be > specifications (they shouldn't, in most cases) or just because they're > wishlist. As a developer, I wish I didn't have to spend time checking my > bug mail just to make sure that well-intentioned but mistaken triagers > aren't taking items off my to-do list that I want to stay there. Something appears to have flagged all inactive bugs as invalid. Apart from generating a ridiculous quantity of email for no obviously good reason, a pile of perfectly valid wishlist bugs or issues that require further work in the rest of the distribution first have suddenly closed. Who on earth thought that this was a good idea? -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Apturl (security) issues and inclusion in Gutsy
On Tue, Sep 18, 2007 at 12:25:00PM +0200, Alexander Sack wrote: > > 2. Repositories added through apturl could provide packages included > > in Ubuntu but with higher version numbers with malicious code. > > ... this is a feature, not an issue. I'm really not convinced by that. We shouldn't be making it easier for users to replace important system files, and we certainly shouldn't be making it easier for arbitrary third parties to encourage them to do so. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Default option for volume hotkeys (speakers/headphones)
On Sun, Aug 26, 2007 at 03:33:06PM +1200, Aaron Whitehouse wrote: > 3) Cleverly determine whether headphones are plugged in. If they are, > make the hotkeys control "Headphones" and if they are not, control > "Master". 4) Alter the kernel so that it binds the headphone and master channel together. That's how it's handled on various pieces of hardware. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: liblicense, a way for creative commons stuff at desktop
On Wed, Aug 22, 2007 at 01:32:01AM +0530, shirish wrote: >Please lemme know whichever way is cool. Btw it would be cool > if we could package it in gutsy. Somebody wants to take a shot at it > ;) We're past feature freeze, so it's unlikely for gutsy. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: too much of a time hence marking a bug invalid
On Mon, Aug 20, 2007 at 06:51:48PM +0530, shirish wrote: > Please look at > https://bugs.launchpad.net/ubuntu/+source/totem/+bug/128899 I am > looking at it purely from a user's point of view. You filed the bug on the 28th of July. On the 7th of August (just over a week later), the maintainer examined it. Today you were asked for some more information, you provided it and then (as far as I can tell) you marked the bug invalid. Bear in mind that you are running a development version of an operating system, and things are expected to break at certain points. Some of these things are much more important than others, and so people will target those first. It's unsurprising that a relatively unimportant bug such as this will be looked at with lower priority than "My panel keeps crashing" or "Openoffice deleted my home directory", but it doesn't alter whether the bug is valid or not. Plese do not mark real bugs as invalid just because you feel that it's taken too long. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Support for multimedia/internet keyboards
On Wed, Aug 15, 2007 at 11:54:24AM +0200, Pascal De Vuyst wrote: > Currently Ubuntu provides no out of the box support for the extra multimedia > or > internet keys on desktop keyboards. For any keyboard that sends the same events as used on the Microsoft multimedia keyboards (which is most of them), this should work out of the box. Logitech ones, not so much. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Checksums Done Right
On Sun, Jul 01, 2007 at 02:59:03PM -0700, [EMAIL PROTECTED] wrote: > > With modern hardware the sensible thing to do is just to boot from CD. > > With modern hardware shutting a dom0 down might mean taking out 10+ > active, virtualized servers (in a HA environment it means live migrating > those other servers). Assuming your dom0 is secure, rebooting only the > domU you wish to check is sufficient and ideal. I expect tools to emerge > that will allow one to analyze/validate a domU's kernel, loaded modules, > and memory from the dom0 but until then shutting down individual domUs > will have to do. Yes, if you're already running in a virtualised environment then providing a mechanism for checking the system makes sense. I'm just not sure it's a compelling reason to move from a non-virtualised system to a virtualised system. On the other hand, it ought to be possible (in most cases) to skip using LVM snapshots. There's plenty of existing userspace filesystem code that can read files from raw block devices (think grub, for instance), so you should be able to scan the filesystem from the dom0 without shutting it down or using LVM. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Checksums Done Right
On Sat, Jun 30, 2007 at 04:21:11PM -0700, [EMAIL PROTECTED] wrote: > Escalation of privileges is one attack, yes. Although the type of "attack" > I'm talking about is for users that already have the ability to write a > root-owned binary. I'm describing more of a DoS attack that basically just > keeps admins scratching their heads. It doesn't have to be a useful > collision to cause headaches. The main point is that md5 is broken and > should be retired[1] (note: that's the "R" in RSA writing that "md5 is > broken" not just me). MD4 was broken in terms of finding collisions some time ago, but it's still not possible to trivially find MD4 collisions for an arbitrary existing file. The fact that (for carefully selected files) it's possible to find collisions for MD5 doesn't mean that it's generally broken or needs replacing immediately. > > So the real benefit is that you can do this on a live system, rather > > than having to reboot to known-good media? > > Potentially, yes. Of course I envision malicious kernel modules being > created that remove themselves from the filesystem while running then at > the last minute before shutting down write what's necessary to load > themselves on boot again. In that case you'd have to shutdown the system > to be certain. With modern hardware the sensible thing to do is just to boot from CD. There's no mechanism for fudging that unless you invoke virtualisation or firmware modification, both of which are still (highly) theoretical attacks. > Then I invite you to join the ongoing "blue pill" debate. That is really > outside the scope of CDR but still an interesting attack vector > none-the-less. We assume you get the "high ground" first. There's absolutely no way of implementing virtualisation without making certain instructions take an extra few cycles. Sensitive enough measurements ought to be able to pick that up. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Checksums Done Right
On Sat, Jun 30, 2007 at 09:14:17AM -0700, [EMAIL PROTECTED] wrote: > Ahh, you are correct. I was thinking of kernel-based rootkits being > common. I have no evidence that states collision attacks are currently > common. To clarify, it's trivially easy, using freely available source > code[1] (31 secs/file now), to attack a system so that some valid > executables have the same checksum as the vendors distributed copy but do > something completely unexpected. If nothing else changes with those files > (permissions, size, owner, group, time) it would easily fool many admins. Right, but being able to create a collision isn't the same as being able to create a *useful* collision. You need to be able to alter the functionality of the program in a very specific way in order to use it to escalate privileges. I'm not aware of anyone being able to demonstrate that with arbitrary executables yet. > The way we run our dom0s is that they are not listening on the network, > they have no users (other than admins), run little (mainly ssh-client) > non-base install software, and they are physically secure. We have not yet > seen a domU -> dom0 escalation attack (anyone else?). It may come > eventually but thankfully it's not here yet. We could also build Xen from > source, and examine the Xen diffs in great detail, but we aren't *that* > paranoid, yet. Really the only known way to compromise a system and kernel > in this environment is to control the mirror/media, control the Xen build > environment or, control ring -1 (think "blue pill"[2] - heh installing Xen > inside an already virtualized system would quickly degrade the quality of > life). So the real benefit is that you can do this on a live system, rather than having to reboot to known-good media? (I'm sceptical about the idea of attackers being able to virtualise a system without anybody noticing. Latency of privileged instructions would change in a pretty obvious way) -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: future of xserver-xorg-video-(i810|intel)?
On Fri, Jun 29, 2007 at 08:58:56PM +0200, Andreas Schildbach wrote: > As the i810 driver cannot handle odd resolutions well, when will the > default be switched over to intel? Will this happen at all for Gutsy? There's still a small number of important regressions in -intel compared to -i810. These are being fixed quite rapidly, so I'd hope we'll be able to cover it for gutsy. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Checksums Done Right
On Fri, Jun 29, 2007 at 05:04:14PM -0700, Scott Beardsley wrote: > Most Ubuntu packages (95% - my estimate) come with MD5 checksums at the > file level (in a file called md5sums in control.tar.gz). debsums uses > these (well actually a cache stored in /var/lib/dpkg/info/*.md5sums) for > doing a *rough* verification that what is installed matches what > *should* be installed. This is great until md5 collision attacks[1] and > kernel-based rootkits are used on your system (common these days). Do you have any references to the use of md5 collision attacks being common? The Wang and Yu attack requires the binaries to be the same size and to differ only in very controlled ways. It's not difficult to construct examples of collisions, but I'm not aware of anyone demonstrating the ability to replace an arbitrary binary with a trojaned one with the same md5sum. > We have been working on a to-be-open-sourced product we are calling > Checksums Done Right (CDR). A colleague gave a talk last week that > included some notes about CDR[2]. Basically we've processed the md5sums > files in dapper, edgy, and feisty and dumped it into a database. When we > update our mirror we update our database. The mirror seems like the best > place to offer this type of verification service. We have used it to > verify binaries on Xen installations by taking LVM snapshots of the > virtualized machine and sending checksums to the mirror using ssh all > from the dom0. Our tests show that we can verify a system installation > (libraries, binaries, and kernel modules) of up to 12k files in around 4 > seconds. This theoretically scales to 5k full machine scans per mirror > per day. It's possible that I'm missing the point here, but what guarantees do you have that you can trust your Dom0? -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Request for Comments (RFC): Potential change in setting sound card volume on boot
On Tue, Jun 26, 2007 at 09:20:59PM -0400, Daniel T. Chen wrote: > (4) Additionally, users seem split on whether a default login sound in a > graphical environment should be muted[4] or audible. > > To alleviate the issues given above, I am requesting comments from > developers and users alike regarding the following proposal that changes > alsa-utils's initscript usage[0]: I'm not sure this alleviates number 4 - surely it will always result in the login sound being muted? mixer_app won't be started until some time after the sound has started playing. It also leaves us with the issue of what to do with the GDM sound. Muting that by default would be an accessibility problem... -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Draining the font swamp
On Mon, May 21, 2007 at 03:09:12PM -0400, Phillip Susi wrote: > Matthew Garrett wrote: > >Measurements have shown that over pretty much any sort of common > >network, latency is more of a problem than bandwidth. Server-side fonts > >require multiple round-trips between the server and the client for > >rendering, whereas client-side fonts only require the initial display. > >Performance-wise, we have the XRender extension for precisely this sort > >of situation. > > How does server side fonts require more round trips? It should amount > to a single message that specifies what font to use, what text to > render, and where. But it doesn't. I don't think that we can gain significant value from this thread until you've familiarised yourself with the X spec. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Draining the font swamp
On Mon, May 21, 2007 at 10:52:46AM -0400, Phillip Susi wrote: > 3) Performance suffers. The X server is in the best position to render > fonts using any hardware acceleration provided by the video card, and > allows for those fonts to be shared by all applications, reducing > duplication and waste. Also for remote X sessions, you want the fonts > rendered on the server so much less data needs exchanged between the > client and server. Measurements have shown that over pretty much any sort of common network, latency is more of a problem than bandwidth. Server-side fonts require multiple round-trips between the server and the client for rendering, whereas client-side fonts only require the initial display. Performance-wise, we have the XRender extension for precisely this sort of situation. > Other than the fact that the client side implementations have advanced > beyond the X server ones in recent times, is there any advantage to > client side font rendering over server side? If not, then we should > push to bring the client side advancements back into the server where > font rendering belongs. Font choice and layout is hard, and doesn't become any easier just because you've moved that code to a binary that runs as root. Nobody is going to argue in favour of putting a layout engine like Pango in the X server, and most of the rest of the stack is similarly well outside the scope of the X server. The client-side font revolution happened 5 years ago, and we've ended up with massively improved font support as a result. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: we should set a grub password by default
On Tue, May 15, 2007 at 07:23:41PM +0200, Sven wrote: > In short terms: I propose that during grub setup/configuration the grub > password in menu.lst is activated by default. Please let me explain why. Providing a grub password by default risks giving people the impression that the system is secure, while in fact there are several other steps that would be required for that to be true (disabling CD drive booting, BIOS password, physical security of machine to prevent BIOS being reset or drives removed). Instead, we should make it easy for people to learn what needs to be done to make a system secure. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: GetDeb.net vs. Backports...
On Thu, May 03, 2007 at 12:43:07AM -0400, Michael R. Head wrote: > This appears to be resolved, with standard source and debdiff links for > each release of the packages. (I don't know when this happened, but when > I visited the site today, I noticed that the change had been made). Indeed, this seems to have been fix now, so I'm entirely happy. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: libata access to pata drives breaks dma/32bit mode?
On Tue, May 01, 2007 at 12:11:21PM -0400, Phillip Susi wrote: > Are you saying that libata is supposed to always enable 32bit dma transfers? When the drive and controller are both capable of it, yes. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: libata access to pata drives breaks dma/32bit mode?
On Fri, Apr 27, 2007 at 03:05:18PM -0400, Phillip Susi wrote: > I've been seeing several complaints lately from users who have upgraded > to feisty and their ide disks are now being handled by libata. It seems > that hdparm does not work on the devices created by the libata driver, > so how are you supposed to enable 32bit and dma transfers? Several > users have been having trouble because they can not find a way to enable > them and thus, have poor performance under feisty. As has been the case for a long time, the kernel and drive are supposed to negotiate the appropriate speeds themselves. If that's not happening, please file a bug against the kernel. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Call for Release Candidate testing (again)
On Sat, Apr 21, 2007 at 08:09:04AM +0800, Wenzhuo Zhang wrote: > Matthew Garrett wrote: > > No, I don't think that the ordering of devices is a bug - the bug is > > that we're assuming that the Linux device ordering bears some sort of > > relation to the Bios device ordering, which is simply not true. Further, > > Perhaps it is not a bug as far as the kernel is concerned; but it is > definitely a bug in the context of Ubuntu installer, because it will > result in unbootable systems for desktops with add-on PCI IDE controller > cards. Like I said, that has nothing whatsoever to do with the device ordering in the kernel. What's relevant is the BIOS drive ordering, and it's necessary for the installer to use that information rather than assuming that the kernel drive ordering bears any relation to it. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Call for Release Candidate testing (again)
On Wed, Apr 18, 2007 at 09:20:01PM +0800, Wenzhuo Zhang wrote: > Thanks for your reply. However, I knew that hda becomes sda when using > the linux-ata drivers instead of the good old IDE drivers. My problem is > that the primary boot drive is recognized as sdb, and Grub gets installed > onto the MBR of the tertiary master hard drive, which is incorrectly > recognized as sda. The sequence of drives detected by the linux-ata > drivers is unpredictable in my setup.This is clearly a bug. No, I don't think that the ordering of devices is a bug - the bug is that we're assuming that the Linux device ordering bears some sort of relation to the Bios device ordering, which is simply not true. Further, I don't believe that there's any way we can make it true. We should be parsing edd data instead, which lets us work out how the bios ordering corresponds to PCI devices. However, for some reason EDD seems to have been switched off in the Ubuntu kernels. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Where to post feature request for Kernel patches
On Thu, Apr 05, 2007 at 11:54:26AM +0200, Hendrik Baecker wrote: > Hi List, > > I just want to know where I can post a feature request for a kernel > patch to get IBMs Ultrabay into "supported" state. It won't be supported until there's safe PATA hotplug support, which I intend to look into at some stage. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Ubuntu and ext4 file system
On Tue, Mar 27, 2007 at 08:06:05PM +1000, Chris Jones wrote: > When is Ubuntu going to support ext4 file system? When it's not likely to corrupt your files (ie, when it's no longer flagged as EXPERIMENTAL) - to the best of my knowledge, there's still no guarantee that the ext4 on-disk format is stable, so formatting a drive as ext4 now may require you to back up and restore for later kernels. > Will it make it into Feisty Fawn final build? No. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: Hello + laptop hotkeys
On Sun, Mar 18, 2007 at 03:58:26PM +, yelo_3 wrote: > So the question is: is there now a unified method to control wireless power? No. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: GetDeb.net vs. Backports...
On Mon, Mar 05, 2007 at 03:45:12PM -0800, Matt Zimmerman wrote: > I'd never heard of it until now. I'm CCing the MOTU Council, who I hope can > provide a channel to begin a conversation with them. At the moment, there's no source available with the downloads. The site maintainer has indicated that he believes pointing to the upstream website is adequate for this, which is not an opinion that agrees with the FSF (or, indeed, any sane reading of the GPL as far as I can tell). I'm told that at least one copyright holder has requested that the site be made compliant or his software removed - so far, there's been no response to that. I'd suggest that if any form of cooperation is envisaged, we ensure that the site is fully GPL-compliant first. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
Re: TV Out SVideo support
On Tue, Feb 13, 2007 at 03:37:57PM +, (``-_-´´) -- Fernando wrote: > Hya. > I would like to know if there has been any new development for this topic? > This is the last thing that I need to have working, so I can say that > I can do anything on Linux that I did on Windows. As Matt says, this is heavily dependent upon your video driver. Things are getting better, especially with the move to xrandr 1.2. -- Matthew Garrett | [EMAIL PROTECTED] -- Ubuntu-devel-discuss mailing list Ubuntu-devel-discuss@lists.ubuntu.com Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel-discuss
