[Bug 1703264] [NEW] [USB-Audio - USB Device 0x46d:0x819, recording] "chipmunk" noise on logitech webcam input when video also recording
Public bug reported: On upgrade to 17.04, my webcam stopped properly working, in the sense that when video chatting with others, they hear a chipmunk-like noise. I used the word chipmunk because I saw this: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/858412 This issue happens with every video chat application I've tried: hangout (chrome), viber, also cheese. I went through some Ubuntu wiki page about fixing sound issues which made "cheese" record once ok after some force- reset of audio, but not the second time (some crashes seemed to happen as well, where the audio was completely disabled globally until restart). I know the above is not very precise, so here's what I can tell for sure: When video chatting, the audio from my logitech webcam (C210) sounds like a chipmunk after I upgraded to Ubuntu 17.04. This is with the 4.10 kernel. If I boot the same Ubuntu with kernel 4.8 (leftover from Ubuntu 16.10), it works correctly! The above experiments suggest that there is nothing wrong with my config files or package installation. But that it's a regression in the kernel/drivers. ProblemType: Bug DistroRelease: Ubuntu 17.04 Package: alsa-base 1.0.25+dfsg-0ubuntu5 ProcVersionSignature: Ubuntu 4.10.0-26.30-generic 4.10.17 Uname: Linux 4.10.0-26-generic x86_64 NonfreeKernelModules: nvidia_uvm nvidia_drm nvidia_modeset nvidia ApportVersion: 2.20.4-0ubuntu4.1 Architecture: amd64 AudioDevicesInUse: USERPID ACCESS COMMAND /dev/snd/controlC2: sciabaz1860 F pulseaudio /dev/snd/controlC1: sciabaz1860 F pulseaudio /dev/snd/controlC0: sciabaz1860 F pulseaudio CurrentDesktop: Unity:Unity7 Date: Sun Jul 9 18:58:57 2017 InstallationDate: Installed on 2017-04-23 (76 days ago) InstallationMedia: Ubuntu 16.10 "Yakkety Yak" - Release amd64 (20161012.2) PackageArchitecture: all SourcePackage: alsa-driver Symptom: audio Symptom_AlsaRecordingTest: ALSA recording test through plughw:U0x46d0x819 failed Symptom_Card: Webcam C210 - USB Device 0x46d:0x819 Symptom_Type: Digital clip or distortion, or "overdriven" sound Title: [USB-Audio - USB Device 0x46d:0x819, recording] Sound is distorted UpgradeStatus: Upgraded to zesty on 2017-07-05 (4 days ago) dmi.bios.date: 02/23/2016 dmi.bios.vendor: American Megatrends Inc. dmi.bios.version: C.60 dmi.board.asset.tag: Default string dmi.board.name: B150 PC MATE (MS-7971) dmi.board.vendor: MSI dmi.board.version: 2.0 dmi.chassis.asset.tag: Default string dmi.chassis.type: 3 dmi.chassis.vendor: MSI dmi.chassis.version: 2.0 dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvrC.60:bd02/23/2016:svnMSI:pnMS-7971:pvr2.0:rvnMSI:rnB150PCMATE(MS-7971):rvr2.0:cvnMSI:ct3:cvr2.0: dmi.product.name: MS-7971 dmi.product.version: 2.0 dmi.sys.vendor: MSI mtime.conffile..etc.modprobe.d.alsa-base.conf: 2017-04-23T21:54:23.554412 ** Affects: alsa-driver (Ubuntu) Importance: Undecided Status: New ** Tags: amd64 apport-bug zesty -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1703264 Title: [USB-Audio - USB Device 0x46d:0x819, recording] "chipmunk" noise on logitech webcam input when video also recording To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/alsa-driver/+bug/1703264/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1374383] Re: Clicking on "Files" opens new window, rather than switch to already open one
Since 16.04 (I think), where Files changed its behavior to show mounted drives in another window (a bad choice in my opinion, but whatever) this problem is not reproducible anymore. I marked as fixed thus. ** Changed in: unity (Ubuntu) Status: Confirmed => Fix Released -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1374383 Title: Clicking on "Files" opens new window, rather than switch to already open one To manage notifications about this bug go to: https://bugs.launchpad.net/nautilus/+bug/1374383/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1550939] Re: Tab on Index makes evince go crazy
Not reproducible on newer evince. Marked as fixed. ** Changed in: evince (Ubuntu) Status: New => Fix Released -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1550939 Title: Tab on Index makes evince go crazy To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/evince/+bug/1550939/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1425578] Re: Window title sometimes gets rendered underneath controls
I meant 17.04 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1425578 Title: Window title sometimes gets rendered underneath controls To manage notifications about this bug go to: https://bugs.launchpad.net/hundredpapercuts/+bug/1425578/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1506704] Re: Thin font, nautilus doesn't integrate with Unity.
I haven't seen this problem for a long time now. Must be fixed already! -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1506704 Title: Thin font, nautilus doesn't integrate with Unity. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/1506704/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1425578] Re: Window title sometimes gets rendered underneath controls
Still present in 16.10. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1425578 Title: Window title sometimes gets rendered underneath controls To manage notifications about this bug go to: https://bugs.launchpad.net/hundredpapercuts/+bug/1425578/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1515771] Re: Occasional hiccup when opening nautilus
Since 16.04 (I think), where Files changed its behavior to show mounted drives in another window (a bad choice in my opinion, but whatever) this problem is not reproducible anymore. I marked as fixed thus. ** Changed in: nautilus (Ubuntu) Status: New => Fix Released -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1515771 Title: Occasional hiccup when opening nautilus To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1515771/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1375276] Re: Menus get in the way of maximization
Still present in 17.04 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1375276 Title: Menus get in the way of maximization To manage notifications about this bug go to: https://bugs.launchpad.net/unity/+bug/1375276/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1418494] Re: volume bar thinks mouse is pressed if released outside area
Cannot reproduce in 17.04. Must be fixed already. ** Changed in: ido (Ubuntu) Status: New => Fix Released -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1418494 Title: volume bar thinks mouse is pressed if released outside area To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/ido/+bug/1418494/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1506704] Re: Thin font, nautilus doesn't integrate with Unity.
@Adolfo, that was it! For some reason gnome-settings-daemon was uninstalled on my upgrade to 16.04, and on boot Ubuntu was complaining about an error with it. I installed it, re-logged in and everything is now correct. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1506704 Title: Thin font, nautilus doesn't integrate with Unity. To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gnome-settings-daemon/+bug/1506704/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1462267] [NEW] Desktop icon alignment `floor()`s instead of `round()`
Public bug reported: When dragging to arrange icons on the desktop, there is an invisible grid that the icons get snapped to. The x-axis has larger distance between grid lines and this bug is more visible in that direction. Steps to reproduce: 1. Take an icon 2. Drag a few pixels to left 3. Release icon Expected behavior: - You are more or less still in the same position, so the icon should go back to where it was before dragging Observed behavior: - The icon jumps a long way to the left You can recreate this with "one pixel up" as well, where the icon jumps up, although not as dramatically as with the x-axis. The behavior is the same regardless of the icon size. I am reporting this as a bug, because it is highly annoying to move an icon to more or less where you want, only to find it jumps back. I think the correct behavior would be for the icon to snap to the closest grid lines, rather than the one on its left/up. --- Note: this is not a fresh install. $ lsb_release -rd Description:Ubuntu 14.04.2 LTS Release:14.04 $ apt-cache policy nautilus nautilus: Installed: 1:3.10.1-0ubuntu9.8 Candidate: 1:3.10.1-0ubuntu9.8 Version table: *** 1:3.10.1-0ubuntu9.8 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty-updates/main i386 Packages 100 /var/lib/dpkg/status 1:3.10.1-0ubuntu8 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty/main i386 Packages Note sure if desktop is nautilus or unity: $ apt-cache policy unity unity: Installed: 7.2.5+14.04.20150521.1-0ubuntu1 Candidate: 7.2.5+14.04.20150521.1-0ubuntu1 Version table: *** 7.2.5+14.04.20150521.1-0ubuntu1 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty-updates/main i386 Packages 100 /var/lib/dpkg/status 7.2.2+14.04.20140714-0ubuntu1.1 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty-security/main i386 Packages 7.2.0+14.04.20140416-0ubuntu1 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty/main i386 Packages ProblemType: Bug DistroRelease: Ubuntu 14.04 Package: nautilus 1:3.10.1-0ubuntu9.8 Uname: Linux 3.16.7-rtai+ i686 NonfreeKernelModules: rtai_rtdm rtai_mbx rtai_fifos rtai_shm rtai_sem rtai_sched rtai_hal ApportVersion: 2.14.1-0ubuntu3.11 Architecture: i386 CurrentDesktop: Unity Date: Fri Jun 5 10:45:02 2015 ExecutablePath: /usr/bin/nautilus GsettingsChanges: b'org.gnome.nautilus.list-view' b'default-column-order' b"['name', 'size', 'type', 'date_modified', 'date_accessed', 'owner', 'group', 'permissions', 'mime_type', 'where']" InstallationDate: Installed on 2014-07-04 (335 days ago) InstallationMedia: Ubuntu 14.04 LTS "Trusty Tahr" - Release i386 (20140417) SourcePackage: nautilus UpgradeStatus: No upgrade log present (probably fresh install) ** Affects: nautilus (Ubuntu) Importance: Undecided Status: New ** Tags: trusty vivid -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1462267 Title: Desktop icon alignment `floor()`s instead of `round()` To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1462267/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1374383] Re: Clicking on "Files" opens new window, rather than switch to already open one
A reminder that this bug still exists! Can someone at least confirm it? Here are exact steps to reproduce reliably for me: 1. Close all "Files" windows 2. Attach a USB disk 3. Open a "Files" window by clicking on its icon in the left bar 4. Click on the USB drive. You should see your USB files now 5. Click on another open window on the left bar 6. Click back on the "Files" icon on the left bar Expected behavior: - The single "Files" window that is open (currently showing the USB driver) comes back up Observed behavior: - A new "Files" window is opened ** Also affects: nautilus Importance: Undecided Status: New -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1374383 Title: Clicking on "Files" opens new window, rather than switch to already open one To manage notifications about this bug go to: https://bugs.launchpad.net/nautilus/+bug/1374383/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1462267] Re: Desktop icon alignment `floor()`s instead of `round()`
There you go: https://bugzilla.gnome.org/show_bug.cgi?id=750446 ** Bug watch added: GNOME Bug Tracker #750446 https://bugzilla.gnome.org/show_bug.cgi?id=750446 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1462267 Title: Desktop icon alignment `floor()`s instead of `round()` To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1462267/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1424608] Re: double mouse cursor after custom cursor
Could you give a link to such a website? Also, what version of Ubuntu and Firefox? -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1424608 Title: double mouse cursor after custom cursor To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/firefox/+bug/1424608/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1424800] Re: crashed after copyed 1 file (as root)
Please try to be more descriptive. What was the command that you executed? I tried $ gksu cp some_files some_directory and nothing went wrong. Is that what you did? -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1424800 Title: crashed after copyed 1 file (as root) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gksu/+bug/1424800/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1425991] Re: Networks I have never connected to should be confined to the "More Networks" folder
I completely agree with what you are saying. If I wanted to manually select my network every time, I would also be frustrated with having unwanted networks in the way. However, I do see a problem. First of all, remember that most people have the network automatically get connected to, so when they do click on the network button, they often want to connect to a new network, for example on a laptop in an airport. With that in mind, there is a usability problem with your suggestion. The issue stems from the fact that people don't read anything. An average person has his network automatically get connected to, so he is likely not familiar with what's inside indicater-network. In fact this is the case for the majority of people. Now, imagine this person wants to connect to a new network. With your suggestion, he opens indicator-network and sees some text, and he is going to completely ignore them (such as "Connection Information", "VPN Connections" etc). >From experience, this person may even know that generally what those texts he just ignored do must have something to do with configuration, which is not what he wants at the moment. What he wants at the moment is to quickly be able to see the name of the new network he wants to connect to. But it's not there. Is the router on? Does my wireless card suck? Am I too far from the router? Do I hate Ubuntu? No, everything seems ok. So why don't I see my damn network? Oh, look there is something there that says "More Networks". WTF. In short, it's a nuisance to the majority of the people who, when they click on indicator-network, want to select a network they have never connected to (because the ones they always connect to, automatically do so, so there is no need to open indicator-network). --- That said, I certainly agree with you. However, I have a different solution. I'd suggest to change the layout of indicator-network from this: Bla Bla Known Network 1 Known Network 2 Unknown Network 1 Unknown Network 2 Unknown Network 3 Unknown Network 4 Unknown Network 5 More Bla Bla More Bla Bla Edit Connections to this: Bla Bla Known Network 1 Known Network 2 More Bla Bla More Bla Bla Edit Connections --- Unknown Network 1 Unknown Network 2 Unknown Network 3 Unknown Network 4 Unknown Network 5 This way, all networks are visible, so the network you are searching for is quickly findable. On the other hand, switching between known networks is not frustrating because the unknown ones are not close-by, or worse interleaved. Building on this idea, and reducing clutter, indicator-network could look like this: Bla Bla Known Network 1 Known Network 2 More Bla Bla More Bla Bla Edit Connections --- Unknown Network 1 Unknown Network 2 V Where `V` is a clickable "down arrow" that once clicked (or hovered on), would expand indicator-network to list all visible Unknown Networks (like the previous case). In this design, indicator-network should show only strong unknown networks (say, higher than >=3 bars), because they are the ones that the user is likely looking for. If none exist, it should show the top 1 or 2 strongest networks (for example 2 if the top 2 strongest are close in strength, and only 1 if it is much stronger than the second strongest). The weak networks, which are not likely what you are looking for, would be hidden under the down arrow. I belive the down arrow should be clear enough for the general user to show there is something more below, and it's not text, so it's easily understandable (because again, people don't read, so any text is of next to no use). What do you think? ** Changed in: indicator-network (Ubuntu) Status: New => Confirmed -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1425991 Title: Networks I have never connected to should be confined to the "More Networks" folder To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/indicator-network/+bug/1425991/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1424948] Re: Evince doesn't show "µ"
With evince 3.10.3 on Ubuntu 14.04 I can see the µ's alright. I will try with 14.10 later. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1424948 Title: Evince doesn't show "µ" To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/fontconfig/+bug/1424948/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1515771] [NEW] Occasional hiccup when opening nautilus
Public bug reported: Since the update to Ubuntu 14.04 a few days ago, (today is Nov 12, 2015), there are occasional hiccups when opening nautilus. That is, sometimes it takes tens of seconds for a new window to open and the nautilus windows that are already open don't respond either. I unfortunately don't have a repeatable way to reproduce this. A relevant piece of information is that I may have a nautilus window open at a `smb://` address. $ lsb_release -rd Description:Ubuntu 14.04.3 LTS Release:14.04 $ apt-cache policy nautilus nautilus: Installed: 1:3.10.1-0ubuntu9.11 Candidate: 1:3.10.1-0ubuntu9.11 Version table: *** 1:3.10.1-0ubuntu9.11 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty-updates/main amd64 Packages 100 /var/lib/dpkg/status 1:3.10.1-0ubuntu8 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages $ uname -a Linux 3.16.0-53-generic #72~14.04.1-Ubuntu SMP Fri Nov 6 18:17:23 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux ** Affects: nautilus (Ubuntu) Importance: Undecided Status: New -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1515771 Title: Occasional hiccup when opening nautilus To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1515771/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1515772] Re: Occasional hiccup when opening nautilus
I swear I didn't post this twice! Please close as duplicate of #1515771 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1515772 Title: Occasional hiccup when opening nautilus To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1515772/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1515772] [NEW] Occasional hiccup when opening nautilus
Public bug reported: Since the update to Ubuntu 14.04 a few days ago, (today is Nov 12, 2015), there are occasional hiccups when opening nautilus. That is, sometimes it takes tens of seconds for a new window to open and the nautilus windows that are already open don't respond either. I unfortunately don't have a repeatable way to reproduce this. A relevant piece of information is that I may have a nautilus window open at a `smb://` address. $ lsb_release -rd Description:Ubuntu 14.04.3 LTS Release:14.04 $ apt-cache policy nautilus nautilus: Installed: 1:3.10.1-0ubuntu9.11 Candidate: 1:3.10.1-0ubuntu9.11 Version table: *** 1:3.10.1-0ubuntu9.11 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty-updates/main amd64 Packages 100 /var/lib/dpkg/status 1:3.10.1-0ubuntu8 0 500 http://us.archive.ubuntu.com/ubuntu/ trusty/main amd64 Packages $ uname -a Linux 3.16.0-53-generic #72~14.04.1-Ubuntu SMP Fri Nov 6 18:17:23 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux ** Affects: nautilus (Ubuntu) Importance: Undecided Status: New -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1515772 Title: Occasional hiccup when opening nautilus To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/1515772/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1445239] [NEW] apt's lists/partial fills disk
Public bug reported: Since I upgraded to vivid, apt fills the disk space at /var/lib/apt/lists/partial. This was triggered by the automatic update every day, and as a solution I killed the processes named `http` spawned by apt, and removed the offending file from lists/partial manually. Until now, `apt-get update` and `apt-get upgrade` had been working. However, today, when I tried `apt-get update`, it got stuck in some file, continuously increasing its size when at 100%. I tried inspecting the file to see if there is a repetition of data, and there seems to be, as I have suspected. First of all, the file is it.archive.ubuntu .com_ubuntu_dists_vivid_main_binary-i386_Packages.bz2 this time, but I'm not sure if the previous time it was also this file. I used `tail` and `hexdump` to figure out whether a certain pattern shows up in the file. Here are some examples: $ tail -c 20 it.archive.ubuntu.com_ubuntu_dists_vivid_main_binary-i386_Packages.bz2 | hexdump | grep '\' c80 8d9e 63ad bbdf 40a7 3f3e 2b4a 9b95 aa5a 0010c80 8d9e 63ad bbdf 40a7 3f3e 2b4a 9b95 aa5a 0020c80 8d9e 63ad bbdf 40a7 3f3e 2b4a 9b95 aa5a 0030c80 8d9e 63ad bbdf 40a7 3f3e 2b4a 9b95 aa5a $ tail -c 20 it.archive.ubuntu.com_ubuntu_dists_vivid_main_binary-i386_Packages.bz2 | hexdump | grep '\<9b95\>' c80 8d9e 63ad bbdf 40a7 3f3e 2b4a 9b95 aa5a 0010c80 8d9e 63ad bbdf 40a7 3f3e 2b4a 9b95 aa5a 0020c80 8d9e 63ad bbdf 40a7 3f3e 2b4a 9b95 aa5a 0030c80 8d9e 63ad bbdf 40a7 3f3e 2b4a 9b95 aa5a $ tail -c 20 it.archive.ubuntu.com_ubuntu_dists_vivid_main_binary-i386_Packages.bz2 | hexdump | grep '\<1234\>' 0002630 fd50 b08d 7082 1234 85b3 a61e 2921 e0cd 0012630 fd50 b08d 7082 1234 85b3 a61e 2921 e0cd 0022630 fd50 b08d 7082 1234 85b3 a61e 2921 e0cd $ tail -c 20 it.archive.ubuntu.com_ubuntu_dists_vivid_main_binary-i386_Packages.bz2 | hexdump | grep '\<76a8\>' 000fa90 747d 76a8 078a 67b6 60d9 83f5 5ff3 787c 001fa90 747d 76a8 078a 67b6 60d9 83f5 5ff3 787c 002fa90 747d 76a8 078a 67b6 60d9 83f5 5ff3 787c As you can see, it seems that after the error, a certain pattern seems to repeat every exactly 0x1 bytes. I attached a copy of those 0x1 bytes that keep repeating (from an arbitrary offset). --- $ lsb_release -rd Description:Ubuntu Vivid Vervet (development branch) Release:15.04 $ apt-cache policy apt apt: Installed: 1.0.9.7ubuntu4 Candidate: 1.0.9.7ubuntu4 Version table: *** 1.0.9.7ubuntu4 0 500 http://it.archive.ubuntu.com/ubuntu/ vivid/main amd64 Packages 100 /var/lib/dpkg/status --- Do let me know if I can test further to see what could be wrong. I'll keep an eye next times the behavior repeat to make sure the file is the same / the pattern is the same and I'll report back. If you think the problem could be the Italy server, let me know and I'll try with a different one. ProblemType: Bug DistroRelease: Ubuntu 15.04 Package: apt 1.0.9.7ubuntu4 ProcVersionSignature: Ubuntu 3.19.0-13.13-generic 3.19.3 Uname: Linux 3.19.0-13-generic x86_64 NonfreeKernelModules: nvidia ApportVersion: 2.17-0ubuntu2 Architecture: amd64 CurrentDesktop: Unity Date: Fri Apr 17 00:04:05 2015 SourcePackage: apt UpgradeStatus: Upgraded to vivid on 2015-03-28 (19 days ago) ** Affects: apt (Ubuntu) Importance: Undecided Status: New ** Tags: amd64 apport-bug vivid ** Attachment added: "The 0x1 bytes that repeat indefinitely during apt's update" https://bugs.launchpad.net/bugs/1445239/+attachment/4377852/+files/apt-error-repeating-bytes -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1445239 Title: apt's lists/partial fills disk To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1445239/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1445239] Re: apt's lists/partial fills disk
This is the next day, and this time the offending file is: it.archive.ubuntu.com_ubuntu_dists_vivid_restricted_binary- amd64_Packages.bz2 The repeating data is still 0x1 bytes, but the contents are different (another copy attached). ** Attachment added: "The 0x1 bytes repeating data for another case of this error" https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1445239/+attachment/4378477/+files/apt-error-repeating-bytes-2 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1445239 Title: apt's lists/partial fills disk To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1445239/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1445239] Re: apt's lists/partial fills disk
I changed the server to US and the problem doesn't seem to occur any more. Perhaps this is an issue with the Italy server, but nevertheless it is strange that apt would continuously write to the disk due to an error on the server. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1445239 Title: apt's lists/partial fills disk To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/apt/+bug/1445239/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1445239] Re: apt's lists/partial fills disk
Hi Brian, Sorry for my lack of responsiveness, but I'd essentially been moving countries and I'm no longer in Italy to begin with. On another computer where I'm typing this response, this issue doesn't happen to begin with (Canada servers). Hopefully others would be able to test this out. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1445239 Title: apt's lists/partial fills disk To manage notifications about this bug go to: https://bugs.launchpad.net/apt/+bug/1445239/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1424800] Re: crashed after copied 1 file (as root)
** Package changed: gksu (Ubuntu) => gnome-commander (Ubuntu) -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1424800 Title: crashed after copied 1 file (as root) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gnome-commander/+bug/1424800/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1424948] Re: Evince doesn't show "µ"
Evince 3.16.1 on Ubuntu 15.10 still contains this bug. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1424948 Title: Evince doesn't show "µ" To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/fontconfig/+bug/1424948/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1374383] Re: Clicking on "Files" opens new window, rather than switch to already open one
... previously-unmounted *drives* ... -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1374383 Title: Clicking on "Files" opens new window, rather than switch to already open one To manage notifications about this bug go to: https://bugs.launchpad.net/nautilus/+bug/1374383/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1374383] Re: Clicking on "Files" opens new window, rather than switch to already open one
On Ubuntu 15.10, this still happens to me with previously-unmounted drivers. Steps to reproduce: - Close all instances of "Files" - Open a new "Files" window, and click to mount a windows partition (one that is not previously mounted) - Click on another open window, such as firefox - Click back on "Files" Expected behavior: - The "Files" window previously opened should be selected Observed behabior: - A new "Files" window is opened as if there were none previously. Alt- tab/Alt-backtick find the window correctly nevertheless. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1374383 Title: Clicking on "Files" opens new window, rather than switch to already open one To manage notifications about this bug go to: https://bugs.launchpad.net/nautilus/+bug/1374383/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1462267] Re: Desktop icon alignment `floor()`s instead of `round()`
Finally tracked it down myself. Here is the result: https://git.gnome.org/browse/nautilus/commit/?id=984438f ** Changed in: nautilus (Ubuntu) Status: Triaged => Fix Committed -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1462267 Title: Desktop icon alignment `floor()`s instead of `round()` To manage notifications about this bug go to: https://bugs.launchpad.net/nautilus/+bug/1462267/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1550939] [NEW] Tab on Index makes evince go crazy
Public bug reported: Steps to reproduce: - Open a pdf with Index - Click on a section in the Index - Press Tab Expected behavior: - I don't know. I accidentally clicked Tab instead of Alt+Tab Observed behavior: - Evince very quickly goes through the Index, section after section. I didn't find a way to stop the madness except closing evince and reopening the pdf. I tested this with the Vulkan specifications (https://www.khronos.org/registry/vulkan/specs/1.0/pdf/vkspec.pdf) and the C11 standard (http://www.open- std.org/jtc1/sc22/wg14/www/docs/n1570.pdf) ProblemType: Bug DistroRelease: Ubuntu 15.10 Package: evince 3.16.1-0ubuntu1 ProcVersionSignature: Ubuntu 4.2.0-30.36-generic 4.2.8-ckt3 Uname: Linux 4.2.0-30-generic x86_64 NonfreeKernelModules: nvidia ApportVersion: 2.19.1-0ubuntu5 Architecture: amd64 CurrentDesktop: Unity Date: Sun Feb 28 10:14:14 2016 ExecutablePath: /usr/bin/evince ProcEnviron: XDG_RUNTIME_DIR= SHELL=/bin/bash LANGUAGE=en_US PATH=(custom, no user) LANG=en_US.UTF-8 SourcePackage: evince UpgradeStatus: Upgraded to wily on 2015-10-22 (129 days ago) ** Affects: evince (Ubuntu) Importance: Undecided Status: New ** Tags: amd64 apport-bug wily -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1550939 Title: Tab on Index makes evince go crazy To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/evince/+bug/1550939/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1550939] Re: Tab on Index makes evince go crazy
What do you mean by current? I tried it on Ubuntu 14.04 (evince 3.10.3) and the bug is not present. With my Ubuntu 15.10 it _is_ though. Hopefully someone else would try it and we'd see if it is reproducible by others. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1550939 Title: Tab on Index makes evince go crazy To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/evince/+bug/1550939/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1161934] Re: Font bottm line artifact after middle of screen
This issue doesn't exist in 14.04. ** Changed in: gnome-terminal (Ubuntu) Status: New => Incomplete ** Changed in: gnome-terminal (Ubuntu) Status: Incomplete => Fix Released -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1161934 Title: Font bottm line artifact after middle of screen To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gnome-terminal/+bug/1161934/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1161941] Re: git-diff losing lines
This issue doesn't exist in Ubuntu 14.04, but I'm not sure if it still affects 12.04. ** Changed in: gnome-terminal (Ubuntu) Status: Confirmed => Fix Released -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1161941 Title: git-diff losing lines To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gnome-terminal/+bug/1161941/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1425952] Re: gedit
Did you perhaps use `sudo`? With graphical applications you need to use `gksudo`. See this question and its answers: http://askubuntu.com/q/376426/31099 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1425952 Title: gedit To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/xorg/+bug/1425952/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1425664] Re: grep(1) man page for GNU grep 2.16 should have PCRE but has PRCE
** Changed in: grep (Ubuntu) Status: New => Confirmed -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1425664 Title: grep(1) man page for GNU grep 2.16 should have PCRE but has PRCE To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/grep/+bug/1425664/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1425400] Re: I am using ubuntu13.10 version via vmware.but i can't update latest 14.10 version ..help me to upgrade ubuntu at my system , Thanks
Please explain what happens. What have you tried? What where the error outputs if any? Just saying it doesn't work cannot make anyone understand how to fix it. Also, from 13.10, which is not a long-term-support (LTS) release, you cannot immediately upgrade to 14.10 as far as I know. You need to upgrade to 14.04 first, and then from there to 14.10. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1425400 Title: I am using ubuntu13.10 version via vmware.but i can't update latest 14.10 version ..help me to upgrade ubuntu at my system ,Thanks To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/ubuntu-release-upgrader/+bug/1425400/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1425578] Re: Window title sometimes gets rendered underneath controls
I confirm that this happens also in 14.04. It doesn't work by clicking on the "Workspace Switcher", but by switching workspaces with keyboard (CTRL + ALT + Arrow Key), and to a workspace where there is nothing. ** Changed in: unity (Ubuntu) Status: New => Confirmed -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1425578 Title: Window title sometimes gets rendered underneath controls To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1425578/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1425104] Re: can't sign in for reviewing
Works fine for me. What is your Ubuntu version? -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1425104 Title: can't sign in for reviewing To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/software-center/+bug/1425104/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1412001] [NEW] alt+backtick doesn't focus on selected one of many windows previously clicked
Public bug reported: This a user experience issue and easily reproducible. Open multiple instances of your favorite application. Then click on Unity's icon for the app when you are already focused on one instance of it. This opens up a tile of the instances of the application which you can select from with the mouse. Now while that tile is shown, use ATL+backtick to select the window you want instead of clicking with the mouse. The result is that the tile window doesn't go away. If you use ALT+backtick and switch to ALT+tab (or start from ALT+tab from the start) and select another kind of application, then the tile goes away correctly and the other application comes in the front. I can verify that the window you would select with ALT+backtick does actually get to the top of the view and is marked as the "last visited instance", because if you switch to another application (using ALT+tab) and click back on the group of the original application you would see the one you previously selected with ALT+backtick. --- Here are step-by-step instructions. - Open a couple of terminals - Open firefox - Click on the terminal icon * This should open a tile showing the various terminal windows you have open * If that doesn't happen, click again (you were not on a terminal window before) - Use ALT+backtick to select one of the terminal windows Expected behavior: The tile of terminals goes away and the selected terminal is focused Observed behavior: The tile of terminals stays on the screen ProblemType: Bug DistroRelease: Ubuntu 14.10 Package: unity 7.3.1+14.10.20141016-0ubuntu1 ProcVersionSignature: Ubuntu 3.16.0-29.39-generic 3.16.7-ckt2 Uname: Linux 3.16.0-29-generic x86_64 NonfreeKernelModules: nvidia ApportVersion: 2.14.7-0ubuntu8 Architecture: amd64 CompizPlugins: [core,bailer,detection,composite,opengl,decor,place,grid,mousepoll,compiztoolbox,vpswitch,regex,gnomecompat,imgpng,snap,move,unitymtgrabhandles,resize,session,wall,animation,workarounds,fade,expo,scale,ezoom,unityshell] CurrentDesktop: Unity Date: Sat Jan 17 16:32:05 2015 ExecutablePath: /usr/bin/compiz ProcEnviron: PATH=(custom, no user) LANGUAGE=en_US XDG_RUNTIME_DIR= LANG=en_US.UTF-8 SHELL=/bin/bash SourcePackage: unity UpgradeStatus: Upgraded to utopic on 2014-11-01 (76 days ago) ** Affects: unity (Ubuntu) Importance: Undecided Status: New ** Tags: unity utopic -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1412001 Title: alt+backtick doesn't focus on selected one of many windows previously clicked To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1412001/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1397258] Re: nvidia-331 331.89-0ubuntu5: nvidia-331 kernel module failed to build
Looking at the DKMS log automatically appended, it seems to have nothing to do with the one I linked. It seems that right after compiling one of the files, the file has gone missing. I have plenty of disk space on that partition. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1397258 Title: nvidia-331 331.89-0ubuntu5: nvidia-331 kernel module failed to build To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-331/+bug/1397258/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1397258] [NEW] nvidia-331 331.89-0ubuntu5: nvidia-331 kernel module failed to build
Public bug reported: Ubuntu asked me to report this and I can't seem to find the related DKMS log file none of the /var/lib/dkms log files contain an "error:". Bug #1366686 seems like this could be a duplicate of, but I can't tell at this moment. In particular, this comment (https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics- drivers-331/+bug/1366686/comments/3) may have some merit. I had been using the most up-to-date driver downloaded from the website of Nvidia for a couple of years until I finally got fed up with recompiling it on every minor kernel update and I switched to the Ubuntu-managed package about 4~5 months ago. ProblemType: Package DistroRelease: Ubuntu 14.10 Package: nvidia-331 331.89-0ubuntu5 ProcVersionSignature: Ubuntu 3.16.0-25.33-generic 3.16.7 Uname: Linux 3.16.0-25-generic x86_64 NonfreeKernelModules: nvidia ApportVersion: 2.14.7-0ubuntu8 Architecture: amd64 DKMSKernelVersion: 3.16.0-25-generic Date: Thu Nov 27 23:05:08 2014 PackageVersion: 331.89-0ubuntu5 SourcePackage: nvidia-graphics-drivers-331 Title: nvidia-331 331.89-0ubuntu5: nvidia-331 kernel module failed to build UpgradeStatus: Upgraded to utopic on 2014-11-01 (26 days ago) ** Affects: nvidia-graphics-drivers-331 (Ubuntu) Importance: Undecided Status: New ** Tags: amd64 apport-package utopic -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1397258 Title: nvidia-331 331.89-0ubuntu5: nvidia-331 kernel module failed to build To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-331/+bug/1397258/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1412001] Re: alt+backtick doesn't focus on selected one of many windows previously clicked
Exactly. It happened for me when I clicked on the terminal icons by mistake, while at (almost) the same time I used ALT+` for switching windows the way I usually do. I agree that simply closing the spread (which I called "tile" in the OP) is the solution. This is what happens when ALT+TAB is used to switch to whole other program, and it was as I had expected. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1412001 Title: alt+backtick doesn't focus on selected one of many windows previously clicked To manage notifications about this bug go to: https://bugs.launchpad.net/unity/+bug/1412001/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1418494] [NEW] volume bar thinks mouse is pressed if released outside area
Public bug reported: Steps to reproduce: - Open indicator-sound (click on volume icon) - Press the mouse in volume slide (don't release) - Move mouse to outside the slide widget (not on another button, or outside indicator-sound) - Release the mouse - Optional: click on the indicator-sound window where there is no button. Nothing happens, but you can click as much as you want. (Even if there is a button, the behavior is the same) - Move mouse back to volume slide Expected behavior: volume slider doesn't change with mouse motion Observed behavior: volume slider follows mouse motion as if mouse is still pressed --- I am reporting this bug here because particularly with indicator-sound, the volume slider is quite thin, so if you are not precise, you could release the mouse easily outside the area. What's more, sometimes you want to lower the volume because some music you don't want starts playing, so this could happen: - Lower the volume with a focus on changing the music - Moving fast, the mouse button release happens outside the slider area - Change the music - Move back to the slider with the intention of raising volume - BAM, the slider still thinks the mouse is pressed, and the volume may potentially be raised to maximum depending on where your mouse is. --- If I may suggest (based on guessing completely), the slider widget needs to check the state of the mouse button on mouse motion event. ProblemType: Bug DistroRelease: Ubuntu 14.10 Package: indicator-sound 12.10.2+14.10.20141010-0ubuntu1 ProcVersionSignature: Ubuntu 3.16.0-30.40-generic 3.16.7-ckt3 Uname: Linux 3.16.0-30-generic x86_64 NonfreeKernelModules: nvidia ApportVersion: 2.14.7-0ubuntu8.1 Architecture: amd64 CurrentDesktop: Unity Date: Thu Feb 5 12:02:30 2015 SourcePackage: indicator-sound UpgradeStatus: Upgraded to utopic on 2014-11-01 (95 days ago) ** Affects: indicator-sound (Ubuntu) Importance: Undecided Status: New ** Tags: amd64 apport-bug utopic -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1418494 Title: volume bar thinks mouse is pressed if released outside area To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/indicator-sound/+bug/1418494/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1420514] [NEW] scrollbar sees reduced mouse movement
Public bug reported: I am not using overlay scrollbar, so just normal scrollbars. Steps to reproduce the issue: - Open a program with large scrollable content, where handle is small (for example, a long pdf in evince, or even the sound volume scrollbar of indicater-sound - It doesn't happen with Firefox) - Take the scrollbar (left mouse press on scrollbar handle) - **Very slowly** start moving the mouse for a second or two - Move the mouse normally Expected behavior: The scrollbar handle moves with the mouse with the same speed Observed behavior: The scrollbar handle moves with greatly reduced speed compared to the movement of the mouse ProblemType: Bug DistroRelease: Ubuntu 14.10 Package: unity 7.3.1+14.10.20141016-0ubuntu1 ProcVersionSignature: Ubuntu 3.16.0-30.40-generic 3.16.7-ckt3 Uname: Linux 3.16.0-30-generic x86_64 NonfreeKernelModules: nvidia ApportVersion: 2.14.7-0ubuntu8.1 Architecture: amd64 CompizPlugins: [core,bailer,detection,composite,opengl,decor,place,grid,mousepoll,compiztoolbox,vpswitch,regex,gnomecompat,imgpng,snap,move,unitymtgrabhandles,resize,session,wall,animation,workarounds,fade,expo,scale,ezoom,unityshell] CurrentDesktop: Unity Date: Tue Feb 10 21:40:21 2015 ExecutablePath: /usr/bin/compiz ProcEnviron: PATH=(custom, no user) LANGUAGE=en_US XDG_RUNTIME_DIR= LANG=en_US.UTF-8 SHELL=/bin/bash SourcePackage: unity UpgradeStatus: Upgraded to utopic on 2014-11-01 (100 days ago) ** Affects: unity (Ubuntu) Importance: Undecided Status: New ** Tags: amd64 apport-bug utopic -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1420514 Title: scrollbar sees reduced mouse movement To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1420514/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1373874] [NEW] Malformed expression after bin/oct/dec/hex conversion
Public bug reported: Steps to produce error: 1. Go to _Programming_ mode, 2. Write whatever number and press enter, 3. Change representation (for example, from Decimal to Hexadecimal, or any other), - You get the number in a new representation (in bold), 4. Write an operation (for example *3) You would see: N*3 where N is bold. Pressing enter, results in "Malformed expression". If you delete the *3 and press enter, you still get the same message. If instead of step 4, you simply press enter, you also get this message. $ lsb_release -rd Description:Ubuntu 14.04.1 LTS Release:14.04 $ apt-cache policy gcalctool gcalctool: Installed: 1:3.10.2-0ubuntu1.1 Candidate: 1:3.10.2-0ubuntu1.1 Version table: *** 1:3.10.2-0ubuntu1.1 0 500 http://it.archive.ubuntu.com/ubuntu/ trusty-updates/universe i386 Packages 100 /var/lib/dpkg/status 1:3.10.2-0ubuntu1 0 500 http://it.archive.ubuntu.com/ubuntu/ trusty/universe i386 Packages ** Affects: gnome-calculator (Ubuntu) Importance: Undecided Status: New ** Summary changed: - Malformed expression when after bin/oct/dec/hex conversion + Malformed expression after bin/oct/dec/hex conversion -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1373874 Title: Malformed expression after bin/oct/dec/hex conversion To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gnome-calculator/+bug/1373874/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1374383] [NEW] Clicking on "Files" opens new window, rather than switch to already open one
Public bug reported: Although I can currently recreate the bug with "Files" (Nautilius, is it?), I remember having seen such a thing before with other windows, although only with previous Ubuntu versions. The bug is as follows. In the current desktop, I open Files (and I see the arrows on the left saying how many of such windows are open). It only happens if "Files" is open once. I click on a partition that is not automatically mounted on startup (e.g. a Windows partition) (I haven't tested with one that automatically mounts but is not a part of /), regardless of whether it was previously mounted or not. When I switch away from that window (e.g. click on Firefox), then I click back on its icon, I get a new "Files" window, rather than the old one I had open, which was viewing the windows partition. If you can't replicate this, try CTRL+L (leave the cursor on the path box) on that window and then switch away and back. Note that with ALT+TAB and ALT+` I can see the window alright. Also, if there is more than one "Files" window, whether all looking at windows partitions or not, there is no problem, even if the first such window was the last to be viewed, clicking on the "Files" icon correctly returns to that window. In summary, when "Files" is opened once and viewing a windows partition, clicking on a different application and back on its icon opens a new "Files" window rather than switch to the old one. ProblemType: Bug DistroRelease: Ubuntu 14.04 Package: unity 7.2.2+14.04.20140714-0ubuntu1.1 ProcVersionSignature: Ubuntu 3.13.0-36.63-generic 3.13.11.6 Uname: Linux 3.13.0-36-generic i686 ApportVersion: 2.14.1-0ubuntu3.4 Architecture: i386 CompizPlugins: [core,bailer,detection,composite,opengl,decor,mousepoll,vpswitch,regex,animation,snap,expo,move,compiztoolbox,place,grid,imgpng,gnomecompat,wall,ezoom,workarounds,staticswitcher,resize,fade,unitymtgrabhandles,scale,session,unityshell] CurrentDesktop: Unity Date: Fri Sep 26 12:01:15 2014 ExecutablePath: /usr/bin/compiz InstallationDate: Installed on 2014-07-04 (83 days ago) InstallationMedia: Ubuntu 14.04 LTS "Trusty Tahr" - Release i386 (20140417) SourcePackage: unity UpgradeStatus: No upgrade log present (probably fresh install) --- P.S. regarding info collected by `ubuntu-bug`: - This is NOT a fresh install. ** Affects: unity (Ubuntu) Importance: Undecided Status: New ** Tags: apport-bug i386 trusty ** Description changed: Although I can currently recreate the bug with "Files" (Nautilius, is it?), I remember having seen such a thing before with other windows, although only with previous Ubuntu versions. The bug is as follows. In the current desktop, I open Files (and I see the arrows on the left saying how many of such windows are open). It only happens if "Files" is open once. I click on a partition that is not automatically mounted on startup (e.g. a Windows partition) (I haven't tested with one that automatically mounts but is not a part of /), regardless of whether it was previously mounted or not. When I switch away from that window (e.g. click on Firefox), then I click back on its icon, I get a new "Files" window, rather than the old one I had open, which was viewing the windows partition. If you can't replicate this, try CTRL+L (leave the cursor on the path box) on that window and then switch away and back. Note that with ALT+TAB and ALT+` I can see the window alright. Also, if there is more than one "Files" window, whether all looking at windows partitions or not, there is no problem, even if the first such window was the last to be viewed, clicking on the "Files" icon correctly returns to that window. In summary, when "Files" is opened once and viewing a windows partition, clicking on a different application and back on its icon opens a new "Files" window rather than switch to the old one. ProblemType: Bug DistroRelease: Ubuntu 14.04 Package: unity 7.2.2+14.04.20140714-0ubuntu1.1 ProcVersionSignature: Ubuntu 3.13.0-36.63-generic 3.13.11.6 Uname: Linux 3.13.0-36-generic i686 ApportVersion: 2.14.1-0ubuntu3.4 Architecture: i386 CompizPlugins: [core,bailer,detection,composite,opengl,decor,mousepoll,vpswitch,regex,animation,snap,expo,move,compiztoolbox,place,grid,imgpng,gnomecompat,wall,ezoom,workarounds,staticswitcher,resize,fade,unitymtgrabhandles,scale,session,unityshell] CurrentDesktop: Unity Date: Fri Sep 26 12:01:15 2014 ExecutablePath: /usr/bin/compiz InstallationDate: Installed on 2014-07-04 (83 days ago) InstallationMedia: Ubuntu 14.04 LTS "Trusty Tahr" - Release i386 (20140417) SourcePackage: unity UpgradeStatus: No upgrade log present (probably fresh install) + + --- + + P.S. regarding info collected by `ubuntu-bug`: + + - This is NOT a fresh install. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1374383 Title: Clicking on "
[Bug 1375276] [NEW] Menus get in the way of maximization
Public bug reported: Steps to reproduce: 1. Set the following configuration: In "Appearance" settings, under "Behavior" tab, where it asks "Show the menus for a window", select "In the window's title bar". 2. Open any unmaximized window with menus, say gvim. The menus of the application are hidden on the top bar which looks nice and clean. Now try to absent-mindedly maximize the window by double clicking the top bar. 3. Usually unamximized windows are small and each program usually has at least 5~6 menu items. Therefore more than half of the top bar is occupied by the menu items. Double clicking on that range opens and immediately closes a menu item. I doubt this is ever anyone's intention! On the contrary, if I double click on the top bar I expect the window to maximize. If suddenly menu items appear under my cursor, I wouldn't care; I would still expect double click to maximize the window. Suggested solution: If a menu item is double clicked, (un)maximize the window. This works also when the menus are in the menu bar, which is the desired behavior for unmaximizing if the window is already maximized. The side effect is that if the menus are in the window's title bar and the menu bar is double-clicked, the window gets maximized. If this is not a desired side effect, it can be guarded against. ProblemType: Bug DistroRelease: Ubuntu 14.04 Package: unity 7.2.2+14.04.20140714-0ubuntu1.1 ProcVersionSignature: Ubuntu 3.13.0-36.63-generic 3.13.11.6 Uname: Linux 3.13.0-36-generic i686 ApportVersion: 2.14.1-0ubuntu3.4 Architecture: i386 CompizPlugins: [core,bailer,detection,composite,opengl,decor,mousepoll,vpswitch,regex,animation,snap,expo,move,compiztoolbox,place,grid,imgpng,gnomecompat,wall,ezoom,workarounds,staticswitcher,resize,fade,unitymtgrabhandles,scale,session,unityshell] CurrentDesktop: Unity Date: Mon Sep 29 15:52:08 2014 ExecutablePath: /usr/bin/compiz InstallationDate: Installed on 2014-07-04 (87 days ago) InstallationMedia: Ubuntu 14.04 LTS "Trusty Tahr" - Release i386 (20140417) SourcePackage: unity UpgradeStatus: No upgrade log present (probably fresh install) P.S. This is NOT a fresh install. ** Affects: unity (Ubuntu) Importance: Undecided Status: New ** Tags: apport-bug i386 trusty ** Description changed: Steps to reproduce: 1. Set the following configuration: In "Appearance" settings, under "Behavior" tab, where it asks "Show the menus for a window", select "In the window's title bar". 2. Open any unmaximized window with menus, say gvim. The menus of the application are hidden on the top bar which looks nice and clean. Now try to absent-mindedly maximize the window by double clicking the top bar. 3. Usually unamximized windows are small and each program usually has at least 5~6 menu items. Therefore more than half of the top bar is occupied by the menu items. Double clicking on that range opens and immediately closes a menu item. I doubt this is ever anyone's intention! On the contrary, if I double click on the top bar I expect the window to maximize. If suddenly menu items appear under my cursor, I wouldn't care; I would still expect double click to maximize the window. Suggested solution: If a menu item is double clicked, (un)maximize the window. This works also when the menus are in the menu bar, which is the desired behavior for unmaximizing if the window is already maximized. The side effect is that if the menus are in the window's title bar and the menu bar is double-clicked, the window gets maximized. If this is not a desired side effect, it can be guarded against. ProblemType: Bug DistroRelease: Ubuntu 14.04 Package: unity 7.2.2+14.04.20140714-0ubuntu1.1 ProcVersionSignature: Ubuntu 3.13.0-36.63-generic 3.13.11.6 Uname: Linux 3.13.0-36-generic i686 ApportVersion: 2.14.1-0ubuntu3.4 Architecture: i386 CompizPlugins: [core,bailer,detection,composite,opengl,decor,mousepoll,vpswitch,regex,animation,snap,expo,move,compiztoolbox,place,grid,imgpng,gnomecompat,wall,ezoom,workarounds,staticswitcher,resize,fade,unitymtgrabhandles,scale,session,unityshell] CurrentDesktop: Unity Date: Mon Sep 29 15:52:08 2014 ExecutablePath: /usr/bin/compiz InstallationDate: Installed on 2014-07-04 (87 days ago) InstallationMedia: Ubuntu 14.04 LTS "Trusty Tahr" - Release i386 (20140417) SourcePackage: unity UpgradeStatus: No upgrade log present (probably fresh install) + + P.S. This is NOT a fresh install. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1375276 Title: Menus get in the way of maximization To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/unity/+bug/1375276/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1391510] [NEW] sscanf extremely slow on large strings
Public bug reported:
This is a bug on glib, and I'm not sure if it applies upstream too
(although I suspect it does).
The bug is that `sscanf` (`vsscanf` actually) is very slow on large
strings. It _seems_ like `vsscanf` is first trying to find the end of
the string (with `rawmemchr`), which is expensive given that most of the
string is not going to be read.
Here's a test code:
#include
#include
#include
#include
#define N 5
static int _debug_helper(const char *src, int *a, int *n)
{
#if 1
return sscanf(src, "%d%n", a, n);
#else
int i = 0;
int r = 0;
int rseen = 0;
while (isspace(src[i])) ++i;
while (isdigit(src[i])) { rseen = 1; r = r * 10 + src[i++] - '0'; }
*a = r;
*n = i;
return rseen;
#endif
}
int main()
{
int i;
int a;
int n;
int so_far = 0;
char *big_string = malloc(N * 4 + 1);
for (i = 0; i < N; ++i)
strcpy(big_string + i * 4, "123 ");
big_string[N * 4] = '\0';
while (1)
{
if (_debug_helper(big_string + so_far, &a, &n) != 1)
break;
so_far += n;
}
return 0;
}
Compile with: gcc -Wall -g -O0 main.c
Running this code with `N = 5` and using `sscanf`, I get:
$ time ./a.out
real0m1.602s
user 0m1.596s
sys 0m0.000s
Running it with `N = 5` and using the substitute code, I get:
$ time ./a.out
real0m0.004s
user 0m0.000s
sys 0m0.000s
Which is considrably smaller. Note that this shows that the part with
`malloc` and initialization take very small time. Running callgrind
shows that almost all of the time when using `sscanf` is spent in
`rawmemchr`. Indeed, using gdb and randomly hitting CTRL+C, you always
end up with a stack trace like this:
#0 __rawmemchr_ia32 () at ../sysdeps/i386/i686/multiarch/../../rawmemchr.S:167
#1 0xb7e78a06 in _IO_str_init_static_internal () at strops.c:44
#2 0xb7e5c857 in __GI___isoc99_vsscanf () at isoc99_vsscanf.c:41
#3 0xb7e5c7cf in __isoc99_sscanf () at isoc99_sscanf.c:31
#4 0x08048494 in _debug_helper () at main.c:11
#5 0x08048517 in main () at main.c:41
This means that `rawmemchr` is slowing down `sscanf` by an unnecessary
degree. To further prove this point (and to confirm my guess that
`rawmemchr` is reading the whole string), here are a couple more tests:
With `N = 25000` and using `sscanf`:
$ time ./a.out
real0m0.407s
user 0m0.404s
sys 0m0.000s
With `N = 12500` and using `sscanf`:
$ time ./a.out
real0m0.106s
user 0m0.104s
sys 0m0.000s
This clearly shows an O(N^2) behavior. The main loop of the program is
`O(N)`, which means `sscanf` is running at `O(N)`. For large `N`, this
is significant. On the other hand, the actual behavior of `sscanf`
should be to read from the string according to the format string and no
more, which in this case (using `%d` and "123" as values) is of constant
time.
---
Note: I'm on Ubuntu 14.04, using gcc 4.8.2. Most importantly, this is
glibc 2.19.
** Affects: glibc (Ubuntu)
Importance: Undecided
Status: New
** Tags: performance
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1391510
Title:
sscanf extremely slow on large strings
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/1391510/+subscriptions
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1391510] Re: sscanf extremely slow on large strings
** Description changed:
This is a bug on glib, and I'm not sure if it applies upstream too
(although I suspect it does).
The bug is that `sscanf` (`vsscanf` actually) is very slow on large
strings. It _seems_ like `vsscanf` is first trying to find the end of
the string (with `rawmemchr`), which is expensive given that most of the
string is not going to be read.
Here's a test code:
#include
#include
#include
#include
+ #include
#define N 5
static int _debug_helper(const char *src, int *a, int *n)
{
#if 1
- return sscanf(src, "%d%n", a, n);
+ return sscanf(src, "%d%n", a, n);
#else
- int i = 0;
- int r = 0;
- int rseen = 0;
+ char *end;
- while (isspace(src[i])) ++i;
- while (isdigit(src[i])) { rseen = 1; r = r * 10 + src[i++] - '0'; }
+ errno = 0;
+ *a = strtol(src, &end, 10);
+ *n = end - src;
- *a = r;
- *n = i;
-
- return rseen;
+ return errno == 0 && *n > 0;
#endif
}
int main()
{
- int i;
- int a;
- int n;
- int so_far = 0;
+ int i;
+ int a;
+ int n;
+ int so_far = 0;
- char *big_string = malloc(N * 4 + 1);
- for (i = 0; i < N; ++i)
- strcpy(big_string + i * 4, "123 ");
- big_string[N * 4] = '\0';
+ char *big_string = malloc(N * 4 + 1);
+ for (i = 0; i < N; ++i)
+ strcpy(big_string + i * 4, "123 ");
+ big_string[N * 4] = '\0';
- while (1)
- {
- if (_debug_helper(big_string + so_far, &a, &n) != 1)
- break;
- so_far += n;
- }
+ while (1)
+ {
+ if (_debug_helper(big_string + so_far, &a, &n) != 1)
+ break;
+ so_far += n;
+ }
- return 0;
+ return 0;
}
Compile with: gcc -Wall -g -O0 main.c
Running this code with `N = 5` and using `sscanf`, I get:
$ time ./a.out
real0m1.602s
- user 0m1.596s
+ user0m1.596s
sys 0m0.000s
Running it with `N = 5` and using the substitute code, I get:
$ time ./a.out
- real0m0.004s
- user 0m0.000s
+ real0m0.002s
+ user0m0.000s
sys 0m0.000s
Which is considrably smaller. Note that this shows that the part with
`malloc` and initialization take very small time. Running callgrind
shows that almost all of the time when using `sscanf` is spent in
`rawmemchr`. Indeed, using gdb and randomly hitting CTRL+C, you always
end up with a stack trace like this:
#0 __rawmemchr_ia32 () at
../sysdeps/i386/i686/multiarch/../../rawmemchr.S:167
#1 0xb7e78a06 in _IO_str_init_static_internal () at strops.c:44
#2 0xb7e5c857 in __GI___isoc99_vsscanf () at isoc99_vsscanf.c:41
#3 0xb7e5c7cf in __isoc99_sscanf () at isoc99_sscanf.c:31
#4 0x08048494 in _debug_helper () at main.c:11
#5 0x08048517 in main () at main.c:41
This means that `rawmemchr` is slowing down `sscanf` by an unnecessary
degree. To further prove this point (and to confirm my guess that
`rawmemchr` is reading the whole string), here are a couple more tests:
With `N = 25000` and using `sscanf`:
$ time ./a.out
real0m0.407s
- user 0m0.404s
+ user0m0.404s
sys 0m0.000s
With `N = 12500` and using `sscanf`:
$ time ./a.out
real0m0.106s
- user 0m0.104s
+ user0m0.104s
sys 0m0.000s
This clearly shows an O(N^2) behavior. The main loop of the program is
`O(N)`, which means `sscanf` is running at `O(N)`. For large `N`, this
is significant. On the other hand, the actual behavior of `sscanf`
should be to read from the string according to the format string and no
more, which in this case (using `%d` and "123" as values) is of constant
time.
---
Note: I'm on Ubuntu 14.04, using gcc 4.8.2. Most importantly, this is
glibc 2.19.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1391510
Title:
sscanf extremely slow on large strings
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/glibc/+bug/1391510/+subscriptions
--
ubuntu-bugs mailing list
ubuntu-bugs@lists.ubuntu.com
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1391510] Re: sscanf extremely slow on large strings
** Description changed:
This is a bug on glib, and I'm not sure if it applies upstream too
(although I suspect it does).
The bug is that `sscanf` (`vsscanf` actually) is very slow on large
strings. It _seems_ like `vsscanf` is first trying to find the end of
the string (with `rawmemchr`), which is expensive given that most of the
string is not going to be read.
Here's a test code:
#include
#include
#include
#include
#include
#define N 5
static int _debug_helper(const char *src, int *a, int *n)
{
#if 1
return sscanf(src, "%d%n", a, n);
#else
- char *end;
+ char *end;
- errno = 0;
- *a = strtol(src, &end, 10);
- *n = end - src;
+ errno = 0;
+ *a = strtol(src, &end, 10);
+ *n = end - src;
- return errno == 0 && *n > 0;
+ return errno == 0 && *n > 0;
#endif
}
int main()
{
int i;
int a;
int n;
int so_far = 0;
char *big_string = malloc(N * 4 + 1);
for (i = 0; i < N; ++i)
strcpy(big_string + i * 4, "123 ");
big_string[N * 4] = '\0';
while (1)
{
if (_debug_helper(big_string + so_far, &a, &n) != 1)
break;
so_far += n;
}
return 0;
}
Compile with: gcc -Wall -g -O0 main.c
Running this code with `N = 5` and using `sscanf`, I get:
$ time ./a.out
real0m1.602s
user0m1.596s
sys 0m0.000s
Running it with `N = 5` and using the substitute code, I get:
$ time ./a.out
real0m0.002s
user0m0.000s
sys 0m0.000s
Which is considrably smaller. Note that this shows that the part with
`malloc` and initialization take very small time. Running callgrind
shows that almost all of the time when using `sscanf` is spent in
`rawmemchr`. Indeed, using gdb and randomly hitting CTRL+C, you always
end up with a stack trace like this:
#0 __rawmemchr_ia32 () at
../sysdeps/i386/i686/multiarch/../../rawmemchr.S:167
#1 0xb7e78a06 in _IO_str_init_static_internal () at strops.c:44
#2 0xb7e5c857 in __GI___isoc99_vsscanf () at isoc99_vsscanf.c:41
#3 0xb7e5c7cf in __isoc99_sscanf () at isoc99_sscanf.c:31
#4 0x08048494 in _debug_helper () at main.c:11
#5 0x08048517 in main () at main.c:41
This means that `rawmemchr` is slowing down `sscanf` by an unnecessary
degree. To further prove this point (and to confirm my guess that
`rawmemchr` is reading the whole string), here are a couple more tests:
With `N = 25000` and using `sscanf`:
$ time ./a.out
real0m0.407s
user0m0.404s
sys 0m0.000s
With `N = 12500` and using `sscanf`:
$ time ./a.out
real0m0.106s
user0m0.104s
sys 0m0.000s
- This clearly shows an O(N^2) behavior. The main loop of the program is
+ This clearly shows an `O(N^2)` behavior. The main loop of the program is
`O(N)`, which means `sscanf` is running at `O(N)`. For large `N`, this
is significant. On the other hand, the actual behavior of `sscanf`
should be to read from the string according to the format string and no
more, which in this case (using `%d` and "123" as values) is of constant
time.
---
Note: I'm on Ubuntu 14.04, using gcc 4.8.2. Most importantly, this is
glibc 2.19.
** Description changed:
This is a bug on glib, and I'm not sure if it applies upstream too
- (although I suspect it does).
+ (although I suspect it does). Upstream bug link:
+
+ https://sourceware.org/bugzilla/show_bug.cgi?id=17577
The bug is that `sscanf` (`vsscanf` actually) is very slow on large
strings. It _seems_ like `vsscanf` is first trying to find the end of
the string (with `rawmemchr`), which is expensive given that most of the
string is not going to be read.
Here's a test code:
#include
#include
#include
#include
#include
#define N 5
static int _debug_helper(const char *src, int *a, int *n)
{
#if 1
return sscanf(src, "%d%n", a, n);
#else
char *end;
errno = 0;
*a = strtol(src, &end, 10);
*n = end - src;
return errno == 0 && *n > 0;
#endif
}
int main()
{
int i;
int a;
int n;
int so_far = 0;
char *big_string = malloc(N * 4 + 1);
for (i = 0; i < N; ++i)
strcpy(big_string + i * 4, "123 ");
big_string[N * 4] = '\0';
while (1)
{
if (_debug_helper(big_string + so_far, &a, &n) != 1)
break;
so_far += n;
}
return 0;
}
Compile with: gcc -Wall -g -O0 main.c
Running this code with `N = 5` and using `sscanf`, I get:
$ time ./a.out
real0m1.602s
user0m1.596s
sys 0m0.000s
Running it with `N = 5` and using the substitute code, I get:
$ time ./a.out
real0m0.002s
user0m0.000s
sys 0m0.000s
Which is considrably smaller. Note that this shows that the part with
`malloc` and initialization take very small time. Running callgrind
shows that almost all of the time when using `sscanf` is spent in
`rawmemchr`
[Bug 1161934] [NEW] Font bottm line artifact after middle of screen
Public bug reported: I'm using Ubuntu 12.10 (updated to current date of writing this report) under 1440x900 resolution. I have indicator-appmenu removed, so in my maximized terminal I see the menus below the main bar on the top. I use the font "Ubuntu Mono" with size 12, other parameters set to default. The cursor shape is Block. The bug is that the last line of the terminal seems to be just a little too low, somehow printing characters below a defined window or something. The symptom of the bug is that in a maximized terminal, once the cursor passes the middle column, it will leave artifacts behind on the last line (one pixel width?) which doesn't get cleared up. I cannot take a screenshot of this, because any refresh of the screen (such as minimize and back, or the flash from Ubuntu's screenshot program) makes the artifacts go away. However, I took a snapshot with a camera and attached the image. If I have a different terminal window size, such as when indicator- appmenu is installed (so there is some more height available for the terminal), this does not happen. Neither does it happen in an unmaximized window. The reason being that unmaximized windows make the height a multiple of the line height gnome-terminal likes, and in the case of other window sizes because the calculated "last line" of the terminal is luckily placed a bit higher than the bottom edge. My suspicion is that when calculating how many lines fit in the given height of the terminal, there is an off-by-one-pixel error. Additional information: $ lsb_release -rd: Description:Ubuntu 12.04.2 LTS Release:12.04 $ apt-cache policy gnome-terminal gnome-terminal: Installed: 3.4.1.1-0ubuntu1 Candidate: 3.4.1.1-0ubuntu1 Version table: *** 3.4.1.1-0ubuntu1 0 500 http://it.archive.ubuntu.com/ubuntu/ precise/main i386 Packages 100 /var/lib/dpkg/status ** Affects: gnome-terminal (Ubuntu) Importance: Undecided Status: New ** Attachment added: "terminal-cursor-artifact.jpg" https://bugs.launchpad.net/bugs/1161934/+attachment/3606476/+files/terminal-cursor-artifact.jpg -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1161934 Title: Font bottm line artifact after middle of screen To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gnome-terminal/+bug/1161934/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
[Bug 1161941] [NEW] git-diff losing lines
Public bug reported: I'm not sure if this is indeed a terminal bug, a `git` bug, a `less` bug or else. But since my terminal is having other issues: https://bugs.launchpad.net/ubuntu/+source/gnome-terminal/+bug/1161934 , I thought it maybe related. The problem happens when I issue a `git diff` on my repository. The diff command has no configs and is using the defaults. The output is colored. The issue only happens in maximized terminal windows and only the first time I issue `git diff`. The issue happens with different terminal fonts and sizes. The effect is that once I use page down to reach the bottom of the diff, and then use page up to go back, some lines seemingly randomly (but always the same lines) disappear. If I diff again, this doesn't happen anymore. The issue is repeatable. If I open another terminal, maximize it and do `git diff`, page down to the bottom and then page up again I see the exact same lines missing. Attached, you can see how the initial output is and how it looks after the lines go missing. Unlike my other bug report, the issue is not a drawing issue. The lines actually become empty. Refreshing, scrolling up and then or anything I do doesn't make the lines come back. Again, this could be an issue with `less` as `git diff` gives the output through `less` (I guess). After quitting `less`, the output remains as it is, with the missing lines on the screen. Calling `git diff` again doesn't cause this issue. Additional information: $ lsb_release -rd: Description: Ubuntu 12.04.2 LTS Release: 12.04 $ apt-cache policy gnome-terminal gnome-terminal: Installed: 3.4.1.1-0ubuntu1 Candidate: 3.4.1.1-0ubuntu1 Version table: *** 3.4.1.1-0ubuntu1 0 500 http://it.archive.ubuntu.com/ubuntu/ precise/main i386 Packages 100 /var/lib/dpkg/status ** Affects: gnome-terminal (Ubuntu) Importance: Undecided Status: New ** Attachment added: "missing-lines.tar.bz2" https://bugs.launchpad.net/bugs/1161941/+attachment/3606494/+files/missing-lines.tar.bz2 -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1161941 Title: git-diff losing lines To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/gnome-terminal/+bug/1161941/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
