Bug#898941: caffe-cpu: Very different results for gpu and cpu calculations due to possibly problematic BLAS dependency
Package: caffe-cpu Severity: important Dear Maintainer, I'm getting very different results in cpu mode compared to gpu mode, when using the libblas3 (version 3.7.0-2) package. After installing the libopenblas-base package (version 0.2.19-3) the problems disappeared. I don't know whether this is a dependency problem of the package or it is a known problem in libblas3. I was able to uninstall and install libopenblas-base while libblas3 was still being installed, which is a little bit confusing to me. Note that I reported the bug initially here: https://github.com/twhui/MSG-Net/issues/2 Note also that this bug has been found with a recompiled package with matlab support enabled. I used this howto https://github.com/BVLC/caffe/blob/master/docs/install_apt_debian.md, and changed only the matlab options. However, I don't think that the behaviour is visible in matlab only, though I didn't try this. Kind Regards Christoph -- System Information: Debian Release: 9.4 APT prefers stable APT policy: (1000, 'stable'), (900, 'stable'), (500, 'stable-updates') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 4.9.0-6-amd64 (SMP w/8 CPU cores) Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8), LANGUAGE=en_US:en (charmap=UTF-8) Shell: /bin/sh linked to /bin/dash Init: systemd (via /run/systemd/system) Versions of packages caffe-cpu depends on: pn caffe-tools-cpu ii libblas3 [libblas.so.3] 3.7.0-2 pn libcaffe-cpu1 ii libopenblas-base [libblas.so.3] 0.2.19-3 pn python3-caffe-cpu caffe-cpu recommends no packages. Versions of packages caffe-cpu suggests: pn caffe-doc pn libcaffe-cpu-dev
Bug#883935: linux-image-4.9.0-4-amd64: i915 GPU hang after every suspend or hibernate (Hang on render ring)
Package: src:linux Version: 4.9.65-3 Followup-For: Bug #883935 Dear Maintainer, I have the same issue, and I find it very annoying. As a workaround I use kernel linux-image-4.12.0-1-amd64, which mostly manages to wake correctly from hibernation. But honestly, I'm a bit overstrained in using a testing kernel on an otherwise stable system. I was hoping that the recent update of 4.9.0 kernel maybe fixed the issue, but unfortunately not. It would be very nice if this could be fixed with a future update. Thank you in advance! Christoph -- Package-specific info: ** Version: Linux version 4.9.0-4-amd64 (debian-ker...@lists.debian.org) (gcc version 6.3.0 20170516 (Debian 6.3.0-18) ) #1 SMP Debian 4.9.65-3 (2017-12-03) ** Command line: BOOT_IMAGE=/boot/vmlinuz-4.9.0-4-amd64 root=UUID=39cb1350-d7bf-43e6-9a89-f1821f0ae9b9 ro quiet i915.preliminary_hw_support=1 ** Tainted: PUO (4161) * Proprietary module has been loaded. * Userspace-defined naughtiness. * Out-of-tree module has been loaded. ** Kernel log: [ 167.358244] ata1.00: supports DRM functions and may not be fully accessible [ 167.358695] ata1.00: configured for UDMA/133 [ 167.383405] usb 1-4: reset high-speed USB device number 2 using xhci_hcd [ 167.643093] usb 1-7: reset full-speed USB device number 5 using xhci_hcd [ 167.903455] usb 1-6: reset full-speed USB device number 3 using xhci_hcd [ 167.910793] psmouse serio1: synaptics: queried max coordinates: x [..5676], y [..4758] [ 167.946771] psmouse serio1: synaptics: queried min coordinates: x [1266..], y [1096..] [ 168.163211] usb 1-9: reset full-speed USB device number 8 using xhci_hcd [ 168.422910] usb 1-8: reset high-speed USB device number 7 using xhci_hcd [ 168.830965] usb 1-4.2: reset low-speed USB device number 6 using xhci_hcd [ 169.387447] usb 1-4.1: reset low-speed USB device number 4 using xhci_hcd [ 169.720828] PM: restore of devices complete after 2684.862 msecs [ 169.721308] usb 1-7:1.0: rebind failed: -517 [ 169.721313] usb 1-7:1.1: rebind failed: -517 [ 169.721665] PM: Image restored successfully. [ 169.721774] PM: Basic memory bitmaps freed [ 169.721780] Restarting tasks ... done. [ 169.731277] thermal thermal_zone2: failed to read out thermal zone (-5) [ 169.739769] Bluetooth: hci0: Bootloader revision 0.0 build 2 week 52 2014 [ 169.746776] Bluetooth: hci0: Device revision is 5 [ 169.746777] Bluetooth: hci0: Secure boot is enabled [ 169.746778] Bluetooth: hci0: OTP lock is enabled [ 169.746779] Bluetooth: hci0: API lock is enabled [ 169.746779] Bluetooth: hci0: Debug lock is disabled [ 169.746780] Bluetooth: hci0: Minimum firmware build 1 week 10 2014 [ 169.751150] bluetooth hci0: firmware: direct-loading firmware intel/ibt-11-5.sfi [ 169.751153] Bluetooth: hci0: Found device firmware: intel/ibt-11-5.sfi [ 169.865935] e1000e: eth1 NIC Link is Down [ 169.867438] IPv6: ADDRCONF(NETDEV_UP): eth1: link is not ready [ 170.075048] IPv6: ADDRCONF(NETDEV_UP): eth1: link is not ready [ 170.075720] IPv6: ADDRCONF(NETDEV_UP): wlp3s0: link is not ready [ 170.079558] iwlwifi :03:00.0: L1 Enabled - LTR Enabled [ 170.082036] iwlwifi :03:00.0: L1 Enabled - LTR Enabled [ 170.212136] iwlwifi :03:00.0: L1 Enabled - LTR Enabled [ 170.212626] iwlwifi :03:00.0: L1 Enabled - LTR Enabled [ 170.309080] IPv6: ADDRCONF(NETDEV_UP): wlp3s0: link is not ready [ 170.321546] iwlwifi :03:00.0: L1 Enabled - LTR Enabled [ 170.323628] iwlwifi :03:00.0: L1 Enabled - LTR Enabled [ 170.458950] iwlwifi :03:00.0: L1 Enabled - LTR Enabled [ 170.459265] iwlwifi :03:00.0: L1 Enabled - LTR Enabled [ 170.491370] [drm] RC6 on [ 170.553392] IPv6: ADDRCONF(NETDEV_UP): wlp3s0: link is not ready [ 170.603365] IPv6: ADDRCONF(NETDEV_UP): wlp3s0: link is not ready [ 170.629381] psmouse serio2: trackpoint: IBM TrackPoint firmware: 0x0e, buttons: 3/3 [ 170.825772] input: TPPS/2 IBM TrackPoint as /devices/platform/i8042/serio1/serio2/input/input23 [ 171.338681] Bluetooth: hci0: Waiting for firmware download to complete [ 171.338772] Bluetooth: hci0: Firmware loaded in 1562047 usecs [ 171.338958] Bluetooth: hci0: Waiting for device to boot [ 171.349737] Bluetooth: hci0: Device booted in 10636 usecs [ 171.349930] bluetooth hci0: firmware: direct-loading firmware intel/ibt-11-5.ddc [ 171.349932] Bluetooth: hci0: Found Intel DDC parameters: intel/ibt-11-5.ddc [ 171.353746] Bluetooth: hci0: Applying Intel DDC parameters completed [ 174.173465] IPv6: ADDRCONF(NETDEV_UP): wlp3s0: link is not ready [ 182.459087] [drm] GPU HANG: ecode 9:0:0x81ba3c75, in Xorg [709], reason: Hang on render ring, action: reset [ 182.459089] [drm] GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace. [ 182.459090] [drm] Please file a _new_ bug report on bugs.freedesktop.org against DRI -> DRM/Intel [ 182.459091] [drm] drm/i915 developers can then reassign to the right component if it's not a kernel issue. [ 182.459092] [drm] The
Bug#648477: ath9k wireless: frequent connection drops (TKIP MIC failure)
On 11/11/11 23:26, Jonathan Nieder wrote: Hi Christoph, Christoph Wiedemann wrote: I observe frequent (1 per minute) connection drops with a new SONY VAIO notebook. [...] 02:00.0 Network controller [0280]: Atheros Communications Inc. AR9485 Wireless Network Adapter [168c:0032] (rev 01) Disabling hwcrypt as suggested in http://www.mail-archive.com/ath9k- de...@lists.ath9k.org/msg04221.html seems to fix the problem Thanks. There have been some AR9485 fixes upstream recently. Could you test v3.1 from experimental? If that fails, please report it upstream to ath9k-de...@lists.ath9k.org, cc-ing linux-wirel...@vger.kernel.org and either me or this bug log so we can track it. Hi Jonathan, linux-image-3.1.0-1-amd64 indeed seems to fix the problem. I tested for half an hour now without any connection drop issues. Thanks and sorry for bothering you... Best Regards Christoph -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
