Bug#616689: jessie still not working
I will try what you mention, to include '-x' in the script. Something I tried, by looking in forums, was to add these 3 lines at the end of the script lvm vgchange -ay activate_vg "$ROOT" activate_vg "$resume" But that trick sometimes doesn't work. That '-x' will help me have more info. thanks! 2014-03-19 13:50 GMT-03:00 Bas Wijnen : > I'm running unstable, and have been affected by this bug for a few weeks > now, which is really annoying. After reading this bug report, I decided > to set the -x flag on /usr/share/initramfs-tools/scripts/local-top/lvm2, > the problem was quickly found: this script refuses any path which does > not start with /dev/mapper/. The root device is given as > /dev/disk/by-uuid/*, so it is not activated. > > Simple workaround: change the grub commandline from root=UUID=* to > root=/dev/mapper/vg-root (fill in your volume group and root logical > volume name). Then it will activate and boot normally. > > I expected that this should be done by changing /etc/fstab and running > update-grub, but grub will put UUID values in /boot/grub/grub.conf > anyway. So it must be done in /boot/grub/grub.conf, and repeated every > time update-grub is run (which isn't ideal, but a lot better than typing > commands every time the system boots). > > I'm guessing this bug was triggered by a change in grub to always write > UUIDs to its config. Still, it should be fixed by changing the > initramfs-tools script to make it allow the UUID-based path. > > Thanks, > Bas > -- Wido
Bug#616689: jessie still not working
I just installed a fresh Jessie from mini daily (the image was created on 2/12 which is the latest available). this lvm issue remains, lvm2 package was intalled when I built this machine, so that fix did not work. I was able to get to the system running 'vgchange -ay', but I would appreciate if there would be a nicest solution than running a command just to boot. thanks! -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
Hi again. I've checked for kernel parameters, and I have none: $ cat /proc/cmdline BOOT_IMAGE=/vmlinuz-3.2.0-2-686-pae root=UUID=d4fe6cec-1f8d-4767-8b4b-4646660c1f31 ro quiet 2012/5/10 Wido > Oh, ok. I'm at work know, I'll check when I arrive home at night. > > However, I'm using the stock kernel without any custom paramer. I'm using > it 'as is', in other words =) so I don't think I'm gonna find something > there. > > > 2012/5/10 Ben Hutchings > >> On Thu, 2012-05-10 at 10:14 -0300, Wido wrote: >> > Sorry I haven't answered before, I've missed your mail in my inbox :$ >> > >> > >> > I have checked that kernel parameter using 'sysctl -a | grep iso', but >> > haven't find such option. Did I searched what you asked? >> [...] >> >> By kernel parameter, I mean a parameter on the kernel command line (set >> by the boot loader, and readable with 'cat /proc/cmdline'). >> >> Ben. >> >> -- >> Ben Hutchings >> Life is what happens to you while you're busy making other plans. >> - John >> Lennon >> > > > > -- > Wido > -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
Oh, ok. I'm at work know, I'll check when I arrive home at night. However, I'm using the stock kernel without any custom paramer. I'm using it 'as is', in other words =) so I don't think I'm gonna find something there. 2012/5/10 Ben Hutchings > On Thu, 2012-05-10 at 10:14 -0300, Wido wrote: > > Sorry I haven't answered before, I've missed your mail in my inbox :$ > > > > > > I have checked that kernel parameter using 'sysctl -a | grep iso', but > > haven't find such option. Did I searched what you asked? > [...] > > By kernel parameter, I mean a parameter on the kernel command line (set > by the boot loader, and readable with 'cat /proc/cmdline'). > > Ben. > > -- > Ben Hutchings > Life is what happens to you while you're busy making other plans. > - John Lennon > -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
Sorry I haven't answered before, I've missed your mail in my inbox :$ I have checked that kernel parameter using 'sysctl -a | grep iso', but haven't find such option. Did I searched what you asked? I'm more fan of K stuff, but currently I use E17 compiled by myself, login using KDM. This setup has been the same for the last 5 years at least and I never used affinity for any of them. At this moment I've hacked my system using a cron entry that changes all processes affinity to 'f' every minute. I'm kind of lost with this :S 2012/5/2 Ben Hutchings > On Wed, 2012-05-02 at 20:21 -0300, Wido wrote: > > Hi again, > > > > > > Ok, I can confirm this is recurrent. I have rebooted my desktop > > without changing anything (but checking default_smp_affinitty is set > > to 'f'). > > Are you talking about /proc/irq/default_smp_affinity? That only affects > IRQ handlers, not tasks. > > > After reboot, I checked smp affinitty again, it's still in 'f'. > > However, all the apps I start keep having the same behaviour and they > > start with affinitty 8. > [...] > > The only way I can see that this would happen is: > > - Using the kernel parameter isolcpus > - Setting affinity for a process that is the ancestor of your session, > e.g. gdm > > Ben. > > > -- > Ben Hutchings > Design a system any fool can use, and only a fool will want to use it. > -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
Hi again, Ok, I can confirm this is recurrent. I have rebooted my desktop without changing anything (but checking default_smp_affinitty is set to 'f'). After reboot, I checked smp affinitty again, it's still in 'f'. However, all the apps I start keep having the same behaviour and they start with affinitty 8. El martes 1 de mayo de 2012, Wido escribió: > Never, I haven't. Actually, I did knew it was possible to set the > affinity, but I was never curious about it until I got this. and checked > the cpu_smp_affinity and was always set to 'f' > > I did, however, installed some time ago 'ulatencyd' and 'schedtool', but I > uninstalled them (--purge remove) before creating the bug (and yes, the > system was rebooted after remove the packages). > > Would it be possible that those packages screwed something else? After > changing all my processes affinity, I haven't rebooted the system, so I > don't know if I'm going to get this again or not. Will check again tomorrow. > > cheers > > 2012/4/30 Ben Hutchings 'b...@decadent.org.uk');>> > >> On Mon, 2012-04-30 at 12:32 -0300, Wido wrote: >> > Ok, not a kernel bug! >> > >> > >> > My last question 'default taskset is set to mask 8, is that ok?' is >> > the key. I have a 4 cores system and mask 8 means the 4th core (the >> > one beeing actually used). >> >> Right, it selects CPU 3 only (according to Linux numbering). >> >> > I changed all the proccesses affinity to 'f' and then the processes >> > started to spread in all 4 cores, as it should. >> > >> > >> > Funniest part is, default_smp_affinity is set to 'f', so I don't know >> > why proccesses start with affinity 8 >> >> I don't know. Did you explicitly configure the affinity of anything >> before this? >> >> Ben. >> >> -- >> Ben Hutchings >> Design a system any fool can use, and only a fool will want to use it. >> > > > > -- > Wido > -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
Never, I haven't. Actually, I did knew it was possible to set the affinity, but I was never curious about it until I got this. and checked the cpu_smp_affinity and was always set to 'f' I did, however, installed some time ago 'ulatencyd' and 'schedtool', but I uninstalled them (--purge remove) before creating the bug (and yes, the system was rebooted after remove the packages). Would it be possible that those packages screwed something else? After changing all my processes affinity, I haven't rebooted the system, so I don't know if I'm going to get this again or not. Will check again tomorrow. cheers 2012/4/30 Ben Hutchings > On Mon, 2012-04-30 at 12:32 -0300, Wido wrote: > > Ok, not a kernel bug! > > > > > > My last question 'default taskset is set to mask 8, is that ok?' is > > the key. I have a 4 cores system and mask 8 means the 4th core (the > > one beeing actually used). > > Right, it selects CPU 3 only (according to Linux numbering). > > > I changed all the proccesses affinity to 'f' and then the processes > > started to spread in all 4 cores, as it should. > > > > > > Funniest part is, default_smp_affinity is set to 'f', so I don't know > > why proccesses start with affinity 8 > > I don't know. Did you explicitly configure the affinity of anything > before this? > > Ben. > > -- > Ben Hutchings > Design a system any fool can use, and only a fool will want to use it. > -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
Ok, not a kernel bug! My last question 'default taskset is set to mask 8, is that ok?' is the key. I have a 4 cores system and mask 8 means the 4th core (the one beeing actually used). I changed all the proccesses affinity to 'f' and then the processes started to spread in all 4 cores, as it should. Funniest part is, default_smp_affinity is set to 'f', so I don't know why proccesses start with affinity 8 2012/4/30 Wido > Hi again > > 2012/4/29 Ben Hutchings > >> On Sun, Apr 29, 2012 at 01:44:56PM -0300, Wido wrote: >> > I tend to follow new kernel feature from 'the H online' and I don't >> recall >> > seeing something like this was ever mentioned. I know current default >> sched >> > is the BFS, but it doesn't mention such behavior. >> >> The scheduler is CFS. BFS is someone else's project outside of >> mainline Linux. > > Then I misunderstood what I read. I know it's from Con Kolivas, I just > though it is set as default. My bad :D > > >> > A little test I did was opening a youtube video (I use chrome), the >> cpu was >> > almost 100% but, in htop. around 70% was show as low pri. If I run >> another >> > program, lets say VirtualBox, they start to share the same core, >> instead of >> > using another one that is free. >> >> That does sound wrong. However it is possible that the CPU frequency >> is being changed so that that single core still has enough cycles to >> run both programs without slowing them down. > > My CPU has the ability to powerscale, but almost always is running at max > speed. I have a gadget to play with cpufreq =) > > >> > This behavior seems buggy to me. I've been using Debian since Woody, >> this >> > desktop has almost 1 and a half year, and this started to happen when I >> > installed 3.1 >> >> Here's how you can test this: >> >> 1. Start the video playing >> 2. Measure how long it takes to boot a particular VM in VirtualBox >> 3. Repeat this another 4 times, so you have a total of 5 runs >> 4. Use taskset to force the video player and VirtualBox onto separate >> cores >> 5. Repeat another 5 times >> >> By comparing the two sets of 5 times, we can see whether the scheduler >> should be using more than one core. (Also, if the video plays more >> smoothly in step 5, then the scheduler should be using two cores. >> But this is not as easy to measure objectively.) >> >> You can substitute whatever else you're interested in for steps 1 and >> 2, just as long as it is something repeatable. > > I didn't need to run 5 times. The processes tends to bond to core 4, I was > playing a flash game and copying files from another pc (using samba) and > when I moved chrome and dolphin to cores 1 and 2, respectively, they > started working a LOT better, from almost non responsive to really smooth. > > SAR is not showing much yet, will leave the PC on and will run some more > things (with and without tasksel) to show differences on the cores usage. > But apart from that, not sure what other info can I send. > > > > Ben. >> >> -- >> Ben Hutchings >> We get into the habit of living before acquiring the habit of thinking. >> - Albert >> Camus >> > > BTW, default taskset is set to mask 8, is that ok? > > cheers > -- > Wido > -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
Hi again 2012/4/29 Ben Hutchings > On Sun, Apr 29, 2012 at 01:44:56PM -0300, Wido wrote: > > I tend to follow new kernel feature from 'the H online' and I don't > recall > > seeing something like this was ever mentioned. I know current default > sched > > is the BFS, but it doesn't mention such behavior. > > The scheduler is CFS. BFS is someone else's project outside of > mainline Linux. Then I misunderstood what I read. I know it's from Con Kolivas, I just though it is set as default. My bad :D > > A little test I did was opening a youtube video (I use chrome), the cpu > was > > almost 100% but, in htop. around 70% was show as low pri. If I run > another > > program, lets say VirtualBox, they start to share the same core, instead > of > > using another one that is free. > > That does sound wrong. However it is possible that the CPU frequency > is being changed so that that single core still has enough cycles to > run both programs without slowing them down. My CPU has the ability to powerscale, but almost always is running at max speed. I have a gadget to play with cpufreq =) > > This behavior seems buggy to me. I've been using Debian since Woody, this > > desktop has almost 1 and a half year, and this started to happen when I > > installed 3.1 > > Here's how you can test this: > > 1. Start the video playing > 2. Measure how long it takes to boot a particular VM in VirtualBox > 3. Repeat this another 4 times, so you have a total of 5 runs > 4. Use taskset to force the video player and VirtualBox onto separate > cores > 5. Repeat another 5 times > > By comparing the two sets of 5 times, we can see whether the scheduler > should be using more than one core. (Also, if the video plays more > smoothly in step 5, then the scheduler should be using two cores. > But this is not as easy to measure objectively.) > > You can substitute whatever else you're interested in for steps 1 and > 2, just as long as it is something repeatable. I didn't need to run 5 times. The processes tends to bond to core 4, I was playing a flash game and copying files from another pc (using samba) and when I moved chrome and dolphin to cores 1 and 2, respectively, they started working a LOT better, from almost non responsive to really smooth. SAR is not showing much yet, will leave the PC on and will run some more things (with and without tasksel) to show differences on the cores usage. But apart from that, not sure what other info can I send. Ben. > > -- > Ben Hutchings > We get into the habit of living before acquiring the habit of thinking. > - Albert Camus > BTW, default taskset is set to mask 8, is that ok? cheers -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
I tend to follow new kernel feature from 'the H online' and I don't recall seeing something like this was ever mentioned. I know current default sched is the BFS, but it doesn't mention such behavior. A little test I did was opening a youtube video (I use chrome), the cpu was almost 100% but, in htop. around 70% was show as low pri. If I run another program, lets say VirtualBox, they start to share the same core, instead of using another one that is free. This behavior seems buggy to me. I've been using Debian since Woody, this desktop has almost 1 and a half year, and this started to happen when I installed 3.1 cheers 2012/4/29 Ben Hutchings > On Sun, 2012-04-29 at 01:25 -0300, Wido wrote: > > Package: linux-2.6 > > Version: 3.2.15-1 > > Severity: important > > > > Dear Maintainer, > > I'm seeing some unusual behaviour in my 4 cores desktop and can't figure > out > > what's going on. Currently I'm using latest 'testing' kernel, but I've > noticed > > this behavior in all 3.x debian stock kernel series. > > > > Even when the kernels have SMP enabled, only one of my 4 cores gets all > the CPU > > affinity, distributing the load only when it's at 100%. > [...] > > Why do you think this is a bug? > > Ben. > > -- > Ben Hutchings > I haven't lost my mind; it's backed up on tape somewhere. > -- Wido
Bug#670797: linux-image-3.2.0-2-686-pae: CPU load not being distributed
Package: linux-2.6 Version: 3.2.15-1 Severity: important Dear Maintainer, I'm seeing some unusual behaviour in my 4 cores desktop and can't figure out what's going on. Currently I'm using latest 'testing' kernel, but I've noticed this behavior in all 3.x debian stock kernel series. Even when the kernels have SMP enabled, only one of my 4 cores gets all the CPU affinity, distributing the load only when it's at 100%. I have checked for SMP_default_afinity, but it's in F (default), which I've read it's ok. I recently installed 'sar' to get more detailed information, but here are the outputs from mpstat and /proc/interrupts, they clearly show processing differences among all CPU cores: $ mpstat -P ALL Linux 3.2.0-2-686-pae (frankie) 29/04/12_i686_ (4 CPU) 01:03:26 CPU%usr %nice%sys %iowait%irq %soft %steal %guest %idle 01:03:26 all2,68 13,793,127,240,000,560,00 0,00 72,61 01:03:26 01,030,830,775,940,000,000,00 0,00 91,42 01:03:26 10,650,680,515,210,000,010,00 0,00 92,95 01:03:26 20,751,100,635,760,000,170,00 0,00 91,60 01:03:26 38,28 52,46 10,55 12,060,002,050,00 0,00 14,60 $ cat /proc/interrupts CPU0 CPU1 CPU2 CPU3 0: 87287 12806 143287 IO-APIC-edge timer 1: 0 3153 3351 IO-APIC-edge i8042 4: 0 0 0 2 IO-APIC-edge 6: 0 0 38 1334 IO-APIC-edge floppy 7: 1 0 0 0 IO-APIC-edge parport0 8: 0 0 2 54 IO-APIC-edge rtc0 9: 0 0 0 0 IO-APIC-fasteoi acpi 14: 1100 2774 46679 IO-APIC-edge pata_atiixp 15: 0 0 0 0 IO-APIC-edge pata_atiixp 16: 1 23816 32680 IO-APIC-fasteoi ohci_hcd:usb3, ohci_hcd:usb4, snd_hda_intel 17: 0 3209 21995 IO-APIC-fasteoi ehci_hcd:usb1 18: 0 0 1 52 IO-APIC-fasteoi ohci_hcd:usb5, ohci_hcd:usb6, ohci_hcd:usb7, fglrx[0]@PCI:1:5:0 19: 0 0 0 19 IO-APIC-fasteoi ehci_hcd:usb2, snd_hda_intel 21: 1210 7109 621617 IO-APIC-fasteoi ath 22: 4106 3942 112678 IO-APIC-fasteoi ahci 42: 0 0 0 0 PCI-MSI-edge eth0 NMI: 9 8 9137 Non-maskable interrupts LOC: 57319 43568 42688 624135 Local timer interrupts SPU: 0 0 0 0 Spurious interrupts PMI: 9 8 9137 Performance monitoring interrupts IWI: 0 0 0 0 IRQ work interrupts RES: 228406 119635 175374 296599 Rescheduling interrupts CAL: 4245 4228 4225539 Function call interrupts TLB: 2111 2331 2262 4288 TLB shootdowns TRM: 0 0 0 0 Thermal event interrupts THR: 0 0 0 0 Threshold APIC interrupts MCE: 0 0 0 0 Machine check exceptions MCP: 7 7 7 7 Machine check polls ERR: 1 MIS: 0 This is my processor: # cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 16 model : 2 model name : AMD Phenom(tm) 9600 Quad-Core Processor stepping: 2 microcode : 0x183 cpu MHz : 2300.000 cache size : 512 KB physical id : 0 siblings: 4 core id : 0 cpu cores : 4 apicid : 0 initial apicid : 0 fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 5 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm 3dnowext 3dnow constant_tsc nonstop_tsc extd_apicid pni monitor cx16 popcnt lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs npt lbrv svm_lock bogomips: 4830.18 clflush size: 64 cache_alignment : 64 address sizes : 48 bits physical, 48 bits virtual power management: ts ttp tm stc 100mhzsteps hwpstate thanks in advance! -- Package-specific info: ** Version: Linux version 3.2.0-2-686-pae (Debian 3.2.15-1) (debian-kernel@lists.debian.org) (gcc version 4.6.3 (Debian 4.6.3-3) ) #1 SMP Sun Apr 15 17:56:31 UTC 2012 ** Command line: BOOT_IMAGE=/vmlinuz-3.2.0-2-686-pae root=UUID=d4fe6
