[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-12 09:35 --- Strange... Have you tried also to compile the latest CVS ? Yes, maybe it is a library problem. I will recompile regularly to see if updating changes something. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: RESOLVED creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-12 05:35 --- Using Thomas's kernel + mencoder recompiled has better performance, but not nearly what I used to have. I used to be able to (easily) get 30fps recording; when I first saw this problem after an update, it dropped drastically (it couldn't even sustain 15fps). Now it's pretty steady at ~20fps if nothing else is running, but can't do any better. If it makes a shred of difference, I'm using the following command, which has worked for months and months and months just fine: mencoder -tv on:driver=v4l:width=496:height=384:chanlist=us-cable:channel=3:norm=ntsc -oac copy -ovc lavc -lavcopts vcodec=mjpeg -endpos 02:20:00 Well, maybe this isn't a kernel problem and is really a library issue. I don't know: any ideas how to track it down? This computer tracks cooker more-or-less in real time. I don't just have old versions of everything lying around. =) Anyway, as far as this particular bug goes, 'cat /proc/interrupts' still shows only CPU0 getting IRQs (which still seems fishy to me) ... --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: RESOLVED creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-12 05:04 --- Recompiling mplayer (tried both in plf mode and not, tried with optimizations and not) didn't help. I'll try Thomas' test kernel and see if it helps. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: RESOLVED creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-12 04:32 --- Eric, I can confirm what you've written here: I tried reverting to as many kernels as I could try, and they all had the same (bad) performance. In addition, after using smp_affinity to make IRQ's route to both CPUs, mencoder performance didn't improve. For now, I'll see if I can get back the performance I need by recompiling from source here. However, it's still perturbing to see that by default IRQ's are not being routed to both processors when the APIC is enabled. Even if it's not the cause of this particular performance problem, it's still something that's changed for the worse. Perhaps we should open a bug just about that issue? --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: RESOLVED creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 [EMAIL PROTECTED] changed: What|Removed |Added Status|NEW |RESOLVED Resolution||INVALID --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 23:22 --- I confirm : using an old version of the libavcodec CVS, I do not have the performance drop. Meanwhile, I have updated to the latest gcc, maybe it is related. So I resolve the bug to INVALID. Wes, contact me if you think that this bug has to be reopened anyway. I suggest you try to recompile from source the 0.90RC4 release of MPlayer after upgarding cooker. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: RESOLVED creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 23:06 --- I have made extensive tests and I am sure this is NOT related to ACPI, and NOT to the kernel !! Here are my tests : I installed different versions of Mandrake, then compiled mplayer from source and tested mencoder on the same vob file sample, with the same parameters, using mpeg4 from libavcodec. On Mdk 9.0 (kernel 2.4.19-16) : 24 fps On Mdk9.1 RC2 (kernel 2.4.21), no ACPI : 11 fps Then On Mdk 9.1RC2 I installed the 2.4.29-16 kernel from 9.0 and booted on it. The test gave me 11 fps !! So the old kernel on the new 9.1RC2 system had the same performance drop. My theory is that it is NOT related to the kernel, but to mencoder or libavcodec, which has its performances crippled using the new glibc or gcc which is in the 9.1RC2. I don't know how to correct this problem, but probably the solution belongs in the mailing list of MPlayer or ffmpeg projects. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 19:30 --- I tried booting with acpi=off as someone suggested; it doesn't make any difference. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 18:31 --- Are you sure this is related to ACPI ? I don't use ACPI, my machine is a single Athlon processor and I have exactly the same performance drop using mencoder. I will also post some tests asap. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 17:31 --- Okay, work machine, SMP, same behavior as at home. EXCEPT: this work machine is 9.0, but with the 2.4.19-24mdksmp kernel! $ cat /proc/interrupts CPU0 CPU1 0: 52397687 0IO-APIC-edge timer 1: 265376 0IO-APIC-edge keyboard 2: 0 0 XT-PIC cascade 3: 1 0IO-APIC-edge serial 4: 1 0IO-APIC-edge serial 6: 81 0IO-APIC-edge floppy 8: 53359184 0IO-APIC-edge rtc 15: 78446 1IO-APIC-edge ide1 16: 2 0 IO-APIC-level ohci1394 17: 21516 0 IO-APIC-level Intel 82801BA-ICH2 19:1317838 0 IO-APIC-level usb-uhci 22: 614901 0 IO-APIC-level aic7xxx 23: 65467297 0 IO-APIC-level usb-uhci, eth0 NMI: 0 0 LOC: 52397233 52397231 ERR: 0 MIS: 0 Notice APIC is enabled; I can do the same cat trick to /proc/irq/x/smp_affinity and get the same behavior I mentioned before (CPU1 starts handling some interrupts). This machine (work)'s info: $ cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 15 model : 2 model name : Intel(R) XEON(TM) CPU 2.40GHz stepping: 4 cpu MHz : 2392.762 cache size : 512 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm bogomips: 4771.02 processor : 1 vendor_id : GenuineIntel cpu family : 15 model : 2 model name : Intel(R) XEON(TM) CPU 2.40GHz stepping: 4 cpu MHz : 2392.762 cache size : 512 KB fdiv_bug: no hlt_bug : no f00f_bug: no coma_bug: no fpu : yes fpu_exception : yes cpuid level : 2 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm bogomips: 4784.12 $ cat /var/log/dmesg Linux version 2.4.19-24mdksmp ([EMAIL PROTECTED]) (gcc version 3.2 (Mandrake Linux 9.0 3.2-1mdk)) #1 SMP Thu Jan 30 10:32:17 MST 2003 BIOS-provided physical RAM map: BIOS-e820: - 000a (usable) BIOS-e820: 000f - 0010 (reserved) BIOS-e820: 0010 - 7ff77000 (usable) BIOS-e820: 7ff77000 - 7ff79000 (ACPI NVS) BIOS-e820: 7ff79000 - 8000 (reserved) BIOS-e820: fec0 - fec1 (reserved) BIOS-e820: fee0 - fee1 (reserved) BIOS-e820: ffb0 - 0001 (reserved) Warning only 896MB will be used. Use a HIGHMEM enabled kernel. 896MB LOWMEM available. found SMP MP-table at 000fe710 hm, page 000fe000 reserved twice. hm, page 000ff000 reserved twice. hm, page 000f reserved twice. Advanced speculative caching feature not present On node 0 totalpages: 229376 zone(0): 4096 pages. zone(1): 225280 pages. zone(2): 0 pages. Intel MultiProcessor Specification v1.4 Virtual Wire compatibility mode. OEM ID: DELL Product ID: WS 530 APIC at: 0xFEE0 Processor #0 Unknown CPU [15:2] APIC version 20 Processor #2 Unknown CPU [15:2] APIC version 20 I/O APIC #4 Version 32 at 0xFEC0. Processors: 2 Kernel command line: BOOT_IMAGE=2419-24smp ro root=801 devfs=mount hdd=ide-scsi ide_setup: hdd=ide-scsi Initializing CPU#0 Detected 2392.762 MHz processor. Console: colour VGA+ 80x25 Calibrating delay loop... 4771.02 BogoMIPS Memory: 904556k/917504k available (1315k kernel code, 12564k reserved, 487k data, 140k init, 0k highmem) Dentry cache hash table entries: 131072 (order: 8, 1048576 bytes) Inode cache hash table entries: 65536 (order: 7, 524288 bytes) Mount-cache hash table entries: 16384 (order: 5, 131072 bytes) Buffer-cache hash table entries: 65536 (order: 6, 262144 bytes) Page-cache hash table entries: 262144 (order: 8, 1048576 bytes) CPU: Before vendor init, caps: 3febfbff , vendor = 0 CPU: L1 I cache: 12K, L1 D cache: 8K CPU: L2 cache: 512K CPU: Physical Processor ID: 0 CPU: After vendor init, caps: 3febfbff Intel machine check architecture supported. Intel machine check reporting enabled on CPU#0. CPU: After generic, caps: 3febfbff CPU: Common caps: 3febfbff Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 17:25 --- Everyone - I will re-post the details I sent to cooker before, and try to give more details tonight, I promise. =) I am tied up at work right now - I did post a lot of [fairly] detailed info in the cooker list yesterday; I had tried to make a bug but bugzilla was giving me problems (internal server errors) at the time. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
Re: [Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
But you could try a 9.0 kernel?
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 17:10 --- No, I can't, I don't have a copy of 9.0 around and I don't have space to install one. IIRC, vorbis encoding speed was the same in 9.0. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 16:34 --- Interesting. So : either it is CPU related, either you had already a problematic kernel when you first used mencoder. Can you try on a machine with mandrake 9.0 ? I would think this is CPU related. I wish Wesley Landaker could indicate his specs and how he could diagnose this CPU0 interrupt issue. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 16:20 --- I have a script to record from my TV card in realtime, using mencoder. It encoded at 25fps when I wrote the script a couple of weeks ago, and it encodes at 25fps now. (I use the standard kernel, not smp or multimedia or anything, version -13mdk). --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 15:52 --- The best test is with mencoder. Try encoding an unencrypted VOB file, or TV if you have a TV card, and compare fps between 9.0 and cooker. Here is a script to encode a vob file in mpeg4 with libavcodec (adjust SCALE and CROP values to your movie): #!/bin/sh SCALE=560:240 CROP=712:428:2:72 CIBLE=/space/video.avi SOURCE=/space/dvd.vob LAVCOPT=vcodec=mpeg4:vbitrate=800:v4me:vhq mencoder -nosound -o /dev/null -ovc lavc -lavcopts $LAVCOPT:vpass=1 -vop scale=$SCALE,crop=$CROP -zoom -sws 2 $SOURCE && mencoder -nosound -o $CIBLE - ovc lavc -lavcopts $LAVCOPT:vpass=2 -vop scale=$SCALE,crop=$CROP -zoom -sws 2 $SOURCE This one will encode TV on the fly (for French tv, adjust values/norm to your country): #!/bin/sh case $1 in 1 ) freq=647.25 chaine=tf1 ;; 2 ) freq=687.25 chaine=france2 ;; 3 ) freq=735.25 chaine=france3 ;; 4 ) freq=823.25 chaine=canal+ ;; 5 ) freq=671.25 chaine=arte ;; 6 ) freq=799.25 chaine=m6 ;; 7 ) freq=775.25 chaine=tv5 ;; * ) echo "unknown channel" exit ;; esac echo "Press ENTER to start" read mencoder -tv on:Driver=v4l:input=0:norm=SECAM:freq=$freq:width=768:height=576 - vop pp=hb/vb/dr/al/lb -oac mp3lame -ovc lavc -lavcopts vcodec=mjpeg -o ${chaine}.avi For encoding in vorbis, I noticed it using transcode, I don't know if this is reproducible with other softwares. My CPU is an AthlonXP2400+. --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.
[Cooker] [Bug 3172] [kernel] all interrupts routed to CPU0 cripples kernel performance
http://qa.mandrakesoft.com/show_bug.cgi?id=3172 --- Additional Comments From [EMAIL PROTECTED] 2003-03-11 15:33 --- Can't confirm for Vorbis encoding on my machine, but I can't quite tell from the description if this bug is specific to certain processor setups. I just have a single processor Duron system. Vorbis encoding is the same speed currently as it always was. Are there more detailed steps to reproduce, or anything? --- You are receiving this mail because: --- You are on the CC list for the bug, or are watching someone who is. --- Reminder: --- assigned_to: [EMAIL PROTECTED] status: NEW creation_date: description: This bug has been signaled on cooker mailing list, I signal it on behalf of Wesley J Landaker, who could not reach Bugzilla: > IO-APIC is enabled on my system, however all interrupts are routed to > CPU0. This didn't used to be the case, and I see a *drastic* > performance hit because of this (for instance, using mencoder to > capture real-time TV I used to get 35fps, now I can't even get > 20fps). It has been confirme by several people with different kernels: with standard kernel 2.4.21, I observed a very drastic performance problem with mencoder, encoding a video or TV, or encoding sound in oggvorbis (about 5 times slower). This happens also with the smp kernel, and this is even worse with the multimedia kernel.