|
Anyone experiencing general stuttering while transfering data over USB2.0?
System specs:
:: Code :: System: Host: host Kernel: 3.2.0-4.dmz.1-liquorix-686 i686 (32 bit, gcc: 4.6.2)
Desktop: Xfce 4.8.3 (Gtk 2.24.6) Distro: Ubuntu 11.10 oneiric Machine: Mobo: ASUSTeK model: P5PE-VM version: Rev 1.03G serial: 0 Bios: American Megatrends version: 1501 date: 12/26/2007 CPU: Single core Intel Pentium 4 CPU (-HT-) clocked at 2400.00 MHz Graphics: Card: Intel 82865G Integrated Graphics Controller bus-ID: 00:02.0 X.org: 1.10.4 driver: intel tty size: 80x24 Advanced Data: N/A for root Network: Card: Marvell 88E8001 Gigabit Ethernet Controller driver: skge ver: 1.14 port: e800 bus-ID: 01:0d.0 Drives: HDD Total Size: 480.1GB (4.8% used) Info: Processes: 118 Uptime: 8 min Memory: 354.9/1000.3MB Runlevel: 2 Gcc sys: 4.6.1 Client: Shell inxi: 1.7.28 So I'm experiencing some general stuttering (mouse stutters, even video playback stutters, but not sound) while I'm copying a couple of GBs of data over a USB2.0 FAT32 flash drive (ehci mode), while watching a measly xvid video. CPU usage stays well below 50%, but still the suttering occurs. This is the first time I have observed this while using Liquorix kernel. This is not the case with the stock Ubuntu kernel. Even without watching a video, by simply moving the mouse while the copy is in progress the symptom may be easily triggered. Could somebody give any hints on where to look to troubleshoot this stuttering? Back to top |
|||||
|
UPDATE:
So I did some further tests and found out that the stuttering started when I enabled frequency scaling in my procie, which was previously disabled because cpufreq-related modules weren't automatically loading in Liquorix. Whereas in the stock kernel the cpufreq-related modules are built-in and doesn't affect the performance of the system while transferring data over USB2.0 and playing back video, the same cannot be said of Liquorix by having frequency scaling. Where can I look in my system to troubleshoot this suttering? I would really like to use Liquorix and frequency scaling at the same time. The stuttering briefly makes the system perform like a Pentium MMX of the old days, unable to handle concurrent tasks, even without taxing the CPU usage. Back to top |
|||||
|
As observed in enable Threadirqs in liquorix, can you try booting with threadirqs in the kernel parameters? This will be a good test to see if there are any compatibility issues.
And also, I'll be integrating the cpufreq modules in the next kernel package, I just need more changes to collect before releasing a new version. Back to top |
|||||
|
Thank you for the suggestion, but it didn't help with the stuttering.
Regarding integration of cpufreq-related modules in Liquorix in the next release, it seems that the Ubuntu kernel, even though it also makes use of acpi-cpufreq module, doesn't seem to make use of the different cpufreq scaling governor modules. See the following for comparison: ls /sys/modules/ (Ubuntu): :: Code :: 8250
ac97_bus acpi acpi_cpufreq apparmor ata_generic ata_piix battery binfmt_misc block brd cpufreq_nforce2 debug_core dns_resolver drm drm_kms_helper edd efivars ehci_hcd eisa_bus fuse gx_suspmod i2c_algo_bit i8042 i915 intel_idle ip6table_filter ip6_tables ip6t_LOG ip6t_rt iptable_filter ip_tables ipt_LOG ipt_REJECT ipv6 kernel keyboard kgdboc libata longhaul loop lp md_mod mmc_core mousedev netpoll nf_conntrack nf_conntrack_broadcast nf_conntrack_ftp nf_conntrack_ipv4 nf_conntrack_ipv6 nf_conntrack_netbios_ns nf_defrag_ipv4 nf_defrag_ipv6 nf_nat nf_nat_ftp parport parport_pc pata_acpi pata_sis pcie_aspm pciehp pci_hotplug pdc_adma pnp powernow_k7 ppdev ppp_generic printk processor psmouse rcutree rfkill scsi_mod serio_raw sg shpchp skge snd snd_ac97_codec snd_intel8x0 snd_page_alloc snd_pcm snd_rawmidi snd_seq snd_seq_device snd_seq_midi snd_seq_midi_event snd_timer soundcore speedstep_lib speedstep_smi spurious sr_mod tcp_cubic thermal tpm uhci_hcd uinput usbcore video virtual_root vt x_tables xt_addrtype xt_hl xt_limit xt_state xt_tcpudp xz_dec ls /sys/modules/ (Liquorix): :: Code :: 8250
aacraid ac97_bus acpi acpi_cpufreq af_packet agpgart alix apparmor ata_generic ata_piix binfmt_misc block button cdrom cpufreq_ondemand cpuidle crc16 dns_resolver drm drm_kms_helper ehci_hcd eisa_bus evdev ext4 freq_table fuse gpio_cs5535 i2c_algo_bit i2c_core i8042 i915 intel_agp intel_gtt intel_idle ip6_tables iptable_filter ip_tables ipt_LOG ipt_REJECT jbd2 kernel keyboard libata loop lp mbcache mousedev mperf netpoll nf_conntrack nf_conntrack_broadcast nf_conntrack_ftp nf_conntrack_ipv4 nf_conntrack_netbios_ns nf_defrag_ipv4 nf_nat nf_nat_ftp parport parport_pc pata_acpi pcie_aspm pci_hotplug ppdev printk processor psmouse pstore rcutree rng_core rtc_cmos scsi_mod sd_mod serio_raw sg shpchp skge snd snd_ac97_codec snd_intel8x0 snd_mixer_oss snd_page_alloc snd_pcm snd_pcm_oss snd_rawmidi snd_seq snd_seq_device snd_seq_dummy snd_seq_midi snd_seq_midi_event snd_seq_oss snd_timer soundcore spurious sr_mod tcp_vegas uhci_hcd usbcore video virtual_root vt x_tables xt_addrtype xt_limit xt_state xt_tcpudp xz_dec Back to top |
|||||
|
Can you try 3.2-5.dmz.1?
Back to top |
|||||
|
Using 3.2-5.dmz.1, the problem with stuttering is gone.
Looks like statically linking the cpufreq-related modules fixed it, or could it be because of some other update? I would like to know your thoughts on the matter. Thank you. Back to top |
|||||
|
All times are GMT - 8 Hours
|
|||||