gmane.linux.rt.user

Re: [PATCH v2 2/4] sched: track the next-highest priority on each runqueue

Gregory Haskins — 2008-12-04T04:22:23

Crap. Trying to fall asleep tonight, I realized this is a bug I think. Looks like I will need a v3 It should be "prio <= rt_rq->highest_prio.next" or we can miss updating .next properly.

Re: [RFC PATCH 0/3] sched: track next-highest priority (was "Series short discription")

Gregory Haskins — 2008-12-03T20:39:13

Hi Randy, Indeed. I fat-fingered that one on the original patch submission and now it persists in the thread for all eternity for all who reply to the original ;) I have fixed it above. Sorry for the confusion. -Greg -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: 2.6.27.5 RT patch status (26rt11)

(null — 2008-11-30T18:19:02

In article <20081110234740.GB18819< at >windriver.com>, Paul Gortmaker windriver.com> wrote: " I've updated from the 2.6.27-rc8 patch series based on 26rt9 to be " based on 2.6.27.5 kernel.org and 26rt11 preempt_rt. [..] " git://opensource1.windriver.com/people/paulg/rt/linux-2.6 " " Note the patched tree is on the branch v2.6.27.5-26rt11 and that " master is just kernel.org master. $ head -6 Makefile VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 27 EXTRAVERSION = .5-26rt11 NAME = Trembling Tortoise $ make [...] CC [M] drivers/acpi/wmi.o CC [M] drivers/acpi/asus_acpi.o CC [M] drivers/acpi/toshiba_acpi.o CC drivers/ata/libata-core.o CC drivers/ata/libata-scsi.o CC drivers/ata/libata-eh.o CC drivers/ata/libata-sff.o CC drivers/ata/libata-pmp.o CC drivers/ata/libata-acpi.o LD drivers/ata/libata.o CC drivers/ata/sata_nv.o LD drivers/ata/built-in.o LD drivers/atm/built-in.o LD drivers/auxdisplay/built-in.o CC [M] drive

RFC: Latency reducing TCP modifications for thin-stream interactive applications

Andreas Petlund — 2008-11-27T13:39:41

A wide range of Internet-based services that use reliable transport protocols display what we call thin-stream properties. This means that the application sends data with such a low rate that the retransmission mechanisms of the transport protocol are not fully effective. In time-dependent scenarios (like online games, control systems or some sensor networks) where the user experience depends on the data delivery latency, packet loss can be devastating for the service quality. Extreme latencies are caused by TCP's dependency on the arrival of new data from the application to trigger retransmissions effectively through fast retransmit instead of waiting for long timeouts. After analyzing a large number of time-dependent interactive applications, we have seen that they often produce thin streams (as described above) and also stay with this traffic pattern throughout its entire lifespan. The combination of time-dependency and the fact that the streams provoke high latencies when using TCP is unfo

Re: strange getrusage timer problem with rt-kernel

Cheng Renquan — 2008-11-26T03:55:23

-- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

strange getrusage timer problem with rt-kernel

2008-11-26T03:52:28

I'm testing 2.6.26.6-rt11 (is that the latest?) kernel, using tiotest from tiobench (http://sourceforge.net/projects/tiobench/), a classic multi-thread IO stress tool, it reported a strange problem, with: root:~/tiobench# time ./tiotest -R -d /dev/mapper/dg5-lv2 -t 1 -f 1000 -r 102400 tiotest: tiotest.c:609: add_timer: Assertion end_time->tv_sec >= start_time->tv_sec' failed.? this problem never happened in mainstream kernels. Regards, -- Cheng Renquan, Shenzhen, China -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: 2.6.27.5 RT patch status (26rt11)

Arnaldo Carvalho de Melo — 2008-11-21T17:56:52

Em Fri, Nov 21, 2008 at 06:25:59PM +0100, Alessio Igor Bogani escreveu: Nope, it was just once, after fixing that NR_CPUS problem I had that trucated output bug, then rebooting got to this one. But then I was sidetracked with other stuff, will try to get back to this soon. - Arnaldo -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: 2.6.27.5 RT patch status (26rt11)

Alessio Igor Bogani — 2008-11-21T17:25:59

Dear Sir Carvalho de Melo, 2008/11/20 Arnaldo Carvalho de Melo redhat.com>: [...] Do you incur in this bug systematically every time boot your machine? Thanks! Ciao, Alessio -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: 2.6.27.5 RT patch status (26rt11)

Arnaldo Carvalho de Melo — 2008-11-20T14:44:51

Another one, this time with more context: (gdb) l *0xffffffff8105d5d9 0xffffffff8105d5d9 is in wakeup_next_waiter (/home/acme/git/paulg-rt-linux-2.6/kernel/rtmutex_common.h:77). 72{ 73struct rt_mutex_waiter *w; 74 75w = plist_first_entry(&lock->wait_list, struct rt_mutex_waiter, 76 list_entry); 77BUG_ON(w->lock != lock); <-------------------------- 78 79return w; 80} 81 (gdb) Linux version 2.6.27.5-26rt11 (acme< at >mica.ghostprotocols.net) (gcc version 4.1.2 20071124 (Red Hat 4.1.2-42)) #8 Command line: ro root=LABEL=/ rhgb console=ttyS1,57600 console=tty0 idle=poll KERNEL supported cpus:al. Use ~? for help] Intel GenuineIntel AMD AuthenticAMD Centaur CentaurHauls BIOS-provided physical RAM map: BIOS-e820: 0000000000000000 - 00000000000a0000 (usable) BIOS-e820: 0000000000100000 - 00000000cffa8000 (usable) BIOS-e820: 00000000cffa8000 - 00000000cffb7c00 (ACPI data) BIOS-e820: 00000000cffb7c00 - 00000000d0000000 (reserved) BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved

Re: 2.6.27.5 RT patch status (26rt11)

Arnaldo Carvalho de Melo — 2008-11-20T14:19:26

Em Thu, Nov 20, 2008 at 02:43:09PM +0100, Alessio Igor Bogani escreveu: Wouldn't it be better to use for_each_possible_cpu()? But yes, it fixes this problem, we go further then die with: F2 = Setup Phoenix ROM BIOS PLUS Version 1.10 1.2.0 F10 = Utility Mode Copyright 1985-1988 Phoenix Technologies Ltd. F11 = Boot Menu Copyright 1990-2006 Dell Inc. F12 = PXE Boot All Rights Reserved Dell System PowerEdge 1900 www.dell.com Two 2.00 GHz Dual-core Processors, Bus Speed: 1333 MHz, L2 Cache: 4 MB System Memory Size: 4.0 GB, System Memory Speed: 533 MHz PowerEdge Expandable RAID Controller BIOS Copyright(c) 2006 LSI Logic Corporation Press to Run Configuration Utility HA -0 (Bus 2 Dev 14) PERC 5/i Integrated FW package: 5.1.1-0040 Booting 'Red Hat Enterprise Linux (realtime) (2.6.27.5-26rt11)' 1 Logical Drive(s) found on the host adapter. rootg(hd0,0)ive(s) handled by BIOS Filesystem type is ext2fs, part

Re: 2.6.27.5 RT patch status (26rt11)

Arnaldo Carvalho de Melo — 2008-11-20T13:46:12

Em Thu, Nov 20, 2008 at 02:43:09PM +0100, Alessio Igor Bogani escreveu: Thanks, looks like this should fix this, I'll test and report back. - Arnaldo -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: 2.6.27.5 RT patch status (26rt11)

Arnaldo Carvalho de Melo — 2008-11-20T13:43:33

Em Thu, Nov 20, 2008 at 11:29:10AM -0200, Arnaldo Carvalho de Melo escreveu: (gdb) l *0xffffffff8151d8b8 0xffffffff8151d8b8 is in fork_init (/home/acme/git/paulg-rt-linux-2.6/kernel/fork.c:233). 228init_task.signal->rlim[RLIMIT_NPROC].rlim_max = max_threads/2; 229init_task.signal->rlim[RLIMIT_SIGPENDING] = 230init_task.signal->rlim[RLIMIT_NPROC]; 231 232for (i = 0; i < NR_CPUS; i++) 233INIT_LIST_HEAD(&per_cpu(delayed_drop_list, i)); 234} 235 236int __attribute__((weak)) arch_dup_task_struct(struct task_struct *dst, 237 struct task_struct *src) (gdb) Here: 232for (i = 0; i < NR_CPUS; i++) 233 INIT_LIST_HEAD(&per_cpu(delayed_drop_list, i)); static DEFINE_PER_CPU(struct list_head, delayed_drop_list); This comes from patch preempt-realtime-mmdrop-delayed.patch, that is dbcb144fbb8adca0a98c4812e08da4e27e2bfadb in your tree. - Arnaldo -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kerne

Re: 2.6.27.5 RT patch status (26rt11)

Alessio Igor Bogani — 2008-11-20T13:43:09

Dear Sir Carvalho de Melo, 2008/11/20 Arnaldo Carvalho de Melo redhat.com>: [...] [...] [...] Could you try this patch, please? Thanks! Ciao, Alessio Index: linux-source-2.6.27/kernel/fork.c =================================================================== --- linux-source-2.6.27.orig/kernel/fork.c 2008-11-11 10:50:21.000000000 +0100 +++ linux-source-2.6.27/kernel/fork.c 2008-11-11 10:50:40.000000000 +0100 < at >< at > -229,7 +229,7 < at >< at > init_task.signal->rlim[RLIMIT_SIGPENDING] = init_task.signal->rlim[RLIMIT_NPROC]; - for (i = 0; i < NR_CPUS; i++) + for (i = 0; i < nr_cpu_ids; i++) INIT_LIST_HEAD(&per_cpu(delayed_drop_list, i)); } -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: 2.6.27.5 RT patch status (26rt11)

Arnaldo Carvalho de Melo — 2008-11-20T13:29:10

Em Thu, Nov 20, 2008 at 11:25:44AM -0200, Arnaldo Carvalho de Melo escreveu: kernel config http://oops.ghostprotocols.net:81/acme/v2.6.27.5-26rt11.config - Arnaldo -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: 2.6.27.5 RT patch status (26rt11)

Arnaldo Carvalho de Melo — 2008-11-20T13:25:44

Em Mon, Nov 10, 2008 at 06:47:40PM -0500, Paul Gortmaker escreveu: I hit one, DELL PE1900, dual dual core xeon, megasas disk controler, bnx2 NIC: ipmitool> sol activate [SOL Session operational. Use ~? for help] Initializing cgroup subsys cpuset Linux version 2.6.27.5-26rt11 (acme< at >mica.ghostprotocols.net) (gcc version 4.1.2 20071124 (Red Hat 4.1.2-42)) #1 SMP PREEMPT RT Thu Nov 20 10:51:37 BRST 2008 Command line: ro root=LABEL=/ rhgb console=ttyS1,57600 console=tty0 idle=poll KERNEL supported cpus: Intel GenuineIntel AMD AuthenticAMD Centaur CentaurHauls BIOS-provided physical RAM map: BIOS-e820: 0000000000000000 - 00000000000a0000 (usable) BIOS-e820: 0000000000100000 - 00000000cffa8000 (usable) BIOS-e820: 00000000cffa8000 - 00000000cffb7c00 (ACPI data) BIOS-e820: 00000000cffb7c00 - 00000000d0000000 (reserved) BIOS-e820: 00000000e0000000 - 00000000f0000000 (reserved) BIOS-e820: 00000000fe000000 - 0000000100000000 (reserved) BIOS-e820: 0000000100000000 - 0000000130000000 (usable) using polling

Did anyone tested core i7?

Amir Aharon — 2008-11-18T12:48:27

Hi, Did anyone tested linux-2.6.24.7-rt21 on the new Intel core i7 ? -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

few question about the platform tested and in use with config_preempt_rt...

Amir Aharon — 2008-11-16T13:59:37

Hi all, I have few questions regarding the platform tested with the config_preempt_rt: 1. When was last time the table at updated? 2. Was the worse case latency measured under stress? If so which stress test was used? 3. What kind of network tests has been used? Does the worse case latency valid for network application (RT low latency audio streaming)? 3. Were there any additional hardware tweaking done on the tested system (e.g. bios option tweaking such as turning off fan control)? 4. Were there any additional kernel option tweaking done, e.g. - clock source, tick-less kernel? 5. Which kernel modules were loaded on the measured system (USB module, keyboard controller, Ethernet driver etc...)? 6. What bios version was used? Was it the latest version? 7. Could you share the contents of /proc/interrupts file of the tested system? Thanks, Amir. -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a

RE: config_preempt_RT patch platform support question.

Amir Aharon — 2008-11-16T13:43:16

Hi Carsten, Good new - 2.6.24.7-rt21 works perfectly on my Xeon x5482 3.2GHz, could not reproduce any of the kernel panic I had. Thanks for the tip. -----Original Message----- From: linux-rt-users-owner< at >vger.kernel.org [mailto:linux-rt-users-owner< at >vger.kernel.org] On Behalf Of Amir Aharon Sent: Thursday, November 13, 2008 3:00 PM To: Carsten Emde Cc: RT-Users Subject: RE: config_preempt_RT patch platform support question. Here is the kernel dump: [] post_schedule_rt [] finish_task_switch [] __schedule [] check_preempt_curr_rt [] try_to_wake_up [] wake_up_process [] e1000_intr_msi [] handle_IRQ_event [] preempt_schedule [] do_irqd [] do_irqd [] kthread [] kthread [] kernel_thread_helper EIP: push_rt_task Node: IRQ-217[1433] exited with preempt_count 2 I have got this when I run my app on octa core xeon (Xeon x5482 3.2GHz). I'll run my app with linux-2.6.24.7-rt21 during the weekend and see where it goes... Thanks, Amir. -----Original Message----- From: Carsten Emde [mailto:Carsten.Emde< at >osadl.org

RE: config_preempt_RT patch platform support question.

Amir Aharon — 2008-11-13T13:00:15

Here is the kernel dump: [] post_schedule_rt [] finish_task_switch [] __schedule [] check_preempt_curr_rt [] try_to_wake_up [] wake_up_process [] e1000_intr_msi [] handle_IRQ_event [] preempt_schedule [] do_irqd [] do_irqd [] kthread [] kthread [] kernel_thread_helper EIP: push_rt_task Node: IRQ-217[1433] exited with preempt_count 2 I have got this when I run my app on octa core xeon (Xeon x5482 3.2GHz). I'll run my app with linux-2.6.24.7-rt21 during the weekend and see where it goes... Thanks, Amir. -----Original Message----- From: Carsten Emde [mailto:Carsten.Emde< at >osadl.org] Sent: Thursday, November 13, 2008 9:42 AM To: Amir Aharon Cc: RT-Users Subject: Re: config_preempt_RT patch platform support question. Amir, The PREEMPT_RT patches for kernel 2.6.26 are still under development. Please post the various kernel messages you are observing. BTW: The most recent 2.6.26 patch is 2.6.26.6-rt11. For the time being, linux-2.6.24.7-rt21 is believed to be stable and ready for production. This version

Strange device behavior on SMP system with CONFIG_PREEMPT_RT patch

Denis Borisevich — 2008-11-13T10:20:17

Hi everybody, I noticed a strange behavior of my PCI->RS422 adapter (MOXA CP-132 using mxer kernel module) on SMP x86 system with kernel 2.6.26.6-rt11. This PCI board has 2 RS422/RS485 ports. When I try to send cyclicaly some ammount of data more than 16 bytes (it's internal buffer is 16 bytes) from the second port to the first the data is lost at random iteration. If I try to send from first port to the second everything works fine no matter which ammount of data I write to the port. It seems that interrupts get lost. Can somebody help me with this problem? What can cause such a behavior? Thanks in advance, Denis -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html

Re: config_preempt_RT patch platform support question.

Carsten Emde — 2008-11-13T07:41:49

Amir, The PREEMPT_RT patches for kernel 2.6.26 are still under development. Please post the various kernel messages you are observing. BTW: The most recent 2.6.26 patch is 2.6.26.6-rt11. For the time being, linux-2.6.24.7-rt21 is believed to be stable and ready for production. This version has been tested heavily under x86 (slow/fast machines, single/dual/quad core, 32/64 bit) and under various PPC platforms. It has been tested less heavily on ARMs and MIPSes. There are no known issues on x86 platforms. Would you mind to run 2.6.24.7-rt21 on your machines? Carsten. -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo< at >vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html