[IRQ] IRQ affinity not working properly?

2021-01-29 Thread Chris Friesen
Hi, I'm not subscribed to the list, please cc me on replies. I have a CentOS 7 linux system with 48 logical CPUs and a number of Intel NICs running the i40e driver. It was booted with irqaffinity=0-1,24-25 in the kernel boot args, resulting in /proc/irq/default_smp_affinity showing ",030

Re: IRQs in /proc/irq/* that aren't listed in /proc/interrupts?

2020-10-12 Thread Chris Friesen
On 10/12/2020 1:37 PM, Thomas Gleixner wrote: On Mon, Oct 12 2020 at 12:40, Chris Friesen wrote: On one of my X86-64 systems /proc/interrupts starts with the following interrupts (per-cpu info snipped): 0: IR-IO-APIC-edge timer 4: IR-IO-APIC-edge serial 8

IRQs in /proc/irq/* that aren't listed in /proc/interrupts?

2020-10-12 Thread Chris Friesen
Hi, On one of my X86-64 systems /proc/interrupts starts with the following interrupts (per-cpu info snipped): 0: IR-IO-APIC-edge timer 4: IR-IO-APIC-edge serial 8: IR-IO-APIC-edge rtc0 9: IR-IO-APIC-fasteoi acpi 17: IR-IO-APIC-fasteoi eh

[RT] should pm_qos_resume_latency_us on one CPU affect latency on another?

2019-08-13 Thread Chris Friesen
Hi all, Just wondering if what I'm seeing is expected. I'm using the CentOS 7 RT kernel with boot args of "skew_tick=1 irqaffinity=0 rcu_nocbs=1-27 nohz_full=1-27" among others. Normally if I run cyclictest it sets /dev/cpu_dma_latency to zero. This gives worst-case latency around 6usec.

[RT] hit recently-fixed PREEMPT_RT CFS-bandwidth timer locking issue in the wild

2019-07-26 Thread Chris Friesen
Hi all, I thought people might be interested to hear that we recently hit the bug fixed by git commit c0ad4aa4d8 on multiple lab systems running the RHEL 7 "kernel-rt" kernel. (But I think other versions are at risk as well.) Interestingly, when the bug hit the system just hung completely.

Re: "swap_free: Bad swap file entry" and "BUG: Bad page map in process" but no swap configured

2017-11-21 Thread Chris Friesen
Huaitong Han 2016-10-12 0:02 GMT+08:00 Chris Friesen : On 10/08/2016 02:05 AM, Hillf Danton wrote: On Friday, October 07, 2016 5:01 AM Chris Friesen I have Linux host running as a kvm hypervisor. It's running CentOS. (So the kernel is based on 3.10 but with loads of stuff backport

Re: [Qemu-devel] kvm bug in __rmap_clear_dirty during live migration

2017-02-24 Thread Chris Friesen
On 02/23/2017 08:23 PM, Herongguang (Stephen) wrote: On 2017/2/22 22:43, Paolo Bonzini wrote: Hopefully Gaohuai and Rongguang can help with this too. Paolo . Yes, we are looking into and testing this. I think this can result in any memory corruption, if VM1 writes its PML buffer into VM2

Re: "swap_free: Bad swap file entry" and "BUG: Bad page map in process" but no swap configured

2016-10-11 Thread Chris Friesen
On 10/08/2016 02:05 AM, Hillf Danton wrote: On Friday, October 07, 2016 5:01 AM Chris Friesen I have Linux host running as a kvm hypervisor. It's running CentOS. (So the kernel is based on 3.10 but with loads of stuff backported by RedHat.) I realize this is not a mainline kernel, but

"swap_free: Bad swap file entry" and "BUG: Bad page map in process" but no swap configured

2016-10-06 Thread Chris Friesen
I have Linux host running as a kvm hypervisor. It's running CentOS. (So the kernel is based on 3.10 but with loads of stuff backported by RedHat.) I realize this is not a mainline kernel, but I was wondering if anyone is aware of similar issues that had been fixed in mainline. When doing

help? usage of indirect per-cpu variables

2016-09-27 Thread Chris Friesen
Hi, I'm trying to wrap my head around indirect percpu variables, and I'm hoping someone can school me on how they work. For example, in mm/slub.c we have "struct kmem_cache *s". s->cpu_slab is a per-cpu variable, so we access it with something like: c = raw_cpu_ptr(s->cpu_slab);

why would we be spending a whole msec running just the cascade() function?

2016-06-24 Thread Chris Friesen
Hi, I'm trying to get an idea why we would be spending a whole millisecond running the cascade() routine. This is on a CentOS 7 kernel (config modified as per below) so feel free to send me off to the other support fora if you like. Running cyclictest gave the following: cyclicte-29932 14

[tip:sched/core] sched/cputime: Fix steal_account_process_tick() to always return jiffies

2016-03-08 Thread tip-bot for Chris Friesen
Commit-ID: f9c904b7613b8b4c85b10cd6b33ad41b2843fa9d Gitweb: http://git.kernel.org/tip/f9c904b7613b8b4c85b10cd6b33ad41b2843fa9d Author: Chris Friesen AuthorDate: Sat, 5 Mar 2016 23:18:48 -0600 Committer: Ingo Molnar CommitDate: Tue, 8 Mar 2016 12:24:56 +0100 sched/cputime: Fix

[PATCH v2] sched/cputime: steal_account_process_tick() should return jiffies

2016-03-05 Thread Chris Friesen
time and only account it once it's worth a jiffy. (Thanks to Frederic Weisbecker for suggestions to fix a bug in my first version of the patch.) Signed-off-by: Chris Friesen --- kernel/sched/cputime.c | 14 +++--- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/kernel/

Re: [PATCH] steal_account_process_tick() should return jiffies

2016-03-05 Thread Chris Friesen
On 03/05/2016 07:19 AM, Frederic Weisbecker wrote: On Sat, Mar 05, 2016 at 11:27:01AM +0100, Thomas Gleixner wrote: Chris, On Fri, 4 Mar 2016, Chris Friesen wrote: First of all the subject line should contain a subsystem prefix, i.e. "sched/cputime:" The

[PATCH] steal_account_process_tick() should return jiffies

2016-03-04 Thread Chris Friesen
to change steal_account_process_tick() to always return jiffies. If CONFIG_VIRT_CPU_ACCOUNTING_GEN is not enabled then this is a no-op. As far as I can tell this bug has been present since commit dee08a72. Signed-off-by: Chris Friesen --- kernel/sched/cputime.c | 2 +- 1 file changed, 1

Re: question about logic of steal_account_process_tick() ?

2016-03-04 Thread Chris Friesen
On 03/04/2016 01:51 PM, Chris Friesen wrote: The thing is, steal_account_process_tick() returns units of cputime, which I think is nanoseconds on x86_64. So if we have a tiny amount of stolen time it seems like that will prevent a whole tick from being accounted into user/system/idle. I feel

question about logic of steal_account_process_tick() ?

2016-03-04 Thread Chris Friesen
I'm trying to wrap my head around how steal_account_process_tick() interacts with account_process_tick(). Suppose we have CONFIG_VIRT_CPU_ACCOUNTING_GEN=y and CONFIG_NO_HZ_IDLE, with a cpu hog on cpu0 to prevent it going idle. As I understand it, account_process_tick() will be called once per

Re: weird /proc/stat output with newer (4.1, 4.2) kernels in kvm guest on 3.10 host?

2016-03-03 Thread Chris Friesen
On 03/02/2016 11:12 AM, Chris Friesen wrote: I'm running a 3.10-based host with libvirt 1.2.12 and qemu 2.2. Running a Fedora23 cloud image as a guest, the "cpu" lines in /proc/stat seem to be hardly changing: [fedora@fedora23 boot]$ uptime 17:01:50 up 44 min, 1 user, loa

weird /proc/stat output with newer (4.1, 4.2) kernels in kvm guest on 3.10 host?

2016-03-02 Thread Chris Friesen
Hi, I'm running a 3.10-based host with libvirt 1.2.12 and qemu 2.2. Running a Fedora23 cloud image as a guest, the "cpu" lines in /proc/stat seem to be hardly changing: [fedora@fedora23 boot]$ uptime 17:01:50 up 44 min, 1 user, load average: 3.00, 2.99, 2.79 [fedora@fedora23 boot]$ grep c

Re: question about cpusets vs sched_setaffinity()

2015-12-11 Thread Chris Friesen
On 12/11/2015 04:15 PM, Jason Baron wrote: On 12/10/2015 04:30 PM, Chris Friesen wrote: If I put a task into a cpuset and then call sched_setaffinity() on it, it will be affined to the intersection of the two sets of cpus. (Those specified on the set, and those specified in the syscall

question about cpusets vs sched_setaffinity()

2015-12-10 Thread Chris Friesen
Hi, I've got a question about the interaction between cpusets and sched_setaffinity(). If I put a task into a cpuset and then call sched_setaffinity() on it, it will be affined to the intersection of the two sets of cpus. (Those specified on the set, and those specified in the syscall.) Ho

weird interaction between kvm and NO_HZ_FULL?

2015-03-20 Thread Chris Friesen
Hi, I'm running 3.10 (yeah, I know) and I'm playing with CONFIG_NO_HZ_FULL. I'm getting a strange result where some CPUs are able to turn off local timer interrupts and others aren't. Is there a known interaction between kvm-based VMs and CONFIG_NO_HZ_FULL? Background: I've got an x86-64 s

Re: absurdly high "optimal_io_size" on Seagate SAS disk

2014-11-07 Thread Chris Friesen
On 11/07/2014 01:17 PM, Martin K. Petersen wrote: I'd suggest trying /dev/sgN instead. That seems to work. Much appreciated. And it's now showing an "optimal_io_size" of 0, so I think the issue is dealt with. Thanks for all the help, it's been educational. :) Chris -- To unsubscribe from

Re: absurdly high "optimal_io_size" on Seagate SAS disk

2014-11-07 Thread Chris Friesen
On 11/07/2014 10:25 AM, Martin K. Petersen wrote: >>>>>> "Chris" == Chris Friesen writes: > > Chris, > > Chris> Also, I think it's wrong for filesystems and userspace to use it > Chris> for alignment. In E.4 and E.5 in the "sb

Re: absurdly high "optimal_io_size" on Seagate SAS disk

2014-11-07 Thread Chris Friesen
On 11/07/2014 11:42 AM, Martin K. Petersen wrote: "Martin" == Martin K Petersen writes: Martin> I know there was a bug open with Seagate. I assume it has been Martin> fixed in their latest firmware. Seagate confirms that this issue was fixed about a year ago. Will provide more data when I hav

Re: absurdly high "optimal_io_size" on Seagate SAS disk

2014-11-06 Thread Chris Friesen
On 11/06/2014 07:56 PM, Martin K. Petersen wrote: "Chris" == Chris Friesen writes: Chris, Chris> For a RAID card I expect it would be related to chunk size or Chris> stripe width or something...but even then I would expect to be Chris> able to cap it at 100MB or so. O

Re: absurdly high "optimal_io_size" on Seagate SAS disk

2014-11-06 Thread Chris Friesen
On 11/06/2014 12:12 PM, Martin K. Petersen wrote: "Chris" == Chris Friesen writes: Chris> That'd work, but is it the best way to go? I mean, I found one Chris> report of a similar problem on an SSD (model number unknown). In Chris> that case it was a near-UINT_MAX va

Re: absurdly high "optimal_io_size" on Seagate SAS disk

2014-11-06 Thread Chris Friesen
On 11/06/2014 11:34 AM, Martin K. Petersen wrote: "Chris" == Chris Friesen writes: Chris> Perhaps the ST900MM0026 should be blacklisted as well? Sure. I'll widen the net a bit for that Seagate model. That'd work, but is it the best way to go? I mean, I found o

Re: absurdly high "optimal_io_size" on Seagate SAS disk

2014-11-06 Thread Chris Friesen
On 11/06/2014 10:47 AM, Chris Friesen wrote: Hi, I'm running a modified 3.4-stable on relatively recent X86 server-class hardware. I recently installed a Seagate ST900MM0026 (900GB 2.5in 10K SAS drive) and it's reporting a value of 4294966784 for optimal_io_size. The other param

absurdly high "optimal_io_size" on Seagate SAS disk

2014-11-06 Thread Chris Friesen
Hi, I'm running a modified 3.4-stable on relatively recent X86 server-class hardware. I recently installed a Seagate ST900MM0026 (900GB 2.5in 10K SAS drive) and it's reporting a value of 4294966784 for optimal_io_size. The other parameters look normal though: /sys/block/sda/queue/hw_secto

Re: semantics of reader/writer semaphores in rt patch

2014-10-27 Thread Chris Friesen
On 10/25/2014 04:19 PM, Thomas Gleixner wrote: On Thu, 23 Oct 2014, Chris Friesen wrote: I recently noticed that when CONFIG_PREEMPT_RT_FULL is enabled we the semantics change. From "include/linux/rwsem_rt.h": * Note that the semantics are different from the usual * Linux r

semantics of reader/writer semaphores in rt patch?

2014-10-24 Thread Chris Friesen
I recently noticed that when CONFIG_PREEMPT_RT_FULL is enabled the semantics change. From "include/linux/rwsem_rt.h": * Note that the semantics are different from the usual * Linux rw-sems, in PREEMPT_RT mode we do not allow * multiple readers to hold the lock at once, we only allow * a re

[tracing] trying to make sense of trace output, can't figure out where time is going

2014-09-03 Thread Chris Friesen
ve disabled speedstep, CPU C states, and NMI on error. Is there a different tracer that would give more insight? The irqs-off or preemption-off tracers perhaps? Please CC me as I'm not subscribed to the list. Thanks, Chris Friesen -- To unsubscribe from this list: send the line "unsubs

Re: [PATCH/RFC] Re: recvmmsg() timeout behavior strangeness [RESEND]

2014-05-28 Thread Chris Friesen
On 05/28/2014 01:50 PM, 'Arnaldo Carvalho de Melo' wrote: What is being discussed here is how to return the EFAULT that may happen _after_ datagram processing, be it interrupted by an EFAULT, signal, or plain returning all that was requested, with no errors. This EFAULT _after_ datagram process

[bug?] can't unmount filesystem after running "exportfs -u" -- dangling refcount?

2014-03-21 Thread Chris Friesen
I'm seeing some odd behaviour on an NFS server. We're running 3.4.82 and have a drbd-replicated filesystem (active/standby) that is exported via NFS (explicitly v3, and /proc/fs/nfsd is mounted). On a server switchover we take down the NFS server IP address, run "exportfs -u" on the exports

Re: mmap vs fs cache

2013-03-08 Thread Chris Friesen
On 03/08/2013 09:00 AM, Howard Chu wrote: First obvious conclusion - kswapd is being too aggressive. When free memory hits the low watermark, the reclaim shrinks slapd down from 25GB to 18-19GB, while the page cache still contains ~7GB of unmapped pages. Ideally I'd like a tuning knob so I can s

Re: mmap vs fs cache

2013-03-08 Thread Chris Friesen
On 03/08/2013 03:40 AM, Howard Chu wrote: There is no way that a process that is accessing only 30GB of a mmap should be able to fill up 32GB of RAM. There's nothing else running on the machine, I've killed or suspended everything else in userland besides a couple shells running top and vmstat.

Re: [GIT PULL] Load keys from signed PE binaries

2013-02-28 Thread Chris Friesen
On 02/28/2013 01:57 AM, Florian Weimer wrote: In any case, there's another reading of the UEFI Secure Boot requirements: you may run any code you wish after calling ExitBootServices(). That could be an unsigned, traditional GRUB. But this will not generally address the issue of dual-booting Wi

Re: [GIT PULL] Load keys from signed PE binaries

2013-02-27 Thread Chris Friesen
On 02/27/2013 11:59 AM, Theodore Ts'o wrote: On Wed, Feb 27, 2013 at 11:36:09AM -0600, Chris Friesen wrote: ... At this point you've got a running infected Win8 install that is running on Secure Boot hardware but is actually running malware. Admittedly this would be tricky to do rel

Re: [GIT PULL] Load keys from signed PE binaries

2013-02-27 Thread Chris Friesen
On 02/27/2013 09:24 AM, Theodore Ts'o wrote: On Tue, Feb 26, 2013 at 11:54:51AM -0500, Peter Jones wrote: No, no, no. Quit saying nobody knows. We've got a pretty good idea - we've got a contract with them, and it says they provide the signing service, and under circumstances where the thing b

Re: [GIT PULL] Load keys from signed PE binaries

2013-02-26 Thread Chris Friesen
On 02/26/2013 03:40 PM, Florian Weimer wrote: * Chris Friesen: On 02/25/2013 10:14 AM, Matthew Garrett wrote: Windows 8 will not load unsigned drivers if Secure Boot is enabled. For reference: http://msdn.microsoft.com/en-us/library/windows/desktop/hh848062%28v=vs.85%29.aspx Thanks. Do

Re: [GIT PULL] Load keys from signed PE binaries

2013-02-25 Thread Chris Friesen
On 02/25/2013 10:14 AM, Matthew Garrett wrote: On Mon, Feb 25, 2013 at 04:50:50PM +0100, Florian Weimer wrote: * Matthew Garrett: On Mon, Feb 25, 2013 at 03:46:14PM +0100, Florian Weimer wrote: You could just drop the requirement that ring 0 code must be signed. I don't think Windows 8 enfor

Re: Read I/O starvation with writeback RAID controller

2013-02-22 Thread Chris Friesen
On 02/22/2013 02:35 PM, Jan Engelhardt wrote: On Friday 2013-02-22 20:28, Martin Svec wrote: Yes, I've already tried the ROW scheduler. It helped for some low iodepths depending on quantum settings but generally didn't solve the problem. I think the key issue is that none of the schedulers can

Re: IPsec AH use of ahash

2013-01-21 Thread Chris Friesen
On 01/21/2013 09:31 AM, Tom St Denis wrote: - Original Message - From: "Steven Rostedt" To: "Tom St Denis" Cc: "David Dillow", "Borislav Petkov", linux-kernel@vger.kernel.org, net...@vger.kernel.org Sent: Monday, 21 January, 2013 10:28:33 AM Subject: Re: IPsec AH use of ahash When I sen

Re: [sqlite] light weight write barriers

2012-11-15 Thread Chris Friesen
On 11/15/2012 11:06 AM, Ryan Johnson wrote: The easiest way to implement this fsync would involve three things: 1. Schedule writes for all dirty pages in the fs cache that belong to the affected file, wait for the device to report success, issue a cache flush to the device (or request ordering c

Re: scsi target, likely GPL violation

2012-11-07 Thread Chris Friesen
On 11/07/2012 07:02 PM, Jon Mason wrote: I'm not a lawyer, nor do I play one on TV, but if I understand the GPL correctly, RTS only needs to provide the relevant source to their customers upon request. Not quite. Assuming the GPL applies, and that they have modified the code, then they must e

Re: [RFC] Second attempt at kernel secure boot support

2012-11-06 Thread Chris Friesen
On 11/06/2012 01:56 AM, Florian Weimer wrote: Personally, I think the only way out of this mess is to teach users how to disable Secure Boot. If you're going to go that far, why not just get them to install a RedHat (or SuSE, or Ubuntu, or whoever) key and use that instead? Secure boot does

Re: [RFC] Second attempt at kernel secure boot support

2012-11-05 Thread Chris Friesen
On 11/05/2012 09:31 AM, Jiri Kosina wrote: I had a naive idea of just putting in-kernel verification of a complete ELF binary passed to kernel by userspace, and if the signature matches, jumping to it. Would work for elf-x86_64 nicely I guess, but we'd lose a lot of other functionality currently

Re: Fwd: Nice processes prevent frequency increases - possible scheduler regression (known good in 2.6.35)

2012-11-05 Thread Chris Friesen
On 11/03/2012 09:40 AM, Michal Zatloukal wrote: On Sat, Nov 3, 2012 at 12:48 PM, Mike Galbraith wrote: On Sat, 2012-11-03 at 04:33 -0700, Mike Galbraith wrote: On Fri, 2012-11-02 at 21:09 +0100, Michal Zatloukal wrote: Your nice 19 tasks receiving 'too much' CPU when there are other runnabl

Re: [RFC] Second attempt at kernel secure boot support

2012-11-02 Thread Chris Friesen
On 11/02/2012 04:03 PM, Eric W. Biederman wrote: Matthew Garrett writes: On Fri, Nov 02, 2012 at 01:49:25AM -0700, Eric W. Biederman wrote: When the goal is to secure Linux I don't see how any of this helps. Windows 8 compromises are already available so if we turn most of these arguments ar

Re: [RFC] Second attempt at kernel secure boot support

2012-11-02 Thread Chris Friesen
On 11/02/2012 09:48 AM, Vivek Goyal wrote: On Thu, Nov 01, 2012 at 03:02:25PM -0600, Chris Friesen wrote: With secure boot enabled, then the kernel should refuse to let an unsigned kexec load new images, and kexec itself should refuse to load unsigned images. Yep, good in theory. Now that

Re: [RFC] Second attempt at kernel secure boot support

2012-11-01 Thread Chris Friesen
On 11/01/2012 02:27 PM, Pavel Machek wrote: Could someone write down exact requirements for Linux kernel to be signed by Microsoft? Because thats apparently what you want, and I don't think crippling kexec/suspend is enough. As I understand it, the kernel won't be signed by Microsoft. Rathe

Re: [RFC] Second attempt at kernel secure boot support

2012-10-31 Thread Chris Friesen
On 10/31/2012 02:14 PM, Oliver Neukum wrote: On Wednesday 31 October 2012 17:39:19 Alan Cox wrote: On Wed, 31 Oct 2012 17:17:43 + Matthew Garrett wrote: On Wed, Oct 31, 2012 at 05:21:21PM +, Alan Cox wrote: On Wed, 31 Oct 2012 17:10:48 + Matthew Garrett wrote: The kernel is sig

Re: Linix-3.6.3 sda, sdb drives in reverse order (with a USB 2.0 drives and a monolithic kernel configuration)

2012-10-26 Thread Chris Friesen
On 10/26/2012 01:43 PM, Wallak wrote: Chris Friesen wrote: On 10/25/2012 04:49 PM, Wallak wrote: I've a very annoying behavior with the linux-3.6.x kernels release, and a monolithic configuration. The USB 2.0 drives are mapped first with /dev/sda, /dev/sdb... devices, and than the SATA

Re: Linix-3.6.3 sda, sdb drives in reverse order (with a USB 2.0 drives and a monolithic kernel configuration)

2012-10-25 Thread Chris Friesen
On 10/25/2012 04:49 PM, Wallak wrote: I've a very annoying behavior with the linux-3.6.x kernels release, and a monolithic configuration. The USB 2.0 drives are mapped first with /dev/sda, /dev/sdb... devices, and than the SATA AHCI drives come after. This is out of order with the BIOS configurat

Re: Local DoS through write heavy I/O on CFQ & Deadline

2012-10-18 Thread Chris Friesen
On 10/18/2012 03:28 PM, Jan Kara wrote: Yeah, ionice has its limitations. The problem is that all buffered writes happen just into memory (so completely independently of ionice settings). Subsequent writing of dirty memory to disk happens using flusher thread which is a kernel process and it

Help! How to get arguments to last function call of blocked task?

2012-09-16 Thread Chris Friesen
t;+24>:push %r12 0x815a9eba <+26>:mov$0x0,%al 0x815a9ebc <+28>:push %rbx I'm interested in the r12 value going into schedule(). How do I identify it in my stack above? Thanks, Chris -- Chris Friesen Software Designer 3500 Carling Av

Re: Drop support for x86-32

2012-08-24 Thread Chris Friesen
On 08/24/2012 11:05 AM, wbrana wrote: On 8/24/12, Martin Nybo Andersen wrote: What I'd hate even more is rendering my old working hardware useless by removing x86-32 support from the kernel. To reason the removal by saying "Microsoft plans to do it" just makes me go bonkers... Your old hardw

Re: Drop support for x86-32

2012-08-24 Thread Chris Friesen
On 08/24/2012 10:14 AM, Martin Nybo Andersen wrote: (And I'm still not sure why one would run 32-bit applications on a 64-bit architecture...) There are several architectures (powerpc comes to mind) where 32-bit userspace on 64-bit kernel is the norm because it offers performance advantages d

Re: [discussion]sched: a rough proposal to enable power saving in scheduler

2012-08-17 Thread Chris Friesen
On 08/17/2012 01:50 PM, Matthew Garrett wrote: On Fri, Aug 17, 2012 at 01:45:09PM -0600, Chris Friesen wrote: On 08/17/2012 12:47 PM, Matthew Garrett wrote: The datasheet for the Xeon E5 (my variant at least) says it doesn't do C7 so never powers down the LLC. However, as you said ea

Re: [discussion]sched: a rough proposal to enable power saving in scheduler

2012-08-17 Thread Chris Friesen
On 08/17/2012 12:47 PM, Matthew Garrett wrote: On Fri, Aug 17, 2012 at 11:44:03AM -0700, Arjan van de Ven wrote: On 8/17/2012 11:41 AM, Matthew Garrett wrote: On Thu, Aug 16, 2012 at 07:01:25AM -0700, Arjan van de Ven wrote: this is ... a dubiously general statement. for good power, at least

Re: allocation of contiguous memory in kernel mode

2012-08-13 Thread Chris Friesen
On 08/09/2012 09:47 AM, J.Hwan Kim wrote: Hi, everyone I wish to alloc physically contiguous memory in kernel mode. I used the __get_free_pages(gfp_mask, order) function, but it fails occasionary. I guess it is caused by memory fragmentation. My code is kernel driver module which cannot use boot

Re: mellanox mlx4_core and SR-IOV

2012-08-10 Thread Chris Friesen
On 08/03/2012 02:33 AM, Lukas Hejtmanek wrote: I also tried OFED package from Mellanox which seems to have better SR-IOV support (at least mlx4_ib does not complain that SR-IOV is not supported). However, it does not work when SR-IOV enabled: Last I heard they were not officially providing supp

Re: Huge performance degradation for UDP between 2.4.17 and 2.6

2012-08-02 Thread Chris Friesen
On 08/02/2012 06:27 AM, leroy christophe wrote: Hi I'm having a big issue with UDP. Using a powerpc board (MPC860). With our board running kernel 2.4.17, I'm able to send 16 voice packets (UDP, 96 bytes per packet) in 11 seconds. With the same board running either Kernel 2.6.35.14 or Kerne

Re: MODULE_LICENSE("GPL")??

2012-08-02 Thread Chris Friesen
On 08/02/2012 06:19 AM, Mark Hounschell wrote: This particular driver does in fact build cleanly after changing the GPL to PROPRIETARY. I haven't actually purchased the product yet so am unable to load it, but can I assume that if I don't have a build issue when specifying PROPRIETARY, that

Re: MODULE_LICENSE("GPL")??

2012-08-01 Thread Chris Friesen
On 08/01/2012 03:24 PM, Mark Hounschell wrote: What would happen if NVIDIA used this define in their proprietary driver? I ask because I am currently in a situation where I believe I may be about to use a product that may be doing this very thing. We had to sign a license agreement to get the ker

Re: CFS vs. cpufreq/cstates vs. latency

2012-07-24 Thread Chris Friesen
On 07/17/2012 05:23 PM, Rik van Riel wrote: While tracking down a latency issue with communication between KVM guests, we ran into a very interesting issue, an interplay of CFS and power saving code. About 3/4 of the 230us latency came from CPUs waking up out of C-states. Disabling C states redu

Re: CFS vs. cpufreq/cstates vs. latency

2012-07-17 Thread Chris Friesen
On 07/17/2012 08:23 AM, Rik van Riel wrote: > Specifically, waking up some process requires that the CPU > which is running the wakeup is already in C0 state. If the > CPU on which the to-be-woken task ran last is in a deep C > state, it may make sense to simply run the woken up task > on the loca

Re: AF_BUS socket address family

2012-07-06 Thread Chris Friesen
On Saturday 2012-06-30 01:12, Vincent Sanders wrote: Firstly it is intended is an interprocess mechanism and not to rely on a configured IP system, indeed one of its primary usages is to provide mechanism for various tools to set up IP networking. Using IP as a localhost IPC is not uncommon (ind

Re: [PATCH] USB: mark USB drivers as being GPL only

2008-02-05 Thread Chris Friesen
David Newall wrote: That being said, a module can be written such that it only dynamically links with the kernel. Ndiswrapper is an example of how this can be done: None of the drivers that work under ndiswrapper make any direct use of the kernel, not in any way, indeed a wrapper could be wri

Re: [PATCH] USB: mark USB drivers as being GPL only

2008-02-05 Thread Chris Friesen
Marcel Holtmann wrote: If the developers say that this symbol can only be used in GPL code (and with EXPORT_SYMBOL_GPL it is quite clear) then you have to obey to that license or don't use this symbol at all. If you use that symbol inside non-GPL (meaning you link at runtime) then you are in vi

Re: ndiswrapper and GPL-only symbols redux

2008-01-31 Thread Chris Friesen
David Newall wrote: This idea that some symbols may only be dynamically bound to GPL code is fallacy. As I understand it, the point of EXPORT_SYMBOL_GPL is not so much the technical restriction (as you say, the module can lie or the user can patch the kernel) but to indicate that the kernel d

Re: ndiswrapper and GPL-only symbols redux

2008-01-30 Thread Chris Friesen
Adrian Bunk wrote: IANAL, but I have serious doubts whether putting some glue layer between the GPL'ed code and the code with a not GPL compatible licence is really a legally effictive way of circumventing the GPL. It may depend on the details of the "code with a not GPL compatible licence".

Re: questions on NAPI processing latency and dropped network packets

2008-01-21 Thread Chris Friesen
Eric Dumazet wrote: Chris Friesen a écrit : I've done some further digging, and it appears that one of the problems we may be facing is very high instantaneous traffic rates. Instrumentation showed up to 222K packets/sec for short periods (at least 1.1 ms, possibly longer), althoug

Re: questions on NAPI processing latency and dropped network packets

2008-01-21 Thread Chris Friesen
Ben Greear wrote: Chris Friesen wrote: Is there anything else we can do to minimize the latency of network packet processing and avoid having to crank the rx ring size up so high? Why is it such a big deal to crank up the rx queue length? Seems like a perfectly normal way to handle bursts

Re: questions on NAPI processing latency and dropped network packets

2008-01-21 Thread Chris Friesen
I've done some further digging, and it appears that one of the problems we may be facing is very high instantaneous traffic rates. Instrumentation showed up to 222K packets/sec for short periods (at least 1.1 ms, possibly longer), although the long-term average is down around 14-16K packets/se

Re: Why is the kfree() argument const?

2008-01-18 Thread Chris Friesen
David Schwartz wrote: 2) The 'kfree' operation changes the logical state of the object pointed to, as the object goes from existent to non-existent. I don't think that kfree() itself changes the state of the object. It doesn't call a destructor or anything like that, so the object itself mu

Re: questions on NAPI processing latency and dropped network packets

2008-01-15 Thread Chris Friesen
Radoslaw Szkodzinski (AstralStorm) wrote: On Tue, 15 Jan 2008 08:47:07 -0600 "Chris Friesen" <[EMAIL PROTECTED]> wrote: Some of our hardware is not supported on mainline, so we need per-kernel version patches to even bring up the blade. The blades netboot via a jumbo-f

Re: questions on NAPI processing latency and dropped network packets

2008-01-15 Thread Chris Friesen
Jarek Poplawski wrote: IMHO, checking this with a current stable, which probably you are going to do some day, anyway, should be 100% acceptable: giving some input to netdev, while still working for yourself. While I would love to do this, it's not that simple. Some of our hardware is not sup

Re: questions on NAPI processing latency and dropped network packets

2008-01-14 Thread Chris Friesen
Eric Dumazet wrote: Chris Friesen a écrit : Based on the profiling information we're spending time in sctp_endpoint_lookup_assoc() which doesn't actually use hashes, so I can't see how the hash would be related. I'm pretty new to SCTP though, so I may be missing someth

Re: questions on NAPI processing latency and dropped network packets

2008-01-14 Thread Chris Friesen
Eric Dumazet wrote: Chris Friesen a écrit : Based on profiling and instrumentation it seems like the cost of sctp_endpoint_lookup_assoc() more than triples, which means that the amount of time that bottom halves are disabled in that function also triples. Any idea of the size of sctp hash

Re: questions on NAPI processing latency and dropped network packets

2008-01-14 Thread Chris Friesen
David Miller wrote: From: "Chris Friesen" <[EMAIL PROTECTED]> Date: Fri, 11 Jan 2008 08:59:26 -0600 I'd love to work on newer kernels, but we have a commitment to our customers to support multiple releases for a significant amount of time. And by asking here for peop

Re: questions on NAPI processing latency and dropped network packets

2008-01-14 Thread Chris Friesen
Ray Lee wrote: On Jan 10, 2008 9:24 AM, Chris Friesen <[EMAIL PROTECTED]> wrote: After a recent userspace app change, we've started seeing packets being dropped by the ethernet hardware (e1000, NAPI is enabled). The error/dropped/fifo counts are going up in ethtool: Can you r

Re: questions on NAPI processing latency and dropped network packets

2008-01-11 Thread Chris Friesen
David Miller wrote: You have to be kidding, coming here for help with a nearly 4 year old kernel. I figured it couldn't hurt to ask...if I can't ask the original authors, who else is there? I'd love to work on newer kernels, but we have a commitment to our customers to support multiple rel

Re: questions on NAPI processing latency and dropped network packets

2008-01-10 Thread Chris Friesen
James Chapman wrote: What's changed in your application? Any real-time threads in there? From the top output below, looks like SigtranServices is consuming all your CPU... There are two cpus, and SigtranServices is multithreaded with many threads. Most of these threads are affined to cpu0

Re: questions on NAPI processing latency and dropped network packets

2008-01-10 Thread Chris Friesen
Kok, Auke wrote: You're using 2.6.10... you can always replace the e1000 module with the out-of-tree version from e1000.sf.net, this might help a bit - the version in the 2.6.10 kernel is very very old. Do you have any reason to believe this would improve things? It seems like the problem l

questions on NAPI processing latency and dropped network packets

2008-01-10 Thread Chris Friesen
Hi all, I've got an issue that's popped up with a deployed system running 2.6.10. I'm looking for some help figuring out why incoming network packets aren't being processed fast enough. After a recent userspace app change, we've started seeing packets being dropped by the ethernet hardware

Re: [JANITOR PROPOSAL] Switch ioctl functions to ->unlocked_ioctl

2008-01-09 Thread Chris Friesen
Alasdair G Kergon wrote: On Wed, Jan 09, 2008 at 11:46:03PM +0100, Andi Kleen wrote: struct inode *inode = file->f_dentry->d_inode; And oops if that's not defined? Isn't this basically identical to what was being passed in to .ioctl()? Chris -- To unsubscribe from this list: send the lin

looking for help with mpt fusion driver log: "FAULT code = 1804h"

2008-01-08 Thread Chris Friesen
Hi all, We're seeing the following on startup: Fusion MPT base driver 3.02.55 Copyright (c) 1999-2005 LSI Logic Corporation Fusion MPT SAS Host driver 3.02.55 mptbase: Initiating ioc0 bringup mptbase: ioc0: WARNING - IOC is in FAULT state!!! FAULT code = 1804h mptbase: ioc0: ERROR -

Re: "ip neigh show" not showing arp cache entries?

2007-12-18 Thread Chris Friesen
Herbert Xu wrote: Chris Friesen <[EMAIL PROTECTED]> wrote: However, if I specifically try to print out one of the missing entries, it shows up: [EMAIL PROTECTED]:/root> /tmp/ip neigh show 192.168.24.81 192.168.24.81 dev bond2 lladdr 00:01:af:14:e9:8a REACHABLE What about

Re: "ip neigh show" not showing arp cache entries?

2007-12-17 Thread Chris Friesen
Patrick McHardy wrote: From a kernel perspective there are only complete dumps, the filtering is done by iproute. So the fact that it shows them when querying specifically implies there is a bug in the iproute neighbour filter. Does it work if you omit "all" from the ip neigh show command? Om

Re: "ip neigh show" not showing arp cache entries?

2007-12-17 Thread Chris Friesen
YOSHIFUJI Hideaki / 吉藤英明 wrote: In article <[EMAIL PROTECTED]> (at Wed, 12 Dec 2007 15:57:08 -0600), "Chris Friesen" <[EMAIL PROTECTED]> says: You may try other versions of this command http://devresources.linux-foundation.org/dev/iproute2/download/ They appear to

Re: "ip neigh show" not showing arp cache entries?

2007-12-12 Thread Chris Friesen
Eric Dumazet wrote: And what is the version of "ip" command you have on this machine ? ip -V iproute2-ss051107 You may try other versions of this command http://devresources.linux-foundation.org/dev/iproute2/download/ They appear to be numbered by kernel version, and the above version is

Re: "ip neigh show" not showing arp cache entries?

2007-12-12 Thread Chris Friesen
Eric Dumazet wrote: Chris Friesen a écrit : Is this expected behaviour? Probably not... Still a 2.6.14 kernel ? Yep. Embedded hardware, so I'm unable to test with a more recent kernel. Could you send the result of : strace ip neigh show I've attached two strace runs, one of

Re: "ip neigh show" not showing arp cache entries?

2007-12-12 Thread Chris Friesen
I retested it on an x86 machine and am seeing similar problems. First, "arp" gives the arp table as expected: [EMAIL PROTECTED]:/tftpboot/cnp/0-0-5-0/0-0-5-0> arp -n Address HWtype HWaddress Flags Mask Iface 172.24.0.9 ether 00:03:CC:51:06:5E C

Re: "ip neigh show" not showing arp cache entries?

2007-12-10 Thread Chris Friesen
Chris Friesen wrote: I'm seeing some strange behaviour on a 2.6.14 ppc64 system. If I run "ip neigh show" it prints out nothing, but if I run "arp" then I see the other nodes on the local network. [EMAIL PROTECTED]:/root> ip neigh show [EMAIL PROT

"ip neigh show" not showing arp cache entries?

2007-12-10 Thread Chris Friesen
I'm seeing some strange behaviour on a 2.6.14 ppc64 system. If I run "ip neigh show" it prints out nothing, but if I run "arp" then I see the other nodes on the local network. [EMAIL PROTECTED]:/root> ip neigh show [EMAIL PROTECTED]:/root> arp -n Address HWtype HWaddress

Re: New Address Family: Inter Process Networking (IPN)

2007-12-10 Thread Chris Friesen
David Miller wrote: The kernel supports much more than 32 groups, see nlk->groups which is a bitmap which can be sized to arbitrary sizes. nlk->nl_groups is for backwards compatability only. netlink_change_ngroups() does the bitmap resizing when necessary. Thanks for the explanation. Given

Re: New Address Family: Inter Process Networking (IPN)

2007-12-06 Thread Chris Friesen
David Miller wrote: From: "Chris Friesen" <[EMAIL PROTECTED]> Date: Thu, 06 Dec 2007 14:36:54 -0600 One problem we ran into was that there are only 32 multicast groups per netlink protocol family. I'm pretty sure we've removed this limitation. As of 2.6.23 nl_

Re: New Address Family: Inter Process Networking (IPN)

2007-12-06 Thread Chris Friesen
Andi Kleen wrote: On Thu, Dec 06, 2007 at 05:02:40PM -0600, Chris Friesen wrote: I just reran on a 3.2GHZ P4 running 2.6.11 (Fedora Core 4). 42% latency increase. Sounds like something that should be looked into. I know of no principal reasons for that. For stream sockets, unix gives

  1   2   3   4   >