Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1509677

Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling feature

From Thomas Gleixner <tglx@linutronix.de>
Newsgroups linux.kernel
Subject Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling feature
Date 2016-10-26 20:20 +0200
Message-ID <swAXU-7td-11@gated-at.bofh.it> (permalink)
References (1 earlier) <sutHc-52m-25@gated-at.bofh.it> <swu66-2w2-35@gated-at.bofh.it> <swuz7-2VL-15@gated-at.bofh.it> <swuz7-2VL-13@gated-at.bofh.it> <swAbw-6Tj-31@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


[Multipart message — attachments visible in raw view] - view raw

On Wed, 26 Oct 2016, Tim Chen wrote:
> On Wed, 2016-10-26 at 13:24 +0200, Thomas Gleixner wrote:
> > > There were reservations on the multi-socket case of ITMT, maybe it would
> > > help to spell those out in great detail here. That is, have the comment
> > > explain the policy instead of simply stating what the code does (which
> > > is always bad comment policy, you can read the code just fine).
> > What is the objection for multi sockets? If it improves the behaviour then
> > why would this be a bad thing for multi sockets?
> 
> For multi-socket (server system), it is much more likely that they will
> have multiple cpus in a socket busy and not run in turbo mode. So the extra
> work in migrating the workload to the one with extra headroom will
> not make use of those headroom in that scenario.  I will update the comment
> to reflect this policy.

So on a single socket server system the extra work does not matter, right?
Don't tell me that single socket server systems are irrelevant. Intel is
actively promoting single socket CPUs, like XEON D, for high densitiy
servers...

Instead of handwaving arguments I prefer a proper analysis of what the
overhead is and why it is not a good thing for loaded servers in general.

Then instead of slapping half baken heuristics into the code, we should sit
down and think a bit harder about it.

Thanks,

	tglx

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH v6 0/9] Support Intel® Turbo Boost Max Technology 3.0 Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
  [PATCH v6 8/9] acpi: bus: Set _OSC for diverse core support Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
  [PATCH v6 2/9] x86/topology: Provide topology_num_packages() Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
  [PATCH v6 4/9] x86: Enable Intel Turbo Boost Max Technology 3.0 Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
  [PATCH v6 3/9] x86/topology: Define x86's arch_update_cpu_topology Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
  [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling feature Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
    Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Thomas Gleixner <tglx@linutronix.de> - 2016-10-26 13:00 +0200
      Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-26 20:10 +0200
        Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Thomas Gleixner <tglx@linutronix.de> - 2016-10-26 20:20 +0200
          Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-26 21:40 +0200
    Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Thomas Gleixner <tglx@linutronix.de> - 2016-10-26 13:00 +0200
      Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Thomas Gleixner <tglx@linutronix.de> - 2016-10-26 13:30 +0200
        Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-26 19:30 +0200
          Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Thomas Gleixner <tglx@linutronix.de> - 2016-10-26 20:20 +0200
            Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-27 21:40 +0200
      Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling feature Peter Zijlstra <peterz@infradead.org> - 2016-10-26 13:30 +0200
      Re: [PATCH v6 5/9] x86/sysctl: Add sysctl for ITMT scheduling  feature Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-26 20:10 +0200
  [PATCH v6 1/9] sched: Extend scheduler's asym packing Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
    Re: [PATCH v6 1/9] sched: Extend scheduler's asym packing Thomas Gleixner <tglx@linutronix.de> - 2016-10-26 12:40 +0200
      Re: [PATCH v6 1/9] sched: Extend scheduler's asym packing Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-26 20:20 +0200
        Re: [PATCH v6 1/9] sched: Extend scheduler's asym packing Thomas Gleixner <tglx@linutronix.de> - 2016-10-26 20:30 +0200
  [PATCH v6 7/9] acpi: bus: Enable HWP CPPC objects Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
  [PATCH v6 9/9] cpufreq: intel_pstate: Use CPPC to get max performance Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200
  [PATCH v6 6/9] x86/sched: Add SD_ASYM_PACKING flags to x86 ITMT CPU Tim Chen <tim.c.chen@linux.intel.com> - 2016-10-21 00:10 +0200

csiph-web