All of lore.kernel.org
 help / color / mirror / Atom feed
* [dpdk-dev] [Bug 450] wrong report for cycles/pkt or pps
@ 2020-04-14  7:36 bugzilla
  0 siblings, 0 replies; only message in thread
From: bugzilla @ 2020-04-14  7:36 UTC (permalink / raw)
  To: dev

https://bugs.dpdk.org/show_bug.cgi?id=450

            Bug ID: 450
           Summary: wrong report for cycles/pkt or pps
           Product: DPDK
           Version: 20.02
          Hardware: ARM
                OS: All
            Status: UNCONFIRMED
          Severity: normal
          Priority: Normal
         Component: testpmd
          Assignee: dev@dpdk.org
          Reporter: ogerlitz@mellanox.com
  Target Milestone: ---

According to doc/guides/prog_guide/profile_app.rst RTE_ARM_EAL_RDTSC_USE_PMU
has to be used over ARM CPU to get correct cyc/pkt report. This involves using
OOT (non upstream) kernel module which is problematic and reported to be
non-stable. 

Moreover, once the PMU configuration is set, the pps report done by testpmd is
wrong, so this doesn't really help, as one wants to see both pps and cyc/pkt.

If the PMU configuration is not set, indeed the cyc/pkt report is wrong, but
this is due to using a timer which is not running in the CPU frequency. If the
CPU frequency would be used we can get a correct cyc/pkt report without the PMU
config.

-- 
You are receiving this mail because:
You are the assignee for the bug.

^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2020-04-14  7:36 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-04-14  7:36 [dpdk-dev] [Bug 450] wrong report for cycles/pkt or pps bugzilla

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.