On Mon, Mar 7, 2016, at 14:28, Stuart Bennett wrote: > On 03/03/16 14:48, Henrique de Moraes Holschuh wrote: > > Argh. Is this a motherboard from a non-joke vendor ? If so, please > > open a support case and tell them you are hitting a "severe performance > > issue" that looks like the Xeon E5-v3 processor erratum HSE106, and that
> Motherboard is an Intel S2600WTT, which hopefully qualifies as non-joke. Intel's server motherboard division is usually good about fixing firmware, yes. As long as the board is still in active support status, at least. > As it happens there is a published update including microcode 0x36 and > the management engine changelog mentions various things relating to > power limiting. Before complaining to them I'll apply the update. Thanks. Let's hope it fixes the issue... > On the other hand the machine is four hours away, and there's an > argument to be made that OS updates shouldn't cause an immense > performance regression, regardless of BIOS version. Well, Intel listed it as an erratum, so it appears that they not only agree with you, but that their microcode division is actually acknowledging it as a problem. Let's hope this means they will design future microcode updates to avoid such issues... > > $ rgrep . /sys/devices/virtual/powercap ... > > $ rgrep . /sys/devices/virtual/thermal ... Thanks. Does anything change on that output when the processor misbehaves? If you capture the output of the above commands to a file while the processor is working on the old microcode, and then when it misbehaves on the new microcode, a "diff -u" on the two files should show any relevant changes... -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique de Moraes Holschuh <[email protected]>
