RE: x86_mce: mce_start uses number of phsical cores instead oflogical cores
From: Ming Lei
Date: Fri May 10 2013 - 16:10:21 EST
I used intel edac error injector and saw the same problem. I actually wrote down the core numbers and I saw mce got to 0-5 and 12-17, but not the others. I have 2 sockets, 24 logical cores. Below is the trace I put into mce code. The core number is after "#".
Ming
344 :344 #4 ** 802097241816 (207303152230.v1) (207303152334) 4294874599 :24::::: mce_start do_machine_check
345 :345 #16 ** 802097241876 (207303152404.v1) (207303152426) 4294874599 :12:16:1:4:4: mce_start do_machine_check
346 :346 #0 ** 802097241914 (207303152271.v1) (207303152343) 4294874599 :24::::: mce_start do_machine_check
347 :347 #1 * 802097242074 (207303152515.v1) (207303152599) 4294874599 :8:-4755801206503178081:256::: mce_no_way_out do_machine_check
348 :348 #13 * 802097242098 (207303152512.v1) (207303152552) 4294874599 :7::::: mce_no_way_out do_machine_check
349 :349 #3 * 802097242282 (207303152630.v1) (207303152679) 4294874599 :7::::: mce_no_way_out do_machine_check
350 :350 #14 ** 802097242342 (207303152452.v1) (207303152520) 4294874599 :12:16:1:4:4: mce_start do_machine_check
351 :351 #2 * 802097242366 (207303152458.v1) (207303152537) 4294874599 :8:-4755801206503178081:256::: mce_no_way_out do_machine_check
352 :352 #0 ** 802097242774 (207303152627.v1) (207303152676) 4294874599 :12:16:1:4:4: mce_start do_machine_check
353 :353 #12 ** 802097242838 (207303152829.v1) (207303152853) 4294874599 :24::::: mce_start do_machine_check
354 :354 #15 ** 802097242890 (207303152676.v1) (207303152707) 4294874599 :24::::: mce_start do_machine_check
355 :355 #4 ** 802097243056 (207303152747.v1) (207303152825) 4294874599 :12:16:1:4:4: mce_start do_machine_check
356 :356 #2 ** 802097243386 (207303152881.v1) (207303153006) 4294874599 :24::::: mce_start do_machine_check
357 :357 #17 ** 802097243546 (207303152953.v1) (207303153023) 4294874599 :24::::: mce_start do_machine_check
358 :358 #5 ** 802097243566 (207303152963.v1) (207303153041) 4294874599 :24::::: mce_start do_machine_check
359 :359 #15 ** 802097243922 (207303153107.v1) (207303153193) 4294874599 :12:21:1:9:9: mce_start do_machine_check
360 :360 #3 * 802097243994 (207303153342.v1) (207303153356) 4294874599 :8:-4755801206503178081:256::: mce_no_way_out do_machine_check
361 :361 #13 * 802097244074 (207303153175.v1) (207303153242) 4294874599 :8:-4755801206503178081:256::: mce_no_way_out do_machine_check
362 :362 #1 ** 802097244050 (207303153167.v1) (207303153229) 4294874599 :24::::: mce_start do_machine_check
363 :363 #12 ** 802097244174 (207303153212.v1) (207303153284) 4294874599 :12:22:1:9:9: mce_start do_machine_check
364 :364 #2 ** 802097244490 (207303153347.v1) (207303153419) 4294874599 :12:22:1:10:10: mce_start do_machine_check
365 :365 #1 ** 802097244746 (207303153452.v1) (207303153521) 4294874599 :12:22:1:10:10: mce_start do_machine_check
366 :366 #5 ** 802097244834 (207303153488.v1) (207303153558) 4294874599 :12:22:1:10:10: mce_start do_machine_check
367 :367 #17 ** 802097244902 (207303153645.v1) (207303153665) 4294874599 :12:22:1:10:10: mce_start do_machine_check
368 :368 #3 ** 802097245130 (207303153611.v1) (207303153680) 4294874599 :24::::: mce_start do_machine_check
369 :369 #13 ** 802097245302 (207303153681.v1) (207303153760) 4294874599 :24::::: mce_start do_machine_check
370 :370 #3 ** 802097245710 (207303153857.v1) (207303153979) 4294874599 :12:24:1:12:12: mce_start do_machine_check
371 :371 #13 ** 802097246234 (207303154072.v1) (207303154141) 4294874599 :12:24:1:12:12: mce_start do_machine_check
372 :372 #15 *** 802097246542 (207303154201.v1) (207303154283) 4294874599 :12:5:::: mce_start do_machine_check
373 :373 #3 *** 802097246614 (207303154539.v1) (207303154565) 4294874599 :12:11:::: mce_start do_machine_check
374 :374 #2 *** 802097246678 (207303154265.v1) (207303154331) 4294874599 :12:9:::: mce_start do_machine_check
375 :375 #13 *** 802097246794 (207303154313.v1) (207303154376) 4294874599 :12:12:::: mce_start do_machine_check
376 :376 #1 *** 802097246814 (207303154325.v1) (207303154388) 4294874599 :12:10:::: mce_start do_machine_check
377 :377 #0 *** 802097246898 (207303154350.v1) (207303154420) 4294874599 :12:4:::: mce_start do_machine_check
378 :378 #12 *** 802097246966 (207303154614.v1) (207303154640) 4294874599 :12:6:::: mce_start do_machine_check
379 :379 #4 *** 802097247044 (207303154416.v1) (207303154481) 4294874599 :12:3:::: mce_start do_machine_check
380 :380 #16 *** 802097247064 (207303154429.v1) (207303154494) 4294874599 :12:1:::: mce_start do_machine_check
381 :381 #17 *** 802097247226 (207303154669.v1) (207303154696) 4294874599 :12:7:::: mce_start do_machine_check
382 :382 #14 *** 802097247250 (207303154495.v1) (207303154575) 4294874599 :12:2:::: mce_start do_machine_check
383 :383 #5 *** 802097247574 (207303154632.v1) (207303154666) 4294874599 :12:8:::: mce_start do_machine_check
384 :384 #16 **** 802097247812 (207303154735.v1) (207303154768) 4294874599 :12:1:::: mce_start do_machine_check
385 :385 #16 *** 802097258184 (207303159067.v1) (207303159094) 4294874599 :8:-4755801206503178081:6::: do_machine_check machine_check
386 :386 #16 * 802097260944 (207303160222.v1) (207303160255) 4294874599 :1:2000000000:1::: mce_end do_machine_check
387 :387 #14 **** 802097261950 (207303160640.v1) (207303160714) 4294874599 :12:2:::: mce_start do_machine_check
388 :388 #16 ** 802097262056 (207303160686.v1) (207303160750) 4294874599 :12::::: mce_end do_machine_check
389 :389 #14 *** 802097263530 (207303161304.v1) (207303161334) 4294874599 :8:-4755801206503178081:6::: do_machine_check machine_check
390 :390 #14 * 802097265926 (207303162305.v1) (207303162331) 4294874599 :2:2000000000:2::: mce_end do_machine_check
391 :391 #4 **** 802097266672 (207303162615.v1) (207303162645) 4294874599 :12:3:::: mce_start do_machine_check
392 :392 #4 *** 802097267796 (207303163087.v1) (207303163119) 4294874599 :8:-4755801206503178081:6::: do_machine_check machine_check
393 :393 #4 * 802097269420 (207303163764.v1) (207303163794) 4294874599 :3:2000000000:3::: mce_end do_machine_check
394 :394 #0 **** 802097270254 (207303164111.v1) (207303164139) 4294874599 :12:4:::: mce_start do_machine_check
395 :395 #0 *** 802097271566 (207303164659.v1) (207303164726) 4294874599 :8:-4755801206503178081:6::: do_machine_check machine_check
396 :396 #0 * 802097273954 (207303165660.v1) (207303165690) 4294874599 :4:2000000000:4::: mce_end do_machine_check
397 :397 #15 **** 802097275214 (207303166183.v1) (207303166211) 4294874599 :12:5:::: mce_start do_machine_check
398 :398 #15 *** 802097276598 (207303166764.v1) (207303166826) 4294874599 :8:-4755801206503178081:6::: do_machine_check machine_check
399 :399 #15 * 802097278818 (207303167688.v1) (207303167720) 4294874599 :5:2000000000:5::: mce_end do_machine_check
400 :400 #12 **** 802097279702 (207303168057.v1) (207303168122) 4294874599 :12:6:::: mce_start do_machine_check
-----Original Message-----
From: Luck, Tony [mailto:tony.luck@xxxxxxxxx]
Sent: Friday, May 10, 2013 12:10 PM
To: Ming Lei; linux-kernel@xxxxxxxxxxxxxxx
Cc: mchehab@xxxxxxxxxx; bp@xxxxxxxxx
Subject: RE: x86_mce: mce_start uses number of phsical cores instead of logical cores
> With hyperthread turns on, the num_online_cpus reports the number of all logical cores.
> What I found in testing is only half the cores receives the mce broadcast, so I assume only the physical cores get broadcast.
See Intel Software Developer Manual Volume 3B Section 15.10.4.1, 3rd bullet:
o For processors on which CPUID reports DisplayFamily_DisplayModel as 06H_0EH and onward, an MCA signal is
broadcast to all logical processors in the system
Your E-5645 processors are a lot newer than this cut-off version - so they should broadcast to all your threads.
You are seeing something very strange. It would be interesting to know *which* 12 cpus show up for your machine check. Perhaps you are seeing all the hyperthreads from one socket and none from the other?
I still suspect that something is strange in the EDAC error injection side of this problem and that you are not getting a h/w initiated INT#18 event.
-Tony
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/