Use real number of cores for default -par, ignore virtual cores #6361

1. 
   laanwj commented at 4:03 pm on July 1, 2015: member

   To determine the default for -par, the number of script verification threads, use boost::thread::physical_concurrency() which counts only physical cores, not virtual cores.

   Virtual cores are roughly a set of cached registers to avoid context switches while threading, they cannot actually perform work, so spawning a verification thread for them could even reduce efficiency and will put undue load on the system.

   Should fix issue #6358, as well as some other reported system overload issues, especially on Intel processors.

   The function was only introduced in boost 1.56, so provide a utility function GetNumCores to fall back for older Boost versions.

2. Use real number of cores for default -par, ignore virtual cores

   To determine the default for `-par`, the number of script verification
   threads, use [boost::thread::physical_concurrency()](http://www.boost.org/doc/libs/1_58_0/doc/html/thread/thread_management.html#thread.thread_management.thread.physical_concurrency)
   which counts only physical cores, not virtual cores.
   
   Virtual cores are roughly a set of cached registers to avoid context
   switches while threading, they cannot actually perform work, so spawning
   a verification thread for them could even reduce efficiency and will put
   undue load on the system.
   
   Should fix issue #6358, as well as some other reported system overload
   issues, especially on Intel processors.
   
   The function was only introduced in boost 1.56, so provide a utility
   function `GetNumCores` to fall back for older Boost versions.

   47162673c7
3. laanwj added the label Bug on Jul 1, 2015
4. 
   laanwj commented at 4:22 pm on July 1, 2015: member

   We’ll also have to bump the boost version in depends for this to work (currently 1.55), but I remember that was already the plan for 0.12.
5. 
   theuni commented at 4:30 am on July 2, 2015: member

   @laanwj yes, we need to go ahead with the boost bump. I’m getting ready to leave for a few days for a conference, but I’ll make the boost/qt bumps first priority when I get back. Ideally those should be done very early in the cycle so we have some testing.

   Taking that one step further, I’ll check the other deps while I’m at it and see if they need bumps too.

6. 
   luke-jr commented at 5:21 am on July 2, 2015: member

   utACK
7. 
   paveljanik commented at 6:04 am on July 2, 2015: contributor

   ACK
8. laanwj merged this on Jul 2, 2015
9. laanwj closed this on Jul 2, 2015

   ---

10. laanwj referenced this in commit d7ada03b8a on Jul 2, 2015
11. 
    sipa commented at 11:13 pm on July 9, 2015: member

    Posthumous ACK.
12. luke-jr referenced this in commit 5c166950eb on Jan 10, 2016
13. zkbot referenced this in commit df07f9ad23 on Feb 15, 2017
14. zkbot referenced this in commit 0a80a1cea3 on Jun 16, 2017
15. laanwj referenced this in commit bc679829e2 on Mar 6, 2018
16. furszy referenced this in commit 4ed15cc69d on Jun 8, 2020
17. PastaPastaPasta referenced this in commit 098ccb3aaa on Jun 10, 2020
18. PastaPastaPasta referenced this in commit a12c3950ac on Jun 12, 2020
19. PastaPastaPasta referenced this in commit d40a7323cb on Jun 13, 2020
20. PastaPastaPasta referenced this in commit 1a40ff1e6e on Jun 14, 2020
21. PastaPastaPasta referenced this in commit a00da7fb0e on Jun 14, 2020
22. MarcoFalke locked this on Sep 8, 2021

Contributors
laanwj theuni luke-jr paveljanik sipa

Labels
Bug