On Tue, 2005-11-08 at 07:01 -0500, Ernest L. Williams Jr. wrote:
> On Tue, 2005-11-08 at 05:46 -0600, Marty Kraimer wrote:
> > On Nov 8, 2005, at 5:30 AM, Ernest L. Williams Jr. wrote:
> > >
> > >
> > > The following is new for Linux:
> > > ======================================================================
> > > ==
> > >
> > > # Use POSIX thread priority scheduling (YES or NO)
> > > USE_POSIX_THREAD_PRIORITY_SCHEDULING = YES
> > > ======================================================================
> > > ==
> > >
> > > However, I think there are tests in <..>/src/libCom/test, right?
> > >
> > > I hoping for an advantage on linux-based softIOCs.
> > >
> > >
> >
> > Yes that test works.
> >
> > Also on any ioc run
> >
> > epicsThreadShowAll
> >
> > If the OSS priorities have values, i.e. not all 0, then you are
> > getting priority scheduling.
> >
> > Remember that you must run the ioc as root.
>
>
> Awesome:
>
> Here are the results for a linux-x86_64 softIOC:
> ===================================================================
> Normal User:
> epics> epicsThreadShowAll
> NAME EPICS ID PTHREAD ID OSIPRI OSSPRI STATE
> _main_ 0x505d20 0 0 0 OK
> errlog 0x50b470 1074010464 10 0 OK
> taskwd 0x57a990 1074280800 10 0 OK
> timerQueue 0x511ff0 1074813280 70 0 OK
> cbLow 0x578810 1075870048 59 0 OK
> cbMedium 0x580090 1075341664 64 0 OK
> cbHigh 0x584220 1077455200 71 0 OK
> dbCaLink 0x5845d0 1076398432 50 0 OK
> scanOnce 0x587af0 1078511968 70 0 OK
> scan10 0x588460 1079568736 60 0 OK
> scan5 0x5887c0 1080625504 61 0 OK
> scan2 0x588b20 1081682272 62 0 OK
> scan1 0x589380 1077983584 63 0 OK
> scan0.5 0x5895c0 1082739040 64 0 OK
> scan0.2 0x589920 1083795808 65 0 OK
> scan0.1 0x589c80 1084852576 66 0 OK
> CAC-event 0x5ba2c0 1084324192 52 0 OK
> CAS-TCP 0x5c7270 1083267424 16 0 OK
> CAS-beacon 0x50b600 1073875296 14 0 OK
> CAS-UDP 0x50a420 1076926816 12 0 OK
> epics>
> =====================================================================
>
> =====================================================================
> Root User:
> epics> epicsThreadShowAll
> NAME EPICS ID PTHREAD ID OSIPRI OSSPRI STATE
> _main_ 0x505d20 0 0 0 OK
> errlog 0x50b190 1073875296 10 10 OK
> taskwd 0x57a0a0 1074010464 10 10 OK
> timerQueue 0x57bce0 1074276704 70 69 OK
> cbLow 0x57b5b0 1074805088 59 58 OK
> cbMedium 0x511e70 1075333472 64 63 OK
> cbHigh 0x5120b0 1075861856 71 70 OK
> dbCaLink 0x511540 1076390240 50 50 OK
> scanOnce 0x5873d0 1076918624 70 69 OK
> scan10 0x587bd0 1077447008 60 59 OK
> scan5 0x587dc0 1077975392 61 60 OK
> scan2 0x587fb0 1078503776 62 61 OK
> scan1 0x5881a0 1079032160 63 62 OK
> scan0.5 0x588390 1079560544 64 63 OK
> scan0.2 0x588580 1080088928 65 64 OK
> scan0.1 0x588770 1080617312 66 65 OK
> CAC-event 0x5b8b10 1080883552 52 51 OK
> CAS-TCP 0x5c5a30 1081149792 16 16 OK
> CAS-beacon 0x5c5c40 1081284960 14 14 OK
> CAS-UDP 0x5c5e60 1081551200 12 12 OK
> epics>
> ====================================================================
Oops. I get an exception when exiting the IOC as root:
epics> epicsThreadShowAll
NAME EPICS ID PTHREAD ID OSIPRI OSSPRI STATE
_main_ 0x505d20 0 0 0 OK
errlog 0x50b190 1073875296 10 10 OK
taskwd 0x57a0a0 1074010464 10 10 OK
timerQueue 0x57bce0 1074276704 70 69 OK
cbLow 0x57b5b0 1074805088 59 58 OK
cbMedium 0x511e70 1075333472 64 63 OK
cbHigh 0x5120b0 1075861856 71 70 OK
dbCaLink 0x511540 1076390240 50 50 OK
scanOnce 0x5873d0 1076918624 70 69 OK
scan10 0x587bd0 1077447008 60 59 OK
scan5 0x587dc0 1077975392 61 60 OK
scan2 0x587fb0 1078503776 62 61 OK
scan1 0x5881a0 1079032160 63 62 OK
scan0.5 0x588390 1079560544 64 63 OK
scan0.2 0x588580 1080088928 65 64 OK
scan0.1 0x588770 1080617312 66 65 OK
CAC-event 0x5b8b10 1080883552 52 51 OK
CAS-TCP 0x5c5a30 1081149792 16 16 OK
CAS-beacon 0x5c5c40 1081284960 14 14 OK
CAS-UDP 0x5c5e60 1081551200 12 12 OK
epics> exit
epicsThread: Unknown C++ exception in thread "timerQueue" at Tue Nov 08
2005 07:01:33.72157000
Segmentation fault
[root@sns-hp-eval1 iocamd64]#
I will run all of this again on 32-bit linux
>
>
> Thanks,
> Ernest
>
>
>
>
> >
> > Marty
>
- References:
- 3.14.8 and mrkSoftTest Ralph Lange
- Re: 3.14.8 and mrkSoftTest Marty Kraimer
- Re: 3.14.8 and mrkSoftTest Ernest L. Williams Jr.
- Re: 3.14.8 and mrkSoftTest Marty Kraimer
- Re: 3.14.8 and mrkSoftTest Ernest L. Williams Jr.
- Navigate by Date:
- Prev:
Re: 3.14.8 and mrkSoftTest Ernest L. Williams Jr.
- Next:
More testing on linux-x86_64 Ernest L. Williams Jr.
- Index:
2002
2003
2004
<2005>
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
- Navigate by Thread:
- Prev:
Re: 3.14.8 and mrkSoftTest Ernest L. Williams Jr.
- Next:
3.14.8: blockingSockTest - assert() fails. Ralph Lange
- Index:
2002
2003
2004
<2005>
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
|