Re: QNX

From: Zdenek Salvet <salvet@ANXUR.DCS.MUNI.CZ>
Subject: Re: QNX
Date: Tue, 6 Feb 1996 16:19:10 GMT

Next Article (by Author): Re: Sdileni modemu v NT "ing. Zbynek Knop"
Previous Article (by Author): Re: Virii Zdenek Maruna
Top of Thread: Re: QNX Science Systems CR
Next in Thread: Re: QNX Jan Kasprzak
Articles sorted by: [Date] [Author] [Subject]

Science Systems CR <scisys@scisys.anet.cz> writes:
>>
>>         Doporucoval bych zacit u news:comp.os.qnx. Jako OS pro aplikace
>> v realnem case je QNX dobry, ale bezny UN*Xovy vykon tohoto systemu
>> je _velmi_ slaby (treba prepinani kontextu, propustnost roury, ...).
 
>Hmmm. Toto je myslim _velmi_ odvazne tvrzeni. Mohl byste prosim
>dolozit, jak jste k tomuto zaveru dosel?
>Jinak to vypada takto (verze 4.22):
 
>            Context [microsec]     Interrupt latency [microsec]
>P133           3.1                        4.3
>P100           3.9                        4.4
>486DX4         5.1                        4.9
>386/33        25                         14
 
Predpokladam, ze je to yield() a ne prepnuti unixovych procesu
pri cekani na nejaky zdroj.
 
Obecne: hodnoceni "_velmi_ slaby" je urcite prehnane, ale kdo chce
klasicky UNIX, ma lepsi varianty. Kdyz jsem naposledy s QNX experimentoval
(asi v. 4.2), nejvetsim problemem byla absence virtualni pameti
ve smyslu odkladani na disk - takovy system je pro bezne pouziti
o dost drazsi nez je nutne.
 
--
Zdenek Salvet                                              salvet@ics.muni.cz
----------------------------------------------------------------------------
           If God had meant for us to be in the Army,
         we would have been born with green, baggy skin.

Go to listserv.cesnet.cz LWGate Home Page.