Date: Fri, 29 May 2015 20:54:16 +0200 (CEST) From: Sebastian Huber <sebastian.huber@embedded-brains.de> To: Poul-Henning Kamp <phk@freebsd.org> Cc: freebsd-hackers@freebsd.org Subject: Re: Problem with timecounters and memory model Message-ID: <1056382604.51315.1432925656612.JavaMail.zimbra@embedded-brains.de> In-Reply-To: <85619.1432920847@critter.freebsd.dk> References: <55686300.3080100@embedded-brains.de> <85086.1432906578@critter.freebsd.dk> <55686C0F.9000101@embedded-brains.de> <85121.1432906871@critter.freebsd.dk> <55686D31.1060408@embedded-brains.de> <85619.1432920847@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
----- Poul-Henning Kamp <phk@freebsd.org> schrieb: > -------- > In message <55686D31.1060408@embedded-brains.de>, Sebastian Huber writes: >=20 > >> How often do you "wind" the timecounter ? > > > >More often than necessary. About 1000 times a second with only one=3D20 > >timehand (this is a hack to support legacy systems). The test case is: >=20 > If you only have one timehand, I can't see how it can work without > locks ? In RTEMS we have two build time configurations: uni-processor and SMP. In = the uni-processor configuration we have only one timehand. The windup func= tion is called in interrupt context with interrupts disabled. We have the = one timehand limitation since not every supported hardware provides a free = running counter. In the SMP configuration we use the default 10 timehands. --=20 Sebastian Huber, embedded brains GmbH Address : Dornierstr. 4, D-82178 Puchheim, Germany Phone : +49 89 189 47 41-16 Fax : +49 89 189 47 41-09 E-Mail : sebastian.huber at embedded-brains.de PGP : Public key available on request. Diese Nachricht ist keine gesch=C3=A4ftliche Mitteilung im Sinne des EHUG.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1056382604.51315.1432925656612.JavaMail.zimbra>