Date: Fri, 22 Jun 2007 10:24:08 -0700 From: Nate Lawson <nate@root.org> To: Kostik Belousov <kostikbel@gmail.com> Cc: cvs-src@freebsd.org, src-committers@freebsd.org, cvs-all@freebsd.org Subject: Re: cvs commit: src/etc devd.conf rc.resume rc.suspend src/sys/i386/acpica acpi_machdep.c src/usr.sbin/acpi/acpiconf acpiconf.c src/sys/dev/acpica acpi.c acpiio.h acpivar.h Message-ID: <467C05B8.6090105@root.org> In-Reply-To: <20070622034555.GD2268@deviant.kiev.zoral.com.ua> References: <200706212250.l5LMobq4038270@repoman.freebsd.org> <20070622034555.GD2268@deviant.kiev.zoral.com.ua>
next in thread | previous in thread | raw e-mail | index | archive | help
Kostik Belousov wrote: > On Thu, Jun 21, 2007 at 10:50:37PM +0000, Nate Lawson wrote: >> njl 2007-06-21 22:50:37 UTC >> >> FreeBSD src repository >> >> Modified files: >> etc devd.conf rc.suspend rc.resume >> sys/i386/acpica acpi_machdep.c >> usr.sbin/acpi/acpiconf acpiconf.c >> sys/dev/acpica acpi.c acpiio.h acpivar.h >> Log: >> Update the suspend/resume user API while maintaining backwards compat. >> >> Improvements: >> * /dev/apm is clonable, allowing multiple listeners for suspend events. > Did you evaluate/tested destroy_dev_sched() KPI for your needs ? Comments on your patch coming via private email. > I do not see any protection for several opens going parallel on given cdev. > I.e., I think that si_drv1 assignment in open() could cause problems. > Generally, as I know, system does not enforce any serialization between > calls to cdev methods for mp-safe drivers. Right, something your patch doesn't implement that we need is for SI_CHEAPCLONE to really mean "no export to devfs". -- Nate
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?467C05B8.6090105>