Date: Thu, 29 Sep 2011 11:48:25 +0300 From: Alexander Motin <mav@FreeBSD.org> To: Adrian Chadd <adrian@freebsd.org> Cc: src-committers@freebsd.org, svn-src-user@freebsd.org Subject: Re: svn commit: r225860 - user/adrian/if_ath_tx/sys/mips/mips Message-ID: <4E8430D9.3060602@FreeBSD.org> In-Reply-To: <CAJ-VmomYAuY-90mWppn5buh6kT8vvno22n5SOudVkGiPTtc%2BGA@mail.gmail.com> References: <201109290310.p8T3AgZO001901@svn.freebsd.org> <4E84222F.7050101@FreeBSD.org> <CAJ-VmokRB-w4fLEp2xnS=XN2N3EUc3qgjBRiKTQWvPrtoGtyLA@mail.gmail.com> <CAJ-Vmom5LH6PdHVfFOQoN73=SjOmyuiLqMN5N9%2BhvjV1cz3GpQ@mail.gmail.com> <CAJ-VmomYAuY-90mWppn5buh6kT8vvno22n5SOudVkGiPTtc%2BGA@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Adrian Chadd wrote: > .. yes, spinlock_enter() / spinlock_exit() in mips does save/restore > the interrupt context. > So if they were disabled, they should continue being disabled. Ah, sorry, probably I've mixed it with something else. But still I see no reason to cover all the critical section with the disabled interrupts. It will just increase race window before the wait instruction and increase problem hit probability. -- Alexander Motin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4E8430D9.3060602>