Date: Thu, 16 Aug 2001 18:26:32 +0900 (JST) From: Kenjiro Cho <kjc@csl.sony.co.jp> To: phk@critter.freebsd.dk Cc: julian@elischer.org, current@FreeBSD.ORG Subject: Re: Stack hogs revisited.. Message-ID: <20010816.182632.85398408.kjc@csl.sony.co.jp> In-Reply-To: <93495.997945131@critter> References: <Pine.BSF.4.21.0108151751240.14235-100000@InterJet.elischer.org> <93495.997945131@critter>
next in thread | previous in thread | raw e-mail | index | archive | help
Poul-Henning Kamp wrote: > In message <Pine.BSF.4.21.0108151751240.14235-100000@InterJet.elischer.org>, Ju > lian Elischer writes: > > >who owns the midway driver? (gotta go look at that.. 2.3 k?) (what I a > >midway?) > >I'm wondering if people who ran that card ever noticed real strange > >behaviour sometimes.. All it would take is a few interupt contexts on top > >of that and someone's signal information starts to get fried.. > > Midway is part of the "chuck-ATM" stack. > > It's unowned I belive. I imported the en driver and am still able to test it. The large frame is due to a 1KB boundary DMA test at the probe time. The stack size is allocated considering the kernel stack limit. (the driver is one of the first PCI drivers that heavily use DMA and we had a lot of weird DMA problems with the first generation PCI motherboards/chipsets.) So, I believe this driver is safe, but you can reduce the frame size by disabling NBURSTS in sys/dev/en/midway.c. -Kenjiro To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010816.182632.85398408.kjc>