Date: Fri, 11 Aug 2006 11:06:00 -0700 From: "Jack Vogel" <jfvogel@gmail.com> To: "Landon Fuller" <landonf@opendarwin.org> Cc: Daniel Ryslink <daniel.ryslink@col.cz>, freebsd-net@freebsd.org Subject: Re: Problems with em interfaces on FreeBSD 6.1 Message-ID: <2a41acea0608111106t6a6dcc84l35eb2cae5dbd42c1@mail.gmail.com> In-Reply-To: <B0C441A8-22F7-45C8-B847-B0BDA7DE7779@opendarwin.org> References: <20060811100536.V80282@k2.vol.cz> <20060811111240.GD96644@FreeBSD.org> <20060811133531.D80282@k2.vol.cz> <20060811125825.GH96644@cell.sick.ru> <2a41acea0608110922h4bed63b1ke09f91b610819805@mail.gmail.com> <B0C441A8-22F7-45C8-B847-B0BDA7DE7779@opendarwin.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 8/11/06, Landon Fuller <landonf@opendarwin.org> wrote: > > We saw this issue here on SMP systems running 6.1; I've been meaning > to set up a reproduction case in the lab and dig into the issue further. > Disabling the mpsafe network stack (debug.mpsafenet=0) is our > temporary work-around; rwatson mentioned that this has the effect of > forcing the interrupt handler for if_em to not run in parallel with > the transmit code, which is likely what caused the problem to disappear. Hmmm, I have code running right now that is required for new hardware, but I think will benefit all, that only has the hardware writeback and interrupt on the EOP descriptor. As it is now we have every descriptor written back. I havent had it put thru extensive tests yet, but I expect it to significantly reduce interrupts. I hope to be submitting this stuff together with TSO support to CURRENT soon, stay tuned. Jack Intel LAD
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2a41acea0608111106t6a6dcc84l35eb2cae5dbd42c1>