Date: Sat, 26 Feb 2011 11:59:26 +0800 From: "beezarliu" <beezarliu@yahoo.com.cn> To: "Arnaud Lacombe" <lacombar@gmail.com> Cc: "freebsd-net@freebsd.org" <freebsd-net@freebsd.org>, Jack Vogel <jfvogel@gmail.com>, "bug-followup@FreeBSD.org" <bug-followup@freebsd.org> Subject: Re: Re: Re: kern/150516: [em] e1000 receive queue handling problem Message-ID: <201102261159157631491@yahoo.com.cn> References: <AANLkTi=HW4XqZ3PTB69fL5h-Qpqpae4C2sfLLdJSFxrs@mail.gmail.com>, <201102231218136253955@yahoo.com.cn>, <AANLkTinD%2B7Xo0RRZH818%2BM9cLKNYhC0eKAsVZPJbR1GK@mail.gmail.com>, <201102231404144686577@yahoo.com.cn>, <AANLkTik3x0qzffoizHMhfsOrYVG9UWKzyY7wf3pKO5QL@mail.gmail.com>, <AANLkTim0NFT2GrhMq77=K-=-4Q6xnh7saHy3c-Bid03N@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2011-02-26 07:52:54, Arnaud Lacombe wrote: >Hi, > >What is the point to invent a complex logic to detected a situation >the chip warn you about ? > >The attached patch has currently survived longer than anything I've >been ever tested, without dirty hack, like raising `nmbclusters'. Sorry, I didn't 'invent' the logic, it's just what rx ring queue works in hardware. You provided another way to detect the hang, which doesn't mean others are meaningless. Raising nmbcluster is not best way, but it can ease the memory shortage, and allcate more for network, which also improve system performance. Beezar
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201102261159157631491>