Date: Wed, 10 Mar 2010 08:21:30 -0600 (CST) From: Mark Tinguely <tinguely@casselton.net> To: gjb@semihalf.com, tinguely@casselton.net Cc: freebsd-arm@freebsd.org Subject: Re: Performance of SheevaPlug on 8-stable Message-ID: <201003101421.o2AELUng048879@casselton.net> In-Reply-To: <4B97A568.5080101@semihalf.com>
next in thread | previous in thread | raw e-mail | index | archive | help
<deletes of my stuff> > > There are a few data dangling allocations that I would like to see > > closed from the multiple kernel allocation fix. *IN THEORY, IF* a page > > is allocated via the arm_nocache (DMA COHERENT) or a sendfile, then > > it is never marked as unallocated. *IN THEORY*, if that page is used > > again, then we could falsely believe that page is being shared and > > we turn off the cache, eventhough it is not shared. > > > > http://www.casselton.net/~tinguely/arm_pmap_unmanaged.diff > > > > * Disclaimer: I am not sure if DMA COHERENT nor sendfiles are used in > > the Sheeva implementation. This is a theoritical observation of a side > > effect of the multiple kernel mapping patch that we did just before > > FreeBSD 8-release. > > I instrumented code with KTRs and your theory is correct. Kernel reuse page > which was previouly mapped via arm_nocache. Your patch should be applied > to -current. > > grzesiek Thank-you. I would appreciate it if someone would commit that patch. It has been on my mind since FreeBSD 8.0-release. This patch should help some data buffers because we do not incorrectly think this page is still mapped to a KVA. I don't think this our I/O performance problem solution, Off-list, I have mentioned the belief that it may be beneficial to think of not turning off the cache DMA_COHERENT even for ARMv4/ARMv5. --Mark Tinguely.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201003101421.o2AELUng048879>