From mboxrd@z Thu Jan 1 00:00:00 1970 From: William Lee Irwin III Subject: Re: [PATCH netdev-2.6 7/10] ixgb: Replace kmalloc with vmalloc to allocate driver local data structures Date: Fri, 29 Oct 2004 11:29:26 -0700 Sender: netdev-bounce@oss.sgi.com Message-ID: <20041029182926.GF12934@holomorphy.com> References: <468F3FDA28AA87429AD807992E22D07E0312853C@orsmsx408> <20041029130817.GA12301@infradead.org> <418277B7.7000705@pobox.com> <20041029173636.GB12934@holomorphy.com> <41828B5D.7090908@pobox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Christoph Hellwig , "Venkatesan, Ganesh" , netdev , Andrew Morton Return-path: To: Jeff Garzik Content-Disposition: inline In-Reply-To: <41828B5D.7090908@pobox.com> Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org William Lee Irwin III wrote: >> vmalloc is relatively highly penalized. For instance, global TLB >> flushes associated with vfree() and so on. Aggregate vmallocspace >> limitations are a big concern. I probably wouldn't use vmallocspace >> for much besides infrequent allocations, e.g. things allocated during >> driver initialization. One per driver instance won't really hurt much >> because vmallocspace limits prevent large numbers of drivers from being >> simultaneously loaded as it stands now anyway. On Fri, Oct 29, 2004 at 02:26:37PM -0400, Jeff Garzik wrote: > That's precisely what we're contemplating -- a once per driver > allocation that remains static for the entire driver runtime (descriptors) An infrastructural change that does this would probably need to be considered more carefully. I think you can probably just do it. -- wli