From - Thu Sep 25 07:44:53 1997 Path: news.mitre.org!blanket.mitre.org!agate!howland.erols.net!newsfeed.direct.ca!news.he.net!newshub.cts.com!newsfeed.cts.com!cmkrnl!jeh From: jeh@cmkrnl.com (Jamie Hanrahan) Newsgroups: comp.os.ms-windows.programmer.nt.kernel-mode Subject: Re: MmMapIoSpace() and KeFlushIoBuffers() Message-ID: <1997Sep25.022000.8224@cmkrnl> Date: 25 Sep 97 02:20:00 PST References: <01bcc792$f16e1f90$3d436382@bs50798> <34279026.3364287@spock> <01bcc95c$baa68260$d320de80@eminhize> Organization: Kernel Mode Systems, San Diego, CA Lines: 26 In article <01bcc95c$baa68260$d320de80@eminhize>, "Terry Eminhizer" writes: > Stephan Wolf wrote in article > <34279026.3364287@spock>... >> On 22 Sep 1997 20:08:30 GMT, "Brad Sahr" wrote: >> >> >My PCI device has some onboard memory. I use MmMapIoSpace() with >> >the CacheEnable parameter set to FALSE to map the device memory >> >to nonpaged system space. >> ... >> > Does my driver need to issue >> >KeFlushIoBuffers prior to writing and subsequent to reading the device >> >memory? >> > > You shouldn't need to worry with KeFlushIOBuffers - my understanding is > that it really only copies memory buffered by the use of MapRegisters - > which you aren't using anyways (32bit PCI - 32bit system memory). You're thinking of MmMapIoSpace here. --- Jamie Hanrahan, Kernel Mode Systems, San Diego CA Internet: jeh@cmkrnl.com (JH645) CompuServe: 74140,2055 drivers, internals, networks, applications, and training for VMS and Windows NT NT driver FAQ, links, and other information: http://www.cmkrnl.com/ If you post a reply in news, please don't e-mail it too.