From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ingo Molnar Subject: Re: [Bug #12872] pwc mmap always fails with EAGAIN Date: Sun, 22 Mar 2009 21:05:21 +0100 Message-ID: <20090322200521.GA5355@elte.hu> References: <20090322142837.2df0d93d@lxorguk.ukuu.org.uk> <20090322144516.GA13129@elte.hu> <200903222058.12924.M4rkusXXL@web.de> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: <200903222058.12924.M4rkusXXL-S0/GAf8tV78@public.gmane.org> Sender: kernel-testers-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Markus , Suresh Siddha Cc: linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Alan Cox , "Rafael J. Wysocki" , Kernel Testers List * Markus wrote: > > > > Still valid for 2.6.29-rc8-git5. > > > > > > And not really likely to change until someone with a pwc camera adds > lots > > > of printks to the code to work out why this happens. > > > > Or does a bisection to figure out the bad commit, because the report > > says that 2.6.28 worked. > > > > The 'howto bisect' portion of: > > > > http://www.kernel.org/doc/local/git-quick.html > > > Well I did a bisect: > be03d9e8022030c16abf534e33e185bfc3d40eef is first bad commit > commit be03d9e8022030c16abf534e33e185bfc3d40eef > Author: Suresh Siddha > Date: Wed Feb 11 11:20:23 2009 -0800 > > x86, pat: fix warn_on_once() while mapping 0-1MB range with /dev/mem > > Jeff Mahoney reported: > > > With Suse's hwinfo tool, on -tip: > > WARNING: at arch/x86/mm/pat.c:637 reserve_pfn_range+0x5b/0x26d() > > reserve_pfn_range() is not tracking the memory range below 1MB > as non-RAM and as such is inconsistent with similar checks in > reserve_memtype() and free_memtype() > > Rename the pagerange_is_ram() to pat_pagerange_is_ram() and add the > "track legacy 1MB region as non RAM" condition. > > And also, fix reserve_pfn_range() to return -EINVAL, when the pfn > range is RAM. This is to be consistent with this API design. > > Reported-and-tested-by: Jeff Mahoney > Signed-off-by: Suresh Siddha > Signed-off-by: Venkatesh Pallipadi > Signed-off-by: Ingo Molnar > > :040000 040000 feb3bf21d762e885284cb1bedb5d2c6c0d16811b > b4c2143775386ce15dd49b6749c1f7913d8beb00 M arch > > So I ran "git bisect reset ; git revert > be03d9e8022030c16abf534e33e185bfc3d40eef" and rebuild. That worked! > > I dont know if this is valid to do, but its working fine. > Its an amd64 system with CONFIG_X86_PAT=y. > > Need some feedback ;) First thing to check would be whether the latest round of PAT fixes (not yet upstream) fix the problem: http://people.redhat.com/mingo/tip.git/README Suresh, any ideas? Ingo