From mboxrd@z Thu Jan  1 00:00:00 1970
From: Wei Liu <liuw@liuw.name>
Subject: Re: mmap in PV xen-4.0.1
Date: Fri, 12 Aug 2011 12:26:08 +0800
Message-ID: <20110812042608.GA2760@limbo>
References: <CACeEFf77eOLLNgsJwEYdzeiuxm+qDcO2zs09FK-Waas0sxcwLQ@mail.gmail.com>
	<CACeEFf5yDb3dh+M3nycmYMaW5Vf+6C57QXZjXEBQZhrwkTHkHg@mail.gmail.com>
	<20110810091256.GA1537@limbo>
	<CACeEFf6HoD10k5PUP7gNBNteyY=Wh3XTJCOZEmvs+Us+WfahNw@mail.gmail.com>
	<20110810192546.GA2875@dumpdata.com>
	<CACeEFf5B08guZ2XNKGcuUSP-W=NiqMHxT3o0ghi5ky7QJOeRKg@mail.gmail.com>
	<20110811013104.GB1889@limbo>
	<CACeEFf72bXGhgb=O9U-eavV7yqBNAepx_EZ+Lp5nPFa7KK_zeQ@mail.gmail.com>
	<CACeEFf7X1cCwPxDfaCGuUGF0qTnrBO_jnMa_UNd_Ni2=AosAoA@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Return-path: <xen-devel-bounces@lists.xensource.com>
Content-Disposition: inline
In-Reply-To: <CACeEFf7X1cCwPxDfaCGuUGF0qTnrBO_jnMa_UNd_Ni2=AosAoA@mail.gmail.com>
List-Unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
List-Post: <mailto:xen-devel@lists.xensource.com>
List-Help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-Subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>,
	<mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
Sender: xen-devel-bounces@lists.xensource.com
Errors-To: xen-devel-bounces@lists.xensource.com
To: Eric Camachat <eric.camachat@gmail.com>
Cc: xen-devel@lists.xensource.com, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
List-Id: xen-devel@lists.xenproject.org

On Thu, Aug 11, 2011 at 10:11:20AM -0700, Eric Camachat wrote:
> 
> Let us back to my original concern:
> Why remap_pfn_range() woks with my own device node, but deesn't work
> with /dev/mem node?
> This behavior is confused me.
> 
> /Eric

Well, I slightly modify your code -- remove the ioctl and hardcoded
*base into mapper, and run on my box in dom0 (Xen 4.2 + 2.6.39 PVOPS),
I get following output and kernel log:

paddr = 0x6d368000
  mem = 0xffffffffffffffff
  map = 0x7fa928e88000
map[0]= 1234
map[1]= abcd

[  405.039120] memtest: registering /dev/memtest (886)
[  405.039125] memtest: size of phys_addr_t is 8 bytes
[  405.039132] memtest: _vbase = 0xffff88006d368000
[  405.039134] memtest: _pbase = 0x6d368000
[  405.039135] memtest: _mbase = 0x6fbaa000
[  781.414747] Program mapper tried to access /dev/mem between 6d368000->6d36c000.
[  781.414764] memtest_vma_open: virt 0x7f327a275000, phys 0x6d368000
[  781.414942] mapper[2744]: segfault at ffffffffffffffff ip 00000000004008d7 sp 00007fff233638e0 error 4 in mapper[400000+1000]
[  781.415064] memtest_vma_close
[  891.350796] Program mapper tried to access /dev/mem between 6d368000->6d36c000.
[  891.350813] memtest_vma_open: virt 0x7fa928e88000, phys 0x6d368000
[  891.350987] mapper[2811]: segfault at ffffffffffffffff ip 00000000004008d7 sp 00007fffc154efd0 error 4 in mapper[400000+1000]
[  891.351102] memtest_vma_close

Looking into the kernel source, a check in range_is_allowed failed so
I just can't map /dev/mem with specified range (-EPERM).

But please note that my output of map[0] and map[1] are correct, while
your output is not correct ('4C4C4C4C'), which you ignored
previously. Why not make sure your mapper works correctly? Or can you
try newer kernel and Xen?

Wei.