[RESEND,2/2] ARM: kexec: Check segment memory addresses

Commit Message

Aaro Koskinen Oct. 16, 2012, 5:09 p.m. UTC

On Tue, Oct 16, 2012 at 05:32:26PM +0100, Will Deacon wrote:
> On Tue, Oct 16, 2012 at 05:10:21PM +0100, Aaro Koskinen wrote:
> > On Thu, Sep 20, 2012 at 05:03:38PM +0100, Will Deacon wrote:
> > > From: Matthew Leach <[e-mail]>
> > > 
> > > Ensure that the memory regions that are set within the segments
> > > correspond to physical contiguous memory regions.
> > 
> > This patch seems to break kexec for me in 3.7-rc1 (commit
> > c564df4db85aac8d1d65a56176a0a25f46138064, bisected).
> > 
> > kexec -l fails with the following output:
> > 
> > kexec_load failed: Invalid argument
> > entry       = 0x80008000 flags = 280000
> > nr_segments = 2
> > segment[0].buf   = 0xc48008
> > segment[0].bufsz = 230
> > segment[0].mem   = 0x80001000
> > segment[0].memsz = 1000
> > segment[1].buf   = 0xb6a9b008
> > segment[1].bufsz = 3a4668
> > segment[1].mem   = 0x80008000
> > segment[1].memsz = 3a5000
> > Could not load the kexec kernel: n800-zImage
> > 
> > The board is OMAP2420 / Nokia N800.
> 
> Interesting, it sounds like kexec thinks that you don't have contiguous
> memory from 0x80008000 to 0x803ad000. Can you provide some more information
> about your physical memory map please?

Well, I think it's because the patch is wrong. Shouldn't it be:


A.

Comments

Will Deacon Oct. 16, 2012, 5:16 p.m. UTC | #1

On Tue, Oct 16, 2012 at 06:09:00PM +0100, Aaro Koskinen wrote:
> On Tue, Oct 16, 2012 at 05:32:26PM +0100, Will Deacon wrote:
> > Interesting, it sounds like kexec thinks that you don't have contiguous
> > memory from 0x80008000 to 0x803ad000. Can you provide some more information
> > about your physical memory map please?
> 
> Well, I think it's because the patch is wrong. Shouldn't it be:
> 
> diff --git a/arch/arm/kernel/machine_kexec.c b/arch/arm/kernel/machine_kexec.c
> index e29c333..a80192e 100644
> --- a/arch/arm/kernel/machine_kexec.c
> +++ b/arch/arm/kernel/machine_kexec.c
> @@ -47,7 +47,7 @@ int machine_kexec_prepare(struct kimage *image)
>  
>  		err = memblock_is_region_memory(current_segment->mem,
>  						current_segment->memsz);
> -		if (err)
> +		if (!err)
>  			return - EINVAL;
>  
>  		err = get_user(header, (__be32*)current_segment->buf);

Oops, that's a howler! Thanks for spotting it. We should probably reflow the
code a bit because !err sounds like everything should be ok.

Fancy reworking the patch or do you want me to take care of this?

Cheers,

Will

Patch

diff --git a/arch/arm/kernel/machine_kexec.c b/arch/arm/kernel/machine_kexec.c
index e29c333..a80192e 100644
--- a/arch/arm/kernel/machine_kexec.c
+++ b/arch/arm/kernel/machine_kexec.c
@@ -47,7 +47,7 @@  int machine_kexec_prepare(struct kimage *image)
 
 		err = memblock_is_region_memory(current_segment->mem,
 						current_segment->memsz);
-		if (err)
+		if (!err)
 			return - EINVAL;
 
 		err = get_user(header, (__be32*)current_segment->buf);