[meta-xilinx] devmem2 accessing AXI (M00_AXI) hangs the board
Moritz Fischer
moritz.fischer at ettus.com
Mon Oct 17 16:31:20 PDT 2016
Giordon,
On Wed, Oct 12, 2016 at 2:25 PM, Giordon Stark <kratsg at gmail.com> wrote:
> Hi,
>
> I have an image based on zc706 that includes python + devmem2 recipes
>
> root at gfex-prototype2:~# devmem2 -h
> /dev/mem opened.
> Memory mapped at address 0xb6f36000.
> Read at address 0x00000000 (0xb6f36000): 0xE59F0000
>
> I can see devmem2 functioning correctly here. When I access the memory
> associated with M00_AXI (see attached screenshot):
>
> root at gfex-prototype2:~# devmem2 0x43c00000 w
> /dev/mem opened.
> Memory mapped at address 0xb6f5e000.
> (hangs)
>
> I wonder if this is because of a failure in the AXI transfer? See attached
> image. M00_AXI does not have an entry in my device tree either, but I don't
> think that's a problem.
It might be your clock not being turned off, or your HDL for the slave
not working correctly.
Check your clock settings in /sys/kernel/debug/clocks/summary etc and
make sure the FCLKx going
to your custom logic is on and check your HDL for warnings ...
Moritz
PS: Do NOT send HTML email to lists
More information about the meta-xilinx
mailing list