Re: [Qemu-arm] [PATCH v5 9/9] tests: add a m25p80 test

[Greg Kurz]

On Fri, 1 Jul 2016 19:30:30 +0200
Cédric Le Goater <address@hidden> wrote:

> On 07/01/2016 07:18 PM, Peter Maydell wrote:
> > On 28 June 2016 at 19:24, Cédric Le Goater <address@hidden> wrote:  
> >> This test uses the palmetto platform and the AST2400 SPI controller to
> >> test the m25p80 flash module device model. The flash model is defined
> >> by the platform (n25q256a) and it would be nice to find way to control
> >> it, using a property probably.
> >>
> >> Signed-off-by: Cédric Le Goater <address@hidden>
> >> Reviewed-by: Peter Maydell <address@hidden>
> >> ---
> >>  
> > 
> > This test fails on ppc64be:
> > 
> > TEST: tests/m25p80-test... (pid=65123)
> >   /arm/m25p80/read_jedec:                                              OK
> >   /arm/m25p80/erase_sector:                                            OK
> >   /arm/m25p80/erase_all:                                               **
> > ERROR:/home/pm215/qemu/tests/m25p80-test.c:162:test_erase_all:
> > assertion failed (page[i] == 0x0):
> > (0xffffffff == 0x00000000)
> > FAIL
> > GTester: last random seed: R02S54b2016fda21b092e18d7a23a2db86ba
> > (pid=65128)
> >   /arm/m25p80/write_page:                                              **
> > ERROR:/home/pm215/qemu/tests/m25p80-test.c:200:test_write_page:
> > assertion failed (page[i] == my_page_addr + i * 4): (0x00000000 ==
> > 0x01400000)
> > FAIL
> > GTester: last random seed: R02S8708910d6b72f700bc41e9340a516239
> > (pid=65133)
> > FAIL: tests/m25p80-test  
> 
> yes ... I am not sure how to fix this :/ 
> 
> I started with a patch using qtest_big_endian() and I found that 
> this one was fixing the problem : 
> 
>       https://lists.gnu.org/archive/html/qemu-devel/2016-06/msg07876.html
> 
> but it feels wrong. The interesting part is that the guest fully 
> boots on a ppc64be. We need an endian shaman for this. Greg ? 
> 

Heh ! Looking at the Cc list I guess you have chances to find the
shaman you're looking for :)

Anyway, doing bswap32() systematically like in the patch mentioned above
looks weird indeed.

> Thanks,
> 
> C.