qoriq_e6500_32 with smp - compiler error

[Joel Sherrill]

On Fri, Sep 25, 2020, 2:36 PM Gedare Bloom <gedare at rtems.org> wrote:

> When I build the qoriq_e6500_32 with SMP enabled in the new build
> system, I hit this error:
> [  77/4070] Compiling bsps/powerpc/shared/cache/cache.c
> In file included from ../../../bsps/powerpc/shared/cache/cache.c:319:
> ../../../bsps/powerpc/include/../../../bsps/shared/cache/cacheimpl.h:92:2:
> error: #error "CPU_DATA_CACHE_ALIGNMENT (" ## CPU_DATA_CACHE_ALIGNMENT
> ## ") is greater than CPU_CACHE_LINE_BYTES (" ## CPU_CACHE_LINE_BYTES
> ## ")"
>    92 | #error "CPU_DATA_CACHE_ALIGNMENT (" ##
> CPU_DATA_CACHE_ALIGNMENT ## ") is greater than CPU_CACHE_LINE_BYTES ("
> ## CPU_CACHE_LINE_BYTES ## ")"
>       |  ^~~~~
> In file included from ../../../bsps/powerpc/shared/cache/cache.c:319:
> ../../../bsps/powerpc/include/../../../bsps/shared/cache/cacheimpl.h:96:2:
> error: #error "CPU_INSTRUCTION_CACHE_ALIGNMENT is greater than
> CPU_CACHE_LINE_BYTES"
>    96 | #error "CPU_INSTRUCTION_CACHE_ALIGNMENT is greater than
> CPU_CACHE_LINE_BYTES"
>       |  ^~~~~
>
> I haven't been able to dig into it. The autotools build doesn't hit
> the same problem.
>

Piling on. There is a coverity issue reported in the same file which is not
likely related. My quick analysis was that a 64 byte cache alignment made
some test in an if redundant. If in the area fixing something, take a look.
:)

--joel

> _______________________________________________
> devel mailing list
> devel at rtems.org
> http://lists.rtems.org/mailman/listinfo/devel
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rtems.org/pipermail/devel/attachments/20200925/451ea664/attachment.html>