On 08/02/17 19:21, Andrew Cooper wrote:
On 08/02/17 19:13, Julien Grall wrote:
Hi Andrew,

On 08/02/17 19:10, Andrew Cooper wrote:
c/s 11c397c broke the ARM build by introducing a common ACCESS_ONCE()
which is
different to the definiton in smmu.c

Forgot this one s/definiton/definition/


The SMMU code included a scalar typecheck, which is worth keeping in the
common case, given ACCESS_ONCE()'s restrictions.  However, express the
typecheck differently so as to avoid Coverity compliants about unused

s/compliants/complaint/

In this case, it is multiple individual complains about unused
individual variables, so "complaints" is scans perfectly well.

An alternative would be "to avoid Coverity complaining about..." if you
prefer?

Sorry I was flagging the typo "i" and "a" inverted and not the plural.



variables.

OOI, the variable is marked as "__maybe_unused", so why Coverity would
complaint?

The entire purpose of Coverity is to second guess what the programmer
actually wrote when it looks suspicious.

As for this specific example, I believe that the annotation doesn't even
survive into the GCC Abstract Syntax Tree, which means Coverity doesn't
get to see it.  Even if it does, the complaint of "This variable is
unused - why do you need it?" is still valid.

Oh, thank you for the explanation.

Cheers,

--
Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel

Reply via email to