Re: [Xen-devel] [PATCH v3 4/4] xen: use SYMBOL everywhere

[Julien Grall <julien.grall@xxxxxxx>]

Hi Stefano,

On 11/8/18 10:27 PM, Stefano Stabellini wrote:
> On Thu, 8 Nov 2018, Jan Beulich wrote:
> > > > > On 06.11.18 at 23:05, <sstabellini@xxxxxxxxxx> wrote:
> > > Use SYMBOL everywhere _stext, _etext, etc. are used. Technically, it
> > > is required when comparing and subtracting pointers [1], but use it
> > > everywhere to avoid confusion.
> > I think using it when not needed is causing more confusion. Also
> > why would you then not use it on all other data symbols? The
> > patch will end up quite a bit more reasonable in size once you drop
> > the unnecessary changes.
> OK, I am happy to do that. It will probably be better that way.
>
>
> > > ---
> > >   xen/arch/arm/alternative.c          |  7 ++--
> > >   xen/arch/arm/arm32/livepatch.c      |  2 +-
> > >   xen/arch/arm/arm64/livepatch.c      |  2 +-
> > >   xen/arch/arm/domain_build.c         |  2 +-
> > >   xen/arch/arm/livepatch.c            |  6 +--
> > >   xen/arch/arm/mm.c                   | 17 ++++----
> > >   xen/arch/arm/setup.c                |  8 ++--
> > >   xen/arch/x86/setup.c                | 79 +++++++++++++++++++------------------
> > >   xen/arch/x86/tboot.c                | 12 +++---
> > >   xen/arch/x86/x86_64/machine_kexec.c |  4 +-
> > >   xen/drivers/vpci/vpci.c             |  7 +++-
> > >   xen/include/asm-arm/grant_table.h   |  3 +-
> > >   xen/include/asm-arm/mm.h            |  4 +-
> > >   xen/include/asm-x86/mm.h            |  4 +-
> > >   xen/include/xen/kernel.h            | 24 +++++------
> > >   15 files changed, 97 insertions(+), 84 deletions(-)
> > Just like for v2: Did you really check you caught them all? The vPCI
> > ones I had pointed at back then were only an example. Another
> > example now is xen/common/kernel.c:_cmdline_parse().
> It is difficult to catch them all. Any suggestion on how to make sure
> there are no leftover (other than waiting for the next QAVerify scan)?
The webpage [1] seems to suggest coverity would be able to catch the 
undefined behavior fixed in that patch.
I am not sure what version of coverity is used to analyze Xen, but it 
probably worth to have a try.
Cheers,

[1] 
https://wiki.sei.cmu.edu/confluence/display/c/ARR36-C.+Do+not+subtract+or+compare+two+pointers+that+do+not+refer+to+the+same+array
--
Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel