Re: [Qemu-trivial] [PATCH 01/29] coccinelle: add a script to enforce qemu/osdep.h macros usage

[Eric Blake <eblake@redhat.com>]

[-- Attachment #1.1: Type: text/plain, Size: 1589 bytes --]

On 07/18/2017 01:09 AM, Philippe Mathieu-Daudé wrote:
> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>

Might be worth mentioning which macros in the commit message body, so
that a grep of 'git log' spots this commit easier.

> ---
>  scripts/coccinelle/use_osdep.cocci | 60 ++++++++++++++++++++++++++++++++++++++
>  1 file changed, 60 insertions(+)
>  create mode 100644 scripts/coccinelle/use_osdep.cocci
> 
> +// docker run --rm -v `pwd`:`pwd` -w `pwd` philmd/coccinelle \
> +//     --macro-file scripts/cocci-macro-file.h \
> +//     --sp-file scripts/coccinelle/add_osdep.cocci \
> +//     --keep-comments --in-place \
> +//     --use-gitgrep --dir .

Is '--use-gitgrep' a relatively new option to spatch (not present in my
F26 coccinelle-1.0.6-7), or is it a directive specific to the docker
coccinelle package you are using?  It's somewhat nice if there is a
clean separation between which arguments are for spatch and which are
for docker.

> +
> +// Use QEMU_IS_ALIGNED()
> +@@
> +typedef uintptr_t;
> +uintptr_t ptr;
> +expression n, m;
> +@@
> +(
> +-ptr % m == 0
> ++QEMU_PTR_IS_ALIGNED(ptr, m)
> +|
> +-n % m == 0
> ++QEMU_IS_ALIGNED(n, m)
> +)

Is it worth also trying to flag '(n & (m - 1)) == 0' (for when m is a
power of 2)?

> +
> +// Use QEMU_ALIGN_DOWN()
> +@@
> +expression n, m;
> +@@
> +-n / m * m
> ++QEMU_ALIGN_DOWN(n, m)

Is it worth also trying to catch (n >> m) << m?

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 619 bytes --]