Re: [RFC PATCH 2/2] target/arm: Constify lot of helpers taking CPUARMState argument

[Philippe Mathieu-Daudé]

On 17/1/25 13:42, Daniel Henrique Barboza wrote:
> On 1/16/25 8:04 PM, Philippe Mathieu-Daudé wrote:
> > When methods don't modify the CPUARMState* argument,
> > we can mark it const. This allow enforcing places where
> > the CPU env shouldn't be modified at all,
> >
> > Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> > ---
>
> We should use 'const' more often in general IMO. It forces a better 
> usage and
> understanding of our own APIs.
>
>
> Reviewed-by: Daniel Henrique Barboza <dbarboza@ventanamicro.com>

Thanks!

> > I went via the "modify one and fix until it builds" path,
> > and the result seemed trivial enough, but can try to split
> > if requested.

I actually did split and will respin.

> > ---
> >   target/arm/cpu-features.h |  2 +-
> >   target/arm/cpu.h          | 71 ++++++++++++++++++++-------------------
> >   target/arm/internals.h    | 10 +++---
> >   target/arm/helper.c       | 25 +++++++-------
> >   target/arm/ptw.c          |  2 +-
> >   target/arm/tcg/m_helper.c |  8 ++---
> >   target/arm/vfp_helper.c   |  6 ++--
> >   7 files changed, 63 insertions(+), 61 deletions(-)