On Fri, Sep 20, 2024 at 02:40:22PM +0100, Frediano Ziglio via Grub-devel wrote:
> On Thu, Sep 19, 2024 at 11:03 PM Sergii Dmytruk
> <sergii.dmyt...@3mdeb.com> wrote:
> >
> > From: Krystian Hebel <krystian.he...@3mdeb.com>
> >
> > Subsequent patches will use those macros and constant.
> >
>
> Minor, but "Define GRUB_PAGE_MASK constant and GRUB_PAGE_{UP, DOWN}
> macros" subject sounds a bit confusing to me. I mean, at the end they
> are all defined as macros so why mixing constant and macros?
The latter two are function-like macros while the first one isn't and
always expands to the same value. That's how I interpreted the title.
> > Signed-off-by: Krystian Hebel <krystian.he...@3mdeb.com>
> > Signed-off-by: Sergii Dmytruk <sergii.dmyt...@3mdeb.com>
> > ---
> > include/grub/i386/memory.h | 5 +++++
> > 1 file changed, 5 insertions(+)
> >
> > diff --git a/include/grub/i386/memory.h b/include/grub/i386/memory.h
> > index c64529630..56f64855b 100644
> > --- a/include/grub/i386/memory.h
> > +++ b/include/grub/i386/memory.h
> > @@ -22,6 +22,7 @@
> >
> > #define GRUB_PAGE_SHIFT 12
> > #define GRUB_PAGE_SIZE (1UL << GRUB_PAGE_SHIFT)
> > +#define GRUB_PAGE_MASK (~(GRUB_PAGE_SIZE - 1))
> >
>
> Why not use signed numbers so it could extend as needed?
I don't think signed numbers were avoided on purpose, masks just tend to
be unsigned (e.g., to avoid UB on shifts). Can do
#define GRUB_PAGE_MASK (~((1L << GRUB_PAGE_SHIFT) - 1))
instead.
> > /* The flag for protected mode. */
> > #define GRUB_MEMORY_CPU_CR0_PE_ON 0x1
> > @@ -43,8 +44,12 @@
> >
> > #define GRUB_MMAP_MALLOC_LOW 1
> >
> > +#include <grub/misc.h>
> > #include <grub/types.h>
> >
> > +#define GRUB_PAGE_UP(p) ALIGN_UP (p, GRUB_PAGE_SIZE)
> > +#define GRUB_PAGE_DOWN(p) ALIGN_DOWN (p, GRUB_PAGE_SIZE)
> > +
> > struct grub_e820_mmap_entry
> > {
> > grub_uint64_t addr;
>
> Frediano
Sergii
_______________________________________________
Grub-devel mailing list
Grub-devel@gnu.org
https://lists.gnu.org/mailman/listinfo/grub-devel
