Re: [PATCH] xen: simplify bitmap_to_xenctl_bitmap for little endian

[Stefano Stabellini <sstabellini@xxxxxxxxxx>]

On Mon, 24 Mar 2025, Jan Beulich wrote:
> On 22.03.2025 00:09, Stefano Stabellini wrote:
> > On Thu, 20 Mar 2025, Jan Beulich wrote:
> >> On 20.03.2025 01:57, Stefano Stabellini wrote:
> >>> On Wed, 19 Mar 2025, Jan Beulich wrote:
> >>>> What about xenctl_bitmap_to_bitmap()?
> >>>  
> >>> Let me see first if I managed to handle bitmap_to_xenctl_bitmap well.
> >>
> >> Well, the code looks correct to me, but the description now has gone
> >> stale. I also wonder whether with that extra restriction the optimization
> >> is then actually worth it. Just one further nit:
> > 
> > Hi Jan, you make a good point. I tried to come up with a better
> > approach. What do you think of this?
> > 
> > 
> > diff --git a/xen/common/bitmap.c b/xen/common/bitmap.c
> > index 3da63a32a6..2f448693c3 100644
> > --- a/xen/common/bitmap.c
> > +++ b/xen/common/bitmap.c
> > @@ -52,7 +52,7 @@ static void clamp_last_byte(uint8_t *bp, unsigned int 
> > nbits)
> >     unsigned int remainder = nbits % 8;
> >  
> >     if (remainder)
> > -           bp[nbits/8] &= (1U << remainder) - 1;
> > +           *bp &= (1U << remainder) - 1;
> >  }
> >  
> >  int __bitmap_empty(const unsigned long *bitmap, unsigned int bits)
> > @@ -338,7 +338,6 @@ static void bitmap_long_to_byte(uint8_t *bp, const 
> > unsigned long *lp,
> >                     nbits -= 8;
> >             }
> >     }
> > -   clamp_last_byte(bp, nbits);
> >  }
> >  
> >  static void bitmap_byte_to_long(unsigned long *lp, const uint8_t *bp,
> > @@ -363,7 +362,6 @@ static void bitmap_long_to_byte(uint8_t *bp, const 
> > unsigned long *lp,
> >                             unsigned int nbits)
> >  {
> >     memcpy(bp, lp, DIV_ROUND_UP(nbits, BITS_PER_BYTE));
> > -   clamp_last_byte(bp, nbits);
> >  }
> >  
> >  static void bitmap_byte_to_long(unsigned long *lp, const uint8_t *bp,
> > @@ -384,21 +382,40 @@ int bitmap_to_xenctl_bitmap(struct xenctl_bitmap 
> > *xenctl_bitmap,
> >      uint8_t zero = 0;
> >      int err = 0;
> >      unsigned int xen_bytes = DIV_ROUND_UP(nbits, BITS_PER_BYTE);
> > -    uint8_t *bytemap = xmalloc_array(uint8_t, xen_bytes);
> > -
> > -    if ( !bytemap )
> > -        return -ENOMEM;
> > +    uint8_t last;
> >  
> >      guest_bytes = DIV_ROUND_UP(xenctl_bitmap->nr_bits, BITS_PER_BYTE);
> >      copy_bytes  = min(guest_bytes, xen_bytes);
> >  
> > -    bitmap_long_to_byte(bytemap, bitmap, nbits);
> > +    if ( IS_ENABLED(__BIG_ENDIAN) )
> > +    {
> > +        uint8_t *bytemap = xmalloc_array(uint8_t, xen_bytes);
> >  
> > -    if ( copy_bytes &&
> > -         copy_to_guest(xenctl_bitmap->bitmap, bytemap, copy_bytes) )
> > -        err = -EFAULT;
> > +        if ( !bytemap )
> > +            return -ENOMEM;
> >  
> > -    xfree(bytemap);
> > +        bitmap_long_to_byte(bytemap, bitmap, nbits);
> > +        last = bytemap[nbits/8];
> 
> Same style nit as before.
> 
> > +        if ( copy_bytes &&
> 
> copy_bytes > 1
> 
> > +             copy_to_guest(xenctl_bitmap->bitmap, bytemap, copy_bytes - 1) 
> > )
> > +            err = -EFAULT;
> > +
> > +        xfree(bytemap);
> > +    }
> > +    else
> > +    {
> > +        const uint8_t *bytemap = (const uint8_t *)bitmap;
> > +        last = bytemap[nbits/8];
> > +
> > +        if ( copy_bytes &&
> > +             copy_to_guest(xenctl_bitmap->bitmap, bytemap, copy_bytes - 1) 
> > )
> > +            err = -EFAULT;
> 
> The two identical instances would imo better stay common, even if this may
> require another function-scope variable (to invoke xfree() on after the
> copy-out).

That's not possible because bytemap is defined differently in the two
cases so it has to be defined within the if block.

I addressed everything else, I'll send v3 as a separate patch


> > +    }
> > +
> > +    clamp_last_byte(&last, nbits);
> > +    if ( copy_to_guest_offset(xenctl_bitmap->bitmap, copy_bytes - 1, 
> > &last, 1) )
> > +        err = -EFAULT;
> 
> Careful here in particular when copy_bytes == 0.
> 
> Jan
>