Re: [PATCH v3 02/18] system/memory: Better describe @plen argument of flatview_translate()

[Philippe Mathieu-Daudé <philmd@xxxxxxxxxx>]

Hi Thomas,

On 30/9/25 10:24, Thomas Huth wrote:
> On 30/09/2025 10.21, Philippe Mathieu-Daudé wrote:
> > flatview_translate()'s @plen argument is output-only and can be NULL.
> >
> > When Xen is enabled, only update @plen_out when non-NULL.
> >
> > Signed-off-by: Philippe Mathieu-Daudé <philmd@xxxxxxxxxx>
> > ---
> >   include/system/memory.h | 5 +++--
> >   system/physmem.c        | 9 +++++----
> >   2 files changed, 8 insertions(+), 6 deletions(-)
> >
> > diff --git a/include/system/memory.h b/include/system/memory.h
> > index aa85fc27a10..3e5bf3ef05e 100644
> > --- a/include/system/memory.h
> > +++ b/include/system/memory.h
> > @@ -2992,13 +2992,14 @@ IOMMUTLBEntry 
> > address_space_get_iotlb_entry(AddressSpace *as, hwaddr addr,
> >    * @addr: address within that address space
> >    * @xlat: pointer to address within the returned memory region 
> > section's
> >    * #MemoryRegion.
> > - * @len: pointer to length
> > + * @plen_out: pointer to valid read/write length of the translated 
> > address.
> > + *            It can be @NULL when we don't care about it.
> >    * @is_write: indicates the transfer direction
> >    * @attrs: memory attributes
> >    */
> >   MemoryRegion *flatview_translate(FlatView *fv,
> >                                    hwaddr addr, hwaddr *xlat,
> > -                                 hwaddr *len, bool is_write,
> > +                                 hwaddr *plen_out, bool is_write,
> >                                    MemTxAttrs attrs);
> >   static inline MemoryRegion *address_space_translate(AddressSpace *as,
> > diff --git a/system/physmem.c b/system/physmem.c
> > index 8a8be3a80e2..86422f294e2 100644
> > --- a/system/physmem.c
> > +++ b/system/physmem.c
> > @@ -566,7 +566,7 @@ iotlb_fail:
> >   /* Called from RCU critical section */
> >   MemoryRegion *flatview_translate(FlatView *fv, hwaddr addr, hwaddr 
> > *xlat,
> > -                                 hwaddr *plen, bool is_write,
> > +                                 hwaddr *plen_out, bool is_write,
> >                                    MemTxAttrs attrs)
> >   {
> >       MemoryRegion *mr;
> > @@ -574,13 +574,14 @@ MemoryRegion *flatview_translate(FlatView *fv, 
> > hwaddr addr, hwaddr *xlat,
> >       AddressSpace *as = NULL;
> >       /* This can be MMIO, so setup MMIO bit. */
> > -    section = flatview_do_translate(fv, addr, xlat, plen, NULL,
> > +    section = flatview_do_translate(fv, addr, xlat, plen_out, NULL,
> >                                       is_write, true, &as, attrs);
> >       mr = section.mr;
> > -    if (xen_enabled() && memory_access_is_direct(mr, is_write, attrs)) {
> > +    if (xen_enabled() && plen_out && memory_access_is_direct(mr, 
> > is_write,
> > +                                                             attrs)) {
> >           hwaddr page = ((addr & TARGET_PAGE_MASK) + TARGET_PAGE_SIZE) 
> > - addr;
> > -        *plen = MIN(page, *plen);
> > +        *plen_out = MIN(page, *plen_out);
> >       }
> My question from the previous version is still unanswered:
>
> https://lore.kernel.org/qemu- 
> devel/22ff756a-51a2-43f4-8fe1-05f17ff4a371@xxxxxxxxxx/
This patches
- checks for plen not being NULL
- describes it as
  "When Xen is enabled, only update @plen_out when non-NULL."
- mention that in the updated flatview_translate() documentation
  "It can be @NULL when we don't care about it." as documented for
  the flatview_do_translate() callee in commit d5e5fafd11b ("exec:
  add page_mask for flatview_do_translate")

before:

  it was not clear whether we can pass plen=NULL without having
  to look at the code.

after:

  no change when plen is not NULL, we can pass plen=NULL safely
  (it is documented).

I shouldn't be understanding your original question, do you mind
rewording it?

Thanks,

Phil.