Re: [PATCH v4 2/5] block: add partition parser for U-Boot uImage.FIT

[Matthew Wilcox]

On Thu, Nov 10, 2022 at 01:52:43AM +0000, Daniel Golle wrote:
> On Wed, Nov 09, 2022 at 05:21:01PM +0000, Matthew Wilcox wrote:
> > On Wed, Nov 09, 2022 at 02:36:11PM +0000, Daniel Golle wrote:
> > > On Wed, Nov 09, 2022 at 01:58:29PM +0000, Matthew Wilcox wrote:
> > > > ... actually, why can't you call read_part_sector() and avoid all of
> > > > this?
> > > 
> > > I've tried that before and the problem is that read_part_sector()
> > > returns a pointer to one sector (typically 512 bytes) of data.
> > > And this pointer should not be accesses beyond sector boundaries,
> > > right? You'd have to call read_part_sector() again for the next
> > > sector.
> > > 
> > > The FIT structure, however, usually exceeds the size of one sector,
> > > and having a continous memory area covering the structure as a whole
> > > is crucial for libfdt to do its job.
> > > 
> > > I could, of course, use read_part_sector() to copy all sectors
> > > covering the FIT structure into a buffer, but that seemed strange
> > > given that read_part_sector() actually used read_mapping_page()
> > > (and now uses read_mapping_folio()) internally and then returns a
> > > pointer to the offset within the page/folio. So why not read it in one
> > > piece in first place instead of having it first split up to sectors
> > > by read_part_sector() just to then having to reassemble it into a
> > > continous buffer again.
> > 
> > Are you guaranteed that it's "sufficiently" aligned on storage so
> > that it fits entirely within a single page?  If not, you'll have
> > to copy it, vmap it, or fix libfdt to handle a segmented buffer.
> 
> Yes, for the uImage.FIT to be usable for the partition parser it has
> to be page-aligned.
> 
> There is a check which makes sure that this is the case:
> > +	/* uImage.FIT should be aligned to page boundaries */
> > +	if (fit_start_sector % (1 << (PAGE_SHIFT - SECTOR_SHIFT)))
> > +		return 0;
> 
> In case of mtdblock or ubiblock devices, the image always starts at
> offset 0, so this is never a problem.
> In case of the image being stored in a GPT partition, one has to make
> sure that the start sector of the partition is page aligned, otherwise
> the above check will fail and the partition parser will bail out.

OK.  Then I think open coding it is the right idea, just with all
the cruft removed ;-)  I looked at extracting parts of
read_part_sector() into read_part_page(), but it ended up being
a two line function that wasn't terribly useful.