[PATCH v2 02/17] boot: Try all bootmeths on the final partition

[Tom Rini]

On Wed, Oct 08, 2025 at 10:52:17AM -0600, Simon Glass wrote:
> Hi Tom,
> 
> On Wed, 8 Oct 2025 at 10:28, Tom Rini <trini at konsulko.com> wrote:
> >
> > On Wed, Oct 08, 2025 at 10:20:43AM -0600, Simon Glass wrote:
> > > Hi Tom,
> > >
> > > On Wed, 8 Oct 2025 at 08:57, Tom Rini <trini at konsulko.com> wrote:
> > > >
> > > > On Wed, Oct 08, 2025 at 08:45:54AM -0600, Simon Glass wrote:
> > > > > Hi Tom,
> > > > >
> > > > > On Wed, 8 Oct 2025 at 08:34, Tom Rini <trini at konsulko.com> wrote:
> > > > > >
> > > > > > On Wed, Oct 08, 2025 at 06:53:44AM -0600, Simon Glass wrote:
> > > > > > > Hi Tom,
> > > > > > >
> > > > > > > On Tue, 7 Oct 2025 at 12:13, Tom Rini <trini at konsulko.com> wrote:
> > > > > > > >
> > > > > > > > On Wed, Oct 01, 2025 at 03:26:27PM -0600, Simon Glass wrote:
> > > > > > > >
> > > > > > > > > At present when one bootmeth fails on the final partition, the next
> > > > > > > > > bootmeth is not tried. Adjust the logic to go to the next bootmeth,
> > > > > > > > > which is the more natural behaviour.
> > > > > > > > >
> > > > > > > > > Signed-off-by: Simon Glass <sjg at chromium.org>
> > > > > > > >
> > > > > > > > My feedback on v1:
> > > > > > > > https://lore.kernel.org/u-boot/20250331174144.GI93000@bill-the-cat/
> > > > > > > > was that we need to do a documentation change when doing a behavior
> > > > > > > > change. And I cannot recall if this change here is because of reports of
> > > > > > > > problems or not.
> > > > > > >
> > > > > > > This is a long time ago, but I think it might have been an
> > > > > > > inconsistency that I found myself. Otherwise I would probably have
> > > > > > > mentioned the person who reported it.
> > > > > > >
> > > > > > > The documentation already implies the logic in this patch, as I see
> > > > > > > it. But please suggest any changes.
> > > > > >
> > > > > > OK, I guess it's because the commit message doesn't explain the bug
> > > > > > clearly enough. Is this a correct description of the patch:
> > > > > >
> > > > > > At present, normally when one bootmeth fails on a partition, we move on
> > > > > > and try the next bootmeth. However, this was not the case for the final
> > > > > > partition due to a bug. Rework the logic so that all partitions are
> > > > > > treated the same.
> > > > >
> > > > > We can certainly say that if you like. It is fair to say that the
> > > > > algorithm is still being evolved as new things come up. Most of the
> > > > > behaviour has been driven by new tests that I've written. The global
> > > > > bootmeth is the biggest change since hunters were introduced, but I
> > > > > suspect there will be more in future.
> > > > >
> > > > > I'm heading out later today but can probably send v3 series (with the
> > > > > tweaks you've given) tomorrow. After that I'm not sure as I'm on
> > > > > holiday for two weeks.
> > > >
> > > > Peter's feedback should be addressed too for a v3 (and not just set
> > > > aside).
> > >
> > > Is that the email just received, or something from a different series
> > > that I maybe missed from a while back?
> >
> > It was yesterday:
> > https://lore.kernel.org/u-boot/CALeDE9Ovemog2AGN_Li8Ju12QKHOC0D30V2=KG00f0KsQ5aPkg@mail.gmail.com/T/#m34f6e2c51fdfb275ba9930bfdc2ff349ee542b51
> 
> OK, I replied. Would you like me to send the 5 separate patches he
> mentions as a separate series?

I think given what Andre said on the call (and that he'll test v2 to see
if it addresses things) and what you said to Peter now, there's some
misunderstandings of what the issues are still. So yes, splitting just
those patches out to start with (and not reposting the rest) would be a
good next step.

-- 
Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 228 bytes
Desc: not available
URL: <https://lists.denx.de/pipermail/u-boot/attachments/20251008/f9b3c6bb/attachment.sig>