[Buildroot] [PATCH v9 03/11] boot/uboot: add support for building the TI K3 DM into U-Boot

[Yann E. MORIN]

Andreas, All,

On 2023-06-25 09:02 +0200, Yann E. MORIN spake thusly:
> On 2023-06-22 11:02 -0500, Andreas Dannenberg via buildroot spake thusly:
[--SNIP--]
> > +ifeq ($(BR2_TARGET_UBOOT_NEEDS_TI_K3_DM),y)
> > +# Currently supports the FW from Git tag 08.06.00.006 by default
> > +TI_K3_DM_VERSION = 340194800a581baf976360386dfc7b5acab8d948
> > +TI_K3_DM_SITE = https://git.ti.com/processor-firmware/ti-linux-firmware/blobs/raw/$(TI_K3_DM_VERSION)/ti-dm/$(BR2_TARGET_UBOOT_TI_K3_DM_SOC)
> > +TI_K3_DM_SOURCE = ipc_echo_testb_mcu1_0_release_strip.xer5f
> > +# This is not really nice but disable the hash check for the DM FW file. Main
> > +# reason is all those DM FW files for different SoCs have the same(!) name in
> > +# the Git repository they reside in, so it would be more difficult to distinguish
> > +# between them for hash checking purposes. To work around this let's just
> > +# rely and trust the official Git repo at ti.com is known-good which is also
> > +# accessed through a secure transport.
> > +BR_NO_CHECK_HASH_FOR += $(TI_K3_DM_SOURCE)
> > +UBOOT_EXTRA_DOWNLOADS = $(TI_K3_DM_SITE)/$(TI_K3_DM_SOURCE)
> > +UBOOT_MAKE_OPTS += DM=$(UBOOT_DL_DIR)/$(TI_K3_DM_SOURCE)
> > +endif
> 
> I again am not very fond of this injecting extra downloads n uboot. I'd
> rather we have a separate package, ti-linux-firmware. That way, it can
> be used by both uboot and ti-k3-image-gen.
> 
> The ti-linux-firmware repository is not small (git clone reports 750MiB
> today) but the repository exposes snapshots, so that is a bit more easy
> on the networks connection (althoiugh we should take the .gz, not the
> .xz, as the on-the-fly compression is much slower for xz than gz).

Well, the .tar.gz is 432MiB, so not so much an improvement over the
750MiB of the repository, and the same download size will have to be
done again when we later bump the version, so I think it is better to
do a git clone...

The other pro of doing a git clone, is that we do have a known-working
way of generating reproducible arhives locally, while relying on the
remote site do generate the archive risk generating non-reproducible
archives (e.g. if they update their git version, there is no guarantee
the archive is reproducible across git versions; we've been bitten quite
hard early in the year when github (and now gitlab?) broke the whole
world that was checking hashes).

Regards,
Yann E. MORIN.

> Then it is again just a matter of ti-linux-firmware copying the
> appropriate blobs into $(BINARIES_DIR) and then uboot grabbing what's
> necessary.
> 
> Regards,
> Yann E. MORIN.
> 
> >  ifeq ($(BR2_TARGET_UBOOT_NEEDS_OPENSBI),y)
> >  UBOOT_DEPENDENCIES += opensbi
> >  UBOOT_MAKE_OPTS += OPENSBI=$(BINARIES_DIR)/fw_dynamic.bin
> > -- 
> > 2.34.1
> > 
> > _______________________________________________
> > buildroot mailing list
> > buildroot at buildroot.org
> > https://lists.buildroot.org/mailman/listinfo/buildroot
> 
> -- 
> .-----------------.--------------------.------------------.--------------------.
> |  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
> | +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
> | +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
> | http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
> '------------------------------^-------^------------------^--------------------'
> _______________________________________________
> buildroot mailing list
> buildroot at buildroot.org
> https://lists.buildroot.org/mailman/listinfo/buildroot

-- 
.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 561 099 427 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'