[PATCH v5 00/44] More tidy-ups of Kconfig options

[Simon Glass]

(trimming cc list)

Hi Tom,

On Sat, 11 Mar 2023 at 07:29, Tom Rini <trini at konsulko.com> wrote:
>
> On Mon, Mar 06, 2023 at 11:20:42AM -0700, Simon Glass wrote:
> > Hi Tom,
> >
> > On Fri, 3 Mar 2023 at 16:43, Tom Rini <trini at konsulko.com> wrote:
> > >
> > > On Wed, Feb 22, 2023 at 09:33:41AM -0700, Simon Glass wrote:
> > >
> > > > This series was split out of the old 'split config' splc series. It
> > > > contains clean-up patches which do not depend on split config.
> > > >
> > > > This is available at u-boot-dm/spld-working
> > > >
> > > > The size changes look pretty good: https://paste.debian.net/1271742/
> > > >
> > > > The remaining patches will move into a new 'splg' series (G for Good).
> > > >
> > > > [1] https://patchwork.ozlabs.org/project/uboot/list/?series=341504&state=*
> > > >
> > > > Changes in v5:
> > > > - Fix reply typo
> > > > - Change approach and expand notes after more investigation
> > > > - Drop FSL_ISBC_KEY_EXT patch as it changes the size
> > > > - Drop PHY_CADENCE_SIERRA patch as it changes the size
> > > >
> > > > Changes in v4:
> > > > - Avoid use of def_bool
> > > > - Modify to get rid of def_bool
> > > > - Adjust Kconfig ordering
> > > > - Just fix the typo
> > > > - Reduce and rename commit
> > > > - Reduce and rename commit
> > > > - Fix 'wanderboard' typo
> > > > - Reduce and rename commit
> > > >
> > > > Changes in v3:
> > > > - Add new patch to disable QFW bootmeth in SPL
> > > > - Move the option down to the non-SPL part of drivers/Makefile
> > > > - Correct 'VPL' typo
> > > > - Use a consistent format for the comment
> > > > - Fix a transitory build error with sandbox_spl
> > > > - Add a new patch to disallow commands in SPL
> > > >
> > > > Changes in v2:
> > > > - Rebase to previous series
> > > >
> > > > Simon Glass (44):
> > > >   mtd: Drop unused kb9202_nand driver
> > > >   mtd: Drop unused CONFIG_ONENAND_U_BOOT
> > > >   sh4: Drop unused twl6030 driver
> > > >   moveconfig: Update to detect / correct missing SPL Kconfigs
> > > >   bootstd: Disable QFW bootmeth in SPL
> > > >   Correct SPL uses of ARCH_MVEBU
> > > >   Correct SPL uses of DISPLAY_AER_FULL
> > > >   Correct SPL uses of MULTIPLEXER
> > > >   Correct SPL use of PG_WCOM_UBOOT_UPDATE_SUPPORTED
> > > >   Correct SPL uses of PHY_FIXED
> > > >   boot: Add Kconfigs for BOOTMETH_VBE_REQUEST
> > > >   Correct SPL use of DM_RNG
> > > >   lib: Add a Kconfig for SPL_BZIP2
> > > >   moveconfig: Various minor improvements
> > > >   sandbox: Expand size for VPL image
> > > >   event: Add Kconfig options for SPL
> > > >   bootstd: Correct 'VPL' typo
> > > >   env: Avoid checking ENV_IS_IN when env disabled
> > > >   env: Allow VPL environment to be nowhere
> > > >   lib: Add VPL options for SHA1 and SHA256
> > > >   x86: Use string functions for all 32-bit builds
> > > >   lib: Fix build condition for tiny-printf
> > > >   sandbox: Tidy up RTC options
> > > >   sandbox: Use the generic VPL option to enable VPL
> > > >   sandbox: Tidy up I2C options
> > > >   fixdep: Add support for VPL
> > > >   fixdep: Refactor to make testing easier
> > > >   fixdep: Add some tests for parse_config_line()
> > > >   test: Add SPL versions of the TEST_KCONFIG options
> > > >   lib: Add an SPL config for LIB_UUID
> > > >   test: Tidy up sandbox handling in test-main
> > > >   x86: Fix up use of X86_32BIT_INIT and X86_64 options
> > > >   Add VPL options for BLOBLIST
> > > >   rockchip: Avoid checking environment without ENV_SUPPORT
> > > >   freescale: Drop old pre-DM_ETH code
> > > >   imx: Use SATA instead of CMD_SATA
> > > >   net: Add an SPL config for atheros
> > > >   freescale: Fix odd use of ESDHCI_QUIRK_BROKEN_TIMEOUT_VALUE
> > > >   serial: Support ns16550 driver in TPL
> > > >   dm: Add a TPL symbol for simple-bus
> > > >   x86: coral: Add missing TPL options
> > > >   power: wandboard: Add a missing CONFIG
> > > >   venice: Simplify conditions for network init
> > > >   command: Don't allow commands in SPL
> > >
> > > I've applied almost all of this series. I've left the moveconfig.py
> > > stuff for now as I'm not sure it's useful until split config. I've
> > > followed up about the X86_32BIT_INIT patch as I do want to see that
> > > boot-tested, and I've left the ns16650 TPL patch as both that feel like
> > > "split config makes this an issue" rather than an issue we have today
> > > and fwiw, the dependencies are wrong in that there's no TPL for omap2
> > > which I didn't reply directly to, I only noticed as reviewing the series
> > > locally before merging.
> >
> > OK, thanks for that. Do you think it is worth picking up the fixdep
> > ones also? The VPL thing is a bug and the others are for tests.
> >
> > I will see if I can get some time by next week to do another spin. How
> > can we get more eyes on the Kconfig-language proposal?
>
> Sorry, I should have chimed in sooner. I was thinking perhaps the next
> step would be to re-sync our Kconfig support with 6.2 or whatever, and
> see what the reaction is to the refactoring patches.

OK. I don't have much of a clue about this. I thought perhaps you
might have more idea.

I'll send a patch which syncs it. It sort-of builds a few boards but
is not right.

Regards,
Simon