[U-Boot] ARM64 Allwinner Binary Size

[Mark Kettenis]

> From: Andre Przywara <andre.przywara at arm.com>
> Date: Tue, 19 Dec 2017 14:17:37 +0000
> 
> Hi,
> 
> On 19/12/17 13:51, Mark Kettenis wrote:
> >> From: Andre Przywara <andre.przywara at arm.com>
> >> Date: Tue, 19 Dec 2017 13:38:59 +0000
> >>
> >> Hi Maxime,
> >>
> >> thanks for having a look!
> >>
> >> On 19/12/17 13:12, Maxime Ripard wrote:
> >>> On Tue, Dec 05, 2017 at 10:28:20AM +0000, Andre Przywara wrote:
> >>>> So even though the actual u-boot.bin for 64-bit boards is still somewhat
> >>>> below the limit (~480KB), adding the ATF image (~32KB) pushes it over
> >>>> the edge. So since v2017.11 u-boot.itb is already too big for the
> >>>> traditional MMC env location.
> >>>
> >>> So I've had a quick look about what could go possibly go away in our
> >>> current armv8 config (using the pine64+ defconfig). Let me know if
> >>> some are actually vitals:
> >>>
> >>>  - FIT_ENABLE_SHA256_SUPPORT
> >>>  - CONSOLE_MUX
> >>>  - CMD_CRC32
> >>>  - CMD_LZMADEC
> >>>  - CMD_UNZIP
> >>>  - CMD_LOADB
> >>>  - CMD_LOADS
> >>>  - CMD_MISC (actually implementing the command sleep)
> >>>  - ISO_PARTITION (yes. For CDROMs.)
> >>
> >> As Alex mentioned, this is needed for some installer images, which come
> >> as ISOs. So if possible, we should keep this in.
> >>
> >>>  - VIDEO_BPP8, VIDEO_BPP16
> >>>  - VIDEO_ANSI
> >>>  - SHA256
> >>>  - LZMA
> >>
> >> From just looking at the names I am fine with the rest gone. But let me
> >> test tonight if there are any side effects.
> >>
> >> Some of them seem useful, but I would leave enabling them to the actual
> >> users. If someone needs it, they can enable them and loose the raw MMC
> >> environment. I think this is a fair trade-off.
> >>
> >>> Removing those options make the u-boot.itb binary size going from
> >>> 516kB to 478kB, making it functional again *and* allowing us to enable
> >>> the DT overlays that seem way more important than any feature
> >>> mentionned above (and bumps the size to 483kB).
> >>
> >> How important is the raw MMC environment for the ARM64 boards, actually?
> >> Most of the rationale for the 32-bit side seemed to apply to legacy use
> >> cases only. Do we have reports/complaints from 64-bit users?
> > 
> > For me/us (OpenBSD) the environment is still important.  I have many
> > setups where U-Boot lives on a uSD card but the installed OS lives on
> > a USB device.  In that scenario I set boot_targets to boot the EFI
> > bootloader and OS off the USB disk.  This is very helpfull for testing
> > new versions of U-Boot as I can simply swap the uSD card.  But for
> > some setups this is essential as OpenBSD doesn't support the SD/MCC
> > controller on all ARM hardware yet (but we do support it on
> > Allwinner).
> 
> I see, but I wasn't arguing for dropping the environment altogether,
> more for supporting FAT environments *only*.
> So how important is preserving existing environments over a firmware
> update in your scenario? I think this is the killer question here, isn't
> it? I'm inclined to just drop raw MMC environment support from sunxi64
> boards and then enjoy the ~450KB more worth of code, until we hit the
> first MB boundary.

I wouldn't consider preserving the environment as crucial.

> I have builds with all (DDR3) A64 board DTs in the binary [1], which
> would be larger than 504K anyway.

Oh, that's some cool work on the ATF side.  Need to check that out.

> Cheers,
> Andre.
> 
> [1] https://github.com/apritzel/pine64/commit/ee12bea43