Hi Tom,
On Tue, 15 Dec 2020 at 09:28, Simon Glass sjg@chromium.org wrote:
Hi Tom,
On Tue, 15 Dec 2020 at 07:06, Tom Rini trini@konsulko.com wrote:
On Mon, Dec 14, 2020 at 08:50:53AM -0700, Simon Glass wrote:
Hi Tom,
https://gitlab.denx.de/u-boot/custodians/u-boot-dm/-/pipelines/5567
Note this is for the 'next' tree.
It includes the rename series which would be good to get in early.
Regards, Simon
The following changes since commit ddaa94978583d07ec515e7226e397221d8cc44c8:
Merge tag 'efi-next' of https://gitlab.denx.de/u-boot/custodians/u-boot-efi into next (2020-12-10 13:54:33 -0500)
are available in the Git repository at:
git://git.denx.de/u-boot-dm.git tags/dm-pull-14dec20
for you to fetch changes up to b7bbd553de0d9752f919dfc616f560f6f2504c14:
checkpatch: Add warnings for unexpected struct names (2020-12-13 16:51:09 -0700)
In general, this is very good and we see small reductions almost everywhere. In a few cases however we see: 22: dm: core: Combine the flattree and livetree binding code aarch64: (for 1/1 boards) all -212.0 spl/u-boot-spl:all +40.0 spl/u-boot-spl:text +40.0 text -212.0 px30-core-ctouch2-px30: all -212 spl/u-boot-spl:all +40 spl/u-boot-spl:text +40 text -212 u-boot: add: 1/-1, grow: 1/-3 bytes: 124/-336 (-212) function old new delta ofnode_is_enabled - 92 +92 dm_scan_fdt_node 196 228 +32 dm_scan_fdt 52 32 -20 dm_scan_fdt_dev 104 20 -84 dm_extended_scan_fdt 236 132 -104 static.dm_scan_fdt_live 128 - -128 spl-u-boot-spl: add: 3/0, grow: 0/-4 bytes: 116/-76 (40) function old new delta ofnode_next_subnode - 40 +40 ofnode_first_subnode - 40 +40 ofnode_is_enabled - 36 +36 dm_scan_fdt 20 16 -4 dm_scan_fdt_dev 36 20 -16 dm_scan_fdt_node 168 148 -20 dm_extended_scan_fdt 168 132 -36
Is there anything we can do about that? That said:
The problem here is that we are now using ofnode versions of those three functions. The only thing I can do is inline them. It does make things a little more complicated, but I think it is worth it. I'll send a patch.
Applied to u-boot/next, thanks!
I'll take a look. I may have missed a condition.
BTW I am hoping to get a series out around EOM with the next evolution of of-platdata. It reduces the SPL code size by a decent amount.
Regards, Simon