Re: [PATCH 0/8] riscv: spacemit: k1: add clock reset drivers, switch to upstream DT

[Yao Zi <me@ziyao.cc>]

On Sun, May 10, 2026 at 09:44:14PM +0200, Vincent Legoll wrote:
> Hello Guodong Xu,
> 
> On 5/10/26 14:06, Guodong Xu wrote:
> > Currently U-Boot's SpacemiT K1 ships its own arch/riscv/dts/k1.dtsi and a
> > spacemit,k1-reset binding that has no node in the upstream
> > (kernel-mainline) K1 dts. After list discussion [1], this series
> > moves U-Boot to the upstream DT before further K1 work lands.
> > 
> > In the upstream DT, each K1 syscon provides both clocks and resets.
> > The kernel's clock driver spawns the reset child via the auxiliary
> > bus. This patchset does the same in U-Boot: each clock driver
> > spawns a UCLASS_RESET sibling on the same ofnode.
> > 
> > Patch 1/8 adds the K1 clock driver.
> > Patch 2/8 adds a new syscon-bound reset driver with no DT of_match. At this
> >            point, the legacy reset driver and local dts are still there.
> > Patch 3/8 has each per-syscon clock driver spawn the reset child on the
> >            same ofnode, the way the kernel uses auxiliary devices.
> > Patch 4/8 enables the clock driver.
> > Patch 5/8 switches BPI-F3 to dts/upstream/.
> > Patch 6/8, 7/8 and 8/8 clean up the legacy reset driver and local dts files.
> > 
> > Old and new reset drivers coexist from 2/8 through 6/8 so the series
> > is bisectable.
> > 
> > The K1 SPL bring-up [2] and PIN/SPI [3] series will be rebased on
> > this and resent once it lands.
> > 
> > Link: https://lore.kernel.org/u-boot/20260507140558.GU1614990@bill-the-cat/ [1]
> > Link: https://lore.kernel.org/u-boot/20260325223232.1553212-1-raymondmaoca@gmail.com/ [2]
> > Link: https://lore.kernel.org/u-boot/20260422143112.1329478-1-raymondmaoca@gmail.com/ [3]
> > 
> > [...]
> > 
> > ---
> > base-commit: 24db98cdf911b6ca362209e674bf9412441c1095
> > change-id: 20260510-b4-k1-clk-reset-upstream-dts-cbf1f3a79cfb
> 
> 
> I tried to apply this locally but failed:
> 
> $ git log --oneline -1
> 24db98cdf91 (HEAD -> test) Merge tag 'u-boot-imx-next-20260318' of
> https://gitlab.denx.de/u-boot/custodians/u-boot-imx into next
> 
> $ git status
> On branch test
> nothing to commit, working tree clean
> 
> $ wget https://lore.kernel.org/u-boot/20260510-b4-k1-clk-reset-upstream-dts-v1-0-db0b0503ee44@riscstar.com/t.mbox.gz
> \
> 
>  -O ~/Downloads/PATCH-0-8-riscv-spacemit-k1-add-clock-reset-drivers-switch-to-upstream-DT.mbox.gz
> 
> 
> $ gzip -dc ~/Downloads/PATCH-0-8-riscv-spacemit-k1-add-clock-reset-drivers-switch-to-upstream-DT.mbox.gz
> | git am
> Applying: riscv: spacemit: k1: add clock reset drivers, switch to upstream
> DT
> error: No valid patches in input (allow with "--allow-empty")
> Patch failed at 0001 riscv: spacemit: k1: add clock reset drivers, switch to
> upstream DT
> hint: Use 'git am --show-current-patch=diff' to see the failed patch
> hint: When you have resolved this problem, run "git am --continue".
> hint: If you prefer to skip this patch, run "git am --skip" instead.
> hint: To restore the original branch and stop patching, run "git am
> --abort".
> hint: Disable this message with "git config set advice.mergeConflict false"
> 
> I had a similar failure with the mbox downloaded with:

Because the first mail in the mbox file is the cover letter, not a real
patch, thus contains no code changes and couldn't be applied by default.

Try

	b4 shazam 20260510-b4-k1-clk-reset-upstream-dts-v1-0-db0b0503ee44@riscstar.com

instead, which applies the series cleanly at least on my side. With
git am you might skip application of the first mail manually by

	git am --skip

and continue with rest of the series.

> $ b4 mbox https://lore.kernel.org/u-boot/20260510-b4-k1-clk-reset-upstream-dts-v1-0-db0b0503ee44@riscstar.com/T/#t
> 
> 
> I also tried to apply to master branch without success
> 
> Is there something I'm doing wrong ?
> 
> Regards
> 

Regards,
Yao Zi