Skip to content

arm64: dts: exynos: google: gs201-lynx: enable UFS - #91

Merged
derekross merged 1 commit into
derekross:omamobilefrom
latentoperator:omamobile-gs201-lynx-ufs
Oct 27, 2026
Merged

derekross merged 1 commit into
derekross:omamobilefrom
latentoperator:omamobile-gs201-lynx-ufs

Conversation

@latentoperator

Copy link
Copy Markdown
Contributor

What and why

Enables UFS on the Pixel 7a (lynx), so the board can mount a root filesystem from internal storage. One commit, on top of today's omamobile (the board file from #43 and the framebuffer from #44 are in).

It turns on ufs_0 and ufs_0_phy and gives the host a VCC. On stock that rail is driven from GPIO gpp0-1 and the bootloader leaves it on, as on cheetah. Like cheetah and felix, the board file describes it as an always-on fixed regulator until there is a gs201 GPIO driver. bootargs gains omamobile.ufs-probe, as on cheetah, so the initramfs reports the disks it finds.

The omarchy-mobile side (the storage guard and the UFS notes for lynx) is in derekross/omarchy-mobile#88. Tracking issue: derekross/omarchy-mobile#36.

Device and build

  • Phone: Pixel 7a (lynx), MP1.0, bootloader lynx-17.0-15199429, SK hynix HN8T05DEHKX073 (128 GB)
  • The phone runs 7.3.0-rc6-omamobile-push1-g38b5a04d67e0 from slot B, with the Arch root on userdata. That build is on an older omamobile, but its UFS nodes and bootargs flag are the same as this commit's.

On the phone

  • Power mode changed to : FAST series_B G_4 L_2: HS-G4 on two lanes, as on stock.
  • Four LUNs (119G, 32M, 32M and 4M in lsblk) and their 55 partitions appear. The root filesystem is /dev/sda31 (ext4).
  • No UFS errors, aborts or resets in dmesg after 2 h 20 min up with the desktop session running.
  • No vdd-hba, vccq or vccq2 supplies are described, and ufshcd logs that it assumes them on.

Checks

tools/kernel-pr-check --build OUT --merge origin/omamobile from omarchy-mobile (d50af30), with dtschema installed:

ok - merges cleanly onto 29a8d419a (test merge b50c02527)
ok - 1 commits signed off by their authors, none by an AI
ok - subjects have subsystem prefixes
ok - checkpatch --strict: no errors (1 warnings and checks to read)
#   3a31f75eb arm64: dts: exynos: google: gs201-lynx: enable UFS
#     WARNING:LONG_LINE: line length of 139 exceeds 100 columns
ok - policy (tools/pr-check): identifiers, blobs
ok - C builds: no C files touched # SKIP
ok - 6 device trees build with W=1 and dtbs_check, no new warnings

The LONG_LINE warning is the bootargs string, which cheetah's board file also runs past 100 columns.

I used an AI assistant (Claude Code) for this; the commit carries Assisted-by: LLM.

🤖 Generated with Claude Code

Enable the UFS host and PHY. lynx's VCC is on gpp0-1 and left on by
the bootloader, as on cheetah, and there is no gs201 GPIO driver yet, so
describe it as an always-on fixed regulator. Add omamobile.ufs-probe so
the initramfs reports the disks it finds.

Tested on a Pixel 7a from slot B: the link comes up at HS-G4 on two
lanes, as on stock, and the four LUNs and their 55 partitions appear
with no UFS errors.

Assisted-by: LLM
Signed-off-by: Chris Gray <chrisgray70@gmail.com>
@derekross

Copy link
Copy Markdown
Owner

Thank you, @latentoperator. This is the right size for an enable: one commit, the board file only, and the nodes are the same as the Pixel 7 Pro's and the Fold's (ufs_0 with the always-on fixed VCC, ufs_0_phy, omamobile.ufs-probe in bootargs). The VCC matches Google's gs201-ufs.dtsi (gpp0-1, boot-on), and lynx has no UFS override of its own there. Because the node comes from gs201.dtsi, lynx also gets the gs201 host's system-suspend level 5 without anything in this commit.

Here kernel-pr-check --build --merge onto today's omamobile (29a8d41) passes: the sign-off, the prefix, checkpatch (only the bootargs long line, as on the Pixel 7 Pro), and the six device trees with W=1 and no new warnings. CI ran dtbs_check on this head, also clean. No other board's tree includes gs201-lynx.dts, so the other phones' DTBs can't change. On the safety side, the storage guard's allowlist for lynx has been on omarchy-mobile's main since the descriptor merged, and it is the same six names your notes show the guard leaving writable on the phone.

Two small notes, neither holding this:

  • The boot you describe is from g38b5a04d67e0, a local tree, and the UFS host driver has had a core change on omamobile since (47fa323c35dc, CPU latency on hosts whose clocks stay on). The hardware facts stand. When the phone next runs a kernel built from omamobile with this commit, please put that uname -r, the Power mode changed to : FAST series_B G_4 L_2 line and a dmesg | grep -ci ufshcd_err count into lynx's NOTES.
  • The Pixel 7 Pro's bootargs also carry mem_sleep_default=s2idle, because on gs201 the kernel offers deep and deep never wakes. I've asked for the package-side drop-in on drm: google: 120 Hz on the Pixel 9a (per-panel DSIM timers, gs-decon on zumapro) #88, which covers it; a later lynx bootargs commit could add the argument too.

I'll merge this as is.

@derekross
derekross merged commit 394ff7c into derekross:omamobile Oct 27, 2026
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants