From 651eb9d464db0e33d905c87abbd237402357a27a Mon Sep 17 00:00:00 2001 From: Craig Jennings Date: Mon, 16 Nov 2026 15:52:24 -0600 Subject: docs: correct the strix firmware signal and add the tty1 autostart test Strix Halo's SMU firmware ships in the BIOS, so a linux-firmware bump can never move the SMC version. The strix-soak-watch addendum now points at BIOS updates instead. The manual testing doc gains the check for the tightened Hyprland autostart guard. --- docs/workflows/strix-soak-watch.org | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'docs/workflows/strix-soak-watch.org') diff --git a/docs/workflows/strix-soak-watch.org b/docs/workflows/strix-soak-watch.org index b86921d..d4fc4d7 100644 --- a/docs/workflows/strix-soak-watch.org +++ b/docs/workflows/strix-soak-watch.org @@ -65,6 +65,8 @@ The two adjacent allocation-fix patches (0002, 0003) may land separately from Cr The canonical fix arrives as a PMFW/SMU firmware update, NOT a kernel patch. +Correction (2026-11-16): on Strix Halo the SMU firmware (the PMFW that reports as SMC 100.6.0) ships inside the system BIOS, not in =linux-firmware=. =amdgpu/smu_14_0_2.bin= belongs to the RDNA4 discrete GPU. So a =linux-firmware= bump can never move the SMC version, and the check below can't fire through it. The real vector is a BIOS update via =fwupdmgr= (ratio had System Firmware 0.0.3.4 → 0.0.3.6 pending on 2026-11-16): re-read SMC after any BIOS flash. =linux-firmware= still carries the Strix VPE/VCN/MES/PSP blobs. The 20261019 package replaced =vpe_6_1_1.bin= with no release notes, so note VPE firmware changes as related-but-unproven. + #+begin_src bash # Pre-bump SMC version (from running system, BEFORE topgrade) sudo cat /sys/kernel/debug/dri/*/amdgpu_firmware_info | grep "SMC feature" @@ -79,7 +81,7 @@ pacman -Ql linux-firmware | grep -iE "smu_14_0_2|psp_14_0|strix" # Reboot is required to load the new firmware. Re-check SMC version post-reboot. #+end_src -A bump from =100.6.0= to anything higher is the canonical Strix-fix signal — that means AMD shipped new SMU firmware. Once that lands, the workaround kernel is no longer needed (in theory; see "Outcomes" below for the validation path). +A bump from =100.6.0= to anything higher (expected after a BIOS flash, per the correction above) is the canonical Strix-fix signal — that means AMD shipped new SMU firmware. Once that lands, the workaround kernel is no longer needed (in theory; see "Outcomes" below for the validation path). ** 3. Crupi's forum thread -- cgit v1.2.3