memory-leaks

commit cb4aa84e13bc99d0983608712ea3be8790e54af6
parent 4fae3ce112ee4037baee65b57318b5edf5224e82
Author: Kévin Le Gouguec <kevin.legouguec@gmail.com>
Date:   Thu,  8 Jan 2026 08:38:08 +0100

Put something resembling closure on the record

I know things could have been worse, but it still feels like a Bad
Ending.

Diffstat:
M
guides/sysadmin/machines/amdahl30/killing-time.org
 | 
22
++++++++++++++++++++++

1 file changed, 22 insertions(+), 0 deletions(-)
diff --git a/guides/sysadmin/machines/amdahl30/killing-time.org b/guides/sysadmin/machines/amdahl30/killing-time.org
@@ -483,6 +483,28 @@ tl;dr
     - except =libdrm_radeon1-32bit=: installed; no changes
   - installed proton-ge: no changes
 - 6.11.7 → 6.11.8 update had [[https://build.opensuse.org/request/show/1224199][a lot going on]]
+* The unsatisfying ending?
+On March 16 2025, installed Debian (then Bookworm) on the NVMe disk
+(KDE Plasma, Wayland, kernel 6.1).
+
+Installed Steam; installed Hades Ⅱ; game ran great; played like no
+tomorrow.  Hissed at Trixie, stubbornly kept playing on Bookworm.  All
+the while, kept daily-driving the Tumbleweed installation on the SATA
+disk; just rebooting into my "BookwormCube" for Hades Ⅱ sessions.
+
+In September–October, caught & handled =SIGSKONG= on Tumbleweed; from
+October 25 through November, rebooted into the BookwormCube® to play
+Hades Ⅱ's endgame & chase cheevos.
+
+Bottomline: ending meh, comparatively with Hades Ⅰ?  Game still rules.
+
+Tech-wise, for all my efforts debugging this stutter, I learned
+nothing, beside a truckload of ineffective system knobs.  Last checked
+Tumbleweed's performance [[https://gitlab.freedesktop.org/mesa/mesa/-/issues/11300#note_3003120][in August]] (kernel 6.15.6;
+kernel-firmware-amdgpu 20250708; Mesa 25.1.5 & 25.3.0-devel with
+=nogttspill=); no improvements.
+
+/Fin/ 🎬
 * Dramatis parametra
 There is not a single parameter I can rule out, months into this
 farce.  Condensed summary of the whole affair: