Systematic Crash when changing time scale on ChromeOS/Linux

[ythurel]

Installation / version OpenHantek6022:

OS: [Chrome OS | using linux]
Installed through flatpak or directly using available .deb, bug is the same.
Using 6022BE

Steps to reproduce the behavior:

Launch OpenHantek6022 + detect USB 6022BE.
timescale is at 1ms (default). It is possible to go lower (100us) without any issue.
As soon as i try to reach 5 ms with 2 channels active (or 10ms with 1 channel active), complete crash.
Many deb version tested, same result (3.3.3 > now)
Message indicates memory allocation issue or equivalent. Not always the same message.

Computer Env

• OpenHantek version: All version tested except the latest in work
• OS: [Chrome OS | using linux]
• Screen resolution (tested, no impact)

Scope device (please complete the following information):

• Device [e.g. Hantek 6022BE, ...]
• Program top line [e.g. OpenHantek6022 (Build ...) - Device ... (FW ...)]
• Input signal: [e.g. calibration out 1 kHz, sine wave 1 Veff @ 50 Hz, ...]
• Probe setting: [e.g. no probe, probe X1, probe X10, ...]

Additional context
Note the problem exists ONLY on linux inside Chrome OS.
With pure Debian 12 or 13, everything works fine.

[ythurel]

I tried to solve the issue without success for the moment.
This kind of message perhaps suggest (IA opinion) that USB autosuspend on chromeos is the issue.

QT_QPA_PLATFORM=xcb /usr/bin/OpenHantek
Graphic: 4.5 (Compatibility Profile) Mesa 22.3.6 - GLSL version 1.20
bulkReadMulti: Getting sample data failed: "No such device (it may have been disconnected)" OpenHantek6022
controlWrite: stop sampling failed: "No such device (it may have been disconnected)"
has stopped after 17 s
malloc(): unsorted double linked list corrupted
Aborted (core dumped)

If i select one channel, i can increase timebase more before crash (1x 10ms=crash, éx 5ms=crash). IA thinks it could come from ChromeOS Crostini limitations on USB timing, with ChromeOS blocking direct USB access to Hantek device.
Hope it could help