-
Notifications
You must be signed in to change notification settings - Fork 5
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[*] rtsx_usb depends badly on autosuspend #183
Comments
Same as #168 but your title is better |
SD_CN_N is pulled low with the tray, so Linux constantly tries to scan for an SD Card. It wastes power and prevents S3 entry (it's frozen after around 13 seconds). |
For anyone else tracking this issue, I removed my SD tray and now S3 entry only takes a second. |
So the tray pulls SD_CD# low constantly, right? Isn't this basically a hardware design flaw (i.e. hardware bug)? |
According to realtek, no. Tray should pull it low then the driver should timeout after 100ms. |
Okay cool, this is quite relaxing to know... Thank you! |
Pulling tray out also fixed my unreliable Suspend issue (sometimes it does suspend, sometimes it does not). I tried disabling the card reader in the coreboot - that did not work. I then disabled the USB device via I am on Fedora Silverblue 6.9.10-200.fc40.x86_64 |
I tried removing the sdcard tray and disabling the usb device, but my starlite never gets to sleep. When I tell it sleep it turns the screen black and keep the led on (no blinking just on all time like when the device is on). Whatever I wait doesn't seem to do anything until I force it to power down pressing the power button. I see a similar behavior sometimes when I leave it on for a while without touching it even though I disabled the auto suspension, it should only lock, but maybe this is a bug in gnome. Same behavior also when I turn off the device, it doesn't turn off completely it stays with the screen off and the led on until I press the power button for few seconds or it drains completely the battery. I'm using debian Trixie (testing) with kernel 6.10.6 and I just upgraded the firmware to 24.08 but the problem is still there. In Maybe a different issue? Should I open another issue? |
What did you do? Definitely different, could just be a Debian bug - are you able to test with a Live USB with Ubuntu? |
I run:
I just tried with the latest ubuntu 24.04.2 live USB and the same behavior. I waited for the desktop to be fully loaded, removed the sdcard tray and clicked on the menu to 'suspend'. The screen becomes black and the led keeps on (no blinking signaling suspension). I see the keyboard light turns on if I click on any key, but neither the power button or any key brings the starlite back to life. |
Ping support - that's something else |
I have the same issue with a long suspend switching. Workaround with unbinding the usb device from AndrzejTunkiel makes suspend instant. |
Same, I've been choosing suspend from the menu, waiting 30 secs and then doing it again to get it to sleep. Thanks for the work around with tee. |
Workaround until upstreamed: https://github.com/StarLabsLtd/rtsx_usb |
Noticed 2 things:-
So, tested last night with a MicroSD card in the slot, charged Starlite up to 75% and put it into Suspend overnight. These are still workarounds but hopefully will also provide another way to get suspend working until the software upstream is updated. |
I've got a bit different numbers. The sleep is For me that is a bit high. I'm using the |
Very nice! My |
What kernel # will this land in? |
6.13 I believe |
Today I installed Fedora 41 with kernel 6.13.5, deactivated the unbind command at boot, and restarted.... Sleep is working promptly without the workaround! Thanks. |
Entering into S3 is super slow, it takes about 15-20 seconds.
At first I thought it was caused by the iwlwifi module, but even if I remove it before sleep, it still takes a long time.
This happens both on a Fedora 40 live ISO and on an up-to-date installation (Linux 6.9.7-gentoo-dist / linux-firmware 20240610 / firmware 2024.06)
dmesg
The text was updated successfully, but these errors were encountered: