s9 only supports a single report, first byte must be set to 0x0

[ghost]

System Information:

rscott@home ~ $ udevadm info --query=all /sys/devices/pci0000:00/0000:00:14.0/usb3/3-5/3-5.2/
P: /devices/pci0000:00/0000:00:14.0/usb3/3-5/3-5.2
M: 3-5.2
R: 2
U: usb
T: usb_device
D: c 189:272
N: bus/usb/003/017
L: 0
V: usb
E: DEVPATH=/devices/pci0000:00/0000:00:14.0/usb3/3-5/3-5.2
E: SUBSYSTEM=usb
E: DEVNAME=/dev/bus/usb/003/017
E: DEVTYPE=usb_device
E: DRIVER=usb
E: PRODUCT=1038/12c2/74
E: TYPE=0/0/0
E: BUSNUM=003
E: DEVNUM=017
E: MAJOR=189
E: MINOR=272
E: USEC_INITIALIZED=10274180
E: ID_VENDOR=SteelSeries
E: ID_VENDOR_ENC=SteelSeries
E: ID_VENDOR_ID=1038
E: ID_MODEL=SteelSeries_Arctis_9
E: ID_MODEL_ENC=SteelSeries\x20Arctis\x209
E: ID_MODEL_ID=12c2
E: ID_REVISION=0074
E: ID_SERIAL=SteelSeries_SteelSeries_Arctis_9
E: ID_BUS=usb
E: ID_USB_INTERFACES=:030000:
E: ID_VENDOR_FROM_DATABASE=SteelSeries ApS
E: ID_PATH=pci-0000:00:14.0-usb-0:5.2
E: ID_PATH_TAG=pci-0000_00_14_0-usb-0_5_2
E: ID_FOR_SEAT=usb-pci-0000_00_14_0-usb-0_5_2
E: TAGS=:seat:
E: CURRENT_TAGS=:seat:

rscott@home ~ $ sudo emerge \=dev-libs/hidapi-0.11.0 -av

These are the packages that would be merged, in order:

Calculating dependencies... done!
[ebuild   R    ] dev-libs/hidapi-0.11.0::gentoo  USE="doc (-fox)" ABI_X86="(64) -32 (-x32)" 0 KiB

Total: 1 package (1 reinstall), Size of downloads: 0 KiB

Would you like to merge these packages? [Yes/No]

rscott@home ~ $ lsusb | grep SteelSeries
Bus 003 Device 017: ID 1038:12c2 SteelSeries ApS SteelSeries Arctis 9
Bus 003 Device 019: ID 1038:12c4 SteelSeries ApS SteelSeries Arctis 9

rscott@home ~ $ uname -a
Linux home 6.0.12-zen1-x86_64 #1 ZEN SMP PREEMPT_DYNAMIC Thu Dec 29 15:31:50 GMT 2022 x86_64 Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz GenuineIntel GNU/Linux
# Gentoo Linux

I wanted to retrieve battery percentage - kept timing out.

When debugging function arctis_9_read_device_status,
I found out that Report ID '20x0',
was the same as Report DATA '20x0'.
I changed the first byte to 0x0 and it worked.

I read online that the Report ID must be 0x0 for devices which only support a single report
http://hidapi-d.dpldocs.info/hidapi.bindings.hid_write.html

I went a step further and changed all bytes to 0x0 (and swapped some with the second byte), and this worked.

I sense there is perhaps some deprecation with newer versions of hidapi?

[Sapd]

Good work!

I sense there is perhaps some deprecation with newer versions of hidapi?

Or also some change in the kernel regarding USB. Linux was/is actually very tolerable when not sending the absolutely correct package (ie it even accept it when sending the wrong length), while on the other hand Windows is very strict.

Given the commit before on the file, it seems that indeed it was not correct from the beginning. Will merge it for now (but its always good if someone also tests it on some other OS).