Timeline for How USB bus number and device number been assigned?
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 3 at 20:42 | comment | added | Angelo Dureghello | Don't take any offense, it was just a technical note, take it as a detail if you prefer. In 2016 there was plenty of SoC with integrated controller, imx51 i.e. is from 2009. It's still true that a cpu cannot directly talk with a bus, but there is no mandatory need of pci in case a SoC with integrated controller. Not sure how to modify the post, feel free to make use of this note if you want. | |
| Sep 16 at 12:39 | comment | added | grochmal | @AngeloDureghello - Thank you for the very useful comment about USB SoC integrated on a 2016 answer. If you desperately need the pedantry, please give a note that there were no commercially available SoCs in 2016. Otherwise please make yourself useful and make an edit to the answer to add info about the SoCs. Will be quite grateful for it. Wouldn't hold it - it is indeed useful info :+1: | |
| Sep 11 at 14:47 | comment | added | Angelo Dureghello | Wrong. See SoC that are cpu's with usb cpontroller integrated generally without any PCI bus. Scheme is mainly Controller(can be on pci) -> Bus -> devices. | |
| S Mar 26, 2021 at 3:08 | history | suggested | Milan | CC BY-SA 4.0 | Improved formatting and fixed grammar. |
| Mar 23, 2021 at 18:33 | review | Suggested edits | |||
| S Mar 26, 2021 at 3:08 | |||||
| Apr 4, 2019 at 7:53 | history | edited | Toby Speight | CC BY-SA 4.0 | Typos and terminology |
| Jul 26, 2016 at 18:15 | comment | added | Conan | Thank you @grochmal, I think I made mistake when test overflow. The later tests showed to me when overflow, the counter indeed searching from the lower number. | |
| Jul 26, 2016 at 18:07 | comment | added | grochmal | @Conan - Hmm... well, how can i say it: I just don't know. I never tried to overflow the device counter as you did. Then again, it is hardly needed to know a device number beforehand. For example, if you are trying to find a USB drive when it connects you should be doing it by filesystem label or UUID (which even udev more-or-less understands). As to understand the numbering for the sake of learning, i believe that the only place that has the info is the kernel code. | |
| Jul 26, 2016 at 15:00 | comment | added | Conan | so what will happen when the max number of device number is used. let's say I keep reconnecting my device. The device number will soon goes to the max. I actually did a test, looks like it is not a simple counter. It is not returning back to 001, instead the kernel is reusing the largest few device number. Is there a easy way to change this behavior? | |
| Jul 22, 2016 at 22:09 | vote | accept | Conan | ||
| Jul 22, 2016 at 4:24 | history | edited | grochmal | CC BY-SA 3.0 | typos everywhere |
| Jul 22, 2016 at 4:11 | history | answered | grochmal | CC BY-SA 3.0 |