Your message dated Thu, 20 Feb 2025 15:17:24 +0100 (CET)
with message-id <20250220141724.df321be2...@eldamar.lan>
and subject line Closing this bug (BTS maintenance for src:linux bugs)
has caused the Debian Bug report #1032042,
regarding linux: gpio in falling-edge mode returns both falling /and/ rising
edge events and /both/ are tagged as falling [Raspberry Pi Zero W, usb-host]
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
1032042: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1032042
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: linux-image-5.10.0-18-rpi
Version: 5.10.140-1
Severity: normal
Tags: upstream
Dear Maintainer,
In my application, I have hard-wired a USB device,
and would like to run the USB controller in host mode
to avoid needing the OTG ID pin.
On Foundation systems this is the default, and OTG mode is only enabled
with dtoverlay=dwc2 in config.txt; the upstream device trees, which
Debian uses, use OTG by default since
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=bcc76c4014dce4e3834dbd5b7f6593cbcfbfebe0
To that end, I applied the following, effectively reverting that:
-- >8 --
$ diff -u ./arch/arm/boot/dts/bcm2835-rpi-zero-w.dts{.orig,}
--- ./arch/arm/boot/dts/bcm2835-rpi-zero-w.dts.orig 2023-02-26
21:44:40.206721896 +0100
+++ ./arch/arm/boot/dts/bcm2835-rpi-zero-w.dts 2023-02-26 21:44:50.919200574
+0100
@@ -6,7 +6,7 @@
/dts-v1/;
#include "bcm2835.dtsi"
#include "bcm2835-rpi.dtsi"
-#include "bcm283x-rpi-usb-otg.dtsi"
+#include "bcm283x-rpi-usb-host.dtsi"
/ {
compatible = "raspberrypi,model-zero-w", "brcm,bcm2835";
-- >8 --
and built the device tree.
(I also did that to the 6.2-rc1 device tree as a test,
and they function identically.)
This does work, inasmuch as the device is enumerated, but in my
application I also use GPIO in GPIO_V2_LINE_FLAG_INPUT |
GPIO_V2_LINE_FLAG_BIAS_PULL_UP | GPIO_V2_LINE_FLAG_EDGE_FALLING mode
(or, per lsgpio
line 5: "GPIO5" "ORNO-OR-WE-505-gpio.c" [used, input, pull-up,
falling-edge]
line 6: "GPIO6" "ORNO-OR-WE-505-gpio.c" [used, input, pull-up,
falling-edge]
).
The GPIO lines are shorted to ground temporarily, then are let float.
Please consult the following diagram w.r.t. what happens
(for simplicity, I described the pulses as 1Hz/36ms, that doesn't matter).
time line correct host both
0s 3.3V
1s 0 falling falling falling
1s36ms 3.3V falling rising
2s 0 falling falling falling
2s36ms 3.3V falling rising
Wherein "correct" is what I observe with the upstream firmware (usb-otg),
"host" with the usb-hostised firmware,
and "both" with either firmware if I subscribe
to GPIO_V2_LINE_FLAG_INPUT |
GPIO_V2_LINE_FLAG_BIAS_PULL_UP |
GPIO_V2_LINE_FLAG_EDGE_FALLING |
GPIO_V2_LINE_FLAG_EDGE_RISING
("[used, input, pull-up,
both-edges]" in lsgpio parlance)
So, in short, "in usb-host mode, when subscribed to falling edges,
both edges are returned, but labelled as falling" I guess?
Will try to update to a sid kernel and see what falls out.
Best,
наб
-- System Information:
/etc/raspi-image-id:
image based on revision: 7dcbdbc (Drop the builds for Buster,
2021-12-06) and build on 2022-10-12 12:35 (UTC)
Raspberry Pi Zero W
signature.asc
Description: PGP signature
--- End Message ---
--- Begin Message ---
Hi
This bug was filed for a very old kernel or the bug is old itself
without resolution.
If you can reproduce it with
- the current version in unstable/testing
- the latest kernel from backports
please reopen the bug, see https://www.debian.org/Bugs/server-control
for details.
Regards,
Salvatore
--- End Message ---
