On Fri, 2023-06-23 at 10:37 +0200, Benjamin Tissoires wrote:

On Jun 21 2023, Greg KH wrote:

quoted

On Wed, Jun 21, 2023 at 11:42:30AM +0200, Benjamin Tissoires wrote:

quoted

Make the code looks less like Pascal.

Extract the internal code inside a helper function, fix the
initialization of the parameters used in the helper function
(`hidpp->answer_available` was not reset and `*response` wasn't
too),
and use a `do {...} while();` loop.

Fixes: 586e8fede795 ("HID: logitech-hidpp: Retry commands when
device is busy")
Cc: stable@vger.kernel.org
Signed-off-by: Benjamin Tissoires <redacted>
---
as requested by
https://lore.kernel.org/all/CAHk-=wiMbF38KCNhPFiargenpSBoecSXTLQACKS2UMyo_Vu2ww@mail.gmail.com/ (local)
This is a rewrite of that particular piece of code.
---
 drivers/hid/hid-logitech-hidpp.c | 102 +++++++++++++++++++++++--
--------------
 1 file changed, 61 insertions(+), 41 deletions(-)

diff --git a/drivers/hid/hid-logitech-hidpp.c b/drivers/hid/hid-

logitech-hidpp.c
index dfe8e09a18de..3d1ffe199f08 100644

--- a/drivers/hid/hid-logitech-hidpp.c
+++ b/drivers/hid/hid-logitech-hidpp.c

@@ -275,21 +275,20 @@ static int __hidpp_send_report(struct

hid_device *hdev,
 }
 
 /*
- * hidpp_send_message_sync() returns 0 in case of success, and
something else
- * in case of a failure.
- * - If ' something else' is positive, that means that an error
has been raised
- *   by the protocol itself.
- * - If ' something else' is negative, that means that we had a
classic error
- *   (-ENOMEM, -EPIPE, etc...)
+ * Effectively send the message to the device, waiting for its
answer.
+ *
+ * Must be called with hidpp->send_mutex locked
+ *
+ * Same return protocol than hidpp_send_message_sync():
+ * - success on 0
+ * - negative error means transport error
+ * - positive value means protocol error
  */
-static int hidpp_send_message_sync(struct hidpp_device *hidpp,
+static int __do_hidpp_send_message_sync(struct hidpp_device
*hidpp,
        struct hidpp_report *message,
        struct hidpp_report *response)

__must_hold(&hidpp->send_mutex)  ?

Good point. I'll add this in v2.

I'm still waiting for some feedback from the people who particpated
in
the original BZ, but the new bug is harder to reproduce. Anyway,
there
is no rush IMO.

The problem is only ever going to show up in very limited circumstances
after the logic fix was applied.

You need a hardware problem (such as the controller being too busy to
answer) to trigger the problems fixed by this patch. I don't see a way
to reliably reproduce it unless you inject that hardware error.