Re: kern/58435: synaptics_aux_mid_button_scroll not working

[Robert Elz <kre%munnari.OZ.AU@localhost>]

The following reply was made to PR kern/58435; it has been noted by GNATS.

From: Robert Elz <kre%munnari.OZ.AU@localhost>
To: "Marc Coquand" <marc@mccd.space>
Cc: gnats-bugs%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost
Subject: Re: kern/58435: synaptics_aux_mid_button_scroll not working
Date: Fri, 19 Jul 2024 00:21:28 +0700

     Date:        Thu, 18 Jul 2024 10:11:19 -0500
     From:        "Marc Coquand" <marc@mccd.space>
     Message-ID:  <D2SREUPAW7DX.2069PCA6P3T4Q@mccd.space>
 
   | I tested your changes and they also work!
 
 That's good.
 
   | I am very new to kernel development and trying to learn more, could I
   | ask how you came to your solution?
 
 Reading the code, and guesswork... along with the solution that you provided.
 
 The button= line that was a little later, in combination with your changed
 value, suggested that the 5 button bits are in some way duplicated in
 the struct - the code for the 2nd case was treating them the same (what
 that line does) but the scrolling code you had a problem with wasn't.
 
 Your change swapped use of the lower bits (or one of them anyway) for the
 equivalent bit in the other set ((0x40 >> 5) == 0x2) - so it seemed likely
 to me that combining the bits for the scrolling case would work if it does
 for the other case (and this is right at the end of the function, so what
 is left in the "buttons" variable cannot matter).
 
 But someone with synaptics knowledge should check this, or at least someone
 with access to some kind of documentation.
 
 kre