Re: [PATCH 5/7] Input: libps2 - add debugging statements
From: Dmitry Torokhov
Date: Mon Jan 22 2018 - 13:33:59 EST
H Ulrik,
On Sun, Jan 21, 2018 at 09:22:31PM +0100, ulrik.debie-os@xxxxxxxxx wrote:
>
> Hi,
>
> the patch series looks nice, but best to also include the procedure
> described here in Documentation/input somewhere. I would not expect
> everyone to dig into git commit logs to find the howto :)
Right now we do not have any notes in kernel on debugging PS/2 devices,
although we have a bunch of tools (ps2emu), etc. I wonder if I can get
Benjamin or Lyude or Hans or Peter write something up as they've been
doing a lot of touchpad bringups...
Thanks.
>
> Best regards,
> Ulrik
>
>
> On Fri, Jan 19, 2018 at 11:41:09AM -0800, Dmitry Torokhov wrote:
> > Date: Fri, 19 Jan 2018 11:41:09 -0800
> > From: Dmitry Torokhov <dmitry.torokhov@xxxxxxxxx>
> > To: Benjamin Tissoires <benjamin.tissoires@xxxxxxxxxx>, Hans de Goede
> > <hdegoede@xxxxxxxxxx>, Lyude Paul <lyude@xxxxxxxxxx>
> > Cc: linux-input@xxxxxxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx
> > Subject: [PATCH 5/7] Input: libps2 - add debugging statements
> > X-Mailer: git-send-email 2.16.0.rc1.238.g530d649a79-goog
> > X-Mailing-List: linux-input@xxxxxxxxxxxxxxx
> >
> > Debugging via i8042.debug and analyzing raw PS/2 data stream may be
> > cumbersome as you need to locate the boundaries of commands, decipher the
> > sliced commands, etc, etc. Let's add a bit more high level debug statements
> > for ps2_sendbyte(), ps2_command(), and ps2_sliced_command().
> >
> > We do not introduce a new module parameter, but rater rely on the kernel
> > having dynamic debug facility enabled (which most everyone has nowadays).
> > Enable with:
> >
> > echo "file libps2.c +pf" > /sys/kernel/debug/dynamic_debug/control
> >
> > or add "libps2.dyndbg=+pf" to the kernel command line.
> >
> > Signed-off-by: Dmitry Torokhov <dmitry.torokhov@xxxxxxxxx>
> > ---
> > drivers/input/serio/libps2.c | 52 +++++++++++++++++++++++++++++++++-----------
> > 1 file changed, 39 insertions(+), 13 deletions(-)
> >
> > diff --git a/drivers/input/serio/libps2.c b/drivers/input/serio/libps2.c
> > index e96ae477f0b56..82befae4dab04 100644
> > --- a/drivers/input/serio/libps2.c
> > +++ b/drivers/input/serio/libps2.c
> > @@ -26,22 +26,20 @@ MODULE_AUTHOR("Dmitry Torokhov <dtor@xxxxxxx>");
> > MODULE_DESCRIPTION("PS/2 driver library");
> > MODULE_LICENSE("GPL");
> >
> > -/*
> > - * ps2_sendbyte() sends a byte to the device and waits for acknowledge.
> > - * It doesn't handle retransmission, though it could - because if there
> > - * is a need for retransmissions device has to be replaced anyway.
> > - *
> > - * ps2_sendbyte() can only be called from a process context.
> > - */
> > -
> > -int ps2_sendbyte(struct ps2dev *ps2dev, u8 byte, unsigned int timeout)
> > +static int ps2_do_sendbyte(struct ps2dev *ps2dev, u8 byte, unsigned int timeout)
> > {
> > + int error;
> > +
> > serio_pause_rx(ps2dev->serio);
> > ps2dev->nak = 1;
> > ps2dev->flags |= PS2_FLAG_ACK;
> > serio_continue_rx(ps2dev->serio);
> >
> > - if (serio_write(ps2dev->serio, byte) == 0)
> > + error = serio_write(ps2dev->serio, byte);
> > + if (error)
> > + dev_dbg(&ps2dev->serio->dev,
> > + "failed to write %#02x: %d\n", byte, error);
> > + else
> > wait_event_timeout(ps2dev->wait,
> > !(ps2dev->flags & PS2_FLAG_ACK),
> > msecs_to_jiffies(timeout));
> > @@ -52,6 +50,24 @@ int ps2_sendbyte(struct ps2dev *ps2dev, u8 byte, unsigned int timeout)
> >
> > return -ps2dev->nak;
> > }
> > +
> > +/*
> > + * ps2_sendbyte() sends a byte to the device and waits for acknowledge.
> > + * It doesn't handle retransmission, though it could - because if there
> > + * is a need for retransmissions device has to be replaced anyway.
> > + *
> > + * ps2_sendbyte() can only be called from a process context.
> > + */
> > +
> > +int ps2_sendbyte(struct ps2dev *ps2dev, u8 byte, unsigned int timeout)
> > +{
> > + int retval;
> > +
> > + retval = ps2_do_sendbyte(ps2dev, byte, timeout);
> > + dev_dbg(&ps2dev->serio->dev, "%02x - %x\n", byte, ps2dev->nak);
> > +
> > + return retval;
> > +}
> > EXPORT_SYMBOL(ps2_sendbyte);
> >
> > void ps2_begin_command(struct ps2dev *ps2dev)
> > @@ -186,6 +202,7 @@ int __ps2_command(struct ps2dev *ps2dev, u8 *param, unsigned int command)
> > unsigned int receive = (command >> 8) & 0xf;
> > int rc = -1;
> > int i;
> > + u8 send_param[16];
> >
> > if (receive > sizeof(ps2dev->cmdbuf)) {
> > WARN_ON(1);
> > @@ -197,6 +214,8 @@ int __ps2_command(struct ps2dev *ps2dev, u8 *param, unsigned int command)
> > return -1;
> > }
> >
> > + memcpy(send_param, param, send);
> > +
> > serio_pause_rx(ps2dev->serio);
> > ps2dev->flags = command == PS2_CMD_GETID ? PS2_FLAG_WAITID : 0;
> > ps2dev->cmdcnt = receive;
> > @@ -210,14 +229,14 @@ int __ps2_command(struct ps2dev *ps2dev, u8 *param, unsigned int command)
> > * ACKing the reset command, and so it can take a long
> > * time before the ACK arrives.
> > */
> > - if (ps2_sendbyte(ps2dev, command & 0xff,
> > - command == PS2_CMD_RESET_BAT ? 1000 : 200)) {
> > + if (ps2_do_sendbyte(ps2dev, command & 0xff,
> > + command == PS2_CMD_RESET_BAT ? 1000 : 200)) {
> > serio_pause_rx(ps2dev->serio);
> > goto out_reset_flags;
> > }
> >
> > for (i = 0; i < send; i++) {
> > - if (ps2_sendbyte(ps2dev, param[i], 200)) {
> > + if (ps2_do_sendbyte(ps2dev, param[i], 200)) {
> > serio_pause_rx(ps2dev->serio);
> > goto out_reset_flags;
> > }
> > @@ -253,6 +272,12 @@ int __ps2_command(struct ps2dev *ps2dev, u8 *param, unsigned int command)
> > ps2dev->flags = 0;
> > serio_continue_rx(ps2dev->serio);
> >
> > + dev_dbg(&ps2dev->serio->dev,
> > + "%02x [%*ph] - %x/%08lx [%*ph]\n",
> > + command & 0xff, send, send_param,
> > + ps2dev->nak, ps2dev->flags,
> > + receive, param ?: send_param);
> > +
> > return rc;
> > }
> > EXPORT_SYMBOL(__ps2_command);
> > @@ -296,6 +321,7 @@ int ps2_sliced_command(struct ps2dev *ps2dev, u8 command)
> > }
> >
> > out:
> > + dev_dbg(&ps2dev->serio->dev, "%02x - %d\n", command, retval);
> > ps2_end_command(ps2dev);
> > return retval;
> > }
> > --
> > 2.16.0.rc1.238.g530d649a79-goog
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-input" in
> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
--
Dmitry