diff options
author | Dmitry Torokhov <dmitry.torokhov@gmail.com> | 2015-12-03 00:50:42 (GMT) |
---|---|---|
committer | Dmitry Torokhov <dmitry.torokhov@gmail.com> | 2015-12-17 23:24:04 (GMT) |
commit | 24a06f3e3ab640490f49c8e8812d9550d6f15e92 (patch) | |
tree | 190b929c1373bdfdd7a09b91792609cb037aae07 /net | |
parent | 2b6f39e9ee7b6a9fa98f6047b05733053876fdbe (diff) | |
download | linux-24a06f3e3ab640490f49c8e8812d9550d6f15e92.tar.xz |
Input: psmouse - clean up Cypress probe
When Cypress protocol support is disabled cypress_init() is a stub that
always returns -ENOSYS, so there is not point in testing for
CONFIG_MOUSE_PS2_CYPRESS after we decided that we are dealing with a
Cypress device. Also, we should only be calling cypress_detect() when
set_properties argument is "true", like with other protocols.
There is a slight change in behavior to make follow-up patches more
uniform: when we detect Cypress but its initialization fails, instead of
immediately returning PSMOUSE_PS2 protocol we now continue trying
IntelliMouse [Explorer]. Given that Cypress devices only have issue with
Sentelic probes probing Imtellimouse should be safe.
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Tested-by: Marcin Sochacki <msochacki+kernel@gmail.com>
Tested-by: Till <till2.schaefer@uni-dortmund.de>
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Diffstat (limited to 'net')
0 files changed, 0 insertions, 0 deletions