Make sure that the index we pass into long_clear_bit is never higher than the
number of bits in the single long we use as array here.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
The hwdb was a quick and easy way to add entries for new devices but we need
more and more info about the various types of devices. It's better to move
this to a proper database.
Side-effect: this also gives us an easy way to point to the list of devices
supported and makes checking for e.g. missing svgs easier.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
We need to read the directory and the profiles only once.
Nothing tells us how many profiles are supported, so use the specifications
from Logitech website, and add a hwdb entry for those.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
Just logging and error and continuing is not enough, we'll eventually
dereference garbage values.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
profile->dpi_modes[i].led[] is an array of boolean.
Given that the callers (hidpp10_fill_dpi_modes_*()) already are doing the
comparison against 0x02, I guess this is just copy/paste error.
Fixes#152
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
I fixed resolution cycle up actually being resultion cycle down and added the missing cycle actions for resolutions and profiles.
Signed-off-by: Clément Vuchener <clement.vuchener@gmail.com>
In hidpp10_get_hidpp_notifications(), we want all three bytes from
parameters[], not parameters[0] twice. This matches
hidpp10_set_hidpp_notifications().
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Was off by one, the message content only includes one byte of extra information
before the serial number. And add a BE 32-bit helper while we're at it.
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
On the G9 the response to a 0x63 is a short report. The G9 does
not have individual resolutions for X and Y and thus only needs
to report a single value.
The set version of 0x63 should probably also be updated. However,
while my testing of such a change worked better than with long
reports there are still strange side effects of setting resolution,
like turning off all the LED's. LGS also seems to do a full profile
read/write instead of using 0x63 when changing the resolution.
This allows to read but not write profiles from the G9.
The dpi mode sturct is a little different as the mouse only supports
symetrical dpi for x/y. Given the current naming a suffix of _sym
was the best I could come up with.
The G9 profile has some additional unknown bytes compared to the
others. Unfortunately the content of them varies for the 4 first
profiles and the last. So adding a default value for the like with
the G700 is not possible. For this reason I dropped implementing
the write part for now.
Something else to address is that this mouse only has 10 buttons
compared to the usual 13 that we have in PROFILE_NUM_BUTTONS. For
now I just hardcoded it in the profile and hidpp10_get_profile.
Previously, we would put a random address in the macro value, meaning we
lose the content of the macros.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Macros are somewhat properly parsed in hidpp10.
The driver, however, supports only key press/release and delays.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
unknown3 in G700 is the constant string "LGS02". This string is a marker
for LGS to specify that the rest of the data is the names of the profiles
and macros.
SetPoint doesn't set those, so they are considered as metadata.
In libratbag, we will parse those properly and always set them, no matter
if the user used SetPoint or not.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
The mouse accept them, so they are not invalid.
Only the profile directory seems to be checking for the CRC.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
The complexity when writing the profile to ROM at first allows:
- to prevent the user to temper with a non full profile
- in case of a USB failure (disconnect or loss of power for the batteries)
we don't use partially overwrite the profile.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
To not have a giant patch that writes profiles to the HID++1.0 mice,
this first patch only convert the user profile into the actual data
that needs to be sent to the mouse.
We use hidpp_error to output it but no end user will ever see that
because it will be removed in a later commit.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
The G500 uses only one bit in the usage code (thus the holes in the table).
So for the G500, RATBAG_BUTTON_ACTION_SPECIAL_RESOLUTION_CYCLE_UP is
the same as RATBAG_BUTTON_ACTION_SPECIAL_RESOLUTION_DOWN for instance.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Store them in a table in hidpp10_device.
When the dpi list is filled in, we need to invalidate
the profiles.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Preparation patch to add support for more than one profile type.
No functional changes
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
Allows to specify a type of profile and speed up the discovery of
hidpp10 devices when they are not actual gaming with a known profile.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
If the crc is wrong, the device will ignore the page.
This technically multiply by 6 the time to access the device, but this is
because we do not cache anything currently in hidpp10.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
The G700 has 5 available profiles, bump HIDPP10_NUM_PROFILES.
On the other hand, hidpp10_get_device_info() doesn't fill anything...
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@redhat.com>
The command get_profile can either return an index in the directory (in
the page 0x01 of the memory) or an absolute address. If the address is
returned, it is unclear whether the directory is updated.
This commit also includes changes from Peter:
hidpp10: if the profile index exceeds the number of directory index, use 0
hidpp10: factor out directory fetching
hidpp10: use the profile directory only in hidpp10_get_profile
Fixes#26
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Each hidpp10 device should have a udev property named either
RATBAG_HIDPP10_DPI or RATBAG_HIDPP10_DPI_LIST. These properties ensure
that we know the mapping between the raw value of the dpi reported by the
mouse and the actual value.
Older devices (using the scheme 1 or 2 of hidpp10_get_current_resolution)
have to use the DPI_LIST to be able to map 0x80+.
Newer devices (using scheme 3) use the DPI setting with min, max, step.
Signed-off-by: Benjamin Tissoires <benjamin.tissoires@gmail.com>
