summaryrefslogtreecommitdiffstats
path: root/drivers/hid/usbhid/hid-quirks.c
diff options
context:
space:
mode:
authorJiri Kosina <jkosina@suse.cz>2010-02-03 16:11:12 +0100
committerJiri Kosina <jkosina@suse.cz>2010-02-03 16:23:45 +0100
commit2dbf209d7a7ab94266b936bd2da6a4026c279992 (patch)
tree0f98440abb782b96787816d94b194a1a0d4df1e2 /drivers/hid/usbhid/hid-quirks.c
parentf54405db66fbec11679241daefd16fd8291a5762 (diff)
downloadlinux-2dbf209d7a7ab94266b936bd2da6a4026c279992.tar.bz2
HID: make full-fledged hid-bus drivers properly selectable
For historical reasons, we don't have most of the in-tree drivers residing on hid-bus properly selectable in kernel configuration unless CONFIG_EMBEDDED is set. This has been introduced on Linus' request from 14 Oct === As to the Kconfig options - do they really add so much space that you need to ask for the quirks? You didn't use to. Can you make the questions depend on EMBEDDED, or at least on the HID_COMPAT thing or whatever? === This still makes perfect sense for small and tiny drivers, which just fix report descriptors, fix up HID->input mappings that slightly violates HUT standard, send one extra packet to the device that is needed before it becomes functional, etc. Since then, we have been gathering more and more HID-bus drivers, which are full-fledged drivers. For these, the size argument becomes more valid. Plus the devices are much more special than "just violates HID specification in this one or two tiny unimportant points". Therefore I am marking such drivers as properly selectable no matter the setting of CONFIG_EMBEDDED, while keeping all the small and tiny ones compiled by default. Signed-off-by: Jiri Kosina <jkosina@suse.cz>
Diffstat (limited to 'drivers/hid/usbhid/hid-quirks.c')
0 files changed, 0 insertions, 0 deletions