Bugzilla Bug 157790

if x86_64-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../..  -I../..  -fPIC
-Wall -mtune=k8 -O2 -pipe -Wno-unused-function -MT joy.o -MD -MP -MF
".deps/joy.Tpo" -c -o joy.o joy.c; \
        then mv -f ".deps/joy.Tpo" ".deps/joy.Po"; else rm -f ".deps/joy.Tpo";
exit 1; fi
if x86_64-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../..  -I../..  -fPIC
-Wall -mtune=k8 -O2 -pipe -Wno-unused-function -MT lb216.o -MD -MP -MF
".deps/lb216.Tpo" -c -o lb216.o lb216.c; \
        then mv -f ".deps/lb216.Tpo" ".deps/lb216.Po"; else rm -f
".deps/lb216.Tpo"; exit 1; fi
if x86_64-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../..  -I../..  -fPIC
-Wall -mtune=k8 -O2 -pipe -Wno-unused-function -MT lcdm001.o -MD -MP -MF
".deps/lcdm001.Tpo" -c -o lcdm001.o lcdm001.c; \
        then mv -f ".deps/lcdm001.Tpo" ".deps/lcdm001.Po"; else rm -f
".deps/lcdm001.Tpo"; exit 1; fi
if x86_64-pc-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../..  -I../..  -fPIC
-Wall -mtune=k8 -O2 -pipe -Wno-unused-function -MT lcterm.o -MD -MP -MF
".deps/lcterm.Tpo" -c -o lcterm.o lcterm.c; \
        then mv -f ".deps/lcterm.Tpo" ".deps/lcterm.Po"; else rm -f
".deps/lcterm.Tpo"; exit 1; fi
In file included from /usr/include/linux/i2c-dev.h:28,
                 from hd44780-i2c.c:62:
/usr/include/linux/i2c.h:66: error: array type has incomplete element type
make[3]: *** [hd44780-i2c.o] Error 1
make[3]: *** Waiting for unfinished jobs....
make[3]: Leaving directory
`/var/tmp/portage/lcdproc-0.5.1/work/lcdproc-0.5.1/server/drivers'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory
`/var/tmp/portage/lcdproc-0.5.1/work/lcdproc-0.5.1/server'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/var/tmp/portage/lcdproc-0.5.1/work/lcdproc-0.5.1'
make: *** [all] Error 2

!!! ERROR: app-misc/lcdproc-0.5.1 failed.

Not only amd64; see Bug 148072 Comment #8 and following.

This bug is caused by <linux-headers-2.6.17 and gcc-4.
It's a dupe of bug #146407, which I reopened therefore.

Might be useful to share that you can set LCDPROC_DRIVERS in make.conf and only
build the driver you require.  In my case that avoided the compile problems. 

Isn't that stated in the info message when you emerge lcdproc (that you can
choose which drivers to compile)?

Also, which drivers did you disable to avoid this problem?

Anyway, it's not an lcdproc bug, but it's kernel<->gcc triggered by lcdproc.

*** This bug has been marked as a duplicate of 146407 ***