The kbuild test robot pointed out some build errors on ARCH=ia64 of this type:
drivers/staging/comedi/drivers/ni_pcimio.c:1105:42: error: macro "writew" passed 3 arguments, but takes just 2 It appears the read[lwb]/write[lwb] callabacks in struct ni_private are getting treated as macros if the architecture has these symbols defined as macros. For some resone these errors don't appear on ARCH=x86. Regardless, this series removes all the i/o callbacks from struct ni_private. The driver specific i/o handling is now handled in ni_mio_common.c based on if the driver has allocated the 'mite' device. If so the board uses memory mpped i/o (read[lwb]/write[lwb]), otherwise the board uses port mapped i/o (in[lwb]/out[lwb]). H Hartley Sweeten (6): staging: comedi: ni_stc.h: add a 'is_m_series' flag to private data staging: comedi: ni_stc.h: cleanup ni_e_series_enable_second_irq() staging: comedi: ni_mio_common.c: wrap the stc_read/write operations staging: comedi: ni_mio_common.c: wrap the read/write operations staging: comedi: ni_stc.h: remove read/write callbacks from ni_private staging: comedi: ni_stc.h: remove stc read/write callbacks from ni_private drivers/staging/comedi/drivers/ni_atmio.c | 85 +- drivers/staging/comedi/drivers/ni_mio_common.c | 1559 ++++++++++++++---------- drivers/staging/comedi/drivers/ni_mio_cs.c | 83 -- drivers/staging/comedi/drivers/ni_pcimio.c | 397 +----- drivers/staging/comedi/drivers/ni_stc.h | 14 +- 5 files changed, 960 insertions(+), 1178 deletions(-) -- 1.9.3 _______________________________________________ devel mailing list de...@linuxdriverproject.org http://driverdev.linuxdriverproject.org/mailman/listinfo/driverdev-devel
