AFAICS, fallout from repacing include of blkdev.h with include of bio.h.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

---

diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c

index 71b986b..7ae18ea 100644

--- a/drivers/mmc/host/mmc_spi.c

+++ b/drivers/mmc/host/mmc_spi.c

@@ -30,6 +30,7 @@

 #include <linux/dma-mapping.h>

 #include <linux/crc7.h>

 #include <linux/crc-itu-t.h>

+#include <linux/scatterlist.h>
 #include <linux/mmc/host.h>

 #include <linux/mmc/mmc.h>		/* for R1_SPI_* bit values */

-

Out of curiousity, which architecture(s) need this?
-

Umm...   IIRC, m32r had been the first build to step into that, at which point

the missing include had been added; the rest had reached that point later, so

I'm not sure which ones would trigger the same crap.  Not hard to test,

though...
Aha.  m68k as well.

-

FWIW this driver originally started out on ColdFire (m68k derived),

a few years back.  :)
Every time I seee a case where architectures don't act the same

with respect to #includes, I count it as an unpleasant surprise.

Not one that's very easy to avoid, unfortunately, but I think it's

been happening less often lately.
Thanks for the fix.
- Dave
-

Yesterday I saw a similar issue with drivers/md/dm-mpath-rdac.c, but it

disappeared during the night, and I didn't see an obvious change that
Simple rule of thumb: he who uses fields of struct xxx should include

the header file that defines struct xxx...
Gr{oetje,eeting}s,
						Geert
--

Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
In personal conversations with technical people, I call myself a hacker. But

when I'm talking to journalists I just say "programmer" or something like that.

							    -- Linus Torvalds

-