4.0.1 build on 1.4T

[der Mouse <mouse%Rodents-Montreal.ORG@localhost>]

Just recently, I tried doing a build (build.sh -U build style) of the
4.0.1 source tree on one of my 1.4T SPARCs.

It failed (with a somewhat surprising message).  Is this something that
would be considered a problem worth fixing, or is 1.4T sufficiently old
and nonconforming to whatever specs the build machinery cares about
that we don't care about build failures when that's the underlying OS?

In case it matters, here's a (severely cut-down) summary of the failure:

dependall ===> tools/compat
[...]
checking for unistd.h... (cached) yes
checking for size_t... no
checking for id_t... no
checking for long long... no
checking for u_long... no
checking for u_char... no
checking for u_short... no
checking for u_int... no
checking for u_quad_t... no
checking for socklen_t... no
checking for uint8_t... no
checking for u_int8_t... no
configure: error: cannot find a suitable type for uint8_t

*** Failed target:  config.cache
*** Failed command: CC=cc CFLAGS=-O LDFLAGS= /bin/sh 
/home/mouse/4.0.1/src/tools/compat/configure --cache-file=config.cache
*** Error code 1
[...failures cascading up the make recursion chain...]
nbmake: stopped in /home/mouse/4.0.1/src

ERROR: Failed to make build
*** BUILD ABORTED ***

I find these failures odd, because I know perfectly well that the
compilation environment does indeed support size_t and long long, and
think (but would check before reporting as a bug) that it supports most
of the other types claimed nonexistent in the quote above.

/~\ The ASCII                             Mouse
\ / Ribbon Campaign
 X  Against HTML                mouse%rodents-montreal.org@localhost
/ \ Email!           7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B