In 2.6.0-test5, the README still says "Make sure you have gcc 2.95.3
available." It is telling the truth.
SuSE 8.2 includes gcc 3.3 20030226 (prerelease). Up to the point of
aborting, it diagnoses a ton of "warning: comparison between signed and
unsigned", but nonetheless compiles many source files and continues. But in
aic7(censored)_core.c, line 76, it diagnoses "warning: `num_chip_names'
defined but not used". Even though this is also just a warning, obviously
it threatens disastrous consequences :-s (sarcasm). make[3] quits with
Error 1 and make[2] and make[1] quit with Error 2.
(My previous tests of 2.6.0-test[1-5] used the version of gcc that was in
SuSE 8.1.)
Norman Diamond writes:
> In 2.6.0-test5, the README still says "Make sure you have gcc 2.95.3
> available." It is telling the truth.
>
> SuSE 8.2 includes gcc 3.3 20030226 (prerelease). Up to the point of
> aborting, it diagnoses a ton of "warning: comparison between signed and
> unsigned", but nonetheless compiles many source files and continues. But in
> aic7(censored)_core.c, line 76, it diagnoses "warning: `num_chip_names'
> defined but not used". Even though this is also just a warning, obviously
> it threatens disastrous consequences :-s (sarcasm). make[3] quits with
> Error 1 and make[2] and make[1] quit with Error 2.
Look for and delete any -Werror in aic7xxx' Makefile.
I've seen people being bit by this before.