Re: A far fetched goal to clean up code quality

[Al Viro]

On Wed, Aug 15, 2018 at 08:33:36PM +0200, Martin Schroeder wrote:
> I propose using following set of GCC flags to enforce code quality in
> the long run.
> 
>     -Wall -Wextra -Werror\
>     -std=gnu11\
>     -pedantic \
>     -Wchar-subscripts\
>     -Wformat\
>     -Wformat-nonliteral\
>     -Wformat-security\
>     -Wmissing-braces\
>     -Wparentheses\
>     -Wsequence-point\
>     -Wswitch\
>     -Wtrigraphs\
>     -Wno-unused-function\
>     -Wunused-label\
>     -Wno-unused-parameter\
>     -Wunused-variable\
>     -Wunused-value\
>     -Wuninitialized\
>     -Wdiv-by-zero\
>     -Wfloat-equal\
>     -Wdouble-promotion\
>     -fsingle-precision-constant\
>     -Wshadow\
>     -Wpointer-arith\
>     -Wwrite-strings\
>     -Wconversion\
>     -Wredundant-decls\
>     -Wunreachable-code\
>     -Winline\
>     -Wenum-compare \
>     -Wlong-long\
>     -Wchar-subscripts

I will probably regret it, but... do you even know C?  And have you ever
bothered to read the documentation describing all those oh-so-fine options,
or have you simply aped the list from some... place?

For example, this:
'-Wlong-long'
     Warn if 'long long' type is used.  This is enabled by either
     '-Wpedantic' or '-Wtraditional' in ISO C90 and C++98 modes.  To
     inhibit the warning messages, use '-Wno-long-long'.
and the warning it triggers is <drumrolls>
	"ISO C90 does not support âlong longâ [-Wlong-long]".

Now, could the esteemed sir be arsed to explain the meaning of "-std=gnu11"?
Take your time and use your own words.  Then explain just how would that
warning be relevant to chosen C dialect.  For an extra credit, explain
just how the hell is one supposed to represent 64bit arithmetical data on
32bit architectures without using the dreaded long long.

Free advice: do try to understand what you are proposing when you are
setting lofty goals for everybody else to toil for.

Al "I loathe aspiring PHBs" Viro, in rather foul mood at the moment...