auxv: partial revert of r366207, cast buflen to unsigned int where needed

[kevans]

Summary

The warning generated pre-rS366207 is actually a sign comparison warning:

error: comparison of integers of different signs: 'unsigned long' and 'int' [-Werror,-Wsign-compare]
                        if (strlcpy(buf, execpath, buflen) >= buflen)

Revert parts that affected other lines and just cast this to unsigned int.

The buflen < 0 -> EINVAL has been kept despite no longer serving any purposes w.r.t. sign-extension because I do believe it's the right thing to do: "The provided buffer was not the right size for the requested item."

The original warning is confirmed to still be gone with an env WARNS=6 make WITHOUT_TESTS=yes.

[kevans]

Quick wrapping fix... casting to unsigned int pushed it past 80.

[asomers]

Why does Clang treat the final argument of memcpy differently than the final argument of strlcpy?

[kevans]

In D26631#593185, @asomers wrote:

Why does Clang treat the final argument of memcpy differently than the final argument of strlcpy?

In what context? In this case, it's not a problem with the final argument but rather with the comparison of the result of strlcpy and signed int.

[asomers]

Oh, I see. Still, it seems like Clang ought to be complaining about the sign-extension when we pass an int to a function expecting a size_t. Maybe that warning is globally disabled.

[kevans]

Ah, right, sorry- I 100% agree... I checked it with a small trivial test-case, and neither gcc9 nor in-base llvm seem to have any warning about possible unintended sign extension with -Wall -Wextra at least.