[PATCH] add unused flag check arguments to toolchain flag detection

[PATCH] add unused flag check arguments to toolchain flag detection

[Shiz]

[-- Attachment #1: Type: text/plain, Size: 446 bytes --]


this checks whether -Werror=unknown-warning-option and
-Werror=unused-command-line-argument are supported by the toolchain. if so,
it adds those flags to the tryflag and tryldflag checks in ./configure.

so far only clang supports these flags, but clang is also the only toolchain
I've encountered that ignores unrecognized arguments instead of erroring out.
---
 configure | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-add-unused-flag-check-arguments-to-toolchain-flag-de.patch --]
[-- Type: text/x-patch; name="0001-add-unused-flag-check-arguments-to-toolchain-flag-de.patch", Size: 1432 bytes --]

diff --git a/configure b/configure
index 143dc92..08acef0 100755
--- a/configure
+++ b/configure
@@ -80,7 +80,7 @@ fi
 tryflag () {
 printf "checking whether compiler accepts %s... " "$2"
 echo "typedef int x;" > "$tmpc"
-if $CC $2 -c -o /dev/null "$tmpc" >/dev/null 2>&1 ; then
+if $CC $CFLAGS_TRY $2 -c -o /dev/null "$tmpc" >/dev/null 2>&1 ; then
 printf "yes\n"
 eval "$1=\"\${$1} \$2\""
 eval "$1=\${$1# }"
@@ -94,7 +94,7 @@ fi
 tryldflag () {
 printf "checking whether linker accepts %s... " "$2"
 echo "typedef int x;" > "$tmpc"
-if $CC -nostdlib -shared "$2" -o /dev/null "$tmpc" >/dev/null 2>&1 ; then
+if $CC $LDFLAGS_TRY -nostdlib -shared "$2" -o /dev/null "$tmpc" >/dev/null 2>&1 ; then
 printf "yes\n"
 eval "$1=\"\${$1} \$2\""
 eval "$1=\${$1# }"
@@ -113,7 +113,9 @@ CFLAGS_C99FSE=
 CFLAGS_AUTO=
 CFLAGS_MEMOPS=
 CFLAGS_NOSSP=
+CFLAGS_TRY=
 LDFLAGS_AUTO=
+LDFLAGS_TRY=
 OPTIMIZE_GLOBS=
 prefix=/usr/local/musl
 exec_prefix='$(prefix)'
@@ -205,6 +207,14 @@ exit 1
 fi
 
 #
+# Figure out options to force errors on unknown flags.
+#
+tryflag   CFLAGS_TRY  -Werror=unknown-warning-option
+tryflag   CFLAGS_TRY  -Werror=unused-command-line-argument
+tryldflag LDFLAGS_TRY -Wl,-Werror=unknown-warning-option
+tryldflag LDFLAGS_TRY -Wl,-Werror=unused-command-line-argument
+
+#
 # Need to know if the compiler is gcc to decide whether to build the
 # musl-gcc wrapper, and for critical bug detection in some gcc versions.
 #

Re: [PATCH] add unused flag check arguments to toolchain flag detection

[Rich Felker]

On Thu, May 28, 2015 at 04:50:10AM +0200, Shiz wrote:
> 
> this checks whether -Werror=unknown-warning-option and
> -Werror=unused-command-line-argument are supported by the toolchain. if so,
> it adds those flags to the tryflag and tryldflag checks in ./configure.
> 
> so far only clang supports these flags, but clang is also the only toolchain
> I've encountered that ignores unrecognized arguments instead of erroring out.
> ---
>  configure | 14 ++++++++++++--
>  1 file changed, 12 insertions(+), 2 deletions(-)

Thanks! The patch looks mostly good (except the -Wl, part that's
wrong). I have some proposed changes to the message. The shortlog
message should be clear on the what and where of the change:

configure: work around compilers that only warn for unknown options

Then, if not already clear, the body should express more about why and
how, such as:

some compilers (such as clang) accept unknown options without error,
but then print warnings on each invocation, cluttering the build
output and burying meaningful warnings. this patch makes configure's
tryflag and tryldflag functions use additional options to turn the
unknown-option warnings into errors, if available, but only at check
time. these options are not output in config.mak to avoid the risk of
spurious build breakage; if they work, they will have already done
their job at configure time.

FYI cparser/firm also supports -Werror=unknown-warning-option so it's
not just clang.

Rich

Re: [PATCH] add unused flag check arguments to toolchain flag detection

[Mark Cilissen]

> On 28 May 2015, at 05:32, Rich Felker <dalias@libc.org> wrote:
> 
> Thanks! The patch looks mostly good (except the -Wl, part that's
> wrong). I have some proposed changes to the message. The shortlog
> message should be clear on the what and where of the change:

Ah yes, the -Wl, part was a local experiment I forgot to revert locally.
Thanks for the commit message feedback, a v2 is on its way.

-S