compile errors : qversion and ac_nonexistent.h

[aus9]

hi

EDIT lower post deals with ac_nonexistent.h

I can write basic code but not so good with a config log error.

Leaping ahead, on a previous build my new package does work, but I want to know
A) is an error for qversion something to worry about it or not
B) if concerned, how to fix please

Fragments of my .configure that refer to qversion are:

# Provide some information about the compiler.
$as_echo "$as_me:${as_lineno-$LINENO}: checking for C compiler version" >&5
set X $ac_compile
ac_compiler=$2
for ac_option in --version -v -V -qversion; do
  { { ac_try="$ac_compiler $ac_option >&5"
case "(($ac_try" in
  *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
  *) ac_try_echo=$ac_try;;
esac
eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\""
$as_echo "$ac_try_echo"; } >&5
  (eval "$ac_compiler $ac_option >&5") 2>conftest.err
  ac_status=$?
  if test -s conftest.err; then
    sed '10a\
... rest of stderr output deleted ...
         10q' conftest.err >conftest.er1
    cat conftest.er1 >&5
  fi
  rm -f conftest.er1 conftest.err
  $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
  test $ac_status = 0; }
done

and

# Provide some information about the compiler.
$as_echo "$as_me:${as_lineno-$LINENO}: checking for C compiler version" >&5
set X $ac_compile
ac_compiler=$2
for ac_option in --version -v -V -qversion; do
  { { ac_try="$ac_compiler $ac_option >&5"
case "(($ac_try" in
  *\"* | *\`* | *\\*) ac_try_echo=\$ac_try;;
  *) ac_try_echo=$ac_try;;
esac
eval ac_try_echo="\"\$as_me:${as_lineno-$LINENO}: $ac_try_echo\""
$as_echo "$ac_try_echo"; } >&5
  (eval "$ac_compiler $ac_option >&5") 2>conftest.err
  ac_status=$?
  if test -s conftest.err; then
    sed '10a\
... rest of stderr output deleted ...
         10q' conftest.err >conftest.er1
    cat conftest.er1 >&5
  fi
  rm -f conftest.er1 conftest.err
  $as_echo "$as_me:${as_lineno-$LINENO}: \$? = $ac_status" >&5
  test $ac_status = 0; }
done

fragment of my config.log errors are:

gcc version 4.6.1 (GCC)
configure:3179: $? = 0
configure:3168: gcc -V >&5
gcc: error: unrecognized option '-V'
gcc: fatal error: no input files
compilation terminated.
configure:3179: $? = 1
configure:3168: gcc -qversion >&5
gcc: error: unrecognized option '-qversion'
gcc: fatal error: no input files
compilation terminated.

and

gcc version 4.6.1 (GCC)
configure:4154: $? = 0
configure:4143: gcc -V >&5
gcc: error: unrecognized option '-V'
gcc: fatal error: no input files
compilation terminated.
configure:4154: $? = 1
configure:4143: gcc -qversion >&5
gcc: error: unrecognized option '-qversion'
gcc: fatal error: no input files
compilation terminated.

now this is at early stage, and I have not resolved some other package issues relating to the new ffmpeg so I am not concerned about them at this stage.

attachments if needed

thanks for reading

ahh the forum does not allow me to post a configure file that is over 128 kb

knowing that ompldr is blocked for some users lets try this link to my configure?

http://aus9-test.googlecode.com/files/configure

[Rich]

Hi aus9

for ac_option in --version -v -V -qversion; do

The config script is just trying to determine the version of the installed compiler. Gcc uses  --version , I saw  -qversion
listed on an IBM website for one of their compilers. I don't think it's a problem.

[aus9]

Rich

thanks, I have had this error for 2 packages and I did a search for qversion so am happy this post exists in case other newbies catch it

cheers

[aus9]

ok in case newbies like me find the next error in their config.log

fatal error: ac_nonexistent.h: No such file or directory

google suggests
http://www.linuxquestions.org/questions/linux-software-2/prelink-configuration-error-fatal-error-ac_nonexistent-h-no-such-file-or-direct-939408/

fragment

The configure script checks the result and depending on the result will continue or fail, and also will generate makefiles and other files. conftest.c is a generated test c source file to drive a test case. The link I gave you tried to explain the test case related to this ac_nonexistent.h: how your compiler detects and rejects broken source code that refers to a non-existent header. This is a test case among the others and it must fail. The script configure will catch that error and that's why it will proceed ahead (because this is the expected behavior)

So it looks like I can ignore both errors and move on to infinity and beyond!