[llvm-commits] [PATCH] LNT: enable parsing of cc1-line on ARM-linux platform.

Kristof Beyls kristof.beyls at arm.com
Thu Nov 1 10:49:36 PDT 2012


Hi Daniel,

The attached one-line patch allows me to use LNT to start a run of the test
suite with GCC as a compiler on an ARMv7-linux platform.

My understanding is that LNT tries to figure out what the cc1 binary is by
calling the compiler (e.g. /usr/bin/gcc) as follows:
/usr/bin/gcc -v -E -x c /dev/null -###

On an x86 machine, this seems to print out a number of lines, including the
following, that is parsed by LNT:
 "/usr/lib/gcc/x86_64-linux-gnu/4.4.3/cc1" "-E" "-quiet" "-v" "/dev/null"
"-D_FORTIFY_SOURCE=2" "-mtune=generic" "-fstack-protector"

On the ARMv7-linux machine I tried it on, the above command results in the
following line containing cc1 to be printed:
 /usr/lib/gcc/arm-linux-gnueabihf/4.6/cc1 -E -quiet -v -imultilib .
-imultiarch arm-linux-gnueabihf /dev/null "-march=armv7-a"
"-mfloat-abi=hard" "-mfpu=vfpv3-d16" -mthumb -fstack-protector

The regular expression used in LNT to extract the cc1 path is the following,
which does expect the double quotes to always be present around cc1:
r' "([^"]*)".*"-E".*'

The attached patch adapts the regular expression to correctly parse the
cc1-line whether the double quotes are present or not.

This is my first patch to LNT, so even though it seems trivially simple,
could you review it before I commit?

Thanks,

Kristof
-------------- next part --------------
A non-text attachment was scrubbed...
Name: parse_cc1.patch
Type: application/octet-stream
Size: 578 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20121101/3348e8ae/attachment.obj>


More information about the llvm-commits mailing list