Bug#487396: Missing link error on ARM

2008-06-21 Thread Enrico Zini
Package: g++-4.3 Version: 4.3.1-2 Severity: normal Hello, while trying to investigate why libept does not build on ARM, I came up with the attached source code. On any system, it gives a link time error while building it. On ARM, only when invoking g++, only with -fPIC, it links fine, then runn

Bug#487406: Code not compiled properly on ARM

2008-06-21 Thread Enrico Zini
Package: g++-4.3 Version: 4.3.1-2 Severity: normal Hello, the attached code seems to be the minimal test case to reproduce the current build failure of libept on ARM. Normally, that code should not do anything, but on ARM, it segfaults. Ciao, Enrico -- System Information: Debian Release: le

[Bug rtl-optimization/323] optimized code gives strange floating point results

2008-06-21 Thread pepalogik at seznam dot cz
--- Comment #112 from pepalogik at seznam dot cz 2008-06-21 22:38 --- (In reply to comment #111) > Concerning the standards: The x87 FPU does obey the IEEE754-1985 standard, > which *allows* extended precision, and double precision is *available*. It's true that double *precision* is av

Results for 4.2.4 (Debian 4.2.4-2) libjava testsuite on arm-unknown-linux-gnueabi

2008-06-21 Thread Matthias Klose
LAST_UPDATED: Tue May 20 12:16:07 UTC 2008 (revision 135609) Native configuration is arm-unknown-linux-gnueabi === libjava tests === Running target unix FAIL: Throw_2 execution - source compiled test FAIL: Throw_2 -findirect-dispatch execution - source compiled test FAIL: Throw_

[Bug rtl-optimization/323] optimized code gives strange floating point results

2008-06-21 Thread vincent at vinc17 dot org
--- Comment #113 from vincent at vinc17 dot org 2008-06-22 00:52 --- (In reply to comment #112) > It's true that double *precision* is available on x87. But not the *IEEE-754 > "double precision" type*. It is available when storing a result to memory. > Beside the precision of mantissa