GNAT/RTEMS Status with 4.5.0 20091012 r152668

Joel Sherrill joel.sherrill@oarcorp.com
Tue Oct 13 13:48:00 GMT 2009


Hi,

ACATS results for  (4.5.0 20091012 r152668) for
MIPS, SPARC, PowerPC, x86 and m68k/Coldfire.  This is
the first time I have results for m68k/Coldfire. :)

+ ARM - does not build, regression, PR40775

+ General Failures

c380004 fails on all RTEMS targets.  It requires
a very large stack to pass.

c34006g fails on mips, powerpc, sparc, and m68k.
It passes on x86.  This is a failure detected by
the test itself "INCORRECT SELECTION (ASSIGNMENT)".

c953002 fails on x86 and SPARC.  It runs too long and
has to be killed with partial output.

+ MIPS - 3 failures, 1 unique.  c760002.  Run ends
with a lot of these messages from jmr3904 simulator:

HILO: MFHI: MF at 0x8803f6c8 following OP at 0x8800858c corrupted by MT 
at 0x880085ac

+ PowerPC: 2 failures, 0 unique to this target.

+ SPARC: 4 failures, 1 target unique.
  c87b39a - error detected
      raised CONSTRAINT_ERROR : c87b39a.adb:93 discriminant check failed

+ x86 - 19 failures, 17 target unique.
  Qemu seems to have FPU precision issues.  17 FP tests fail
  with precision messages.  They run on real hardware.

+ m68k/coldfire - 24 failures, 21 target unique.
  13 are csx5aXX or cxg20XX tests which are likely the same qemu FPU
  precision issue as found on the x86.

  c85006a - partial output, runs too long
  c97205b, c97305c
     raised CONSTRAINT_ERROR : report.adb:109 range check failed
  c974001, c974013
     Abortable part did not execute.
  cxb3004 cxb3013
     die with stack trace

-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
   Support Available             (256) 722-9985




More information about the Gcc mailing list