[llvm-testresults] buildbot failure in smooshlab on llvm-gcc-i386-darwin9

daniel_dunbar at apple.com daniel_dunbar at apple.com
Sat Feb 27 16:47:00 PST 2010


The Buildbot has detected a new failure of llvm-gcc-i386-darwin9 on smooshlab.
Full details are available at:
 http://smooshlab.apple.com:8010/builders/llvm-gcc-i386-darwin9/builds/3918

Buildbot URL: http://smooshlab.apple.com:8010/

Buildslave for this Build: smoosh-03.apple.com

Build Reason: 
Build Source Stamp: 97355
Blamelist: andersca,djg,lattner

BUILD FAILED: failed compile.llvm-gcc.stage1

sincerely,
 -The Buildbot


================================================================================

CHANGES:
File: lib/CodeGen/MachineLICM.cpp
At: Sat 27 Feb 2010 16:14:03
Changed By: djg
Comments: Don't unconditionally suppress hoisting of instructions with implicit
defs or uses. The regular def and use checking below covers them, and
can be more precise. It's safe to hoist an instruction with a dead
implicit def if the register isn't live into the loop header.
Properties: 




File: lib/CodeGen/CGVtable.cpp
At: Sat 27 Feb 2010 16:14:03
Changed By: andersca
Comments: Add new function.Properties: 




Files:
 lib/Target/X86/X86InstrInfo.cpp
 lib/Target/X86/X86InstrSSE.td
 lib/Target/X86/X86RegisterInfo.cpp
 lib/Target/X86/X86RegisterInfo.h
 lib/Target/X86/X86RegisterInfo.td
At: Sat 27 Feb 2010 16:19:01
Changed By: djg
Comments: Implement XMM subregs.

Extracting the low element of a vector is now done with EXTRACT_SUBREG,
and the zero-extension performed by load movss is now modeled with
SUBREG_TO_REG, and so on.

Register-to-register movss and movsd are no longer considered copies;
they are two-address instructions which insert a scalar into a vector.
Properties: 




Files:
 utils/TableGen/CodeGenDAGPatterns.cpp
 utils/TableGen/CodeGenDAGPatterns.h
 utils/TableGen/DAGISelMatcher.cpp
At: Sat 27 Feb 2010 16:24:02
Changed By: lattner
Comments: Generalize my hack to use SDNodeInfo to find out when a
node is always guaranteed to have a particular type 
instead of hacking in ISD::STORE explicitly.  This allows
us to use implied types for a broad range of nodes, even
target specific ones.
Properties: 




LOGS:
Last 10 lines of 'stdio':
	{standard input}:unknown:Undefined local symbol L_fprintf$stub
	{standard input}:unknown:Undefined local symbol LBB2_9
	{standard input}:unknown:Undefined local symbol L_strlen$stub
	make[4]: *** [libgcc/./_gcov.o] Error 1
	make[4]: *** Waiting for unfinished jobs....
	make[3]: *** [stmp-multilib] Error 2
	rm fsf-funding.pod gcov.pod gfdl.pod cpp.pod gpl.pod gcc.pod
	make[2]: *** [all-stage1-gcc] Error 2
	make[1]: *** [stage1-bubble] Error 2
	make: *** [all] Error 2

Last 10 lines of 'warnings':
	warning: structure `VEC_cp_token_position_heap' used but not defined
	warning: structure `pointer_set_t' used but not defined
	warning: structure `c_arg_info' used but not defined
	warning: structure `c_switch' used but not defined
	warning: structure `et_node' used but not defined
	warning: structure `loop' used but not defined
	warning: structure `ipa_reference_vars_info_d' used but not defined
	warning: structure `reg_info_def' used but not defined
	warning: structure `value_set' used but not defined
	../../llvm-gcc.src/gcc/sched-vis.c:628: warning: no previous prototype for 'print_insn'




More information about the llvm-testresults mailing list