ValueMap::entry_for incorrectly used negation to deal with negative hashcodes which doesn't work for MININT. The is to use unsigned for the computation.

I did a test PRT submission last night, and got a failure in C1. I
didn't change any C1 files so I'm pretty confident that this was not
caused by any changes I made.

The failure is:

solaris_i586_jdk1.6_fastdebug_c1_runThese_Xcomp_2 reliability log details
http://analemma.sfbay.sun.com/net/prt-web.sfbay/prt-workspaces/20051121224815.steved.escape_analysis_ws/solaris_i486_compiler1/fastdebug/runThese_Xcomp_2-reliability-test-fastdebug.log

[Exit:javasoft.sqe.tests.lang.fpl038.fpl03806mDDAS1.fpl03806mDDAS1]
......................................
[Enter:javasoft.sqe.tests.lang.fpl038.fpl03806mDDAS10.fpl03806mDDAS10]
# To suppress the following error report, specify this argument
# after -XX: or in .hotspotrc: SuppressErrorAt=/c1_ValueMap.hpp:33]
#
# An unexpected error has been detected by Java Runtime Environment:
#
# Internal Error (/net/prt-solx86-q1-2/PrtBuildDir/workspace/src/share/vm/c1/c1_ValueMap.hpp, 33), pid=7570, tid=8
#
# Java VM: Java HotSpot(TM) Client VM (20051121224815.steved.escape_analysis_ws-debug compiled mode)
#
# Error: assert(0 <= i && i < length(),"index out of bounds")
# An error report file with more information is saved as hs_err_pid7570.log
#
# If you would like to submit a bug report, please visit:
# http://java.sun.com/webapps/bugreport/crash.jsp
#
Current thread is 8
Dumping core ...