Results for 2.97 20001120 (experimental) testsuite on sparc-sun-solaris2.8 with -mcpu=ultrasparc -m64

[Brad Lucier]

This message does not contain the actual test results, because they
were just bounced by the gcc.gnu.org mail handler for being over
200,000 bytes in size!

But I do have a question:  The results of gcc testing on this target
are not so great with 32-bit code (89 failures in gcc), see, e.g.,

http://gcc.gnu.org/ml/gcc-testresults/2000-11/msg00185.html

but they are absolutely abysmal for 64-bit code (with -mcpu=ultrasparc -m64).

Is Jakub's work on the subreg-byte branch an attempt to fix this?

Brad Lucier