Share via


LKG H&L going great

Wow what a change a few days can make. I'm doing super well on my
LKG H&L project. I finally got some help from another developer on
the team. He really helped by running builds on various architectures.
We were able to get ia64 completed first because of it. And we found a
simple mistake in one of our changes. I had commented out only the x86
version of a function. That would have cost us a full day because of
turnaround times on Gibraltar test runs. Now I'm fairly certain that
all chip types will build the same.

Tonight's Gibraltar build and test run should complete the full
process. I don't know what the quality will be, but I'm optimistic that
it will build and run tests. I know our hacky fix for PSAPI.dll's
EnumProcesses will cause 2 tests to fail. But that's ok because the bug
will not prevent us from getting into RI #16. If we don't get into #16
then our payload for RI #17 will be in jeopardy. I know it sounds
strange, but after RI #17 there is a "new feature" freeze; after 17 we
can only do bug fixes.