Re: Patch for PR c/12373

[Ian Lance Taylor <ian at wasabisystems dot com>]

Mark Mitchell <mark@codesourcery.com> writes:

> >This PR is currently targeted at 3.4.0.  Mark, let me know if you want
> >this patch on the 3.4 branch.  The patch should never introduce a new
> >crash.  But I suppose it is conceivable that it could introduce a case
> >where the compiler currently crashes but, after this patch, it
> >generates bad code.  I think that is unlikely in this case.
> >
> OK for 3.4.0.  Thanks!

Patch checked in, PR closed.

I just want to comment that this PR was opened last September, six
months ago.  It's severity was set to critical, which was quite
reasonable since this is an ICE on valid code.  I started looking at
last night, and I found the problem and wrote a patch in about an
hour.  I haven't done much work in the frontend.  I'm sure any other
gcc hacker could have found and fixed this problem in the same amount
of time.

So we had a span of six months during which no gcc hacker took an hour
to look at a critical PR, a PR with a clean test case, a PR which was
a regression from the last release and which therefore was a potential
problem for a lot of people after the release was made.

I don't think that's very good.  Wasabi Systems is hardly the largest
company working on gcc.  Obviously we can't control volunteers, but
perhaps some other company could assign somebody to spend some time
working on PRs.

(Partially mitigating the above is the fact that the PR was assigned
to Zack, and was reported against the sh-elf target which is not one
of the most popular.  As it happens, though, within the PR was a
report that it also affected Alpha, and thus potentially other
targets.  And the PR was not in code which Zack wrote.  So I think it
is still quite reasonable to expect that somebody would have taken a
look at the PR over a six month period.)

Ian