Re: Patch: Update jacks.xfail for the tree-ssa merge

[Ranjit Mathew <rmathew at gmail dot com>]

> Ranjit> As noted elsewhere, we now have 30 XPASSes and 5 FAILs
> Ranjit> in Jacks after the tree-ssa merge.
> Ranjit> Looks like another update to jacks.xfail is in order.
> Ranjit> Is the following OK?
> 
> Do we know why we now have more FAILs?

Yes.

All of the failures:

  16.1.7-simple-definite-unassignment-pass-1
  16.2.10-definite-unassign-pass-3
  16.2.11-definite-unassign-pass-3
  16.2.9-definite-unassign-pass-3
  non-jls-jsr41.2-nesting-1

arise because we now no longer optimise away an if-else statement
if the condition is constant. This change was done as a part of the
original gimplification of Java (see patch_if_else_statement):

  http://gcc.gnu.org/ml/java-patches/2003-q2/msg00409.html

Reverting the change to patch_if_else_statement() fixes all
of these failures and produces no new failures.

Unfortunately, it also reduces the number of XPASSes from 30
to just 2:

  15.21-equal-3
  9.2-implicit-4

So we can't simply throw away the dead portion of an if-else
statement with a constant condition.

That is why I proposed that we just update the jacks.xfail
file - unless of course someone wants to implement tracking 
definite [un]assignment of a variable under various conditions,
including the ones shown by the failing tests above.

Sorry, I should have explained all this in my earlier mail.

Ranjit.