This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [tree-ssa] Improve removal of statements when removing blocks


In message <wvlvfu23pgj.fsf@prospero.boston.redhat.com>, Jason Merrill writes:
 >On Wed, 16 Jul 2003 19:11:55 -0600, law@redhat.com wrote:
 >
 >> Unfortunately, the code which linearizes a COND_EXPR which is always
 >> true requires an ELSE clause which is empty -- a COMPOUND_EXPR with
 >> both operands being empty statements does not qualify as empty, nor does
 >> a chain of COMPOUND_EXPRs where all the leafs are empty statements.
 >
 >This is why we should be checking TREE_SIDE_EFFECTS rather than
 >IS_EMPTY_STMT in most cases.
Agreed.   However, I'm pretty sure that wouldn't help in this case --
unless we propagate up changes in TREE_SIDE_EFFECTS as we delete
unreachable statments with side effects.

jeff




Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]