This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: PR 23551: why should we coalesce inlined variables?
On May 10, 2007, "Richard Guenther" <richard.guenther@gmail.com> wrote:
> On 5/10/07, Alexandre Oliva <aoliva@redhat.com> wrote:
>> We're discarding useful debug info by SSA-coalescing inlined
>> variables. The reason we refrain from coalescing non-inlined
>> variables is for better debug info. No other pass pays attention to
>> DECL_FROM_INLINE, and it doesn't look like we generate significantly
>> worse code if we refrain from coalescing these variables. So...
>>
>> May I check this in if it passes bootstrap and regression testing on
>> x86_64-linux-gnu?
>>
>> :ADDPATCH debug:
> I would expect this to cause a noticable memory usage increase on testcases
> like tramp3d. Can you check that, too? (c.f.
> www.suse.de/~rguenther/tramp3d/tramp3d-v4.cpp.gz)
> So I guess while it may look reasonable for C, it isn't really for C++.
Sorry that this took so long.
I'm not sure whether you meant compile-time or run-time memory usage,
so I measured both.
I haven't detected any significant changes to stack sizes, only one
change from 32 to 48 bytes and one from 64 to 80. The ones with big
stack use remained absolutely identical.
As for build time, I've observed a 300k- (0.05%) increase in max mem
use before GC on x86_64-linux-gnu:
{GC 649910k -> 641578k}
{GC 650182k -> 641804k}
Not exactly earth-shattering, is it?
--
Alexandre Oliva http://www.lsd.ic.unicamp.br/~oliva/
FSF Latin America Board Member http://www.fsfla.org/
Red Hat Compiler Engineer aoliva@{redhat.com, gcc.gnu.org}
Free Software Evangelist oliva@{lsd.ic.unicamp.br, gnu.org}