[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#684926: marked as done (linux: disable CONFIG_CC_OPTIMIZE_FOR_SIZE)



Your message dated Thu, 25 Oct 2012 16:12:26 -0700
with message-id <20121025231226.GA1758@elie.Belkin>
and subject line Re: linux: disable CONFIG_CC_OPTIMIZE_FOR_SIZE
has caused the Debian Bug report #684926,
regarding linux: disable CONFIG_CC_OPTIMIZE_FOR_SIZE
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
684926: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=684926
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message ---
Source: linux
Severity: wishlist


Hi.

Is there any reasons why you've enabled CONFIG_CC_OPTIMIZE_FOR_SIZE?

>From the help:
> Enabling this option will pass "-Os" instead of "-O2" to gcc resulting in a smaller kernel.

>From gcc:
> Optimize for size.  -Os enables all -O2 optimizations that do not
> typically increase code size.  It also performs further
> optimizations designed to reduce code size.
> -Os disables the following optimization flags: -falign-functions
> -falign-jumps  -falign-loops -falign-labels  -freorder-blocks
> -freorder-blocks-and-partition -fprefetch-loop-arrays
> -ftree-vect-loop-version


So we loose all kinds of optimisations just for storage, which really doesn't
count on all but embedded systems.


Cheers,
Chris.

--- End Message ---
--- Begin Message ---
Hi,

Christoph Anton Mitterer wrote:
> On Tue, 2012-08-14 at 21:57 +0100, Ben Hutchings wrote:

>>> Is there any reasons why you've enabled CONFIG_CC_OPTIMIZE_FOR_SIZE?
[...]
>> Larger code results in more I-cache misses, so it can be slower after
>> all.
>> Let's see your benchmark results.
>
> Well.. I guess making usable benchmarks there is really difficult,..

Not really. :)  All it would take is one important use case that gets
faster to get the conversation started.

Until then, I can't see much reason to pursue this, so closing.

Thanks,
Jonathan

--- End Message ---

Reply to: