Re: overhead of bfd_{get,put}*()

[David Miller <davem at davemloft dot net>]

From: Tristan Gingold <gingold@adacore.com>
Date: Tue, 23 Feb 2010 12:34:04 +0100

> On Feb 23, 2010, at 2:38 AM, David Miller wrote:
> 
>> I don't know what to do about this.  It's a more than %10 performance
>> hit on this test case.  But I assume that these routines might have
>> to access potentially unaligned data right?
>
> I do think so.  Maybe we can create unaligned version of them, so
> that everyone will be happy ?

We could also just check the alignment of the pointer in the
existing handlers and execute the existing code if the pointer
is unaligned.

>>  Unfortunately I was not
>> able to hit any such case by running the full testsuite, it passed
>> with my hacks in there :-)
> 
> For which target(s) did you try ?

sparc32 and sparc64