Re: CVS commit: src/sys/lib/libsa

[Valery Ushakov <uwe%stderr.spb.ru@localhost>]

On Tue, Apr 18, 2017 at 11:28:21 +0900, Kimihiro Nonaka wrote:

> On Tue, Apr 18, 2017 at 3:26 AM, Valery Ushakov <uwe%stderr.spb.ru@localhost> wrote:
> 
> > What is really intended here?  Do we need shstr set so that we can
> > test for ".SUNW_ctf" section name later?
> 
> That's right.

Ok, after re-reading that bit a few more times I convinced myself it's
actually ~correct.  Sorry for the noise, that code is not exactly well
organized to begin with.

Thanks!


PS: While here, I'd like to note for the record that instead of
reading .shstrtab twice (which is expensive with compressed kernels),
we should read it to the local buffer and BCOPY() it to maxp if
LOAD_SYM.  We should also not treat failure to read .shstrtab as
fatal.  But I'm not shaving that yak now.

-uwe