Re: [Gdbheads] Re: Feb's patch resolution rate

[Andrew Cagney]

> > Hmm, I should have also counted who reviewed Daniels patches as I now
> > > suspect the results would be similar.
> > >
> > > If anything it should be Elena and you [david] maintaining the symbol table.

Jim, lets stick to this question.  Would the pair:

        Elena
        David

make for a better symbol table leads than:

        JimB
        Elena

The symbol table is a critical piece of infrastructure and as such needs 
a steady hand, and a long term commitment by developers.  Both through 
overhaul, and through a disciplined and diligent review of patches. 
What the area does not need is random tugs in differing directions based 
on the wim of "maintainer of the day".

The critical importance of this can be seen with the architecture 
vector.  In the good old days where being a Cygnus employee meant 
blanket write, people would continually and randomly add architecture 
dependent macros (back then there wasn't an architecture object).  This 
attitude spilled through to the frame code, the register code, and many 
other areas where the result was a convoluted mess of interfaces.

Now, through discipline and review, we see a very different situation - 
 the number of [non-deprecated] interfaces is dropping (5.3 peaked at 
139, vs 104 in current).  Since I'm responsible for this area I'm 
expected to not just review an occasional and selective patch, rather I 
need to ensure that all patches are viewed in a timely manner and in 
addition, where necessary, respond to problems addressing them as we as 
a group see fit (cf my most recent addition of gdbarch_register_pre_init).

Adding more reviewers to the mix, be it by spreading responsibility more 
thinly, or by giving core-developers carte blanche, is not going to 
service this need.

> So I didn't see how it was in GDB's
> interests to remove myself from the list of people able to approve
> symtab patches.

GDB needs people that _do_ review symtab patches, not people that are 
listed as being "able to approve" symtab patches.  There is a very real 
difference.

Denying this is not being honest with either themselves or the GDB 
community, they are treating their position as one of power and glory, 
and not responsibility.

Andrew