No subject

passing simple values by value, and structures by reference.  In
particular, I'd guess that it considers a one-element struct a simple
value.  Does that help?

> As of now I wish to play with the code and would be making a git repo @
> gitorious [ I think its quiet late] . So can you please suggest me a way to
> test and play with code as you are an expert on the matter.
> [ Also my prime moto here is to understand the code genration and not the
> underlying facts in the code so is that idea wrong? ]

"git clone" the existing XCB repos, and hack on the bits you want.  If
you want us to see some intermediate results, you can post a patch or
push the modified repo(s) somewhere we can see.  Note, though, as Peter
Harris said, that you should start from full clones of the existing
repositories.

And yes, you absolutely want to focus on the code generation and not the
underlying code it generates.  You don't need to understand all the
details of the X protocol or specific protocol requests; you just need
to work out what the code generator generates for specific XML, where it
goes wrong (generating the wrong code), and how to fix it.

> I also wanted to talk something regarding to Bug
> 34037<https://bugs.freedesktop.org/show_bug.cgi?id=34037>I tend to see
> it everyday but I am unable to escalate. I am not able to
> understand the bug much even with the provided description. :( Where should
> I look for the bug in the code?
> 
> If you remember, once you said "I think you will find that the bug you
> linked to (fixing XCB's padding) represents a summer's worth of work by
> itself; plan on spending the bulk of your time there, and don't worry about
> anything else.  If you get past that and still want more to do, we can
> easily come up with something. :)"
> So,
> [ Am I going very slow? Please speak to me bluntly as this is very
> important, with context to deadlines, work and assessments as I really want
> to know and would pace up more to give my best, By now almost 2 weeks are
> coming to an end and I feel very unsatisfied with my work, though still I am
> not sure, if I am judging my self very brutally or sympathetically]
>
> Also, I am unable to set my deadlines, please guide me on that
>     Should I first concentrate on compilation and code generation? or what?

You have a large and complex code generator to work with, and the right
way to start figuring out the details you need to understand involves
doing experiments to see how it works.  Figuring out how to get started,
running the experiments, and understanding where to go from there will
take time.  Don't rush through it feeling like you've gotten behind
schedule.  As long as you keep working steadily on getting the general
idea of the code generation, understanding the effect of the bug you
want to fix (what code it generates versus what code it *should*
generate), and figuring out how to implement the padding rules I
mentioned in a previous email to you, you shouldn't have any trouble
getting done within the SoC timeframe.

- Josh Triplett