<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Tue, Jan 6, 2015 at 8:06 PM, Sahasrabuddhe, Sameer <span dir="ltr"><<a href="mailto:sameer.sahasrabuddhe@amd.com" target="_blank">sameer.sahasrabuddhe@amd.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div bgcolor="#FFFFFF" text="#000000"><span class="">
<br>
<div>On 1/7/2015 8:59 AM, Chandler Carruth
wrote:<br>
</div>
<blockquote type="cite">
<div dir="ltr">
<div class="gmail_extra"><br>
<div class="gmail_quote">Essentially, I think
target-independent optimizations are still attractive, but
we might want to just force them to go through an actual
target-implemented API to interpret the scopes rather than
making the interpretation work from first principles. I just
worry that the targets are going to be too different and we
may fail to accurately predict future targets' needs.</div>
</div>
</div>
</blockquote>
<br></span>
If we have a target-implemented API, then just opaque numbers should
also be sufficient, right? For the API, all we care about is queries
that interesting optimizations will want answered from the target.
This could be at the instruction level: "is it okay to remove this
atomic store with scope n1 that is immediately followed by atomic
store with scope n2?". Or it could be at the scope level: "does
scope n2 include scope n1"?</div></blockquote><div><br></div><div>I think it is significantly more friendly (and easier to debug mistakes) if the textual IR uses human readable names. We already have a hard time due to the totally opaque nature of address spaces -- there are magical address spaces for segment stuff in x86.</div><div><br></div><div>The strings are only opaque to the target-independent optimizer. While integers and strings are equally friendly to the code in the target, strings are significantly more friendly to humans reading the IR.</div><div><br></div><div><br></div><div>The other advantage is that it makes it much harder to accidentally write code that relies on the particular values for the integers. =]</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div bgcolor="#FFFFFF" text="#000000"><span class=""><br>
<br>
<blockquote type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<div>I think the "strings" can be made relatively clean.</div>
<div><br>
</div>
<div>What I'm imagining is something very much like the
target-specific attributes which are just strings and left
to the target to interpret, but are cleanly factored so
that the strings are wrapped up in a nice opaque attribute
that is used as the sigil everywhere in the IR. We could
do this with metadata, and technically this fits the model
of metadata if we make the interpretation of the absence
of metadata be "system". However, I'm quite hesitant to
rely on metadata here as it hasn't always ended up working
so well for us. ;]</div>
</div>
</div>
</div>
</blockquote>
<br></span>
Metadata was the first thing to be considered internally at AMD. But
it was quickly shot down because the Research guys were unwilling to
accept the possibility of scope being lost and replaced by a default
"system" scope. Current models are useful only when all atomic
accesses for a given location use the same scope throughout the
application, i.e., all threads running on all agents. So it is not
okay for the compiler to "promote" the scope in just one kernel
unless it has access to the entire application; the result is
undefined. This is true for OpenCL source as well as HSAIL target.
This may change in the near furture:<br>
<br>
HRF-Relaxed: Adapting HRF to the complexities of industrial
heterogeneous memory models<br>
<a href="http://benedictgaster.org/?page_id=278" target="_blank">http://benedictgaster.org/?page_id=278</a><br>
<br>
But even then, it will be difficult to say if the same models can be
applied to heterogeneous systems that don't resemble OpenCL or
HSAIL.</div></blockquote><div><br></div><div>Yea, I'm not really surprised by this.</div><div><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div bgcolor="#FFFFFF" text="#000000"><span class=""><br>
<br>
<blockquote type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<div>I'd be interested in your thoughts and others' thoughts
on how me might encode an opaque string-based scope
effectively. If we can find a reasonably clean way of
doing it, it seems like the best approach at this point:</div>
<div><br>
</div>
<div>- It ensures we have no bitcode stability problems.</div>
<div>- It makes it easy to define a small number of
IR-specified values like
system/crossthread/allthreads/whatever and singlethread,
and doing so isn't ever awkward due to any kind of
baked-in ordering.</div>
<div>- In practice in the real world, every target is
probably going to just take this and map it to an enum
that clearly spells out the rank for their target, so I
suspect it won't actually increase the complexity of
things much.</div>
</div>
</div>
</div>
</blockquote>
<br></span>
I seem to be missing something here about the need for strings. If
they are opaque anyway, and they are represented by sigils, then the
sigils themselves are all that matter, right? Then the encoding is
just a number...</div></blockquote><div><br></div><div>See above for why I'd prefer not to use a raw number in the IR.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div bgcolor="#FFFFFF" text="#000000"><span class=""><br>
<br>
<blockquote type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<div> </div>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> But while the topic is wide open,
here's another possibly whacky approach: we let the scopes
be integers, and add a "scope layout" string similar to
data-layout. The string encodes the ordering of the
integers. If it is empty, then simple numerical
comparisons are sufficient. Else the string spells out the
exact ordering to be used. Any known current target will
be happy with the first option. If some target inserts an
intermediate scope in the future, then that version
switches from empty to a fully specified string. The best
part is that we don't even need to do this right now, and
only come up with a "scope layout" spec when we really hit
the problem for some future target.</blockquote>
</div>
<br>
This isn't a bad approach, but it seems even more complex. I
think I'd rather go with the fairly boring one where the IR
just encodes enough data for the target to answer queries
about the relationship between scopes.</div>
</div>
</blockquote>
<br></span>
I am not really championing scope layout strings over a
target-implemented API, but it seems less work to me rather than
more. The relationship between scopes is just an SWO, and it can be
represented as a graph. A practical target will have a very small
number of scopes, say not more than 16. It should be possible to
encode this into a graphviz-style string. Then instead of having
every target implement an API, they just have to specify the
relationship as a string. <br></div></blockquote><div><br></div><div>I see where you're going here, and it sounds feasible, but it honestly seems much *more* work and certainly more complex for the IR. We can always add such a representation to communicate the relationships if it becomes important, but I'd rather communicate via a boring target API to start with I think.</div></div></div></div>