new pragma name ideas? (was: defunctionalization)
On Thu, Jul 18, 2013 at 5:10 PM, Carter Schonwald < carter.schonwald@gmail.com> wrote:
So the idea here to make it possible to have a function that can be specialized at certain types, and explicitly inlined at specific use sites, but ghc otherwise will not inline it? Cool!
one thought: might it be simpler to instead have something like EXPLICIT-INLINABLE, rather that requiring the juxtaposition of two pragmas which "seem" contradictory?
I like that idea. How about SPECIALISABLE? This is a nod to the possibility that GHC might someday automatically specialize an imported function. Or EXPOSE? Or EXTERNAL? Bikeshed activate!
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 18/07/13 23:18, Nicolas Frisby wrote:
On Thu, Jul 18, 2013 at 5:10 PM, Carter Schonwald < carter.schonwald@gmail.com> wrote:
So the idea here to make it possible to have a function that can be specialized at certain types, and explicitly inlined at specific use sites, but ghc otherwise will not inline it? Cool!
one thought: might it be simpler to instead have something like EXPLICIT-INLINABLE, rather that requiring the juxtaposition of two pragmas which "seem" contradictory?
I like that idea. How about SPECIALISABLE? This is a nod to the possibility that GHC might someday automatically specialize an imported function.
Or EXPOSE? Or EXTERNAL? Bikeshed activate!
_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs
Maybe we need a BIKESHED pragma that will let us alias any extensions within the project to anything we like. - -- Mateusz K. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.20 (GNU/Linux) iQIcBAEBAgAGBQJR6G1uAAoJEM1mucMq2pqXrbYP/RXWS0yeI6Ewy6hx35rmqxPG TfCK5U9KjW9WkvWXZqJq407Z0Qluiif0UIfMFy8oujj3RnjJUKN+1Co/zVdYxyiD 0LGdS4SxDQO4aRTd1uRAjaMGzVUA4xkAp+lQNI9jKN0ZojMiSQHXr4FoThkNTV1r wWUlYwTeDJlm1HJgqcNMIjhegeG1GJrmz3i5BZgW7WBI65TL/8MfwEozFO8jjtzD R44gF0sIg7rWEfzO3jFuhr+F3w9vwZ24KzAFPLjAIJlmB/YiBK7WwiIXI0tQhdg/ /72vVgQQMGsmSauZ0j2hjfImGDU45wLZ21w26/kkn7F6k9k+6BRFsedI1qI0Bf7V UqTG4oWLaZ5+EYQhvLH4VjYtH7h3UgGacHiYHcKHnmL49rr6GjDWQqBE8pieAtkQ ckhTw+fpMPQZd6T6J5OX5VNsnWcEG4C89PPCvYL9Jvw8tq6uzcTdRbxc+Thu3Plr I6hmcM5Ibz0n+nvqWLzsY/4voQb/CtiZWQQZ+c2VbIx1HpWsKTH0NFdHX891faTn ijsm8wsO0YcyvtlFPfZxKoQfb1GiBN+ttdlYStS7i15YmDujl7GYlXeHz0nr9FYv nNpHoNkz+U+fH3Neeh4epCOGxUhkAPWG/u1tMTAxQad0JiLpedKxB/IMBen+16pb IT4deYFnfsQjp/HAQfHQ =h5Nz -----END PGP SIGNATURE-----
So the idea here to make it possible to have a function that can be specialized at certain types, and explicitly inlined at specific use sites, but ghc otherwise will not inline it? Cool!
That's almost exactly what INLINABLE means. I agree that SPECIALISABLE would have been a better name.
The only difference between INLINABLE and what you say is that GHC is *free* to inline an INLINABLE if it thinks it'd be a good idea, whereas you want a promise that it will never do so. (I'm not sure why.) But they are pretty close already.
My suggestion
* Rename INLINABLE to SPECIALISABLE (deprecating the former)
* Allow SPECIALISABLE in conjunction with the existing NOINLINE
Simon
From: ghc-devs [mailto:ghc-devs-bounces@haskell.org] On Behalf Of Nicolas Frisby
Sent: 18 July 2013 23:19
To: ghc-devs@haskell.org
Subject: new pragma name ideas? (was: defunctionalization)
On Thu, Jul 18, 2013 at 5:10 PM, Carter Schonwald
participants (3)
-
Mateusz Kowalczyk -
Nicolas Frisby -
Simon Peyton-Jones