From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham autolearn_force=no version=3.4.4 X-Google-Thread: 103376,3e26dfa741e64e5f X-Google-Attributes: gid103376,public X-Google-Language: ENGLISH,ASCII Path: g2news1.google.com!postnews.google.com!g14g2000cwa.googlegroups.com!not-for-mail From: "Ludovic Brenta" Newsgroups: comp.lang.ada Subject: Re: GNAT GPL 2005 Edition is now available Date: 20 Sep 2005 03:28:58 -0700 Organization: http://groups.google.com Message-ID: <1127212138.672063.41070@g14g2000cwa.googlegroups.com> References: <432919be$0$10539$4d4eb98e@read.news.fr.uu.net> <1126773856.876636.265130@g49g2000cwa.googlegroups.com> <1126794444.071827.191320@o13g2000cwo.googlegroups.com> <877jdecrhj.fsf@mid.deneb.enyo.de> <877jde2vyx.fsf@ludovic-brenta.org> <87zmqa9vpy.fsf@mid.deneb.enyo.de> <87y85u1f34.fsf@ludovic-brenta.org> <87aci8m5jz.fsf@mid.deneb.enyo.de> NNTP-Posting-Host: 212.190.145.10 Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable X-Trace: posting.google.com 1127212143 31042 127.0.0.1 (20 Sep 2005 10:29:03 GMT) X-Complaints-To: groups-abuse@google.com NNTP-Posting-Date: Tue, 20 Sep 2005 10:29:03 +0000 (UTC) In-Reply-To: <87aci8m5jz.fsf@mid.deneb.enyo.de> User-Agent: G2/0.2 X-HTTP-UserAgent: Mozilla/5.0 (X11; U; SunOS sun4u; fr-FR; rv:1.6) Gecko/20040116,gzip(gfe),gzip(gfe) X-HTTP-Via: 1.1 SEVPXS01 Complaints-To: groups-abuse@google.com Injection-Info: g14g2000cwa.googlegroups.com; posting-host=212.190.145.10; posting-account=ZjNXewwAAADyBPkwI57_UcX8yKfXWOss Xref: g2news1.google.com comp.lang.ada:4945 Date: 2005-09-20T03:28:58-07:00 List-Id: Florian Weimer a =E9crit : [tasking on all processors on SMP systems] > Ah, I saw the problems on a quad-Xeon machine (with hyperthreading), > Linux 2.6.0-test9 (slightly patched, don't ask) and NTPL. The latter > is probably the culprit. I don't think NPTL is the culprit, as I backported support for it into gnat 3.15p-10 from gnat-3.4. And I use native threads (as opposed to FSU threads) in gnat. It could be something stupid like a uniprocessor kernel? Don't laugh, I know from experience that it happens :) Maybe I should try it again on a 2.6 kernel on an SMP system. I think I did that a while ago, and it worked. > * Ludovic Brenta: >> GNAT GPL 2005 Edition is based on GCC 3.4.4 (with several patches). >> It may or may not ICE on your code like GCC 3.4 does. I suggest >> you try it and decide for yourself which one suits you best. > > Okay, downloaded it and ran it through my testsuite. It passes, on > x86 at least. However, there are two quite unsettling warnings: one > about an initialized variable of type Exception_Occurrence (the RM > guarantees that objects of this type are initialized to > Null_Occurrence), and a claim that a certain expression is always > false, which doesn't seem to match reality. (Both issues are also > present in the GCC 4 codebase.) Could you now summarise your results? I'm curious to know how the four compare: gnat, gnat-3.4, gnat-4.0 and GNAT GPL. If you could post a "quality assessment matrix" with i386 and amd64, that would be very helpful for everyone here. > There's also a new restriction on pragma Inline_Always: it can't be > applied to dispatching subprograms. This doesn't make much sense > because often, the tag is known at compile time, so cross-compilation > unit inlining would still be possible. Inline_Always was nonstandard, but if I remember correctly it is being introduced into Ada 200y. In any case, you cannot rely on it in Ada 95 programs. But you are correct that inlining non-dispatching calls should be feasible. --=20 Ludovic Brenta.