From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gnu.wildebeest.org (wildebeest.demon.nl [212.238.236.112]) by sourceware.org (Postfix) with ESMTPS id E544E38515E8 for ; Mon, 16 Aug 2021 21:03:27 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org E544E38515E8 Authentication-Results: sourceware.org; dmarc=none (p=none dis=none) header.from=klomp.org Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=klomp.org Received: from reform (unknown [172.31.128.41]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by gnu.wildebeest.org (Postfix) with ESMTPSA id ECE4F302FBB1; Mon, 16 Aug 2021 23:03:20 +0200 (CEST) Received: by reform (Postfix, from userid 1000) id BFBF12E81EBF; Mon, 16 Aug 2021 23:03:19 +0200 (CEST) Date: Mon, 16 Aug 2021 23:03:19 +0200 From: Mark Wielaard To: dkm@kataplop.net, gcc-rust@gcc.gnu.org Subject: Re: [PATCH] Add support for const bool and const float Message-ID: References: <20210815195526.59273-1-mark@klomp.org> <0e8cd730cb7867bd1970b20d99a55f1c@kataplop.net> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: X-Spam-Status: No, score=-5.1 required=5.0 tests=BAYES_00, KAM_DMARC_STATUS, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.4 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on server2.sourceware.org X-BeenThere: gcc-rust@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: gcc-rust mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 16 Aug 2021 21:03:38 -0000 Hi, On Mon, Aug 16, 2021 at 10:52:07AM +0200, Mark Wielaard wrote: > On Mon, 2021-08-16 at 07:48 +0000, dkm@kataplop.net wrote: > > Looks like tests are not OK, at least in the github action. > > Are you talking about this patch (which I cannot find on github) or > "Use builtin bool instead of creating new bool types for > ComparisonExpr"? On github I can only see that one has a red cross > which says "build-and-check" but doesn't give any more information > except that "Check regressions" failed, but without any logs ("Sign in > for the full log view"). > > > Can't test but maybe you can confirm this issue ? > > > > # of unexpected failures 14 > > On my setup, Debian arm64 with gcc 10.2, for both patches the newly > added testcase fails without the patch and succeeds with the patch. > I'll try to find some other setup. What is the configuration for the > machine where you see the new unexpected failures (and with which > patch)? On irc we tracked down the confusion. This is about the other patch "Use builtin bool instead of creating new bool types for ComparisonExpr". But rebased on top of current master. Which includes the testcase rust/compile/torture/ifunaryexpr.rs So the above failures where: FAIL: rust/compile/torture/ifunaryexpr.rs -O0 (internal compiler error) FAIL: rust/compile/torture/ifunaryexpr.rs -O0 (test for excess errors) FAIL: rust/compile/torture/ifunaryexpr.rs -O1 (internal compiler error) FAIL: rust/compile/torture/ifunaryexpr.rs -O1 (test for excess errors) FAIL: rust/compile/torture/ifunaryexpr.rs -O2 (internal compiler error) FAIL: rust/compile/torture/ifunaryexpr.rs -O2 (test for excess errors) FAIL: rust/compile/torture/ifunaryexpr.rs -O3 -g (internal compiler error) FAIL: rust/compile/torture/ifunaryexpr.rs -O3 -g (test for excess errors) FAIL: rust/compile/torture/ifunaryexpr.rs -Os (internal compiler error) FAIL: rust/compile/torture/ifunaryexpr.rs -Os (test for excess errors) FAIL: rust/compile/torture/ifunaryexpr.rs -O2 -flto -fno-use-linker-plugin -flto-partition=none (internal compiler error) FAIL: rust/compile/torture/ifunaryexpr.rs -O2 -flto -fno-use-linker-plugin -flto-partition=none (test for excess errors) FAIL: rust/compile/torture/ifunaryexpr.rs -O2 -flto -fuse-linker-plugin -fno-fat-lto-objects (internal compiler error) FAIL: rust/compile/torture/ifunaryexpr.rs -O2 -flto -fuse-linker-plugin -fno-fat-lto-objects (test for excess errors) Now having replicated it locally the rust.log shows: /srv/gccrs/gccrs/gcc/testsuite/rust/compile/torture/ifunaryexpr.rs: In function 'main': /srv/gccrs/gccrs/gcc/testsuite/rust/compile/torture/ifunaryexpr.rs:8:5: error: mismatching comparison operand types const i32 bool if (n != 0) goto ; else goto ; /srv/gccrs/gccrs/gcc/testsuite/rust/compile/torture/ifunaryexpr.rs:8:5: internal compiler error: 'verify _gimple' failed 0xf38fbd verify_gimple_in_seq(gimple*) ../../gccrs/gcc/tree-cfg.c:5157 0xc72346 gimplify_body(tree_node*, bool) ../../gccrs/gcc/gimplify.c:15401 0xc724cd gimplify_function_tree(tree_node*) ../../gccrs/gcc/gimplify.c:15472 0xab0dc7 cgraph_node::analyze() ../../gccrs/gcc/cgraphunit.c:670 0xab38b7 analyze_functions ../../gccrs/gcc/cgraphunit.c:1236 0xab454d symbol_table::finalize_compilation_unit() ../../gccrs/gcc/cgraphunit.c:2514 So, my patch created bad gimple. I'll try to track it down. Cheers, Mark