From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.220.29]) by sourceware.org (Postfix) with ESMTPS id 25B10383A833 for ; Tue, 27 Jul 2021 08:45:02 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org 25B10383A833 Authentication-Results: sourceware.org; dmarc=none (p=none dis=none) header.from=suse.de Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=suse.de Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id 273281FD59; Tue, 27 Jul 2021 08:45:01 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1627375501; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=6QlEL6ANwaA5cf2Nc1G4B9ctgQJ1Ez3ZrfK++nHeF+g=; b=tswchRlmwUKI/hjdewcQ7yE3I2YEgBfoqB8UDsLPEby3NWIqEG/14Baxo9MLT/xjnWTb0h Kh0yGvffpBIsGR2apht91NMtLW7UA8E1Dt67aEA+cSAQNUaTayvt4CYEPxSVcACGQGM1IB xKSkUwRyb5eipEk1phWIhsLW82OmbEo= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1627375501; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=6QlEL6ANwaA5cf2Nc1G4B9ctgQJ1Ez3ZrfK++nHeF+g=; b=JhQyle1ISyX3laRXPRts00LFID9YT6vyWMdPP+zqyNmL6o2Q1L9qCcBsAxT7oGDNGoXhjs Z5ast+vstEs5VeDA== Received: from imap2.suse-dmz.suse.de (imap2.suse-dmz.suse.de [192.168.254.74]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-521) server-digest SHA512) (No client certificate requested) by imap2.suse-dmz.suse.de (Postfix) with ESMTPS id 0758A1377D; Tue, 27 Jul 2021 08:45:01 +0000 (UTC) Received: from dovecot-director2.suse.de ([192.168.254.65]) by imap2.suse-dmz.suse.de with ESMTPSA id w8iiAI3H/2CoawAAMHmgww (envelope-from ); Tue, 27 Jul 2021 08:45:01 +0000 Date: Tue, 27 Jul 2021 10:45:00 +0200 (CEST) From: Richard Biener To: Jeff Law cc: gcc-patches@gcc.gnu.org, msebor@redhat.com Subject: Re: [PATCH] tree-optimization/101573 - improve uninit warning at -O0 In-Reply-To: <58095323-cc9f-a9fa-694b-70061acba908@gmail.com> Message-ID: <85923o4-rsp3-rrr9-295q-8o46sq328n3@fhfr.qr> References: <3on3509q-309q-8562-r54n-qr20o3ns1p5q@fhfr.qr> <58095323-cc9f-a9fa-694b-70061acba908@gmail.com> MIME-Version: 1.0 X-Spam-Status: No, score=-11.0 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, KAM_NUMSUBJECT, 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 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8BIT X-Content-Filtered-By: Mailman/MimeDel 2.1.29 X-BeenThere: gcc-patches@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-patches mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 27 Jul 2021 08:45:04 -0000 On Fri, 23 Jul 2021, Jeff Law wrote: > > > On 7/22/2021 6:34 AM, Richard Biener wrote: > > We can improve uninit warnings from the early pass by looking > > at PHI arguments on fallthru edges that are uninitialized and > > have uses that are before a possible loop exit. This catches > > some cases earlier that we'd only warn in a more confusing > > way after early inlining as seen by testcase adjustments. > > > > It introduces > > > > FAIL: gcc.dg/uninit-23.c (test for excess errors) > > > > where we additionally warn > > > > gcc.dg/uninit-23.c:21:13: warning: 't4' is used uninitialized > > [-Wuninitialized] > > > > which I think is OK even if it's not obvious that the new > > warning is an improvement when you look at the obvious source. > > > > Somehow for all cases I never get the `'foo' was declared here` > > notes, I didn't dig why that happens but it's odd. > > > > Bootstrapped and tested on x86_64-unknown-linux-gnu. > > > > Any comments? > > > > Thanks, > > Richard. > > > > 2021-07-22 Richard Biener > > > > PR tree-optimization/101573 > > * tree-ssa-uninit.c (warn_uninitialized_vars): Look at > > uninitialized PHI arg defs in some constrained cases. > > (execute_early_warn_uninitialized): Calculate dominators. > > > > * gcc.dg/uninit-pr101573.c: New testcase. > > * gcc.dg/uninit-15-O0.c: Adjust. > > * gcc.dg/uninit-15.c: Likewise. > > * gcc.dg/uninit-23.c: Likewise. > > * c-c++-common/uninit-17.c: Likewise. > OK.  Like Martin I think the new code in a function would be easier to read.  > So if you could factor the bits into a new function it'd be appreciated. > > I wouldn't be terribly surprised if we find additional fallout on other > targets, but we can fault in those fixes. The following is what I applied then. Thanks, Richard. >From 8fb649db98f3cf49c7d4fd5e566318d2e5ef9756 Mon Sep 17 00:00:00 2001 From: Richard Biener Date: Thu, 22 Jul 2021 12:26:16 +0200 Subject: [PATCH] tree-optimization/101573 - improve uninit warning at -O0 To: gcc-patches@gcc.gnu.org We can improve uninit warnings from the early pass by looking at PHI arguments on fallthru edges that are uninitialized and have uses that are before a possible loop exit. This catches some cases earlier that we'd only warn in a more confusing way after early inlining as seen by testcase adjustments. It introduces FAIL: gcc.dg/uninit-23.c (test for excess errors) where we additionally warn gcc.dg/uninit-23.c:21:13: warning: 't4' is used uninitialized [-Wuninitialized] which I think is OK even if it's not obvious that the new warning is an improvement when you look at the obvious source. Somehow for all cases I never get the `'foo' was declared here` notes, I didn't dig why that happens but it's odd. 2021-07-22 Richard Biener PR tree-optimization/101573 * tree-ssa-uninit.c (warn_uninit_phi_uses): New function looking at uninitialized PHI arg defs in some constrained cases. (warn_uninitialized_vars): Call it. (execute_early_warn_uninitialized): Calculate dominators. * gcc.dg/uninit-pr101573.c: New testcase. * gcc.dg/uninit-15-O0.c: Adjust. * gcc.dg/uninit-15.c: Likewise. * gcc.dg/uninit-23.c: Likewise. * c-c++-common/uninit-17.c: Likewise. --- gcc/testsuite/c-c++-common/uninit-17.c | 6 +-- gcc/testsuite/gcc.dg/uninit-15-O0.c | 2 +- gcc/testsuite/gcc.dg/uninit-15.c | 10 ++-- gcc/testsuite/gcc.dg/uninit-23.c | 2 +- gcc/testsuite/gcc.dg/uninit-pr101573.c | 10 ++++ gcc/tree-ssa-uninit.c | 75 ++++++++++++++++++++++++++ 6 files changed, 95 insertions(+), 10 deletions(-) create mode 100644 gcc/testsuite/gcc.dg/uninit-pr101573.c diff --git a/gcc/testsuite/c-c++-common/uninit-17.c b/gcc/testsuite/c-c++-common/uninit-17.c index fd773da78ad..b5495366c5b 100644 --- a/gcc/testsuite/c-c++-common/uninit-17.c +++ b/gcc/testsuite/c-c++-common/uninit-17.c @@ -9,11 +9,11 @@ static void bar(int a, int *ptr) { do { - int b; /* { dg-message "declared" } */ + int b; if (b < 40) { - ptr[0] = b; /* { dg-warning "may be used uninitialized" } */ + ptr[0] = b; } - b += 1; + b += 1; /* { dg-warning "is used uninitialized" } */ ptr++; } while (--a != 0); diff --git a/gcc/testsuite/gcc.dg/uninit-15-O0.c b/gcc/testsuite/gcc.dg/uninit-15-O0.c index a3fd2b63ba7..36d96348617 100644 --- a/gcc/testsuite/gcc.dg/uninit-15-O0.c +++ b/gcc/testsuite/gcc.dg/uninit-15-O0.c @@ -15,6 +15,6 @@ void baz(); void bar() { int j; /* { dg-message "was declared here" {} { xfail *-*-* } } */ - for (; foo(j); ++j) + for (; foo(j); ++j) /* { dg-warning "is used uninitialized" } */ baz(); } diff --git a/gcc/testsuite/gcc.dg/uninit-15.c b/gcc/testsuite/gcc.dg/uninit-15.c index 8ee10c27aba..85cb116f349 100644 --- a/gcc/testsuite/gcc.dg/uninit-15.c +++ b/gcc/testsuite/gcc.dg/uninit-15.c @@ -1,7 +1,7 @@ /* PR tree-optimization/17506 - We issue an uninitialized variable warning at a wrong location at + We used to issue an uninitialized variable warning at a wrong location at line 11, which is very confusing. Make sure we print out a note to - make it less confusing. (not xfailed alternative) + make it less confusing. (xfailed alternative) But it is of course ok if we warn in bar about uninitialized use of j. (not xfailed alternative) */ /* { dg-do compile } */ @@ -10,7 +10,7 @@ inline int foo (int i) { - if (i) /* { dg-warning "used uninitialized" } */ + if (i) /* { dg-warning "used uninitialized" "" { xfail *-*-* } } */ return 1; return 0; } @@ -20,7 +20,7 @@ void baz (void); void bar (void) { - int j; /* { dg-message "note: 'j' was declared here" "" } */ - for (; foo (j); ++j) /* { dg-warning "'j' is used uninitialized" "" { xfail *-*-* } } */ + int j; /* { dg-message "note: 'j' was declared here" "" { xfail *-*-* } } */ + for (; foo (j); ++j) /* { dg-warning "'j' is used uninitialized" } */ baz (); } diff --git a/gcc/testsuite/gcc.dg/uninit-23.c b/gcc/testsuite/gcc.dg/uninit-23.c index d64eb7d2ee9..87b4e989b53 100644 --- a/gcc/testsuite/gcc.dg/uninit-23.c +++ b/gcc/testsuite/gcc.dg/uninit-23.c @@ -18,7 +18,7 @@ ql (void) int *t4 = go; /* { dg-warning "is used uninitialized" } */ l1: - *t4 = (*t4 != 0) ? 0 : 2; + *t4 = (*t4 != 0) ? 0 : 2; /* { dg-warning "is used uninitialized" } */ } if (ij != 0) diff --git a/gcc/testsuite/gcc.dg/uninit-pr101573.c b/gcc/testsuite/gcc.dg/uninit-pr101573.c new file mode 100644 index 00000000000..a574844b791 --- /dev/null +++ b/gcc/testsuite/gcc.dg/uninit-pr101573.c @@ -0,0 +1,10 @@ +/* { dg-do compile } */ +/* { dg-options "-O0 -Wuninitialized" } */ + +int main(int argc, char **argv) +{ + int a; + for(; a < 5; ++a) /* { dg-warning "is used uninitialized" } */ + ; + return 0; +} diff --git a/gcc/tree-ssa-uninit.c b/gcc/tree-ssa-uninit.c index 148f3c2b31d..718b32691c1 100644 --- a/gcc/tree-ssa-uninit.c +++ b/gcc/tree-ssa-uninit.c @@ -638,6 +638,76 @@ maybe_warn_pass_by_reference (gcall *stmt, wlimits &wlims) wlims.always_executed = save_always_executed; } +/* Warn about an uninitialized PHI argument on the fallthru path to + an always executed block BB. */ + +static void +warn_uninit_phi_uses (basic_block bb) +{ + edge_iterator ei; + edge e, found = NULL, found_back = NULL; + /* Look for a fallthru and possibly a single backedge. */ + FOR_EACH_EDGE (e, ei, bb->preds) + { + /* Ignore backedges. */ + if (dominated_by_p (CDI_DOMINATORS, e->src, bb)) + { + if (found_back) + { + found = NULL; + break; + } + found_back = e; + continue; + } + if (found) + { + found = NULL; + break; + } + found = e; + } + if (!found) + return; + + basic_block succ = single_succ (ENTRY_BLOCK_PTR_FOR_FN (cfun)); + for (gphi_iterator si = gsi_start_phis (bb); !gsi_end_p (si); + gsi_next (&si)) + { + gphi *phi = si.phi (); + tree def = PHI_ARG_DEF_FROM_EDGE (phi, found); + if (TREE_CODE (def) != SSA_NAME + || !SSA_NAME_IS_DEFAULT_DEF (def) + || virtual_operand_p (def)) + continue; + /* If there's a default def on the fallthru edge PHI + value and there's a use that post-dominates entry + then that use is uninitialized and we can warn. */ + imm_use_iterator iter; + use_operand_p use_p; + gimple *use_stmt = NULL; + FOR_EACH_IMM_USE_FAST (use_p, iter, gimple_phi_result (phi)) + { + use_stmt = USE_STMT (use_p); + if (gimple_location (use_stmt) != UNKNOWN_LOCATION + && dominated_by_p (CDI_POST_DOMINATORS, succ, + gimple_bb (use_stmt)) + /* If we found a non-fallthru edge make sure the + use is inside the loop, otherwise the backedge + can serve as initialization. */ + && (!found_back + || dominated_by_p (CDI_DOMINATORS, found_back->src, + gimple_bb (use_stmt)))) + break; + use_stmt = NULL; + } + if (use_stmt) + warn_uninit (OPT_Wuninitialized, def, SSA_NAME_VAR (def), + SSA_NAME_VAR (def), + "%qD is used uninitialized", use_stmt, + UNKNOWN_LOCATION); + } +} static unsigned int warn_uninitialized_vars (bool wmaybe_uninit) @@ -652,6 +722,10 @@ warn_uninitialized_vars (bool wmaybe_uninit) { basic_block succ = single_succ (ENTRY_BLOCK_PTR_FOR_FN (cfun)); wlims.always_executed = dominated_by_p (CDI_POST_DOMINATORS, succ, bb); + + if (wlims.always_executed) + warn_uninit_phi_uses (bb); + for (gsi = gsi_start_bb (bb); !gsi_end_p (gsi); gsi_next (&gsi)) { gimple *stmt = gsi_stmt (gsi); @@ -3135,6 +3209,7 @@ execute_early_warn_uninitialized (void) optimization we want to warn about possible uninitialized as late as possible, thus don't do it here. However, without optimization we need to warn here about "may be uninitialized". */ + calculate_dominance_info (CDI_DOMINATORS); calculate_dominance_info (CDI_POST_DOMINATORS); warn_uninitialized_vars (/*warn_maybe_uninitialized=*/!optimize); -- 2.26.2