From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.io!.POSTED.blaine.gmane.org!not-for-mail From: "Basil L. Contovounesios" via "Bug reports for GNU Emacs, the Swiss army knife of text editors" Newsgroups: gmane.emacs.bugs Subject: bug#58975: 29.0.50; noverlay fails to build with --enable-checking=structs Date: Thu, 03 Nov 2022 14:26:50 +0200 Message-ID: <87zgd8dy0l.fsf@tcd.ie> References: <874jvg3n8z.fsf@tcd.ie> Reply-To: "Basil L. Contovounesios" Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="=-=-=" Injection-Info: ciao.gmane.io; posting-host="blaine.gmane.org:116.202.254.214"; logging-data="32876"; mail-complaints-to="usenet@ciao.gmane.io" User-Agent: Gnus/5.13 (Gnus v5.13) Cc: Matt Armstrong , 58975@debbugs.gnu.org To: Stefan Monnier Original-X-From: bug-gnu-emacs-bounces+geb-bug-gnu-emacs=m.gmane-mx.org@gnu.org Thu Nov 03 13:28:42 2022 Return-path: Envelope-to: geb-bug-gnu-emacs@m.gmane-mx.org Original-Received: from lists.gnu.org ([209.51.188.17]) by ciao.gmane.io with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1oqZKr-0008OV-W9 for geb-bug-gnu-emacs@m.gmane-mx.org; Thu, 03 Nov 2022 13:28:42 +0100 Original-Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1oqZKX-00077j-9p; Thu, 03 Nov 2022 08:28:23 -0400 Original-Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1oqZKF-00074g-U5 for bug-gnu-emacs@gnu.org; Thu, 03 Nov 2022 08:28:04 -0400 Original-Received: from debbugs.gnu.org ([209.51.188.43]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1oqZKF-0006jF-5c for bug-gnu-emacs@gnu.org; Thu, 03 Nov 2022 08:28:03 -0400 Original-Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1oqZKE-0004MP-HH for bug-gnu-emacs@gnu.org; Thu, 03 Nov 2022 08:28:02 -0400 X-Loop: help-debbugs@gnu.org Resent-From: "Basil L. Contovounesios" Original-Sender: "Debbugs-submit" Resent-CC: bug-gnu-emacs@gnu.org Resent-Date: Thu, 03 Nov 2022 12:28:02 +0000 Resent-Message-ID: Resent-Sender: help-debbugs@gnu.org X-GNU-PR-Message: followup 58975 X-GNU-PR-Package: emacs X-GNU-PR-Keywords: patch Original-Received: via spool by 58975-submit@debbugs.gnu.org id=B58975.166747842316684 (code B ref 58975); Thu, 03 Nov 2022 12:28:02 +0000 Original-Received: (at 58975) by debbugs.gnu.org; 3 Nov 2022 12:27:03 +0000 Original-Received: from localhost ([127.0.0.1]:48304 helo=debbugs.gnu.org) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1oqZJG-0004Ks-Eb for submit@debbugs.gnu.org; Thu, 03 Nov 2022 08:27:03 -0400 Original-Received: from mail-ej1-f50.google.com ([209.85.218.50]:42910) by debbugs.gnu.org with esmtp (Exim 4.84_2) (envelope-from ) id 1oqZJD-0004KV-85 for 58975@debbugs.gnu.org; Thu, 03 Nov 2022 08:27:01 -0400 Original-Received: by mail-ej1-f50.google.com with SMTP id y14so4711824ejd.9 for <58975@debbugs.gnu.org>; Thu, 03 Nov 2022 05:26:59 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=tcd.ie; s=google21; h=mime-version:user-agent:message-id:date:references:in-reply-to :subject:cc:to:from:from:to:cc:subject:date:message-id:reply-to; bh=BWHgE2U5zVp17cxqOJR2hCbfhYpz0fixYaAoD0KT8tM=; b=MgzPxQ/J8E4cLG4MTwPNwDo2Gbebu5+hMEoeI3fLG9HGiiw/asjC/A/JwOOEuMUUF5 ytEwWwKAo+EVPtwOjz2dSxm9M6cAWAsIR3XrSwSkYctK+TfC7BhOpA8ExFWgg+tc1TP8 Y5j6oZhg95egw6wCJBspbG8QVqeBf7GirqYPwa4qNHvbI0iJF7AVEivzscXPYjKN3NrX dW0MtS4WP2YXeiF4kxK3eLqndcLVkn1c22fQOP49UCY31ffuhLREks5dChVYnT+Heu2o odIHTf11O9LhV+FuMsGVAEBxPdCr1W+/pplUz00ebTQokEEijPWhlrjKI9aZNKIsgC0N UoNw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=mime-version:user-agent:message-id:date:references:in-reply-to :subject:cc:to:from:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=BWHgE2U5zVp17cxqOJR2hCbfhYpz0fixYaAoD0KT8tM=; b=beyN6ZKYmqSIda9VQZMA2voGaulwAaBtVAQIN+4CrSbLW9KnhxwCPB75MvDqvCcmo7 NFBJgHKHsMpBaKmCi+vj+zwOfz4TL55JHb79ZB+4lvsjWLvmI7nm50c1pVntize7P993 J9++3pEPIC8qMhbaaYQAmLaQqk39V49DQJHnmcd1YBr0PgbX5XM90AKr7/bAU1qxcqfA GJ9298f5yqgCXbeUIQavxvSZjD3CXaOf6/tWcdsQaPbzbN/QKAnZnzVFyQIa3DXlez8j HmMKDqOt+OudbLAUGWZlGFa5Y8bE6wv7N+dAvzECtos91oMqpVExq1XS7weUCFpMxVPr RRFg== X-Gm-Message-State: ACrzQf1ylU5uZU1L/SPmGfCzwx8d7VUKZhXZUuDVceWR3JajyixNhdcR 3TdbFHmGtr4Mql9kwDI4SZZ4Rw== X-Google-Smtp-Source: AMsMyM6NsSNj03K5cPk6BiLygnS45Lw1Ci8gXIM34+7O+Nb9YJ9XY89t4ijYHJH4fuXM7+DmimWjkw== X-Received: by 2002:a17:907:6d0f:b0:7ad:c0d5:c7b5 with SMTP id sa15-20020a1709076d0f00b007adc0d5c7b5mr25224083ejc.262.1667478412999; Thu, 03 Nov 2022 05:26:52 -0700 (PDT) Original-Received: from localhost ([2a02:587:320c:8829:23:8156:16ed:40c2]) by smtp.gmail.com with ESMTPSA id 27-20020a170906059b00b0077a201f6d1esm438727ejn.87.2022.11.03.05.26.51 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 03 Nov 2022 05:26:51 -0700 (PDT) In-Reply-To: (Stefan Monnier's message of "Wed, 02 Nov 2022 21:55:07 -0400") X-BeenThere: debbugs-submit@debbugs.gnu.org X-Mailman-Version: 2.1.18 Precedence: list X-BeenThere: bug-gnu-emacs@gnu.org List-Id: "Bug reports for GNU Emacs, the Swiss army knife of text editors" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Original-Sender: "bug-gnu-emacs" Errors-To: bug-gnu-emacs-bounces+geb-bug-gnu-emacs=m.gmane-mx.org@gnu.org Xref: news.gmane.io gmane.emacs.bugs:246954 Archived-At: --=-=-= Content-Type: text/plain Stefan Monnier [2022-11-02 21:55 -0400] wrote: >> Is this the right fix? > > The patch looks good, thanks. > >> BTW, every time itree_create is called via add_buffer_overlay, it in >> turn calls itree_init, > > Oh, that's not right. `itree_init` should be called once only, from > `emacs.c`, like things like `init_lread`. Like this? --=-=-= Content-Type: text/x-diff Content-Disposition: attachment; filename=0001-Port-interval-trees-to-enable-checking-structs.patch >From 321484838251ec8cea4cf296e3b8b0134ba892bc Mon Sep 17 00:00:00 2001 From: "Basil L. Contovounesios" Date: Wed, 2 Nov 2022 03:50:38 +0200 Subject: [PATCH] Port interval trees to --enable-checking=structs Some names under the interval_* namespace were renamed under the itree_* namespace in commits: 0. f421b58db5 of 2022-10-19 "Prefix all itree.h type names with itree_". 1. 37a1145410 of 2022-10-19 "Rename all exported itree.h functions with the itree_ prefix" Further, some values still referenced in commentary were removed in commits: 2. 258e618364 of 2022-10-17 "Delete the itree_null sentinel node, use NULL everywhere." 3. 2c4a3910b3 of 2022-10-02 "itree: Use a single iterator object" * src/emacs.c (main): Allocate global itree iterator once and for all. * src/alloc.c (mark_overlay): * src/buffer.c (set_overlays_multibyte): * src/itree.c (itree_destroy): Update commentary. (interval_stack_ensure_space, itree_insert_gap): Prefer unsigned-to-unsigned comparisons over signed-to-unsigned. (interval_stack_push_flagged, interval_tree_insert) (interval_tree_contains, itree_iterator_start) (itree_iterator_finish, itree_iterator_next, itree_iterator_narrow): Improve assertions. (itree_init): Rename... (init_itree): ...to this, for consistency with other global init functions. (itree_create): Stop leaking a global iterator allocation on each call. (interval_tree_init): Complete renames of interval_tree -> itree_tree and interval_tree_clear -> itree_clear. (interval_tree_remove_fix): Fix indentation. * src/itree.h: Declare init_itree. (ITREE_FOREACH): Fix typo in commentary. * src/pdumper.c [CHECK_STRUCTS] (dump_interval_node): Use the correct name in the HASH condition and #error message. (dump_overlay, dump_buffer): Update HASH (bug#58975). --- src/alloc.c | 2 +- src/buffer.c | 2 +- src/emacs.c | 2 ++ src/itree.c | 42 +++++++++++++++++------------------------- src/itree.h | 3 ++- src/pdumper.c | 8 ++++---- 6 files changed, 27 insertions(+), 32 deletions(-) diff --git a/src/alloc.c b/src/alloc.c index f69c65dedc..6862cf916f 100644 --- a/src/alloc.c +++ b/src/alloc.c @@ -6508,7 +6508,7 @@ mark_char_table (struct Lisp_Vector *ptr, enum pvec_type pvectype) static void mark_overlay (struct Lisp_Overlay *ov) { - /* We don't mark the `interval_node` object, because it is managed manually + /* We don't mark the `itree_node` object, because it is managed manually rather than by the GC. */ eassert (BASE_EQ (ov->interval->data, make_lisp_ptr (ov, Lisp_Vectorlike))); set_vectorlike_marked (&ov->header); diff --git a/src/buffer.c b/src/buffer.c index 3129aa2890..3b0e6f1f9a 100644 --- a/src/buffer.c +++ b/src/buffer.c @@ -982,7 +982,7 @@ set_overlays_multibyte (bool multibyte) struct itree_tree *tree = current_buffer->overlays; const intmax_t size = itree_size (tree); - /* We can't use `interval_node_set_region` at the same time + /* We can't use `itree_node_set_region` at the same time as we iterate over the itree, so we need an auxiliary storage to keep the list of nodes. */ USE_SAFE_ALLOCA; diff --git a/src/emacs.c b/src/emacs.c index 8ad70fecd4..40ba0db340 100644 --- a/src/emacs.c +++ b/src/emacs.c @@ -82,6 +82,7 @@ #define MAIN_PROGRAM #endif /* HAVE_WINDOW_SYSTEM */ #include "bignum.h" +#include "itree.h" #include "intervals.h" #include "character.h" #include "buffer.h" @@ -1931,6 +1932,7 @@ main (int argc, char **argv) running_asynch_code = 0; init_random (); init_xfaces (); + init_itree (); #if defined HAVE_JSON && !defined WINDOWSNT init_json (); diff --git a/src/itree.c b/src/itree.c index bd4e8cc574..3137cb6358 100644 --- a/src/itree.c +++ b/src/itree.c @@ -191,7 +191,7 @@ interval_stack_clear (struct interval_stack *stack) } static inline void -interval_stack_ensure_space (struct interval_stack *stack, intmax_t nelements) +interval_stack_ensure_space (struct interval_stack *stack, uintmax_t nelements) { if (nelements > stack->size) { @@ -207,7 +207,7 @@ interval_stack_ensure_space (struct interval_stack *stack, intmax_t nelements) interval_stack_push_flagged (struct interval_stack *stack, struct itree_node *node, bool flag) { - eassert (node && node != NULL); + eassert (node); /* FIXME: While the stack used in the iterator is bounded by the tree depth and could be easily pre-allocated to a large enough size to avoid @@ -287,8 +287,8 @@ itree_iterator_create (struct itree_tree *tree) return g; } -static void -itree_init (void) +void +init_itree (void) { iter = itree_iterator_create (NULL); } @@ -555,16 +555,11 @@ itree_node_end (struct itree_tree *tree, return node->end; } -/* Allocate an interval_tree. Free with interval_tree_destroy. */ +/* Allocate an itree_tree. Free with itree_destroy. */ struct itree_tree * itree_create (void) { - /* FIXME? Maybe avoid the initialization of itree_null in the same - way that is used to call mem_init in alloc.c? It's not really - important though. */ - itree_init (); - struct itree_tree *tree = xmalloc (sizeof (*tree)); itree_clear (tree); return tree; @@ -584,10 +579,9 @@ itree_clear (struct itree_tree *tree) /* Initialize a pre-allocated tree (presumably on the stack). */ static void -interval_tree_init (struct interval_tree *tree) +interval_tree_init (struct itree_tree *tree) { - interval_tree_clear (tree); - /* tree->iter = itree_iterator_create (tree); */ + itree_clear (tree); } #endif @@ -596,8 +590,6 @@ interval_tree_init (struct interval_tree *tree) itree_destroy (struct itree_tree *tree) { eassert (tree->root == NULL); - /* if (tree->iter) - * itree_iterator_destroy (tree->iter); */ xfree (tree); } @@ -775,7 +767,7 @@ interval_tree_insert_fix (struct itree_tree *tree, static void interval_tree_insert (struct itree_tree *tree, struct itree_node *node) { - eassert (node->begin <= node->end && node != NULL); + eassert (node && node->begin <= node->end); /* FIXME: The assertion below fails because `delete_all_overlays` doesn't set left/right/parent to NULL. */ /* eassert (node->left == NULL && node->right == NULL @@ -868,7 +860,7 @@ itree_node_set_region (struct itree_tree *tree, static bool interval_tree_contains (struct itree_tree *tree, struct itree_node *node) { - eassert (node); + eassert (iter && node); struct itree_node *other; ITREE_FOREACH (other, tree, node->begin, PTRDIFF_MAX, ASCENDING) if (other == node) @@ -912,7 +904,7 @@ interval_tree_remove_fix (struct itree_tree *tree, if (parent == NULL) eassert (node == tree->root); else - eassert (node == NULL || node->parent == parent); + eassert (node == NULL || node->parent == parent); while (parent != NULL && null_safe_is_black (node)) { @@ -1151,7 +1143,7 @@ itree_iterator_start (struct itree_tree *tree, ptrdiff_t begin, ptrdiff_t end, enum itree_order order, const char *file, int line) { - /* struct itree_iterator *iter = tree->iter; */ + eassert (iter); if (iter->running) { fprintf (stderr, @@ -1179,7 +1171,7 @@ itree_iterator_start (struct itree_tree *tree, ptrdiff_t begin, void itree_iterator_finish (struct itree_iterator *iter) { - eassert (iter->running); + eassert (iter && iter->running); iter->running = false; } @@ -1212,7 +1204,7 @@ itree_insert_gap (struct itree_tree *tree, && (node->begin != node->end || node->rear_advance)) interval_stack_push (saved, node); } - for (int i = 0; i < saved->length; ++i) + for (size_t i = 0; i < saved->length; ++i) itree_remove (tree, nav_nodeptr (saved->nodes[i])); /* We can't use an iterator here, because we can't effectively @@ -1352,7 +1344,7 @@ interval_node_intersects (const struct itree_node *node, struct itree_node * itree_iterator_next (struct itree_iterator *g) { - eassert (g->running); + eassert (g && g->running); struct itree_node *const null = NULL; struct itree_node *node; @@ -1424,9 +1416,9 @@ itree_iterator_next (struct itree_iterator *g) itree_iterator_narrow (struct itree_iterator *g, ptrdiff_t begin, ptrdiff_t end) { - eassert (g->running); + eassert (g && g->running); eassert (begin >= g->begin); eassert (end <= g->end); - g->begin = max (begin, g->begin); - g->end = min (end, g->end); + g->begin = max (begin, g->begin); + g->end = min (end, g->end); } diff --git a/src/itree.h b/src/itree.h index c6b68d3667..49a0333f34 100644 --- a/src/itree.h +++ b/src/itree.h @@ -106,6 +106,7 @@ #define ITREE_H ITREE_PRE_ORDER, }; +extern void init_itree (void); extern void itree_node_init (struct itree_node *, bool, bool, Lisp_Object); extern ptrdiff_t itree_node_begin (struct itree_tree *, struct itree_node *); extern ptrdiff_t itree_node_end (struct itree_tree *, struct itree_node *); @@ -147,7 +148,7 @@ #define ITREE_H BEWARE: - The expression T may be evaluated more than once, so make sure - it is cheap a pure. + it is cheap and pure. - Only a single iteration can happen at a time, so make sure none of the code within the loop can start another tree iteration, i.e. it shouldn't be able to run ELisp code, nor GC since GC can run ELisp by way diff --git a/src/pdumper.c b/src/pdumper.c index d6ae57afb2..0a5d96dbb7 100644 --- a/src/pdumper.c +++ b/src/pdumper.c @@ -2137,8 +2137,8 @@ dump_marker (struct dump_context *ctx, const struct Lisp_Marker *marker) dump_interval_node (struct dump_context *ctx, struct itree_node *node, dump_off parent_offset) { -#if CHECK_STRUCTS && !defined (HASH_interval_node_5765524F7E) -# error "interval_node changed. See CHECK_STRUCTS comment in config.h." +#if CHECK_STRUCTS && !defined (HASH_itree_node_50DE304F13) +# error "itree_node changed. See CHECK_STRUCTS comment in config.h." #endif struct itree_node out; dump_object_start (ctx, &out, sizeof (out)); @@ -2179,7 +2179,7 @@ dump_interval_node (struct dump_context *ctx, struct itree_node *node, static dump_off dump_overlay (struct dump_context *ctx, const struct Lisp_Overlay *overlay) { -#if CHECK_STRUCTS && !defined (HASH_Lisp_Overlay_1CD4249AEC) +#if CHECK_STRUCTS && !defined (HASH_Lisp_Overlay_EB4C05D8D2) # error "Lisp_Overlay changed. See CHECK_STRUCTS comment in config.h." #endif START_DUMP_PVEC (ctx, &overlay->header, struct Lisp_Overlay, out); @@ -2748,7 +2748,7 @@ dump_hash_table (struct dump_context *ctx, static dump_off dump_buffer (struct dump_context *ctx, const struct buffer *in_buffer) { -#if CHECK_STRUCTS && !defined HASH_buffer_F0F08347A5 +#if CHECK_STRUCTS && !defined HASH_buffer_193CAA5E45 # error "buffer changed. See CHECK_STRUCTS comment in config.h." #endif struct buffer munged_buffer = *in_buffer; -- 2.35.1 --=-=-= Content-Type: text/plain >> which calls itree_iterator_create, which xmallocs a new >> itree_iterator, interval_stack, and itree_nodes. Is this memory leak >> going to rust my computer? > > No, Emacs is pure C and ELisp, But not pure enough to avoid side-effects such as headaches. > so it doesn't suffer from Rust. Not in this lifetime anyway ;). Thanks, -- Basil --=-=-=--