From patchwork Mon Aug 14 09:42:57 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Florian Weimer X-Patchwork-Id: 74082 Return-Path: X-Original-To: patchwork@sourceware.org Delivered-To: patchwork@sourceware.org Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id C79093858C3A for ; Mon, 14 Aug 2023 09:43:24 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org C79093858C3A DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1692006204; bh=gNigZGx7A2ft+/kvK9Yj/4FLiUAI3cFH3Zkqp3DrWOs=; h=To:Cc:Subject:Date:List-Id:List-Unsubscribe:List-Archive: List-Post:List-Help:List-Subscribe:From:Reply-To:From; b=M3loS8H+1T1fvekL7/vx0RTsCx7gmmhFRvmPy0hClFYRZrlwOSCLI1jRYBYPhGCiB 0nJh1LfdozwADZveZvo4zlstb7d2OnEa3EXCQPu9clmES3m7Wgk8JAUenSvVHn38dL XBI6yV8nKWnlBHMF64SjVcVMcG4LMdjvsgIsYW80= X-Original-To: libc-alpha@sourceware.org Delivered-To: libc-alpha@sourceware.org Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by sourceware.org (Postfix) with ESMTPS id 42E303858C36 for ; Mon, 14 Aug 2023 09:43:01 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 42E303858C36 Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-531-sRtCrhzCN7WH9JIjDPkKGA-1; Mon, 14 Aug 2023 05:42:59 -0400 X-MC-Unique: sRtCrhzCN7WH9JIjDPkKGA-1 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.rdu2.redhat.com [10.11.54.7]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 1F64C101A52E for ; Mon, 14 Aug 2023 09:42:59 +0000 (UTC) Received: from oldenburg.str.redhat.com (unknown [10.2.16.33]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 81DE6140E950; Mon, 14 Aug 2023 09:42:58 +0000 (UTC) To: libc-alpha@sourceware.org Cc: DJ Delorie Subject: [PATCH] malloc: Remove bin scanning from memalign (bug 30723) Date: Mon, 14 Aug 2023 11:42:57 +0200 Message-ID: <87y1iearj2.fsf@oldenburg.str.redhat.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/28.2 (gnu/linux) MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.1 on 10.11.54.7 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com X-Spam-Status: No, score=-10.2 required=5.0 tests=BAYES_00, DKIMWL_WL_HIGH, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, KAM_ASCII_DIVIDERS, RCVD_IN_DNSWL_NONE, RCVD_IN_MSPIKE_H4, RCVD_IN_MSPIKE_WL, SPF_HELO_NONE, SPF_NONE, TXREP autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org X-BeenThere: libc-alpha@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Libc-alpha mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-Patchwork-Original-From: Florian Weimer via Libc-alpha From: Florian Weimer Reply-To: Florian Weimer Errors-To: libc-alpha-bounces+patchwork=sourceware.org@sourceware.org Sender: "Libc-alpha" On the test workload (mpv --cache=yes with VP9 video decoding), the bin scanning has a very poor success rate (less than 2%). The tcache scanning has about 50% success rate, so keep that. Update comments in malloc/tst-memalign-2 to indicate the purpose of the tests. Even with the scanning removed, the additional merging opportunities since commit 542b1105852568c3ebc712225ae78b ("malloc: Enable merging of remainders in memalign (bug 30723)") are sufficient to pass the existing large bins test. Remove leftover variables from _int_free from refactoring in the same commit. Reviewed-by: DJ Delorie --- v4: Fix -Wall issues. malloc/malloc.c | 169 ++---------------------------------------------- malloc/tst-memalign-2.c | 7 +- 2 files changed, 10 insertions(+), 166 deletions(-) base-commit: 542b1105852568c3ebc712225ae78b8c8ba31a78 diff --git a/malloc/malloc.c b/malloc/malloc.c index 948f9759af..d0bbbf3710 100644 --- a/malloc/malloc.c +++ b/malloc/malloc.c @@ -4488,12 +4488,6 @@ _int_free (mstate av, mchunkptr p, int have_lock) { INTERNAL_SIZE_T size; /* its size */ mfastbinptr *fb; /* associated fastbin */ - mchunkptr nextchunk; /* next contiguous chunk */ - INTERNAL_SIZE_T nextsize; /* its size */ - int nextinuse; /* true if nextchunk is used */ - INTERNAL_SIZE_T prevsize; /* size of previous contiguous chunk */ - mchunkptr bck; /* misc temp for linking */ - mchunkptr fwd; /* misc temp for linking */ size = chunksize (p); @@ -5032,42 +5026,6 @@ _int_realloc (mstate av, mchunkptr oldp, INTERNAL_SIZE_T oldsize, ------------------------------ memalign ------------------------------ */ -/* Returns 0 if the chunk is not and does not contain the requested - aligned sub-chunk, else returns the amount of "waste" from - trimming. NB is the *chunk* byte size, not the user byte - size. */ -static size_t -chunk_ok_for_memalign (mchunkptr p, size_t alignment, size_t nb) -{ - void *m = chunk2mem (p); - INTERNAL_SIZE_T size = chunksize (p); - void *aligned_m = m; - - if (__glibc_unlikely (misaligned_chunk (p))) - malloc_printerr ("_int_memalign(): unaligned chunk detected"); - - aligned_m = PTR_ALIGN_UP (m, alignment); - - INTERNAL_SIZE_T front_extra = (intptr_t) aligned_m - (intptr_t) m; - - /* We can't trim off the front as it's too small. */ - if (front_extra > 0 && front_extra < MINSIZE) - return 0; - - /* If it's a perfect fit, it's an exception to the return value rule - (we would return zero waste, which looks like "not usable"), so - handle it here by returning a small non-zero value instead. */ - if (size == nb && front_extra == 0) - return 1; - - /* If the block we need fits in the chunk, calculate total waste. */ - if (size > nb + front_extra) - return size - nb; - - /* Can't use this chunk. */ - return 0; -} - /* BYTES is user requested bytes, not requested chunksize bytes. */ static void * _int_memalign (mstate av, size_t alignment, size_t bytes) @@ -5082,7 +5040,6 @@ _int_memalign (mstate av, size_t alignment, size_t bytes) mchunkptr remainder; /* spare room at end to split off */ unsigned long remainder_size; /* its size */ INTERNAL_SIZE_T size; - mchunkptr victim; nb = checked_request2size (bytes); if (nb == 0) @@ -5101,129 +5058,13 @@ _int_memalign (mstate av, size_t alignment, size_t bytes) we don't find anything in those bins, the common malloc code will scan starting at 2x. */ - /* This will be set if we found a candidate chunk. */ - victim = NULL; + /* Call malloc with worst case padding to hit alignment. */ + m = (char *) (_int_malloc (av, nb + alignment + MINSIZE)); - /* Fast bins are singly-linked, hard to remove a chunk from the middle - and unlikely to meet our alignment requirements. We have not done - any experimentation with searching for aligned fastbins. */ + if (m == 0) + return 0; /* propagate failure */ - if (av != NULL) - { - int first_bin_index; - int first_largebin_index; - int last_bin_index; - - if (in_smallbin_range (nb)) - first_bin_index = smallbin_index (nb); - else - first_bin_index = largebin_index (nb); - - if (in_smallbin_range (nb * 2)) - last_bin_index = smallbin_index (nb * 2); - else - last_bin_index = largebin_index (nb * 2); - - first_largebin_index = largebin_index (MIN_LARGE_SIZE); - - int victim_index; /* its bin index */ - - for (victim_index = first_bin_index; - victim_index < last_bin_index; - victim_index ++) - { - victim = NULL; - - if (victim_index < first_largebin_index) - { - /* Check small bins. Small bin chunks are doubly-linked despite - being the same size. */ - - mchunkptr fwd; /* misc temp for linking */ - mchunkptr bck; /* misc temp for linking */ - - bck = bin_at (av, victim_index); - fwd = bck->fd; - while (fwd != bck) - { - if (chunk_ok_for_memalign (fwd, alignment, nb) > 0) - { - victim = fwd; - - /* Unlink it */ - victim->fd->bk = victim->bk; - victim->bk->fd = victim->fd; - break; - } - - fwd = fwd->fd; - } - } - else - { - /* Check large bins. */ - mchunkptr fwd; /* misc temp for linking */ - mchunkptr bck; /* misc temp for linking */ - mchunkptr best = NULL; - size_t best_size = 0; - - bck = bin_at (av, victim_index); - fwd = bck->fd; - - while (fwd != bck) - { - int extra; - - if (chunksize (fwd) < nb) - break; - extra = chunk_ok_for_memalign (fwd, alignment, nb); - if (extra > 0 - && (extra <= best_size || best == NULL)) - { - best = fwd; - best_size = extra; - } - - fwd = fwd->fd; - } - victim = best; - - if (victim != NULL) - { - unlink_chunk (av, victim); - break; - } - } - - if (victim != NULL) - break; - } - } - - /* Strategy: find a spot within that chunk that meets the alignment - request, and then possibly free the leading and trailing space. - This strategy is incredibly costly and can lead to external - fragmentation if header and footer chunks are unused. */ - - if (victim != NULL) - { - p = victim; - m = chunk2mem (p); - set_inuse (p); - if (av != &main_arena) - set_non_main_arena (p); - } - else - { - /* Call malloc with worst case padding to hit alignment. */ - - m = (char *) (_int_malloc (av, nb + alignment + MINSIZE)); - - if (m == 0) - return 0; /* propagate failure */ - - p = mem2chunk (m); - } + p = mem2chunk (m); if ((((unsigned long) (m)) % alignment) != 0) /* misaligned */ { diff --git a/malloc/tst-memalign-2.c b/malloc/tst-memalign-2.c index f229283dbf..ecd6fa249e 100644 --- a/malloc/tst-memalign-2.c +++ b/malloc/tst-memalign-2.c @@ -86,7 +86,8 @@ do_test (void) TEST_VERIFY (tcache_allocs[i].ptr1 == tcache_allocs[i].ptr2); } - /* Test for non-head tcache hits. */ + /* Test for non-head tcache hits. This exercises the memalign + scanning code to find matching allocations. */ for (i = 0; i < array_length (ptr); ++ i) { if (i == 4) @@ -113,7 +114,9 @@ do_test (void) free (p); TEST_VERIFY (count > 0); - /* Large bins test. */ + /* Large bins test. This verifies that the over-allocated parts + that memalign releases for future allocations can be reused by + memalign itself at least in some cases. */ for (i = 0; i < LN; ++ i) {