From patchwork Wed Sep 20 07:41:09 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Etienne Cordonnier X-Patchwork-Id: 30763 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from aws-us-west-2-korg-lkml-1.web.codeaurora.org (localhost.localdomain [127.0.0.1]) by smtp.lore.kernel.org (Postfix) with ESMTP id 545BECE79AC for ; Wed, 20 Sep 2023 07:41:22 +0000 (UTC) Received: from mail-lj1-f178.google.com (mail-lj1-f178.google.com [209.85.208.178]) by mx.groups.io with SMTP id smtpd.web10.31670.1695195678035418331 for ; Wed, 20 Sep 2023 00:41:18 -0700 Authentication-Results: mx.groups.io; dkim=pass header.i=@snap.com header.s=google header.b=Yr/Mavql; spf=pass (domain: snapchat.com, ip: 209.85.208.178, mailfrom: ecordonnier@snapchat.com) Received: by mail-lj1-f178.google.com with SMTP id 38308e7fff4ca-2c02e232c48so22507651fa.1 for ; Wed, 20 Sep 2023 00:41:17 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=snap.com; s=google; t=1695195676; x=1695800476; darn=lists.openembedded.org; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:from:to:cc:subject:date:message-id:reply-to; bh=nI1wRwLArerz0eE1/8QFwnCKBuZb0fs6yFtQQE9QcTg=; b=Yr/Mavql+kZXtqpqbyoIe9UtF2J0S4XKRPjYkAWoysYv+65QhHBvh+5fGCQjtOPJah ksbYcskpNT1Rb2Eyz+EahTmuDbyFY7eD5z2dwBTAEvNB9RXpJ9YS2rste7H6ea8Gw2m/ 4cj2gi/ZvG0r6n1sS7WSNtHHHRffK4yGBTJBM= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1695195676; x=1695800476; h=content-transfer-encoding:mime-version:message-id:date:subject:cc :to:from:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=nI1wRwLArerz0eE1/8QFwnCKBuZb0fs6yFtQQE9QcTg=; b=OtSld1bOc95KQoWuvApKA2ObFaoIfIKAm3AWZZdSSsJP+ronu5kjj20tXQekeomfcE E3VVvEO49HpdS08prxMTHd+vPWQMVnVZwH7cTpX3ONWsBuOQxmbl5ltX0Ws/ForHyKDk 9dn2Fsb4qvtxmdKabt14Z3Wk6H/48A/2MVq0eJ8+9WaEKOCp3L0UarIeMdLfeFS/xJT7 aKi3LXWZb8BIrQP8r5Yuo9QcX5mGAYYjiXFaICosxjOFsE+5/SvoqDUxVChDtaKyK8RE LhNCoYTGTCxl9dIqiuRD0nMDNLszmxUfL8RHiMK6YHGPelVtrUxNrThDq30K1rzeA+2v lMJw== X-Gm-Message-State: AOJu0YwFQCbd/0eG0R6ByR8lozph9eszKzr7LFNhXNsQdOznPsktY3bs gcuAxR13bqebWosM54nEVl54+e4bUE1uXPuSLe0= X-Google-Smtp-Source: AGHT+IGZHTCBBkaLdwayd3jUPcdGmEIelK9YAaqZ99z4FApMz96jiYRjiVKR3yssirw20YhC5Dk+5w== X-Received: by 2002:a2e:998c:0:b0:2c0:2ef8:9716 with SMTP id w12-20020a2e998c000000b002c02ef89716mr1479983lji.1.1695195675593; Wed, 20 Sep 2023 00:41:15 -0700 (PDT) Received: from lj8k2dq3.sc-core.net ([85.237.126.22]) by smtp.gmail.com with ESMTPSA id p8-20020a1709061b4800b0099c53c44083sm8902610ejg.79.2023.09.20.00.41.14 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 20 Sep 2023 00:41:15 -0700 (PDT) From: ecordonnier@snap.com To: bitbake-devel@lists.openembedded.org Cc: Etienne Cordonnier Subject: [bitbake-devel][PATCH v2] avoid some unnecessary copies Date: Wed, 20 Sep 2023 09:41:09 +0200 Message-Id: <20230920074109.4047252-1-ecordonnier@snap.com> X-Mailer: git-send-email 2.36.1.vfs.0.0 MIME-Version: 1.0 List-Id: X-Webhook-Received: from li982-79.members.linode.com [45.33.32.79] by aws-us-west-2-korg-lkml-1.web.codeaurora.org with HTTPS for ; Wed, 20 Sep 2023 07:41:22 -0000 X-Groupsio-URL: https://lists.openembedded.org/g/bitbake-devel/message/15087 From: Etienne Cordonnier declaring queue=b"" creates an object of types bytes(). bytes() is an immutable object, and therefore doing "self.queue = self.queue + r" creates a new object containing "self.queue" concatenated with "r". On my test setup, we are passing 180MB of data of "workerdata" to the bitbake-worker, so those copies significantly slow down the initialization of the bitbake-worker. Rather use bytearray() which a mutable type, and use extend() to avoid copies. In my test setup, byterray.extend() is 10.000 times faster than copying the queue, for a queue size of 180MB. Signed-off-by: Etienne Cordonnier --- bin/bitbake-worker | 14 +++++++------- lib/bb/runqueue.py | 4 ++-- 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/bin/bitbake-worker b/bin/bitbake-worker index 451e6926..609e276f 100755 --- a/bin/bitbake-worker +++ b/bin/bitbake-worker @@ -91,19 +91,19 @@ def worker_fire_prepickled(event): worker_thread_exit = False def worker_flush(worker_queue): - worker_queue_int = b"" + worker_queue_int = bytearray() global worker_pipe, worker_thread_exit while True: try: - worker_queue_int = worker_queue_int + worker_queue.get(True, 1) + worker_queue_int.extend(worker_queue.get(True, 1)) except queue.Empty: pass while (worker_queue_int or not worker_queue.empty()): try: (_, ready, _) = select.select([], [worker_pipe], [], 1) if not worker_queue.empty(): - worker_queue_int = worker_queue_int + worker_queue.get() + worker_queue_int.extend(worker_queue.get()) written = os.write(worker_pipe, worker_queue_int) worker_queue_int = worker_queue_int[written:] except (IOError, OSError) as e: @@ -346,12 +346,12 @@ class runQueueWorkerPipe(): if pipeout: pipeout.close() bb.utils.nonblockingfd(self.input) - self.queue = b"" + self.queue = bytearray() def read(self): start = len(self.queue) try: - self.queue = self.queue + (self.input.read(102400) or b"") + self.queue.extend(self.input.read(102400) or b"") except (OSError, IOError) as e: if e.errno != errno.EAGAIN: raise @@ -379,7 +379,7 @@ class BitbakeWorker(object): def __init__(self, din): self.input = din bb.utils.nonblockingfd(self.input) - self.queue = b"" + self.queue = bytearray() self.cookercfg = None self.databuilder = None self.data = None @@ -413,7 +413,7 @@ class BitbakeWorker(object): if len(r) == 0: # EOF on pipe, server must have terminated self.sigterm_exception(signal.SIGTERM, None) - self.queue = self.queue + r + self.queue.extend(r) except (OSError, IOError): pass if len(self.queue): diff --git a/lib/bb/runqueue.py b/lib/bb/runqueue.py index c88d7129..fde6e4e9 100644 --- a/lib/bb/runqueue.py +++ b/lib/bb/runqueue.py @@ -3159,7 +3159,7 @@ class runQueuePipe(): if pipeout: pipeout.close() bb.utils.nonblockingfd(self.input) - self.queue = b"" + self.queue = bytearray() self.d = d self.rq = rq self.rqexec = rqexec @@ -3178,7 +3178,7 @@ class runQueuePipe(): start = len(self.queue) try: - self.queue = self.queue + (self.input.read(102400) or b"") + self.queue.extend((self.input.read(102400) or b"")) except (OSError, IOError) as e: if e.errno != errno.EAGAIN: raise