From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail3.ecloud.global (mail3.ecloud.global [135.181.139.185]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id CE6E22E7196 for ; Mon, 13 Oct 2025 09:33:03 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=pass smtp.client-ip=135.181.139.185 ARC-Seal:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1760347985; cv=pass; b=cPlDwqMWddIpKnOB/Eu0UsNX5g7xxDXmFQ8/xSeZ+Jda4B9oBUjqYm40GqMyj4aSdsHICo9Bi54G7NKB5vqocPxkDd/P/C7guXARHXRhpCAzFiRy8Gq2qn/wmBi9rSoZh6GFJ1utYNzV0oMGiS/y9BHrt43irhsqU3Nfwxk2BPk= ARC-Message-Signature:i=2; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1760347985; c=relaxed/simple; bh=7e4oPSbyT6o1XKtkgDqQchHA30cNRFihW0o+KCSVfM0=; h=MIME-Version:Date:Content-Type:From:Message-ID:Subject:To:Cc; b=FKN5TGM99d2/ZYL+1YCm6gOX80+NxlRNYSVB7mMSJQtcXIT0+UxbBQ4ohwGpx30fHx0LsR2gCKh76AntJoUK2YW6Xi+T2Nznul76f5DguIURlxVnv0KQrPqFoz/5oDk0ykkWfXxy1NNmDWXN7o9gLoXi1LCwN92uf44OwULWaxM= ARC-Authentication-Results:i=2; smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=murena.io; spf=pass smtp.mailfrom=murena.io; dkim=pass (1024-bit key) header.d=murena.io header.i=@murena.io header.b=KWZmnklH; arc=pass smtp.client-ip=135.181.139.185 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=reject dis=none) header.from=murena.io Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=murena.io Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=murena.io header.i=@murena.io header.b="KWZmnklH" Received: from authenticated-user (mail3.ecloud.global [127.0.0.1]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mail3.ecloud.global (Postfix) with ESMTPSA id 3988688A3EC; Mon, 13 Oct 2025 09:22:00 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=murena.io; s=mail3; t=1760347320; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=7e4oPSbyT6o1XKtkgDqQchHA30cNRFihW0o+KCSVfM0=; b=KWZmnklHrgb53JpNCiU46XtaJg4zhgannNk2NRAkhanaIXcUrRdg9OCwx5WFf20Un+jGUn XwOFcrLl1QQF/qN+mwIy58sIwn1dhR8+jFcSVrP6CDRnCtugphqkfaDfNsGXyqJDb8+yKP 0TNR9KUDxj7vKZ3J6jgAsoE0ffjkPA0= ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=murena.io; s=mail3; t=1760347320; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: content-transfer-encoding:content-transfer-encoding; bh=7e4oPSbyT6o1XKtkgDqQchHA30cNRFihW0o+KCSVfM0=; b=BrzcjlUU3iNXAcHY+ElA6LJFefQR08+H1hXRQJRcng3/dP2w6jWjScGktlBbof8nhH5oFC 0s2lgzMiVKywXlXNrUBwHgHnV1aXr1NlW//J1Z1SskROKnaKQSckQitmms4U3GQhD/vrNa vkS+UDF0drWoxNoG8EkZLtWOnWeMgYU= ARC-Authentication-Results: i=1; mail3.ecloud.global; auth=pass smtp.mailfrom=craftfever@murena.io ARC-Seal: i=1; s=mail3; d=murena.io; t=1760347320; a=rsa-sha256; cv=none; b=PwxP0AJIGQ/zfae6OZPyxKRCEAvpNf8EfcnkrLjQIBKFO2tS2RnKuNHcH5drsPOS9Z1VhZ O7NVGyHPg8bPUnXFEb+uvNtlzDvXRUz/0yXFrSTpP5uJFD83/3kfXOcg9kbSZPUHwKwHjL I40Fy8Kl5ivXS/7kQUO8jMXV5qtZAzg= Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Date: Mon, 13 Oct 2025 09:22:00 +0000 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: craftfever@murena.io Message-ID: <020cf8de6e773bb78ba7614ef250129f11a63781@murena.io> TLS-Required: No Subject: [Regerssion] [KSM] KSM CPU overhead in 6.16+ kernel compared to <=6.15 versions ("folio_walk_start" kernel object overhead) To: akpm@linux-foundation.org, david@redhat.com, xu.xin16@zte.com.cn, chengming.zhou@linux.dev Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org, regressions@lists.linux.dev I've posted about that problem already on bigzilla (#220599), but maintai= ners asked to post issues on maillist. The problem with freezes during KSM page scanning with certain processes = like Chromium with huge virtual memory size amount was fized in 6.17.1 co= mpared to 6.16.x/6.17, but problem with huge CPU overhead is present ther= e. Compared to Linux <=3D6.15, where the overhead is much lighter anad th= ere no much CPU consuming during KSM scanning, there is "folio_walk_start= " kernel object is present (which I reviewed with "perf top" command) tha= t is not present in versions <=3D6.15 during KSM work and which is in wor= k starting from Linux 6.16. This method very resource-consuming compared = to algorithm used in <=3D6.15 versions. Is there a kernel parameter to di= sable it or it needs more optimization? I'm using MemoryKSM setting in systemd in user@.service for KSM process m= erging and it very light on <=3D6.15 ver., but CPU consuming on 6.16+ (6.= 17.1 without freezes) due to reasons said above.