From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-ed1-f42.google.com (mail-ed1-f42.google.com [209.85.208.42]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id C6431367 for ; Thu, 2 Jan 2025 00:07:00 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.208.42 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1735776422; cv=none; b=NR1s81u7dtQjWJeiQE8zUk4BrA/mFUWbvKC3U/CdZs0NwrxTBPNpe4m9opo+nQJihHFwIjerU+7ZiFzYdkdhJSLZpl+4Jfi6CTJYK7ep7m3hKVybC9zgrUp+G5/GNwXtwQu0vUFc+1o4pLggANZyGeCPb5+DiuFF8h6g97cgTZY= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1735776422; c=relaxed/simple; bh=SZff9vj+Bup5V+2kFqBxTidurmz0gHd34ibhHzGpozU=; h=Message-ID:Date:MIME-Version:Subject:To:Cc:References:From: In-Reply-To:Content-Type; b=Esfo9UqYXx5Aw9sXOiEQfbU6xCShgbt+5vXKB0wQQyyA5cblJdKXIGCm056vvTTpc9cC0jDogVGPNEDAbsXQdWAwVWQpVrB+AYesT6xD7x4reH8dwCMVrs+E1VKoaWcehKX+Srm+eyjzgFH2Du45FmlJ0Kh91rii/xbgTCJP1c4= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=W4QAqSWv; arc=none smtp.client-ip=209.85.208.42 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="W4QAqSWv" Received: by mail-ed1-f42.google.com with SMTP id 4fb4d7f45d1cf-5d4e2aa7ea9so19800412a12.2 for ; Wed, 01 Jan 2025 16:07:00 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20230601; t=1735776419; x=1736381219; darn=vger.kernel.org; h=content-transfer-encoding:in-reply-to:autocrypt:from :content-language:references:cc:to:subject:user-agent:mime-version :date:message-id:from:to:cc:subject:date:message-id:reply-to; bh=y713/f0f785zEA94Opwqb6fiMwYNpnqQhjKjw8eunkg=; b=W4QAqSWvIm7VgyL+2aDjdh9vJcXAc2dQP/316M1QpZBZbXFj48ACibUbNzXS0O83KZ /yMNLFDi6tTcrXWBpqMjDy9xTR1BtmdSu/1hlr63tHXQYBNVDOo3nIPLLiIfZMsykUyz u3KBQypBJvS0/INiinpIZo7H2Gd75Vm3sEMumxEMBro83Agb/p9S2q6NwbpoEjyc3ZSX CmX2x36Lu/kZ6zrl4lcIeMzGLf2qtORY4JeIFyg0wot3XbxAEX/G7H7Dub7v8fXBggU+ ceXDJIJI2KoduIxKohaUMxWcmn9zkhGpYUSsupOSegGg9xX7fmdY6gsGP0e0e5mZMrP1 B4BQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1735776419; x=1736381219; h=content-transfer-encoding:in-reply-to:autocrypt:from :content-language:references:cc:to:subject:user-agent:mime-version :date:message-id:x-gm-message-state:from:to:cc:subject:date :message-id:reply-to; bh=y713/f0f785zEA94Opwqb6fiMwYNpnqQhjKjw8eunkg=; b=r4Fp4UuCSo1vn4NY9hb/Xq8ArimkeGCHwyWB5YpPiyCdJ95fBYdW0kcd65OITaMgrf NCcetjpRZD8lnl5NENHHunWEuCH1JURqCz5EdnRUrMQCJlhwPaYZdJsb9NYYkN2oZXqO ks5WFnWi1G/6UA3y9FOo3hu4BAra23k4tKul4t476R8VTh7s4HYo0MbF+hvHwt3LQEnD p7DzaCX7xEKjlxbec7o9cSwSLeGM/T18DAHsH5Z9mVj3vrcwfOhAa36r7BYVjF0Rgxzb y+PGyZ4Q8X7qVtqDnQ/YVoERo4VM+UdkaJLWUlS5NkfIUSRF+5/hwKjr92hWZI1zxToS d+Tg== X-Forwarded-Encrypted: i=1; AJvYcCWiqRzYLcsjI7Mo4FPwv7+8bQS+qGZ8Ym6uHhbKMJg1DBIGdq0YeKBCVGs3dr9cTgD0nLUqzFjLem6MKzY=@vger.kernel.org X-Gm-Message-State: AOJu0YzTvhbg7jtps9V10JPJBDGXddP/gl/9EXm3KppNUrgNOJJ8pO2u /0bA0TfgYwtP8TgHj0Zm+uEh/EFeO6jisADc/8cEE2Kr0fz1CHi0 X-Gm-Gg: ASbGnculf3T20iORGnL0CZgPiROjdO08LGrMtPTCvp7jFmSX1iF/PkO3RbImG4b6gZD dal3TTv+ItqITPzrmzd5mUhLLY0+RpFseUmYwOvBrUxia/OR3jEeu9keFKbJbV1Qrki1upUKI4v KF4TjNVmkPwUSln3EHhXZfgQFN8LWrnNjO+5v9qXt259U+fHNmnmSTDsDYhWERBvKrEB6HaYQs5 xUd9XPbirFjOE/aJu/rF/Z6D64FRA5dhxX0Em5GvYOzn1CeSWQGwGbQ+RglY32mgWjYzItv X-Google-Smtp-Source: AGHT+IFrVlvPbI0GMuQjWuu5OKI1dbZ6JGftY4K+Vml6sfN5jZqaYrvpfSH+kghh3UCCEYi+h8/ynQ== X-Received: by 2002:a17:907:6d0c:b0:aac:212d:a883 with SMTP id a640c23a62f3a-aac34451c4fmr3364567566b.45.1735776418706; Wed, 01 Jan 2025 16:06:58 -0800 (PST) Received: from [132.69.243.32] ([132.69.243.32]) by smtp.gmail.com with ESMTPSA id 4fb4d7f45d1cf-5d80665253dsm17619697a12.0.2025.01.01.16.06.55 (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Wed, 01 Jan 2025 16:06:57 -0800 (PST) Message-ID: <3736e1f1-231c-4060-9fdc-579c11600720@gmail.com> Date: Thu, 2 Jan 2025 02:06:54 +0200 Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 User-Agent: Mozilla Thunderbird Subject: Re: [PATCH 09/12] x86/mm: enable broadcast TLB invalidation for multi-threaded processes To: Rik van Riel , Karim Manaouil Cc: the arch/x86 maintainers , Linux Kernel Mailing List , kernel-team@meta.com, Dave Hansen , luto@kernel.org, peterz@infradead.org, Thomas Gleixner , Ingo Molnar , Borislav Petkov , "H. Peter Anvin" , Andrew Morton , zhengqi.arch@bytedance.com, "open list:MEMORY MANAGEMENT" References: <20241230175550.4046587-1-riel@surriel.com> <20241230175550.4046587-10-riel@surriel.com> <95b614577f5475a919c878d6906d721004c83584.camel@surriel.com> <24c9b4b6-e07b-4cee-aa7f-6f317a1b7ef6@gmail.com> <20250101161556.lniuomixy75vmj5g@ed.ac.uk> <7be55c4eb33aac4f9b99c70bfd4578934fbdd0ea.camel@surriel.com> Content-Language: en-US From: Nadav Amit Autocrypt: addr=nadav.amit@gmail.com; keydata= xsFNBF3nSCQBEACinGNWCLb3Iptvg066zPUQCFfR0nIX+j/pjxlLd094zOKLJ7Z1wKNNTQgb +Y1GQYwKurK1hFHISPc1T7xk2twzov5niFqQp1/E0I6t/92rxo7C26tKMrCHgqXpFXXBaOvq t0QN8B83o3sfS3OZyc0F5hWoJijhKQ+h7dJHMPpBBeTWiZzYVlMNz/wRzYrh1nvesbGJtV1w Wk96qoQAOagczRaSRWouJ1O2JbHqQarLJG657hqtPZctQ1mkXQcJedT1c8ejWBdIiU0rIpV/ R2uTgM5NSjVrY5OxiS2G4C/bDcI4Q6ApWxqu1zRGffKWRFJdQvttSv+JbW2q6OCvJReJs1i6 nTovpfJb3g9d7U76LtAw7HE5iIl5CmyFoTMMH+sDc5PG7JRboiofku9DWeDGK4rB2s97GuA9 BXJDHBbH1AI+Y0H6oz/Dri1sQSayUU8Xj/jCkwxpwDle1qYYE0UM4PqDTUoeejURR/h/AREt bCecfxNUYNz2jY8PUL0YXj6yex2p1otMy5sy7djrTqhC02AXLQ3Oab2UJuMZnf4cTbN2sGXu CRXcCSsoyhxIFtwpgdVIDqkPMCOEPvUwy9hWs0N9Xy+jOebpH6FYxq83pP3IhyBdaoexoyGL IkSzrO4+0bVk5voh5SIYVkLEooMSEQT9rfSdrJmRRuM7S/PowwARAQABzSFOYWRhdiBBbWl0 IDxuYWRhdi5hbWl0QGdtYWlsLmNvbT7CwZcEEwEIAEECGwMFCwkIBwIGFQoJCAsCBBYCAwEC HgECF4ACGQEWIQRIkveiyXlibH/D0jUdoCpJptoPWgUCZWMLXAUJCz4qOAAKCRAdoCpJptoP WoWjD/0d8I+qQIHp2Eq7Whi+idkGDJaaBksRw2r0+XTzRj0CsCFJS7oMWudQaYaI2bTW9uha zfcmpnkm5aOVoxGOGJUJSy0H4of6lShE+7caoz1pdX+y15STij3+RvUjcGn0N+Mpxrnj6anm ArH5KOwzQJDiMF48ntQ2EQAFJ5X6YZ6Ed9N4I88XPeOtWaU21Q5tZI/ASHrQTH7BMukkVe2U PkSjn205vbrCfcC4Ohu9HXC8ey6l5txomFaTIRvaz1La3W8gQHhT2D2QeNnJFRudN5hQJNIE OHFi6QmPVIbI/uYD82h5uesiufycV2kvuIACmvs2T3tpaxSY9HOJF/yAhqg28hKE4wKQIwh2 wzZ+ywDeZ5TboqTjTU9oPHuRTxYItWmevJgQDwOYcssYI4wOdXvu/bh5VvhBgJxgkOOH7j8N efzqooL/2/pD/1JAO9GjPEWbuRx0dZh5tcMq/RYFFhdZQNFYmqdDyY0+y6ntxEo25ScvVLfR GqhpyJkN9+cqJZfg3IolbWoZxkSLtiCBge/EDRRoLJpjpsLpBmMmQcH/50WgQcoL5ObVlc1I ft1sKfzZvIyqXZfu3YcWgvOXfVXaMw1dXQ8ffwgyCSIsKvA5+Y45knd72SHny5z3gi1YsMc1 lm24v+OfxfmZMmJQ7TXBDl0JkOMnQ3A+KDb0oGAI6c7BTQRd50gkARAAt/NqfDSMph828JIn VOck27zsyT12AHRLIhsCCatw+zYbdMSLTW6hd/sChStdndAQlqOOW3mpm6bA+jl59c4OhkBW 8fDKgaWjVq8kBm4Z61zhzk3oXMmxgQ3cqMEfN36PCpcfs5MtLiMoa2JxaqFMQbKkmcZqz8fw qGIID/SMI5qdk+yWZSmR+6RdzXfppEIeqs78qTFxaEl+0MlAcgo2AjrMPLIjNK3k4jNAVCix KjMfkkPk6xVmP5Q7cr1st0xTTa1APt0y966oa6CZIdHY76D/jLpjJlfalctmEAxmyBqK0ZkK WDkEF+qCqqZ7oSkC1Q0ogBuBffIZb5srleqMz++IutiE4r/LELs7RanXXCRK5JYzZc/uAnxN fCJzqKR5/JxoSnCh0yQNU4FHa78DeD/SLnCu8JwTGkcX9ZUW/ZQ+5Nsimy+SlZb0UyaWA+vr Q7+F7OAL6PFsE8v+Q2u2EHwa8e2KmsL0sC+enBXbaogLFFfKi4I13vOk+76nCZTNetRhAubM etLe1PZZ0IDzgGhjFg+cc/jptq3qYjK+3jzhdKaqmoibAoea3E/5HsTVVbxUiYEdk2crTqf9 Mr5S5AzQfH19xU7uUduBvjN3L0dRKP0wJEqgCIB5gZBjD8AYmKSgdEujDAPQNXTJgwB99C+r F0XoHEmO6m9vI6djzDsAEQEAAcLBfAQYAQgAJgIbDBYhBEiS96LJeWJsf8PSNR2gKkmm2g9a BQJlYwtcBQkLPio4AAoJEB2gKkmm2g9ah7wP/jP18+dznONUt3Nga6XKYxxlVmVnMlqHx6SB jDWTwFzcQ6UeVVUmODWBSLI+iqg5djkgsTcHwI668qhIyyWaXC9ld+ONhUaE4NH1PKcJ8cdG vzhdTG8lM9mwmnJriTfih59OLI2yKpml/et44gvsfQpmFOyrdH9u6xPai5h3KyOoXD+Q3EzQ KEiHqs2x0PX72Jz4iVKEanP+qxEwzarUUcum94ecGilYY4jfxhfkgwBWjmC7ExbVTSEKIx3Q zdDbDiAxhAWymSisP3lbYqRo8mkg8r2emNan5L5Xh1b/62zmMO+p6gzucV7vO0VJxlKuUkFW 77Ug+X+hkDedvp2oRBz+wKtBxwSxlTM4toD49Jn111OMS6R5opquZP+/SJH1njSKY9MY3D3D 57QDpRUvqbqMDdqZqmYlMxntxTJ+0yksR61/t62F/bSp4XLxbgwCXMm73TxGrU73ngaS/Nxz Pv28VEcCGXZqFS2sbsbBSEjGgPhXf73YlI1fYiAFiaKHKvlofX+xNLeRHr9/+fL4JWPFurGi WYMyVA5Fv1sfefXhdEv9qNlhyNHxxB3vEpxkEOdo0JUhNfIcxe/w5iOfCUAtwUfvijvHUFHh U5QeQckJZTApjJ4S/Wa1ef8eS/LBsvG9xMNG3Ns+M+xR6unu6gTuzUNp7Zsxgmp8hs0racLB In-Reply-To: <7be55c4eb33aac4f9b99c70bfd4578934fbdd0ea.camel@surriel.com> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit On 01/01/2025 18:23, Rik van Riel wrote: > On Wed, 2025-01-01 at 16:15 +0000, Karim Manaouil wrote: >> On Wed, Jan 01, 2025 at 05:20:01PM +0200, Nadav Amit wrote: >>> >>> >>> On 01/01/2025 6:42, Rik van Riel wrote: >>>> Fixed for the next version. >>> >>> Thanks Rik, >>> >>> Admittedly, I don't feel great about my overall last review - it >>> mostly >>> focused on style and common BKMs. >>> >>> I still don't quite get the entire logic. To name one thing that I >>> don't >>> understand: why do we need broadcast_asid_list and the complicated >>> games of >>> syncing it with broadcast_asid_used. Why wouldn't >>> broadcast_asid_used >>> suffice? >> >> If I uderstand correctly from Rik's patch, I think the list is needed >> to >> save the flush for only when we run out of the ASID space (wrap >> around). >> Without the list, whenever the ASID bit is cleared, you also have to >> flush >> the TLBs. > > That's exactly it. > > The list will only contain processes that are active on > multiple CPUs, and hit a TLB flush "at the right moment" > to be assigned a broadcast ASID, which will be true for > essentially every process that does a lot of TLB flushes > and is long lived. > > However, something like a kernel build has lots of > short lived, single threaded processes, for which we > should not be using broadcast TLB flushing, and which > will not need to remove themselves from the list at > exit time. Thank you Karim and Rik for the patient explanations. But IIUC, it does seem a bit overly complicated (and when I say complicated, I mostly refer to traversing the broadcast_asid_list and its overhead). It seems to me that basically you want to have two pieces of data that can easily be kept in bitmaps. 1. (existing) broadcast_asid_used - those that are currently "busy" and cannot be allocated. 2. (new) broadcast_asid_pending_flush - those that are logically free but were still not flushed. This would allow removing broadcast_asid_list. Then in reset_broadcast_asid_space(), you just do something like: bitmap_andnot(broadcast_asid_used, broadcast_asid_used, broadcast_asid_pending_flush, MAX_ASID_AVAILABLE); bitmap_clear(broadcast_asid_pending_flush, MAX_ASID_AVAILABLE); Seems to me as simpler to understand, faster, and may even allow in the future to avoid locking in fast-paths (would require different ordering and some thought). Of course, I might be missing something...