From patchwork Sat Oct 26 22:08:26 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jens Axboe X-Patchwork-Id: 13852378 Received: from mail-pl1-f173.google.com (mail-pl1-f173.google.com [209.85.214.173]) (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 C632A42AB1 for ; Sat, 26 Oct 2024 22:23:56 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.173 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981440; cv=none; b=BOvFLKh4DFDftlqKZOfai8GufAQSUqchXdq2TZ/hnrVEMtY4/eBqvhUtw+7RHC9pQOeYCLa9oHbY5uOW0iykyfGMdE5rdUSINHf1OuMJE/sgPLPqIcVvylS/Tk/vSHY/TggMU8Om/or/dIb4xC5bqe+Inkib+915cPV6Tp6pgaE= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981440; c=relaxed/simple; bh=RrDXVmFVz1DJrLc0BSNRTXBdkG6HbImKxPgyJQkLKTQ=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=tEV1VyhYUEU3/gYFfrG0Be1XBfZdyxtfxNjAxhH8Nlgo2+8iQFUVhYFJtTQlYxoQAJTb+k/1C6nSLwqxx08kK/3qBvWqhTHn0jeRXestdB6E6IKGyw8HCHDI/vNNYGjDRD9N1/5osENTDjiQ/7JMh+xEI0c411jZe2iDJSz8Q98= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk; spf=pass smtp.mailfrom=kernel.dk; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b=vCBDDfTP; arc=none smtp.client-ip=209.85.214.173 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=kernel.dk Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b="vCBDDfTP" Received: by mail-pl1-f173.google.com with SMTP id d9443c01a7336-20c693b68f5so32310915ad.1 for ; Sat, 26 Oct 2024 15:23:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20230601.gappssmtp.com; s=20230601; t=1729981435; x=1730586235; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=RwJIDvb53GRUtCJpMgJ66hIkVw8Alhl/3YCSjG18yfU=; b=vCBDDfTP4ysshkAbKQaaBFtt79fTHxTELlex1+A7jW0E9PKArNHt9UrhL67s1bSQjZ 54av+8kXj4KWp6m0dMak23b7CFMywJsAvcDT70hynoVhf0mctBw/g3feer5moKHjKl7n eX2ARPfs/5iHq0M8ZPciDqigpjCOddroPEamk5Y2J9zW+CUceRowWhQSmDARtcc4STvQ jba3Mklp/cn354wATYiEDRWnnGxKQPHPiG9lchgc78eQib4wnKzILCQlCG/+arVojZBk FuGfFvWbDyvb1NemYfYUa/vzAS2HKXmuw/BN2SxL0uTiQprAItNYIF1NMJTM0e4fbsJQ 2VOQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1729981435; x=1730586235; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=RwJIDvb53GRUtCJpMgJ66hIkVw8Alhl/3YCSjG18yfU=; b=di/mkN/48uWVJTpG763Dt9XaxCPnZuoVcRX5Pv14kJdM2iR0WeSy3M3MV1gYIx3N2N 2tEAR863/3EME1vLJP25L/INWiBCwHkTr75wWJtidUqhF8MTEUqheWPcPoJ2QftLFmuw onRhvZxiTAj7WhJqFP8SHX/mmQDAIeyzVNJm14k8WoXSHmls/ZBvQdBKR13cujV66x+k +ysGhAl+jtk3tRJyWj94LXXJvUqyaAjzLFaSF5vmnzgulxHCy28ghuc28chHpHXFIhrn z2A+lps/WxTm7TfhoPPDz3g4kVdBjmuqgbt6eXXDghG1IiLmNgdA8lwLi3gOo6xSBb74 BykQ== X-Gm-Message-State: AOJu0YxkywWn8UggYX1VWNAP3g13MukjGR3j/2BdMtDcRN888QzPaYkl BW3V2cEJOuKRao+mIUr9FKmE8LlAqIyXoXmEdMjIokbLHEMQ9nnS+TvVdr+1BB1TVFHwvieGBe6 u X-Google-Smtp-Source: AGHT+IFxKLQUiTqYGxd/Zqgdn4dbMwlRysDo6A1e8VwQLViXxmX3CB1yMn8I6wTrffQLCeOgn+jvCw== X-Received: by 2002:a17:902:ccce:b0:20c:f9ec:cd9e with SMTP id d9443c01a7336-210c6c6ef7fmr52715115ad.41.1729981435293; Sat, 26 Oct 2024 15:23:55 -0700 (PDT) Received: from localhost.localdomain ([198.8.77.157]) by smtp.gmail.com with ESMTPSA id d9443c01a7336-210bbf44321sm28134705ad.30.2024.10.26.15.23.53 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 26 Oct 2024 15:23:53 -0700 (PDT) From: Jens Axboe To: io-uring@vger.kernel.org Cc: Jens Axboe Subject: [PATCH 1/7] io_uring/rsrc: move strct io_fixed_file to rsrc.h header Date: Sat, 26 Oct 2024 16:08:26 -0600 Message-ID: <20241026222348.90331-2-axboe@kernel.dk> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20241026222348.90331-1-axboe@kernel.dk> References: <20241026222348.90331-1-axboe@kernel.dk> Precedence: bulk X-Mailing-List: io-uring@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 There's no need for this internal structure to be visible, move it to the private rsrc.h header instead. Signed-off-by: Jens Axboe --- include/linux/io_uring_types.h | 5 ----- io_uring/filetable.h | 1 + io_uring/rsrc.h | 5 +++++ 3 files changed, 6 insertions(+), 5 deletions(-) diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index 841579dcdae9..b61db1e8b639 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -55,11 +55,6 @@ struct io_wq_work { int cancel_seq; }; -struct io_fixed_file { - /* file * with additional FFS_* flags */ - unsigned long file_ptr; -}; - struct io_file_table { struct io_fixed_file *files; unsigned long *bitmap; diff --git a/io_uring/filetable.h b/io_uring/filetable.h index b2435c4dca1f..c027ed4ad68d 100644 --- a/io_uring/filetable.h +++ b/io_uring/filetable.h @@ -4,6 +4,7 @@ #include #include +#include "rsrc.h" bool io_alloc_file_tables(struct io_file_table *table, unsigned nr_files); void io_free_file_tables(struct io_file_table *table); diff --git a/io_uring/rsrc.h b/io_uring/rsrc.h index c50d4be4aa6d..e072fb3ee351 100644 --- a/io_uring/rsrc.h +++ b/io_uring/rsrc.h @@ -40,6 +40,11 @@ struct io_rsrc_node { struct io_rsrc_put item; }; +struct io_fixed_file { + /* file * with additional FFS_* flags */ + unsigned long file_ptr; +}; + struct io_mapped_ubuf { u64 ubuf; unsigned int len; From patchwork Sat Oct 26 22:08:27 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jens Axboe X-Patchwork-Id: 13852379 Received: from mail-pl1-f180.google.com (mail-pl1-f180.google.com [209.85.214.180]) (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 1657D1D461B for ; Sat, 26 Oct 2024 22:23:57 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.180 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981440; cv=none; b=FZopOLoYdih7xs6xSYOpMAuWUvzMQKQGLDpNALGy2Hl91pl/pNya4zjrntQrXSP/QvO3hS59KxPr1wdJrhpYZpQ2ukMizw/urAWmp6pJ0pJzEvY0WMRVdJOz1BSe0P/g4EgBCzn8R3RpNryT/yTqri2bPY8dVf+4fq7dwvAYzQQ= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981440; c=relaxed/simple; bh=2TXmyCtwXZp1vsGW+1a4yPJ5ITbgbVvr2+o2HfpIP0k=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=ltzz8h777qc2eyAGE3bHd8iV7cxM6AzRIiVmMhhnIkUD+JA9UW2OR8u45512uVqImJEbNbHhtXGmdTpvzxzADXh3xhOWcmPcXIymYpGhl1evSBHs9clBvmvMadquuhpBgvLaaL5KeeMfrIEYvt//bhHtR6+RrIQ68UyZcIMuOZY= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk; spf=pass smtp.mailfrom=kernel.dk; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b=cSKm4qD4; arc=none smtp.client-ip=209.85.214.180 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=kernel.dk Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b="cSKm4qD4" Received: by mail-pl1-f180.google.com with SMTP id d9443c01a7336-20cd76c513cso28788815ad.3 for ; Sat, 26 Oct 2024 15:23:57 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20230601.gappssmtp.com; s=20230601; t=1729981437; x=1730586237; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=TVO7eeCq6r97bEPVJLNy2xG86WtG8C+xSLbWchjxhk4=; b=cSKm4qD4BSq5qkleuEhSZ3QdE1p++hc9ADwhgUyKoSIyvuunGdxfG7Ma6KrZ3aLY4R ze9J11JJ7jCjVjcWg7TZtlrav4BxJ+BcdbWCSIzjO366B/wVgbKeORB6jHiZQ/J5PsON +79zlBfWHrNJOb/H2sfhtcIlABsoCnNL3TbHccm3O49oP5shBL1sIig+xB2eZNr3iCm9 hDT67hFcSutMygp8AOk6e1+MWXIV43W6J6B+6aX4XLsv2+YXz12/z5UVuWlM+1BKYtBd aP80YDL/FNrhtOgaWhvo8fakcctvJRCWtiVxh0ld2knZ9e7/jFQfzAsOhs08SMmPIFcz gflA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1729981437; x=1730586237; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=TVO7eeCq6r97bEPVJLNy2xG86WtG8C+xSLbWchjxhk4=; b=Az47+EjtSKoc27ZIYPZWadVYoIHSFc2t+p46EMSKxcghMFNRoD0UU35FdPdCe3TBEA ek7ciaMaliv68gABIyu90FoBWdTPzDzPCHPKLzzpeAaGuMpfJ6EFyC0Kjf6UqKvRaJe9 ZVprfHlTFDm++V+z0q5Kq8BNYODGDXN803VOSQK2bh4kkDMbDU5n7FHB0HxYlxxp2tiu PeuxXqUaB4iSkWl13GoKCeHbpnozskwAYWRAAwYI+fl2j0wnMi340+goxpcZbGtEIFwj LD+xRIu757Dpvn9ggtlt2AVV+9hQnxyVgKlj3oi4Jhd/2ynt3LuxTCJdVRz9RUOtQNhR /7sg== X-Gm-Message-State: AOJu0YzwOjtrVM1WmWVp+Px0WHVtw7gzLvLRjjyr5IbDRs/qvLaXDTT5 RG7e5tII/9z8zYQfIK9iKQZSrL5pnqs39NFmUqyzs2ENr12xdr5A5oN0GKknUhppAdbeDNehBqT x X-Google-Smtp-Source: AGHT+IF8YzUOoSl1dLSZAP29PU0wDGubzK4rhPHg6f5OkBWC1if7nAogYCjfTd8KJPSjZRziTQg3nA== X-Received: by 2002:a17:903:32c8:b0:20b:6a57:bf3a with SMTP id d9443c01a7336-210c686d4a6mr46496445ad.1.1729981436801; Sat, 26 Oct 2024 15:23:56 -0700 (PDT) Received: from localhost.localdomain ([198.8.77.157]) by smtp.gmail.com with ESMTPSA id d9443c01a7336-210bbf44321sm28134705ad.30.2024.10.26.15.23.55 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 26 Oct 2024 15:23:56 -0700 (PDT) From: Jens Axboe To: io-uring@vger.kernel.org Cc: Jens Axboe Subject: [PATCH 2/7] io_uring: specify freeptr usage for SLAB_TYPESAFE_BY_RCU io_kiocb cache Date: Sat, 26 Oct 2024 16:08:27 -0600 Message-ID: <20241026222348.90331-3-axboe@kernel.dk> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20241026222348.90331-1-axboe@kernel.dk> References: <20241026222348.90331-1-axboe@kernel.dk> Precedence: bulk X-Mailing-List: io-uring@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Doesn't matter right now as there's still some bytes left for it, but let's prepare for the io_kiocb potentially growing and add a specific freeptr offset for it. Signed-off-by: Jens Axboe --- io_uring/io_uring.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 140cd47fbdb3..187d4a6b8337 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -3785,6 +3785,8 @@ static int __init io_uring_init(void) struct kmem_cache_args kmem_args = { .useroffset = offsetof(struct io_kiocb, cmd.data), .usersize = sizeof_field(struct io_kiocb, cmd.data), + .freeptr_offset = offsetof(struct io_kiocb, work), + .use_freeptr_offset = true, }; #define __BUILD_BUG_VERIFY_OFFSET_SIZE(stype, eoffset, esize, ename) do { \ From patchwork Sat Oct 26 22:08:28 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jens Axboe X-Patchwork-Id: 13852381 Received: from mail-pl1-f180.google.com (mail-pl1-f180.google.com [209.85.214.180]) (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 7A79742AB1 for ; Sat, 26 Oct 2024 22:24:00 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.180 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981445; cv=none; b=a24hVKK7h2huc9o0lhiVQ08X5ft0XcKDPQn4vw8HyLjrf67lSIm13Bn2yPZ9KyswExgkCYT2a85/XFBTVyMAj0+53t9U96rTL+cFercwN5pFO+7MmsKlIFTkMfYEUoJw7oknPh5ZVYB9X+CuYBSbeMQ8E1KV7T0eQCXgJHw0VoM= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981445; c=relaxed/simple; bh=AE2Zcm2K0tljvidbtikWvi4b8YmCGwiXIwWujHRZTPU=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=XJ5ZRuLsiO0sp8XVGmUpwQk0XzWt2CzVdrLVXoiBY/61IIpJ8nWSmy/oDICwlsJV2UeRjpkFkFHbs2YOUgnEm5rUtM3CVMh49t0Tgqbm462+G18uA7xhF2aIVMtTuuD+7P0LvBz6OuXU/Jn1U5IDttIVzE2hIiBJl/0qSNGtQkw= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk; spf=pass smtp.mailfrom=kernel.dk; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b=sH8jRXG3; arc=none smtp.client-ip=209.85.214.180 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=kernel.dk Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b="sH8jRXG3" Received: by mail-pl1-f180.google.com with SMTP id d9443c01a7336-20c803787abso24326965ad.0 for ; Sat, 26 Oct 2024 15:24:00 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20230601.gappssmtp.com; s=20230601; t=1729981439; x=1730586239; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=pq7TNonOdyzWJ/URVlCrhpuSnW0qT1toJkxWQDgHlg4=; b=sH8jRXG3j7egHzw6KYvE9yCwyCqkbb/WVzNWIaC3MBXP8FjA4zy+FxRie2ObNx79iA e6uyl7bmTIMqToS64RQ4/+oWXZjr0rMDRK10iVxbFzuIMpPbxPPI5NP+xiCbcJi874SH x42dp8lYhBmuidgy4PV2z+R+kPOKptVyz3dAdATQbtFY6U6YC4zfZQoa/BBcSK2z6/b8 q0Gj868KevZfqFWYCuUKhmBfOyUy4KYpKM+yYS9NJUwOXw8iGnjkB+hy+4pyOs/hEjxY OinOnSHbOx09ReCNyy9Y4R3N0Gxk4rF4SFr4RpSUxICBlCsGTtkU0bvvboV4p1Su81tV FC9Q== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1729981439; x=1730586239; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=pq7TNonOdyzWJ/URVlCrhpuSnW0qT1toJkxWQDgHlg4=; b=wjWit/ye9qDR19QLLlengo1kIyrasVh3YZ2Iyaq7PrM16B/SBfYn8c+KlYhGih1fxI WSXicmv2JoATUWen1FpHTEZxz0lZlD5iCJi1wLX6EBEvoGgZbVoopjboB5dzc3xUPGyh tEPAK3GgJVHQQyvGZXKwCj3I9l9rvY/SxCCTRKp1Zplc4fAHYpgXmbu6/fBFyJA3W4L1 bXt3x7b6ojS26gK0Zz1IXvipvYagKLGjtviWkS/81gck5scDpleLl4oMAR9mZnPvbEhF i7SCWAkUFBE6T3sF2HmRpDBIXL/CpM3RKu8ijMERSyReDqFzd866RsMkzQg7so+jFwoC FaVw== X-Gm-Message-State: AOJu0YwtczYtu6ULNg6ZnFvJYzVdXeWoyMm12JNtTKIlgkdxyviQLPFG BOI6BKyZPkPAZvZ4Ahp2WJHl34LbSUg0Pl0XFZlJeP4Wv4CiMNPPQEetecdlup21mZCt8+w4LBF w X-Google-Smtp-Source: AGHT+IHUm7oAFghSXjZBPFjzM2V5L4n0KlTxYOTlo5gVx8htanTVaDXvfXFSkJ9iqBMhBQTTz1h1zA== X-Received: by 2002:a17:903:2307:b0:20b:723a:cba1 with SMTP id d9443c01a7336-20fb884fc0dmr160597795ad.1.1729981438692; Sat, 26 Oct 2024 15:23:58 -0700 (PDT) Received: from localhost.localdomain ([198.8.77.157]) by smtp.gmail.com with ESMTPSA id d9443c01a7336-210bbf44321sm28134705ad.30.2024.10.26.15.23.56 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 26 Oct 2024 15:23:57 -0700 (PDT) From: Jens Axboe To: io-uring@vger.kernel.org Cc: Jens Axboe Subject: [PATCH 3/7] io_uring/rsrc: get rid of per-ring io_rsrc_node list Date: Sat, 26 Oct 2024 16:08:28 -0600 Message-ID: <20241026222348.90331-4-axboe@kernel.dk> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20241026222348.90331-1-axboe@kernel.dk> References: <20241026222348.90331-1-axboe@kernel.dk> Precedence: bulk X-Mailing-List: io-uring@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Work in progress, but get rid of the per-ring serialization of resource nodes, like registered buffers and files. Main issue here is that one node can otherwise hold up a bunch of other nodes from getting freed, which is especially a problem for file resource nodes and networked workloads where some descriptors may not see activity in a long time. As an example, instantiate an io_uring ring fd and create a sparse registered file table. Even 2 will do. Then create a socket and register it as fixed file 0, F0. The number of open files in the app is now 5, with 0/1/2 being the usual stdin/out/err, 3 being the ring fd, and 4 being the socket. Register this socket (eg "the listener") in slot 0 of the registered file table. Now add an operation on the socket that uses slot 0. Finally, loop N times, where each loop creates a new socket, registers said socket as a file, then unregisters the socket, and finally closes the socket. This is roughly similar to what a basic accept loop would look like. At the end of this loop, it's not unreasonable to expect that there would still be 5 open files. Each socket created and registered in the loop is also unregistered and closed. But since the listener socket registered first still has references to its resource node due to still being active, each subsequent socket unregistration is stuck behind it for reclaim. Hence 5 + N files are still open at that point, where N is awaiting the final put held up by the listener socket. Rewrite the io_rsrc_node handling to NOT rely on serialization. Struct io_kiocb now gets explicit resource nodes assigned, with each holding a reference to the parent node. A parent node is either of type FILE or BUFFER, which are the two types of nodes that exist. A request can have two nodes assigned, if it's using both registered files and buffers. Since request issue and task_work completion is both under the ring private lock, no atomics are needed to handle these references. It's a simple unlocked inc/dec. As before, the registered buffer or file table each hold a reference as well to the registered nodes. Final put of the node will remove the node and free the underlying resource, eg unmap the buffer or put the file. Outside of removing the stall in resource reclaim described above, it has the following advantages: 1) It's a lot simpler than the previous scheme, and easier to follow. No need to specific quiesce handling anymore. 2) There are no resource node allocations in the fast path, all of that happens at resource registration time. 3) The structs related to resource handling can all get simplified quite a bit, like io_rsrc_node and io_rsrc_data. io_rsrc_put can go away completely. 4) Handling of resource tags is much simpler, and doesn't require persistent storage as it can simply get assigned up front at registration time. Just copy them in one-by-one at registration time and assign to the resource node. The only real downside is that a request is now explicitly limited to pinning 2 resources, one file and one buffer, where before just assigning a resource node to a request would pin all of them. The upside is that it's easier to follow now, as an individual resource is explicitly referenced and assigned to the request. With this in place, the above mentioned example will be using exactly 5 files at the end of the loop, not N. Needs to get broken up a bit and there are certainly rough edges, but that's why it's a work in progress... But it does remove a ton more code than it adds, and passes the liburing tests. Signed-off-by: Jens Axboe --- include/linux/io_uring_types.h | 10 +- io_uring/fdinfo.c | 2 +- io_uring/filetable.c | 52 ++-- io_uring/filetable.h | 25 +- io_uring/io_uring.c | 38 +-- io_uring/net.c | 11 +- io_uring/notif.c | 3 +- io_uring/rsrc.c | 490 +++++++++++++-------------------- io_uring/rsrc.h | 82 ++---- io_uring/rw.c | 8 +- io_uring/uring_cmd.c | 12 +- 11 files changed, 271 insertions(+), 462 deletions(-) diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index b61db1e8b639..93111d87a88a 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -56,7 +56,7 @@ struct io_wq_work { }; struct io_file_table { - struct io_fixed_file *files; + struct io_rsrc_node **nodes; unsigned long *bitmap; unsigned int alloc_hint; }; @@ -264,7 +264,6 @@ struct io_ring_ctx { * Fixed resources fast path, should be accessed only under * uring_lock, and updated through io_uring_register(2) */ - struct io_rsrc_node *rsrc_node; atomic_t cancel_seq; /* @@ -277,7 +276,7 @@ struct io_ring_ctx { struct io_wq_work_list iopoll_list; struct io_file_table file_table; - struct io_mapped_ubuf **user_bufs; + struct io_rsrc_node **user_bufs; unsigned nr_user_files; unsigned nr_user_bufs; @@ -364,10 +363,7 @@ struct io_ring_ctx { struct io_rsrc_data *buf_data; /* protected by ->uring_lock */ - struct list_head rsrc_ref_list; struct io_alloc_cache rsrc_node_cache; - struct wait_queue_head rsrc_quiesce_wq; - unsigned rsrc_quiesce; u32 pers_next; struct xarray personalities; @@ -632,7 +628,7 @@ struct io_kiocb { __poll_t apoll_events; }; - struct io_rsrc_node *rsrc_node; + struct io_rsrc_node *rsrc_nodes[2]; atomic_t refs; bool cancel_seq_set; diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c index a6bac533edbe..064a79475c5f 100644 --- a/io_uring/fdinfo.c +++ b/io_uring/fdinfo.c @@ -176,7 +176,7 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) } seq_printf(m, "UserBufs:\t%u\n", ctx->nr_user_bufs); for (i = 0; has_lock && i < ctx->nr_user_bufs; i++) { - struct io_mapped_ubuf *buf = ctx->user_bufs[i]; + struct io_mapped_ubuf *buf = ctx->user_bufs[i]->buf; seq_printf(m, "%5u: 0x%llx/%u\n", i, buf->ubuf, buf->len); } diff --git a/io_uring/filetable.c b/io_uring/filetable.c index 997c56d32ee6..c18e37b495a1 100644 --- a/io_uring/filetable.c +++ b/io_uring/filetable.c @@ -38,14 +38,14 @@ static int io_file_bitmap_get(struct io_ring_ctx *ctx) bool io_alloc_file_tables(struct io_file_table *table, unsigned nr_files) { - table->files = kvcalloc(nr_files, sizeof(table->files[0]), - GFP_KERNEL_ACCOUNT); - if (unlikely(!table->files)) + table->nodes = kvmalloc_array(nr_files, sizeof(struct io_src_node *), + GFP_KERNEL_ACCOUNT | __GFP_ZERO); + if (unlikely(!table->nodes)) return false; table->bitmap = bitmap_zalloc(nr_files, GFP_KERNEL_ACCOUNT); if (unlikely(!table->bitmap)) { - kvfree(table->files); + kvfree(table->nodes); return false; } @@ -54,9 +54,9 @@ bool io_alloc_file_tables(struct io_file_table *table, unsigned nr_files) void io_free_file_tables(struct io_file_table *table) { - kvfree(table->files); + kvfree(table->nodes); bitmap_free(table->bitmap); - table->files = NULL; + table->nodes = NULL; table->bitmap = NULL; } @@ -64,8 +64,7 @@ static int io_install_fixed_file(struct io_ring_ctx *ctx, struct file *file, u32 slot_index) __must_hold(&req->ctx->uring_lock) { - struct io_fixed_file *file_slot; - int ret; + struct io_rsrc_node *node; if (io_is_uring_fops(file)) return -EBADF; @@ -74,22 +73,18 @@ static int io_install_fixed_file(struct io_ring_ctx *ctx, struct file *file, if (slot_index >= ctx->nr_user_files) return -EINVAL; - slot_index = array_index_nospec(slot_index, ctx->nr_user_files); - file_slot = io_fixed_file_slot(&ctx->file_table, slot_index); - - if (file_slot->file_ptr) { - ret = io_queue_rsrc_removal(ctx->file_data, slot_index, - io_slot_file(file_slot)); - if (ret) - return ret; + node = io_rsrc_node_alloc(ctx, ctx->file_data, slot_index, IORING_RSRC_FILE); + if (IS_ERR(node)) + return -ENOMEM; - file_slot->file_ptr = 0; - } else { + slot_index = array_index_nospec(slot_index, ctx->nr_user_files); + if (ctx->file_table.nodes[slot_index]) + io_put_rsrc_node(ctx->file_table.nodes[slot_index]); + else io_file_bitmap_set(&ctx->file_table, slot_index); - } - *io_get_tag_slot(ctx->file_data, slot_index) = 0; - io_fixed_file_set(file_slot, file); + ctx->file_table.nodes[slot_index] = node; + io_fixed_file_set(node, file); return 0; } @@ -134,25 +129,16 @@ int io_fixed_fd_install(struct io_kiocb *req, unsigned int issue_flags, int io_fixed_fd_remove(struct io_ring_ctx *ctx, unsigned int offset) { - struct io_fixed_file *file_slot; - int ret; - if (unlikely(!ctx->file_data)) return -ENXIO; if (offset >= ctx->nr_user_files) return -EINVAL; offset = array_index_nospec(offset, ctx->nr_user_files); - file_slot = io_fixed_file_slot(&ctx->file_table, offset); - if (!file_slot->file_ptr) + if (!ctx->file_table.nodes[offset]) return -EBADF; - - ret = io_queue_rsrc_removal(ctx->file_data, offset, - io_slot_file(file_slot)); - if (ret) - return ret; - - file_slot->file_ptr = 0; + io_put_rsrc_node(ctx->file_table.nodes[offset]); + ctx->file_table.nodes[offset] = NULL; io_file_bitmap_clear(&ctx->file_table, offset); return 0; } diff --git a/io_uring/filetable.h b/io_uring/filetable.h index c027ed4ad68d..47616079abaa 100644 --- a/io_uring/filetable.h +++ b/io_uring/filetable.h @@ -34,36 +34,35 @@ static inline void io_file_bitmap_set(struct io_file_table *table, int bit) table->alloc_hint = bit + 1; } -static inline struct io_fixed_file * -io_fixed_file_slot(struct io_file_table *table, unsigned i) -{ - return &table->files[i]; -} - #define FFS_NOWAIT 0x1UL #define FFS_ISREG 0x2UL #define FFS_MASK ~(FFS_NOWAIT|FFS_ISREG) -static inline unsigned int io_slot_flags(struct io_fixed_file *slot) +static inline unsigned int io_slot_flags(struct io_rsrc_node *node) { - return (slot->file_ptr & ~FFS_MASK) << REQ_F_SUPPORT_NOWAIT_BIT; + + return (node->file_ptr & ~FFS_MASK) << REQ_F_SUPPORT_NOWAIT_BIT; } -static inline struct file *io_slot_file(struct io_fixed_file *slot) +static inline struct file *io_slot_file(struct io_rsrc_node *node) { - return (struct file *)(slot->file_ptr & FFS_MASK); + return (struct file *)(node->file_ptr & FFS_MASK); } static inline struct file *io_file_from_index(struct io_file_table *table, int index) { - return io_slot_file(io_fixed_file_slot(table, index)); + struct io_rsrc_node *node = table->nodes[index]; + + if (node) + return io_slot_file(node); + return NULL; } -static inline void io_fixed_file_set(struct io_fixed_file *file_slot, +static inline void io_fixed_file_set(struct io_rsrc_node *node, struct file *file) { - file_slot->file_ptr = (unsigned long)file | + node->file_ptr = (unsigned long)file | (io_file_get_flags(file) >> REQ_F_SUPPORT_NOWAIT_BIT); } diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 187d4a6b8337..95ad74993a3f 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -333,7 +333,6 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) mutex_init(&ctx->uring_lock); init_waitqueue_head(&ctx->cq_wait); init_waitqueue_head(&ctx->poll_wq); - init_waitqueue_head(&ctx->rsrc_quiesce_wq); spin_lock_init(&ctx->completion_lock); spin_lock_init(&ctx->timeout_lock); INIT_WQ_LIST(&ctx->iopoll_list); @@ -341,7 +340,6 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) INIT_LIST_HEAD(&ctx->defer_list); INIT_LIST_HEAD(&ctx->timeout_list); INIT_LIST_HEAD(&ctx->ltimeout_list); - INIT_LIST_HEAD(&ctx->rsrc_ref_list); init_llist_head(&ctx->work_llist); INIT_LIST_HEAD(&ctx->tctx_list); ctx->submit_state.free_list.next = NULL; @@ -1415,7 +1413,7 @@ static void io_free_batch_list(struct io_ring_ctx *ctx, io_clean_op(req); } io_put_file(req); - io_put_rsrc_node(ctx, req->rsrc_node); + io_req_put_rsrc_nodes(req); io_put_task(req->task); node = req->comp_list.next; @@ -1878,7 +1876,7 @@ inline struct file *io_file_get_fixed(struct io_kiocb *req, int fd, unsigned int issue_flags) { struct io_ring_ctx *ctx = req->ctx; - struct io_fixed_file *slot; + struct io_rsrc_node *node; struct file *file = NULL; io_ring_submit_lock(ctx, issue_flags); @@ -1886,11 +1884,12 @@ inline struct file *io_file_get_fixed(struct io_kiocb *req, int fd, if (unlikely((unsigned int)fd >= ctx->nr_user_files)) goto out; fd = array_index_nospec(fd, ctx->nr_user_files); - slot = io_fixed_file_slot(&ctx->file_table, fd); - if (!req->rsrc_node) - __io_req_set_rsrc_node(req, ctx); - req->flags |= io_slot_flags(slot); - file = io_slot_file(slot); + node = ctx->file_table.nodes[fd]; + if (node) { + io_req_assign_rsrc_node(req, node); + req->flags |= io_slot_flags(node); + file = io_slot_file(node); + } out: io_ring_submit_unlock(ctx, issue_flags); return file; @@ -2036,7 +2035,8 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, req->flags = (__force io_req_flags_t) sqe_flags; req->cqe.user_data = READ_ONCE(sqe->user_data); req->file = NULL; - req->rsrc_node = NULL; + req->rsrc_nodes[IORING_RSRC_FILE] = NULL; + req->rsrc_nodes[IORING_RSRC_BUFFER] = NULL; req->task = current; req->cancel_seq_set = false; @@ -2722,15 +2722,10 @@ static void io_req_caches_free(struct io_ring_ctx *ctx) static __cold void io_ring_ctx_free(struct io_ring_ctx *ctx) { io_sq_thread_finish(ctx); - /* __io_rsrc_put_work() may need uring_lock to progress, wait w/o it */ - if (WARN_ON_ONCE(!list_empty(&ctx->rsrc_ref_list))) - return; mutex_lock(&ctx->uring_lock); - if (ctx->buf_data) - __io_sqe_buffers_unregister(ctx); - if (ctx->file_data) - __io_sqe_files_unregister(ctx); + io_sqe_buffers_unregister(ctx); + io_sqe_files_unregister(ctx); io_cqring_overflow_kill(ctx); io_eventfd_unregister(ctx); io_alloc_cache_free(&ctx->apoll_cache, kfree); @@ -2746,11 +2741,6 @@ static __cold void io_ring_ctx_free(struct io_ring_ctx *ctx) if (ctx->submitter_task) put_task_struct(ctx->submitter_task); - /* there are no registered resources left, nobody uses it */ - if (ctx->rsrc_node) - io_rsrc_node_destroy(ctx, ctx->rsrc_node); - - WARN_ON_ONCE(!list_empty(&ctx->rsrc_ref_list)); WARN_ON_ONCE(!list_empty(&ctx->ltimeout_list)); io_alloc_cache_free(&ctx->rsrc_node_cache, kfree); @@ -3668,10 +3658,6 @@ static __cold int io_uring_create(unsigned entries, struct io_uring_params *p, if (ret) goto err; - ret = io_rsrc_init(ctx); - if (ret) - goto err; - p->features = IORING_FEAT_SINGLE_MMAP | IORING_FEAT_NODROP | IORING_FEAT_SUBMIT_STABLE | IORING_FEAT_RW_CUR_POS | IORING_FEAT_CUR_PERSONALITY | IORING_FEAT_FAST_POLL | diff --git a/io_uring/net.c b/io_uring/net.c index 2040195e33ab..ce1156551d10 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -1342,15 +1342,15 @@ static int io_send_zc_import(struct io_kiocb *req, unsigned int issue_flags) if (sr->flags & IORING_RECVSEND_FIXED_BUF) { struct io_ring_ctx *ctx = req->ctx; - struct io_mapped_ubuf *imu; + struct io_rsrc_node *node; int idx; ret = -EFAULT; io_ring_submit_lock(ctx, issue_flags); if (sr->buf_index < ctx->nr_user_bufs) { idx = array_index_nospec(sr->buf_index, ctx->nr_user_bufs); - imu = READ_ONCE(ctx->user_bufs[idx]); - io_req_set_rsrc_node(sr->notif, ctx); + node = ctx->user_bufs[idx]; + io_req_assign_rsrc_node(sr->notif, node); ret = 0; } io_ring_submit_unlock(ctx, issue_flags); @@ -1358,8 +1358,9 @@ static int io_send_zc_import(struct io_kiocb *req, unsigned int issue_flags) if (unlikely(ret)) return ret; - ret = io_import_fixed(ITER_SOURCE, &kmsg->msg.msg_iter, imu, - (u64)(uintptr_t)sr->buf, sr->len); + ret = io_import_fixed(ITER_SOURCE, &kmsg->msg.msg_iter, + node->buf, (u64)(uintptr_t)sr->buf, + sr->len); if (unlikely(ret)) return ret; kmsg->msg.sg_from_iter = io_sg_from_iter; diff --git a/io_uring/notif.c b/io_uring/notif.c index 28859ae3ee6e..4f02e969cf08 100644 --- a/io_uring/notif.c +++ b/io_uring/notif.c @@ -117,7 +117,8 @@ struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx) notif->file = NULL; notif->task = current; io_get_task_refs(1); - notif->rsrc_node = NULL; + notif->rsrc_nodes[IORING_RSRC_FILE] = NULL; + notif->rsrc_nodes[IORING_RSRC_BUFFER] = NULL; nd = io_notif_to_data(notif); nd->zc_report = false; diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index ca2ec8a018be..9822222a7acb 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -26,10 +26,8 @@ struct io_rsrc_update { u32 offset; }; -static void io_rsrc_buf_put(struct io_ring_ctx *ctx, struct io_rsrc_put *prsrc); -static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, - struct io_mapped_ubuf **pimu, - struct page **last_hpage); +static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx, + struct iovec *iov, int index, struct page **last_hpage); /* only define max */ #define IORING_MAX_FIXED_FILES (1U << 20) @@ -110,13 +108,13 @@ static int io_buffer_validate(struct iovec *iov) return 0; } -static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_mapped_ubuf **slot) +static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_rsrc_node *node) { - struct io_mapped_ubuf *imu = *slot; unsigned int i; - *slot = NULL; - if (imu != &dummy_ubuf) { + if (node->buf != &dummy_ubuf) { + struct io_mapped_ubuf *imu = node->buf; + if (!refcount_dec_and_test(&imu->refs)) return; for (i = 0; i < imu->nr_bvecs; i++) @@ -125,207 +123,63 @@ static void io_buffer_unmap(struct io_ring_ctx *ctx, struct io_mapped_ubuf **slo io_unaccount_mem(ctx, imu->acct_pages); kvfree(imu); } + node->rsrc = NULL; } -static void io_rsrc_put_work(struct io_rsrc_node *node) -{ - struct io_rsrc_put *prsrc = &node->item; - - if (prsrc->tag) - io_post_aux_cqe(node->ctx, prsrc->tag, 0, 0); - - switch (node->type) { - case IORING_RSRC_FILE: - fput(prsrc->file); - break; - case IORING_RSRC_BUFFER: - io_rsrc_buf_put(node->ctx, prsrc); - break; - default: - WARN_ON_ONCE(1); - break; - } -} - -void io_rsrc_node_destroy(struct io_ring_ctx *ctx, struct io_rsrc_node *node) -{ - if (!io_alloc_cache_put(&ctx->rsrc_node_cache, node)) - kfree(node); -} - -void io_rsrc_node_ref_zero(struct io_rsrc_node *node) - __must_hold(&node->ctx->uring_lock) -{ - struct io_ring_ctx *ctx = node->ctx; - - while (!list_empty(&ctx->rsrc_ref_list)) { - node = list_first_entry(&ctx->rsrc_ref_list, - struct io_rsrc_node, node); - /* recycle ref nodes in order */ - if (node->refs) - break; - list_del(&node->node); - - if (likely(!node->empty)) - io_rsrc_put_work(node); - io_rsrc_node_destroy(ctx, node); - } - if (list_empty(&ctx->rsrc_ref_list) && unlikely(ctx->rsrc_quiesce)) - wake_up_all(&ctx->rsrc_quiesce_wq); -} - -struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx) +struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx, + struct io_rsrc_data *data, int index, + int type) { - struct io_rsrc_node *ref_node; + struct io_rsrc_node *node; - ref_node = io_alloc_cache_get(&ctx->rsrc_node_cache); - if (!ref_node) { - ref_node = kzalloc(sizeof(*ref_node), GFP_KERNEL); - if (!ref_node) + node = io_alloc_cache_get(&ctx->rsrc_node_cache); + if (!node) { + node = kzalloc(sizeof(*node), GFP_KERNEL); + if (!node) return NULL; } - ref_node->ctx = ctx; - ref_node->empty = 0; - ref_node->refs = 1; - return ref_node; -} - -__cold static int io_rsrc_ref_quiesce(struct io_rsrc_data *data, - struct io_ring_ctx *ctx) -{ - struct io_rsrc_node *backup; - DEFINE_WAIT(we); - int ret; - - /* As We may drop ->uring_lock, other task may have started quiesce */ - if (data->quiesce) - return -ENXIO; - - backup = io_rsrc_node_alloc(ctx); - if (!backup) - return -ENOMEM; - ctx->rsrc_node->empty = true; - ctx->rsrc_node->type = -1; - list_add_tail(&ctx->rsrc_node->node, &ctx->rsrc_ref_list); - io_put_rsrc_node(ctx, ctx->rsrc_node); - ctx->rsrc_node = backup; - - if (list_empty(&ctx->rsrc_ref_list)) - return 0; - - if (ctx->flags & IORING_SETUP_DEFER_TASKRUN) { - atomic_set(&ctx->cq_wait_nr, 1); - smp_mb(); - } - - ctx->rsrc_quiesce++; - data->quiesce = true; - do { - prepare_to_wait(&ctx->rsrc_quiesce_wq, &we, TASK_INTERRUPTIBLE); - mutex_unlock(&ctx->uring_lock); - - ret = io_run_task_work_sig(ctx); - if (ret < 0) { - finish_wait(&ctx->rsrc_quiesce_wq, &we); - mutex_lock(&ctx->uring_lock); - if (list_empty(&ctx->rsrc_ref_list)) - ret = 0; - break; - } - - schedule(); - mutex_lock(&ctx->uring_lock); - ret = 0; - } while (!list_empty(&ctx->rsrc_ref_list)); - - finish_wait(&ctx->rsrc_quiesce_wq, &we); - data->quiesce = false; - ctx->rsrc_quiesce--; - - if (ctx->flags & IORING_SETUP_DEFER_TASKRUN) { - atomic_set(&ctx->cq_wait_nr, 0); - smp_mb(); - } - return ret; -} - -static void io_free_page_table(void **table, size_t size) -{ - unsigned i, nr_tables = DIV_ROUND_UP(size, PAGE_SIZE); - - for (i = 0; i < nr_tables; i++) - kfree(table[i]); - kfree(table); + node->ctx = ctx; + node->refs = 1; + node->type = type; + node->index = index; + node->tag = 0; + node->rsrc = NULL; + return node; } static void io_rsrc_data_free(struct io_rsrc_data *data) { - size_t size = data->nr * sizeof(data->tags[0][0]); - - if (data->tags) - io_free_page_table((void **)data->tags, size); - kfree(data); -} - -static __cold void **io_alloc_page_table(size_t size) -{ - unsigned i, nr_tables = DIV_ROUND_UP(size, PAGE_SIZE); - size_t init_size = size; - void **table; - - table = kcalloc(nr_tables, sizeof(*table), GFP_KERNEL_ACCOUNT); - if (!table) - return NULL; + int i; - for (i = 0; i < nr_tables; i++) { - unsigned int this_size = min_t(size_t, size, PAGE_SIZE); + for (i = 0; i < data->nr; i++) { + struct io_rsrc_node *node = data->nodes[i]; - table[i] = kzalloc(this_size, GFP_KERNEL_ACCOUNT); - if (!table[i]) { - io_free_page_table(table, init_size); - return NULL; - } - size -= this_size; + io_put_rsrc_node(node); } - return table; + kvfree(data->nodes); + kfree(data); } -__cold static int io_rsrc_data_alloc(struct io_ring_ctx *ctx, int type, - u64 __user *utags, - unsigned nr, struct io_rsrc_data **pdata) +__cold static int io_rsrc_data_alloc(struct io_ring_ctx *ctx, unsigned nr, + struct io_rsrc_data **pdata) { struct io_rsrc_data *data; - int ret = 0; - unsigned i; data = kzalloc(sizeof(*data), GFP_KERNEL); if (!data) return -ENOMEM; - data->tags = (u64 **)io_alloc_page_table(nr * sizeof(data->tags[0][0])); - if (!data->tags) { - kfree(data); + + data->nodes = kvmalloc_array(nr, sizeof(struct io_rsrc_node *), + GFP_KERNEL | __GFP_ZERO); + if (!data->nodes) { + io_rsrc_data_free(data); return -ENOMEM; } data->nr = nr; - data->ctx = ctx; - data->rsrc_type = type; - if (utags) { - ret = -EFAULT; - for (i = 0; i < nr; i++) { - u64 *tag_slot = io_get_tag_slot(data, i); - - if (copy_from_user(tag_slot, &utags[i], - sizeof(*tag_slot))) - goto fail; - } - } *pdata = data; return 0; -fail: - io_rsrc_data_free(data); - return ret; } static int __io_sqe_files_update(struct io_ring_ctx *ctx, @@ -334,8 +188,6 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, { u64 __user *tags = u64_to_user_ptr(up->tags); __s32 __user *fds = u64_to_user_ptr(up->data); - struct io_rsrc_data *data = ctx->file_data; - struct io_fixed_file *file_slot; int fd, i, err = 0; unsigned int done; @@ -360,18 +212,14 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, continue; i = array_index_nospec(up->offset + done, ctx->nr_user_files); - file_slot = io_fixed_file_slot(&ctx->file_table, i); - - if (file_slot->file_ptr) { - err = io_queue_rsrc_removal(data, i, - io_slot_file(file_slot)); - if (err) - break; - file_slot->file_ptr = 0; + if (ctx->file_table.nodes[i]) { + io_put_rsrc_node(ctx->file_table.nodes[i]); + ctx->file_table.nodes[i] = NULL; io_file_bitmap_clear(&ctx->file_table, i); } if (fd != -1) { struct file *file = fget(fd); + struct io_rsrc_node *node; if (!file) { err = -EBADF; @@ -385,8 +233,15 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, err = -EBADF; break; } - *io_get_tag_slot(data, i) = tag; - io_fixed_file_set(file_slot, file); + node = io_rsrc_node_alloc(ctx, ctx->file_data, i, IORING_RSRC_FILE); + if (!node) { + err = -ENOMEM; + fput(file); + break; + } + ctx->file_table.nodes[i] = node; + node->tag = tag; + io_fixed_file_set(node, file); io_file_bitmap_set(&ctx->file_table, i); } } @@ -411,7 +266,7 @@ static int __io_sqe_buffers_update(struct io_ring_ctx *ctx, return -EINVAL; for (done = 0; done < nr_args; done++) { - struct io_mapped_ubuf *imu; + struct io_rsrc_node *node; u64 tag = 0; uvec = u64_to_user_ptr(user_data); @@ -431,23 +286,16 @@ static int __io_sqe_buffers_update(struct io_ring_ctx *ctx, err = -EINVAL; break; } - err = io_sqe_buffer_register(ctx, iov, &imu, &last_hpage); - if (err) - break; - i = array_index_nospec(up->offset + done, ctx->nr_user_bufs); - if (ctx->user_bufs[i] != &dummy_ubuf) { - err = io_queue_rsrc_removal(ctx->buf_data, i, - ctx->user_bufs[i]); - if (unlikely(err)) { - io_buffer_unmap(ctx, &imu); - break; - } - ctx->user_bufs[i] = (struct io_mapped_ubuf *)&dummy_ubuf; + node = io_sqe_buffer_register(ctx, iov, i, &last_hpage); + if (IS_ERR(node)) { + err = PTR_ERR(node); + break; } + io_put_rsrc_node(ctx->user_bufs[i]); - ctx->user_bufs[i] = imu; - *io_get_tag_slot(ctx->buf_data, i) = tag; + ctx->user_bufs[i] = node; + node->tag = tag; if (ctx->compat) user_data += sizeof(struct compat_iovec); else @@ -622,38 +470,47 @@ int io_files_update(struct io_kiocb *req, unsigned int issue_flags) return IOU_OK; } -int io_queue_rsrc_removal(struct io_rsrc_data *data, unsigned idx, void *rsrc) +void io_free_rsrc_node(struct io_rsrc_node *node) { - struct io_ring_ctx *ctx = data->ctx; - struct io_rsrc_node *node = ctx->rsrc_node; - u64 *tag_slot = io_get_tag_slot(data, idx); + struct io_ring_ctx *ctx = node->ctx; - ctx->rsrc_node = io_rsrc_node_alloc(ctx); - if (unlikely(!ctx->rsrc_node)) { - ctx->rsrc_node = node; - return -ENOMEM; + lockdep_assert_held(&ctx->uring_lock); + + if (node->tag) + io_post_aux_cqe(node->ctx, node->tag, 0, 0); + + switch (node->type) { + case IORING_RSRC_FILE: + if (io_slot_file(node)) + fput(io_slot_file(node)); + break; + case IORING_RSRC_BUFFER: + if (node->buf) + io_buffer_unmap(node->ctx, node); + break; + default: + WARN_ON_ONCE(1); + break; } - node->item.rsrc = rsrc; - node->type = data->rsrc_type; - node->item.tag = *tag_slot; - *tag_slot = 0; - list_add_tail(&node->node, &ctx->rsrc_ref_list); - io_put_rsrc_node(ctx, node); - return 0; + if (!io_alloc_cache_put(&ctx->rsrc_node_cache, node)) + kfree(node); } -void __io_sqe_files_unregister(struct io_ring_ctx *ctx) +static void __io_sqe_files_unregister(struct io_ring_ctx *ctx) { int i; + lockdep_assert_held(&ctx->uring_lock); + for (i = 0; i < ctx->nr_user_files; i++) { - struct file *file = io_file_from_index(&ctx->file_table, i); + struct io_rsrc_node *node = ctx->file_table.nodes[i]; - if (!file) - continue; - io_file_bitmap_clear(&ctx->file_table, i); - fput(file); + if (node) { + io_put_rsrc_node(node); + io_file_bitmap_clear(&ctx->file_table, i); + ctx->file_table.nodes[i] = NULL; + } } io_free_file_tables(&ctx->file_table); @@ -665,22 +522,11 @@ void __io_sqe_files_unregister(struct io_ring_ctx *ctx) int io_sqe_files_unregister(struct io_ring_ctx *ctx) { - unsigned nr = ctx->nr_user_files; - int ret; - if (!ctx->file_data) return -ENXIO; - /* - * Quiesce may unlock ->uring_lock, and while it's not held - * prevent new requests using the table. - */ - ctx->nr_user_files = 0; - ret = io_rsrc_ref_quiesce(ctx->file_data, ctx); - ctx->nr_user_files = nr; - if (!ret) - __io_sqe_files_unregister(ctx); - return ret; + __io_sqe_files_unregister(ctx); + return 0; } int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, @@ -699,8 +545,7 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, return -EMFILE; if (nr_args > rlimit(RLIMIT_NOFILE)) return -EMFILE; - ret = io_rsrc_data_alloc(ctx, IORING_RSRC_FILE, tags, nr_args, - &ctx->file_data); + ret = io_rsrc_data_alloc(ctx, nr_args, &ctx->file_data); if (ret) return ret; @@ -711,17 +556,20 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, } for (i = 0; i < nr_args; i++, ctx->nr_user_files++) { - struct io_fixed_file *file_slot; + struct io_rsrc_node *node; + u64 tag = 0; - if (fds && copy_from_user(&fd, &fds[i], sizeof(fd))) { - ret = -EFAULT; + ret = -EFAULT; + if (tags && copy_from_user(&tag, &tags[i], sizeof(tag))) + goto fail; + if (fds && copy_from_user(&fd, &fds[i], sizeof(fd))) goto fail; - } /* allow sparse sets */ if (!fds || fd == -1) { ret = -EINVAL; - if (unlikely(*io_get_tag_slot(ctx->file_data, i))) + if (tag) goto fail; + ctx->file_table.nodes[i] = NULL; continue; } @@ -737,8 +585,16 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, fput(file); goto fail; } - file_slot = io_fixed_file_slot(&ctx->file_table, i); - io_fixed_file_set(file_slot, file); + ret = -ENOMEM; + node = io_rsrc_node_alloc(ctx, ctx->file_data, i, IORING_RSRC_FILE); + if (!node) { + fput(file); + goto fail; + } + if (tag) + node->tag = tag; + ctx->file_table.nodes[i] = node; + io_fixed_file_set(node, file); io_file_bitmap_set(&ctx->file_table, i); } @@ -750,43 +606,30 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, return ret; } -static void io_rsrc_buf_put(struct io_ring_ctx *ctx, struct io_rsrc_put *prsrc) -{ - io_buffer_unmap(ctx, &prsrc->buf); - prsrc->buf = NULL; -} - -void __io_sqe_buffers_unregister(struct io_ring_ctx *ctx) +static void __io_sqe_buffers_unregister(struct io_ring_ctx *ctx) { unsigned int i; - for (i = 0; i < ctx->nr_user_bufs; i++) - io_buffer_unmap(ctx, &ctx->user_bufs[i]); - kfree(ctx->user_bufs); - io_rsrc_data_free(ctx->buf_data); + lockdep_assert_held(&ctx->uring_lock); + + for (i = 0; i < ctx->nr_user_bufs; i++) { + io_put_rsrc_node(ctx->user_bufs[i]); + ctx->user_bufs[i] = NULL; + } + kvfree(ctx->user_bufs); ctx->user_bufs = NULL; + io_rsrc_data_free(ctx->buf_data); ctx->buf_data = NULL; ctx->nr_user_bufs = 0; } int io_sqe_buffers_unregister(struct io_ring_ctx *ctx) { - unsigned nr = ctx->nr_user_bufs; - int ret; - if (!ctx->buf_data) return -ENXIO; - /* - * Quiesce may unlock ->uring_lock, and while it's not held - * prevent new requests using the table. - */ - ctx->nr_user_bufs = 0; - ret = io_rsrc_ref_quiesce(ctx->buf_data, ctx); - ctx->nr_user_bufs = nr; - if (!ret) - __io_sqe_buffers_unregister(ctx); - return ret; + __io_sqe_buffers_unregister(ctx); + return 0; } /* @@ -813,7 +656,8 @@ static bool headpage_already_acct(struct io_ring_ctx *ctx, struct page **pages, /* check previously registered pages */ for (i = 0; i < ctx->nr_user_bufs; i++) { - struct io_mapped_ubuf *imu = ctx->user_bufs[i]; + struct io_rsrc_node *node = ctx->user_bufs[i]; + struct io_mapped_ubuf *imu = node->buf; for (j = 0; j < imu->nr_bvecs; j++) { if (!PageCompound(imu->bvec[j].bv_page)) @@ -950,21 +794,28 @@ static bool io_try_coalesce_buffer(struct page ***pages, int *nr_pages, return io_do_coalesce_buffer(pages, nr_pages, data, nr_folios); } -static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, - struct io_mapped_ubuf **pimu, - struct page **last_hpage) +static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx, + struct iovec *iov, + int index, + struct page **last_hpage) { struct io_mapped_ubuf *imu = NULL; struct page **pages = NULL; + struct io_rsrc_node *node; unsigned long off; size_t size; int ret, nr_pages, i; struct io_imu_folio_data data; bool coalesced; - *pimu = (struct io_mapped_ubuf *)&dummy_ubuf; - if (!iov->iov_base) - return 0; + node = io_rsrc_node_alloc(ctx, ctx->buf_data, index, IORING_RSRC_BUFFER); + if (!node) + return ERR_PTR(-ENOMEM); + + if (!iov->iov_base) { + node->buf = (struct io_mapped_ubuf *) &dummy_ubuf; + return node; + } ret = -ENOMEM; pages = io_pin_pages((unsigned long) iov->iov_base, iov->iov_len, @@ -998,7 +849,7 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, imu->folio_shift = data.folio_shift; refcount_set(&imu->refs, 1); off = (unsigned long) iov->iov_base & ((1UL << imu->folio_shift) - 1); - *pimu = imu; + node->buf = imu; ret = 0; for (i = 0; i < nr_pages; i++) { @@ -1010,10 +861,14 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, size -= vec_len; } done: - if (ret) + if (ret) { kvfree(imu); + if (node) + io_put_rsrc_node(node); + node = ERR_PTR(ret); + } kvfree(pages); - return ret; + return node; } static int io_buffers_map_alloc(struct io_ring_ctx *ctx, unsigned int nr_args) @@ -1037,7 +892,7 @@ int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg, return -EBUSY; if (!nr_args || nr_args > IORING_MAX_REG_BUFFERS) return -EINVAL; - ret = io_rsrc_data_alloc(ctx, IORING_RSRC_BUFFER, tags, nr_args, &data); + ret = io_rsrc_data_alloc(ctx, nr_args, &data); if (ret) return ret; ret = io_buffers_map_alloc(ctx, nr_args); @@ -1050,6 +905,9 @@ int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg, memset(iov, 0, sizeof(*iov)); for (i = 0; i < nr_args; i++, ctx->nr_user_bufs++) { + struct io_rsrc_node *node; + u64 tag = 0; + if (arg) { uvec = (struct iovec __user *) arg; iov = iovec_from_user(uvec, 1, 1, &fast_iov, ctx->compat); @@ -1066,15 +924,24 @@ int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg, arg += sizeof(struct iovec); } - if (!iov->iov_base && *io_get_tag_slot(data, i)) { - ret = -EINVAL; - break; + if (tags) { + if (copy_from_user(&tag, &tags[i], sizeof(tag))) { + ret = -EFAULT; + break; + } + if (tag && !iov->iov_base) { + ret = -EINVAL; + break; + } } - ret = io_sqe_buffer_register(ctx, iov, &ctx->user_bufs[i], - &last_hpage); - if (ret) + node = io_sqe_buffer_register(ctx, iov, i, &last_hpage); + if (IS_ERR(node)) { + ret = PTR_ERR(node); break; + } + node->tag = tag; + ctx->user_bufs[i] = node; } WARN_ON_ONCE(ctx->buf_data); @@ -1148,7 +1015,7 @@ int io_import_fixed(int ddir, struct iov_iter *iter, static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx) { - struct io_mapped_ubuf **user_bufs; + struct io_rsrc_node **user_bufs; struct io_rsrc_data *data; int i, ret, nbufs; @@ -1163,21 +1030,31 @@ static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx nbufs = src_ctx->nr_user_bufs; if (!nbufs) goto out_unlock; - ret = io_rsrc_data_alloc(ctx, IORING_RSRC_BUFFER, NULL, nbufs, &data); + ret = io_rsrc_data_alloc(ctx, nbufs, &data); if (ret) goto out_unlock; ret = -ENOMEM; - user_bufs = kcalloc(nbufs, sizeof(*ctx->user_bufs), GFP_KERNEL); + user_bufs = kvmalloc_array(nbufs, sizeof(struct io_rsrc_node *), + GFP_KERNEL); if (!user_bufs) goto out_free_data; for (i = 0; i < nbufs; i++) { - struct io_mapped_ubuf *src = src_ctx->user_bufs[i]; + struct io_mapped_ubuf *imu = src_ctx->user_bufs[i]->buf; + struct io_rsrc_node *dst_node; - if (src != &dummy_ubuf) - refcount_inc(&src->refs); - user_bufs[i] = src; + dst_node = io_rsrc_node_alloc(ctx, data, i, IORING_RSRC_BUFFER); + if (!dst_node) + goto out_put_free; + + if (imu == &dummy_ubuf) { + dst_node->buf = (struct io_mapped_ubuf *) &dummy_ubuf; + } else { + refcount_inc(&imu->refs); + dst_node->buf = imu; + } + user_bufs[i] = dst_node; } /* Have a ref on the bufs now, drop src lock and re-grab our own lock */ @@ -1190,12 +1067,17 @@ static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx return 0; } + mutex_unlock(&ctx->uring_lock); + mutex_lock(&src_ctx->uring_lock); /* someone raced setting up buffers, dump ours */ - for (i = 0; i < nbufs; i++) - io_buffer_unmap(ctx, &user_bufs[i]); - io_rsrc_data_free(data); - kfree(user_bufs); - return -EBUSY; + ret = -EBUSY; + i = nbufs; +out_put_free: + while (i--) { + io_buffer_unmap(src_ctx, user_bufs[i]); + kfree(user_bufs[i]); + } + kvfree(user_bufs); out_free_data: io_rsrc_data_free(data); out_unlock: diff --git a/io_uring/rsrc.h b/io_uring/rsrc.h index e072fb3ee351..22285169c178 100644 --- a/io_uring/rsrc.h +++ b/io_uring/rsrc.h @@ -11,38 +11,26 @@ enum { IORING_RSRC_FILE = 0, IORING_RSRC_BUFFER = 1, -}; - -struct io_rsrc_put { - u64 tag; - union { - void *rsrc; - struct file *file; - struct io_mapped_ubuf *buf; - }; + IORING_RSRC_INVALID = 2, }; struct io_rsrc_data { - struct io_ring_ctx *ctx; - - u64 **tags; unsigned int nr; - u16 rsrc_type; - bool quiesce; + struct io_rsrc_node **nodes; }; struct io_rsrc_node { struct io_ring_ctx *ctx; int refs; - bool empty; + int index; u16 type; - struct list_head node; - struct io_rsrc_put item; -}; -struct io_fixed_file { - /* file * with additional FFS_* flags */ - unsigned long file_ptr; + u64 tag; + union { + void *rsrc; + unsigned long file_ptr; + struct io_mapped_ubuf *buf; + }; }; struct io_mapped_ubuf { @@ -63,21 +51,19 @@ struct io_imu_folio_data { unsigned int folio_shift; }; -void io_rsrc_node_ref_zero(struct io_rsrc_node *node); -void io_rsrc_node_destroy(struct io_ring_ctx *ctx, struct io_rsrc_node *ref_node); -struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx); -int io_queue_rsrc_removal(struct io_rsrc_data *data, unsigned idx, void *rsrc); +struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx, + struct io_rsrc_data *data, int index, + int type); +void io_free_rsrc_node(struct io_rsrc_node *node); int io_import_fixed(int ddir, struct iov_iter *iter, struct io_mapped_ubuf *imu, u64 buf_addr, size_t len); int io_register_clone_buffers(struct io_ring_ctx *ctx, void __user *arg); -void __io_sqe_buffers_unregister(struct io_ring_ctx *ctx); int io_sqe_buffers_unregister(struct io_ring_ctx *ctx); int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg, unsigned int nr_args, u64 __user *tags); -void __io_sqe_files_unregister(struct io_ring_ctx *ctx); int io_sqe_files_unregister(struct io_ring_ctx *ctx); int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, unsigned nr_args, u64 __user *tags); @@ -89,47 +75,23 @@ int io_register_rsrc_update(struct io_ring_ctx *ctx, void __user *arg, int io_register_rsrc(struct io_ring_ctx *ctx, void __user *arg, unsigned int size, unsigned int type); -static inline void io_put_rsrc_node(struct io_ring_ctx *ctx, struct io_rsrc_node *node) +static inline void io_put_rsrc_node(struct io_rsrc_node *node) { - lockdep_assert_held(&ctx->uring_lock); - if (node && !--node->refs) - io_rsrc_node_ref_zero(node); -} - -static inline void io_charge_rsrc_node(struct io_ring_ctx *ctx, - struct io_rsrc_node *node) -{ - node->refs++; -} - -static inline void __io_req_set_rsrc_node(struct io_kiocb *req, - struct io_ring_ctx *ctx) -{ - lockdep_assert_held(&ctx->uring_lock); - req->rsrc_node = ctx->rsrc_node; - io_charge_rsrc_node(ctx, ctx->rsrc_node); + io_free_rsrc_node(node); } -static inline void io_req_set_rsrc_node(struct io_kiocb *req, - struct io_ring_ctx *ctx) +static inline void io_req_put_rsrc_nodes(struct io_kiocb *req) { - if (!req->rsrc_node) - __io_req_set_rsrc_node(req, ctx); + io_put_rsrc_node(req->rsrc_nodes[IORING_RSRC_FILE]); + io_put_rsrc_node(req->rsrc_nodes[IORING_RSRC_BUFFER]); } -static inline u64 *io_get_tag_slot(struct io_rsrc_data *data, unsigned int idx) +static inline void io_req_assign_rsrc_node(struct io_kiocb *req, + struct io_rsrc_node *node) { - unsigned int off = idx & IO_RSRC_TAG_TABLE_MASK; - unsigned int table_idx = idx >> IO_RSRC_TAG_TABLE_SHIFT; - - return &data->tags[table_idx][off]; -} - -static inline int io_rsrc_init(struct io_ring_ctx *ctx) -{ - ctx->rsrc_node = io_rsrc_node_alloc(ctx); - return ctx->rsrc_node ? 0 : -ENOMEM; + node->refs++; + req->rsrc_nodes[node->type] = node; } int io_files_update(struct io_kiocb *req, unsigned int issue_flags); diff --git a/io_uring/rw.c b/io_uring/rw.c index 8080ffd6d571..65491f4f2c7e 100644 --- a/io_uring/rw.c +++ b/io_uring/rw.c @@ -330,7 +330,7 @@ static int io_prep_rw_fixed(struct io_kiocb *req, const struct io_uring_sqe *sqe { struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw); struct io_ring_ctx *ctx = req->ctx; - struct io_mapped_ubuf *imu; + struct io_rsrc_node *node; struct io_async_rw *io; u16 index; int ret; @@ -342,11 +342,11 @@ static int io_prep_rw_fixed(struct io_kiocb *req, const struct io_uring_sqe *sqe if (unlikely(req->buf_index >= ctx->nr_user_bufs)) return -EFAULT; index = array_index_nospec(req->buf_index, ctx->nr_user_bufs); - imu = ctx->user_bufs[index]; - io_req_set_rsrc_node(req, ctx); + node = ctx->user_bufs[index]; + io_req_assign_rsrc_node(req, node); io = req->async_data; - ret = io_import_fixed(ddir, &io->iter, imu, rw->addr, rw->len); + ret = io_import_fixed(ddir, &io->iter, node->buf, rw->addr, rw->len); iov_iter_save_state(&io->iter, &io->iter_state); return ret; } diff --git a/io_uring/uring_cmd.c b/io_uring/uring_cmd.c index 6994f60d7ec7..0899c71008ae 100644 --- a/io_uring/uring_cmd.c +++ b/io_uring/uring_cmd.c @@ -220,7 +220,7 @@ int io_uring_cmd_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) * being called. This prevents destruction of the mapped buffer * we'll need at actual import time. */ - io_req_set_rsrc_node(req, ctx); + io_req_assign_rsrc_node(req, ctx->user_bufs[req->buf_index]); } ioucmd->cmd_op = READ_ONCE(sqe->cmd_op); @@ -276,15 +276,11 @@ int io_uring_cmd_import_fixed(u64 ubuf, unsigned long len, int rw, struct iov_iter *iter, void *ioucmd) { struct io_kiocb *req = cmd_to_io_kiocb(ioucmd); - struct io_ring_ctx *ctx = req->ctx; + struct io_rsrc_node *node = req->rsrc_nodes[IORING_RSRC_BUFFER]; /* Must have had rsrc_node assigned at prep time */ - if (req->rsrc_node) { - struct io_mapped_ubuf *imu; - - imu = READ_ONCE(ctx->user_bufs[req->buf_index]); - return io_import_fixed(rw, iter, imu, ubuf, len); - } + if (node) + return io_import_fixed(rw, iter, node->buf, ubuf, len); return -EFAULT; } From patchwork Sat Oct 26 22:08:29 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jens Axboe X-Patchwork-Id: 13852380 Received: from mail-pl1-f177.google.com (mail-pl1-f177.google.com [209.85.214.177]) (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 35AAD1D461B for ; Sat, 26 Oct 2024 22:24:01 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.177 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981444; cv=none; b=Ar9s87jqcIHDDc1P6Xe5MqXizM4azFEn1EDQ5x1cq+BrWpUulQAgxZWVV/V8xVVLW7EVE0RRAMitBo5YhxzNJDIqO4I6mgJ4OST5B5M4HBhFH5MIenDHFNkudTfOcRH0muDz1eVcvZNgDv8h5nehpzzCmQL7og++dQ72DhxiFDc= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981444; c=relaxed/simple; bh=HaRHgExVHYvR3LpjtwRGLKGiWieHlxX77HTAFVNc2lk=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=KAPOkGj6XP2If7Nnyv1Ur39GekrMqeD3eXxjtYgi9nAvbr0xKpbZnO3WvzesA+MR+MpHcWVIn6DXASqgJeMQH1nFPLtTQOfnmya0FdYZzoIcDOBxoR8MTXCU0aMRbASiV2yfkwg7MIWCuGK7SGUd6hNE2Tn7RnvTzP/MpgF7NG0= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk; spf=pass smtp.mailfrom=kernel.dk; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b=Dc/lqAq9; arc=none smtp.client-ip=209.85.214.177 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=kernel.dk Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b="Dc/lqAq9" Received: by mail-pl1-f177.google.com with SMTP id d9443c01a7336-20ce65c8e13so26440555ad.1 for ; Sat, 26 Oct 2024 15:24:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20230601.gappssmtp.com; s=20230601; t=1729981441; x=1730586241; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=nJ1XKYylAS5g6MXJsNyLduviD0yVNQeg99x/fT/sC04=; b=Dc/lqAq9X+z2PfyMkJ4BkqnFCmST3bMoVcWYha2F1eEGCHne3gcKztmk6n5H5CWNSR LEMOk5IwizeZa+iMxoLqU0vvuaW070PZ4GO7uwRogHdel+K8Whc7RMZQXnx9YWmB3rlL pAPUSnIvv6/2OVK+lZ9R4iqWzGlZ2ebxMY7n4molmFehjb/COoSN3REJlORSA4J9G2ah OqTQk8yYUMN/Nu0aFWR0kKVZIFWlXDxEQR6v5hCbmz8jiFUK48QpWeuk4u1AqsorcSJp DJhiO5yjyO5ukRKlOutOvFEy9p0A3y5k7LvVhsPY9YgGXKpifERVbP84bUIo+7uRYEeb hyGg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1729981441; x=1730586241; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=nJ1XKYylAS5g6MXJsNyLduviD0yVNQeg99x/fT/sC04=; b=OYT4Jnm0uft71B2s8B1NFmyU+DBX+Tjz9k6eedPaS6xP0Hld3kwmthVHjNksH6lUM0 rzT/0fmdI1NHTLjTHlA0lcng+22TPaZ0yIyEYBVKni/qfeFwNJeWwzK7p3jTmR4t0N4O Ie84L6raOmofL1+LMB726NtSLiNeBOpDiwaOL4cpuP5/BS9ihRzjUlEZERJ7gJAPJWo/ +pz3Ecfbu646sNQF1oSRBqUX/E9q1uSB/lpox0/9iYwiaJ5b7KEufdNIKH0VNzoxcBJb TFBKY0ha0hIm63YiZOjzrhLYkG7stzFkuY18r/eHVNlIxx5NXeHfzILiW5j0s9Shdxhb fUKg== X-Gm-Message-State: AOJu0Yyws4Bv49xHBWX69TiAwWckqpv5oqf5fsYMfjCKBIU6d0RP6GeR WSsDB+7h1+ZgRhf+nFOkLh+eubqvEN6WFRqAg0nsDsoWvG/5AzRU+fAexFGWIZWbwF6c4sUi6dL R X-Google-Smtp-Source: AGHT+IE0xY3wQDfeJVACMmH/M1N1re9d95M8pZ0OMm8G+QPRBIZRlTct7bV3DurAT8Tu+mjk07n/ag== X-Received: by 2002:a17:902:dac9:b0:20c:f3be:2f82 with SMTP id d9443c01a7336-210c6c82bb4mr44970015ad.33.1729981440986; Sat, 26 Oct 2024 15:24:00 -0700 (PDT) Received: from localhost.localdomain ([198.8.77.157]) by smtp.gmail.com with ESMTPSA id d9443c01a7336-210bbf44321sm28134705ad.30.2024.10.26.15.23.58 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 26 Oct 2024 15:23:59 -0700 (PDT) From: Jens Axboe To: io-uring@vger.kernel.org Cc: Jens Axboe Subject: [PATCH 4/7] io_uring/rsrc: get rid of io_rsrc_node allocation cache Date: Sat, 26 Oct 2024 16:08:29 -0600 Message-ID: <20241026222348.90331-5-axboe@kernel.dk> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20241026222348.90331-1-axboe@kernel.dk> References: <20241026222348.90331-1-axboe@kernel.dk> Precedence: bulk X-Mailing-List: io-uring@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 It's not going to be needed in the fast path going forward, so kill it off. Signed-off-by: Jens Axboe --- include/linux/io_uring_types.h | 3 --- io_uring/io_uring.c | 6 +----- io_uring/rsrc.c | 24 +++++++++--------------- 3 files changed, 10 insertions(+), 23 deletions(-) diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index 93111d87a88a..60541da6b875 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -362,9 +362,6 @@ struct io_ring_ctx { struct io_rsrc_data *file_data; struct io_rsrc_data *buf_data; - /* protected by ->uring_lock */ - struct io_alloc_cache rsrc_node_cache; - u32 pers_next; struct xarray personalities; diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 95ad74993a3f..8e24373c1c98 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -312,9 +312,7 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) INIT_LIST_HEAD(&ctx->sqd_list); INIT_LIST_HEAD(&ctx->cq_overflow_list); INIT_LIST_HEAD(&ctx->io_buffers_cache); - ret = io_alloc_cache_init(&ctx->rsrc_node_cache, IO_NODE_ALLOC_CACHE_MAX, - sizeof(struct io_rsrc_node)); - ret |= io_alloc_cache_init(&ctx->apoll_cache, IO_POLL_ALLOC_CACHE_MAX, + ret = io_alloc_cache_init(&ctx->apoll_cache, IO_POLL_ALLOC_CACHE_MAX, sizeof(struct async_poll)); ret |= io_alloc_cache_init(&ctx->netmsg_cache, IO_ALLOC_CACHE_MAX, sizeof(struct io_async_msghdr)); @@ -358,7 +356,6 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) free_ref: percpu_ref_exit(&ctx->refs); err: - io_alloc_cache_free(&ctx->rsrc_node_cache, kfree); io_alloc_cache_free(&ctx->apoll_cache, kfree); io_alloc_cache_free(&ctx->netmsg_cache, io_netmsg_cache_free); io_alloc_cache_free(&ctx->rw_cache, io_rw_cache_free); @@ -2743,7 +2740,6 @@ static __cold void io_ring_ctx_free(struct io_ring_ctx *ctx) WARN_ON_ONCE(!list_empty(&ctx->ltimeout_list)); - io_alloc_cache_free(&ctx->rsrc_node_cache, kfree); if (ctx->mm_account) { mmdrop(ctx->mm_account); ctx->mm_account = NULL; diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index 9822222a7acb..322749c7dee9 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -13,7 +13,6 @@ #include #include "io_uring.h" -#include "alloc_cache.h" #include "openclose.h" #include "rsrc.h" #include "memmap.h" @@ -132,19 +131,15 @@ struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx, { struct io_rsrc_node *node; - node = io_alloc_cache_get(&ctx->rsrc_node_cache); - if (!node) { - node = kzalloc(sizeof(*node), GFP_KERNEL); - if (!node) - return NULL; + node = kzalloc(sizeof(*node), GFP_KERNEL); + if (node) { + node->ctx = ctx; + node->refs = 1; + node->type = type; + node->index = index; + node->tag = 0; + node->rsrc = NULL; } - - node->ctx = ctx; - node->refs = 1; - node->type = type; - node->index = index; - node->tag = 0; - node->rsrc = NULL; return node; } @@ -493,8 +488,7 @@ void io_free_rsrc_node(struct io_rsrc_node *node) break; } - if (!io_alloc_cache_put(&ctx->rsrc_node_cache, node)) - kfree(node); + kfree(node); } static void __io_sqe_files_unregister(struct io_ring_ctx *ctx) From patchwork Sat Oct 26 22:08:30 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jens Axboe X-Patchwork-Id: 13852382 Received: from mail-pl1-f179.google.com (mail-pl1-f179.google.com [209.85.214.179]) (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 97C42146013 for ; Sat, 26 Oct 2024 22:24:03 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.179 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981446; cv=none; b=pcBTGa1dFe20rWtbLpkOgQf/OZBlqIXoagiN4c+VCAj/2JwRGUJ4ZxzVIptTZyTTGp61DSZz0CHmPUIlFQfvN6zcM3Xkk1AJWy6PSPEVwyEV5D+I/Uy59mQlfWVDOOWzZNBk0piS7YWXEN9L3VlQGLS0TKy/TE9ozRMhseA/vSs= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981446; c=relaxed/simple; bh=VDYifqCk0ipg93tPU3mBOhFEfpj5LZgdY2LvbPerBDI=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=oSSQVHahB+9nwhxlMYhFZbu/OgK8ou6psXBm20ivBJCHwYGLpYtXimwj8EsBKIeQ+RIYhIHgGpawGbypPSDAftgAtO6CYBFIHCwi0FNFxt2HFb3fT2lPIdwXNVKGqgcBXhmDDrdOjj0ugoIrqzqPxJAS7ofOyYBASYsIFVwcwCo= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk; spf=pass smtp.mailfrom=kernel.dk; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b=q4tXs60x; arc=none smtp.client-ip=209.85.214.179 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=kernel.dk Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b="q4tXs60x" Received: by mail-pl1-f179.google.com with SMTP id d9443c01a7336-20cdb889222so29153245ad.3 for ; Sat, 26 Oct 2024 15:24:03 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20230601.gappssmtp.com; s=20230601; t=1729981442; x=1730586242; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=JBRwY6ouz20EYxF9IRnXXUNz+wDhcc0tq154Wvig08I=; b=q4tXs60xr/dS9CggKYsjrAUFzRRMttJ82V4E2NhGjSAGOAs1+dr3SRO3yEW/weQ5XP OYqPV/XPVHYSdLRsipZVACZInuPls0g5HVPhzyijwb/Iu+PEURlDwEhAYZL9eD/z6vXT igTgH/mn3JO4/fXx+H+3JzaiSzBk3iPaI0F7faJTXojYc9wweVFaKPqp7623Kt2lEuyU UE9auY78Xw6Hdpw0XQaBUmM33bBgjsYwbA/oxnlZQyRvawDyrR6aJiDjjtmVqrcR3Nff V4lAVUTPTaqcv5UQ/34wM8l+ypvLzuGUDM8rgNZAqA0+BETqmziC6gAuwRtvwVDfshra hUCA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1729981442; x=1730586242; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=JBRwY6ouz20EYxF9IRnXXUNz+wDhcc0tq154Wvig08I=; b=l85amiBRATgz80FKE3HKUqRNjTxBIA3VeeCp7bKEf29RAG+y+lR87G2AybOJt26q8p cpkzvdcZ5+uSxCMvuVOivqCbKV7a4bpSt8KWi9jhAWoxtVYuW8W5NDPULJjlN3BLfmzH Q9W+RUBjSrnKfHj1Gw8KCW7199HCSmE1+LhGtckcqlOhwuP1wyRi+DUT0IWr5hO0JElR +ZViVEJDSOvisvGp3jWRpEjpy/bSrfMIBScqacbfADjPIJ8df1mnyFKpQYhbjoSZKXlI Y2my14wyOuadYQDs3ZPzNKZzi+qjNZ4JWYgVqP0TLg80pv1gDIFh4IWRfyUzXczif3w9 2UOA== X-Gm-Message-State: AOJu0YxSOVMsIk0cVqNMtDCgQcfx3XN+AGuYpNJiqNphEJLWYmdq6Ugq viLj6cLL2xB2oMe3uP+T+N0rwkJEDnVg/gdltEubYXudbtwkGwRdfNSfQ+tag4V66emWiAnBf75 O X-Google-Smtp-Source: AGHT+IFMhkMjE0uLfKbUvWNxtqYKC/1zPZ1aCdVm1qYfvL3OXf4dwjKWweBa9Zq8mLeNf7PaTg1+Uw== X-Received: by 2002:a17:902:db0b:b0:20b:5645:d860 with SMTP id d9443c01a7336-210c6c1ff52mr54193185ad.36.1729981442481; Sat, 26 Oct 2024 15:24:02 -0700 (PDT) Received: from localhost.localdomain ([198.8.77.157]) by smtp.gmail.com with ESMTPSA id d9443c01a7336-210bbf44321sm28134705ad.30.2024.10.26.15.24.01 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 26 Oct 2024 15:24:01 -0700 (PDT) From: Jens Axboe To: io-uring@vger.kernel.org Cc: Jens Axboe Subject: [PATCH 5/7] io_uring/rsrc: add an empty io_rsrc_node for sparse buffer entries Date: Sat, 26 Oct 2024 16:08:30 -0600 Message-ID: <20241026222348.90331-6-axboe@kernel.dk> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20241026222348.90331-1-axboe@kernel.dk> References: <20241026222348.90331-1-axboe@kernel.dk> Precedence: bulk X-Mailing-List: io-uring@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Rather than allocate an io_rsrc_node for an empty/sparse buffer entry, add a const entry that can be used for that. This just needs checking for writing the tag, and the put check needs to check for that sparse node rather than NULL for validity. This avoids allocating rsrc nodes for sparse buffer entries. Signed-off-by: Jens Axboe --- io_uring/io_uring.c | 4 ++-- io_uring/notif.c | 4 ++-- io_uring/rsrc.c | 52 +++++++++++++++++++++++++++------------------ io_uring/rsrc.h | 11 +++++++--- 4 files changed, 43 insertions(+), 28 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 8e24373c1c98..603fd51d170f 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -2032,8 +2032,8 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, req->flags = (__force io_req_flags_t) sqe_flags; req->cqe.user_data = READ_ONCE(sqe->user_data); req->file = NULL; - req->rsrc_nodes[IORING_RSRC_FILE] = NULL; - req->rsrc_nodes[IORING_RSRC_BUFFER] = NULL; + req->rsrc_nodes[IORING_RSRC_FILE] = rsrc_empty_node; + req->rsrc_nodes[IORING_RSRC_BUFFER] = rsrc_empty_node; req->task = current; req->cancel_seq_set = false; diff --git a/io_uring/notif.c b/io_uring/notif.c index 4f02e969cf08..44bf21c0f810 100644 --- a/io_uring/notif.c +++ b/io_uring/notif.c @@ -117,8 +117,8 @@ struct io_kiocb *io_alloc_notif(struct io_ring_ctx *ctx) notif->file = NULL; notif->task = current; io_get_task_refs(1); - notif->rsrc_nodes[IORING_RSRC_FILE] = NULL; - notif->rsrc_nodes[IORING_RSRC_BUFFER] = NULL; + notif->rsrc_nodes[IORING_RSRC_FILE] = rsrc_empty_node; + notif->rsrc_nodes[IORING_RSRC_BUFFER] = rsrc_empty_node; nd = io_notif_to_data(notif); nd->zc_report = false; diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index 322749c7dee9..255618335b4f 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -38,6 +38,11 @@ static const struct io_mapped_ubuf dummy_ubuf = { .len = UINT_MAX, }; +const struct io_rsrc_node empty_node = { + .type = IORING_RSRC_BUFFER, + .buf = (struct io_mapped_ubuf *) &dummy_ubuf, +}; + int __io_account_mem(struct user_struct *user, unsigned long nr_pages) { unsigned long page_limit, cur_pages, new_pages; @@ -150,7 +155,8 @@ static void io_rsrc_data_free(struct io_rsrc_data *data) for (i = 0; i < data->nr; i++) { struct io_rsrc_node *node = data->nodes[i]; - io_put_rsrc_node(node); + if (node) + io_put_rsrc_node(node); } kvfree(data->nodes); kfree(data); @@ -235,7 +241,8 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, break; } ctx->file_table.nodes[i] = node; - node->tag = tag; + if (tag) + node->tag = tag; io_fixed_file_set(node, file); io_file_bitmap_set(&ctx->file_table, i); } @@ -287,10 +294,12 @@ static int __io_sqe_buffers_update(struct io_ring_ctx *ctx, err = PTR_ERR(node); break; } - io_put_rsrc_node(ctx->user_bufs[i]); + if (ctx->user_bufs[i]) + io_put_rsrc_node(ctx->user_bufs[i]); ctx->user_bufs[i] = node; - node->tag = tag; + if (tag) + node->tag = tag; if (ctx->compat) user_data += sizeof(struct compat_iovec); else @@ -607,8 +616,10 @@ static void __io_sqe_buffers_unregister(struct io_ring_ctx *ctx) lockdep_assert_held(&ctx->uring_lock); for (i = 0; i < ctx->nr_user_bufs; i++) { - io_put_rsrc_node(ctx->user_bufs[i]); - ctx->user_bufs[i] = NULL; + if (ctx->user_bufs[i]) { + io_put_rsrc_node(ctx->user_bufs[i]); + ctx->user_bufs[i] = NULL; + } } kvfree(ctx->user_bufs); ctx->user_bufs = NULL; @@ -802,15 +813,13 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx, struct io_imu_folio_data data; bool coalesced; + if (!iov->iov_base) + return rsrc_empty_node; + node = io_rsrc_node_alloc(ctx, ctx->buf_data, index, IORING_RSRC_BUFFER); if (!node) return ERR_PTR(-ENOMEM); - if (!iov->iov_base) { - node->buf = (struct io_mapped_ubuf *) &dummy_ubuf; - return node; - } - ret = -ENOMEM; pages = io_pin_pages((unsigned long) iov->iov_base, iov->iov_len, &nr_pages); @@ -934,7 +943,8 @@ int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg, ret = PTR_ERR(node); break; } - node->tag = tag; + if (tag) + node->tag = tag; ctx->user_bufs[i] = node; } @@ -1035,18 +1045,18 @@ static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx goto out_free_data; for (i = 0; i < nbufs; i++) { - struct io_mapped_ubuf *imu = src_ctx->user_bufs[i]->buf; + struct io_rsrc_node *src_node = src_ctx->user_bufs[i]; struct io_rsrc_node *dst_node; - dst_node = io_rsrc_node_alloc(ctx, data, i, IORING_RSRC_BUFFER); - if (!dst_node) - goto out_put_free; - - if (imu == &dummy_ubuf) { - dst_node->buf = (struct io_mapped_ubuf *) &dummy_ubuf; + if (src_node == rsrc_empty_node) { + dst_node = rsrc_empty_node; } else { - refcount_inc(&imu->refs); - dst_node->buf = imu; + dst_node = io_rsrc_node_alloc(ctx, data, i, IORING_RSRC_BUFFER); + if (!dst_node) + goto out_put_free; + + refcount_inc(&src_node->buf->refs); + dst_node->buf = src_node->buf; } user_bufs[i] = dst_node; } diff --git a/io_uring/rsrc.h b/io_uring/rsrc.h index 22285169c178..69d4111d5f07 100644 --- a/io_uring/rsrc.h +++ b/io_uring/rsrc.h @@ -75,9 +75,12 @@ int io_register_rsrc_update(struct io_ring_ctx *ctx, void __user *arg, int io_register_rsrc(struct io_ring_ctx *ctx, void __user *arg, unsigned int size, unsigned int type); +extern const struct io_rsrc_node empty_node; +#define rsrc_empty_node (struct io_rsrc_node *) &empty_node + static inline void io_put_rsrc_node(struct io_rsrc_node *node) { - if (node && !--node->refs) + if (node != &empty_node && !--node->refs) io_free_rsrc_node(node); } @@ -90,8 +93,10 @@ static inline void io_req_put_rsrc_nodes(struct io_kiocb *req) static inline void io_req_assign_rsrc_node(struct io_kiocb *req, struct io_rsrc_node *node) { - node->refs++; - req->rsrc_nodes[node->type] = node; + if (node != rsrc_empty_node) { + node->refs++; + req->rsrc_nodes[node->type] = node; + } } int io_files_update(struct io_kiocb *req, unsigned int issue_flags); From patchwork Sat Oct 26 22:08:31 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jens Axboe X-Patchwork-Id: 13852383 Received: from mail-pl1-f179.google.com (mail-pl1-f179.google.com [209.85.214.179]) (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 6D0BB13D246 for ; Sat, 26 Oct 2024 22:24:05 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.179 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981447; cv=none; b=IoraH4Ingst0AsYnz1LREuS2GZ1tmZzEAa0JG1qC48X4B0kzirtfLBj4SnUtbZHpsRXcD2mxEHnn4Gs7h0wsAEJBONGqAOLM7mwE4IWWVwtTFZdB9/STDwnVCmUEH4tq4ePrj2l7+Cg7vgY7NwZsUnzqy8OYiNlBo2z3IJz2zws= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981447; c=relaxed/simple; bh=4h2usiU+oTkKcOTaeSE92V+Xzo604kYzN6f9pKvR8sc=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=Xo1JzEocjRbWgFUgR4PEaEoMZHhY+nXsCclIsQkG4ibeZ4jzmy4yF8p+oILjqfguqLGL3kPCggDnaAbtROOFkEFZNWBTzObruL2bL38IOXs8sg3+zT3Ev9h3qGuKJW3VP/Gin+j1g+Nb4Y+ROEkXm4W8/vkB2bDGBoToW9daBCU= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk; spf=pass smtp.mailfrom=kernel.dk; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b=KgESfgPj; arc=none smtp.client-ip=209.85.214.179 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=kernel.dk Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b="KgESfgPj" Received: by mail-pl1-f179.google.com with SMTP id d9443c01a7336-20cdda5cfb6so29069305ad.3 for ; Sat, 26 Oct 2024 15:24:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20230601.gappssmtp.com; s=20230601; t=1729981444; x=1730586244; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=zj0zfjZqT3t67RpWHrpfOHG+T//anNh4xzNNUBis8XY=; b=KgESfgPj7r9ucKVqn5Siz4X2t77TA2npTNfl9cpIq6qA8ZbCGiCSaFKzxoZSgbDnVu pvCZuxHOfXGSW6SqylBuMl/C6XJ97ZN7pF96BiSKOaJe5oQaQGaMTW5aXz5Fm0BDexqj zBIovYePTREDdDnmLmPn01vpAchYqavc6GsPGMenTqCrQKCJ4QBcekSwcdiJx3WUX3UJ iFRDbqpbzmYR3kqoRBLq7tgAWYOdFVEEP0vbSbjTGM3yKByJMRKwzGeAr1MME++dNg0y SxW/5t02xa6P4vnSPp5aRSTTG3SixId5AjNcTWfoi9I6ga2V1mfhhIRO9syXCT+8tolk Zt2w== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1729981444; x=1730586244; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=zj0zfjZqT3t67RpWHrpfOHG+T//anNh4xzNNUBis8XY=; b=XAp+fe35PmQHzDH3+V4m5uDgOYqfJRc8h1loA3txMzlZI9riJhpHRAqqFmglvf9Yt8 HvrFXgN/hhuVerzSbeNW7xsuwYEidGlQ8PWgfdCBH20ZXrrI2Gb8Vwjo74IOGTAuE1SK Wr6lZ7UL7UJD34ipHz4i6y38qUgz2ExhSHipokxouk3M/3/+eLYFuzHMa2l1WZkYl0X1 mjth/KncnY24f+jOr0sPKbkCrH5GTcY3BSehpliQWC6iU69wq2FRZkTQ6k+FBvxo+tG+ D6ihx5e9qLCW0WtP9VWHL9o//9aQUIYKRZ7GGZLEtCsLAhYuY+ttCw5/26LxUPb9oX6M zwVw== X-Gm-Message-State: AOJu0Yw3h2WH3v1EZUww30dOJasnsO4t0Nc0Yg0fHNJOW7CMJ78zd82y dZazv23H5wYifcYzuvNu4xhSWJBgfE04uPv0N9pt/HyQ9dWBlh5tfZ43hZEVmlg1uFN0XtSSpDP C X-Google-Smtp-Source: AGHT+IEiw2reVnl5wOfECrWVUAi9M9fs9XZrKVeGirykLyYuoRv71/F36iruj3rpJD/QTUIHiyFq1Q== X-Received: by 2002:a17:903:41c2:b0:20b:54cc:b34e with SMTP id d9443c01a7336-210c6cff16bmr55105185ad.51.1729981444333; Sat, 26 Oct 2024 15:24:04 -0700 (PDT) Received: from localhost.localdomain ([198.8.77.157]) by smtp.gmail.com with ESMTPSA id d9443c01a7336-210bbf44321sm28134705ad.30.2024.10.26.15.24.02 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 26 Oct 2024 15:24:02 -0700 (PDT) From: Jens Axboe To: io-uring@vger.kernel.org Cc: Jens Axboe Subject: [PATCH 6/7] io_uring: only initialize io_kiocb rsrc_nodes when needed Date: Sat, 26 Oct 2024 16:08:31 -0600 Message-ID: <20241026222348.90331-7-axboe@kernel.dk> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20241026222348.90331-1-axboe@kernel.dk> References: <20241026222348.90331-1-axboe@kernel.dk> Precedence: bulk X-Mailing-List: io-uring@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Add the empty node initializing to the preinit part of the io_kiocb allocation, and reset them if they have been used. Signed-off-by: Jens Axboe --- io_uring/io_uring.c | 4 ++-- io_uring/rsrc.h | 10 ++++++++-- 2 files changed, 10 insertions(+), 4 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 603fd51d170f..0956401acd26 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -947,6 +947,8 @@ void io_req_defer_failed(struct io_kiocb *req, s32 res) static void io_preinit_req(struct io_kiocb *req, struct io_ring_ctx *ctx) { req->ctx = ctx; + req->rsrc_nodes[IORING_RSRC_FILE] = rsrc_empty_node; + req->rsrc_nodes[IORING_RSRC_BUFFER] = rsrc_empty_node; req->link = NULL; req->async_data = NULL; /* not necessary, but safer to zero */ @@ -2032,8 +2034,6 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, req->flags = (__force io_req_flags_t) sqe_flags; req->cqe.user_data = READ_ONCE(sqe->user_data); req->file = NULL; - req->rsrc_nodes[IORING_RSRC_FILE] = rsrc_empty_node; - req->rsrc_nodes[IORING_RSRC_BUFFER] = rsrc_empty_node; req->task = current; req->cancel_seq_set = false; diff --git a/io_uring/rsrc.h b/io_uring/rsrc.h index 69d4111d5f07..d62d086331d2 100644 --- a/io_uring/rsrc.h +++ b/io_uring/rsrc.h @@ -86,8 +86,14 @@ static inline void io_put_rsrc_node(struct io_rsrc_node *node) static inline void io_req_put_rsrc_nodes(struct io_kiocb *req) { - io_put_rsrc_node(req->rsrc_nodes[IORING_RSRC_FILE]); - io_put_rsrc_node(req->rsrc_nodes[IORING_RSRC_BUFFER]); + if (req->rsrc_nodes[IORING_RSRC_FILE] != rsrc_empty_node) { + io_put_rsrc_node(req->rsrc_nodes[IORING_RSRC_FILE]); + req->rsrc_nodes[IORING_RSRC_FILE] = rsrc_empty_node; + } + if (req->rsrc_nodes[IORING_RSRC_BUFFER] != rsrc_empty_node) { + io_put_rsrc_node(req->rsrc_nodes[IORING_RSRC_BUFFER]); + req->rsrc_nodes[IORING_RSRC_BUFFER] = rsrc_empty_node; + } } static inline void io_req_assign_rsrc_node(struct io_kiocb *req, From patchwork Sat Oct 26 22:08:32 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jens Axboe X-Patchwork-Id: 13852384 Received: from mail-pl1-f169.google.com (mail-pl1-f169.google.com [209.85.214.169]) (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 768AF1D86C3 for ; Sat, 26 Oct 2024 22:24:07 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.214.169 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981451; cv=none; b=JtYyYgbAlM0Zpi1/ZzqaxpkZLkJOcmLxiki9TSG8HgwASzXpKYXjO2IHHs7G5cSBqeEsJ/OrrkiWAWQNzm9B3HunQZ+5e0k+daEr+gS8EZnCMO/Ut+rmWcmL/XR8twR0xzQupACLC+73VrDVH7qVdA3YukXOnvPkwBKiPwE0O7k= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1729981451; c=relaxed/simple; bh=AlPwJWblqMsJYf7MuPJH5PaZbVDEreIWG+1X6R6Vlsc=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=Rg1bOeV2pAE1i2qrwFEG2LgWWaACimYmKvOifHbRbCe9ygYerFGmH17Rpu9US8bkjWEJceQDkuuB1NQl2BiNX6Wu9m1hx5MejUZzPGh7gHHLpktZ1CMkt2N/WDxqyCE/cxGAoeVFQZIMvpjOsNCJCayUuDdyG1I7Z5aiCfkhipM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk; spf=pass smtp.mailfrom=kernel.dk; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b=zkPqXV0Z; arc=none smtp.client-ip=209.85.214.169 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=kernel.dk Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=kernel.dk Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=kernel-dk.20230601.gappssmtp.com header.i=@kernel-dk.20230601.gappssmtp.com header.b="zkPqXV0Z" Received: by mail-pl1-f169.google.com with SMTP id d9443c01a7336-20cdda5cfb6so29069395ad.3 for ; Sat, 26 Oct 2024 15:24:07 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=kernel-dk.20230601.gappssmtp.com; s=20230601; t=1729981446; x=1730586246; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=axT0ZGxKlICO0uEH9qWAOOL3daynWEtn+ApVzL5Kl4g=; b=zkPqXV0Z1Pe5YRYoycdPJxgqlp+U3uMN0jyqOAPx2iiWn725kL/jgpfAs+VUDciOgI elR3qFJqllKSyS1KueyHUpRGgt8frRxy091kSTExbcbuLmBS5UOpWEYLmGQFMACjdltR hCaabebaSFADTsE/oNVEVry6ZaS6X6MKKwKzGi3ffb74iYr46NISxD4EJYxJOgzhiDnp H2Py9joe7HN8u+kWPkiaoPe4w+7X3rWVjncCzOapzNTeHQVLzwdciv0m2ru/hwF76PRv gRda9z0n8FuEysHnPEMg7hcIhDbiJ9tDJ7oORTkbFqP+4BCdWBOviUKvUP+CBHOXM2yl xKFQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1729981446; x=1730586246; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=axT0ZGxKlICO0uEH9qWAOOL3daynWEtn+ApVzL5Kl4g=; b=SQIErKk+PuVKp2ABg7X7N2xRPIMddYR+A46BxFIAs95GmYGMr8oxBK+lAt5H3Z6hX1 aUVsCR59xQXlit0VGbHCF5DnhE72n4oYezNDJGJnRNs2rUaLd/GQ2RLw+IKggEGI17BW BNjgEgrUNDuWtCvEsgWI/twIYOZEkGQi9+V0vemqldUAc+berK6w+fECXKovuzDLWlRG KVIMaTwhmA6W9ATVsPRs6jr8doGrSi1r6oOZ2r0qv8kC2KLqwW0DMktQqHETh8bJRiwo IJu/+fUhe4wQv8XgnpLNhbKBNhOBknUB58FmlOFTTJmvHEYWsbuZknpEtlrfucmMJumu djQg== X-Gm-Message-State: AOJu0Yz8s6lFD6T1DuIWEHOJZGiZWAjSLgJm854Q+Esfpe0qlnXmHMJ1 JN+jkLuz1zM8CH5k7SBRwgNc6iZ8sNF8DRhnJcxZkcMnBNCwQo1NP2f6ewQjPCWgrHHNWl4mmfv n X-Google-Smtp-Source: AGHT+IFdSTaS5BpkOQ1UElF9QvT0rJNTlLhMY/fqTZblI9XZ+jv0kXEhUG9M9SdFszN5c7gYiXOFFg== X-Received: by 2002:a17:902:f78a:b0:20b:70b4:69d8 with SMTP id d9443c01a7336-210c6cd3b79mr39312175ad.37.1729981446024; Sat, 26 Oct 2024 15:24:06 -0700 (PDT) Received: from localhost.localdomain ([198.8.77.157]) by smtp.gmail.com with ESMTPSA id d9443c01a7336-210bbf44321sm28134705ad.30.2024.10.26.15.24.04 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sat, 26 Oct 2024 15:24:04 -0700 (PDT) From: Jens Axboe To: io-uring@vger.kernel.org Cc: Jens Axboe Subject: [PATCH 7/7] io_uring/rsrc: unify file and buffer resource tables Date: Sat, 26 Oct 2024 16:08:32 -0600 Message-ID: <20241026222348.90331-8-axboe@kernel.dk> X-Mailer: git-send-email 2.45.2 In-Reply-To: <20241026222348.90331-1-axboe@kernel.dk> References: <20241026222348.90331-1-axboe@kernel.dk> Precedence: bulk X-Mailing-List: io-uring@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 For files, there's nr_user_files/file_table/file_data, and buffers have nr_user_bufs/user_bufs/buf_data. There's no reason why file_table and file_data can't be the same thing, and ditto for the buffer side. That gets rid of more io_ring_ctx state that's in two spots rather than just being in one spot, as it should be. Put all the registered file data in one locations, and ditto on the buffer front. This also avoids having both io_rsrc_data->nodes being an allocated array, and ->user_bufs[] or ->file_table.nodes. There's no reason to have this information duplicated. Keep it in one spot, io_rsrc_data, along with how many resources are available. Signed-off-by: Jens Axboe --- include/linux/io_uring_types.h | 15 ++- io_uring/cancel.c | 4 +- io_uring/fdinfo.c | 10 +- io_uring/filetable.c | 47 ++++---- io_uring/filetable.h | 2 +- io_uring/io_uring.c | 7 +- io_uring/msg_ring.c | 4 +- io_uring/net.c | 6 +- io_uring/opdef.c | 2 + io_uring/register.c | 3 +- io_uring/rsrc.c | 214 ++++++++++----------------------- io_uring/rsrc.h | 17 ++- io_uring/rw.c | 6 +- io_uring/splice.c | 42 +++++-- io_uring/splice.h | 1 + io_uring/uring_cmd.c | 6 +- 16 files changed, 165 insertions(+), 221 deletions(-) diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index 60541da6b875..e8d702b3757f 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -55,8 +55,13 @@ struct io_wq_work { int cancel_seq; }; +struct io_rsrc_data { + unsigned int nr; + struct io_rsrc_node **nodes; +}; + struct io_file_table { - struct io_rsrc_node **nodes; + struct io_rsrc_data data; unsigned long *bitmap; unsigned int alloc_hint; }; @@ -276,9 +281,7 @@ struct io_ring_ctx { struct io_wq_work_list iopoll_list; struct io_file_table file_table; - struct io_rsrc_node **user_bufs; - unsigned nr_user_files; - unsigned nr_user_bufs; + struct io_rsrc_data buf_table; struct io_submit_state submit_state; @@ -358,10 +361,6 @@ struct io_ring_ctx { struct wait_queue_head poll_wq; struct io_restriction restrictions; - /* slow path rsrc auxilary data, used by update/register */ - struct io_rsrc_data *file_data; - struct io_rsrc_data *buf_data; - u32 pers_next; struct xarray personalities; diff --git a/io_uring/cancel.c b/io_uring/cancel.c index cc3475b22ae5..3a2996307025 100644 --- a/io_uring/cancel.c +++ b/io_uring/cancel.c @@ -240,9 +240,9 @@ static int __io_sync_cancel(struct io_uring_task *tctx, /* fixed must be grabbed every time since we drop the uring_lock */ if ((cd->flags & IORING_ASYNC_CANCEL_FD) && (cd->flags & IORING_ASYNC_CANCEL_FD_FIXED)) { - if (unlikely(fd >= ctx->nr_user_files)) + if (unlikely(fd >= ctx->file_table.data.nr)) return -EBADF; - fd = array_index_nospec(fd, ctx->nr_user_files); + fd = array_index_nospec(fd, ctx->file_table.data.nr); cd->file = io_file_from_index(&ctx->file_table, fd); if (!cd->file) return -EBADF; diff --git a/io_uring/fdinfo.c b/io_uring/fdinfo.c index 064a79475c5f..e3f5e9fe5562 100644 --- a/io_uring/fdinfo.c +++ b/io_uring/fdinfo.c @@ -165,8 +165,8 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) seq_printf(m, "SqThreadCpu:\t%d\n", sq_cpu); seq_printf(m, "SqTotalTime:\t%llu\n", sq_total_time); seq_printf(m, "SqWorkTime:\t%llu\n", sq_work_time); - seq_printf(m, "UserFiles:\t%u\n", ctx->nr_user_files); - for (i = 0; has_lock && i < ctx->nr_user_files; i++) { + seq_printf(m, "UserFiles:\t%u\n", ctx->file_table.data.nr); + for (i = 0; has_lock && i < ctx->file_table.data.nr; i++) { struct file *f = io_file_from_index(&ctx->file_table, i); if (f) @@ -174,9 +174,9 @@ __cold void io_uring_show_fdinfo(struct seq_file *m, struct file *file) else seq_printf(m, "%5u: \n", i); } - seq_printf(m, "UserBufs:\t%u\n", ctx->nr_user_bufs); - for (i = 0; has_lock && i < ctx->nr_user_bufs; i++) { - struct io_mapped_ubuf *buf = ctx->user_bufs[i]->buf; + seq_printf(m, "UserBufs:\t%u\n", ctx->buf_table.nr); + for (i = 0; has_lock && i < ctx->buf_table.nr; i++) { + struct io_mapped_ubuf *buf = ctx->buf_table.nodes[i]->buf; seq_printf(m, "%5u: 0x%llx/%u\n", i, buf->ubuf, buf->len); } diff --git a/io_uring/filetable.c b/io_uring/filetable.c index c18e37b495a1..c4c31f8b4f5d 100644 --- a/io_uring/filetable.c +++ b/io_uring/filetable.c @@ -38,25 +38,19 @@ static int io_file_bitmap_get(struct io_ring_ctx *ctx) bool io_alloc_file_tables(struct io_file_table *table, unsigned nr_files) { - table->nodes = kvmalloc_array(nr_files, sizeof(struct io_src_node *), - GFP_KERNEL_ACCOUNT | __GFP_ZERO); - if (unlikely(!table->nodes)) + if (io_rsrc_data_alloc(&table->data, nr_files)) return false; - table->bitmap = bitmap_zalloc(nr_files, GFP_KERNEL_ACCOUNT); - if (unlikely(!table->bitmap)) { - kvfree(table->nodes); - return false; - } - - return true; + if (table->bitmap) + return true; + io_rsrc_data_free(&table->data); + return false; } void io_free_file_tables(struct io_file_table *table) { - kvfree(table->nodes); + io_rsrc_data_free(&table->data); bitmap_free(table->bitmap); - table->nodes = NULL; table->bitmap = NULL; } @@ -68,22 +62,23 @@ static int io_install_fixed_file(struct io_ring_ctx *ctx, struct file *file, if (io_is_uring_fops(file)) return -EBADF; - if (!ctx->file_data) + if (!ctx->file_table.data.nr) return -ENXIO; - if (slot_index >= ctx->nr_user_files) + if (slot_index >= ctx->file_table.data.nr) return -EINVAL; - node = io_rsrc_node_alloc(ctx, ctx->file_data, slot_index, IORING_RSRC_FILE); + node = io_rsrc_node_alloc(ctx, &ctx->file_table.data, slot_index, + IORING_RSRC_FILE); if (IS_ERR(node)) return -ENOMEM; - slot_index = array_index_nospec(slot_index, ctx->nr_user_files); - if (ctx->file_table.nodes[slot_index]) - io_put_rsrc_node(ctx->file_table.nodes[slot_index]); + slot_index = array_index_nospec(slot_index, ctx->file_table.data.nr); + if (ctx->file_table.data.nodes[slot_index]) + io_put_rsrc_node(ctx->file_table.data.nodes[slot_index]); else io_file_bitmap_set(&ctx->file_table, slot_index); - ctx->file_table.nodes[slot_index] = node; + ctx->file_table.data.nodes[slot_index] = node; io_fixed_file_set(node, file); return 0; } @@ -129,16 +124,16 @@ int io_fixed_fd_install(struct io_kiocb *req, unsigned int issue_flags, int io_fixed_fd_remove(struct io_ring_ctx *ctx, unsigned int offset) { - if (unlikely(!ctx->file_data)) + if (unlikely(!ctx->file_table.data.nr)) return -ENXIO; - if (offset >= ctx->nr_user_files) + if (offset >= ctx->file_table.data.nr) return -EINVAL; - offset = array_index_nospec(offset, ctx->nr_user_files); - if (!ctx->file_table.nodes[offset]) + offset = array_index_nospec(offset, ctx->file_table.data.nr); + if (!ctx->file_table.data.nodes[offset]) return -EBADF; - io_put_rsrc_node(ctx->file_table.nodes[offset]); - ctx->file_table.nodes[offset] = NULL; + io_put_rsrc_node(ctx->file_table.data.nodes[offset]); + ctx->file_table.data.nodes[offset] = NULL; io_file_bitmap_clear(&ctx->file_table, offset); return 0; } @@ -153,7 +148,7 @@ int io_register_file_alloc_range(struct io_ring_ctx *ctx, return -EFAULT; if (check_add_overflow(range.off, range.len, &end)) return -EOVERFLOW; - if (range.resv || end > ctx->nr_user_files) + if (range.resv || end > ctx->file_table.data.nr) return -EINVAL; io_file_table_set_alloc_range(ctx, range.off, range.len); diff --git a/io_uring/filetable.h b/io_uring/filetable.h index 47616079abaa..664c31502dbb 100644 --- a/io_uring/filetable.h +++ b/io_uring/filetable.h @@ -52,7 +52,7 @@ static inline struct file *io_slot_file(struct io_rsrc_node *node) static inline struct file *io_file_from_index(struct io_file_table *table, int index) { - struct io_rsrc_node *node = table->nodes[index]; + struct io_rsrc_node *node = table->data.nodes[index]; if (node) return io_slot_file(node); diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 0956401acd26..5842f27d5bdf 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -1879,17 +1879,12 @@ inline struct file *io_file_get_fixed(struct io_kiocb *req, int fd, struct file *file = NULL; io_ring_submit_lock(ctx, issue_flags); - - if (unlikely((unsigned int)fd >= ctx->nr_user_files)) - goto out; - fd = array_index_nospec(fd, ctx->nr_user_files); - node = ctx->file_table.nodes[fd]; + node = io_rsrc_node_lookup(&ctx->file_table.data, fd); if (node) { io_req_assign_rsrc_node(req, node); req->flags |= io_slot_flags(node); file = io_slot_file(node); } -out: io_ring_submit_unlock(ctx, issue_flags); return file; } diff --git a/io_uring/msg_ring.c b/io_uring/msg_ring.c index edea1ffd501c..b90ab3b8f5e0 100644 --- a/io_uring/msg_ring.c +++ b/io_uring/msg_ring.c @@ -180,8 +180,8 @@ static struct file *io_msg_grab_file(struct io_kiocb *req, unsigned int issue_fl int idx = msg->src_fd; io_ring_submit_lock(ctx, issue_flags); - if (likely(idx < ctx->nr_user_files)) { - idx = array_index_nospec(idx, ctx->nr_user_files); + if (likely(idx < ctx->file_table.data.nr)) { + idx = array_index_nospec(idx, ctx->file_table.data.nr); file = io_file_from_index(&ctx->file_table, idx); if (file) get_file(file); diff --git a/io_uring/net.c b/io_uring/net.c index ce1156551d10..3e1f31574abb 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -1347,9 +1347,9 @@ static int io_send_zc_import(struct io_kiocb *req, unsigned int issue_flags) ret = -EFAULT; io_ring_submit_lock(ctx, issue_flags); - if (sr->buf_index < ctx->nr_user_bufs) { - idx = array_index_nospec(sr->buf_index, ctx->nr_user_bufs); - node = ctx->user_bufs[idx]; + if (sr->buf_index < ctx->buf_table.nr) { + idx = array_index_nospec(sr->buf_index, ctx->buf_table.nr); + node = ctx->buf_table.nodes[idx]; io_req_assign_rsrc_node(sr->notif, node); ret = 0; } diff --git a/io_uring/opdef.c b/io_uring/opdef.c index a2be3bbca5ff..3de75eca1c92 100644 --- a/io_uring/opdef.c +++ b/io_uring/opdef.c @@ -641,6 +641,7 @@ const struct io_cold_def io_cold_defs[] = { }, [IORING_OP_SPLICE] = { .name = "SPLICE", + .cleanup = io_splice_cleanup, }, [IORING_OP_PROVIDE_BUFFERS] = { .name = "PROVIDE_BUFFERS", @@ -650,6 +651,7 @@ const struct io_cold_def io_cold_defs[] = { }, [IORING_OP_TEE] = { .name = "TEE", + .cleanup = io_splice_cleanup, }, [IORING_OP_SHUTDOWN] = { .name = "SHUTDOWN", diff --git a/io_uring/register.c b/io_uring/register.c index fc6c94d694b2..3c5a3cfb186b 100644 --- a/io_uring/register.c +++ b/io_uring/register.c @@ -855,7 +855,8 @@ SYSCALL_DEFINE4(io_uring_register, unsigned int, fd, unsigned int, opcode, mutex_lock(&ctx->uring_lock); ret = __io_uring_register(ctx, opcode, arg, nr_args); mutex_unlock(&ctx->uring_lock); - trace_io_uring_register(ctx, opcode, ctx->nr_user_files, ctx->nr_user_bufs, ret); + trace_io_uring_register(ctx, opcode, ctx->file_table.data.nr, + ctx->buf_table.nr, ret); if (!use_registered_ring) fput(file); return ret; diff --git a/io_uring/rsrc.c b/io_uring/rsrc.c index 255618335b4f..20f110877e9c 100644 --- a/io_uring/rsrc.c +++ b/io_uring/rsrc.c @@ -148,39 +148,28 @@ struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx, return node; } -static void io_rsrc_data_free(struct io_rsrc_data *data) +__cold void io_rsrc_data_free(struct io_rsrc_data *data) { - int i; - - for (i = 0; i < data->nr; i++) { - struct io_rsrc_node *node = data->nodes[i]; - - if (node) - io_put_rsrc_node(node); + if (!data->nr) + return; + while (data->nr--) { + if (data->nodes[data->nr]) + io_put_rsrc_node(data->nodes[data->nr]); } kvfree(data->nodes); - kfree(data); + data->nodes = NULL; + data->nr = 0; } -__cold static int io_rsrc_data_alloc(struct io_ring_ctx *ctx, unsigned nr, - struct io_rsrc_data **pdata) +__cold int io_rsrc_data_alloc(struct io_rsrc_data *data, unsigned nr) { - struct io_rsrc_data *data; - - data = kzalloc(sizeof(*data), GFP_KERNEL); - if (!data) - return -ENOMEM; - data->nodes = kvmalloc_array(nr, sizeof(struct io_rsrc_node *), - GFP_KERNEL | __GFP_ZERO); - if (!data->nodes) { - io_rsrc_data_free(data); - return -ENOMEM; + GFP_KERNEL_ACCOUNT | __GFP_ZERO); + if (data->nodes) { + data->nr = nr; + return 0; } - - data->nr = nr; - *pdata = data; - return 0; + return -ENOMEM; } static int __io_sqe_files_update(struct io_ring_ctx *ctx, @@ -192,9 +181,9 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, int fd, i, err = 0; unsigned int done; - if (!ctx->file_data) + if (!ctx->file_table.data.nr) return -ENXIO; - if (up->offset + nr_args > ctx->nr_user_files) + if (up->offset + nr_args > ctx->file_table.data.nr) return -EINVAL; for (done = 0; done < nr_args; done++) { @@ -212,10 +201,10 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, if (fd == IORING_REGISTER_FILES_SKIP) continue; - i = array_index_nospec(up->offset + done, ctx->nr_user_files); - if (ctx->file_table.nodes[i]) { - io_put_rsrc_node(ctx->file_table.nodes[i]); - ctx->file_table.nodes[i] = NULL; + i = array_index_nospec(up->offset + done, ctx->file_table.data.nr); + if (ctx->file_table.data.nodes[i]) { + io_put_rsrc_node(ctx->file_table.data.nodes[i]); + ctx->file_table.data.nodes[i] = NULL; io_file_bitmap_clear(&ctx->file_table, i); } if (fd != -1) { @@ -234,13 +223,14 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, err = -EBADF; break; } - node = io_rsrc_node_alloc(ctx, ctx->file_data, i, IORING_RSRC_FILE); + node = io_rsrc_node_alloc(ctx, &ctx->file_table.data, i, + IORING_RSRC_FILE); if (!node) { err = -ENOMEM; fput(file); break; } - ctx->file_table.nodes[i] = node; + ctx->file_table.data.nodes[i] = node; if (tag) node->tag = tag; io_fixed_file_set(node, file); @@ -262,9 +252,9 @@ static int __io_sqe_buffers_update(struct io_ring_ctx *ctx, __u32 done; int i, err; - if (!ctx->buf_data) + if (!ctx->buf_table.nr) return -ENXIO; - if (up->offset + nr_args > ctx->nr_user_bufs) + if (up->offset + nr_args > ctx->buf_table.nr) return -EINVAL; for (done = 0; done < nr_args; done++) { @@ -288,16 +278,16 @@ static int __io_sqe_buffers_update(struct io_ring_ctx *ctx, err = -EINVAL; break; } - i = array_index_nospec(up->offset + done, ctx->nr_user_bufs); + i = array_index_nospec(up->offset + done, ctx->buf_table.nr); node = io_sqe_buffer_register(ctx, iov, i, &last_hpage); if (IS_ERR(node)) { err = PTR_ERR(node); break; } - if (ctx->user_bufs[i]) - io_put_rsrc_node(ctx->user_bufs[i]); + if (ctx->buf_table.nodes[i]) + io_put_rsrc_node(ctx->buf_table.nodes[i]); - ctx->user_bufs[i] = node; + ctx->buf_table.nodes[i] = node; if (tag) node->tag = tag; if (ctx->compat) @@ -415,7 +405,7 @@ static int io_files_update_with_index_alloc(struct io_kiocb *req, struct file *file; int ret, fd; - if (!req->ctx->file_data) + if (!req->ctx->file_table.data.nr) return -ENXIO; for (done = 0; done < up->nr_args; done++) { @@ -500,35 +490,13 @@ void io_free_rsrc_node(struct io_rsrc_node *node) kfree(node); } -static void __io_sqe_files_unregister(struct io_ring_ctx *ctx) -{ - int i; - - lockdep_assert_held(&ctx->uring_lock); - - for (i = 0; i < ctx->nr_user_files; i++) { - struct io_rsrc_node *node = ctx->file_table.nodes[i]; - - if (node) { - io_put_rsrc_node(node); - io_file_bitmap_clear(&ctx->file_table, i); - ctx->file_table.nodes[i] = NULL; - } - } - - io_free_file_tables(&ctx->file_table); - io_file_table_set_alloc_range(ctx, 0, 0); - io_rsrc_data_free(ctx->file_data); - ctx->file_data = NULL; - ctx->nr_user_files = 0; -} - int io_sqe_files_unregister(struct io_ring_ctx *ctx) { - if (!ctx->file_data) + if (!ctx->file_table.data.nr) return -ENXIO; - __io_sqe_files_unregister(ctx); + io_free_file_tables(&ctx->file_table); + io_file_table_set_alloc_range(ctx, 0, 0); return 0; } @@ -540,7 +508,7 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, int fd, ret; unsigned i; - if (ctx->file_data) + if (ctx->file_table.data.nr) return -EBUSY; if (!nr_args) return -EINVAL; @@ -548,17 +516,10 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, return -EMFILE; if (nr_args > rlimit(RLIMIT_NOFILE)) return -EMFILE; - ret = io_rsrc_data_alloc(ctx, nr_args, &ctx->file_data); - if (ret) - return ret; - - if (!io_alloc_file_tables(&ctx->file_table, nr_args)) { - io_rsrc_data_free(ctx->file_data); - ctx->file_data = NULL; + if (!io_alloc_file_tables(&ctx->file_table, nr_args)) return -ENOMEM; - } - for (i = 0; i < nr_args; i++, ctx->nr_user_files++) { + for (i = 0; i < nr_args; i++) { struct io_rsrc_node *node; u64 tag = 0; @@ -572,7 +533,7 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, ret = -EINVAL; if (tag) goto fail; - ctx->file_table.nodes[i] = NULL; + ctx->file_table.data.nodes[i] = NULL; continue; } @@ -589,51 +550,32 @@ int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, goto fail; } ret = -ENOMEM; - node = io_rsrc_node_alloc(ctx, ctx->file_data, i, IORING_RSRC_FILE); + node = io_rsrc_node_alloc(ctx, &ctx->file_table.data, i, + IORING_RSRC_FILE); if (!node) { fput(file); goto fail; } if (tag) node->tag = tag; - ctx->file_table.nodes[i] = node; + ctx->file_table.data.nodes[i] = node; io_fixed_file_set(node, file); io_file_bitmap_set(&ctx->file_table, i); } /* default it to the whole table */ - io_file_table_set_alloc_range(ctx, 0, ctx->nr_user_files); + io_file_table_set_alloc_range(ctx, 0, ctx->file_table.data.nr); return 0; fail: - __io_sqe_files_unregister(ctx); + io_sqe_files_unregister(ctx); return ret; } -static void __io_sqe_buffers_unregister(struct io_ring_ctx *ctx) -{ - unsigned int i; - - lockdep_assert_held(&ctx->uring_lock); - - for (i = 0; i < ctx->nr_user_bufs; i++) { - if (ctx->user_bufs[i]) { - io_put_rsrc_node(ctx->user_bufs[i]); - ctx->user_bufs[i] = NULL; - } - } - kvfree(ctx->user_bufs); - ctx->user_bufs = NULL; - io_rsrc_data_free(ctx->buf_data); - ctx->buf_data = NULL; - ctx->nr_user_bufs = 0; -} - int io_sqe_buffers_unregister(struct io_ring_ctx *ctx) { - if (!ctx->buf_data) + if (!ctx->buf_table.nr) return -ENXIO; - - __io_sqe_buffers_unregister(ctx); + io_rsrc_data_free(&ctx->buf_table); return 0; } @@ -660,8 +602,8 @@ static bool headpage_already_acct(struct io_ring_ctx *ctx, struct page **pages, } /* check previously registered pages */ - for (i = 0; i < ctx->nr_user_bufs; i++) { - struct io_rsrc_node *node = ctx->user_bufs[i]; + for (i = 0; i < ctx->buf_table.nr; i++) { + struct io_rsrc_node *node = ctx->buf_table.nodes[i]; struct io_mapped_ubuf *imu = node->buf; for (j = 0; j < imu->nr_bvecs; j++) { @@ -816,7 +758,7 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx, if (!iov->iov_base) return rsrc_empty_node; - node = io_rsrc_node_alloc(ctx, ctx->buf_data, index, IORING_RSRC_BUFFER); + node = io_rsrc_node_alloc(ctx, &ctx->buf_table, index, IORING_RSRC_BUFFER); if (!node) return ERR_PTR(-ENOMEM); @@ -874,40 +816,29 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx, return node; } -static int io_buffers_map_alloc(struct io_ring_ctx *ctx, unsigned int nr_args) -{ - ctx->user_bufs = kcalloc(nr_args, sizeof(*ctx->user_bufs), GFP_KERNEL); - return ctx->user_bufs ? 0 : -ENOMEM; -} - int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg, unsigned int nr_args, u64 __user *tags) { struct page *last_hpage = NULL; - struct io_rsrc_data *data; + struct io_rsrc_data data; struct iovec fast_iov, *iov = &fast_iov; const struct iovec __user *uvec; int i, ret; BUILD_BUG_ON(IORING_MAX_REG_BUFFERS >= (1u << 16)); - if (ctx->user_bufs) + if (ctx->buf_table.nr) return -EBUSY; if (!nr_args || nr_args > IORING_MAX_REG_BUFFERS) return -EINVAL; - ret = io_rsrc_data_alloc(ctx, nr_args, &data); + ret = io_rsrc_data_alloc(&data, nr_args); if (ret) return ret; - ret = io_buffers_map_alloc(ctx, nr_args); - if (ret) { - io_rsrc_data_free(data); - return ret; - } if (!arg) memset(iov, 0, sizeof(*iov)); - for (i = 0; i < nr_args; i++, ctx->nr_user_bufs++) { + for (i = 0; i < nr_args; i++) { struct io_rsrc_node *node; u64 tag = 0; @@ -945,14 +876,12 @@ int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg, } if (tag) node->tag = tag; - ctx->user_bufs[i] = node; + data.nodes[i] = node; } - WARN_ON_ONCE(ctx->buf_data); - - ctx->buf_data = data; + ctx->buf_table = data; if (ret) - __io_sqe_buffers_unregister(ctx); + io_sqe_buffers_unregister(ctx); return ret; } @@ -1019,8 +948,7 @@ int io_import_fixed(int ddir, struct iov_iter *iter, static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx) { - struct io_rsrc_node **user_bufs; - struct io_rsrc_data *data; + struct io_rsrc_data data; int i, ret, nbufs; /* @@ -1031,43 +959,35 @@ static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx mutex_lock(&src_ctx->uring_lock); ret = -ENXIO; - nbufs = src_ctx->nr_user_bufs; + nbufs = src_ctx->buf_table.nr; if (!nbufs) goto out_unlock; - ret = io_rsrc_data_alloc(ctx, nbufs, &data); + ret = io_rsrc_data_alloc(&data, nbufs); if (ret) goto out_unlock; - ret = -ENOMEM; - user_bufs = kvmalloc_array(nbufs, sizeof(struct io_rsrc_node *), - GFP_KERNEL); - if (!user_bufs) - goto out_free_data; - for (i = 0; i < nbufs; i++) { - struct io_rsrc_node *src_node = src_ctx->user_bufs[i]; + struct io_rsrc_node *src_node = src_ctx->buf_table.nodes[i]; struct io_rsrc_node *dst_node; if (src_node == rsrc_empty_node) { dst_node = rsrc_empty_node; } else { - dst_node = io_rsrc_node_alloc(ctx, data, i, IORING_RSRC_BUFFER); + dst_node = io_rsrc_node_alloc(ctx, &data, i, IORING_RSRC_BUFFER); if (!dst_node) goto out_put_free; refcount_inc(&src_node->buf->refs); dst_node->buf = src_node->buf; } - user_bufs[i] = dst_node; + data.nodes[i] = dst_node; } /* Have a ref on the bufs now, drop src lock and re-grab our own lock */ mutex_unlock(&src_ctx->uring_lock); mutex_lock(&ctx->uring_lock); - if (!ctx->user_bufs) { - ctx->user_bufs = user_bufs; - ctx->buf_data = data; - ctx->nr_user_bufs = nbufs; + if (!ctx->buf_table.nr) { + ctx->buf_table = data; return 0; } @@ -1078,12 +998,10 @@ static int io_clone_buffers(struct io_ring_ctx *ctx, struct io_ring_ctx *src_ctx i = nbufs; out_put_free: while (i--) { - io_buffer_unmap(src_ctx, user_bufs[i]); - kfree(user_bufs[i]); + io_buffer_unmap(src_ctx, data.nodes[i]); + kfree(data.nodes[i]); } - kvfree(user_bufs); -out_free_data: - io_rsrc_data_free(data); + io_rsrc_data_free(&data); out_unlock: mutex_unlock(&src_ctx->uring_lock); mutex_lock(&ctx->uring_lock); @@ -1104,7 +1022,7 @@ int io_register_clone_buffers(struct io_ring_ctx *ctx, void __user *arg) struct file *file; int ret; - if (ctx->user_bufs || ctx->nr_user_bufs) + if (ctx->buf_table.nr) return -EBUSY; if (copy_from_user(&buf, arg, sizeof(buf))) return -EFAULT; diff --git a/io_uring/rsrc.h b/io_uring/rsrc.h index d62d086331d2..f9f2b254b0c6 100644 --- a/io_uring/rsrc.h +++ b/io_uring/rsrc.h @@ -14,11 +14,6 @@ enum { IORING_RSRC_INVALID = 2, }; -struct io_rsrc_data { - unsigned int nr; - struct io_rsrc_node **nodes; -}; - struct io_rsrc_node { struct io_ring_ctx *ctx; int refs; @@ -55,6 +50,8 @@ struct io_rsrc_node *io_rsrc_node_alloc(struct io_ring_ctx *ctx, struct io_rsrc_data *data, int index, int type); void io_free_rsrc_node(struct io_rsrc_node *node); +void io_rsrc_data_free(struct io_rsrc_data *data); +int io_rsrc_data_alloc(struct io_rsrc_data *data, unsigned nr); int io_import_fixed(int ddir, struct iov_iter *iter, struct io_mapped_ubuf *imu, @@ -78,6 +75,16 @@ int io_register_rsrc(struct io_ring_ctx *ctx, void __user *arg, extern const struct io_rsrc_node empty_node; #define rsrc_empty_node (struct io_rsrc_node *) &empty_node +static inline struct io_rsrc_node *io_rsrc_node_lookup(struct io_rsrc_data *data, + int index) +{ + if (index < data->nr) { + index = array_index_nospec(index, data->nr); + return data->nodes[index]; + } + return NULL; +} + static inline void io_put_rsrc_node(struct io_rsrc_node *node) { if (node != &empty_node && !--node->refs) diff --git a/io_uring/rw.c b/io_uring/rw.c index 65491f4f2c7e..28fff18ebb19 100644 --- a/io_uring/rw.c +++ b/io_uring/rw.c @@ -339,10 +339,10 @@ static int io_prep_rw_fixed(struct io_kiocb *req, const struct io_uring_sqe *sqe if (unlikely(ret)) return ret; - if (unlikely(req->buf_index >= ctx->nr_user_bufs)) + if (unlikely(req->buf_index >= ctx->buf_table.nr)) return -EFAULT; - index = array_index_nospec(req->buf_index, ctx->nr_user_bufs); - node = ctx->user_bufs[index]; + index = array_index_nospec(req->buf_index, ctx->buf_table.nr); + node = ctx->buf_table.nodes[index]; io_req_assign_rsrc_node(req, node); io = req->async_data; diff --git a/io_uring/splice.c b/io_uring/splice.c index 3b659cd23e9d..deeb8bb18651 100644 --- a/io_uring/splice.c +++ b/io_uring/splice.c @@ -21,6 +21,7 @@ struct io_splice { u64 len; int splice_fd_in; unsigned int flags; + struct io_rsrc_node *rsrc_node; }; static int __io_splice_prep(struct io_kiocb *req, @@ -34,6 +35,7 @@ static int __io_splice_prep(struct io_kiocb *req, if (unlikely(sp->flags & ~valid_flags)) return -EINVAL; sp->splice_fd_in = READ_ONCE(sqe->splice_fd_in); + sp->rsrc_node = rsrc_empty_node; req->flags |= REQ_F_FORCE_ASYNC; return 0; } @@ -45,6 +47,36 @@ int io_tee_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return __io_splice_prep(req, sqe); } +void io_splice_cleanup(struct io_kiocb *req) +{ + struct io_splice *sp = io_kiocb_to_cmd(req, struct io_splice); + + io_put_rsrc_node(sp->rsrc_node); +} + +static struct file *io_splice_get_file(struct io_kiocb *req, + unsigned int issue_flags) +{ + struct io_splice *sp = io_kiocb_to_cmd(req, struct io_splice); + struct io_ring_ctx *ctx = req->ctx; + struct io_rsrc_node *node; + struct file *file = NULL; + + if (!(sp->flags & SPLICE_F_FD_IN_FIXED)) + return io_file_get_normal(req, sp->splice_fd_in); + + io_ring_submit_lock(ctx, issue_flags); + node = io_rsrc_node_lookup(&ctx->file_table.data, sp->splice_fd_in); + if (node) { + node->refs++; + sp->rsrc_node = node; + file = io_slot_file(node); + req->flags |= REQ_F_NEED_CLEANUP; + } + io_ring_submit_unlock(ctx, issue_flags); + return file; +} + int io_tee(struct io_kiocb *req, unsigned int issue_flags) { struct io_splice *sp = io_kiocb_to_cmd(req, struct io_splice); @@ -55,10 +87,7 @@ int io_tee(struct io_kiocb *req, unsigned int issue_flags) WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK); - if (sp->flags & SPLICE_F_FD_IN_FIXED) - in = io_file_get_fixed(req, sp->splice_fd_in, issue_flags); - else - in = io_file_get_normal(req, sp->splice_fd_in); + in = io_splice_get_file(req, issue_flags); if (!in) { ret = -EBADF; goto done; @@ -96,10 +125,7 @@ int io_splice(struct io_kiocb *req, unsigned int issue_flags) WARN_ON_ONCE(issue_flags & IO_URING_F_NONBLOCK); - if (sp->flags & SPLICE_F_FD_IN_FIXED) - in = io_file_get_fixed(req, sp->splice_fd_in, issue_flags); - else - in = io_file_get_normal(req, sp->splice_fd_in); + in = io_splice_get_file(req, issue_flags); if (!in) { ret = -EBADF; goto done; diff --git a/io_uring/splice.h b/io_uring/splice.h index 542f94168ad3..b9b2848327fb 100644 --- a/io_uring/splice.h +++ b/io_uring/splice.h @@ -3,5 +3,6 @@ int io_tee_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe); int io_tee(struct io_kiocb *req, unsigned int issue_flags); +void io_splice_cleanup(struct io_kiocb *req); int io_splice_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe); int io_splice(struct io_kiocb *req, unsigned int issue_flags); diff --git a/io_uring/uring_cmd.c b/io_uring/uring_cmd.c index 0899c71008ae..17d5f5004702 100644 --- a/io_uring/uring_cmd.c +++ b/io_uring/uring_cmd.c @@ -212,15 +212,15 @@ int io_uring_cmd_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) u16 index; index = READ_ONCE(sqe->buf_index); - if (unlikely(index >= ctx->nr_user_bufs)) + if (unlikely(index >= ctx->buf_table.nr)) return -EFAULT; - req->buf_index = array_index_nospec(index, ctx->nr_user_bufs); + req->buf_index = array_index_nospec(index, ctx->buf_table.nr); /* * Pi node upfront, prior to io_uring_cmd_import_fixed() * being called. This prevents destruction of the mapped buffer * we'll need at actual import time. */ - io_req_assign_rsrc_node(req, ctx->user_bufs[req->buf_index]); + io_req_assign_rsrc_node(req, ctx->buf_table.nodes[req->buf_index]); } ioucmd->cmd_op = READ_ONCE(sqe->cmd_op);