From patchwork Fri Jun 2 16:08:59 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265574 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 350F3C7EE2A for ; Fri, 2 Jun 2023 16:09:28 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236496AbjFBQJ0 (ORCPT ); Fri, 2 Jun 2023 12:09:26 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34896 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236500AbjFBQJY (ORCPT ); Fri, 2 Jun 2023 12:09:24 -0400 Received: from mail-pf1-x449.google.com (mail-pf1-x449.google.com [IPv6:2607:f8b0:4864:20::449]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 471A61BD for ; Fri, 2 Jun 2023 09:09:22 -0700 (PDT) Received: by mail-pf1-x449.google.com with SMTP id d2e1a72fcca58-6533f8d4c29so683610b3a.0 for ; Fri, 02 Jun 2023 09:09:22 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722162; x=1688314162; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=DpCi+Z+5IrlXYT4x1uPx6ey3Kh6k4RwU4EEsfyAeFCY=; b=HfcMR6iJ3p+XpoWwwGnuCf8DXEruqodHMrgo0qRUl9XKRxAQVZGiN8TxS7KXZ6liYk IHfRjrAnrHvGlvRXdd0vMGLUxjqvqzDBCEZqkZ+AlgA2D+HqWzekq3osmRA2k0du5O5o kCcbGty+UgEm8rxi/OLq7qgVCFwHH13HTXpftBj1XaM0i7tjBO9lKRJTdTnJF5R/3t6c JrkS5ti9XRCTfaa1D8P3ktMUx2ASyp8GqB5XYe9RG7pBvxny7Wq3BWLQtvGrXB/UToqw xyXoWdXWsB9HM0UfeGzOOvYeGlmazG1w/VKQFNAblPuwmKiFaBvQdPac1I4EbR7ShaXp NDtQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722162; x=1688314162; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=DpCi+Z+5IrlXYT4x1uPx6ey3Kh6k4RwU4EEsfyAeFCY=; b=YbSp54Tspv8h4Ko9Kp9Rl2mMNT0pSN4ABFh+NESD67tpOyyomzowORK1weMKGE78D+ 7A+CuK6oKDHz+F04I0+OaKkckOinkyW0boGNj64jLnndgqbp8x13TLBWfXG2dkrhEcY1 Ua5AuTSYR1UhaOQY6jSOwwfHfgp5vTyUOGqbut6qBHKitOPFOLHaajW9pMuTzk1YWk+E dCg+7iPTzT+5Ikhq4FfS22Bl10kV5TZNSHgulB856XPxnWDQyfJZVEfnR94jwiLfshiq E8MGeFGmOES5gldwp4P9kGX+PxEXi5aMcLDUd9d3rTNWy2HMx87iAbVKSwOyhQDi0VgL VcvQ== X-Gm-Message-State: AC+VfDzKN7z6jDQ1NERRBf0yQMcFrZm7G83N9+lZFy7SH0syP+mYs3n0 C5zdOmIl/PyO5YpOnug8PeONLm4ZdlAC X-Google-Smtp-Source: ACHHUZ7r4si2IcQKI55bOSM8SHxJbAhndrnIqDFmjr9vqnHPxyaBa77OMmwMCW0FKU1TUemviXs4kKMtvfhR X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a05:6a00:2183:b0:643:a542:b311 with SMTP id h3-20020a056a00218300b00643a542b311mr4840248pfi.0.1685722161708; Fri, 02 Jun 2023 09:09:21 -0700 (PDT) Date: Fri, 2 Jun 2023 09:08:59 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-2-vipinsh@google.com> Subject: [PATCH v2 01/16] KVM: selftests: Clear dirty logs in user defined chunks sizes in dirty_log_perf_test From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org In dirty_log_perf_test, provide a new option 'k' to specify the size of the chunks and clear dirty memory in chunks in each iteration. If option is not provided then fallback to the old way of clearing whole memslot in one call in each iteration. In production environment whole memslot is rarely cleared in a single call, instead clearing operation is split across multiple calls to reduce time between clearing and sending memory to a remote host. This change mimics the production usecases and allows to get performance numbers based on that. Signed-off-by: Vipin Sharma --- .../selftests/kvm/dirty_log_perf_test.c | 42 +++++++++++++++---- 1 file changed, 34 insertions(+), 8 deletions(-) diff --git a/tools/testing/selftests/kvm/dirty_log_perf_test.c b/tools/testing/selftests/kvm/dirty_log_perf_test.c index e9d6d1aecf89..119ddfc7306e 100644 --- a/tools/testing/selftests/kvm/dirty_log_perf_test.c +++ b/tools/testing/selftests/kvm/dirty_log_perf_test.c @@ -134,6 +134,7 @@ struct test_params { uint32_t write_percent; uint32_t random_seed; bool random_access; + uint64_t clear_chunk_size; }; static void toggle_dirty_logging(struct kvm_vm *vm, int slots, bool enable) @@ -169,16 +170,28 @@ static void get_dirty_log(struct kvm_vm *vm, unsigned long *bitmaps[], int slots } } -static void clear_dirty_log(struct kvm_vm *vm, unsigned long *bitmaps[], - int slots, uint64_t pages_per_slot) +static void clear_dirty_log_in_chunks(struct kvm_vm *vm, + unsigned long *bitmaps[], int slots, + uint64_t pages_per_slot, + uint64_t pages_per_clear) { - int i; + uint64_t from, clear_pages_count; + int i, slot; for (i = 0; i < slots; i++) { - int slot = MEMSTRESS_MEM_SLOT_INDEX + i; - - kvm_vm_clear_dirty_log(vm, slot, bitmaps[i], 0, pages_per_slot); + slot = MEMSTRESS_MEM_SLOT_INDEX + i; + from = 0; + clear_pages_count = pages_per_clear; + + while (from < pages_per_slot) { + if (from + clear_pages_count > pages_per_slot) + clear_pages_count = pages_per_slot - from; + kvm_vm_clear_dirty_log(vm, slot, bitmaps[i], from, + clear_pages_count); + from += clear_pages_count; + } } + } static unsigned long **alloc_bitmaps(int slots, uint64_t pages_per_slot) @@ -215,6 +228,7 @@ static void run_test(enum vm_guest_mode mode, void *arg) uint64_t guest_num_pages; uint64_t host_num_pages; uint64_t pages_per_slot; + uint64_t pages_per_clear; struct timespec start; struct timespec ts_diff; struct timespec get_dirty_log_total = (struct timespec){0}; @@ -235,6 +249,7 @@ static void run_test(enum vm_guest_mode mode, void *arg) guest_num_pages = vm_adjust_num_guest_pages(mode, guest_num_pages); host_num_pages = vm_num_host_pages(mode, guest_num_pages); pages_per_slot = host_num_pages / p->slots; + pages_per_clear = p->clear_chunk_size / getpagesize(); bitmaps = alloc_bitmaps(p->slots, pages_per_slot); @@ -315,7 +330,9 @@ static void run_test(enum vm_guest_mode mode, void *arg) if (dirty_log_manual_caps) { clock_gettime(CLOCK_MONOTONIC, &start); - clear_dirty_log(vm, bitmaps, p->slots, pages_per_slot); + clear_dirty_log_in_chunks(vm, bitmaps, p->slots, + pages_per_slot, + pages_per_clear); ts_diff = timespec_elapsed(start); clear_dirty_log_total = timespec_add(clear_dirty_log_total, ts_diff); @@ -413,6 +430,11 @@ static void help(char *name) " To leave the application task unpinned, drop the final entry:\n\n" " ./dirty_log_perf_test -v 3 -c 22,23,24\n\n" " (default: no pinning)\n"); + printf(" -k: Specify the chunk size in which dirty memory gets cleared\n" + " in memslots in each iteration. If the size is bigger than\n" + " the memslot size then whole memslot is cleared in one call.\n" + " Size must be aligned to the host page size. e.g. 10M or 3G\n" + " (default: UINT64_MAX, clears whole memslot in one call)\n"); puts(""); exit(0); } @@ -428,6 +450,7 @@ int main(int argc, char *argv[]) .slots = 1, .random_seed = 1, .write_percent = 100, + .clear_chunk_size = UINT64_MAX, }; int opt; @@ -438,7 +461,7 @@ int main(int argc, char *argv[]) guest_modes_append_default(); - while ((opt = getopt(argc, argv, "ab:c:eghi:m:nop:r:s:v:x:w:")) != -1) { + while ((opt = getopt(argc, argv, "ab:c:eghi:k:m:nop:r:s:v:x:w:")) != -1) { switch (opt) { case 'a': p.random_access = true; @@ -462,6 +485,9 @@ int main(int argc, char *argv[]) case 'i': p.iterations = atoi_positive("Number of iterations", optarg); break; + case 'k': + p.clear_chunk_size = parse_size(optarg); + break; case 'm': guest_modes_cmdline(optarg); break; From patchwork Fri Jun 2 16:09:00 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265575 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1A4AFC7EE2A for ; Fri, 2 Jun 2023 16:09:31 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236580AbjFBQJ2 (ORCPT ); Fri, 2 Jun 2023 12:09:28 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34954 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236507AbjFBQJ1 (ORCPT ); Fri, 2 Jun 2023 12:09:27 -0400 Received: from mail-yw1-x114a.google.com (mail-yw1-x114a.google.com [IPv6:2607:f8b0:4864:20::114a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id AEBF01B1 for ; Fri, 2 Jun 2023 09:09:24 -0700 (PDT) Received: by mail-yw1-x114a.google.com with SMTP id 00721157ae682-565d1b86a63so31907487b3.0 for ; Fri, 02 Jun 2023 09:09:24 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722164; x=1688314164; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=reSpQux61RJZJqDsTFoDFmSOkfig6/G5ur1FFXdW+Iw=; b=Nery7N09ub9ZwszJcaP2RaQeFUpVfIUClULYoJNCv0nmmaSATwziFhFG6U3dYDLKoQ L9drpqC2yL+EBnwjo9iEoiI5SqrPJ4COUeKCUkto6WScJ4k7QxjCwWO+xu6zR2w8Ms7Y g71MEHJild0QfvuE9EvmodRfRpkB6FSwoCyRei/rwamIW8aYnRWnvssNIV+xig+yToqy 7LLYR4U6K3iWH99NzWXy3BA1U6+xbZoa0eqzZ6Q+yGdcRqc3MVym+ONWDEZ2yy/xmSHV b2XWmusMNnhXq3avpcOVHAJU3vLRZTkYx3P/807P0dyObmLHptHPl4+5txiV/ugQdwi2 WupA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722164; x=1688314164; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=reSpQux61RJZJqDsTFoDFmSOkfig6/G5ur1FFXdW+Iw=; b=LmGLGp8NsdlP4xyZWd2xlOf095TUTbSVduiXOz/jf4xKI4PQzsM3tlgUeq6oc4fYVb c9oDP5LYBHTVvyy+txt6zaUOrsaCwLohNvFab7+MkuNZ+8ed3Wu0eZsop1YETDynjJiJ 5bT9/CLSKfXZ8gi0tC6CPEq2pGB7fADxv2Ble1PjEztdXHM3ArFECHvPXFO7AohvBjzl PT/Sz7zJ5G2bcRijBq4EiVaqcmoZXSl/2c1NEg2UsFgvWftSym/vI5QF6qo49fTgW2G6 mBGBsnWQRXp44RAkEEoIsaVVdHa2xubZlRbZx5mlE+8GUIle66gzU88npv83af+xYOsb mxxA== X-Gm-Message-State: AC+VfDz1kFDwzxEeZlc/5s2C2bRb5T70A80fxSchA4UYo7aa5Ujz8ME3 JiDiGaS94F2FpOBNZKmMXoWusS0Qd4Pg X-Google-Smtp-Source: ACHHUZ4Q3v0w4Q5HlE0q/no23AAAdsXEAJElE4UvgXKrVWGxv34xFISO1VT+CZ5ao9TA8TGy/Rc8XIiLO5th X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a25:cd0a:0:b0:ba8:918a:ceec with SMTP id d10-20020a25cd0a000000b00ba8918aceecmr470216ybf.4.1685722163747; Fri, 02 Jun 2023 09:09:23 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:00 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-3-vipinsh@google.com> Subject: [PATCH v2 02/16] KVM: selftests: Add optional delay between consecutive clear-dirty-log calls From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org In dirty_log_perf_test, add option "-l" to wait between consecutive clear-dirty-log calls. Accept delay from user in milliseconds. If option is not provided then fallback to no wait between clear calls. This allows dirty_log_perf_test to mimic real world use where after clearing dirty memory, some time is spent in transferring memory before making a subsequeunt clear-dirty-log call. Signed-off-by: Vipin Sharma --- .../selftests/kvm/dirty_log_perf_test.c | 35 +++++++++++++++---- 1 file changed, 29 insertions(+), 6 deletions(-) diff --git a/tools/testing/selftests/kvm/dirty_log_perf_test.c b/tools/testing/selftests/kvm/dirty_log_perf_test.c index 119ddfc7306e..2e31f13aaba6 100644 --- a/tools/testing/selftests/kvm/dirty_log_perf_test.c +++ b/tools/testing/selftests/kvm/dirty_log_perf_test.c @@ -135,6 +135,7 @@ struct test_params { uint32_t random_seed; bool random_access; uint64_t clear_chunk_size; + int clear_chunk_wait_time_ms; }; static void toggle_dirty_logging(struct kvm_vm *vm, int slots, bool enable) @@ -173,8 +174,14 @@ static void get_dirty_log(struct kvm_vm *vm, unsigned long *bitmaps[], int slots static void clear_dirty_log_in_chunks(struct kvm_vm *vm, unsigned long *bitmaps[], int slots, uint64_t pages_per_slot, - uint64_t pages_per_clear) + uint64_t pages_per_clear, int wait_ms, + struct timespec *time_taken) { + struct timespec wait = { + .tv_sec = wait_ms / 1000, + .tv_nsec = (wait_ms % 1000) * 1000000ull, + }; + struct timespec start, end; uint64_t from, clear_pages_count; int i, slot; @@ -186,12 +193,17 @@ static void clear_dirty_log_in_chunks(struct kvm_vm *vm, while (from < pages_per_slot) { if (from + clear_pages_count > pages_per_slot) clear_pages_count = pages_per_slot - from; + clock_gettime(CLOCK_MONOTONIC, &start); kvm_vm_clear_dirty_log(vm, slot, bitmaps[i], from, clear_pages_count); + end = timespec_elapsed(start); + *time_taken = timespec_add(*time_taken, end); from += clear_pages_count; + if (wait_ms) + nanosleep(&wait, NULL); + } } - } static unsigned long **alloc_bitmaps(int slots, uint64_t pages_per_slot) @@ -329,11 +341,11 @@ static void run_test(enum vm_guest_mode mode, void *arg) iteration, ts_diff.tv_sec, ts_diff.tv_nsec); if (dirty_log_manual_caps) { - clock_gettime(CLOCK_MONOTONIC, &start); clear_dirty_log_in_chunks(vm, bitmaps, p->slots, pages_per_slot, - pages_per_clear); - ts_diff = timespec_elapsed(start); + pages_per_clear, + p->clear_chunk_wait_time_ms, + &ts_diff); clear_dirty_log_total = timespec_add(clear_dirty_log_total, ts_diff); pr_info("Iteration %d clear dirty log time: %ld.%.9lds\n", @@ -435,6 +447,11 @@ static void help(char *name) " the memslot size then whole memslot is cleared in one call.\n" " Size must be aligned to the host page size. e.g. 10M or 3G\n" " (default: UINT64_MAX, clears whole memslot in one call)\n"); + printf(" -l: Specify time in milliseconds to wait after Clear-Dirty-Log\n" + " call. This allows to mimic use cases where flow is to get\n" + " dirty log followed by multiple clear dirty log calls and\n" + " sending corresponding memory to destination (in this test\n" + " sending will be just idle waiting)\n"); puts(""); exit(0); } @@ -451,6 +468,7 @@ int main(int argc, char *argv[]) .random_seed = 1, .write_percent = 100, .clear_chunk_size = UINT64_MAX, + .clear_chunk_wait_time_ms = 0, }; int opt; @@ -461,7 +479,7 @@ int main(int argc, char *argv[]) guest_modes_append_default(); - while ((opt = getopt(argc, argv, "ab:c:eghi:k:m:nop:r:s:v:x:w:")) != -1) { + while ((opt = getopt(argc, argv, "ab:c:eghi:k:l:m:nop:r:s:v:x:w:")) != -1) { switch (opt) { case 'a': p.random_access = true; @@ -488,6 +506,11 @@ int main(int argc, char *argv[]) case 'k': p.clear_chunk_size = parse_size(optarg); break; + case 'l': + p.clear_chunk_wait_time_ms = + atoi_non_negative("Clear dirty log chunks wait time", + optarg); + break; case 'm': guest_modes_cmdline(optarg); break; From patchwork Fri Jun 2 16:09:01 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265576 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 7BE0EC7EE24 for ; Fri, 2 Jun 2023 16:09:36 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236424AbjFBQJd (ORCPT ); Fri, 2 Jun 2023 12:09:33 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35024 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236598AbjFBQJb (ORCPT ); Fri, 2 Jun 2023 12:09:31 -0400 Received: from mail-pj1-x1049.google.com (mail-pj1-x1049.google.com [IPv6:2607:f8b0:4864:20::1049]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 555EF1BD for ; Fri, 2 Jun 2023 09:09:26 -0700 (PDT) Received: by mail-pj1-x1049.google.com with SMTP id 98e67ed59e1d1-2566b668cc5so1567373a91.0 for ; Fri, 02 Jun 2023 09:09:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722166; x=1688314166; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=U16a7SsqQ7yjyv5J8Y2N4u7Ylkl11H9qatwpuhmZUgo=; b=636YoGkhL1EaD07rsMfPK9rB926Ht76M8pVZszNcRp8cSFBaHFxW4ESJErvKH1m1wQ b8SVVb+49Lm4c7lpuGUxvabzksatMUh3aHtTfXuucGjqsfAbB7x8jB5X+IIkfWd/B+pM hI1Jypk91JlkfEFB2dbGzi8Bjd9QmphZBvbPzeFfXopwyd64oFIEk7hcFeTi49fzsYb0 jMhjapC0h8n44W4QCQ7q/RI4lKIJPywJjobPFbGliUK4+jwpyLj3v9rdAbRrHTavotzv jSbTLKdo3Rd++6WRrKKX3ZluUSmLDy9wL6sxwwb/x/fXUrLPPNX7BCJ2JivFRXiwGY+V XR9A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722166; x=1688314166; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=U16a7SsqQ7yjyv5J8Y2N4u7Ylkl11H9qatwpuhmZUgo=; b=Ivqlij2MD525PZ/Q+5eDSoLs1jFoBdhrC5NevtcY4+djYHXlSkRdAblBVDrmLIvFHt uRRVQR2dM9pX12VA0LuhUM6mP4TV+907tPDZbDCKXaQb2owG9Tn+dSgPuGCP+h5osNo6 fvLZjjV0/sJf/FKHEYxRAsCpn8qnys48BmVFy8QhJSji1eLVRy/Yt0I8rVv0b+aLfHnm WxjMBwZ42cJHKkd4TRk3USZGaPg41EgLCW13pcl88wSoCwEwoDu8HUvoS59tHfbqlnpI irWy3ID4IXxkKeyd2Y9/j3m8DJGfQEMrVFeP5UNxvw8eA7wMwUPn8Vzr6NYLvz7IEDvK VEcQ== X-Gm-Message-State: AC+VfDzz4wHalV60sdUTMTBlF7kp3TJzThR9AbM9zZ5dZlJ1GkMkk/Ii puy6w7YqBCD9FrM/4Iw6MVJOuk1Qr75U X-Google-Smtp-Source: ACHHUZ5ZPLDaNjdv6ql0/d9oDMUgHRd2fE1fBqzKriMYD3yXbiA53bpGux4rVDyMUkeRamgRC/kjMfQm5qU6 X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a17:90a:de04:b0:24e:18ff:5bad with SMTP id m4-20020a17090ade0400b0024e18ff5badmr42850pjv.0.1685722165847; Fri, 02 Jun 2023 09:09:25 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:01 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-4-vipinsh@google.com> Subject: [PATCH v2 03/16] KVM: selftests: Pass the count of read and write accesses from guest to host From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Pass the number of read and write accesses done in the memstress guest code to userspace. These counts will provide a way to measure vCPUs performance during memstress and dirty logging related tests. For example, in dirty_log_perf_test this can be used to measure how much progress vCPUs are able to do while VMM is getting and clearing dirty logs. In dirty_log_perf_test, each vCPU runs once and then waits until iteration value is incremented by main thread, therefore, these access counts will not provide much useful information except for observing read vs write counts. However, in future commits, dirty_log_perf_test behavior will be changed to allow vCPUs to execute independent of userspace iterations. This will mimic real world workload where guest keeps on executing while VMM is collecting and clearing dirty logs separately. With read and write accesses known for each vCPU, impact of get and clear dirty log APIs can be quantified. Note that access counts will not be 100% reliable in knowing vCPUs performances. Few things which can affect vCPU progress: 1. vCPUs are scheduled less by host 2. Userspace operations run for longer time which end up giving vCPUs more time to execute. Signed-off-by: Vipin Sharma --- tools/testing/selftests/kvm/lib/memstress.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/kvm/lib/memstress.c b/tools/testing/selftests/kvm/lib/memstress.c index 5f1d3173c238..ac53cc6e36d7 100644 --- a/tools/testing/selftests/kvm/lib/memstress.c +++ b/tools/testing/selftests/kvm/lib/memstress.c @@ -49,6 +49,8 @@ void memstress_guest_code(uint32_t vcpu_idx) struct memstress_args *args = &memstress_args; struct memstress_vcpu_args *vcpu_args = &args->vcpu_args[vcpu_idx]; struct guest_random_state rand_state; + uint64_t write_access; + uint64_t read_access; uint64_t gva; uint64_t pages; uint64_t addr; @@ -64,6 +66,8 @@ void memstress_guest_code(uint32_t vcpu_idx) GUEST_ASSERT(vcpu_args->vcpu_idx == vcpu_idx); while (true) { + write_access = 0; + read_access = 0; for (i = 0; i < pages; i++) { if (args->random_access) page = guest_random_u32(&rand_state) % pages; @@ -72,13 +76,16 @@ void memstress_guest_code(uint32_t vcpu_idx) addr = gva + (page * args->guest_page_size); - if (guest_random_u32(&rand_state) % 100 < args->write_percent) + if (guest_random_u32(&rand_state) % 100 < args->write_percent) { *(uint64_t *)addr = 0x0123456789ABCDEF; - else + write_access++; + } else { READ_ONCE(*(uint64_t *)addr); + read_access++; + } } - GUEST_SYNC(1); + GUEST_SYNC_ARGS(1, read_access, write_access, 0, 0); } } From patchwork Fri Jun 2 16:09:02 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265607 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5A86DC7EE33 for ; Fri, 2 Jun 2023 16:10:00 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236609AbjFBQJ7 (ORCPT ); Fri, 2 Jun 2023 12:09:59 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35018 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236592AbjFBQJc (ORCPT ); Fri, 2 Jun 2023 12:09:32 -0400 Received: from mail-pg1-x549.google.com (mail-pg1-x549.google.com [IPv6:2607:f8b0:4864:20::549]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 52AE31BB for ; Fri, 2 Jun 2023 09:09:28 -0700 (PDT) Received: by mail-pg1-x549.google.com with SMTP id 41be03b00d2f7-53f44c2566dso973969a12.2 for ; Fri, 02 Jun 2023 09:09:28 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722167; x=1688314167; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=vX8ZCjC9l+xt5cqUnwzlqCkk/7EvJIEZcs67EwYM95Y=; b=ZpYUhbbZC6uUYrReM+ixNdEUwp22WUU7l8j2aahWsJUYqP381L7DaYeopKye6+Q9sn w6jGO0nX6WnAtH7Q3sGlOdc68vECR5PacYtF8uFrv3gPYqUB1KYLtsQIi3wA+701ktoo VcYLyZENUnFYI88cWmJbgLosheXItBqWsjUj1JOOFK7yQMQ1EVYBrKEM74EAZ9G1JJes +Fy3XP3gLD3JszJ1CTZiXeoILpXFoZBpgnn/vdDYBMvk+Dqwv0kvHMMpaOH7tt3QFj+s ke1ao1kRwftxIusnCaBSonu2zz1KVzCGGfUgW7HZsGqYe4xCh5lCKxmVw7qkF3LGd5ny Gpkw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722167; x=1688314167; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=vX8ZCjC9l+xt5cqUnwzlqCkk/7EvJIEZcs67EwYM95Y=; b=Et5cILaqsl6BApuzPypYW0yyYnRP0JSdZ2vmomc1zKJvOGhcdi8RjFnFiCQxI6J6UF si/+WqrSINWcrtdlJbIHCgNc8vnvgep4cDEtpC5QhGL6fXgUFRb8CxZZGjseRG8xM00G 3seP7agcaTAH9rQXQRMe+v/A07Xo+hwgn0XibfPyj16fvVBIof1qQm+LRFQZh7N2aSAh l0nXHOBZ0UqRFuMMagceDNA/zea2qFcdGCfpjesk0DJ68sMu88RkL6tiHWFxj/hYtO9h sWBwr4fuNhNlmhZBycwpgUSwSaYmLA5Xwm/vbb5zGLeh+3/kXRKCuoTOnCoDsb2YndJG rF1A== X-Gm-Message-State: AC+VfDwtcPYfpR67wzoeKlAll1MoPDXMx4G8d8QYAGqaB8eSUKESfqbx wf0NaCU+u9M3ak2JW9VAHncb2AlES29Y X-Google-Smtp-Source: ACHHUZ7w+3zMwOJ66cN0pwH2useIGj7AgTlfLjnw5VFZgi7VE2BNKptqAZGtU12TE7bl7/aJEHHayi6hzCyj X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a63:685:0:b0:530:866e:c3c1 with SMTP id 127-20020a630685000000b00530866ec3c1mr2520235pgg.11.1685722167618; Fri, 02 Jun 2023 09:09:27 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:02 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-5-vipinsh@google.com> Subject: [PATCH v2 04/16] KVM: selftests: Print read-write progress by vCPUs in dirty_log_perf_test From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Fetch count of read and write accesses from guest code and print sum of these values across all vCPUs in dirty_log_perf_test. This data provides progress made by vCPUs during dirty logging operations. Since, vCPUs execute in lockstep with userspace dirty log iterations, this metric is not very interesting. However, in future commits when dirty_log_perf_test can execute vCPUs independently from dirty log iterations then this metric can give good measure of vCPUs performance during dirty logging. Signed-off-by: Vipin Sharma --- .../selftests/kvm/dirty_log_perf_test.c | 18 +++++++++++++++++- 1 file changed, 17 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/kvm/dirty_log_perf_test.c b/tools/testing/selftests/kvm/dirty_log_perf_test.c index 2e31f13aaba6..14b012a0dcb1 100644 --- a/tools/testing/selftests/kvm/dirty_log_perf_test.c +++ b/tools/testing/selftests/kvm/dirty_log_perf_test.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include "kvm_util.h" @@ -66,17 +67,22 @@ static u64 dirty_log_manual_caps; static bool host_quit; static int iteration; static int vcpu_last_completed_iteration[KVM_MAX_VCPUS]; +static atomic_ullong total_reads; +static atomic_ullong total_writes; static void vcpu_worker(struct memstress_vcpu_args *vcpu_args) { struct kvm_vcpu *vcpu = vcpu_args->vcpu; int vcpu_idx = vcpu_args->vcpu_idx; uint64_t pages_count = 0; + uint64_t reads = 0; + uint64_t writes = 0; struct kvm_run *run; struct timespec start; struct timespec ts_diff; struct timespec total = (struct timespec){0}; struct timespec avg; + struct ucall uc = {}; int ret; run = vcpu->run; @@ -89,7 +95,7 @@ static void vcpu_worker(struct memstress_vcpu_args *vcpu_args) ts_diff = timespec_elapsed(start); TEST_ASSERT(ret == 0, "vcpu_run failed: %d\n", ret); - TEST_ASSERT(get_ucall(vcpu, NULL) == UCALL_SYNC, + TEST_ASSERT(get_ucall(vcpu, &uc) == UCALL_SYNC, "Invalid guest sync status: exit_reason=%s\n", exit_reason_str(run->exit_reason)); @@ -101,6 +107,8 @@ static void vcpu_worker(struct memstress_vcpu_args *vcpu_args) if (current_iteration) { pages_count += vcpu_args->pages; total = timespec_add(total, ts_diff); + reads += uc.args[2]; + writes += uc.args[3]; pr_debug("vCPU %d iteration %d dirty memory time: %ld.%.9lds\n", vcpu_idx, current_iteration, ts_diff.tv_sec, ts_diff.tv_nsec); @@ -123,6 +131,8 @@ static void vcpu_worker(struct memstress_vcpu_args *vcpu_args) pr_debug("\nvCPU %d dirtied 0x%lx pages over %d iterations in %ld.%.9lds. (Avg %ld.%.9lds/iteration)\n", vcpu_idx, pages_count, vcpu_last_completed_iteration[vcpu_idx], total.tv_sec, total.tv_nsec, avg.tv_sec, avg.tv_nsec); + atomic_fetch_add(&total_reads, reads); + atomic_fetch_add(&total_writes, writes); } struct test_params { @@ -270,6 +280,8 @@ static void run_test(enum vm_guest_mode mode, void *arg) dirty_log_manual_caps); arch_setup_vm(vm, nr_vcpus); + atomic_store(&total_reads, 0); + atomic_store(&total_writes, 0); /* Start the iterations */ iteration = 0; @@ -388,6 +400,10 @@ static void run_test(enum vm_guest_mode mode, void *arg) clear_dirty_log_total.tv_nsec, avg.tv_sec, avg.tv_nsec); } + pr_info("Total pages touched: %llu (Reads: %llu, Writes: %llu)\n", + atomic_load(&total_reads) + atomic_load(&total_writes), + atomic_load(&total_reads), atomic_load(&total_writes)); + free_bitmaps(bitmaps, p->slots); arch_cleanup_vm(vm); memstress_destroy_vm(vm); From patchwork Fri Jun 2 16:09:03 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265608 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4630DC7EE24 for ; Fri, 2 Jun 2023 16:10:02 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236656AbjFBQKA (ORCPT ); Fri, 2 Jun 2023 12:10:00 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35042 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S235957AbjFBQJm (ORCPT ); Fri, 2 Jun 2023 12:09:42 -0400 Received: from mail-pg1-x549.google.com (mail-pg1-x549.google.com [IPv6:2607:f8b0:4864:20::549]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 50332E51 for ; Fri, 2 Jun 2023 09:09:30 -0700 (PDT) Received: by mail-pg1-x549.google.com with SMTP id 41be03b00d2f7-517bad1b8c5so2085632a12.0 for ; Fri, 02 Jun 2023 09:09:30 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722169; x=1688314169; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=TCZIk7VgxCOZOGj9BrulapCebBiXjQ5+VJfBT1Ieq9g=; b=48Jv9Ndg6W2FWtv0TNF+EvHQkuzevnf0de3wNHQhiINnd1y+TJ3ZHZNsYvgiBjl6lh 9WUrwiuBOv0A4GYPSawEV6yVx5LbPQL0U+gQGnFWAFVUnyrF1z1qm1xlbl6HRjy6NNwE 8XB2SVK1dbUaEeLrM/H8JHaqBSIup+A6xNFpzpaufkcVOP+VAaTUukUdmKNrTQcISuZt yTtWhk1kuZr3BXzyN9NP6mKp7IAXkKM1nK9WQaTXO/PtqFm7I9EIaNt4fcwETtWcECAj 9ktEJ+8oLB8jJgpGPXgzImfVTKXWHcOIgbQS6xiesRZonieLTn89FjSta0ZVX6QUoqee mZLw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722169; x=1688314169; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=TCZIk7VgxCOZOGj9BrulapCebBiXjQ5+VJfBT1Ieq9g=; b=Vo8vJ6znuTdZK5782taSyu3TawwJ/CB06HtLaLVrCRGaLT7KKIJOhzwKQmXwNL6ukw a2aNDogk7kGJJgQfVqIou4f0j71flpMyMHalnsBVxqVmWJ7rdmdrT+I6Ae2Up3yLKzKo 3VcdYGzgCJBWQGjTicQRQ0XJ8zUwOg37VgVqrLBiQt6ChxPYr6Qqk5aMpm8IHZO4JTVg 8ksqXoJlHc61PFTBTUbGczPkedk9+bxX/YwA+Wk/FRlAQc7tLjBd5Vb09ryH/Ca4M889 TUPuBLPzS4wcwgQN9zMNfdtssV2ATTpHYnnuIFXknYn78axtpqbn2CyWnN/YyK9XEUps YyuQ== X-Gm-Message-State: AC+VfDwsk6xEyzYTemQ54Mz5EQYXLHfwTjnjGQEnPW4OSw82Tc3Zlckk 0UDcnGrU44KkWgEqabb5JWnEBRLA2+QM X-Google-Smtp-Source: ACHHUZ5eYF/RV9lQrAgLH67VBGXqFnNNXk0G34PjM3lxGa6/KnJxp9P54a/FAO5X5SRVmoZXvlFdh+lSImCS X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a63:d044:0:b0:53f:32cf:bcd1 with SMTP id s4-20020a63d044000000b0053f32cfbcd1mr2492124pgi.5.1685722169678; Fri, 02 Jun 2023 09:09:29 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:03 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-6-vipinsh@google.com> Subject: [PATCH v2 05/16] KVM: selftests: Allow independent execution of vCPUs in dirty_log_perf_test From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Give users command line option (-j) to execute vCPUs independently of dirty log iterations after initialization is complete. This change makes dirty_log_perf_test behave like real world workflows where guest vCPUs keep on executing while VMM collects and clear dirty logs. Total pages touched during execution of test will give good estimate of how vCPUs are performing while dirty logging is enabled. Signed-off-by: Vipin Sharma --- .../selftests/kvm/dirty_log_perf_test.c | 64 +++++++++++++------ 1 file changed, 44 insertions(+), 20 deletions(-) diff --git a/tools/testing/selftests/kvm/dirty_log_perf_test.c b/tools/testing/selftests/kvm/dirty_log_perf_test.c index 14b012a0dcb1..fbf973d6cc66 100644 --- a/tools/testing/selftests/kvm/dirty_log_perf_test.c +++ b/tools/testing/selftests/kvm/dirty_log_perf_test.c @@ -69,6 +69,7 @@ static int iteration; static int vcpu_last_completed_iteration[KVM_MAX_VCPUS]; static atomic_ullong total_reads; static atomic_ullong total_writes; +static bool lockstep_iterations; static void vcpu_worker(struct memstress_vcpu_args *vcpu_args) { @@ -83,12 +84,16 @@ static void vcpu_worker(struct memstress_vcpu_args *vcpu_args) struct timespec total = (struct timespec){0}; struct timespec avg; struct ucall uc = {}; + int current_iteration = -1; int ret; run = vcpu->run; while (!READ_ONCE(host_quit)) { - int current_iteration = READ_ONCE(iteration); + if (lockstep_iterations) + current_iteration = READ_ONCE(iteration); + else + current_iteration++; clock_gettime(CLOCK_MONOTONIC, &start); ret = _vcpu_run(vcpu); @@ -118,13 +123,19 @@ static void vcpu_worker(struct memstress_vcpu_args *vcpu_args) ts_diff.tv_nsec); } - /* - * Keep running the guest while dirty logging is being disabled - * (iteration is negative) so that vCPUs are accessing memory - * for the entire duration of zapping collapsible SPTEs. - */ - while (current_iteration == READ_ONCE(iteration) && - READ_ONCE(iteration) >= 0 && !READ_ONCE(host_quit)) {} + if (lockstep_iterations) { + /* + * Keep running the guest while dirty logging is being disabled + * (iteration is negative) so that vCPUs are accessing memory + * for the entire duration of zapping collapsible SPTEs. + */ + while (current_iteration == READ_ONCE(iteration) && + READ_ONCE(iteration) >= 0 && !READ_ONCE(host_quit)) + ; + } else { + while (!READ_ONCE(iteration) && !READ_ONCE(host_quit)) + ; + } } avg = timespec_div(total, vcpu_last_completed_iteration[vcpu_idx]); @@ -332,18 +343,20 @@ static void run_test(enum vm_guest_mode mode, void *arg) clock_gettime(CLOCK_MONOTONIC, &start); iteration++; - pr_debug("Starting iteration %d\n", iteration); - for (i = 0; i < nr_vcpus; i++) { - while (READ_ONCE(vcpu_last_completed_iteration[i]) - != iteration) - ; + if (lockstep_iterations) { + pr_debug("Starting iteration %d\n", iteration); + for (i = 0; i < nr_vcpus; i++) { + while (READ_ONCE(vcpu_last_completed_iteration[i]) + != iteration) + ; + } + + ts_diff = timespec_elapsed(start); + vcpu_dirty_total = timespec_add(vcpu_dirty_total, ts_diff); + pr_info("Iteration %d dirty memory time: %ld.%.9lds\n", + iteration, ts_diff.tv_sec, ts_diff.tv_nsec); } - ts_diff = timespec_elapsed(start); - vcpu_dirty_total = timespec_add(vcpu_dirty_total, ts_diff); - pr_info("Iteration %d dirty memory time: %ld.%.9lds\n", - iteration, ts_diff.tv_sec, ts_diff.tv_nsec); - clock_gettime(CLOCK_MONOTONIC, &start); get_dirty_log(vm, bitmaps, p->slots); ts_diff = timespec_elapsed(start); @@ -365,6 +378,10 @@ static void run_test(enum vm_guest_mode mode, void *arg) } } + /* Block further vCPUs execution */ + if (!lockstep_iterations) + WRITE_ONCE(iteration, 0); + /* * Run vCPUs while dirty logging is being disabled to stress disabling * in terms of both performance and correctness. Opt-in via command @@ -458,6 +475,10 @@ static void help(char *name) " To leave the application task unpinned, drop the final entry:\n\n" " ./dirty_log_perf_test -v 3 -c 22,23,24\n\n" " (default: no pinning)\n"); + printf(" -j: Execute vCPUs independent of dirty log iterations\n" + " Independent vCPUs execution will allow them to continuously\n" + " dirty memory while main thread is collecting and clearing\n" + " dirty logs in each iteration.\n"); printf(" -k: Specify the chunk size in which dirty memory gets cleared\n" " in memslots in each iteration. If the size is bigger than\n" " the memslot size then whole memslot is cleared in one call.\n" @@ -492,10 +513,10 @@ int main(int argc, char *argv[]) kvm_check_cap(KVM_CAP_MANUAL_DIRTY_LOG_PROTECT2); dirty_log_manual_caps &= (KVM_DIRTY_LOG_MANUAL_PROTECT_ENABLE | KVM_DIRTY_LOG_INITIALLY_SET); - + lockstep_iterations = true; guest_modes_append_default(); - while ((opt = getopt(argc, argv, "ab:c:eghi:k:l:m:nop:r:s:v:x:w:")) != -1) { + while ((opt = getopt(argc, argv, "ab:c:eghi:jk:l:m:nop:r:s:v:x:w:")) != -1) { switch (opt) { case 'a': p.random_access = true; @@ -519,6 +540,9 @@ int main(int argc, char *argv[]) case 'i': p.iterations = atoi_positive("Number of iterations", optarg); break; + case 'j': + lockstep_iterations = false; + break; case 'k': p.clear_chunk_size = parse_size(optarg); break; From patchwork Fri Jun 2 16:09:04 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265609 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 6512AC7EE32 for ; Fri, 2 Jun 2023 16:10:04 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236678AbjFBQKC (ORCPT ); Fri, 2 Jun 2023 12:10:02 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35362 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236624AbjFBQJn (ORCPT ); Fri, 2 Jun 2023 12:09:43 -0400 Received: from mail-yw1-x1149.google.com (mail-yw1-x1149.google.com [IPv6:2607:f8b0:4864:20::1149]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BD16B1BC for ; Fri, 2 Jun 2023 09:09:32 -0700 (PDT) Received: by mail-yw1-x1149.google.com with SMTP id 00721157ae682-565a336caa0so33671177b3.1 for ; Fri, 02 Jun 2023 09:09:32 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722172; x=1688314172; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=fngtlIXFBOk5ojPhoCZKxdftJsl/PCjzcivfHulUfxM=; b=KTqFTUl8/O/otunDnIcW8NMT9NoZpURoL+fT5x3jr7xLTtmqLTsOpMHELHqceBBNBQ AmDayYQtgKFH5Z55J+qEEznt7CMzcfBLKa9rXdwkcbaxYdDyFrpWK5tIf+OgM6RdtGyq GQdHIK3QOQ2drYG/ZFFk+kHjBy+s90SO6eGGqGaoiIn9r7XkEBIEo/mXJtp3GumBEOA+ p4STDDMfVtsOTK10O8PNGuXazkAWxMkEfK3iCnrl9MiWc43AYLln6h9/+ty9OxynFFMd DAYdFz0PuZJQnqyg+7O9VnZBBnLY/2RVdSpDM0DZZNDzkFP0BkvcK+wZMlJvwm1mfdaV VyLQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722172; x=1688314172; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=fngtlIXFBOk5ojPhoCZKxdftJsl/PCjzcivfHulUfxM=; b=L9IPYdOWhBmzF2MhwHR7T+0gBLSpEE0rR4Wa7NNucS5nFGQyrJBN5JVQshO3hEd5jd Ok0xUAnyG/w3t8HcHmkVirl5w/fa53gpPMrLE3bD4P5jrHnnukXKmIecrovzaPwktH6P woYMI77EJ2Dy7ytMGi9H817ClJdlt3JeDXoreKHo7brJLqwNcVDt5M9NYKXb4n+Z4VbS 1tnmJNJzxn8iIHJ54avodtxivbG2tkbPpLrl5HlESdiAd1xI/vBISEtC6w6QxihTv74c rY6NbJxBGDm7753ftTcwZHXwoHCFEypZ1wQdsCKogB239P56nUg8SY/SZCBJlSDYBHV/ tUKg== X-Gm-Message-State: AC+VfDxh9NN9d9o1EX1xatv032MCgpw2HH4Eqt38kKxbHLPKzzM1qOJ3 d151dULKFRAttqeWnfDkLaJVAV3cHpdg X-Google-Smtp-Source: ACHHUZ6gan5YlBkZGg9KteJiKVhKxkJ5gsh9dOHqcHA/QTL4d+xT++IRNn1eOG3Ha8PAz+KvOVFz0vRB3dhD X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a81:c509:0:b0:55a:3133:86fa with SMTP id k9-20020a81c509000000b0055a313386famr182818ywi.3.1685722171727; Fri, 02 Jun 2023 09:09:31 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:04 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-7-vipinsh@google.com> Subject: [PATCH v2 06/16] KVM: arm64: Correct the kvm_pgtable_stage2_flush() documentation From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Remove _range suffix from kvm_pgtable_stage2_flush_range which is used in documentation of kvm_pgtable_stage2_flush(). There is no function named kvm_pgtable_stage2_flush_range(). Fixes: 93c66b40d728 ("KVM: arm64: Add support for stage-2 cache flushing in generic page-table") Signed-off-by: Vipin Sharma --- arch/arm64/include/asm/kvm_pgtable.h | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h index 850d65f705fa..d542a671c564 100644 --- a/arch/arm64/include/asm/kvm_pgtable.h +++ b/arch/arm64/include/asm/kvm_pgtable.h @@ -657,9 +657,8 @@ int kvm_pgtable_stage2_relax_perms(struct kvm_pgtable *pgt, u64 addr, bool kvm_pgtable_stage2_is_young(struct kvm_pgtable *pgt, u64 addr); /** - * kvm_pgtable_stage2_flush_range() - Clean and invalidate data cache to Point - * of Coherency for guest stage-2 address - * range. + * kvm_pgtable_stage2_flush() - Clean and invalidate data cache to Point of + * Coherency for guest stage-2 address range. * @pgt: Page-table structure initialised by kvm_pgtable_stage2_init*(). * @addr: Intermediate physical address from which to flush. * @size: Size of the range. From patchwork Fri Jun 2 16:09:05 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265610 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id DBCFEC7EE2A for ; Fri, 2 Jun 2023 16:10:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235707AbjFBQKE (ORCPT ); Fri, 2 Jun 2023 12:10:04 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35042 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236715AbjFBQJq (ORCPT ); Fri, 2 Jun 2023 12:09:46 -0400 Received: from mail-pg1-x549.google.com (mail-pg1-x549.google.com [IPv6:2607:f8b0:4864:20::549]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 12E04E6D for ; Fri, 2 Jun 2023 09:09:34 -0700 (PDT) Received: by mail-pg1-x549.google.com with SMTP id 41be03b00d2f7-53425d37fefso973843a12.3 for ; Fri, 02 Jun 2023 09:09:34 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722174; x=1688314174; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=XlXfZDaqwtQ8jb+NBFxxHmMdNa0dhVq7sf+3J9mNCvc=; b=Cav0E0cCrA4YixnMBQKMfIUVj88wiap2lnIB8ag1Iz/8mWp6WkXueCvYRCBiVgJMpL yz6b5GmGWPhdoaqfWLr4eMh+mes1ArSi+nNH2aSvBuPWl2X06K+1JFMN+UK2I8zKDZ+Y An4/Ccw/5HBcgWIwlUsNKHK+GBxiwi6UrW0Rt5OfBwhcgGTE2dlnkWyH08h03FWRL/+p PpYax2iQC4+/Q7s8QR0zA1oOFG5toYWxuVQyxHq17uCmAv8sJvsjFT497G95S2n7vNCo An7ffox8QUfjjPO3WBaFb5yO8HO3WOQ44El3H8raJwdkFn/AASepix+It/ssVM4Hws1C WsIA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722174; x=1688314174; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=XlXfZDaqwtQ8jb+NBFxxHmMdNa0dhVq7sf+3J9mNCvc=; b=hch9qOkWLVcPuLBatlfnCdyjUytJ0h5IScTm90FGAxd26lWJwHQCAwqCc0qPZ3Ufpv UbUZeEg1cAhL7zdI3tHg5W2ZLQIqy5LxqJrJZS5+3Z1ytm20XuS6Ej1YegwBiP0SQbRQ E4e01F84WAEnuu3ajiH1+S2BgU86fLeERZ7elTeFaW4LHRlUzNE3rq2vTqSmpkGYmf5V C6rt1a8SXINzdvapajQj5c54krYd08KDcOzDhk0P3T7GFKvcK1eBAxsb3sOruqIvVhFP hLhUzRgU3cuCQ8tHXMeE8m5uFwhG71yuyII+Qm2sWr1H0XA6Fp56DrhqluRr6Suah48+ ZiDg== X-Gm-Message-State: AC+VfDx2/k8ZIxYIV37RC5vCWmLtqYZLvX7zZNBgtVBjvFOhz3gSvi25 zgvxMHuhkBSpM/5ECA60n+uFIwa3GNQF X-Google-Smtp-Source: ACHHUZ63/E8GjVgWNJSqqa8vS+/kUt+EPMZ2lBAM1R4WdW2tBWK9Y4EulGJYzAllZOblN3IVUjTbhEd8OzIe X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a63:385:0:b0:530:70cb:6da9 with SMTP id 127-20020a630385000000b0053070cb6da9mr2521810pgd.10.1685722173764; Fri, 02 Jun 2023 09:09:33 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:05 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-8-vipinsh@google.com> Subject: [PATCH v2 07/16] KVM: mmu: Move mmu lock/unlock to arch code for clear dirty log From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Move mmu_lock lock and unlock calls from common code in kvm_clear_dirty_log_protect() to arch specific code in kvm_arch_mmu_enable_log_dirty_pt_masked(). None of the other code inside the for loop of kvm_arch_mmu_enable_log_dirty_pt_masked() needs mmu_lock exclusivity apart from the arch specific API call. Future commits will change clear dirty log operations under mmu read lock instead of write lock for ARM and, potentially, x86 architectures. No functional changes intended. Signed-off-by: Vipin Sharma --- arch/arm64/kvm/mmu.c | 2 ++ arch/mips/kvm/mmu.c | 2 ++ arch/riscv/kvm/mmu.c | 2 ++ arch/x86/kvm/mmu/mmu.c | 3 +++ virt/kvm/dirty_ring.c | 2 -- virt/kvm/kvm_main.c | 4 ---- 6 files changed, 9 insertions(+), 6 deletions(-) diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 6db9ef288ec3..0c2c2c0846f1 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -1125,6 +1125,7 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, phys_addr_t start = (base_gfn + __ffs(mask)) << PAGE_SHIFT; phys_addr_t end = (base_gfn + __fls(mask) + 1) << PAGE_SHIFT; + write_lock(&kvm->mmu_lock); lockdep_assert_held_write(&kvm->mmu_lock); stage2_wp_range(&kvm->arch.mmu, start, end); @@ -1139,6 +1140,7 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, */ if (kvm_dirty_log_manual_protect_and_init_set(kvm)) kvm_mmu_split_huge_pages(kvm, start, end); + write_unlock(&kvm->mmu_lock); } static void kvm_send_hwpoison_signal(unsigned long address, short lsb) diff --git a/arch/mips/kvm/mmu.c b/arch/mips/kvm/mmu.c index e8c08988ed37..33c5af333ff9 100644 --- a/arch/mips/kvm/mmu.c +++ b/arch/mips/kvm/mmu.c @@ -419,7 +419,9 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, gfn_t start = base_gfn + __ffs(mask); gfn_t end = base_gfn + __fls(mask); + spin_lock(&kvm->mmu_lock); kvm_mips_mkclean_gpa_pt(kvm, start, end); + spin_unlock(&kvm->mmu_lock); } /* diff --git a/arch/riscv/kvm/mmu.c b/arch/riscv/kvm/mmu.c index f2eb47925806..fe026ff5eb65 100644 --- a/arch/riscv/kvm/mmu.c +++ b/arch/riscv/kvm/mmu.c @@ -399,7 +399,9 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, phys_addr_t start = (base_gfn + __ffs(mask)) << PAGE_SHIFT; phys_addr_t end = (base_gfn + __fls(mask) + 1) << PAGE_SHIFT; + spin_lock(&kvm->mmu_lock); gstage_wp_range(kvm, start, end); + spin_unlock(&kvm->mmu_lock); } void kvm_arch_sync_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot) diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c index c8961f45e3b1..6fff4228e31c 100644 --- a/arch/x86/kvm/mmu/mmu.c +++ b/arch/x86/kvm/mmu/mmu.c @@ -1382,6 +1382,7 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn_offset, unsigned long mask) { + write_lock(&kvm->mmu_lock); /* * Huge pages are NOT write protected when we start dirty logging in * initially-all-set mode; must write protect them here so that they @@ -1412,6 +1413,8 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, kvm_mmu_clear_dirty_pt_masked(kvm, slot, gfn_offset, mask); else kvm_mmu_write_protect_pt_masked(kvm, slot, gfn_offset, mask); + + write_unlock(&kvm->mmu_lock); } int kvm_cpu_dirty_log_size(void) diff --git a/virt/kvm/dirty_ring.c b/virt/kvm/dirty_ring.c index c1cd7dfe4a90..d894c58d2152 100644 --- a/virt/kvm/dirty_ring.c +++ b/virt/kvm/dirty_ring.c @@ -66,9 +66,7 @@ static void kvm_reset_dirty_gfn(struct kvm *kvm, u32 slot, u64 offset, u64 mask) if (!memslot || (offset + __fls(mask)) >= memslot->npages) return; - KVM_MMU_LOCK(kvm); kvm_arch_mmu_enable_log_dirty_pt_masked(kvm, memslot, offset, mask); - KVM_MMU_UNLOCK(kvm); } int kvm_dirty_ring_alloc(struct kvm_dirty_ring *ring, int index, u32 size) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 13aed654111a..747bfa2f1dd3 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -2160,7 +2160,6 @@ static int kvm_get_dirty_log_protect(struct kvm *kvm, struct kvm_dirty_log *log) dirty_bitmap_buffer = kvm_second_dirty_bitmap(memslot); memset(dirty_bitmap_buffer, 0, n); - KVM_MMU_LOCK(kvm); for (i = 0; i < n / sizeof(long); i++) { unsigned long mask; gfn_t offset; @@ -2176,7 +2175,6 @@ static int kvm_get_dirty_log_protect(struct kvm *kvm, struct kvm_dirty_log *log) kvm_arch_mmu_enable_log_dirty_pt_masked(kvm, memslot, offset, mask); } - KVM_MMU_UNLOCK(kvm); } if (flush) @@ -2271,7 +2269,6 @@ static int kvm_clear_dirty_log_protect(struct kvm *kvm, if (copy_from_user(dirty_bitmap_buffer, log->dirty_bitmap, n)) return -EFAULT; - KVM_MMU_LOCK(kvm); for (offset = log->first_page, i = offset / BITS_PER_LONG, n = DIV_ROUND_UP(log->num_pages, BITS_PER_LONG); n--; i++, offset += BITS_PER_LONG) { @@ -2294,7 +2291,6 @@ static int kvm_clear_dirty_log_protect(struct kvm *kvm, offset, mask); } } - KVM_MMU_UNLOCK(kvm); if (flush) kvm_arch_flush_remote_tlbs_memslot(kvm, memslot); From patchwork Fri Jun 2 16:09:06 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265612 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id CA536C7EE32 for ; Fri, 2 Jun 2023 16:10:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235118AbjFBQKG (ORCPT ); Fri, 2 Jun 2023 12:10:06 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35438 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236762AbjFBQJr (ORCPT ); Fri, 2 Jun 2023 12:09:47 -0400 Received: from mail-yw1-x1149.google.com (mail-yw1-x1149.google.com [IPv6:2607:f8b0:4864:20::1149]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 904D710C2 for ; Fri, 2 Jun 2023 09:09:36 -0700 (PDT) Received: by mail-yw1-x1149.google.com with SMTP id 00721157ae682-564fb1018bcso33919707b3.0 for ; Fri, 02 Jun 2023 09:09:36 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722176; x=1688314176; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=GJx+yypKzd3COmz5TQWkFGxt9Sb1HD5xNFJXRHqVvy0=; b=szBGu+EQ2vs8Q3A0ZT1RS79FNLRJxSOVkrDrBBX7rmCKLzKn/b+E4mNymytABzCLif EgUFXaPy7nG7KM6jDjkYBO0vsJLxZ4XBjsoF6X75ZimJDOLxnKGoSfjvnPNLfA9vSHBD s9ihl+kXGelNmLHAJZYRz4+yQ+Fx6J3bT5bOlL9FOfNpgH6MN5wQNIGrZvk2XbL68tHF B/csW1RkKWfzpRPcNxb+4yNUfWvhAhXwGpxjibheJUkN4+CaHLLPZUdUbJv/P8pKuPN/ Cvmf4i9nzVBBvNWz3J0YGn8SEauVZOkYjg+xhiWvEBlTBwkFZHPSrYROctUU+loxAFaS zjKQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722176; x=1688314176; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=GJx+yypKzd3COmz5TQWkFGxt9Sb1HD5xNFJXRHqVvy0=; b=NDWOi09rCw8+M2eQw/435Uzlo+h89iqGphQ4JfmDgYlPARmjlwOWV21nk6xDz3welI aN13h47k+QYRx6IbzrG/8wJ0gqbJw4ANjkOtACGboqOZOgeNjvk0fZFbNoQwdto2QcOG yP1LFaX6PYGgCaz8uBQihcMfLeAhoTee25hZ3quYkd8JKXu4B5PFm+4pwoY+pUxx3F5f WTm+0JKY6JJc1xj/qzGDJWnlZaBgLR85n8fiRI8NFksJp/NOl+P//Oet8jhQNVY2vw5K YDT0ubATjcLMuFtthUzlTMqWFrOrIVtxXXklQ9trsesAbqZm8yWkqY4w17H/YOeuut9w M4EQ== X-Gm-Message-State: AC+VfDxEyCpjnIkCwEsf3U26UNlVTbBO2+AK/pTRlMzRxDZRTCM8lvW0 fXA2r/TlRKFXvCVOCokKbrelp88qbauF X-Google-Smtp-Source: ACHHUZ55GAWUbCX2pBxCQpbjCqZ+MgzCwRW5rUzsUOGzP4VGdNUP2AN+4xPnboHdKBOoO7J5TTwkex+IpQ+T X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a05:6902:100a:b0:bad:600:1833 with SMTP id w10-20020a056902100a00b00bad06001833mr2072241ybt.0.1685722176005; Fri, 02 Jun 2023 09:09:36 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:06 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-9-vipinsh@google.com> Subject: [PATCH v2 08/16] KMV: arm64: Pass page table walker flags to stage2_apply_range_*() From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Allow stage2_apply_range_*() to accept enum kvm_pgtable_walk_flags{} for stage 2 walkers. Pass 0 as the flag value from all of its caller effectively making it a no-op. Page table walker flags will be used in future commits to enable clear-dirty-log operation under MMU read lock. Current users of stage2_apply_range_*() API runs under assumption of holding MMU write lock. Stage2 page table walkers then run under the same assumption. In future commits, when clear-dirty-log is modified to run under MMU read lock then this flag will be used to pass shared page walk intent. No functional changes intended. Signed-off-by: Vipin Sharma --- arch/arm64/include/asm/kvm_pgtable.h | 12 +++++++++--- arch/arm64/kvm/hyp/nvhe/mem_protect.c | 4 ++-- arch/arm64/kvm/hyp/pgtable.c | 16 ++++++++++------ arch/arm64/kvm/mmu.c | 26 ++++++++++++++++---------- 4 files changed, 37 insertions(+), 21 deletions(-) diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h index d542a671c564..8ef7e8f3f054 100644 --- a/arch/arm64/include/asm/kvm_pgtable.h +++ b/arch/arm64/include/asm/kvm_pgtable.h @@ -560,6 +560,7 @@ int kvm_pgtable_stage2_set_owner(struct kvm_pgtable *pgt, u64 addr, u64 size, * @pgt: Page-table structure initialised by kvm_pgtable_stage2_init*(). * @addr: Intermediate physical address from which to remove the mapping. * @size: Size of the mapping. + * @flags: Page-table walker flags. * * The offset of @addr within a page is ignored and @size is rounded-up to * the next page boundary. @@ -572,7 +573,8 @@ int kvm_pgtable_stage2_set_owner(struct kvm_pgtable *pgt, u64 addr, u64 size, * * Return: 0 on success, negative error code on failure. */ -int kvm_pgtable_stage2_unmap(struct kvm_pgtable *pgt, u64 addr, u64 size); +int kvm_pgtable_stage2_unmap(struct kvm_pgtable *pgt, u64 addr, u64 size, + enum kvm_pgtable_walk_flags flags); /** * kvm_pgtable_stage2_wrprotect() - Write-protect guest stage-2 address range @@ -580,6 +582,7 @@ int kvm_pgtable_stage2_unmap(struct kvm_pgtable *pgt, u64 addr, u64 size); * @pgt: Page-table structure initialised by kvm_pgtable_stage2_init*(). * @addr: Intermediate physical address from which to write-protect, * @size: Size of the range. + * @flags: Page-table walker flags. * * The offset of @addr within a page is ignored and @size is rounded-up to * the next page boundary. @@ -590,7 +593,8 @@ int kvm_pgtable_stage2_unmap(struct kvm_pgtable *pgt, u64 addr, u64 size); * * Return: 0 on success, negative error code on failure. */ -int kvm_pgtable_stage2_wrprotect(struct kvm_pgtable *pgt, u64 addr, u64 size); +int kvm_pgtable_stage2_wrprotect(struct kvm_pgtable *pgt, u64 addr, u64 size, + enum kvm_pgtable_walk_flags flags); /** * kvm_pgtable_stage2_mkyoung() - Set the access flag in a page-table entry. @@ -662,13 +666,15 @@ bool kvm_pgtable_stage2_is_young(struct kvm_pgtable *pgt, u64 addr); * @pgt: Page-table structure initialised by kvm_pgtable_stage2_init*(). * @addr: Intermediate physical address from which to flush. * @size: Size of the range. + * @flags: Page-table walker flags. * * The offset of @addr within a page is ignored and @size is rounded-up to * the next page boundary. * * Return: 0 on success, negative error code on failure. */ -int kvm_pgtable_stage2_flush(struct kvm_pgtable *pgt, u64 addr, u64 size); +int kvm_pgtable_stage2_flush(struct kvm_pgtable *pgt, u64 addr, u64 size, + enum kvm_pgtable_walk_flags flags); /** * kvm_pgtable_stage2_split() - Split a range of huge pages into leaf PTEs pointing diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c index d35e75b13ffe..13f5cf5f87c3 100644 --- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c +++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c @@ -333,11 +333,11 @@ static int host_stage2_unmap_dev_all(void) /* Unmap all non-memory regions to recycle the pages */ for (i = 0; i < hyp_memblock_nr; i++, addr = reg->base + reg->size) { reg = &hyp_memory[i]; - ret = kvm_pgtable_stage2_unmap(pgt, addr, reg->base - addr); + ret = kvm_pgtable_stage2_unmap(pgt, addr, reg->base - addr, 0); if (ret) return ret; } - return kvm_pgtable_stage2_unmap(pgt, addr, BIT(pgt->ia_bits) - addr); + return kvm_pgtable_stage2_unmap(pgt, addr, BIT(pgt->ia_bits) - addr, 0); } struct kvm_mem_range { diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 364b68013038..a3a0812b2301 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -1044,12 +1044,14 @@ static int stage2_unmap_walker(const struct kvm_pgtable_visit_ctx *ctx, return 0; } -int kvm_pgtable_stage2_unmap(struct kvm_pgtable *pgt, u64 addr, u64 size) +int kvm_pgtable_stage2_unmap(struct kvm_pgtable *pgt, u64 addr, u64 size, + enum kvm_pgtable_walk_flags flags) { struct kvm_pgtable_walker walker = { .cb = stage2_unmap_walker, .arg = pgt, - .flags = KVM_PGTABLE_WALK_LEAF | KVM_PGTABLE_WALK_TABLE_POST, + .flags = flags | KVM_PGTABLE_WALK_LEAF | + KVM_PGTABLE_WALK_TABLE_POST, }; return kvm_pgtable_walk(pgt, addr, size, &walker); @@ -1128,11 +1130,12 @@ static int stage2_update_leaf_attrs(struct kvm_pgtable *pgt, u64 addr, return 0; } -int kvm_pgtable_stage2_wrprotect(struct kvm_pgtable *pgt, u64 addr, u64 size) +int kvm_pgtable_stage2_wrprotect(struct kvm_pgtable *pgt, u64 addr, u64 size, + enum kvm_pgtable_walk_flags flags) { return stage2_update_leaf_attrs(pgt, addr, size, 0, KVM_PTE_LEAF_ATTR_LO_S2_S2AP_W, - NULL, NULL, 0); + NULL, NULL, flags); } kvm_pte_t kvm_pgtable_stage2_mkyoung(struct kvm_pgtable *pgt, u64 addr) @@ -1213,11 +1216,12 @@ static int stage2_flush_walker(const struct kvm_pgtable_visit_ctx *ctx, return 0; } -int kvm_pgtable_stage2_flush(struct kvm_pgtable *pgt, u64 addr, u64 size) +int kvm_pgtable_stage2_flush(struct kvm_pgtable *pgt, u64 addr, u64 size, + enum kvm_pgtable_walk_flags flags) { struct kvm_pgtable_walker walker = { .cb = stage2_flush_walker, - .flags = KVM_PGTABLE_WALK_LEAF, + .flags = flags | KVM_PGTABLE_WALK_LEAF, .arg = pgt, }; diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 0c2c2c0846f1..1030921d89f8 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -55,7 +55,9 @@ static phys_addr_t stage2_range_addr_end(phys_addr_t addr, phys_addr_t end) */ static int stage2_apply_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, phys_addr_t end, - int (*fn)(struct kvm_pgtable *, u64, u64), + enum kvm_pgtable_walk_flags flags, + int (*fn)(struct kvm_pgtable *, u64, u64, + enum kvm_pgtable_walk_flags), bool resched) { struct kvm *kvm = kvm_s2_mmu_to_kvm(mmu); @@ -68,7 +70,7 @@ static int stage2_apply_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, return -EINVAL; next = stage2_range_addr_end(addr, end); - ret = fn(pgt, addr, next - addr); + ret = fn(pgt, addr, next - addr, flags); if (ret) break; @@ -79,8 +81,8 @@ static int stage2_apply_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, return ret; } -#define stage2_apply_range_resched(mmu, addr, end, fn) \ - stage2_apply_range(mmu, addr, end, fn, true) +#define stage2_apply_range_resched(mmu, addr, end, flags, fn) \ + stage2_apply_range(mmu, addr, end, flags, fn, true) /* * Get the maximum number of page-tables pages needed to split a range @@ -316,7 +318,7 @@ static void __unmap_stage2_range(struct kvm_s2_mmu *mmu, phys_addr_t start, u64 lockdep_assert_held_write(&kvm->mmu_lock); WARN_ON(size & ~PAGE_MASK); - WARN_ON(stage2_apply_range(mmu, start, end, kvm_pgtable_stage2_unmap, + WARN_ON(stage2_apply_range(mmu, start, end, 0, kvm_pgtable_stage2_unmap, may_block)); } @@ -331,7 +333,8 @@ static void stage2_flush_memslot(struct kvm *kvm, phys_addr_t addr = memslot->base_gfn << PAGE_SHIFT; phys_addr_t end = addr + PAGE_SIZE * memslot->npages; - stage2_apply_range_resched(&kvm->arch.mmu, addr, end, kvm_pgtable_stage2_flush); + stage2_apply_range_resched(&kvm->arch.mmu, addr, end, 0, + kvm_pgtable_stage2_flush); } /** @@ -1041,10 +1044,13 @@ int kvm_phys_addr_ioremap(struct kvm *kvm, phys_addr_t guest_ipa, * @mmu: The KVM stage-2 MMU pointer * @addr: Start address of range * @end: End address of range + * @flags: Page-table walker flags. */ -static void stage2_wp_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, phys_addr_t end) +static void stage2_wp_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, phys_addr_t end, + enum kvm_pgtable_walk_flags flags) { - stage2_apply_range_resched(mmu, addr, end, kvm_pgtable_stage2_wrprotect); + stage2_apply_range_resched(mmu, addr, end, flags, + kvm_pgtable_stage2_wrprotect); } /** @@ -1073,7 +1079,7 @@ static void kvm_mmu_wp_memory_region(struct kvm *kvm, int slot) end = (memslot->base_gfn + memslot->npages) << PAGE_SHIFT; write_lock(&kvm->mmu_lock); - stage2_wp_range(&kvm->arch.mmu, start, end); + stage2_wp_range(&kvm->arch.mmu, start, end, 0); write_unlock(&kvm->mmu_lock); kvm_flush_remote_tlbs(kvm); } @@ -1128,7 +1134,7 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, write_lock(&kvm->mmu_lock); lockdep_assert_held_write(&kvm->mmu_lock); - stage2_wp_range(&kvm->arch.mmu, start, end); + stage2_wp_range(&kvm->arch.mmu, start, end, 0); /* * Eager-splitting is done when manual-protect is set. We From patchwork Fri Jun 2 16:09:07 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265611 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1D68DC7EE37 for ; Fri, 2 Jun 2023 16:10:10 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236726AbjFBQKJ (ORCPT ); Fri, 2 Jun 2023 12:10:09 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35362 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236812AbjFBQJw (ORCPT ); Fri, 2 Jun 2023 12:09:52 -0400 Received: from mail-yw1-x1149.google.com (mail-yw1-x1149.google.com [IPv6:2607:f8b0:4864:20::1149]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4063E10DF for ; Fri, 2 Jun 2023 09:09:39 -0700 (PDT) Received: by mail-yw1-x1149.google.com with SMTP id 00721157ae682-565a1788f3fso33704177b3.0 for ; Fri, 02 Jun 2023 09:09:39 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722178; x=1688314178; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=NyrihgmYesBmBKLp4ITVdJ3LGNh6LVDPlJHbJ3BWmKo=; b=6JvB68luEiFeoyKEfjP/g+vMwC/sRddTQaYllS5g+5bpQDmhXIgwwZiHHiYhNY6/hh +N3LQSiNtnyna/XZaiM18ivCXXQ4yuX/15A1AGZwy0Qa/f1hIAf3oKXB2h5AuHkYQgCs xj8Y82H0f8TZvkTqQ0Hx5MWBndbNiI7NCzZ3czEuYGWHs7LAUCirzskxbkVOl+noMZFx tm27fSCTKAwfbeEstcfEbI/nW6D9qAihf0lb2MdszgnmyhfNayDhGNG1boLJ+W+UNY9A PKxN0bCcFoMuNsJAFUnK+mzP9snrdOrZPjuzoVjhjXEuIkdNPWm6zU7rdhcbdNuxMxHK DaOA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722178; x=1688314178; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=NyrihgmYesBmBKLp4ITVdJ3LGNh6LVDPlJHbJ3BWmKo=; b=cckZSMgf5VTlQwerKy4Plf0nyjvfobXQ74G9gscIxUTORh0hM3Ub799GWqohbDQIQQ iCVYDlnIqhLi+Ai0wAVkkb3csexqoQj9/j4uYqsJvTgwWrxyImvGc54ZBs+gsm6jgkcY Wh6Ces/lWSU2aZ+ObBsvuUDNKBu4TWkcGQi96USGA/gNz9of/bQr6GfH9+TyldRtt0QC pGclvxQHuTwvF3H6vjxO5TMlCZU3QO+9E/dvjYPL2H0IG9rD0CRT5vKzXnyfbFEwNR9l ZgtlmLVOGEnSoOs8Xtn59xtp487wM+Y6gRdeM+rTpepr2F2MY2GXzOvanHU47Cgo+ePn cVNA== X-Gm-Message-State: AC+VfDzxEg516Aw8n3HwzLgVpxqcHUnm82tZaATOQPhQEbsp3WpJ31hl jA8NkExQjI2p3QQmUMtndR2WPKljB0IK X-Google-Smtp-Source: ACHHUZ7z74Jy0Y96qFW+7u40ZiTgX4ucCC/5bWAvnbf5ZGF1cU7Xjv0z02cQ2QpbWgJAAII5qTnOdyVeR9zk X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a81:b620:0:b0:561:b8d1:743b with SMTP id u32-20020a81b620000000b00561b8d1743bmr170869ywh.10.1685722178126; Fri, 02 Jun 2023 09:09:38 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:07 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-10-vipinsh@google.com> Subject: [PATCH v2 09/16] KVM: arm64: Document the page table walker actions based on the callback's return value From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Document what the page table walker do when walker callback function returns a value. Current documentation is not correct as negative error of -EAGAIN on a non-shared page table walker doesn't terminate the walker and continues to the next step. There might be a better place to keep this information, for now this documentation will work as a reference guide until a better way is found. Signed-off-by: Vipin Sharma --- arch/arm64/include/asm/kvm_pgtable.h | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h index 8ef7e8f3f054..957bc20dab00 100644 --- a/arch/arm64/include/asm/kvm_pgtable.h +++ b/arch/arm64/include/asm/kvm_pgtable.h @@ -711,8 +711,19 @@ int kvm_pgtable_stage2_split(struct kvm_pgtable *pgt, u64 addr, u64 size, * after invoking the walker callback, allowing the walker to descend into * a newly installed table. * - * Returning a negative error code from the walker callback function will - * terminate the walk immediately with the same error code. + * Depending on the return value from the walker callback function, the page + * table walk will continue or exit the walk. This is also dependent on the + * type of the walker, i.e. shared walker (vCPU fault handlers) or non-shared + * walker. + * + * Walker Type | Callback | Walker action + * -------------|------------------|-------------- + * Non-Shared | 0 | Continue + * Non-Shared | -EAGAIN | Continue + * Non-Shared | Any other | Exit + * -------------|------------------|-------------- + * Shared | 0 | Continue + * Shared | Any other | Exit * * Return: 0 on success, negative error code on failure. */ From patchwork Fri Jun 2 16:09:08 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265613 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 864E9C7EE24 for ; Fri, 2 Jun 2023 16:10:14 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236593AbjFBQKL (ORCPT ); Fri, 2 Jun 2023 12:10:11 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35362 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236850AbjFBQJ5 (ORCPT ); Fri, 2 Jun 2023 12:09:57 -0400 Received: from mail-pg1-x54a.google.com (mail-pg1-x54a.google.com [IPv6:2607:f8b0:4864:20::54a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4660810FE for ; Fri, 2 Jun 2023 09:09:41 -0700 (PDT) Received: by mail-pg1-x54a.google.com with SMTP id 41be03b00d2f7-5343c1d114cso2240725a12.0 for ; Fri, 02 Jun 2023 09:09:41 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722180; x=1688314180; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=i16qvWrTxXmM44zqvAfMO+VQ2ZgLrONxQcP35eFduXw=; b=7epqj+D7b/Vs8MJCaB/Yaig4pdk6Xsa6+PhHha/KiUX7vkzJuwNVRvh/QKSBOYWOoF 9Kl1CGagQi1DXxlGyYvTod7C6g15Jc/lvseccbcVTnZGlpUMuN85YotpNb24SiVNDe/4 jJWg+BewppQ9iRi+5TbG/EFvPtgidVbx51XdS+ZgnSZfKoodj1WI4tk7ctB/2mpU7ujt 5GR5L1lxyi6Yc/dQP61OizppKV4BxJWb97C2tfKTvu6BWLjF7QcXhyPnKGED9xXxbRV0 rkKjEao0Yem9n3mVyvzo3PMl7x2joHF/KT0tNezMLtd5RxYZ9imnBCUvokvJr8Ssuvgw mKhg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722180; x=1688314180; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=i16qvWrTxXmM44zqvAfMO+VQ2ZgLrONxQcP35eFduXw=; b=Ok/GuxhYMKCU8jcQc9hHy4trh3Qx/mFfEEBIrIscrBqvm4F2/5PsMvTbRBE0rAhO17 QXu86lv0vYw8MgYW9naK0t3ZKAqc2DBkC79sdf3f+V6yQWlwlCBnnqA1hNmlKSKXYDt6 i8iHtZFoubGnhw0O66CbjimagoX+9jf4hbvshjyFOZBEqctYtwrBscs0xrhoswBGZI/e 0yKgjbvIZ/Fq1eqynSAj6JE2uK2IX/Ahwl0mZNQYD/Y0wX5a5cfxvbrI6rP2SdNJrvII gnV2v6qMRzZtKYIoh55emJ6cx442urtrEFrflcYpCfN/g5F9DjIYaHQOUKAbSuBAoOiu /tiw== X-Gm-Message-State: AC+VfDxAQqrqzDxWNznr4cvKz3aoAZ74EwxWaza9BsI2nLm1efN03ssC XLsAfg6CABJDNWjt0W3aSiH5HZNFyu7F X-Google-Smtp-Source: ACHHUZ50eFmYtmjd+7phpNHJVAKKYKF0b2Pf24cv0bbiOkOy3oeWBFiQvv39uMwyDgbA4q0xlnPSa5qY2vtt X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a17:903:3251:b0:1b0:6a10:1ba1 with SMTP id ji17-20020a170903325100b001b06a101ba1mr117384plb.13.1685722180001; Fri, 02 Jun 2023 09:09:40 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:08 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-11-vipinsh@google.com> Subject: [PATCH v2 10/16] KVM: arm64: Return -ENOENT if PTE is not valid in stage2_attr_walker From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Return -ENOENT from stage2_attr_walker for invalid PTE. Continue page table walk if walker callback returns -ENOENT outside of the fault handler path else terminate the walk. In fault handler path, similar to -EAGAIN in user_mem_abort, retry guest execution. stage2_attr_walker() is used from multiple places like, write protection, MMU notifier callbacks, and relaxing permission during vCPU faults. This function returns -EAGAIN for different cases: 1. When PTE is not valid. 2. When cmpxchg() fails while setting new SPTE. For non-shared walkers, like write protection and MMU notifier, above 2 cases are just ignored by walker and it moves to the next SPTE. #2 will never happen for non-shared walkers as they don't use cmpxchg() for updating SPTEs. For shared walkers, like vCPU fault handler, above 2 cases results in walk termination. In future commits, clear-dirty-log walker will write protect SPTEs under MMU read lock and use shared page table walker. This will result in two shared page table walkers type, vCPUs fault handler and clear-dirty-log, competing with each other and sometime causing cmpxchg() failure. So, -EAGAIN in clear-dirty-log walker due to cmpxchg() failure must be retried. Whereas, -EAGAIN in the clear-dirty-log due to invalid SPTE must be ignored instead of exiting as per the current logic of shared page table walker. This is not needed for vCPU fault handler which also runs via shared page table walker and terminates walk on getting -EAGAIN due to invalid SPTE. To handle all these scenarios, stage2_attr_walker must return different error codes for invalid SPTEs and cmxchg() failure. -ENOENT for invalid SPTE is chosen because it is not used by any other shared walker. When clear-dirty-log will be changed to use shared page table walker, it will be possible to differentiate cases of retrying, continuing or terminating the walk for shared fault handler and shared clear-dirty-log. Signed-off-by: Vipin Sharma --- arch/arm64/include/asm/kvm_pgtable.h | 1 + arch/arm64/kvm/hyp/pgtable.c | 19 ++++++++++++------- arch/arm64/kvm/mmu.c | 2 +- 3 files changed, 14 insertions(+), 8 deletions(-) diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h index 957bc20dab00..23e7e7851f1d 100644 --- a/arch/arm64/include/asm/kvm_pgtable.h +++ b/arch/arm64/include/asm/kvm_pgtable.h @@ -720,6 +720,7 @@ int kvm_pgtable_stage2_split(struct kvm_pgtable *pgt, u64 addr, u64 size, * -------------|------------------|-------------- * Non-Shared | 0 | Continue * Non-Shared | -EAGAIN | Continue + * Non-Shared | -ENOENT | Continue * Non-Shared | Any other | Exit * -------------|------------------|-------------- * Shared | 0 | Continue diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index a3a0812b2301..bc8c5c4ac1cf 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -186,14 +186,19 @@ static bool kvm_pgtable_walk_continue(const struct kvm_pgtable_walker *walker, /* * Visitor callbacks return EAGAIN when the conditions that led to a * fault are no longer reflected in the page tables due to a race to - * update a PTE. In the context of a fault handler this is interpreted - * as a signal to retry guest execution. + * update a PTE. * - * Ignore the return code altogether for walkers outside a fault handler - * (e.g. write protecting a range of memory) and chug along with the - * page table walk. + * Callbacks can also return ENOENT when PTE which is visited is not + * valid. + * + * In the context of a fault handler interpret these as a signal + * to retry guest execution. + * + * Ignore these return codes altogether for walkers outside a fault + * handler (e.g. write protecting a range of memory) and chug along + * with the page table walk. */ - if (r == -EAGAIN) + if (r == -EAGAIN || r == -ENOENT) return !(walker->flags & KVM_PGTABLE_WALK_HANDLE_FAULT); return !r; @@ -1072,7 +1077,7 @@ static int stage2_attr_walker(const struct kvm_pgtable_visit_ctx *ctx, struct kvm_pgtable_mm_ops *mm_ops = ctx->mm_ops; if (!kvm_pte_valid(ctx->old)) - return -EAGAIN; + return -ENOENT; data->level = ctx->level; data->pte = pte; diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 1030921d89f8..356dc4131023 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -1551,7 +1551,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, read_unlock(&kvm->mmu_lock); kvm_set_pfn_accessed(pfn); kvm_release_pfn_clean(pfn); - return ret != -EAGAIN ? ret : 0; + return (ret != -EAGAIN && ret != -ENOENT) ? ret : 0; } /* Resolve the access fault by making the page young again. */ From patchwork Fri Jun 2 16:09:09 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265614 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 87B82C77B7A for ; Fri, 2 Jun 2023 16:10:51 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235665AbjFBQKs (ORCPT ); Fri, 2 Jun 2023 12:10:48 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35812 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236724AbjFBQK3 (ORCPT ); Fri, 2 Jun 2023 12:10:29 -0400 Received: from mail-pl1-x649.google.com (mail-pl1-x649.google.com [IPv6:2607:f8b0:4864:20::649]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id AFD29E62 for ; Fri, 2 Jun 2023 09:09:56 -0700 (PDT) Received: by mail-pl1-x649.google.com with SMTP id d9443c01a7336-1b0314f0aadso15276985ad.0 for ; Fri, 02 Jun 2023 09:09:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722181; x=1688314181; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=rivHDhk0ioxMpP1XynwGaSZFsNZDBBbqgVAGbR7E/X8=; b=IxpZG4TkYuPd8UtpcDQplKJDDIpAs6lGKWZe95duADLUk5Xly6SRvjrVvCEGB/Fk1a E9X3IJECzkdQ25Bkjejj4IbXek6JqPY58HRA50EOI/YFTzUpFaA/bzGSlAOm+5PEvo+T 6jKSmBS0KSmdPni70wpBlwTPrvyYGZvMtxSC3Fbri9jwwzePk7OY91lpj01vmu9g7PZn a5J+YN0h1Qu/z/w7ubpNpCMDwcQqYWttGGXgoLeySgwUxIZDyVDmv/3Bk/4OJqAcfD0+ gHcpPqAo0vTPvSXpzWrIqDWuTjrALALhGQwaLZCyItTHvmVctEjS/JfpfTvctKcX+1gM QFLQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722181; x=1688314181; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=rivHDhk0ioxMpP1XynwGaSZFsNZDBBbqgVAGbR7E/X8=; b=ZVr52YiNcbwTATQpafF4FDeea2D5EDncprLTRyMZrodBo5JD7GG087ahLGs2jm6W6x GvR1q2dXlfzaCU3+4O6pHSkvMcbt0VVlhywLAD59YfKaM0uZlswkIc6DCA0uIDNpYmkk GYGzRf6O9SAwJ9IlQskLGIuKcY9Gq0xdv/Sb5a+kjoZFTmPIT9YGt/k6bNSXI0wuhix2 7RWldakVh8BYF2kgqq26jrVoFG3dLV7mHK/p7/29n1snJbBdqRmCtuMeoGC68NNHB/Ic YUExKy4pddsQ9yFbW56sz17iMW2nvUQa4TOEZYChdlcSOA/V3IxZ48l513842/ZDabFd H56Q== X-Gm-Message-State: AC+VfDzh2wsVZfOEPqalHupNHB2TTVqkYuirPx20JF1mbfZaB7tZ9ltU aMEvxJy56Z8W3oQaGaxJ6idhWy77A68+ X-Google-Smtp-Source: ACHHUZ4crxpGe15BxcOcaGOZO2WjPbibX9jRmjl7rSrwdD8Z6/aya1uE9CJM9AhjU7HHArKs+z3PEQeSOj8s X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a17:902:e3c5:b0:1b1:c90e:b7aa with SMTP id r5-20020a170902e3c500b001b1c90eb7aamr56727ple.4.1685722181662; Fri, 02 Jun 2023 09:09:41 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:09 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-12-vipinsh@google.com> Subject: [PATCH v2 11/16] KVM: arm64: Use KVM_PGTABLE_WALK_SHARED flag instead of KVM_PGTABLE_WALK_HANDLE_FAULT From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Check against shared page table walker flag instead of fault handler flag when determining if walk should continue or not. vCPU page fault handlers uses shared page walker and there are no other shared page walkers in Arm. This will change in future commit when clear-dirty-log will use shared page walker and continue, retry or terminate logic for a walk will change between shared page walkers. Signed-off-by: Vipin Sharma --- arch/arm64/kvm/hyp/pgtable.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index bc8c5c4ac1cf..7f80e953b502 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -191,7 +191,7 @@ static bool kvm_pgtable_walk_continue(const struct kvm_pgtable_walker *walker, * Callbacks can also return ENOENT when PTE which is visited is not * valid. * - * In the context of a fault handler interpret these as a signal + * In the context of a shared walker interpret these as a signal * to retry guest execution. * * Ignore these return codes altogether for walkers outside a fault @@ -199,7 +199,7 @@ static bool kvm_pgtable_walk_continue(const struct kvm_pgtable_walker *walker, * with the page table walk. */ if (r == -EAGAIN || r == -ENOENT) - return !(walker->flags & KVM_PGTABLE_WALK_HANDLE_FAULT); + return !(walker->flags & KVM_PGTABLE_WALK_SHARED); return !r; } From patchwork Fri Jun 2 16:09:10 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265615 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 48566C7EE2D for ; Fri, 2 Jun 2023 16:11:02 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236616AbjFBQLA (ORCPT ); Fri, 2 Jun 2023 12:11:00 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35822 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236529AbjFBQKl (ORCPT ); Fri, 2 Jun 2023 12:10:41 -0400 Received: from mail-pj1-x104a.google.com (mail-pj1-x104a.google.com [IPv6:2607:f8b0:4864:20::104a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 172F2E76 for ; Fri, 2 Jun 2023 09:10:01 -0700 (PDT) Received: by mail-pj1-x104a.google.com with SMTP id 98e67ed59e1d1-2565bf593e7so845489a91.2 for ; Fri, 02 Jun 2023 09:10:01 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722184; x=1688314184; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=dYi36v/YAwaQRjgMVH+8EY8FA7aBBtQ+BKuLP6eDkvQ=; b=Dw12ocPSe9vJUzYvSsSf9AU3CWoiwTk/XWOJWSQ28tLpoS2cFvmMFQbSf/qpjpxy7v oO7YPmr7slJeCTpuInsqqCsWP0mQqS3ua3oN9mzqlTzskeKCsUc+Eng3akxDWWH0DIw4 jhV0Z/aBx0lDztLRsncoYBKeU0QZUKmq2gyk1zRIGBfBu0xTC7gF0pRwGDV7061++Ztg o5Q/lPpRn/0BOjk3RHl0WBfcrcdb3R1NScbQhD6quOJddTL53tPtBy4zB10NE0mDM/7t Ap9iqqT0iz8St+1DIUrYSZSYDvTKdSKGgplDSGDJzXhKejoEUVmNl+mnTfSIjic7IX8O rhDA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722184; x=1688314184; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=dYi36v/YAwaQRjgMVH+8EY8FA7aBBtQ+BKuLP6eDkvQ=; b=AenINZQOE6w2Gn58Xc+En57g+CkVZeX4WrhRm4ZygObZK6KLyqWpIqvIO7L/yfo+bo Qm7WUAkbPqv4GaBCAjn7UBRB7N8x9/Kh1rc09naZt5zvzHejU2MfLaHVbAUdJ6a+YPa9 gwg6e9V23ynZdVlMrYs+X4CmDmrDeTo201S++/iIN/6KzPpLA8dcXGFYCWCbeMy6fPXJ CTdzeLGNaDdkTnhomoYCAKE3LAPYB4EA3qYHGsQWbSQdPLBmI/QS8CsiaOx91h9p0mDW agMsr2YTjuy0d7Am+w3ouEF23IIH9yyclaAQ2vVj5MoJCELlY4CP8AxtJZd3P3ZYyqVC 01hw== X-Gm-Message-State: AC+VfDwFMQ5FM15VCNvpQzuQXXXGSDjXPzNgo2PDX5SV2Lln0JMwl4ed bDKiL5hYQfYhAHPtVV5c5doNQa1O8qSF X-Google-Smtp-Source: ACHHUZ4+c4f5xcGK/eLSU3xHgu8tOZbN6Cv39+hh+7NNc/cJ0dF1+UQE1gqaCbeHO9GT4u+sOrr9aoJh/Nwo X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a17:90a:c08e:b0:252:733d:15dc with SMTP id o14-20020a17090ac08e00b00252733d15dcmr97816pjs.2.1685722183850; Fri, 02 Jun 2023 09:09:43 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:10 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-13-vipinsh@google.com> Subject: [PATCH v2 12/16] KVM: arm64: Retry shared page table walks outside of fault handler From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org For a shared page walker which is not fault handler, retry the walk if walker callback function returns -EAGAIN, or continue to the next SPTE if callback function return -ENOENT. Update the kvm_pgtable_walk documentation. For fault handler logic remains same, i.e. exit the walk and resume the guest when getting -EAGAIN and -ENOENT errors from walker callback function. Currently, there is no page walker which is shared and not a fault handler, but this will change in future patches when clear-dirty-log walker will use MMU read lock and run via shared walker. Signed-off-by: Vipin Sharma --- arch/arm64/include/asm/kvm_pgtable.h | 23 ++++++++++------- arch/arm64/kvm/hyp/pgtable.c | 38 +++++++++++++++++++++++----- 2 files changed, 46 insertions(+), 15 deletions(-) diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h index 23e7e7851f1d..145be12a5fc2 100644 --- a/arch/arm64/include/asm/kvm_pgtable.h +++ b/arch/arm64/include/asm/kvm_pgtable.h @@ -716,15 +716,20 @@ int kvm_pgtable_stage2_split(struct kvm_pgtable *pgt, u64 addr, u64 size, * type of the walker, i.e. shared walker (vCPU fault handlers) or non-shared * walker. * - * Walker Type | Callback | Walker action - * -------------|------------------|-------------- - * Non-Shared | 0 | Continue - * Non-Shared | -EAGAIN | Continue - * Non-Shared | -ENOENT | Continue - * Non-Shared | Any other | Exit - * -------------|------------------|-------------- - * Shared | 0 | Continue - * Shared | Any other | Exit + * Walker Type | Callback | Walker action + * -----------------------|------------------|-------------- + * Non-Shared | 0 | Continue + * Non-Shared | -EAGAIN | Continue + * Non-Shared | -ENOENT | Continue + * Non-Shared | Any other | Exit + * -----------------------|------------------|-------------- + * Shared | 0 | Continue + * Shared | -EAGAIN | Retry + * Shared | -ENOENT | Continue + * Shared | Any other | Exit + * -----------------------|------------------|-------------- + * Shared (Fault Handler) | 0 | Continue + * Shared (Fault Handler) | Any other | Exit * * Return: 0 on success, negative error code on failure. */ diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 7f80e953b502..23cda3de2dd4 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -191,15 +191,21 @@ static bool kvm_pgtable_walk_continue(const struct kvm_pgtable_walker *walker, * Callbacks can also return ENOENT when PTE which is visited is not * valid. * - * In the context of a shared walker interpret these as a signal + * In the context of a fault handler interpret these as a signal * to retry guest execution. * - * Ignore these return codes altogether for walkers outside a fault - * handler (e.g. write protecting a range of memory) and chug along + * In the context of a shared walker which is not fault handler + * interpret: + * 1. EAGAIN - A signal to retry walk again. + * 2. ENOENT - A signal to ignore and move on to next SPTE. + * + * Ignore these return codes altogether for other walkers and chug along * with the page table walk. */ - if (r == -EAGAIN || r == -ENOENT) + if (r == -EAGAIN) return !(walker->flags & KVM_PGTABLE_WALK_SHARED); + if (r == -ENOENT) + return !(walker->flags & KVM_PGTABLE_WALK_HANDLE_FAULT); return !r; } @@ -260,24 +266,44 @@ static inline int __kvm_pgtable_visit(struct kvm_pgtable_walk_data *data, return ret; } +static bool kvm_pgtable_walk_retry(const struct kvm_pgtable_walker *walker, + int r) +{ + /* + * All shared page table walks where visitor callbacks return -EAGAIN + * should be retried with the exception of fault handler. In case of + * fault handler retry is achieved by resuming the guest. + */ + if (r == -EAGAIN) + return (walker->flags & KVM_PGTABLE_WALK_SHARED) && + !(walker->flags & KVM_PGTABLE_WALK_HANDLE_FAULT); + + return !r; +} + static int __kvm_pgtable_walk(struct kvm_pgtable_walk_data *data, struct kvm_pgtable_mm_ops *mm_ops, kvm_pteref_t pgtable, u32 level) { u32 idx; int ret = 0; + kvm_pteref_t pteref; if (WARN_ON_ONCE(level >= KVM_PGTABLE_MAX_LEVELS)) return -EINVAL; for (idx = kvm_pgtable_idx(data, level); idx < PTRS_PER_PTE; ++idx) { - kvm_pteref_t pteref = &pgtable[idx]; +retry: + pteref = &pgtable[idx]; if (data->addr >= data->end) break; ret = __kvm_pgtable_visit(data, mm_ops, pteref, level); - if (ret) + if (ret) { + if (kvm_pgtable_walk_retry(data->walker, ret)) + goto retry; break; + } } return ret; From patchwork Fri Jun 2 16:09:11 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265616 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 319E3C7EE2E for ; Fri, 2 Jun 2023 16:11:06 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236039AbjFBQLE (ORCPT ); Fri, 2 Jun 2023 12:11:04 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35814 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236615AbjFBQKv (ORCPT ); Fri, 2 Jun 2023 12:10:51 -0400 Received: from mail-yw1-x1149.google.com (mail-yw1-x1149.google.com [IPv6:2607:f8b0:4864:20::1149]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BB6A010D0 for ; Fri, 2 Jun 2023 09:10:06 -0700 (PDT) Received: by mail-yw1-x1149.google.com with SMTP id 00721157ae682-568960f4596so34413587b3.1 for ; Fri, 02 Jun 2023 09:10:06 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722186; x=1688314186; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=iiQKR5r12cEmfxEtqKB501qQ7U5/inaqlNqkSHrg1Dk=; b=pa3mVvglr9ldo0oASWJgWygRLd3hw8lG1pjLeB+Qzm3K3zOU5BQU88aja7wxcgPXwA AwayadZx6shxLAXjuJFrDzlMtsvb67kGNUVBIz8jQ9pcx5Jd6BF7pMABQixgPuk47qo/ QPSMIrvsitPTIbIvBs3Unmlez10KGpvyazPz0vuoWxeM0ddu7MUX6bOpsZv5FX/RHsfI ab0UhQ4BKG9ohwFpekgRN7rsQtHAtMFU2PEbvOsbX5+K7m399sRVLlUcmEJAu36XrUm4 MIK72o4Xoc2dnSjS77GyfH7dzmpmSt4gmenJO3sxGLvRMkP8fiKXjBjSppWa9vRlnGfd FwPA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722186; x=1688314186; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=iiQKR5r12cEmfxEtqKB501qQ7U5/inaqlNqkSHrg1Dk=; b=jtQXkU1MRKuAegVwAHYyCZ4hZh9ud6RS3g13lY1DTYmT0iCpKalHGrUyecReQDEFYV ZGvdBthFgIeAS9bf+aeThoO/l/3f3J7jmWaMGgbMJFAn9Ki6Uzcgx/YMEqz0bNw0ic/N 9EeX4yXDxKMrEOaXJ/xdjWFH705kgonC1D7ODLimLlAv61CGzDcgKsTI91CW1ZPP90hi +2ayuA6QLayljLqOFLh1V5V5kT2fpR/GkWRIciDjYnZvOG1aPajLJV9S2xCEGBBxs8MH E4QLRaIKNqc9E3Ixv1jRTZr1vsOmvRClcCEnaI01WGxqlJGBARxlZgnit/YwLBdoY/Nh lmMQ== X-Gm-Message-State: AC+VfDwnuERw7KaEV9EBv2DuzqAMeOw6y2Y/GyGXxbbW9AsH4o51dZrP 5jyHumzaQB0uZAxNljRkN+sYc+Tk0jPX X-Google-Smtp-Source: ACHHUZ4B/VNE5B0Lzqq5kfdIQx5Z2vYuoQBh8/63RZRhZfWL9mTILlGstZ7p9kQje0NOJUJPGy2sE9Fs2xL/ X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a05:6902:1545:b0:ba8:181b:2558 with SMTP id r5-20020a056902154500b00ba8181b2558mr2255805ybu.4.1685722186031; Fri, 02 Jun 2023 09:09:46 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:11 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-14-vipinsh@google.com> Subject: [PATCH v2 13/16] KVM: arm64: Run clear-dirty-log under MMU read lock From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Take MMU read lock for clearing dirty logs and use shared page table walker. Dirty logs are currently cleared using MMU write locks. This means vCPUs page faults, which takes MMU read lock, will be blocked while dirty logs are being cleared. This causes guest degradation and especially noticeable on VMs with lot of vCPUs. Taking MMU read lock will allow vCPUs to execute parallelly and reduces the impact on vCPUs performance. Signed-off-by: Vipin Sharma --- arch/arm64/kvm/mmu.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 356dc4131023..7c966f6f1a41 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -74,8 +74,12 @@ static int stage2_apply_range(struct kvm_s2_mmu *mmu, phys_addr_t addr, if (ret) break; - if (resched && next != end) - cond_resched_rwlock_write(&kvm->mmu_lock); + if (resched && next != end) { + if (flags & KVM_PGTABLE_WALK_SHARED) + cond_resched_rwlock_read(&kvm->mmu_lock); + else + cond_resched_rwlock_write(&kvm->mmu_lock); + } } while (addr = next, addr != end); return ret; @@ -1131,11 +1135,11 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, phys_addr_t start = (base_gfn + __ffs(mask)) << PAGE_SHIFT; phys_addr_t end = (base_gfn + __fls(mask) + 1) << PAGE_SHIFT; - write_lock(&kvm->mmu_lock); - lockdep_assert_held_write(&kvm->mmu_lock); - - stage2_wp_range(&kvm->arch.mmu, start, end, 0); + read_lock(&kvm->mmu_lock); + stage2_wp_range(&kvm->arch.mmu, start, end, KVM_PGTABLE_WALK_SHARED); + read_unlock(&kvm->mmu_lock); + write_lock(&kvm->mmu_lock); /* * Eager-splitting is done when manual-protect is set. We * also check for initially-all-set because we can avoid From patchwork Fri Jun 2 16:09:12 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265617 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 74CE4C77B7A for ; Fri, 2 Jun 2023 16:11:09 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236773AbjFBQLG (ORCPT ); Fri, 2 Jun 2023 12:11:06 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:35844 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236456AbjFBQKy (ORCPT ); Fri, 2 Jun 2023 12:10:54 -0400 Received: from mail-yb1-xb4a.google.com (mail-yb1-xb4a.google.com [IPv6:2607:f8b0:4864:20::b4a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 6EDF410D4 for ; Fri, 2 Jun 2023 09:10:07 -0700 (PDT) Received: by mail-yb1-xb4a.google.com with SMTP id 3f1490d57ef6-bacfa4eefd2so3129241276.0 for ; Fri, 02 Jun 2023 09:10:07 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722188; x=1688314188; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=ATfYeV4iZLydVfrv7VkfrrcrF6Iyfqli3wOIB/41cFU=; b=Rqyu16keB0kE0WrdwrfeOOljSmmTEf5ym2HasOLzHF8ArvodVvyp0S6Lr0lp8jhqaO yU7vlRlNbGqDi7zvQpbHT6R5zFRE8DckASYhOzfH9V34Q20hh77zuuNe12GxNtHCZXh5 lnbwZ2Svz+2HB81mY68YPRkYswYr7F2GvqW0LS+zjsO1U81VkyPUwY6eD2Ki1IgFqmLS 03LcT6Q3zPmFBx5Lk1wR2QR7kJ2u+CVAlb7S/W7GXhPrjwIAChkVcnGC08nkCFgLZ6+3 IJyon3TZZUbxyP2iFDZRsS+2chcqzC8RuTMktYv6Efxwhi8TOTgmYLyH3MqqUylTOe6h 6cPw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722188; x=1688314188; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=ATfYeV4iZLydVfrv7VkfrrcrF6Iyfqli3wOIB/41cFU=; b=MqsWU46vVyCHqIY9HDPXKNwSobppEv0tTlSxvz2ksmSIzbQYfLtqf7oizmvhY5rDrX 6txi1sTd3ZRS0YF+dF4X49iGKUzcCvqafFEwhnc3s/AU0NXo5VUVdw2bzgAwyIVUzgHi W6y+Eff35Kw9Pzt9XphSc9FdngExG3MsHVl+kwTbwzdAHPvmoeat8RJbL/xrb8MvK4Js B+4DHaeMhuk4DFYASbBtelnxcpDz1j4FqbTP7SeXR4wXcoVtHr9ya34Q1GH943ZHWOcp XGOyzQ+WdoI4pTjq22fzNUnsc48o5KU21y8BvlJABEjpQOfnQX/YPdUHKK/Sjk8N7v3O iWQA== X-Gm-Message-State: AC+VfDzzQnbmz2zdzHAn59UMKZPqYQ2PWfl76XzQ2VtBul3lFP4OQsjc NC4/tedWx6mtQEZlC9XZnx1wASOVWisW X-Google-Smtp-Source: ACHHUZ4IW8kR8j7bn7igS1kcMhvfZS4XncKRF7NLGQ8SgKKpwW4y3+9yGIJ+PiwzD8LO4pWUxsLhYeKpfotT X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a25:e706:0:b0:bac:5d2c:844b with SMTP id e6-20020a25e706000000b00bac5d2c844bmr1305917ybh.8.1685722187745; Fri, 02 Jun 2023 09:09:47 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:12 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-15-vipinsh@google.com> Subject: [PATCH v2 14/16] KVM: arm64: Pass page walker flags from callers of stage 2 split walker From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Pass enum kvm_pgtable_walk_flags{} to kvm_pgtable_stage2_split() walker from its caller. This allows split walker users to specify if they want to run split logic via shared walker or non-shared walker. Signed-off-by: Vipin Sharma --- arch/arm64/include/asm/kvm_pgtable.h | 4 +++- arch/arm64/kvm/hyp/pgtable.c | 5 +++-- arch/arm64/kvm/mmu.c | 2 +- 3 files changed, 7 insertions(+), 4 deletions(-) diff --git a/arch/arm64/include/asm/kvm_pgtable.h b/arch/arm64/include/asm/kvm_pgtable.h index 145be12a5fc2..fbf5c6c509fb 100644 --- a/arch/arm64/include/asm/kvm_pgtable.h +++ b/arch/arm64/include/asm/kvm_pgtable.h @@ -684,6 +684,7 @@ int kvm_pgtable_stage2_flush(struct kvm_pgtable *pgt, u64 addr, u64 size, * @size: Size of the range. * @mc: Cache of pre-allocated and zeroed memory from which to allocate * page-table pages. + * @flags: Page walker flags * * The function tries to split any level 1 or 2 entry that overlaps * with the input range (given by @addr and @size). @@ -693,7 +694,8 @@ int kvm_pgtable_stage2_flush(struct kvm_pgtable *pgt, u64 addr, u64 size, * blocks in the input range as allowed by @mc_capacity. */ int kvm_pgtable_stage2_split(struct kvm_pgtable *pgt, u64 addr, u64 size, - struct kvm_mmu_memory_cache *mc); + struct kvm_mmu_memory_cache *mc, + enum kvm_pgtable_walk_flags flags); /** * kvm_pgtable_walk() - Walk a page-table. diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 23cda3de2dd4..7e84be13d76d 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -1408,11 +1408,12 @@ static int stage2_split_walker(const struct kvm_pgtable_visit_ctx *ctx, } int kvm_pgtable_stage2_split(struct kvm_pgtable *pgt, u64 addr, u64 size, - struct kvm_mmu_memory_cache *mc) + struct kvm_mmu_memory_cache *mc, + enum kvm_pgtable_walk_flags flags) { struct kvm_pgtable_walker walker = { .cb = stage2_split_walker, - .flags = KVM_PGTABLE_WALK_LEAF, + .flags = flags | KVM_PGTABLE_WALK_LEAF, .arg = mc, }; diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 7c966f6f1a41..34d2bd03cf5f 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -153,7 +153,7 @@ static int kvm_mmu_split_huge_pages(struct kvm *kvm, phys_addr_t addr, return -EINVAL; next = __stage2_range_addr_end(addr, end, chunk_size); - ret = kvm_pgtable_stage2_split(pgt, addr, next - addr, cache); + ret = kvm_pgtable_stage2_split(pgt, addr, next - addr, cache, 0); if (ret) break; } while (addr = next, addr != end); From patchwork Fri Jun 2 16:09:13 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265618 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 447C9C7EE24 for ; Fri, 2 Jun 2023 16:11:10 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236698AbjFBQLJ (ORCPT ); Fri, 2 Jun 2023 12:11:09 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:36100 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236792AbjFBQK5 (ORCPT ); Fri, 2 Jun 2023 12:10:57 -0400 Received: from mail-yb1-xb49.google.com (mail-yb1-xb49.google.com [IPv6:2607:f8b0:4864:20::b49]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 4021810DF for ; Fri, 2 Jun 2023 09:10:10 -0700 (PDT) Received: by mail-yb1-xb49.google.com with SMTP id 3f1490d57ef6-ba8a5cbb012so2938144276.3 for ; Fri, 02 Jun 2023 09:10:10 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722190; x=1688314190; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=Hryh0ZWAFdBP/1ensclnMWOs6DFcoKq9t8dEIwwzMjM=; b=yFcSJ5sk1YiYcQktVHUrIraA4vq3gzBg01wawQ+99X99BcERdv6aw79BJglFIsX38X Jmk2e5UaolCr5t1qAVgv5xhRcuwqz8qsPeKTN4dMinvHjEGhpXtPGawN2ICvlDQZGj8D ZmOScsl1sdyn4gxehMyRauugNpIrQjpu1KxL67VdJydhpwrAe8TXpUojYixIDTXpkYLZ L/KcS63c1WkLy8I19gAWFgXf90HgC8c3L1xN5Hze37jt9RE4scxALJ3lb9KGpgKF7m+y IPNOsGvdWwTPZEtqpxeUuo6lpLAcPxZcX9jQeWMRNApjKkg5jyviIXgCTWMDwLFLuVTH Fsfg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722190; x=1688314190; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=Hryh0ZWAFdBP/1ensclnMWOs6DFcoKq9t8dEIwwzMjM=; b=WeRECJxWlu8jck00vbCGy/sXvTXbOLbZdQKSray0Ndi1OBbefw8dLwB4iBSc59tSQV 1vywCzVsd54dsdXJ5IkQ7ldHbhOIm9oMCttgVI5YwJ94vLnEJHoxLBQHRiij/1ujPX79 SqFZn86W2LGNT1p6MTRjQ3fl8gcRia1/MhmYScYt+QgnJ7UrEUeWBNHmOdGuDnjeprH0 5k4mI87QJ+wbXbJIOp7oDv8RiBwHrZS+7R3vcvls8966KS8jmVD6MmT50VxjwM/YBp/H oSFaqnjb7vmVAV6M+yULqmOwpQUOtL4DVPhfLGc10FLXnMkPYDHiaBm2fmfxzkXXbruk k3/Q== X-Gm-Message-State: AC+VfDyzKTzakswwJZ74p08Mh9eOF8eMV0ouINjJFKAt6X/POTzRMB7P esOOeNRc7Ci+J7BHnDzdb9EKuRHZo4fO X-Google-Smtp-Source: ACHHUZ5M9AG9HKhKFMo8prXUgGc/ucGnPOZDLd1OSRiqHslc40DT/QGSesfErKBawc3k99MHt4wj3TuQ4IOn X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a25:bc7:0:b0:ba8:cbd2:61b3 with SMTP id 190-20020a250bc7000000b00ba8cbd261b3mr1223005ybl.5.1685722190144; Fri, 02 Jun 2023 09:09:50 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:13 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-16-vipinsh@google.com> Subject: [PATCH v2 15/16] KVM: arm64: Provide option to pass page walker flag for huge page splits From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Pass enum kvm_pgtable_walk_flags{} to kvm_mmu_split_huge_pages(). Use 0 as the flag value to make it no-op. In future commit kvm_mmu_split_huge_pages() will be used under both MMU read lock and MMU write lock. Flag allows to pass intent to use shared or non-shared page walkers to split the huge pages. Signed-off-by: Vipin Sharma --- arch/arm64/kvm/mmu.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 34d2bd03cf5f..6dd964e3682c 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -118,7 +118,8 @@ static bool need_split_memcache_topup_or_resched(struct kvm *kvm) } static int kvm_mmu_split_huge_pages(struct kvm *kvm, phys_addr_t addr, - phys_addr_t end) + phys_addr_t end, + enum kvm_pgtable_walk_flags flags) { struct kvm_mmu_memory_cache *cache; struct kvm_pgtable *pgt; @@ -153,7 +154,8 @@ static int kvm_mmu_split_huge_pages(struct kvm *kvm, phys_addr_t addr, return -EINVAL; next = __stage2_range_addr_end(addr, end, chunk_size); - ret = kvm_pgtable_stage2_split(pgt, addr, next - addr, cache, 0); + ret = kvm_pgtable_stage2_split(pgt, addr, next - addr, cache, + flags); if (ret) break; } while (addr = next, addr != end); @@ -1112,7 +1114,7 @@ static void kvm_mmu_split_memory_region(struct kvm *kvm, int slot) end = (memslot->base_gfn + memslot->npages) << PAGE_SHIFT; write_lock(&kvm->mmu_lock); - kvm_mmu_split_huge_pages(kvm, start, end); + kvm_mmu_split_huge_pages(kvm, start, end, 0); write_unlock(&kvm->mmu_lock); } @@ -1149,7 +1151,7 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, * again. */ if (kvm_dirty_log_manual_protect_and_init_set(kvm)) - kvm_mmu_split_huge_pages(kvm, start, end); + kvm_mmu_split_huge_pages(kvm, start, end, 0); write_unlock(&kvm->mmu_lock); } From patchwork Fri Jun 2 16:09:14 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Vipin Sharma X-Patchwork-Id: 13265619 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1C494C7EE2A for ; Fri, 2 Jun 2023 16:11:22 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S236888AbjFBQLS (ORCPT ); Fri, 2 Jun 2023 12:11:18 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:36190 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S236801AbjFBQK6 (ORCPT ); Fri, 2 Jun 2023 12:10:58 -0400 Received: from mail-pf1-x44a.google.com (mail-pf1-x44a.google.com [IPv6:2607:f8b0:4864:20::44a]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 0E8951B9 for ; Fri, 2 Jun 2023 09:10:13 -0700 (PDT) Received: by mail-pf1-x44a.google.com with SMTP id d2e1a72fcca58-65320a6e904so380505b3a.0 for ; Fri, 02 Jun 2023 09:10:13 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20221208; t=1685722192; x=1688314192; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:from:to:cc:subject:date:message-id:reply-to; bh=pnaTvciscS5chSJrFdy/XWmGw+v5DL+sKG313yn6vtE=; b=3qM+gPLnKOcFkte53nrFJ4yxYHfRvj6VAQ74elJ+qIzLMj97WYlQnMm+LRqHDvoho4 8AW7RUIjuOvTDqJRc6MGpLfknZTHUVt0ac+jf1e4fznmqTNQB1YeAez2Pbf+vu4gX/wo xA9uw3YiFJ59xCiKuG3PqvgkOpwn1kfOyfDGnV3P6/41gfMjzqB9OHhb0wbZKNfck3ux QuubqPWPIMr2dr+usJ/0oHW/nWAg0PkVCfHTFjYBlz4FbohVSt4JZLXej+/lCcVIkw51 dikq26nIlv8414HoG8qlckcW0gk8fed5eU7B/bJPELllDOCa6K4cYwpu/yZBPdgX/sdW NMYg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1685722192; x=1688314192; h=cc:to:from:subject:message-id:references:mime-version:in-reply-to :date:x-gm-message-state:from:to:cc:subject:date:message-id:reply-to; bh=pnaTvciscS5chSJrFdy/XWmGw+v5DL+sKG313yn6vtE=; b=jxMR3giG8YuG0eU5fWsIjrze72HtzoTTZgYra1OH5A4K4kG0rnJ9QKaAPApjqNDbli NfLUfPAE4I75QDxbmZ8jbsDHgxPeXB91l4qq/A77YxdejWD4yVUdWV2d80TJO62N6ZTh 5awCQxMSqUzd13cI6acseVEJ+3/9icopTqqetayS8ga6oC8HGOskOfQfCyZ1yo3HJiNr QZEUhFVR1frhydK3ELANy6xuquWjall2aB156RffDt/KBiDlltjZ6iX9sXL/c8IYdz+N uX0pvj/672+d9E6U5Atu7HEBPnBCuxDolHaSmZyBgEnIsO33RE9wD41FQjHCizo6r+0Z fzkA== X-Gm-Message-State: AC+VfDznuGjcJ5BSFmNhkUirf4na0CF/uxdhPYVXc1cJc/hdOh8ZyKhI cTKk2Xx+1CJiOA4Ig5A3BTrlzhOk0d3d X-Google-Smtp-Source: ACHHUZ7CrgQcXBBk0scZtBwPutYqwQ0i1jeHWFv03iHedZStcHfAHgjqtsGQJinldbGkFtIY3nwdqwhDBsPu X-Received: from vipin.c.googlers.com ([fda3:e722:ac3:cc00:7f:e700:c0a8:479f]) (user=vipinsh job=sendgmr) by 2002:a05:6a00:2e1a:b0:64f:9e1b:d4a8 with SMTP id fc26-20020a056a002e1a00b0064f9e1bd4a8mr4922796pfb.1.1685722192061; Fri, 02 Jun 2023 09:09:52 -0700 (PDT) Date: Fri, 2 Jun 2023 09:09:14 -0700 In-Reply-To: <20230602160914.4011728-1-vipinsh@google.com> Mime-Version: 1.0 References: <20230602160914.4011728-1-vipinsh@google.com> X-Mailer: git-send-email 2.41.0.rc0.172.g3f132b7071-goog Message-ID: <20230602160914.4011728-17-vipinsh@google.com> Subject: [PATCH v2 16/16] KVM: arm64: Split huge pages during clear-dirty-log under MMU read lock From: Vipin Sharma To: maz@kernel.org, oliver.upton@linux.dev, james.morse@arm.com, suzuki.poulose@arm.com, yuzenghui@huawei.com, catalin.marinas@arm.com, will@kernel.org, chenhuacai@kernel.org, aleksandar.qemu.devel@gmail.com, tsbogend@alpha.franken.de, anup@brainfault.org, atishp@atishpatra.org, paul.walmsley@sifive.com, palmer@dabbelt.com, aou@eecs.berkeley.edu, seanjc@google.com, pbonzini@redhat.com, dmatlack@google.com, ricarkol@google.com Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev, linux-mips@vger.kernel.org, kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org, linux-kselftest@vger.kernel.org, kvm@vger.kernel.org, linux-kernel@vger.kernel.org, Vipin Sharma Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Split huge pages under MMU read lock instead of write when clearing dirty log. Running huge page split under read lock will unblock vCPUs execution and allow whole clear-dirty-log operation run parallelly to vCPUs. Note that splitting huge pages involves two walkers. First walker calls stage2_split_walker() callback on each huge page. This callback will call another walker which creates an unlinked page table. This commit makes first walker as shared page walker which means, -EAGAIN will be retried. Before this patch, -EAGAIN would have been ignored and walker would go to next huge page. In practice this would not happen as the first walker was holding MMU write lock. Inner walker is unchanged as it is working on unlinked page table so no other thread will have access to it. To improve confidence in correctness tested via dirty_log_test. To measure performance improvement tested via dirty_log_perf_test. Set up: ------- Host: ARM Ampere Altra host (64 CPUs, 256 GB memory and single NUMA node) Test VM: 48 vCPU, 192 GB total memory. Ran dirty_log_perf_test for 400 iterations. ./dirty_log_perf_test -k 192G -v 48 -b 4G -m 2 -i 4000 -s anonymous_hugetlb_2mb -j Observation: ------------ +==================+=============================+===================+ | Clear Chunk size | Clear dirty log time change | vCPUs improvement | +==================+=============================+===================+ | 192GB | 56% | 152% | +------------------+-----------------------------+-------------------+ | 1GB | -81% | 72% | +------------------+-----------------------------+-------------------+ When larger chunks are used, clear dirty log time increases due to lots of cmpxchg() but vCPUs are also able to execute parallelly causing better performance of guest. When chunk size is small, read lock is very fast in clearing dirty logs as it is not waiting for MMU write lock and vCPUs are also able to run parallelly. Signed-off-by: Vipin Sharma --- arch/arm64/kvm/mmu.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/arch/arm64/kvm/mmu.c b/arch/arm64/kvm/mmu.c index 6dd964e3682c..aa278f5d27a2 100644 --- a/arch/arm64/kvm/mmu.c +++ b/arch/arm64/kvm/mmu.c @@ -126,7 +126,10 @@ static int kvm_mmu_split_huge_pages(struct kvm *kvm, phys_addr_t addr, int ret, cache_capacity; u64 next, chunk_size; - lockdep_assert_held_write(&kvm->mmu_lock); + if (flags & KVM_PGTABLE_WALK_SHARED) + lockdep_assert_held_read(&kvm->mmu_lock); + else + lockdep_assert_held_write(&kvm->mmu_lock); chunk_size = kvm->arch.mmu.split_page_chunk_size; cache_capacity = kvm_mmu_split_nr_page_tables(chunk_size); @@ -138,13 +141,19 @@ static int kvm_mmu_split_huge_pages(struct kvm *kvm, phys_addr_t addr, do { if (need_split_memcache_topup_or_resched(kvm)) { - write_unlock(&kvm->mmu_lock); + if (flags & KVM_PGTABLE_WALK_SHARED) + read_unlock(&kvm->mmu_lock); + else + write_unlock(&kvm->mmu_lock); cond_resched(); /* Eager page splitting is best-effort. */ ret = __kvm_mmu_topup_memory_cache(cache, cache_capacity, cache_capacity); - write_lock(&kvm->mmu_lock); + if (flags & KVM_PGTABLE_WALK_SHARED) + read_lock(&kvm->mmu_lock); + else + write_lock(&kvm->mmu_lock); if (ret) break; } @@ -1139,9 +1148,7 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, read_lock(&kvm->mmu_lock); stage2_wp_range(&kvm->arch.mmu, start, end, KVM_PGTABLE_WALK_SHARED); - read_unlock(&kvm->mmu_lock); - write_lock(&kvm->mmu_lock); /* * Eager-splitting is done when manual-protect is set. We * also check for initially-all-set because we can avoid @@ -1151,8 +1158,8 @@ void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm, * again. */ if (kvm_dirty_log_manual_protect_and_init_set(kvm)) - kvm_mmu_split_huge_pages(kvm, start, end, 0); - write_unlock(&kvm->mmu_lock); + kvm_mmu_split_huge_pages(kvm, start, end, KVM_PGTABLE_WALK_SHARED); + read_unlock(&kvm->mmu_lock); } static void kvm_send_hwpoison_signal(unsigned long address, short lsb)