From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-9.5 required=3.0 tests=DKIMWL_WL_HIGH,DKIM_SIGNED, DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH, MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_PASS,URIBL_BLOCKED,USER_AGENT_GIT autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1B1DEECE564 for ; Wed, 19 Sep 2018 20:51:39 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id B1D2121533 for ; Wed, 19 Sep 2018 20:51:38 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=pass (2048-bit key) header.d=arista.com header.i=@arista.com header.b="fEac+gJb" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org B1D2121533 Authentication-Results: mail.kernel.org; dmarc=fail (p=quarantine dis=none) header.from=arista.com Authentication-Results: mail.kernel.org; spf=none smtp.mailfrom=linux-kernel-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1733198AbeITCbP (ORCPT ); Wed, 19 Sep 2018 22:31:15 -0400 Received: from mail-ed1-f67.google.com ([209.85.208.67]:40435 "EHLO mail-ed1-f67.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1733021AbeITCan (ORCPT ); Wed, 19 Sep 2018 22:30:43 -0400 Received: by mail-ed1-f67.google.com with SMTP id j62-v6so6018911edd.7 for ; Wed, 19 Sep 2018 13:51:02 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=arista.com; s=googlenew; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=7uOjBHO7Lrm52O0tbCG82wfS7n8jMKfvrIot6rE79ss=; b=fEac+gJbwc70BUi+TQv4ChHED/hfPZxUtEuIplGMLRbQmJ1/zw5xvOqaHM44Ys6sYa F1t+xkV+SjjwgfHvjw6iMcEfFLV2NDf4py+Fj+pWC3pOvIJTI9vHcswtwtmMItiJmuMN 9HiT/rq+ivi0HrUSgdpl0T1gg6pRm6zCURdqKwUCw3rP9Agb83nQ5P+pJc4LO7bz+p+U zFGfQVYzM/M4VUf0QR5sMDZ4ngW1JXxOFNG7BQUs95YhV3aVtkAQmlX5hoWZC9RrluOK Wyht3lcYmsCKrRTQdVCUO6m/EQJ/SnPIbTyf7ZzgA6BMxW5HkRHPHhJJ45jPalHhN5VV yaEQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=7uOjBHO7Lrm52O0tbCG82wfS7n8jMKfvrIot6rE79ss=; b=ihFQyk9/netpcfLeUNDw1n5vV9D3fzAduze6Ksg3q7CXeneBTHv5ZzTcwMtkkcHUnF SlMxqOk4ueKixfvg2J+oi/OlDbhKwSXcvLOKcmS6kCi8a2mzNDtj/NGWHrK8v0dPCrTr 83RyAti4XQSVW9eIvXPSlbKCUWVq1Q0wlyviPTOSXK86ZmZCEpjcVeclndSibYsR15g5 TSudpjHgKFpBkNimJpapln6ohRnlEqPi/bgkqaUL4mrx3At1tJg/D8rC29U91OGvH5qj tJsmugkTRVzgJ+282wSGLP7whH9TEla2HYYKAm2IP0KSyEE/rMG9U5yNBmL11q1FA7m+ 5Gkw== X-Gm-Message-State: APzg51CgTGWy2/qVM1LZSK000EjIJ+B0UWusg7HYCmvOEajHtIyeRqVG zCyO8XKuEbCEIsXyAg8KXri02Qb74C0= X-Google-Smtp-Source: ANB0VdZPAdtSGlptPeJ7tS4cRn9ERsACYu+R9HHufOdAypsrfQcsfJFiAczzjwmZeDloWAvYK99uCw== X-Received: by 2002:a50:eb96:: with SMTP id y22-v6mr59898154edr.38.1537390261475; Wed, 19 Sep 2018 13:51:01 -0700 (PDT) Received: from dhcp.ire.aristanetworks.com ([217.173.96.166]) by smtp.gmail.com with ESMTPSA id t17-v6sm1747729edb.27.2018.09.19.13.51.00 (version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256); Wed, 19 Sep 2018 13:51:00 -0700 (PDT) From: Dmitry Safonov To: linux-kernel@vger.kernel.org Cc: Dmitry Safonov <0x7f454c46@gmail.com>, Dmitry Safonov , Adrian Reber , Andrei Vagin , Andy Lutomirski , Christian Brauner , Cyrill Gorcunov , "Eric W. Biederman" , "H. Peter Anvin" , Ingo Molnar , Jeff Dike , Oleg Nesterov , Pavel Emelyanov , Shuah Khan , Thomas Gleixner , containers@lists.linux-foundation.org, criu@openvz.org, linux-api@vger.kernel.org, x86@kernel.org Subject: [RFC 15/20] timens: Optimize zero-offsets Date: Wed, 19 Sep 2018 21:50:32 +0100 Message-Id: <20180919205037.9574-16-dima@arista.com> X-Mailer: git-send-email 2.13.6 In-Reply-To: <20180919205037.9574-1-dima@arista.com> References: <20180919205037.9574-1-dima@arista.com> Sender: linux-kernel-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Fall through on host or in ns without time set. Add TIMENS_FALLBACK_SYSCALL which might be wired up if timens offsets should be unknown for userspace (will result in fall-back to syscalls). Signed-off-by: Dmitry Safonov --- arch/x86/entry/vdso/vclock_gettime.c | 17 +++++++++++++---- include/linux/timens_offsets.h | 12 ++++++++++-- kernel/time/posix-timers.c | 21 ++++++++++++--------- kernel/time_namespace.c | 2 +- 4 files changed, 36 insertions(+), 16 deletions(-) diff --git a/arch/x86/entry/vdso/vclock_gettime.c b/arch/x86/entry/vdso/vclock_gettime.c index a265e2737a9a..458cb1992e2e 100644 --- a/arch/x86/entry/vdso/vclock_gettime.c +++ b/arch/x86/entry/vdso/vclock_gettime.c @@ -252,17 +252,25 @@ notrace void set_normalized_timespec(struct timespec *ts, time_t sec, s64 nsec) ts->tv_nsec = nsec; } -notrace static __always_inline void monotonic_to_ns(struct timespec *ts) +notrace static __always_inline int monotonic_to_ns(struct timespec *ts) { #ifdef CONFIG_TIME_NS struct timens_offsets *timens = (struct timens_offsets *) &timens_page; struct timespec offset; + /* Optimization: time is the same as on host, return right away */ + if (!(timens->flags & TIMENS_USE_OFFSETS)) + return 0; + + if (timens->flags & TIMENS_FALLBACK_SYSCALL) + return -1; + offset = timespec64_to_timespec(timens->monotonic_time_offset); *ts = timespec_add(*ts, offset); #endif + return 0; } notrace static int __always_inline do_monotonic(struct timespec *ts) @@ -283,8 +291,6 @@ notrace static int __always_inline do_monotonic(struct timespec *ts) ts->tv_sec += __iter_div_u64_rem(ns, NSEC_PER_SEC, &ns); ts->tv_nsec = ns; - monotonic_to_ns(ts); - return mode; } @@ -306,7 +312,6 @@ notrace static void do_monotonic_coarse(struct timespec *ts) ts->tv_sec = gtod->monotonic_time_coarse_sec; ts->tv_nsec = gtod->monotonic_time_coarse_nsec; } while (unlikely(gtod_read_retry(gtod, seq))); - monotonic_to_ns(ts); } notrace int __vdso_clock_gettime(clockid_t clock, struct timespec *ts) @@ -319,12 +324,16 @@ notrace int __vdso_clock_gettime(clockid_t clock, struct timespec *ts) case CLOCK_MONOTONIC: if (do_monotonic(ts) == VCLOCK_NONE) goto fallback; + if (monotonic_to_ns(ts)) + goto fallback; break; case CLOCK_REALTIME_COARSE: do_realtime_coarse(ts); break; case CLOCK_MONOTONIC_COARSE: do_monotonic_coarse(ts); + if (monotonic_to_ns(ts)) + goto fallback; break; default: goto fallback; diff --git a/include/linux/timens_offsets.h b/include/linux/timens_offsets.h index 92a8ea5601eb..8c43e7c3e632 100644 --- a/include/linux/timens_offsets.h +++ b/include/linux/timens_offsets.h @@ -2,6 +2,13 @@ #ifndef _LINUX_TIME_OFFSETS_H #define _LINUX_TIME_OFFSETS_H +enum { + /* We're in namespace - add offsets from vvar */ + TIMENS_USE_OFFSETS = 1, + /* Don't expose host's offsets, fall back to syscall - slow */ + TIMENS_FALLBACK_SYSCALL = 2, /* TODO if anyone actually interested */ +}; + /* * Time offsets need align as they're placed on vvar page, * which should have tail paddings on ia32 vdso. @@ -10,8 +17,9 @@ * to timespec because of a padding occuring between the fields. */ struct timens_offsets { - struct timespec64 monotonic_time_offset __aligned(8); - struct timespec64 monotonic_boottime_offset __aligned(8); + u64 flags; + struct timespec64 monotonic_time_offset __aligned(8); + struct timespec64 monotonic_boottime_offset __aligned(8); }; #endif diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c index 701cb0602b7a..576dbd24c498 100644 --- a/kernel/time/posix-timers.c +++ b/kernel/time/posix-timers.c @@ -210,7 +210,7 @@ static void common_timens_adjust(clockid_t which_clock, struct timespec64 *tp) { struct timens_offsets *ns_offsets = current->nsproxy->time_ns->offsets; - if (!ns_offsets) + if (!ns_offsets || !(ns_offsets->flags & TIMENS_USE_OFFSETS)) return; switch (which_clock) { @@ -234,15 +234,16 @@ static int posix_ktime_set_ts(clockid_t which_clock, struct timens_offsets *ns_offsets = current->nsproxy->time_ns->offsets; struct timespec64 ktp; + if (!ns_offsets) + return -EINVAL; + if (!ns_capable(current->nsproxy->time_ns->user_ns, CAP_SYS_TIME)) return -EPERM; ktime_get_ts64(&ktp); - if (ns_offsets) - ns_offsets->monotonic_time_offset = timespec64_sub(*tp, ktp); - else - return -EINVAL; + ns_offsets->monotonic_time_offset = timespec64_sub(*tp, ktp); + ns_offsets->flags |= TIMENS_USE_OFFSETS; return 0; } @@ -296,15 +297,17 @@ static int posix_set_boottime(clockid_t which_clock, const struct timespec64 *tp struct timens_offsets *ns_offsets = current->nsproxy->time_ns->offsets; struct timespec64 ktp; + if (!ns_offsets) + return -EINVAL; + if (!ns_capable(current->nsproxy->time_ns->user_ns, CAP_SYS_TIME)) return -EPERM; ktime_get_boottime_ts64(&ktp); - if (ns_offsets) - ns_offsets->monotonic_boottime_offset = timespec64_sub(*tp, ktp); - else - return -EINVAL; + ns_offsets->monotonic_boottime_offset = timespec64_sub(*tp, ktp); + ns_offsets->flags |= TIMENS_USE_OFFSETS; + return 0; } diff --git a/kernel/time_namespace.c b/kernel/time_namespace.c index f88ae0e17d92..4052bdcec110 100644 --- a/kernel/time_namespace.c +++ b/kernel/time_namespace.c @@ -171,7 +171,7 @@ static void clock_timens_fixup(int clockid, struct timespec64 *val, bool to_ns) struct timens_offsets *ns_offsets = current->nsproxy->time_ns->offsets; struct timespec64 *offsets = NULL; - if (!ns_offsets) + if (!ns_offsets || !(ns_offsets->flags & TIMENS_USE_OFFSETS)) return; if (val->tv_sec == 0 && val->tv_nsec == 0) -- 2.13.6