From patchwork Wed Mar 29 20:46:32 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?q?Uwe_Kleine-K=C3=B6nig?= X-Patchwork-Id: 13193205 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 48953C74A5B for ; Wed, 29 Mar 2023 20:46:47 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S229564AbjC2Uqq (ORCPT ); Wed, 29 Mar 2023 16:46:46 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:50996 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229868AbjC2Uqp (ORCPT ); Wed, 29 Mar 2023 16:46:45 -0400 Received: from metis.ext.pengutronix.de (metis.ext.pengutronix.de [IPv6:2001:67c:670:201:290:27ff:fe1d:cc33]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 98F261985 for ; Wed, 29 Mar 2023 13:46:44 -0700 (PDT) Received: from drehscheibe.grey.stw.pengutronix.de ([2a0a:edc0:0:c01:1d::a2]) by metis.ext.pengutronix.de with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1phcgk-0001Sl-2y; Wed, 29 Mar 2023 22:46:34 +0200 Received: from [2a0a:edc0:0:900:1d::77] (helo=ptz.office.stw.pengutronix.de) by drehscheibe.grey.stw.pengutronix.de with esmtp (Exim 4.94.2) (envelope-from ) id 1phcgj-007cAT-7l; Wed, 29 Mar 2023 22:46:33 +0200 Received: from ukl by ptz.office.stw.pengutronix.de with local (Exim 4.94.2) (envelope-from ) id 1phcgi-0091Df-FO; Wed, 29 Mar 2023 22:46:32 +0200 Date: Wed, 29 Mar 2023 22:46:32 +0200 From: Uwe =?utf-8?q?Kleine-K=C3=B6nig?= To: Stephen Boyd Cc: Michael Turquette , linux-kernel@vger.kernel.org, Jonathan Corbet , linux-clk@vger.kernel.org, kernel@pengutronix.de, linux-doc@vger.kernel.org Subject: [PATCH v4] clk: expand clk_ignore_unused mechanism to keep only a few clks on Message-ID: <20230329204632.lsiiqf42hrwmn6xm@pengutronix.de> References: <20221026151812.1042052-1-u.kleine-koenig@pengutronix.de> <4d8d412a33a7d63f2ffe6a13194375ed.sboyd@kernel.org> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <4d8d412a33a7d63f2ffe6a13194375ed.sboyd@kernel.org> X-SA-Exim-Connect-IP: 2a0a:edc0:0:c01:1d::a2 X-SA-Exim-Mail-From: ukl@pengutronix.de X-SA-Exim-Scanned: No (on metis.ext.pengutronix.de); SAEximRunCond expanded to false X-PTX-Original-Recipient: linux-clk@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-clk@vger.kernel.org Allow to pass an integer n that results in only keeping n unused clocks enabled. This helps to debug the problem if you only know that clk_ignore_unused helps but you have no clue yet which clock is the culprit. Signed-off-by: Uwe Kleine-König --- Hello Stephen, On Wed, Mar 29, 2023 at 12:49:19PM -0700, Stephen Boyd wrote: > Quoting Uwe Kleine-König (2022-10-26 08:18:12) > > Documentation/driver-api/clk.rst | 4 +++- > > No update to Documentation/admin-guide/kernel-parameters.txt? Fair request. I mentioned it there that you can assign an integer but refer to Documentation/driver-api/clk.rst for the details. > > diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c > > index c3c3f8c07258..356119a7e5fe 100644 > > --- a/drivers/clk/clk.c > > +++ b/drivers/clk/clk.c > > [...] > > @@ -1352,12 +1354,17 @@ static void __init clk_disable_unused_subtree(struct clk_core *core) > > * back to .disable > > */ > > if (clk_core_is_enabled(core)) { > > - trace_clk_disable(core); > > - if (core->ops->disable_unused) > > - core->ops->disable_unused(core->hw); > > - else if (core->ops->disable) > > - core->ops->disable(core->hw); > > - trace_clk_disable_complete(core); > > + if (clk_unused_keep_on) { > > + pr_warn("Keep unused clk \"%s\" on\n", core->name); > > + clk_unused_keep_on -= 1; > > + } else { > > + trace_clk_disable(core); > > We have trace_clk_disable() here. Can you have this tracepoint print to > the kernel log and watch over serial console? That would be faster than > bisecting. Well no, that doesn't work for all the problems where clk_ignore_unused=7 could be useful. Consider that e.g. you know that eth0 is broken, but with clk_ignore_unused is works. So one of the (say) 25 nominally unused clks are required for eth0. But it's not possible to test the network after each of the 25 clk_disable()s. Unless I'm missing something and you can hook a userspace action on a trace line?! > > + if (core->ops->disable_unused) > > + core->ops->disable_unused(core->hw); > > + else if (core->ops->disable) > > + core->ops->disable(core->hw); > > + trace_clk_disable_complete(core); > > + } > > } > > > > unlock_out: > > @@ -1369,9 +1376,17 @@ static void __init clk_disable_unused_subtree(struct clk_core *core) > > } > > > > static bool clk_ignore_unused __initdata; > > -static int __init clk_ignore_unused_setup(char *__unused) > > +static int __init clk_ignore_unused_setup(char *keep) > > { > > - clk_ignore_unused = true; > > + if (*keep == '=') { > > + int ret; > > + > > + ret = kstrtouint(keep + 1, 0, &clk_unused_keep_on); > > + if (ret < 0) > > Could omit 'ret' and just have if (kstrtouint(..)) I don't feel strong, but I think having that on two lines is easier to read. So I kept it as is, but if you insist, I can change. > > + pr_err("Warning: failed to parse clk_ignore_unused parameter, ignoring"); > > Missing newline on printk. fixed. Best regards Uwe .../admin-guide/kernel-parameters.txt | 6 ++-- Documentation/driver-api/clk.rst | 4 ++- drivers/clk/clk.c | 33 ++++++++++++++----- 3 files changed, 32 insertions(+), 11 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 6221a1d057dd..1a378fe94e48 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -600,8 +600,10 @@ force such clocks to be always-on nor does it reserve those clocks in any way. This parameter is useful for debug and development, but should not be needed on a - platform with proper driver support. For more - information, see Documentation/driver-api/clk.rst. + platform with proper driver support. + It can take a value (e.g. clk_ignore_unused=7), to only + disable some clks. For more information, see + Documentation/driver-api/clk.rst. clock= [BUGS=X86-32, HW] gettimeofday clocksource override. [Deprecated] diff --git a/Documentation/driver-api/clk.rst b/Documentation/driver-api/clk.rst index 3cad45d14187..65ae7c3e2b33 100644 --- a/Documentation/driver-api/clk.rst +++ b/Documentation/driver-api/clk.rst @@ -259,7 +259,9 @@ the disabling means that the driver will remain functional while the issues are sorted out. To bypass this disabling, include "clk_ignore_unused" in the bootargs to the -kernel. +kernel. If you pass "clk_ignore_unused=n" (where n is an integer) the first n +found clocks are not disabled which can be useful for bisecting over the unused +clks if you don't know yet which of them is reponsible for your problem. Locking ======= diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c index ae07685c7588..87f605a4f791 100644 --- a/drivers/clk/clk.c +++ b/drivers/clk/clk.c @@ -1343,6 +1343,8 @@ static void __init clk_unprepare_unused_subtree(struct clk_core *core) clk_pm_runtime_put(core); } +static unsigned clk_unused_keep_on __initdata; + static void __init clk_disable_unused_subtree(struct clk_core *core) { struct clk_core *child; @@ -1373,12 +1375,17 @@ static void __init clk_disable_unused_subtree(struct clk_core *core) * back to .disable */ if (clk_core_is_enabled(core)) { - trace_clk_disable(core); - if (core->ops->disable_unused) - core->ops->disable_unused(core->hw); - else if (core->ops->disable) - core->ops->disable(core->hw); - trace_clk_disable_complete(core); + if (clk_unused_keep_on) { + pr_warn("Keep unused clk \"%s\" on\n", core->name); + clk_unused_keep_on -= 1; + } else { + trace_clk_disable(core); + if (core->ops->disable_unused) + core->ops->disable_unused(core->hw); + else if (core->ops->disable) + core->ops->disable(core->hw); + trace_clk_disable_complete(core); + } } unlock_out: @@ -1390,9 +1397,17 @@ static void __init clk_disable_unused_subtree(struct clk_core *core) } static bool clk_ignore_unused __initdata; -static int __init clk_ignore_unused_setup(char *__unused) +static int __init clk_ignore_unused_setup(char *keep) { - clk_ignore_unused = true; + if (*keep == '=') { + int ret; + + ret = kstrtouint(keep + 1, 0, &clk_unused_keep_on); + if (ret < 0) + pr_err("Warning: failed to parse clk_ignore_unused parameter, ignoring\n"); + } else { + clk_ignore_unused = true; + } return 1; } __setup("clk_ignore_unused", clk_ignore_unused_setup); @@ -1404,6 +1419,8 @@ static int __init clk_disable_unused(void) if (clk_ignore_unused) { pr_warn("clk: Not disabling unused clocks\n"); return 0; + } else if (clk_unused_keep_on) { + pr_warn("clk: Not disabling %u unused clocks\n", clk_unused_keep_on); } clk_prepare_lock();