From patchwork Tue Jan 23 01:46:40 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Michael Pratt X-Patchwork-Id: 1889475 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; dkim=pass (2048-bit key; secure) header.d=lists.infradead.org header.i=@lists.infradead.org header.a=rsa-sha256 header.s=bombadil.20210309 header.b=YkZEWbJL; dkim=fail reason="signature verification failed" (2048-bit key; secure) header.d=pm.me header.i=@pm.me header.a=rsa-sha256 header.s=protonmail3 header.b=JYBdeS/7; dkim-atps=neutral Authentication-Results: legolas.ozlabs.org; spf=none (no SPF record) smtp.mailfrom=lists.infradead.org (client-ip=198.137.202.133; helo=bombadil.infradead.org; envelope-from=linux-mtd-bounces+incoming=patchwork.ozlabs.org@lists.infradead.org; receiver=patchwork.ozlabs.org) Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (secp384r1) server-digest SHA384) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4TJqhx1c0nz23f0 for ; Tue, 23 Jan 2024 12:47:29 +1100 (AEDT) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender: Content-Transfer-Encoding:Content-Type:List-Subscribe:List-Help:List-Post: List-Archive:List-Unsubscribe:List-Id:MIME-Version:References:In-Reply-To: Message-ID:Subject:Cc:From:To:Date:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=SpvUwu29DSyNGovIdyiYpaD0QEXIKATO1PQ5F6ZhHDk=; b=YkZEWbJL1M/mvN TwZYPPP59XLOiKo0vMOc6nrFpidhIE7+sHJ0WNzOj//HxWnMjuF9SY/VwP8syEL/jOsz3bJ63WJoz IBn5sup4gydm/ywWlXKdXRK5aUBmjQ/RHbXws0u6i7tC2DdkoGcGjiSV1fWU3cgWzHgekh8PF8/nm Hfb1ykMl0Ap7cLd+ZEWIiIQ1QbrLGTrbAGRBPHo7/LOL6mgLkBzFJ0eOgN4YExs6wuXJmid5sKAlj sESp96tbbu1XPDf67naQp3iFtWvMjHoP4XQAgoO3UybAG5hH+0Jo23MOqdkQXEFwXAXbgcqwe426Y NLM6E5zZySDldaiCoKoA==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.96 #2 (Red Hat Linux)) id 1rS5sR-00EnsM-33; Tue, 23 Jan 2024 01:46:59 +0000 Received: from mail-40133.protonmail.ch ([185.70.40.133]) by bombadil.infradead.org with esmtps (Exim 4.96 #2 (Red Hat Linux)) id 1rS5sN-00EnrA-2b for linux-mtd@lists.infradead.org; Tue, 23 Jan 2024 01:46:57 +0000 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=pm.me; s=protonmail3; t=1705974413; x=1706233613; bh=IuTlLINJgaRw7B3N708S7h3v0HiOAtfm1OMnmzHUN50=; h=Date:To:From:Cc:Subject:Message-ID:In-Reply-To:References: Feedback-ID:From:To:Cc:Date:Subject:Reply-To:Feedback-ID: Message-ID:BIMI-Selector; b=JYBdeS/7qxzIJC0+6PiVAFOXhjyPOiiUuPrrE9l5qdNV1qW9G+nJIBi/H5pusUExW jEhCC3EXSPFrGjBsfXHoPG19+8YeL7BYH85PIm0PxWE1AW58Hf3b0ULPRhSHc2Fq2i nuq7W6Zmnc+5S9uj0FYj7IZej/dqnFMewI9bgdwaYcJYsvOkq/4kyZEk93Y12+RU3y Oc+tzZsVBMaThH35Erhf6F/aAhSVd7sQ3JOkbQCKrD/Oyww/SESsVFNyGM+hKUNT86 XEiBkD6ljwzuVWAOyXPni4Rs/Y7awS1OC7YhPjZkaIAzg0pk7hRPfkaGKtaQQWK8yU 3JDnNtJezGi1g== Date: Tue, 23 Jan 2024 01:46:40 +0000 To: devicetree@vger.kernel.org, gregkh@linuxfoundation.org, linux-kernel@vger.kernel.org, linux-mtd@lists.infradead.org From: Michael Pratt Cc: Michael Pratt , saravanak@google.com, abel.vesa@linaro.org, alexander.stein@ew.tq-group.com, andriy.shevchenko@linux.intel.com, bigunclemax@gmail.com, brgl@bgdev.pl, colin.foster@in-advantage.com, djrscally@gmail.com, dmitry.baryshkov@linaro.org, festevam@gmail.com, fido_max@inbox.ru, frowand.list@gmail.com, geert@linux-m68k.org, heikki.krogerus@linux.intel.com, kernel@pengutronix.de, linus.walleij@linaro.org, linux@roeck-us.net, luca.weiss@fairphone.com, magnus.damm@gmail.com, martin.kepplinger@puri.sm, miquel.raynal@bootlin.com, rafal@milecki.pl, ansuelsmth@gmail.com, richard@nod.at, sakari.ailus@linux.intel.com, sudeep.holla@arm.com, tglx@linutronix.de, tony@atomide.com, vigneshr@ti.com, dianders@chromium.org, jpb@kernel.org, rafael@kernel.org Subject: [PATCH v1 2/4] driver core: fw_devlink: Link to supplier ancestor if no device Message-ID: <20240123014517.5787-3-mcpratt@pm.me> In-Reply-To: <20240123014517.5787-1-mcpratt@pm.me> References: <20240123014517.5787-1-mcpratt@pm.me> Feedback-ID: 27397442:user:proton MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20240122_174656_154183_47DF55C4 X-CRM114-Status: GOOD ( 22.23 ) X-Spam-Score: -0.2 (/) X-Spam-Report: Spam detection software, running on the system "bombadil.infradead.org", has NOT identified this incoming email as spam. The original message has been attached to this so you can view it or label similar future email. If you have any questions, see the administrator of that system for details. Content preview: Driver core currently supports linking to the next parent fwnode, but is not yet handling cases where that parent is also a firmware child node not representing a real device, which can lead to an ind [...] Content analysis details: (-0.2 points, 5.0 required) pts rule name description ---- ---------------------- -------------------------------------------------- 0.0 RCVD_IN_MSPIKE_H5 RBL: Excellent reputation (+5) [185.70.40.133 listed in wl.mailspike.net] -0.0 SPF_PASS SPF: sender matches SPF record -0.0 SPF_HELO_PASS SPF: HELO matches SPF record -0.1 DKIM_VALID_AU Message has a valid DKIM or DK signature from author's domain -0.1 DKIM_VALID Message has at least one valid DKIM or DK signature 0.1 DKIM_SIGNED Message has a DKIM or DK signature, not necessarily valid -0.1 DKIM_VALID_EF Message has a valid DKIM or DK signature from envelope-from domain 0.0 RCVD_IN_MSPIKE_WL Mailspike good senders X-BeenThere: linux-mtd@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-mtd" Errors-To: linux-mtd-bounces+incoming=patchwork.ozlabs.org@lists.infradead.org Driver core currently supports linking to the next parent fwnode, but is not yet handling cases where that parent is also a firmware child node not representing a real device, which can lead to an indefinite deferred probe in some cases. In this case, the fwnode that should actually be linked to is multiple ancestors up which presents a challenge where it is unknown how many ancestors up the node that represents the real probing device is. This makes the usage of fwnode_get_next_parent_dev() insufficient because the real device's fwnode may or may not be an ancestor of the next parent fwnode as well. Introduce flag FWNODE_FLAG_PARENT_IS_DEV in order to mark child firmware nodes of a device as having a parent device that can probe. Allow fwnode link creation to the original supplier fwnode's ancestors when the original supplier fwnode and any fwnodes in between are flagged as FWNODE_FLAG_NOT_DEVICE and/or FWNODE_FLAG_PARENT_IS_DEV with a new function __fwnode_link_add_parents() which then creates the fwnode link to a real device that provides the supplier's function. This depends on other functions to label a supplier fwnode as not a real device, which must be done before the fwnode links are created, and if after that, relevant links to the supplier would have to be deleted and have links recreated, otherwise, the fwnode link would be dropped before the device link is attempted or a fwnode link would not be able to become a device link at all, because they were created before these fwnode flags can have any effect. It also depends on the supplier device to actually probe first in order to have the fwnode flags in place to know for certain which fwnodes are non-probing child nodes of the fwnode for the supplier device. The use case of function __fw_devlink_pickup_dangling_consumers() is designed so that the parameters are always a supplier fwnode and one of it's parent fwnodes, so it is safer to assume and more specific that the flag PARENT_IS_DEV should be added there, rather than declaring the original supplier fwnode as NOT_DEVICE at that point. Because this function is called when the real supplier device probes and recursively calls itself for all child nodes of the device's fwnode, set the new flag here in order to let it propagate down to all descendant nodes, thereby providing the info needed later in order to link to the proper fwnode representing the supplier device. If a fwnode is flagged as FWNODE_FLAG_NOT_DEVICE by the time a device link is to be made with it, but not flagged as FWNODE_FLAG_PARENT_IS_DEV, the link is dropped, otherwise the device link is still made with the original supplier fwnode. Theoretically, we can also handle linking to an ancestor of the supplier fwnode when forming device links, but there are still cases where the necessary fwnode flags are still missing because the real supplier device did not probe yet. Signed-off-by: Michael Pratt --- drivers/base/core.c | 49 ++++++++++++++++++++++++++++++++++++------ include/linux/fwnode.h | 4 ++++ 2 files changed, 47 insertions(+), 6 deletions(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index c05a5f6b0641..7f2652cf5edc 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -92,13 +92,45 @@ static int __fwnode_link_add(struct fwnode_handle *con, return 0; } +static int __fwnode_link_add_parents(struct fwnode_handle *con, + struct fwnode_handle *sup, + u8 flags, bool loop) +{ + int ret = -EINVAL; + struct fwnode_handle *parent; + + fwnode_for_each_parent_node(sup, parent) { + /* Ignore the root node */ + if (fwnode_count_parents(parent) && + fwnode_device_is_available(parent) && + !(parent->flags & FWNODE_FLAG_NOT_DEVICE) && + !(parent->flags & FWNODE_FLAG_PARENT_IS_DEV)) { + ret = __fwnode_link_add(con, parent, flags); + } + + if (!loop && !ret) { + fwnode_handle_put(parent); + return 0; + } + } + + return ret; +} + int fwnode_link_add(struct fwnode_handle *con, struct fwnode_handle *sup) { int ret; mutex_lock(&fwnode_link_lock); - ret = __fwnode_link_add(con, sup, 0); + + if ((sup->flags & FWNODE_FLAG_NOT_DEVICE) && + (sup->flags & FWNODE_FLAG_PARENT_IS_DEV)) + ret = __fwnode_link_add_parents(con, sup, 0, false); + else + ret = __fwnode_link_add(con, sup, 0); + mutex_unlock(&fwnode_link_lock); + return ret; } @@ -218,7 +250,8 @@ static void __fwnode_links_move_consumers(struct fwnode_handle *from, * MANAGED device links to this device, so leave @fwnode and its descendant's * fwnode links alone. * - * Otherwise, move its consumers to the new supplier @new_sup. + * Otherwise, flag descendants of @fwnode as having a parent fwnode for a device + * that has probed and move bad fwnode consumer links from @fwnode to @new_sup. */ static void __fw_devlink_pickup_dangling_consumers(struct fwnode_handle *fwnode, struct fwnode_handle *new_sup) @@ -228,8 +261,11 @@ static void __fw_devlink_pickup_dangling_consumers(struct fwnode_handle *fwnode, if (fwnode->dev && fwnode->dev->driver) return; - fwnode->flags |= FWNODE_FLAG_NOT_DEVICE; - __fwnode_links_move_consumers(fwnode, new_sup); + if (fwnode->flags & FWNODE_FLAG_NOT_DEVICE) + __fwnode_links_move_consumers(fwnode, new_sup); + + fwnode->flags |= FWNODE_FLAG_PARENT_IS_DEV; + new_sup->flags &= ~(FWNODE_FLAG_PARENT_IS_DEV); fwnode_for_each_available_child_node(fwnode, child) __fw_devlink_pickup_dangling_consumers(child, new_sup); @@ -2071,8 +2107,9 @@ static int fw_devlink_create_devlink(struct device *con, device_links_write_unlock(); } - if (sup_handle->flags & FWNODE_FLAG_NOT_DEVICE) - sup_dev = fwnode_get_next_parent_dev(sup_handle); + if ((sup_handle->flags & FWNODE_FLAG_NOT_DEVICE) && + !(sup_handle->flags & FWNODE_FLAG_PARENT_IS_DEV)) + return -EINVAL; else sup_dev = get_dev_from_fwnode(sup_handle); diff --git a/include/linux/fwnode.h b/include/linux/fwnode.h index 2a72f55d26eb..3ed8ba503062 100644 --- a/include/linux/fwnode.h +++ b/include/linux/fwnode.h @@ -30,6 +30,9 @@ struct device; * BEST_EFFORT: The fwnode/device needs to probe early and might be missing some * suppliers. Only enforce ordering with suppliers that have * drivers. + * PARENT_IS_DEV: The fwnode is a child of a fwnode that is or will be populated as a + * struct device, which is more suitable for device links + * than the fwnode child which may never have a struct device. */ #define FWNODE_FLAG_LINKS_ADDED BIT(0) #define FWNODE_FLAG_NOT_DEVICE BIT(1) @@ -37,6 +40,7 @@ struct device; #define FWNODE_FLAG_NEEDS_CHILD_BOUND_ON_ADD BIT(3) #define FWNODE_FLAG_BEST_EFFORT BIT(4) #define FWNODE_FLAG_VISITED BIT(5) +#define FWNODE_FLAG_PARENT_IS_DEV BIT(6) struct fwnode_handle { struct fwnode_handle *secondary;