From patchwork Thu Nov 28 23:20:35 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?q?=C5=81ukasz_Bartosik?= X-Patchwork-Id: 13888365 Received: from mail-lj1-f177.google.com (mail-lj1-f177.google.com [209.85.208.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 504371C82E6 for ; Thu, 28 Nov 2024 23:20:47 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=209.85.208.177 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1732836049; cv=none; b=KAcnf6Eon01Kl5/Qj2gdsku0m48cFhINZ5EF6kDXDNX8mN941X0YqcbOWaouV0hn9nugn3dilTW5uPusp685VP9HdDcpr/tkJT3ojoVHqfRvtynVpF/pUuJGRlVE/a1Xe/rkCqwdZK3ijKj9ax3qwg8krWjNKjRn9S8zxivmuJ4= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1732836049; c=relaxed/simple; bh=tbcZYXU6HcwHtSy5LuzFDa9JvHECIlXLmldxQYaGrnY=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version:Content-Type; b=TVj51yTQ0Uy72GrR3rYIpyere8RrGSYKhLeZzYD1+vkYkeAV6an/eI4GEctvZSmFZ2IEECwm9Qt8weCB7n1rmXLvnbNluv0yUOB6T9RzTKYfTQNFMMhQzQqh199DrGtIvIia3HUDKo2g0L2AvcoKI4hwBnIdj1Xy04jfvuzZdkI= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=chromium.org; spf=pass smtp.mailfrom=chromium.org; dkim=pass (1024-bit key) header.d=chromium.org header.i=@chromium.org header.b=JedQd9e7; arc=none smtp.client-ip=209.85.208.177 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=chromium.org Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=chromium.org Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=chromium.org header.i=@chromium.org header.b="JedQd9e7" Received: by mail-lj1-f177.google.com with SMTP id 38308e7fff4ca-2ffc3f2b3a9so19343861fa.1 for ; Thu, 28 Nov 2024 15:20:47 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=chromium.org; s=google; t=1732836045; x=1733440845; darn=lists.linux.dev; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=NO7pr5IjtQ7DDii6MRKHjZ3m6XUBJXvLyCa0CyvruQQ=; b=JedQd9e72aTnVU3wtwpgvXIfRxQc1dCdOQ6owPg9XX/P8eW464zle9stwoXaPnM2iE fkrQVolS2bFZVn5Wh5ZdTe7THaPbuPrrBm9grM9wVMfjiQxA+ostVfXavSUaqf8sXz1x o7ZsQxJutnkPPDELKjCkQwHeTxSNCysJJijsI= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1732836045; x=1733440845; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=NO7pr5IjtQ7DDii6MRKHjZ3m6XUBJXvLyCa0CyvruQQ=; b=krxt9CFczsksNzuZWZC5QVUsyP9E/asPa9HyaHc7AjAXNNpwIN+2deciTJv3CCnjB5 MLNmyS5X2S6MXz2ypc9jRxUninwcIfe7MOEJJYSog+5e/NTEve0reC4hEOoUmz9AgccV +P2+UMOT0jhx12+X4Vqp8EJnH1LRBiJFnm2TRPTMUDrRmEcjjyc9tJNEVSSXeXC9mE+5 cwiG33XS5/DkLXeonZ1sd9FmiXksPmmkQRkAqf6eFfzY8XNnjwpCvR+IQgGBbLW3X3sL 5Lnzj/bFx4rOJFfrEjuvwZqONdVYAeAQY8+GOPrNWrnYuMKrfDpocxRNLY29//jtjvA0 B/AA== X-Forwarded-Encrypted: i=1; AJvYcCXRupF4lYo68vaxloemwLZY9y3c7UyUR9qw2rZxUMAH/8KrJD/48/CIutAKyYYCigeJKkPhnzfcZR5EM0J5+Pc=@lists.linux.dev X-Gm-Message-State: AOJu0YwdEfEbp5IYSphXAEhac+spV3+2lL1Y1K8nXO8xiIN10nFoB/fS Dpnuuost4ik+Y/CWlapjfGrSoQJlYBX3zFY8yxzb0IfSQJ2vuCrj9cqD4mGG X-Gm-Gg: ASbGncsbh30FlKQVwezsAmvfTmxyLfFZH1rIf28MUf+VDzyxzttoH9IwPjWRiDRMrEP FT/LVrlScGX8HdiKQm/uUiwNFv5lEnjHSfOY0NK0sPq18MwmbLYJhjQjuTyA5TsEJEdi/CprB/E HGd6J1GW0dEBdpxrXDzz6JxOr6woV3LoPTIbjBM/GCNzg2FHnI95s6lmxJJaX4bvbbplx8t/2KN Hm9f/CxHOlQI3YqVgnjX6XodLWwNfrLww/A1qEXeOkmABrvF4GBqbwcP4GH4ys7tzt3OoIKkBGj Oj1uRf7Y7fgMWPO5Fc1Iy1BC+2WFNZ3EOAo= X-Google-Smtp-Source: AGHT+IGmiebKIbZzhcAjX1oLllwlbZw4xh5D+KuARx/TSvORsh+S7wRIiiNwsgWUZtSjqXFexldF4g== X-Received: by 2002:a05:651c:1502:b0:2fc:9674:60b5 with SMTP id 38308e7fff4ca-2ffd60a9703mr75687331fa.25.1732836045246; Thu, 28 Nov 2024 15:20:45 -0800 (PST) Received: from ukaszb-ng.c.googlers.com.com (103.45.147.34.bc.googleusercontent.com. [34.147.45.103]) by smtp.gmail.com with ESMTPSA id a640c23a62f3a-aa5997d5672sm110757566b.77.2024.11.28.15.20.44 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 28 Nov 2024 15:20:44 -0800 (PST) From: =?utf-8?q?=C5=81ukasz_Bartosik?= To: Heikki Krogerus , Greg Kroah-Hartman , Benson Leung , Dmitry Baryshkov Cc: Abhishek Pandit-Subedi , Jameson Thies , Pavan Holla , Tzung-Bi Shih , linux-usb@vger.kernel.org, chrome-platform@lists.linux.dev Subject: [PATCH v8 3/3] usb: typec: cros_ec_ucsi: Recover from write timeouts Date: Thu, 28 Nov 2024 23:20:35 +0000 Message-ID: <20241128232035.1525978-4-ukaszb@chromium.org> X-Mailer: git-send-email 2.47.0.338.g60cca15819-goog In-Reply-To: <20241128232035.1525978-1-ukaszb@chromium.org> References: <20241128232035.1525978-1-ukaszb@chromium.org> Precedence: bulk X-Mailing-List: chrome-platform@lists.linux.dev List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 From: Abhishek Pandit-Subedi In a suspend-resume edge case, the OPM is timing out in ucsi_resume and the PPM is getting stuck waiting for a command complete ack. Add a write timeout recovery task that will get us out of this state. Signed-off-by: Abhishek Pandit-Subedi Signed-off-by: Łukasz Bartosik --- drivers/usb/typec/ucsi/cros_ec_ucsi.c | 88 ++++++++++++++++++++++++++- 1 file changed, 87 insertions(+), 1 deletion(-) diff --git a/drivers/usb/typec/ucsi/cros_ec_ucsi.c b/drivers/usb/typec/ucsi/cros_ec_ucsi.c index c588d9a57643..6daf61e7e62a 100644 --- a/drivers/usb/typec/ucsi/cros_ec_ucsi.c +++ b/drivers/usb/typec/ucsi/cros_ec_ucsi.c @@ -7,6 +7,7 @@ #include #include +#include #include #include #include @@ -29,6 +30,9 @@ */ #define WRITE_TMO_MS 5000 +/* Number of times to attempt recovery from a write timeout before giving up. */ +#define WRITE_TMO_CTR_MAX 5 + struct cros_ucsi_data { struct device *dev; struct ucsi *ucsi; @@ -36,6 +40,8 @@ struct cros_ucsi_data { struct cros_ec_device *ec; struct notifier_block nb; struct work_struct work; + struct delayed_work write_tmo; + int tmo_counter; struct completion complete; unsigned long flags; @@ -99,12 +105,43 @@ static int cros_ucsi_async_control(struct ucsi *ucsi, u64 cmd) return 0; } +static int cros_ucsi_sync_control(struct ucsi *ucsi, u64 cmd) +{ + struct cros_ucsi_data *udata = ucsi_get_drvdata(ucsi); + int ret; + + ret = ucsi_sync_control_common(ucsi, cmd); + if (ret) + goto out; + + /* Successful write. Cancel any pending recovery work. */ + cancel_delayed_work_sync(&udata->write_tmo); + + return 0; +out: + /* EC may return -EBUSY if CCI.busy is set. Convert this to a timeout. + */ + if (ret == -EBUSY) + ret = -ETIMEDOUT; + + /* Schedule recovery attempt when we timeout or tried to send a command + * while still busy. + */ + if (ret == -ETIMEDOUT) { + cancel_delayed_work_sync(&udata->write_tmo); + schedule_delayed_work(&udata->write_tmo, + msecs_to_jiffies(WRITE_TMO_MS)); + } + + return ret; +} + struct ucsi_operations cros_ucsi_ops = { .read_version = cros_ucsi_read_version, .read_cci = cros_ucsi_read_cci, .read_message_in = cros_ucsi_read_message_in, .async_control = cros_ucsi_async_control, - .sync_control = ucsi_sync_control_common, + .sync_control = cros_ucsi_sync_control, }; static void cros_ucsi_work(struct work_struct *work) @@ -118,6 +155,54 @@ static void cros_ucsi_work(struct work_struct *work) ucsi_notify_common(udata->ucsi, cci); } +static void cros_ucsi_write_timeout(struct work_struct *work) +{ + struct cros_ucsi_data *udata = + container_of(work, struct cros_ucsi_data, write_tmo.work); + u32 cci; + u64 cmd; + + if (cros_ucsi_read(udata->ucsi, UCSI_CCI, &cci, sizeof(cci))) { + dev_err(udata->dev, + "Reading CCI failed; no write timeout recovery possible."); + return; + } + + if (cci & UCSI_CCI_BUSY) { + udata->tmo_counter++; + + if (udata->tmo_counter <= WRITE_TMO_CTR_MAX) + schedule_delayed_work(&udata->write_tmo, + msecs_to_jiffies(WRITE_TMO_MS)); + else + dev_err(udata->dev, + "PPM unresponsive - too many write timeouts."); + + return; + } + + /* No longer busy means we can reset our timeout counter. */ + udata->tmo_counter = 0; + + /* Need to ack previous command which may have timed out. */ + if (cci & UCSI_CCI_COMMAND_COMPLETE) { + cmd = UCSI_ACK_CC_CI | UCSI_ACK_COMMAND_COMPLETE; + cros_ucsi_async_control(udata->ucsi, &cmd); + + /* Check again after a few seconds that the system has + * recovered to make sure our async write above was successful. + */ + schedule_delayed_work(&udata->write_tmo, + msecs_to_jiffies(WRITE_TMO_MS)); + return; + } + + /* We recovered from a previous timeout. Treat this as a recovery from + * suspend and call resume. + */ + ucsi_resume(udata->ucsi); +} + static int cros_ucsi_event(struct notifier_block *nb, unsigned long host_event, void *_notify) { @@ -162,6 +247,7 @@ static int cros_ucsi_probe(struct platform_device *pdev) platform_set_drvdata(pdev, udata); INIT_WORK(&udata->work, cros_ucsi_work); + INIT_DELAYED_WORK(&udata->write_tmo, cros_ucsi_write_timeout); init_completion(&udata->complete); udata->ucsi = ucsi_create(dev, &cros_ucsi_ops);