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=-15.9 required=3.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, INCLUDES_CR_TRAILER,INCLUDES_PATCH,MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS, URIBL_BLOCKED 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 B3146C48BDF for ; Thu, 10 Jun 2021 18:40:35 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by mail.kernel.org (Postfix) with ESMTP id 9B77C613F5 for ; Thu, 10 Jun 2021 18:40:35 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231372AbhFJSmb (ORCPT ); Thu, 10 Jun 2021 14:42:31 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45460 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231379AbhFJSmY (ORCPT ); Thu, 10 Jun 2021 14:42:24 -0400 Received: from mail-il1-x135.google.com (mail-il1-x135.google.com [IPv6:2607:f8b0:4864:20::135]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 48C2CC0617A8 for ; Thu, 10 Jun 2021 11:40:28 -0700 (PDT) Received: by mail-il1-x135.google.com with SMTP id z1so2827318ils.0 for ; Thu, 10 Jun 2021 11:40:28 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=chromium.org; s=google; h=mime-version:references:in-reply-to:from:date:message-id:subject:to :cc; bh=Nf6J5VPNhOdTIuNzS3DZB8U4CF2Pa9aEYKvb9flUDmE=; b=L9ROHi2BHpv08ujQ0qbHKjVArHSZnmhGjjBBzyRm68KOimSWOubX9EnG8CQ0OXAVev +CJgYDykJlToHLMoS/FMyD5TmcfVYoP2/XrCIqGk92OfAGYPgPU0piy1c776G5VCviVD uFvX2DGmcDFLBAYbS3HHS+ZqUy0mFxi3w67Xw= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:mime-version:references:in-reply-to:from:date :message-id:subject:to:cc; bh=Nf6J5VPNhOdTIuNzS3DZB8U4CF2Pa9aEYKvb9flUDmE=; b=CFl6Z52P93397hush1axQwSRMdROOMboNWaYDY77DgRWh6OnqqRxS6N8ey6JKX4Y9G 6v2JdlExIU01OQTKvwXOxzCJQUThA3doL0ljkxcDrvAL9OtHCBg3Q/zyM3ZNssB0h8XI t2VRDNlA8ffzC8PFHc3RqbXxLlUqnGlVwVIYTvAaH5OKCCMjtVyqsUcAnM8a531TRbYx +Ani2GVMG1bbIlk9W3CGQhKxMRe6bbYxL4FtS8xU3QvZBxHWri4dunrg4hitfpUXtfIp yMfLhyoaSXiyD66f+2CeH4UCeQW7lcNwOMQwFCIly4FoRQIf1Kiz20pY/gb61l4ki3WZ MxFw== X-Gm-Message-State: AOAM5300/o6vp+J0EIC61ClOSsHfVgEBoxtPx3jWCV9AHMTFE9zOCPdL UZcERqIxtjAKlsWB7Un9dJ7cqPVXO3mzFw== X-Google-Smtp-Source: ABdhPJzWd1wvNKilHx1+Zs6ognkbkdzuQz8atznrWZFZD4RgEsjh/FhR3yh+20B1uOv5RhiTGTD3IQ== X-Received: by 2002:a05:6e02:54e:: with SMTP id i14mr158374ils.129.1623350427474; Thu, 10 Jun 2021 11:40:27 -0700 (PDT) Received: from mail-io1-f43.google.com (mail-io1-f43.google.com. [209.85.166.43]) by smtp.gmail.com with ESMTPSA id e4sm2279669iow.47.2021.06.10.11.40.26 for (version=TLS1_3 cipher=TLS_AES_128_GCM_SHA256 bits=128/128); Thu, 10 Jun 2021 11:40:27 -0700 (PDT) Received: by mail-io1-f43.google.com with SMTP id q7so28172796iob.4 for ; Thu, 10 Jun 2021 11:40:26 -0700 (PDT) X-Received: by 2002:a5e:974d:: with SMTP id h13mr4904715ioq.150.1623350426300; Thu, 10 Jun 2021 11:40:26 -0700 (PDT) MIME-Version: 1.0 References: <20210326095840.364424-1-ribalda@chromium.org> <20210326095840.364424-17-ribalda@chromium.org> In-Reply-To: From: Ricardo Ribalda Date: Thu, 10 Jun 2021 20:40:15 +0200 X-Gmail-Original-Message-ID: Message-ID: Subject: Re: [PATCH v9 16/22] media: uvcvideo: Return -EACCES to inactive controls To: Laurent Pinchart Cc: Mauro Carvalho Chehab , Hans Verkuil , Sergey Senozhatsky , Linux Media Mailing List , Linux Kernel Mailing List , Tomasz Figa Content-Type: text/plain; charset="UTF-8" Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Laurent Thanks for your review! On Thu, 10 Jun 2021 at 19:28, Laurent Pinchart wrote: > > Hi Ricardo, > > Thank you for the patch. > > On Fri, Mar 26, 2021 at 10:58:34AM +0100, Ricardo Ribalda wrote: > > If a control is inactive return -EACCES to let the userspace know that > > the value will not be applied automatically when the control is active > > again. > > Isn't the device supposed to stall the control set in that case, with > the bRequestErrorCode set to "Wrong state", which uvc_query_ctrl() > translates to -EACCES already ? Could you elaborate on why this patch is > needed ? The problem is that the hardware was not returning the equivalent of EACCES, so we had to capture it manually. I will try to revert the patch and capture an error. > > > Suggested-by: Hans Verkuil > > Signed-off-by: Ricardo Ribalda > > --- > > drivers/media/usb/uvc/uvc_ctrl.c | 71 +++++++++++++++++++++----------- > > 1 file changed, 48 insertions(+), 23 deletions(-) > > > > diff --git a/drivers/media/usb/uvc/uvc_ctrl.c b/drivers/media/usb/uvc/uvc_ctrl.c > > index bcebf9d1a46f..d9d4add1e813 100644 > > --- a/drivers/media/usb/uvc/uvc_ctrl.c > > +++ b/drivers/media/usb/uvc/uvc_ctrl.c > > @@ -1082,13 +1082,36 @@ static const char *uvc_map_get_name(const struct uvc_control_mapping *map) > > return "Unknown Control"; > > } > > > > +static bool uvc_ctrl_is_inactive(struct uvc_video_chain *chain, > > + struct uvc_control *ctrl, > > + struct uvc_control_mapping *mapping) > > +{ > > + struct uvc_control_mapping *master_map = NULL; > > + struct uvc_control *master_ctrl = NULL; > > + s32 val; > > + int ret; > > + > > + if (!mapping->master_id) > > + return false; > > + > > + __uvc_find_control(ctrl->entity, mapping->master_id, &master_map, > > + &master_ctrl, 0); > > + > > + if (!master_ctrl || !(master_ctrl->info.flags & UVC_CTRL_FLAG_GET_CUR)) > > + return false; > > + > > + ret = __uvc_ctrl_get(chain, master_ctrl, master_map, &val); > > + if (ret < 0 || val == mapping->master_manual) > > + return false; > > + > > + return true; > > +} > > + > > static int __uvc_query_v4l2_ctrl(struct uvc_video_chain *chain, > > struct uvc_control *ctrl, > > struct uvc_control_mapping *mapping, > > struct v4l2_queryctrl *v4l2_ctrl) > > { > > - struct uvc_control_mapping *master_map = NULL; > > - struct uvc_control *master_ctrl = NULL; > > const struct uvc_menu_info *menu; > > unsigned int i; > > > > @@ -1104,18 +1127,8 @@ static int __uvc_query_v4l2_ctrl(struct uvc_video_chain *chain, > > if (!(ctrl->info.flags & UVC_CTRL_FLAG_SET_CUR)) > > v4l2_ctrl->flags |= V4L2_CTRL_FLAG_READ_ONLY; > > > > - if (mapping->master_id) > > - __uvc_find_control(ctrl->entity, mapping->master_id, > > - &master_map, &master_ctrl, 0); > > - if (master_ctrl && (master_ctrl->info.flags & UVC_CTRL_FLAG_GET_CUR)) { > > - s32 val; > > - int ret = __uvc_ctrl_get(chain, master_ctrl, master_map, &val); > > - if (ret < 0) > > - return ret; > > There's a small change in behaviour here, the driver used to return an > error, now it will ignore it. Is it intentional ? AFAIK The error did not follow the v4l2 spec. You shall always be able to query a ctrl. I will add it to the commit message to make it clear. > > > - > > - if (val != mapping->master_manual) > > - v4l2_ctrl->flags |= V4L2_CTRL_FLAG_INACTIVE; > > - } > > + if (uvc_ctrl_is_inactive(chain, ctrl, mapping)) > > + v4l2_ctrl->flags |= V4L2_CTRL_FLAG_INACTIVE; > > > > if (!ctrl->cached) { > > int ret = uvc_ctrl_populate_cache(chain, ctrl); > > @@ -1638,25 +1651,37 @@ static int uvc_ctrl_commit_entity(struct uvc_device *dev, > > return 0; > > } > > > > -static int uvc_ctrl_find_ctrlidx(struct uvc_entity *entity, > > +static int uvc_ctrl_commit_error(struct uvc_video_chain *chain, > > + struct uvc_entity *entity, > > struct v4l2_ext_controls *ctrls, > > - struct uvc_control *uvc_control) > > + struct uvc_control *err_control, > > + int ret) > > { > > struct uvc_control_mapping *mapping; > > struct uvc_control *ctrl_found; > > unsigned int i; > > > > - if (!entity) > > - return ctrls->count; > > + if (!entity) { > > + ctrls->error_idx = ctrls->count; > > + return ret; > > + } > > > > for (i = 0; i < ctrls->count; i++) { > > __uvc_find_control(entity, ctrls->controls[i].id, &mapping, > > &ctrl_found, 0); > > - if (uvc_control == ctrl_found) > > - return i; > > + if (err_control == ctrl_found) > > + break; > > } > > + ctrls->error_idx = i; > > I think this line should be moved after the next check. Not really, if we cannot find a control, we cannot blame it on control 0 ;) > > > + > > + /* We could not find the control that failed. */ > > + if (i == ctrls->count) > > + return ret; > > > > - return ctrls->count; > > + if (uvc_ctrl_is_inactive(chain, err_control, mapping)) > > + return -EACCES; > > + > > + return ret; > > } > > > > int __uvc_ctrl_commit(struct uvc_fh *handle, int rollback, > > @@ -1679,8 +1704,8 @@ int __uvc_ctrl_commit(struct uvc_fh *handle, int rollback, > > uvc_ctrl_send_events(handle, ctrls->controls, ctrls->count); > > done: > > if (ret < 0 && ctrls) > > - ctrls->error_idx = uvc_ctrl_find_ctrlidx(entity, ctrls, > > - err_ctrl); > > + ret = uvc_ctrl_commit_error(chain, entity, ctrls, err_ctrl, > > + ret); > > mutex_unlock(&chain->ctrl_mutex); > > return ret; > > } > > -- > Regards, > > Laurent Pinchart -- Ricardo Ribalda