All of lore.kernel.org
 help / color / mirror / Atom feed
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
To: Johannes Berg <johannes@sipsolutions.net>,
	<linux-kernel@vger.kernel.org>
Cc: rafael@kernel.org, linux-arm-msm@vger.kernel.org,
	dri-devel@lists.freedesktop.org, swboyd@chromium.org,
	khsieh@codeaurora.org, nganji@codeaurora.org,
	seanpaul@chromium.org, gregkh@linuxfoundation.org,
	dmitry.baryshkov@linaro.org, aravindh@codeaurora.org,
	freedreno@lists.freedesktop.org
Subject: Re: [PATCH] devcoredump: increase the device delete timeout to 10 mins
Date: Tue, 8 Feb 2022 17:55:18 -0800	[thread overview]
Message-ID: <b60d30cf-e435-49c4-a251-b910bc2e94ae@quicinc.com> (raw)
In-Reply-To: <c2a6e29063793eecc5c65d32af9d826544404ecc.camel@sipsolutions.net>

Hi Johannes

On 2/8/2022 1:54 PM, Johannes Berg wrote:
> On Tue, 2022-02-08 at 13:40 -0800, Abhinav Kumar wrote:
>>>
>> I am checking what usermode sees and will get back ( I didnt see an
>> error do most likely it was EOF ). I didnt follow the second part.
> 
> I think probably it got -ENODEV, looking at kernfs_file_read_iter().
> 
>> If the file descriptor read returns EOF, even if we consider them
>> separate how will it resolve this issue?
>>
>> My earlier questions were related to fixing it in devcoredump to detect
>> and fix it there. Are you suggesting to fix in usermode instead? How?
>>
> 
> Yeah, no, you cannot fix it in userspace.
> 
> But I just followed the rabbit hole down kernfs and all, and it looks
> like indeed the read would be cut short with -ENODEV, sorry.
> 
> It doesn't look like there's good API for this, but it seems at least
> from the underlying kernfs POV it should be possible to get_device() in
> open and put_device() in release, so that the device sticks around while
> somebody has the file open? It's entirely virtual, so this should be OK?
> 
> johannes

Are you suggesting something like below?

diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 42dcf96..14203d0 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -32,6 +32,22 @@ static const struct sysfs_ops *sysfs_file_ops(struct 
kernfs_node *kn)
         return kobj->ktype ? kobj->ktype->sysfs_ops : NULL;
  }

+static int sysfs_kf_open(struct kernfs_open_file *of)
+{
+       struct kobject *kobj = of->kn->parent->priv;
+       struct device *dev = kobj_to_dev(kobj);
+
+       get_device(dev);
+}
+
+static void sysfs_kf_release(struct kernfs_open_file *of)
+{
+       struct kobject *kobj = of->kn->parent->priv;
+       struct device *dev = kobj_to_dev(kobj);
+
+       put_device(dev);
+}
+
  /*
   * Reads on sysfs are handled through seq_file, which takes care of hairy
   * details like buffering and seeking.  The following function pipes
@@ -211,6 +227,8 @@ static const struct kernfs_ops sysfs_file_kfops_wo = {
  };

  static const struct kernfs_ops sysfs_file_kfops_rw = {
+       .open       = sysfs_kf_open;
+       .release    = sysfs_kf_release;
         .seq_show       = sysfs_kf_seq_show,
         .write          = sysfs_kf_write,
  };

If so, dont you think this will be a more intrusive change just for the 
sake of devcoredump? Any other way to keep the changes limited to 
devcoredump?

Thanks

Abhinav


WARNING: multiple messages have this Message-ID (diff)
From: Abhinav Kumar <quic_abhinavk@quicinc.com>
To: Johannes Berg <johannes@sipsolutions.net>,
	<linux-kernel@vger.kernel.org>
Cc: <rafael@kernel.org>, <linux-arm-msm@vger.kernel.org>,
	<dri-devel@lists.freedesktop.org>, <swboyd@chromium.org>,
	<khsieh@codeaurora.org>, <nganji@codeaurora.org>,
	<seanpaul@chromium.org>, <gregkh@linuxfoundation.org>,
	<dmitry.baryshkov@linaro.org>, <aravindh@codeaurora.org>,
	<freedreno@lists.freedesktop.org>
Subject: Re: [PATCH] devcoredump: increase the device delete timeout to 10 mins
Date: Tue, 8 Feb 2022 17:55:18 -0800	[thread overview]
Message-ID: <b60d30cf-e435-49c4-a251-b910bc2e94ae@quicinc.com> (raw)
In-Reply-To: <c2a6e29063793eecc5c65d32af9d826544404ecc.camel@sipsolutions.net>

Hi Johannes

On 2/8/2022 1:54 PM, Johannes Berg wrote:
> On Tue, 2022-02-08 at 13:40 -0800, Abhinav Kumar wrote:
>>>
>> I am checking what usermode sees and will get back ( I didnt see an
>> error do most likely it was EOF ). I didnt follow the second part.
> 
> I think probably it got -ENODEV, looking at kernfs_file_read_iter().
> 
>> If the file descriptor read returns EOF, even if we consider them
>> separate how will it resolve this issue?
>>
>> My earlier questions were related to fixing it in devcoredump to detect
>> and fix it there. Are you suggesting to fix in usermode instead? How?
>>
> 
> Yeah, no, you cannot fix it in userspace.
> 
> But I just followed the rabbit hole down kernfs and all, and it looks
> like indeed the read would be cut short with -ENODEV, sorry.
> 
> It doesn't look like there's good API for this, but it seems at least
> from the underlying kernfs POV it should be possible to get_device() in
> open and put_device() in release, so that the device sticks around while
> somebody has the file open? It's entirely virtual, so this should be OK?
> 
> johannes

Are you suggesting something like below?

diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c
index 42dcf96..14203d0 100644
--- a/fs/sysfs/file.c
+++ b/fs/sysfs/file.c
@@ -32,6 +32,22 @@ static const struct sysfs_ops *sysfs_file_ops(struct 
kernfs_node *kn)
         return kobj->ktype ? kobj->ktype->sysfs_ops : NULL;
  }

+static int sysfs_kf_open(struct kernfs_open_file *of)
+{
+       struct kobject *kobj = of->kn->parent->priv;
+       struct device *dev = kobj_to_dev(kobj);
+
+       get_device(dev);
+}
+
+static void sysfs_kf_release(struct kernfs_open_file *of)
+{
+       struct kobject *kobj = of->kn->parent->priv;
+       struct device *dev = kobj_to_dev(kobj);
+
+       put_device(dev);
+}
+
  /*
   * Reads on sysfs are handled through seq_file, which takes care of hairy
   * details like buffering and seeking.  The following function pipes
@@ -211,6 +227,8 @@ static const struct kernfs_ops sysfs_file_kfops_wo = {
  };

  static const struct kernfs_ops sysfs_file_kfops_rw = {
+       .open       = sysfs_kf_open;
+       .release    = sysfs_kf_release;
         .seq_show       = sysfs_kf_seq_show,
         .write          = sysfs_kf_write,
  };

If so, dont you think this will be a more intrusive change just for the 
sake of devcoredump? Any other way to keep the changes limited to 
devcoredump?

Thanks

Abhinav


  reply	other threads:[~2022-02-09  1:55 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-08 19:44 [PATCH] devcoredump: increase the device delete timeout to 10 mins Abhinav Kumar
2022-02-08 19:44 ` Abhinav Kumar
2022-02-08 20:35 ` Johannes Berg
2022-02-08 20:35   ` Johannes Berg
2022-02-08 21:04   ` Abhinav Kumar
2022-02-08 21:04     ` Abhinav Kumar
2022-02-08 21:12     ` Johannes Berg
2022-02-08 21:12       ` Johannes Berg
2022-02-08 21:40       ` Abhinav Kumar
2022-02-08 21:40         ` Abhinav Kumar
2022-02-08 21:54         ` Johannes Berg
2022-02-09  1:55           ` Abhinav Kumar [this message]
2022-02-09  1:55             ` Abhinav Kumar
2022-02-09  7:50             ` Johannes Berg
2022-02-09 16:29               ` Abhinav Kumar
2022-02-09 16:29                 ` Abhinav Kumar
2022-02-11 11:09             ` Greg KH
2022-02-11 11:09               ` Greg KH
2022-02-11 11:09 ` Greg KH
2022-02-11 11:09   ` Greg KH
2022-02-11 18:59   ` Abhinav Kumar
2022-02-11 18:59     ` Abhinav Kumar
2022-02-12  7:04     ` Greg KH
2022-02-12  7:04       ` Greg KH
2022-02-12  7:52       ` Abhinav Kumar
2022-02-12  7:52         ` Abhinav Kumar
2022-02-12  8:24         ` Johannes Berg
2022-02-12  8:24           ` Johannes Berg
2022-02-12  8:35           ` Abhinav Kumar
2022-02-12  8:35             ` Abhinav Kumar
2022-02-28 21:38             ` Abhinav Kumar
2022-02-28 21:38               ` Abhinav Kumar
2022-03-01  6:48               ` David Laight
2022-03-01 17:45                 ` Rob Clark
2022-03-01 17:45                   ` Rob Clark
2022-03-11 11:53                   ` Johannes Berg
2022-03-11 11:53                     ` Johannes Berg
2022-02-12  8:29         ` Greg KH
2022-02-12  8:29           ` Greg KH
2022-02-12  8:33           ` Abhinav Kumar
2022-02-12  8:33             ` Abhinav Kumar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b60d30cf-e435-49c4-a251-b910bc2e94ae@quicinc.com \
    --to=quic_abhinavk@quicinc.com \
    --cc=aravindh@codeaurora.org \
    --cc=dmitry.baryshkov@linaro.org \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=freedreno@lists.freedesktop.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=johannes@sipsolutions.net \
    --cc=khsieh@codeaurora.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=nganji@codeaurora.org \
    --cc=rafael@kernel.org \
    --cc=seanpaul@chromium.org \
    --cc=swboyd@chromium.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.