[PATCH] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

Subsystems: networking drivers, the rest, usb "usbnet" driver framework, usb networking drivers · Maintainers: Andrew Lunn, "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds, Oliver Neukum

STALE2184d

12 messages, 5 authors, 2020-11-02 · open the first message on its own page

[PATCH] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Anant Thazhemadam <hidden>
Date: 2020-10-10 06:59:53

Currently, __usbnet_{read|write}_cmd() use usb_control_msg().
However, this could lead to potential partial reads/writes being
considered valid, and since most of the callers of
usbnet_{read|write}_cmd() don't take partial reads/writes into account
(only checking for negative error number is done), and this can lead to
issues.

However, the new usb_control_msg_{send|recv}() APIs don't allow partial
reads and writes.
Using the new APIs also relaxes the return value checking that must
be done after usbnet_{read|write}_cmd() is called.

Signed-off-by: Anant Thazhemadam <redacted>
---
Since not all callers of usbnet_{read|write}_cmd() check if a complete 
read/write happened, partial reads can go unnoticed.

This issue was briefly mentioned here.
	https://lore.kernel.org/linux-usb/1565777764.25764.4.camel@suse.com/

Using the new API in place of the old one doesn't break anything.
This is mainly because usb_control_msg_{send|recv}() returns 0 on success
and a negative error number on failure (which includes partial reads/writes).

Thus, the error checking condition provided by the present callers of 
usbnet_{read|write}_cmd() for failure (return value < 0 is considered as an 
error) will hold. 
And similarly, the condition checked by some callers for 'success' 
(return value >= 0 && return value < length/size) will also hold.

However, if I have missed out on any caller that this might cause problems with,
please let me know, and I will fix that up as well.

 drivers/net/usb/usbnet.c | 52 ++++++++--------------------------------
 1 file changed, 10 insertions(+), 42 deletions(-)
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index bf6c58240bd4..dd9fe530a374 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -1982,64 +1982,32 @@ EXPORT_SYMBOL(usbnet_link_change);
 static int __usbnet_read_cmd(struct usbnet *dev, u8 cmd, u8 reqtype,
 			     u16 value, u16 index, void *data, u16 size)
 {
-	void *buf = NULL;
-	int err = -ENOMEM;
 
 	netdev_dbg(dev->net, "usbnet_read_cmd cmd=0x%02x reqtype=%02x"
 		   " value=0x%04x index=0x%04x size=%d\n",
 		   cmd, reqtype, value, index, size);
 
-	if (size) {
-		buf = kmalloc(size, GFP_KERNEL);
-		if (!buf)
-			goto out;
-	}
-
-	err = usb_control_msg(dev->udev, usb_rcvctrlpipe(dev->udev, 0),
-			      cmd, reqtype, value, index, buf, size,
-			      USB_CTRL_GET_TIMEOUT);
-	if (err > 0 && err <= size) {
-        if (data)
-            memcpy(data, buf, err);
-        else
-            netdev_dbg(dev->net,
-                "Huh? Data requested but thrown away.\n");
-    }
-	kfree(buf);
-out:
-	return err;
+	return usb_control_msg_recv(dev->udev, 0,
+			      cmd, reqtype, value, index, data, size,
+			      USB_CTRL_GET_TIMEOUT, GFP_KERNEL);
 }
 
 static int __usbnet_write_cmd(struct usbnet *dev, u8 cmd, u8 reqtype,
 			      u16 value, u16 index, const void *data,
 			      u16 size)
 {
-	void *buf = NULL;
-	int err = -ENOMEM;
-
 	netdev_dbg(dev->net, "usbnet_write_cmd cmd=0x%02x reqtype=%02x"
 		   " value=0x%04x index=0x%04x size=%d\n",
 		   cmd, reqtype, value, index, size);
 
-	if (data) {
-		buf = kmemdup(data, size, GFP_KERNEL);
-		if (!buf)
-			goto out;
-	} else {
-        if (size) {
-            WARN_ON_ONCE(1);
-            err = -EINVAL;
-            goto out;
-        }
-    }
-
-	err = usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0),
-			      cmd, reqtype, value, index, buf, size,
-			      USB_CTRL_SET_TIMEOUT);
-	kfree(buf);
+	if (size && !data) {
+		WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
 
-out:
-	return err;
+	return usb_control_msg_send(dev->udev, 0,
+			cmd, reqtype, value, index, data, size,
+			USB_CTRL_SET_TIMEOUT, GPF_KERNEL);
 }
 
 /*
-- 
2.25.1

Re: [PATCH] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Anant Thazhemadam <hidden>
Date: 2020-10-10 22:51:53

On 10/10/20 10:33 pm, Jakub Kicinski wrote:
On Sat, 10 Oct 2020 12:26:23 +0530 Anant Thazhemadam wrote:
quoted
GPF_KERNEL
You haven't even built this, let alone tested :/
I'm really sorry about this.
Turns out, my .config wasn't set generated by make allyesconfig, and thus
this regression went undetected.

I can submit a v2 that doesn't break the build, and which is build tested
properly.
However, I'm not clear on how else I could locally run unit tests pertaining to this
patch. I understand the critical requirement for testing changes and would really
appreciate it if someone could direct me towards some resource or another that
could help me figure that out too.

Once again, I'm really sorry for this oversight.
Thank you for your time.

Thanks,
Anant

Re: [PATCH] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Jakub Kicinski <kuba@kernel.org>
Date: 2020-10-10 23:09:29

On Sat, 10 Oct 2020 12:26:23 +0530 Anant Thazhemadam wrote:
GPF_KERNEL
You haven't even built this, let alone tested :/

Re: [PATCH] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: kernel test robot <hidden>
Date: 2020-10-13 02:40:58

Hi Anant,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on linus/master]
[also build test ERROR on v5.9 next-20201012]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Anant-Thazhemadam/net-usb-usbnet-update-__usbnet_-read-write-_cmd-to-use-new-API/20201010-145950
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 6f2f486d57c4d562cdf4932320b66fbb878ab1c4
config: x86_64-randconfig-a005-20201012 (attached as .config)
compiler: clang version 12.0.0 (https://github.com/llvm/llvm-project 9e72d3eaf38f217698f72cb8fdc969a6e72dad3a)
reproduce (this is a W=1 build):
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # install x86_64 cross compiling tool for clang build
        # apt-get install binutils-x86-64-linux-gnu
        # https://github.com/0day-ci/linux/commit/77da62a1a1fcc73c3d8309c273dfd4d497918c6e
        git remote add linux-review https://github.com/0day-ci/linux
        git fetch --no-tags linux-review Anant-Thazhemadam/net-usb-usbnet-update-__usbnet_-read-write-_cmd-to-use-new-API/20201010-145950
        git checkout 77da62a1a1fcc73c3d8309c273dfd4d497918c6e
        # save the attached .config to linux build tree
        COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross ARCH=x86_64 

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <redacted>

All errors (new ones prefixed by >>):

   drivers/net/usb/usbnet.c:1990:9: error: implicit declaration of function 'usb_control_msg_recv' [-Werror,-Wimplicit-function-declaration]
           return usb_control_msg_recv(dev->udev, 0,
                  ^
   drivers/net/usb/usbnet.c:1990:9: note: did you mean 'usb_control_msg'?
   include/linux/usb.h:1794:12: note: 'usb_control_msg' declared here
   extern int usb_control_msg(struct usb_device *dev, unsigned int pipe,
              ^
   drivers/net/usb/usbnet.c:2008:9: error: implicit declaration of function 'usb_control_msg_send' [-Werror,-Wimplicit-function-declaration]
           return usb_control_msg_send(dev->udev, 0,

5 more lines

quoted
drivers/net/usb/usbnet.c:2010:26: error: use of undeclared identifier 'GPF_KERNEL'
                           USB_CTRL_SET_TIMEOUT, GPF_KERNEL);
                                                 ^
   3 errors generated.

vim +/GPF_KERNEL +2010 drivers/net/usb/usbnet.c

  1994	
  1995	static int __usbnet_write_cmd(struct usbnet *dev, u8 cmd, u8 reqtype,
  1996				      u16 value, u16 index, const void *data,
  1997				      u16 size)
  1998	{
  1999		netdev_dbg(dev->net, "usbnet_write_cmd cmd=0x%02x reqtype=%02x"
  2000			   " value=0x%04x index=0x%04x size=%d\n",
  2001			   cmd, reqtype, value, index, size);
  2002	
  2003		if (size && !data) {
  2004			WARN_ON_ONCE(1);
  2005			return -EINVAL;
  2006		}
  2007	
  2008		return usb_control_msg_send(dev->udev, 0,
  2009				cmd, reqtype, value, index, data, size,
2010				USB_CTRL_SET_TIMEOUT, GPF_KERNEL);
  2011	}
  2012	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[PATCH v2] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Anant Thazhemadam <hidden>
Date: 2020-10-29 13:23:39

Currently, __usbnet_{read|write}_cmd() use usb_control_msg(),
and thus consider potential partial reads/writes being done to 
be perfectly valid.
Quite a few callers of usbnet_{read|write}_cmd() don't enforce
checking for partial reads/writes into account either, automatically
assuming that a complete read/write occurs.

However, the new usb_control_msg_{send|recv}() APIs don't allow partial
reads and writes.
Using the new APIs also relaxes the return value checking that must
be done after usbnet_{read|write}_cmd() is called.

Signed-off-by: Anant Thazhemadam <redacted>
---
Changes in v2:
	* Fix build error

This patch has been compile and build tested with a .config file that
was generated using make allyesconfig, and the build error has been 
fixed.
Unfortunately, I wasn't able to get my hands on a usbnet adapter for testing,
and would appreciate it if someone could do that.

 drivers/net/usb/usbnet.c | 52 ++++++++--------------------------------
 1 file changed, 10 insertions(+), 42 deletions(-)
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index bf6c58240bd4..2f7c7b7f4047 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -1982,64 +1982,32 @@ EXPORT_SYMBOL(usbnet_link_change);
 static int __usbnet_read_cmd(struct usbnet *dev, u8 cmd, u8 reqtype,
 			     u16 value, u16 index, void *data, u16 size)
 {
-	void *buf = NULL;
-	int err = -ENOMEM;
 
 	netdev_dbg(dev->net, "usbnet_read_cmd cmd=0x%02x reqtype=%02x"
 		   " value=0x%04x index=0x%04x size=%d\n",
 		   cmd, reqtype, value, index, size);
 
-	if (size) {
-		buf = kmalloc(size, GFP_KERNEL);
-		if (!buf)
-			goto out;
-	}
-
-	err = usb_control_msg(dev->udev, usb_rcvctrlpipe(dev->udev, 0),
-			      cmd, reqtype, value, index, buf, size,
-			      USB_CTRL_GET_TIMEOUT);
-	if (err > 0 && err <= size) {
-        if (data)
-            memcpy(data, buf, err);
-        else
-            netdev_dbg(dev->net,
-                "Huh? Data requested but thrown away.\n");
-    }
-	kfree(buf);
-out:
-	return err;
+	return usb_control_msg_recv(dev->udev, 0,
+			      cmd, reqtype, value, index, data, size,
+			      USB_CTRL_GET_TIMEOUT, GFP_KERNEL);
 }
 
 static int __usbnet_write_cmd(struct usbnet *dev, u8 cmd, u8 reqtype,
 			      u16 value, u16 index, const void *data,
 			      u16 size)
 {
-	void *buf = NULL;
-	int err = -ENOMEM;
-
 	netdev_dbg(dev->net, "usbnet_write_cmd cmd=0x%02x reqtype=%02x"
 		   " value=0x%04x index=0x%04x size=%d\n",
 		   cmd, reqtype, value, index, size);
 
-	if (data) {
-		buf = kmemdup(data, size, GFP_KERNEL);
-		if (!buf)
-			goto out;
-	} else {
-        if (size) {
-            WARN_ON_ONCE(1);
-            err = -EINVAL;
-            goto out;
-        }
-    }
-
-	err = usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0),
-			      cmd, reqtype, value, index, buf, size,
-			      USB_CTRL_SET_TIMEOUT);
-	kfree(buf);
+	if (size && !data) {
+		WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
 
-out:
-	return err;
+	return usb_control_msg_send(dev->udev, 0,
+			cmd, reqtype, value, index, data, size,
+			USB_CTRL_SET_TIMEOUT, GFP_KERNEL);
 }
 
 /*
-- 
2.25.1

Re: [PATCH v2] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Anant Thazhemadam <hidden>
Date: 2020-10-29 15:17:06

On 29/10/20 6:52 pm, Anant Thazhemadam wrote:
quoted hunk
Currently, __usbnet_{read|write}_cmd() use usb_control_msg(),
and thus consider potential partial reads/writes being done to 
be perfectly valid.
Quite a few callers of usbnet_{read|write}_cmd() don't enforce
checking for partial reads/writes into account either, automatically
assuming that a complete read/write occurs.

However, the new usb_control_msg_{send|recv}() APIs don't allow partial
reads and writes.
Using the new APIs also relaxes the return value checking that must
be done after usbnet_{read|write}_cmd() is called.

Signed-off-by: Anant Thazhemadam <redacted> <mailto:<redacted>>
---
Changes in v2:
	* Fix build error

This patch has been compile and build tested with a .config file that
was generated using make allyesconfig, and the build error has been 
fixed.
Unfortunately, I wasn't able to get my hands on a usbnet adapter for testing,
and would appreciate it if someone could do that.

 drivers/net/usb/usbnet.c | 52 ++++++++--------------------------------
 1 file changed, 10 insertions(+), 42 deletions(-)
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index bf6c58240bd4..2f7c7b7f4047 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -1982,64 +1982,32 @@ EXPORT_SYMBOL(usbnet_link_change);
 static int __usbnet_read_cmd(struct usbnet *dev, u8 cmd, u8 reqtype,
 			     u16 value, u16 index, void *data, u16 size)
 {
-	void *buf = NULL;
-	int err = -ENOMEM;
 
 	netdev_dbg(dev->net, "usbnet_read_cmd cmd=0x%02x reqtype=%02x"
 		   " value=0x%04x index=0x%04x size=%d\n",
 		   cmd, reqtype, value, index, size);

65 more quoted lines

I had a v2 prepared and ready but was told to wait for a week before sending it in,
since usb_control_msg_{send|recv}() that were being used were not present in the
networking tree at the time, and all the trees would be converged by then.
So, just to be on the safer side, I waited for two weeks.
I checked the net tree, and found the APIs there too (defined in usb.h).

However the build seems to fail here,
    https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org/thread/O2BERGN7SYYC6LNOOKNUGPS2IJLDWYT7/

I'm not entirely sure at this point why this is happening, and would appreciate it if
someone could take the time to tell me if and how this might be an issue with my
patch.

Thanks,
Anant

Re: [PATCH v2] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Alan Stern <stern@rowland.harvard.edu>
Date: 2020-10-29 16:32:13

On Thu, Oct 29, 2020 at 08:46:59PM +0530, Anant Thazhemadam wrote:
I had a v2 prepared and ready but was told to wait for a week before sending it in,
since usb_control_msg_{send|recv}() that were being used were not present in the
networking tree at the time, and all the trees would be converged by then.
So, just to be on the safer side, I waited for two weeks.
I checked the net tree, and found the APIs there too (defined in usb.h).

However the build seems to fail here,
    https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org/thread/O2BERGN7SYYC6LNOOKNUGPS2IJLDWYT7/

I'm not entirely sure at this point why this is happening, and would appreciate it if
someone could take the time to tell me if and how this might be an issue with my
patch.
It's happening because the tree you used for building did not include 
the declarations of usb_control_msg_{send|recv}().

It's hard to tell exactly what that tree does contain, because the 
github.com links embedded in the web page you mention above don't work.

Alan Stern

Re: [PATCH v2] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Jakub Kicinski <kuba@kernel.org>
Date: 2020-10-31 21:11:46

On Thu, 29 Oct 2020 18:52:56 +0530 Anant Thazhemadam wrote:
quoted hunk
+	return usb_control_msg_recv(dev->udev, 0,
+			      cmd, reqtype, value, index, data, size,
+			      USB_CTRL_GET_TIMEOUT, GFP_KERNEL);
Please align continuation lines after the opening bracket.

Re: [PATCH v2] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Anant Thazhemadam <hidden>
Date: 2020-10-31 21:31:46

On 01/11/20 2:41 am, Jakub Kicinski wrote:
On Thu, 29 Oct 2020 18:52:56 +0530 Anant Thazhemadam wrote:
quoted
+	return usb_control_msg_recv(dev->udev, 0,
+			      cmd, reqtype, value, index, data, size,
+			      USB_CTRL_GET_TIMEOUT, GFP_KERNEL);
Please align continuation lines after the opening bracket.
I will do that, and send in a v3 right away.

Thanks,
Anant

[PATCH v3] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Anant Thazhemadam <hidden>
Date: 2020-10-31 21:35:42

Currently, __usbnet_{read|write}_cmd() use usb_control_msg().
However, this could lead to potential partial reads/writes being
considered valid, and since most of the callers of
usbnet_{read|write}_cmd() don't take partial reads/writes into account
(only checking for negative error number is done), and this can lead to
issues.

However, the new usb_control_msg_{send|recv}() APIs don't allow partial
reads and writes.
Using the new APIs also relaxes the return value checking that must
be done after usbnet_{read|write}_cmd() is called.

Signed-off-by: Anant Thazhemadam <redacted>
---
Changes in v3:
	* Aligned continuation lines after the opening brackets
Changes in v2:
	* Fix build error

 drivers/net/usb/usbnet.c | 52 ++++++++--------------------------------
 1 file changed, 10 insertions(+), 42 deletions(-)
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index bf6c58240bd4..b2df3417a41c 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -1982,64 +1982,32 @@ EXPORT_SYMBOL(usbnet_link_change);
 static int __usbnet_read_cmd(struct usbnet *dev, u8 cmd, u8 reqtype,
 			     u16 value, u16 index, void *data, u16 size)
 {
-	void *buf = NULL;
-	int err = -ENOMEM;
 
 	netdev_dbg(dev->net, "usbnet_read_cmd cmd=0x%02x reqtype=%02x"
 		   " value=0x%04x index=0x%04x size=%d\n",
 		   cmd, reqtype, value, index, size);
 
-	if (size) {
-		buf = kmalloc(size, GFP_KERNEL);
-		if (!buf)
-			goto out;
-	}
-
-	err = usb_control_msg(dev->udev, usb_rcvctrlpipe(dev->udev, 0),
-			      cmd, reqtype, value, index, buf, size,
-			      USB_CTRL_GET_TIMEOUT);
-	if (err > 0 && err <= size) {
-        if (data)
-            memcpy(data, buf, err);
-        else
-            netdev_dbg(dev->net,
-                "Huh? Data requested but thrown away.\n");
-    }
-	kfree(buf);
-out:
-	return err;
+	return usb_control_msg_recv(dev->udev, 0,
+				    cmd, reqtype, value, index, data, size,
+				    USB_CTRL_GET_TIMEOUT, GFP_KERNEL);
 }
 
 static int __usbnet_write_cmd(struct usbnet *dev, u8 cmd, u8 reqtype,
 			      u16 value, u16 index, const void *data,
 			      u16 size)
 {
-	void *buf = NULL;
-	int err = -ENOMEM;
-
 	netdev_dbg(dev->net, "usbnet_write_cmd cmd=0x%02x reqtype=%02x"
 		   " value=0x%04x index=0x%04x size=%d\n",
 		   cmd, reqtype, value, index, size);
 
-	if (data) {
-		buf = kmemdup(data, size, GFP_KERNEL);
-		if (!buf)
-			goto out;
-	} else {
-        if (size) {
-            WARN_ON_ONCE(1);
-            err = -EINVAL;
-            goto out;
-        }
-    }
-
-	err = usb_control_msg(dev->udev, usb_sndctrlpipe(dev->udev, 0),
-			      cmd, reqtype, value, index, buf, size,
-			      USB_CTRL_SET_TIMEOUT);
-	kfree(buf);
+	if (size && !data) {
+		WARN_ON_ONCE(1);
+		return -EINVAL;
+	}
 
-out:
-	return err;
+	return usb_control_msg_send(dev->udev, 0,
+				    cmd, reqtype, value, index, data, size,
+				    USB_CTRL_SET_TIMEOUT, GFP_KERNEL);
 }
 
 /*
-- 
2.25.1

Re: [PATCH v3] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Oliver Neukum <oneukum@suse.com>
Date: 2020-11-02 09:41:09

Am Sonntag, den 01.11.2020, 03:05 +0530 schrieb Anant Thazhemadam:
Currently, __usbnet_{read|write}_cmd() use usb_control_msg().
However, this could lead to potential partial reads/writes being
considered valid, and since most of the callers of
usbnet_{read|write}_cmd() don't take partial reads/writes into account
(only checking for negative error number is done), and this can lead to
issues.
Hi,

plesae send this as a post of its own. We cannot take a new set
as a reply to an older set.

	Regards
		Oliver

Re: [PATCH v3] net: usb: usbnet: update __usbnet_{read|write}_cmd() to use new API

From: Anant Thazhemadam <hidden>
Date: 2020-11-02 16:41:07

On 02/11/20 3:10 pm, Oliver Neukum wrote:
Am Sonntag, den 01.11.2020, 03:05 +0530 schrieb Anant Thazhemadam:
quoted
Currently, __usbnet_{read|write}_cmd() use usb_control_msg().
However, this could lead to potential partial reads/writes being
considered valid, and since most of the callers of
usbnet_{read|write}_cmd() don't take partial reads/writes into account
(only checking for negative error number is done), and this can lead to
issues.
Hi,

plesae send this as a post of its own. We cannot take a new set
as a reply to an older set.

	Regards
		Oliver
Got it. I will do that.

Thanks,
Anant
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help