usb:composite:fix Provide function data when addressing device with only one interface

PATCH New Lukasz Majewski 5 participants 15 messages Mar 1, 2013 – Jul 1, 2013 USB

Share & tools

Lukasz Majewski

author

2013-03-01 14:30 UTC #0 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).

This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors:
Error obtaining DFU functional descriptor
Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected
...
Warning: Trying default transfer size 1024

This fix allows passing non-standard request to function setup code, where it shall be handled.

Tested at: Trats (exynos4210)
Tested with:DFU and UMS gadgets

Signed-off-by Lukasz Majewski <[email protected]>

Signed-off-by Kyungmin Park <[email protected]>

---
This patch fixes problem pointed by Pantelis Antoniou.
It is supposed to replace following commit added to u-boot-usb repository:

"dfu: Send correct DFU response from composite_setup"
SHA1: fef31049fa06e516945114195eb14ac3549d0ad2

---
 drivers/usb/gadget/composite.c |   19 +++++++++++++++++++
 1 file changed, 19 insertions(+)

drivers/usb/gadget/composite.c +19 -0

index ebb5131..2c5600e 100644
@@ -859,6 +859,25 @@ unknown:
 			if (&f->list == &cdev->config->functions)
 				f = NULL;
 			break;
+		/*
+		 * dfu-util (version 0.5) sets bmRequestType.Receipent = Device
+		 * for non-standard request (w_value = 0x21,
+		 * bRequest = GET_DESCRIPTOR in this case).
+		 * When only one interface is registered (as it is done now),
+		 * then this request shall be handled as it was requested for
+		 * interface.
+		 *
+		 * In the below code it is checked if only one interface is
+		 * present and proper function for it is extracted. Due to that
+		 * function's setup (f->setup) is called to handle this
+		 * special non-standard request.
+		 */
+		case USB_RECIP_DEVICE:
+			debug("cdev->config->next_interface_id: %d intf: %d\n",
+			       cdev->config->next_interface_id, intf);
+			if (cdev->config->next_interface_id == 1)
+				f = cdev->config->interface[intf];
+			break;
 		}
 
 		if (f && f->setup)

Signature · 1 line

1.7.10.4

Marek Vasut

2013-03-01 14:34 UTC #1 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

Dear Lukasz Majewski,

Quoted 15 lines

This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors:
Error obtaining DFU functional descriptor
Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected
...
Warning: Trying default transfer size 1024

Quoted 5 lines

This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210)
Tested with:DFU and UMS gadgets

Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.

Best regards,
Marek Vasut

Lukasz Majewski

author

2013-03-01 15:34 UTC #2 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

Hi Marek,

Quoted 17 lines

Dear Lukasz Majewski,
Quoted 15 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors:
Error obtaining DFU functional descriptor
Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected
...
Warning: Trying default transfer size 1024

Quoted 8 lines

Quoted 5 lines
This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210)
Tested with:DFU and UMS gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.

This is only one patch to replace patch done by Pantelis:

Patch to replace (revert):
dfu: Send correct DFU response from composite_setup
SHA1: fef31049fa06e516945114195eb14ac3549d0ad2

And this shall be replaced with posted patch.

Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:

http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/

Quoted 3 lines

Best regards,
Marek Vasut

Signature · 5 lines

Best regards,

Lukasz Majewski

Samsung R&D Poland (SRPOL) | Linux Platform Group

Marek Vasut

2013-03-05 23:24 UTC #3 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

Dear Lukasz Majewski,

Quoted 34 lines

Hi Marek,
Quoted 17 lines
Dear Lukasz Majewski,
Quoted 15 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors:
Error obtaining DFU functional descriptor
Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected
...
Warning: Trying default transfer size 1024
Quoted 8 lines
Quoted 5 lines
This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210)
Tested with:DFU and UMS gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.
This is only one patch to replace patch done by Pantelis:
Patch to replace (revert):
dfu: Send correct DFU response from composite_setup
SHA1: fef31049fa06e516945114195eb14ac3549d0ad2

Quoted 8 lines

And this shall be replaced with posted patch.
Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:
http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/

Ok, please check the u-boot-usb ... if it's OK, I will send Tom a pullrq (finally). Tom, do you still accept those for .04 ?

Best regards,
Marek Vasut

Tom Rini

2013-03-05 23:44 UTC #4 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 39 lines · Marek Vasut

Dear Lukasz Majewski,
Quoted 37 lines
Hi Marek,
Quoted 16 lines
Dear Lukasz Majewski,
Quoted 14 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors: Error obtaining DFU functional descriptor Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected ... Warning:
Trying default transfer size 1024
Quoted 8 lines
Quoted 5 lines
This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210) Tested with:DFU and UMS
gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.
This is only one patch to replace patch done by Pantelis:
Patch to replace (revert): dfu: Send correct DFU response from composite_setup SHA1: fef31049fa06e516945114195eb14ac3549d0ad2
And this shall be replaced with posted patch.
Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:

Quoted 6 lines

Quoted 3 lines
http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/
Ok, please check the u-boot-usb ... if it's OK, I will send Tom a pullrq (finally). Tom, do you still accept those for .04 ?

As of earlier in the day it still had Pantelis' patch for sending files larger than memory which breaks all filesystem-based writes, so that needs to go. Lukasz and I are discussing how to deal with that.

- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQIcBAEBAgAGBQJRNoNSAAoJENk4IS6UOR1WnDkP+QE7ihcrFz0u5v2wAhkjwtaZ etpl6oL/rB7jA6jK9zDYkyoc/Bf9A9/ffxsfGruP8kzjK2BTwHizs/Hz4I8Z90k6 6g4S7DUY4kx9t20Iyyswwuo8Bs0MXuEyDSS1a3kWXsB1110eUXd6xBsQoOKv5BeK 1UO29ZFMRJAnIpKBSrrGNE0e8xG8fnF1y8/Wos8OAk240DKoMPTF1lZEB5TEkbtN 7xePk4XLGClePj8CH8iIBj2YDo5kjRPUCZ2zSd6niayVrp5On5as+Fn9XTvWiQYG UjLDbeAhNinzRJx9q8sB2dxXq2sSh+8ehtKZgnQCwOuj0bV0Ok4um0FP7DKN1Z+K COvJdKC629rZ9cyc/kLfXtJzEYBJwK+OsLjLHOf7YuuhQpcPcugZWmdc8LZt6QMA aL61JGTCfF80PVO8jAjWkJjniTHTSQRL+Adt91wej9D8e246/SbzFMqjjpTWQvVB RcR6fUwKdcMwT+RKAJybYH8z3tqwiRQXleBPs6NqcRYV7E8AHzpeMUh5v7m2Gz/P 8y+ea4v8Nglq79UnYmeqCHmvMO0ZqUpb8/+hB35xfkyuIHstx3J784NHOaQ7IZqj iJZ/Stf3De+ASddDFeAldoyc3EdT2k98gYUKpitExTOB5EbWKrK781JOEC0q+iSb +xUyux8npF6Rhhf55Rm7
=DAuM
-----END PGP SIGNATURE-----

Marek Vasut

2013-03-05 23:54 UTC #5 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

Dear Tom Rini,

Quoted 44 lines · Marek Vasut

Quoted 43 lines
Dear Lukasz Majewski,
Quoted 40 lines
Hi Marek,
Quoted 16 lines
Dear Lukasz Majewski,
Quoted 14 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors: Error obtaining DFU functional descriptor Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected ... Warning:
Trying default transfer size 1024
Quoted 8 lines
Quoted 5 lines
This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210) Tested with:DFU and UMS
gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.
This is only one patch to replace patch done by Pantelis:
Patch to replace (revert): dfu: Send correct DFU response from composite_setup SHA1: fef31049fa06e516945114195eb14ac3549d0ad2
And this shall be replaced with posted patch.
Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:
http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/

Quoted 6 lines

Quoted 2 lines
Ok, please check the u-boot-usb ... if it's OK, I will send Tom a pullrq (finally). Tom, do you still accept those for .04 ?
As of earlier in the day it still had Pantelis' patch for sending files larger than memory which breaks all filesystem-based writes, so that needs to go. Lukasz and I are discussing how to deal with that.

IIRC I killed that patch.

Best regards,
Marek Vasut

Tom Rini

2013-03-06 00:26 UTC #6 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

Quoted 46 lines · Marek Vasut

Dear Tom Rini,
Quoted 44 lines · Marek Vasut
Quoted 43 lines
Dear Lukasz Majewski,
Quoted 40 lines
Hi Marek,
Quoted 16 lines
Dear Lukasz Majewski,
Quoted 14 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors: Error obtaining DFU functional descriptor Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected ... Warning:
Trying default transfer size 1024
Quoted 8 lines
Quoted 5 lines
This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210) Tested with:DFU and UMS
gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.
This is only one patch to replace patch done by Pantelis:
Patch to replace (revert): dfu: Send correct DFU response from composite_setup SHA1: fef31049fa06e516945114195eb14ac3549d0ad2
And this shall be replaced with posted patch.
Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:
http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/

Quoted 8 lines

Quoted 6 lines
Quoted 2 lines
Ok, please check the u-boot-usb ... if it's OK, I will send Tom a pullrq (finally). Tom, do you still accept those for .04 ?
As of earlier in the day it still had Pantelis' patch for sending files larger than memory which breaks all filesystem-based writes, so that needs to go. Lukasz and I are discussing how to deal with that.
IIRC I killed that patch.

In u-boot-usb/master currently:
commit a09f955db6ad7031bd715a615fb8b90d671a14d4

Author: Pantelis Antoniou <[email protected]>
Date:   Fri Nov 30 08:01:12 2012 +0000

dfu: Support larger than memory transfers.

Is one of the two that needed to go. The patch in this thread replaced the other one that you did drop.

Signature · 8 lines

Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130305/fb20e169/attachment.pgp>

Marek Vasut

2013-03-06 00:55 UTC #7 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

Dear Tom Rini,

Quoted 58 lines · Marek Vasut

Quoted 46 lines
Dear Tom Rini,
Quoted 44 lines · Marek Vasut
Quoted 43 lines
Dear Lukasz Majewski,
Quoted 40 lines
Hi Marek,
Quoted 16 lines
Dear Lukasz Majewski,
Quoted 14 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors: Error obtaining DFU functional descriptor Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected ... Warning:
Trying default transfer size 1024
Quoted 8 lines
Quoted 5 lines
This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210) Tested with:DFU and UMS
gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.
This is only one patch to replace patch done by Pantelis:
Patch to replace (revert): dfu: Send correct DFU response from composite_setup SHA1: fef31049fa06e516945114195eb14ac3549d0ad2
And this shall be replaced with posted patch.
Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:
http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/
Quoted 8 lines
Quoted 6 lines
Quoted 2 lines
Ok, please check the u-boot-usb ... if it's OK, I will send Tom a pullrq (finally). Tom, do you still accept those for .04 ?
As of earlier in the day it still had Pantelis' patch for sending files larger than memory which breaks all filesystem-based writes, so that needs to go. Lukasz and I are discussing how to deal with that.
IIRC I killed that patch.
In u-boot-usb/master currently:
commit a09f955db6ad7031bd715a615fb8b90d671a14d4

Quoted 7 lines

Author: Pantelis Antoniou <[email protected]>
Date:   Fri Nov 30 08:01:12 2012 +0000
dfu: Support larger than memory transfers.
Is one of the two that needed to go. The patch in this thread replaced the other one that you did drop.

True, we now have the proper one applied to my understanding.

Best regards,
Marek Vasut

Tom Rini

2013-03-06 01:38 UTC #8 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Quoted 70 lines · Marek Vasut

Dear Tom Rini,
Quoted 68 lines · Marek Vasut
Quoted 55 lines
Dear Tom Rini,
Quoted 53 lines · Marek Vasut
Quoted 51 lines
Dear Lukasz Majewski,
Quoted 37 lines
Hi Marek,
Quoted 21 lines
Dear Lukasz Majewski,
Quoted 19 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors: Error obtaining DFU functional descriptor Warning: Assuming DFU version 1.0 Warning: Transfer size can not be detected ... Warning: Trying default transfer size 1024
This fix allows passing non-standard request to function setup code, where it shall be handled.
Quoted 6 lines
Quoted 2 lines
Tested at: Trats (exynos4210) Tested with:DFU and
UMS gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.
This is only one patch to replace patch done by
Pantelis:
Patch to replace (revert): dfu: Send correct DFU response from composite_setup SHA1:
fef31049fa06e516945114195eb14ac3549d0ad2
Quoted 8 lines
And this shall be replaced with posted patch.
Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:
http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/
Ok, please check the u-boot-usb ... if it's OK, I will send Tom a pullrq (finally). Tom, do you still accept those for .04 ?
Quoted 6 lines
Quoted 4 lines
As of earlier in the day it still had Pantelis' patch for sending files larger than memory which breaks all filesystem-based writes, so that needs to go. Lukasz and I are discussing how to deal with that.
IIRC I killed that patch.
In u-boot-usb/master currently: commit a09f955db6ad7031bd715a615fb8b90d671a14d4 Author: Pantelis
Antoniou <[email protected]> Date: Fri Nov 30
08:01:12 2012 +0000

Quoted 6 lines

Quoted 4 lines
dfu: Support larger than memory transfers.
Is one of the two that needed to go. The patch in this thread replaced the other one that you did drop.
True, we now have the proper one applied to my understanding.

No, the concept needs re-working as non-filesystem can work in chunks but files must be done all at once.

- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQIcBAEBAgAGBQJRNp4EAAoJENk4IS6UOR1W61YQAJutQ0MUjAY5uw/P7YrY9o5C KPiuxLrTgaTEl6cNeeGWS30YxqX7cz7w/SQd8igKno5sqeoNncggc0CwkAumpxRe baBxb1VVePijS9NFoQC/zMMzJ8+H4Csx9OLYW1EqlYL30DDpNqS7T2aYznQ7hsth wyOA7oasGm5KY6tgsi8e8Qfw/dNER//OWgCx+6KqsUJemg5MPSZCABAKQmtNz6yU jgaXC/z5bx3bfrAa/yf8zaNnbtMLhjcM8Bwhd3oiOsaFgU0eGS/IFJe6gDK9Gpfw NrehYEBmA3xKXROi7ph8c4yHUaPqA62nAPO0s60uTOQ+K6rXeh5eZ1OPe5Y99raG Ka0Dx+GT2zT4PJcMTML8xWcF5o7ugD3kG8YbVyWM8OfT7deuZscvdsHaL6k7KOtM pVrM/lfurHs2wgR4wCLHXCxoRYioHHl4ZgSFTI0RwX/fTphYokX3hloVEMlbRLtR 2QTknQIq7lwRvAOem0CrFczTfVMXtawZ9UDQ6qQ/jgIyQbfyUSyywsTWs2Ts8WSL PTfOgRrTK9cXk5PCT+pKVv1160mYUCu7c0YYCImqmq9T2qw5QvACpc9FXoSpo/9n JJ9tt5awX+2nZhrVGqc+d+cfM957xqMJ2ViD3M7Tw8wwy/076FOBvtDFbvA73pV+ HHh8svg50YA8O0dI9n5F
=s8vQ
-----END PGP SIGNATURE-----

Lukasz Majewski

author

2013-03-06 07:42 UTC #9 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

In reply to Tom Rini

Dear All,

Quoted 49 lines

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 45 lines · Marek Vasut
Dear Lukasz Majewski,
Quoted 32 lines
Hi Marek,
Quoted 16 lines
Dear Lukasz Majewski,
Quoted 14 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors: Error obtaining DFU functional descriptor Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected ... Warning:
Trying default transfer size 1024
Quoted 8 lines
Quoted 5 lines
This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210) Tested with:DFU and UMS
gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.
This is only one patch to replace patch done by Pantelis:
Patch to replace (revert): dfu: Send correct DFU response from composite_setup SHA1: fef31049fa06e516945114195eb14ac3549d0ad2
Quoted 8 lines
And this shall be replaced with posted patch.
Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:
http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/
Ok, please check the u-boot-usb ... if it's OK, I will send Tom a pullrq (finally). Tom, do you still accept those for .04 ?

Marek, thanks for pulling those patches.

Quoted 5 lines

As of earlier in the day it still had Pantelis' patch for sending files larger than memory which breaks all filesystem-based writes, so that needs to go. Lukasz and I are discussing how to deal with that.

The patch about which Tom is speaking is:
"dfu: Support larger than memory transfers."
SHA1: a09f955db6ad7031bd715a615fb8b90d671a14d4

This patch is causing regression (not only on Trats).

I think, that it shall be dropped and replaced by new patch on which Tom (with my little help) is working now.

This patch is not connected to the one, which you have already removed from u-boot-usb tree.
You have removed patch about composite.c DFU fix (which was fixed by:
"usb:composite:fix Provide function data when addressing device with only one interface").

Quoted 20 lines

- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQIcBAEBAgAGBQJRNoNSAAoJENk4IS6UOR1WnDkP+QE7ihcrFz0u5v2wAhkjwtaZ etpl6oL/rB7jA6jK9zDYkyoc/Bf9A9/ffxsfGruP8kzjK2BTwHizs/Hz4I8Z90k6 6g4S7DUY4kx9t20Iyyswwuo8Bs0MXuEyDSS1a3kWXsB1110eUXd6xBsQoOKv5BeK 1UO29ZFMRJAnIpKBSrrGNE0e8xG8fnF1y8/Wos8OAk240DKoMPTF1lZEB5TEkbtN 7xePk4XLGClePj8CH8iIBj2YDo5kjRPUCZ2zSd6niayVrp5On5as+Fn9XTvWiQYG UjLDbeAhNinzRJx9q8sB2dxXq2sSh+8ehtKZgnQCwOuj0bV0Ok4um0FP7DKN1Z+K COvJdKC629rZ9cyc/kLfXtJzEYBJwK+OsLjLHOf7YuuhQpcPcugZWmdc8LZt6QMA aL61JGTCfF80PVO8jAjWkJjniTHTSQRL+Adt91wej9D8e246/SbzFMqjjpTWQvVB RcR6fUwKdcMwT+RKAJybYH8z3tqwiRQXleBPs6NqcRYV7E8AHzpeMUh5v7m2Gz/P 8y+ea4v8Nglq79UnYmeqCHmvMO0ZqUpb8/+hB35xfkyuIHstx3J784NHOaQ7IZqj iJZ/Stf3De+ASddDFeAldoyc3EdT2k98gYUKpitExTOB5EbWKrK781JOEC0q+iSb +xUyux8npF6Rhhf55Rm7
=DAuM
-----END PGP SIGNATURE-----

Signature · 5 lines

Best regards,

Lukasz Majewski

Samsung R&D Poland (SRPOL) | Linux Platform Group

Pantelis Antoniou

2013-03-06 07:45 UTC #10 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

Hi Lukasz,

Quoted 66 lines · Lukasz Majewski

Dear All,
Quoted 49 lines
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Quoted 45 lines · Marek Vasut
Dear Lukasz Majewski,
Quoted 32 lines
Hi Marek,
Quoted 16 lines
Dear Lukasz Majewski,
Quoted 14 lines
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors: Error obtaining DFU functional descriptor Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected ... Warning:
Trying default transfer size 1024
Quoted 8 lines
Quoted 5 lines
This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210) Tested with:DFU and UMS
gadgets
Can you please repost all the patches you want applied on u-boot-usb ? I'm completely lost in all this DFU stuff.
This is only one patch to replace patch done by Pantelis:
Patch to replace (revert): dfu: Send correct DFU response from composite_setup SHA1: fef31049fa06e516945114195eb14ac3549d0ad2
Quoted 8 lines
And this shall be replaced with posted patch.
Another patches (with UMS gadget support - done by Piotr Wilczek), also shall be added to u-boot-usb tree:
http://patchwork.ozlabs.org/patch/219744/
http://patchwork.ozlabs.org/patch/219746/
http://patchwork.ozlabs.org/patch/219745/
Ok, please check the u-boot-usb ... if it's OK, I will send Tom a pullrq (finally). Tom, do you still accept those for .04 ?
Marek, thanks for pulling those patches.
Quoted 5 lines
As of earlier in the day it still had Pantelis' patch for sending files larger than memory which breaks all filesystem-based writes, so that needs to go. Lukasz and I are discussing how to deal with that.
The patch about which Tom is speaking is:
"dfu: Support larger than memory transfers."
SHA1: a09f955db6ad7031bd715a615fb8b90d671a14d4
This patch is causing regression (not only on Trats).

Quoted 9 lines

I think, that it shall be dropped and replaced by new patch on which Tom (with my little help) is working now.
This patch is not connected to the one, which you have already removed from u-boot-usb tree.
You have removed patch about composite.c DFU fix (which was fixed by:
"usb:composite:fix Provide function data when addressing device with only one interface").

Sorry about missing this thread. I've been busy with other stuff.

What kind of regressions are we talking about?

I'm trying to get around to looking into all of this, but this shouldn't be causing all that trouble.

Regards

-- Pantelis

Quoted 31 lines

Quoted 22 lines
- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQIcBAEBAgAGBQJRNoNSAAoJENk4IS6UOR1WnDkP+QE7ihcrFz0u5v2wAhkjwtaZ etpl6oL/rB7jA6jK9zDYkyoc/Bf9A9/ffxsfGruP8kzjK2BTwHizs/Hz4I8Z90k6 6g4S7DUY4kx9t20Iyyswwuo8Bs0MXuEyDSS1a3kWXsB1110eUXd6xBsQoOKv5BeK 1UO29ZFMRJAnIpKBSrrGNE0e8xG8fnF1y8/Wos8OAk240DKoMPTF1lZEB5TEkbtN 7xePk4XLGClePj8CH8iIBj2YDo5kjRPUCZ2zSd6niayVrp5On5as+Fn9XTvWiQYG UjLDbeAhNinzRJx9q8sB2dxXq2sSh+8ehtKZgnQCwOuj0bV0Ok4um0FP7DKN1Z+K COvJdKC629rZ9cyc/kLfXtJzEYBJwK+OsLjLHOf7YuuhQpcPcugZWmdc8LZt6QMA aL61JGTCfF80PVO8jAjWkJjniTHTSQRL+Adt91wej9D8e246/SbzFMqjjpTWQvVB RcR6fUwKdcMwT+RKAJybYH8z3tqwiRQXleBPs6NqcRYV7E8AHzpeMUh5v7m2Gz/P 8y+ea4v8Nglq79UnYmeqCHmvMO0ZqUpb8/+hB35xfkyuIHstx3J784NHOaQ7IZqj iJZ/Stf3De+ASddDFeAldoyc3EdT2k98gYUKpitExTOB5EbWKrK781JOEC0q+iSb +xUyux8npF6Rhhf55Rm7
=DAuM
-----END PGP SIGNATURE-----
Signature · 5 lines
Best regards,

Lukasz Majewski

Samsung R&D Poland (SRPOL) | Linux Platform Group

Tom Rini

2013-03-06 13:03 UTC #11 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 03/06/2013 02:45 AM, Pantelis Antoniou wrote:

[snip]

Quoted 7 lines

Quoted 7 lines
Sorry about missing this thread. I've been busy with other stuff.
What kind of regressions are we talking about?
I'm trying to get around to looking into all of this, but this shouldn't be causing all that trouble.

There's two problems. The first big problem is that file writes become impossible as we don't support appending to files at an offset today. The second problem is that Lukasz is seeing a regression with regards to raw block operations (I think akin to how we have to account for bad blocks, on eMMC we have to deal with lba block sizes).

- --
Tom
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iQIcBAEBAgAGBQJRNz6dAAoJENk4IS6UOR1WsuAQAKZ+tXycZbgIvwa2QnKpU11g eaZiD0JcWUbi1xa1Novy4qLxgnzz+x1cMFE2GtExWywPbrF5K+tdgvPp2H6O04aS PUfw1EdgoFMANkEHfuQx7NjDiS1ye25Ik6SFvVCRxUjprRJ7nyOxRWpzvMpLU3zb al8ByraUiGhciz73tcT2wKBQNAfuwcauW5omJgwSVdhMfxG25HD7DOv+CiBFugs1 g6Gawls5lr4e9C7dSAmI8H8TbC99YbxZ5eBYQbkYWwbH2kd6/7VYfMJQHeqAsgcu Qb/gpXt3pFHgq+zQz0k4RdbN1m1WXcp1o72PV3sKA4ElNsnUzRc1ddJ4NLL+UsfS mVVYkLIK46nnoKsOlEP/g4T0BLi0R0DNkBg4TZ/l2EDRG3MOkpT9OwCcxkYW5Rs1 3PVsajYaMOmYk4LLHK33hY6Xe0cQv8ieQ1VM45djlGdkGX5PjxRyrE82Ewm0jwyS y1B0+naXlvCe15A7eFmYZyjDN/WndRECaAWshyXjHGThXhzoCULQcSUqhItzCPWG 50zcFi4GZz2KRLHOoJkl3AwC4ItMUAIAbIeWR5k2Vh4jQYJc5oFKAQu6pQxT9NEG IQdks3nRkHylXENR5lStTm4lmI6uuSgfOnALOft5vJCDPPlY7Mkb+Jjoy2v4oFMR WpxIAgeJ/4fJrYWUo9kv
=9/Lh
-----END PGP SIGNATURE-----

Tom Rini

2013-03-01 16:46 UTC #12 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

In reply to Lukasz Majewski

Quoted 14 lines · Lukasz Majewski

This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device. Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B) In turn it displays following errors:
Error obtaining DFU functional descriptor
Warning: Assuming DFU version 1.0
Warning: Transfer size can not be detected
...
Warning: Trying default transfer size 1024

Quoted 8 lines

This fix allows passing non-standard request to function setup code, where it shall be handled.
Tested at: Trats (exynos4210)
Tested with:DFU and UMS gadgets
Signed-off-by Lukasz Majewski <[email protected]>
Signed-off-by Kyungmin Park <[email protected]>

Tested-by Tom Rini <[email protected]>

on am335x_evm with this patch replacing Pantelis' patch, in the stack of things required for DFU on this board.

Signature · 8 lines

Tom
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20130301/c3dd77b9/attachment.pgp>

Tormod Volden

2013-06-27 21:36 UTC #13 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

In reply to Lukasz Majewski

Quoted 6 lines · Lukasz Majewski

This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device.

Dear u-boot developers,

The above comment seems to imply that dfu-util is sending non-standard requests. For the record, that would be wrong. dfu-util is sending a standard "get descriptor" request. These always go to the device, not to an interface. The descriptor it is asking for is, although not mandatory per the core USB standard, a class-specific descriptor which is part of the DFU 1.0 standard. So this is all standard.

While on this topic, I would recommend that you include this descriptor in the configuration descriptor set, that is, in the complete configuration descriptor that the host OS usually requests during enumeration. Also this class-specific descriptor belongs here. In that case, the OS and libusb will keep a copy of this descriptor, and dfu-util will not have to send an explicit request (through libusb_get_descriptor()) to retrieve it.

Quoted 2 lines

Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B)

Note that this number is the "Maximum number of bytes that the device can accept per control-write transaction."

The DFU transactions should work fine if a smaller packet size is used - as long as it is larger or equal to the bMaxPacketSize0. Of course, smaller packets will cause more overhead and be less efficient, but you must make sure that your device can handle this. I am not sure from these comments whether the smaller packet size caused errors in your implementation, or if the problem was just reduced performance.

Best regards,
Tormod Volden

Lukasz Majewski

author

2013-07-01 06:28 UTC #14 lore raw

PATCH 1/1 usb:composite:fix Provide function data when addressing device with only one interface

On Thu, 27 Jun 2013 21:36:09 +0000 (UTC), Tormod Volden wrote:

Hi Tormod,

Quoted 20 lines

Quoted 8 lines · Lukasz Majewski
This commit fixes problems with some non-standard requests send with device address instead of interface address (bmRequestType.Receipent field).
This happens with dfu-util (debian version: 0.5), which address non standard requests (like w_value=0x21 and bRequest=GET_DESCRIPTOR) to device.
Dear u-boot developers,
The above comment seems to imply that dfu-util is sending non-standard requests. For the record, that would be wrong. dfu-util is sending a standard "get descriptor" request. These always go to the device, not to an interface. The descriptor it is asking for is, although not mandatory per the core USB standard, a class-specific descriptor which is part of the DFU 1.0 standard. So this is all standard.

Quoted 7 lines

While on this topic, I would recommend that you include this descriptor in the configuration descriptor set, that is, in the complete configuration descriptor that the host OS usually requests during enumeration. Also this class-specific descriptor belongs here. In that case, the OS and libusb will keep a copy of this descriptor, and dfu-util will not have to send an explicit request (through libusb_get_descriptor()) to retrieve it.

Thanks for explanation. The commit about which we are discussing was a (reasonable) fix for the problem.

I need to look into this and came up with proper solution.

Quoted 8 lines

Quoted 5 lines
Without this fix, the above request is STALLED, and hence causes dfu-util to assume some standard configuration (packet size = 1024B instead of 4096B)
Note that this number is the "Maximum number of bytes that the device can accept per control-write transaction."

I need to check this, but AFAIR there was a problem with dfu-util (and probably libusb under the hood) version 0.1 (shipped with "ancient" old stable debian) and 0.5 which is currently in use.

Quoted 8 lines

The DFU transactions should work fine if a smaller packet size is used - as long as it is larger or equal to the bMaxPacketSize0. Of course, smaller packets will cause more overhead and be less efficient, but you must make sure that your device can handle this. I am not sure from these comments whether the smaller packet size caused errors in your implementation, or if the problem was just reduced performance.

The change was driven by the need to improve performance, when downloading large files (with roofts) and to silent warning on the dfu-util side.

Quoted 9 lines

Best regards,
Tormod Volden
_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot

Signature · 5 lines

Best regards,

Lukasz Majewski

Samsung R&D Institute Poland (SRPOL) | Linux Platform Group