netdev - CDC NCM: question regarding dev->rx_urb

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <5e3d8e86e9e6cf283c67e54c86a8939b-EhVcX1dARApbRx0VCw03AF9BdAJSS1xcXV9EAF01UUhGV1kPQV5/H1ZRMl5cQk0CXF9QSFg=-webmailer2@server02.webmailer.hosteurope.de>
Date:	Sat, 26 May 2012 14:39:40 +0200
From:	wp1081184-tech <tech@...aigne.com>
To:	<netdev@...r.kernel.org>
Subject: CDC NCM: question regarding dev->rx_urb_size

 Hi,

 I am facing a strange performance problem with a device. I am using the
 "14-Mar-2012" driver found in linux 3.4.

 The device announces a dwNtbInMaxSize of 32 kB, hence causing 
 ctx->rx_max to be
 32kB and as a result dev->rx_urb_size too, since:
 dev->rx_urb_size = ctx->rx_max; // in cdc_ncm_bind()

 The performance (transfer rate) are however miserable. If I however set 
 manually
 the dev->rx_urb_size to 16kB instead of 32kB, I get very good 
 performance.

 So my questions are:

 1) what is this rx_urb_size setting for? Where is it used (in usbnet?). 
 Could
 you detail?

 2) Why changing from 32kB to 16kB could cause such dramatic performance
 improvement?

 If you need additional info, let me know.

 TIA, Loïc
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html