rebase (was: benchmarking of user-level pcnet32 driver)

Hi, On Sun, Nov 15, 2009 at 10:52:16AM +0800, Da Zheng wrote: > However, the problem is that the code of the driver was checked in to > my Hurd git repository and there are many irrelevant commits in the > history of the branch for developing the user-level driver because I > created the branch f

Re: benchmarking of user-level pcnet32 driver

Hi, On Fri, Nov 13, 2009 at 10:57:06AM +0800, Da Zheng wrote: > Experiment 1: sending 100MB data from the host machine to the Hurd > with TCP. I only calculate data transferring rate (i.e., excluding the > size of packet headers). > in-kernel user-level averag

Re: benchmarking of user-level pcnet32 driver

Hello, Thomas Schwinge wrote: > Hello! > > On Sat, Nov 14, 2009 at 10:27:33AM +0800, Da Zheng wrote: >> I try to upload my code but I don't know how to keep the logs during the >> development. >> Is it possible to check in my code along with the developing logs? > > Sorry, I'm not sure if I'm u

Re: benchmarking of user-level pcnet32 driver

Hi, On Sat, Nov 14, 2009 at 10:27:33AM +0800, Da Zheng wrote: > Thomas Schwinge wrote: > > For new packages (like the user-level driver itself, I suppose), I > > suggest to create a suitably-named branch in the incubator > > repository, > > -

Re: benchmarking of user-level pcnet32 driver

Hello! On Sat, Nov 14, 2009 at 10:27:33AM +0800, Da Zheng wrote: > I try to upload my code but I don't know how to keep the logs during the > development. > Is it possible to check in my code along with the developing logs? Sorry, I'm not sure if I'm understanding your question correctly. Are y

Re: benchmarking of user-level pcnet32 driver

Thomas Schwinge wrote: > For modifications on the existing Hurd or GNU Mach code base, just create > a working branch for each of them. We had enough discussions over the > last weeks with suggestions about how to name and handle such branches, > so that I don't really care anymore how you're goin

Re: benchmarking of user-level pcnet32 driver

Thomas Schwinge wrote: > > For modifications on the existing Hurd or GNU Mach code base, just create > a working branch for each of them. We had enough discussions over the > last weeks with suggestions about how to name and handle such branches, > so that I don't really care anymore how you're g

Re: benchmarking of user-level pcnet32 driver

Hello! On Fri, Nov 13, 2009 at 07:58:57PM +0800, Da Zheng wrote: > Samuel Thibault wrote: > > Da Zheng, le Fri 13 Nov 2009 16:38:34 +0800, a écrit : > >> Arne Babenhauserheide wrote: > >>> Am Freitag, 13. November 2009 03:57:06 schrieb Da Zheng: > Anyway, the informal benchmarking shows us th

Re: benchmarking of user-level pcnet32 driver

Samuel Thibault wrote: > Da Zheng, le Fri 13 Nov 2009 16:38:34 +0800, a écrit : >> Arne Babenhauserheide wrote: >>> Am Freitag, 13. November 2009 03:57:06 schrieb Da Zheng: Anyway, the informal benchmarking shows us that the user-level pcnet32 driver does have relatively good performance

Re: benchmarking of user-level pcnet32 driver

Da Zheng, le Fri 13 Nov 2009 16:38:34 +0800, a écrit : > Arne Babenhauserheide wrote: > > Am Freitag, 13. November 2009 03:57:06 schrieb Da Zheng: > >> Anyway, the informal benchmarking shows us that the user-level pcnet32 > >> driver does have relatively good performance compared with the in-kern

Re: benchmarking of user-level pcnet32 driver

Arne Babenhauserheide wrote: > Am Freitag, 13. November 2009 03:57:06 schrieb Da Zheng: >> Anyway, the informal benchmarking shows us that the user-level pcnet32 >> driver does have relatively good performance compared with the in-kernel >> driver. > > That sounds great! > > Is your port alrea

Re: benchmarking of user-level pcnet32 driver

Am Freitag, 13. November 2009 03:57:06 schrieb Da Zheng: > Anyway, the informal benchmarking shows us that the user-level pcnet32 > driver does have relatively good performance compared with the in-kernel > driver. That sounds great! Is your port already publicly available somewhere? Best wi

benchmarking of user-level pcnet32 driver

Hello, I finished porting pcnet32 driver from Mach to the user space and here is the brief report of performance measurement. It was just a rough performance measurement. The Hurd runs in a VMWare virtual machine and it only receives data. The data sender is in the host machine, which runs Mac