[Click] e1000-5.7.6 polling driver

[Beyers Cronje]

Hi Eddie,

Eish, I should've checked the repository, could've spared me a couple
of gray hair :)

There are some differences as I've tried to use as much of Intel's
code as possible so it'll be easier for others to follow. In example:

e1000_rx_poll is roughly based on e1000_clean_rx_irq
e1000_rx_refill on e1000_alloc_rx_buffers
etc..

Attached is the diffs between your e1000_main.c and mine obtained by:

diff -Naur e1000_main.c.eddie e1000_main.c.beyers > e1000_main.diff

If you want to have a look at all the sources let me know and I'll
forward it to you.

Beyers


On Thu, 10 Mar 2005 05:06:39 -0800, Eddie Kohler <kohler at cs.ucla.edu> wrote:
> Hi Beyers,
> 
> Very cool!  So we've got a version of the 5.7.6 driver checked into the Click
> repository at this point, in drivers/e1000-5.x.  I'd be interested in the diffs
> between our versions.
> 
> Eddie
> 
> 
> Beyers Cronje wrote:
> > Hi,
> >
> > I've got a working click polling driver based on Intel's latest e1000
> > v5.7.6 driver. Anyone interested in testing/using the driver pop me an
> > email.
> >
> > Stability seems good. Current interface stats runnin in polling mode:
> > RX packets:519393113 errors:25 dropped:0 overruns:0 frame:25
> > TX shows 0 errors.
> >
> > (Not sure where the 25 frame errors are from, but minute in comparison
> > with RX packets)
> >
> > I havent done much performance testing, but should theoretically have
> > similar performance to the click supplied polling driver.
> >
> > Tested on Linux 2.4.26 with 82541GI and 82545GM controllers (both
> > controllers not supported on the supplied click driver).
> >
> > The driver should provide polling support for the following controllers:
> >
> > E1000_DEV_ID_82542               0x1000
> > E1000_DEV_ID_82543GC_FIBER       0x1001
> > E1000_DEV_ID_82543GC_COPPER      0x1004
> > E1000_DEV_ID_82544EI_COPPER      0x1008
> > E1000_DEV_ID_82544EI_FIBER       0x1009
> > E1000_DEV_ID_82544GC_COPPER      0x100C
> > E1000_DEV_ID_82544GC_LOM         0x100D
> > E1000_DEV_ID_82540EM             0x100E
> > E1000_DEV_ID_82540EM_LOM         0x1015
> > E1000_DEV_ID_82540EP_LOM         0x1016
> > E1000_DEV_ID_82540EP             0x1017
> > E1000_DEV_ID_82540EP_LP          0x101E
> > E1000_DEV_ID_82545EM_COPPER      0x100F
> > E1000_DEV_ID_82545EM_FIBER       0x1011
> > E1000_DEV_ID_82545GM_COPPER      0x1026
> > E1000_DEV_ID_82545GM_FIBER       0x1027
> > E1000_DEV_ID_82545GM_SERDES      0x1028
> > E1000_DEV_ID_82546EB_COPPER      0x1010
> > E1000_DEV_ID_82546EB_FIBER       0x1012
> > E1000_DEV_ID_82546EB_QUAD_COPPER 0x101D
> > E1000_DEV_ID_82541EI             0x1013
> > E1000_DEV_ID_82541EI_MOBILE      0x1018
> > E1000_DEV_ID_82541ER             0x1078
> > E1000_DEV_ID_82547GI             0x1075
> > E1000_DEV_ID_82541GI             0x1076
> > E1000_DEV_ID_82541GI_MOBILE      0x1077
> > E1000_DEV_ID_82541GI_LF          0x107C
> > E1000_DEV_ID_82546GB_COPPER      0x1079
> > E1000_DEV_ID_82546GB_FIBER       0x107A
> > E1000_DEV_ID_82546GB_SERDES      0x107B
> > E1000_DEV_ID_82546GB_PCIE        0x108A
> > E1000_DEV_ID_82547EI             0x1019
> >
> > Regards
> >
> > Beyers Cronje
> > _______________________________________________
> > click mailing list
> > click at amsterdam.lcs.mit.edu
> > https://amsterdam.lcs.mit.edu/mailman/listinfo/click
> 
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: e1000_main.diff
Type: text/x-patch
Size: 22439 bytes
Desc: not available
Url : https://amsterdam.lcs.mit.edu/pipermail/click/attachments/20050310/7217ce82/e1000_main-0001.bin