> The good news is that this one works fine. I can't
> make it stuck the NIC. The bad news is that I did try
> to find out what change from the -8 version is
> responsible for this, but I wasn't able to. Actually,
> I think it may have been the trigger line in the
> send() function.
>
> Thanks a lot for fixing the driver.
>
>
> annis
Thank you very much for testing rge repeatedly.
When I try to integrate the fixes into opensolaris, I'd like to explain to Sun
engineers which fix did worked for your nic. So, would you help me again ?
I prepared rge-6888015-10, in which I inserted #ifdef - #endif around each
change I did. So you can easily select the changes in -9.
See Makefile.config in the tar ball. I guess one of the macros below in
Makefile.config make rge-9 work for you.
FIX_SMALL_BUFFERS
FIX_COMMAND_REG
FIX_TXDESC_ATOMIC
FIX_DMA_UNLIMIT
FIX_PCI_MEM_SPACE
If you want to choose a macro related a fix, define it with -D.
If you don't, undifine it with -U.
If you define all macros, the compiled binary will be the same with -9.
Fisrt of all, please try the binary without all macros above to ensure
the problem happen.
Then, try the binary with all macros to ensure the problem was gone.
Then, the binary with only one of the macro defined.
You can get rge-6888015-10 from here:
http://homepage2.nifty.com/mrym3/taiyodo/rge-6888015-10.tar.gz
best regards,
-masa
--
This message posted from opensolaris.org
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org
