Jaromir Dolecek wrote:

> Lars Heidieker wrote:
> 
>>should have the FXPF_HAS_RESUME_BUG removed as this breaks at least the 
>>case PCI_PRODUCT_INTEL_PRO_100_VM_6 interface if it is a 82562EZ
>>if there is any chip that needs the pacth this case should be broken 
>>into two.
> 
> 
> Does the chip work with both things (sc_rev = 1 and FXPF_HAS_RESUME_BUG)
> removed, or is the sc_rev = 1 part still needed for VM_6 ?
> 
> Jaromir 

Actually as tried it out i only removed the section where the patch is 
applied from sys/dev/ic/i82557.c and the chip worked without a problem.
That section is triggered by the FXPF_HAS_RESUME_BUG flag.
I haven't tested the sc_rev flag but I think it should not be needed to 
be set to 1 any longer as the chip is a lot newer and should therefor 
support the hardware flow control.

Lars