A primary concern seems to be solved (4k frame size). However, it is not really a patch yet, just a dirty workaround (just replaced ETH_DATA_LEN with a value 4000, my switch is limited to 4k)
486c486 < #define NV_PKTLIMIT_1_ETH_DATA_LEN /* hard limit not known */ --- > #define NV_PKTLIMIT_1 4000 /* hard limit not known */ I will try to improve the MTU detection in driver, but it can take some time before I really understand what each function does. I do not know if this workaround forces a value to the 'initiated' phy, or just forces a wrong MTU to be used on standard nvidia cards. the results are worth considering: dd -if=//server/share of=/dev/null went from 91 to 112 mbps (and this is no file cache, the pc has 2gb ram (>50% used by GUI) and the source file is 30 ...) any help developing a patch may be useful. Is there some documentation about this driver module? -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
