Re: some kernel hacking tasks
Sepherosa Ziehau <[email protected]>
| Newsgroups | gmane.os.dragonfly-bsd.kernel |
|---|---|
| Message-ID | <CAMOc5cyffgHevh5_Mi59W2LwLp5EtKQFE4P4Pw2fbpMW+ety-w@mail.gmail.com> |
On Tue, Oct 2, 2012 at 5:57 AM, Antonio Huete Jimenez <[email protected]> wrote: > Hi Sephe, > > Here's a patch for bpf: > http://island.quantumachine.net/~antonioh/archive/patches/bpf_token01.diff > Not sure if I'm missing something. I have looked at the patch. I think there are still something missing: - bpf code actually assumes that mplock is held on the device node interfaces, so you probably need to protect them w/ the bpf token. - bpfattach/bpfdetach should probably be protected by bpf token - Macros like BPF_TAP() and ETHER_BPF_MTAP() probably will need the check-hold_token-recheck sequence, e.g.: if (ifp->if_bpf) { bpf_gettoken(); if (ifp->if_bpf) bpf_tap(...); bpf_reltoken(); } Best Regards, sephe > > I've tested it with netperf in one ssh session and tcpdump in another > (thanks sjg) > > Cheers, > Antonio Huete > > 2012/9/29 Sepherosa Ziehau <[email protected]>: >> Hi all, >> >> I have several kernel hacking tasks; any helping hands are welcome. >> - Replace bpf's mplock w/ its own token >> - Support busdma(9) in some network drivers, the most commonly >> used/available ones: >> o dc(4) >> o fxp(4) >> o vr(4) >> There are actually more drivers that do not support busdma(9): >> find /usr/src/sys/dev/netif -type f -name "*.c" -exec grep -l >> "contigmalloc" "{}" ";" >> >> I could help to review, test and commit the patch. >> >> Best Regards, >> sephe >> >> -- >> Tomorrow Will Never Die -- Tomorrow Will Never Die