[Bug 148807] [panic] "panic: sbdrop" and "panic: sbsndptr: sockbuf _ and mbuf _ clashing" (8.1-RELEASE/10.1-STABLE/11-CURRENT)

[bugzilla-noreply at freebsd.org]

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=148807

--- Comment #35 from slw at zxy.spb.ru ---
(In reply to Hiren Panchasara from comment #34)
> which value of 'm' should I trust? is it null in frame #22 or not? it seems like null in the frames above it also.

Зartially. ether_input call with m set to 0xfffff8014eee7600 (and this is first
m for next invocation of further functions), do one (or more, w/ different m,
need access to vmcore by kgdb and analyse 0xfffff8014eee7600 for answer)
iteration w/  and call netisr_dispatch with passed m as second argument (in
%rsi register). All next invocation can don't preserve %rsi (or %rdx in case of
m passed as 3'th argument) and backtrace can incorrectly decode arguments call.

Just realyty check: frame #19, ether_input_internal (ifp=<optimized out>,
m=0x0), line 483:

        if (m->m_len < ETHER_HDR_LEN) {

MUST occur kernel panic if m realy NULL.

This is just incorrect decoding of arguments.

-- 
You are receiving this mail because:
You are the assignee for the bug.