The DP83820/83821 has an undocumented limitation concerning jumbo frames
and TX checksum offload. In order for TX checksum offload to work, the
outgoing frame must fit entirely within the TX FIFO, which is 8192 bytes
in size. This isn't a problem, until you try to send a 9000-byte frame,
at which point the TX DMA engine goes to sleep. It turns out that if
you want to send a jumbo frame larger than 8170 bytes (8192 - 64), you
have to turn off the TX checksum support.
As a workaround, I changed nge_ioctl() so that if the user selects an
MTU larger than 8152 bytes, we clear the if_hwassist flags. The flags
will be set again once the MTU is reduced to a smaller value.
same as it did before the IPv6 commit. Probably irrelevent to freebsd,
but I think it was useful on other OS's. Also makes a few minor style chgs.
Reviewed by: freebsd-print@bostonradio.org
MFC after: 1 week
more sensible/understandable. 'from'->'from_host' 'host'->'local_host'
'fromb'->'frombuf' 'fromhost'->'origin_host' and a local-variable
named 'host'->'hostbuf'. This fixes some compile-time warnings about
local variables shadowing global variables.
Other than renaming variables, the only actual code changes are to call
strlcpy() instead of strncpy() when setting those (renamed) variables,
and that 'from_ip' is now a strdup()-created buffer instead of being a
static buffer compiled in as 1025 bytes.
Reviewed by: freebsd-print@bostonradio.org (an earlier version)
MFC after: 1 week
- Use db_printf() instead of printf().
- Clean up decode_syscall() to use regular if-then-else rather than goto's.
- Use the same method of parsing PID's for per-process traces as the x86
code does: that is, if the address passed in is not a valid kernel
address, treat it is a decimal pid.
- If the pid of the current process is specified, fall back to using the
"default" parameters for the trace as curproc's pcb is not valid at this
point.
MFC after: 1 week
we want the checksums calculated on a per-packet basis using control bits
in the extsts field of the DMA descriptor structure. For TX, the chip
seems to want these bits set in the field of the first descriptor in
a fragment chain, not the last.
Removed the ambiguity in -s, -f, -p and -i flags handling.
Basically, there are four displays (except others):
1. PCB display.
2. Protocol statistics display. (-s)
3. Interface statistics display. (-i)
4. Per-interface protocol statistics display. (-i -s)
All of the above except 3) can be limited to a particular
protocol family (-f) or a single protocol (-p).
Some examples:
1. netstat -f inet -- show PCBs of all INET protocols
2. netstat -p udp -- show PCB of UDP protocol only (NEW!)
3. netstat -s -- show protocol statistics for all families
4. netstat -s -f inet -- show INET protocols statistics
5. netstat -s -p icmp -- show ICMP protocol statistics
This is a work in progress. Manpage has been fixed slightly,
but is still incomplete.
warnings which come up for various routines that have a parameter which
is also called 'name'.
Reviewed by: freebsd-print@bostonradio.org
MFC after: 1 week
possible when writing:
[EFBIG] An attempt was made to write a file that exceeds
the process's file size limit or the maximum file
size.
[EPERM] An append-only flag is set on the file, but the
caller is attempting to write before the current
end of file.
announcement from JKH (back when he was still working for Lotus)
http://minnie.tuhs.org/cgi-bin/newsread?23150
This also means that if, as Microsoft say, Linux is a Cancer, FreeBSD is
a Scorpio.
