Messages in this thread | | | Subject | Re: 2.2.0-pre1/Token-Ring/IPX = bad | Date | Fri, 1 Jan 1999 01:37:34 -0600 (CST) | From | "Gilbert Ramirez Jr." <> |
| |
As Alan Cox said: > > > Once I attempt an 'slist' with this kernel, the Linux box keeps sending NCP > > packets to my nearest file server. I sent > 80K packets in 5 seconds. > > Interesting. The NCP packets come from the user space daemon, so it could > be an NCP daemon originated problem, or maybe a driver problem. Im not sure > IPX itself can trip this. > > Do you have a box with an IBM tr card you can try to eliminate IPX. Also can > you strace the "slist" and see if strace shows it bersekely sending 80K packets
I will have to wait to go back into work to try ibmtr; I have the card but I have to physically plug it into the network.
However, I did try the "strace slist 2> out" test, since I could do that remotely.
With kernel 2.0.36, my output file was 490 lines long (mostly select() and recv()'s). With 2.2.0-pre2, the slist completed instead of hanging (perhaps because there's less network traffic at work at midnight), but my slist output is 1946 lines long, again select()'s and recv()'s. These two tests were done back-to-back, so there should be no traffic change on the network.
grep -c select output-2.0.36 = 152 select() calls grep -c select output-2.2.0-pre2 = 880 select() calls
Something is amiss. 2.2.0 requires 5 times as many select() calls to complete the same slist.
To give you a flavor of the output files, here's the last 20 lines of each file. Notice the much larger number in the left { x, y } statement in the 2.2.0 strace. Of what relevance is that number? The entire 2.2.0 strace output has that number (990000).
`tail -20 output-2.0.36` ------------------------ select(4, [3], [], [], {1, 0}) = 1 (in [3], left {1, 0}) recv(3, "33I\31\1\0\0\0\233\260\250\305\0"..., 4070, 0) = 138 write(1, "UH-DATA1 "..., 75) = 75 send(3, "\"\"J\31\1\0\27\0\t7g\22\0\32\0\4"..., 18, 0) = 18 select(4, [3], [], [], {3, 0}) = 1 (in [3], left {3, 0}) recv(3, "33J\31\1\0\374\0", 8, 0x2) = 8 select(4, [3], [], [], {1, 0}) = 1 (in [3], left {1, 0}) recv(3, "33J\31\1\0\374\0", 4070, 0) = 8 send(3, "UUK\31\1\0\0", 7, 0) = 7 select(4, [3], [], [], {3, 0}) = 1 (in [3], left {3, 0}) recv(3, "33K\31\1\0\0\0", 8, 0x2) = 8 select(4, [3], [], [], {1, 0}) = 1 (in [3], left {1, 0}) recv(3, "33K\31\1\0\0\0", 4070, 0) = 8 close(3) = 0 close(4) = 0 kill(270, SIGTERM) = 0 --- SIGCHLD (Child exited) --- wait4(-1, NULL, 0, NULL) = 270 munmap(0x400ca000, 4096) = 0 _exit(0) = ?
`tail -20 output-2.2.0-pre2` ---------------------------- send(3, "UUK\4\1\0\0", 7, 0) = 7 select(4, [3], [], [], {3, 0}) = 1 (in [3], left {2, 990000}) recv(3, "33J\4\1\0\374\0", 8, 0x2) = 8 select(4, [3], [], [], {1, 0}) = 1 (in [3], left {1, 0}) recv(3, "33J\4\1\0\374\0", 8, 0) = 8 select(4, [3], [], [], {3, 0}) = 1 (in [3], left {2, 990000}) recv(3, "33J\4\1\0\374\0", 8, 0x2) = 8 select(4, [3], [], [], {1, 0}) = 1 (in [3], left {1, 0}) recv(3, "33J\4\1\0\374\0", 8, 0) = 8 select(4, [3], [], [], {3, 0}) = 1 (in [3], left {3, 0}) recv(3, "33K\4\1\0\0\0", 8, 0x2) = 8 select(4, [3], [], [], {1, 0}) = 1 (in [3], left {1, 0}) recv(3, "33K\4\1\0\0\0", 4070, 0) = 8 close(3) = 0 close(4) = 0 kill(569, SIGTERM) = 0 --- SIGCHLD (Child exited) --- wait4(-1, NULL, 0, NULL) = 569 munmap(0x400ca000, 4096) = 0 _exit(0) = ?
I will attempt the ibmtr vs. oltr test as soon as possible.
Just to remind other readers on what's happening, my 2.2.0-pre2 machine spews NCP (IPX) packets at an incredible rate (at least 3K packets/sec) when using 'slist' with the Olicom TR driver. I don't know who is the culprit yet. 2.0.36 behaves. Unfortunately, I had not tried any 2.1.x kernels.
--gilbert
-- Gilbert Ramirez Voice: +1 210 358 4032 Technical Services Fax: +1 210 358 1122 University Health System San Antonio, Texas, USA
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |