Messages in this thread | | | Date | Sat, 2 May 1998 04:15:25 -0700 | From | "David S. Miller" <> | Subject | Re: 2.1.99: NULL pointer dereference in tcp_v4_rehash |
| |
Date: Sat, 2 May 1998 13:11:34 +0200 From: Steffen Zahn <zahn@berlin.snafu.de>
Since egcs-1.0.2 appears to be the default Linux compiler, at least if one looks at http://sunsite.unc.edu/pub/Linux/GCC, it may be better to find such wrong code generation problems. I have produced a combined source/disassembler of the file linux/net/ipv4/tcp_ipv4.c and placed it at http://www.snafu.de/~zahn/tcp_ipv4.txt.gz
Ok, thanks I'll look at it.
I could recompile the kernel with another compiler, but since I have no way to reproduce the bug this could only be useful if I happen to see the bug again in a kernel compiled with another compiler. In that case we would know that the bug is not compiler-related.
Yes, but the whole point of this exercise is to determine whether it is a egcs code gen. bug or a bug in the networking. There are just too many variables at the moment, I use gcc-2.7.2 on the Intel on my development machine, you've compiled the kernel with egcs-1.0.2, so the first step is to see if the compiler is the issue.
Later, David S. Miller davem@dm.cobaltmicro.com
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
| |