| 1 |
On 4/20/07, Timothy Redaelli <drizzt@g.o> wrote: |
| 2 |
> Can you try to recompile the kernel with gcc 4 but with -O0? |
| 3 |
> cd /usr/src/sys/i386/compile/GENERIC ; COPTFLAGS=-O0 make |
| 4 |
|
| 5 |
I can confirm that compiling with gcc-4 and COPTFLAGS=-O0 produces a |
| 6 |
working kernel (i.e. no malformed tcp packets). This was after I |
| 7 |
found that the gcc41 patch is also not causing the problem. After |
| 8 |
getting a working kernel with gcc-4 I went back and used the default |
| 9 |
COPTFLAGS again and it broke again. So I think this gives us a fair |
| 10 |
amount of certainty that the problem is caused by gcc-4 optimizations. |
| 11 |
|
| 12 |
As to the reason why some FreeBSD users experience this and other do |
| 13 |
not, I am not sure. The symptoms (broken TCP traffic to some sites) |
| 14 |
depend entirely upon the sites you are connecting to and how your |
| 15 |
connections are routed. So it is possible that somebody could have |
| 16 |
this problem by not realize it. |
| 17 |
|
| 18 |
However, using tcpdump I can tell that some machines apparently do not |
| 19 |
experience this problem, so it may be processor-specific (which makes |
| 20 |
sense given that it is an optimization problem). I use a Pentium M. |
| 21 |
What are other people using that have been affected by the problem? |
| 22 |
|
| 23 |
-- |
| 24 |
Nathan Smith |
| 25 |
ndansmith@×××××.com |
| 26 |
-- |
| 27 |
gentoo-bsd@g.o mailing list |
Archives