On Thursday, May 15, 2003, at 05:30 PM, David S. Miller wrote:
I can only guess that you're getting a miscompile because of how little contextual information is included in your report.
For example, what propted you to add this cast? What debugging did you see that led you to put this cast there? Why didn't you mention these issues of "why" when presenting this patch?
Sorry for the lack of further explanation.
The patch above was the result of some tinkering to try to figure out what was going on, and since it seemed to work, I posted it as more of a hint... I agree that it doesn't make sense, so perhaps I was getting a miscompilation.
The specific tinkering I did was:
printk'd proto in pfkey_msg2xfrm_state before calling xfrm_get_type: it was 51.
printk'd proto in xfrm_get_type: it was 0.
I saw the problem on 2.5.67 and 2.5.69, and in both cases the patch fixed it.
However, I just tried doing setkey add on 2.5.69-bk10, and it worked fine,
no fix needed. So I dunno...
- : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
