David, Yoshifuji-san, Herbert, I appreciate your feedback.
I made an another patch that simply replaced __kfree_skb() in exit path with kfree_skb(). I tested it overnight with a chat benchmark tool and my test program, which can reproduce the original problem. As a result, I didn't see any problem. (For example, neither oops nor memory leak happened.) I will post the patch a few moments later. Please take a look at it. Thanks, Masa David Miller wrote: > From: YOSHIFUJI Hideaki <[EMAIL PROTECTED]> > Date: Wed, 24 Jan 2007 13:37:25 +0900 (JST) > > >> In article <[EMAIL PROTECTED]> (at Wed, 24 Jan 2007 15:31:47 +1100), Herbert >> Xu <[EMAIL PROTECTED]> says: >> >> >>> Masayuki Nakagawa <[EMAIL PROTECTED]> wrote: >>> >>>> I suggest to use kfree_skb() instead of __kfree_skb(). >>>> >>> I agree. In fact please do it for all paths in that function, i.e., >>> just change __kfree_skb to kfree_skb rather than adding a special case >>> for this path. >>> >> I do think so, too. >> > > So do I, but initially I want to push his basic patch in > so that I can push the same exact thing into -stable to > fix this bug. > > So if you make the subsequent change, please make it relative > to the original patch. > > Thank you. > - > To unsubscribe from this list: send the line "unsubscribe netdev" in > the body of a message to [EMAIL PROTECTED] > More majordomo info at http://vger.kernel.org/majordomo-info.html > > - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
