From: "Ilpo_Järvinen" <[EMAIL PROTECTED]> Date: Sat, 26 May 2007 11:36:01 +0300
> Previously TCP had a transitional state during which reno > counted segments that are already below the current window into > sacked_out, which is now prevented. Re-try now unconditional > S+L catching (I wonder if we could get that BUG_ON place > pinpointed more exactly in oops because now inlining makes it > lose its original context as they always seem to be in > tcp_ack, #define helps?). > > Beware, this change is not a trivial one and might have some > unexpected side-effects under obscure conditions since state > tracking is to happen much later on and the reno sack counting > was highly depending on the current state. > > This approach conservatively calls just remove_sack and leaves > reset_sack() calls alone. The best solution to the whole problem > would be to first calculate the new sacked_out fully (this patch > does not move reno_sack_reset calls from original sites and thus > does not implement this). However, that would require very > invasive change to fastretrans_alert (perhaps even slicing it to > two halves). Alternatively, all callers of tcp_packets_in_flight > (i.e., users that depend on sacked_out) should be postponed > until the new sacked_out has been calculated but it isn't any > simpler alternative. > > Signed-off-by: Ilpo Järvinen <[EMAIL PROTECTED]> So basically the idea behind this patch is to do the update of the fake RENO sakcs in clean_rtx_queue instead of fixing it up at the very last moment right before we invoke tcp_try_to_undo_partial(). I like this patch and I can't find any holes in the idea. But some things have changed in the meantime and this patch (and probably 9/9 too) don't apply cleanly. Could you respin these against current tcp-2.6 so I can apply them? Thanks. - 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
