On Sun, Mar 17, 2002 at 02:35:17PM +0900, Takuya Satoh wrote: > > On Sat, Mar 16, 2002 at 04:52:22PM +0900, Takuya Satoh wrote: > > > Hi, > > > Does the old FTOS target zero the ECN bits? I mean if I use FTOS in the > > > output chain of an ECN enabled host to set some bits except the last two > ECN > > > bits (e.g. 0x14) does it mean they will be zeroed? In my understanding > FTOS > > > comes after ECN which sets the first of the two last bits to "1" when > > > enabled. > > > > As IPTOS_TOS_MASK is 0x1e (00011110) and the ECN bits are the lsb's, it > > overwrites > > Isn't this the mask for TOS ? Mask for FTOS should be 0xff (11111111).
Yes, the FTOS target is however comparing (old_value && 0x1e) with new_value, which doesn't make sense to me. Anyway, I'll now remove the FTOS target and put ECN into CVS later. > Similarly mask for DSCP is 11111100 and mask for the future ECN is 00000011 > or am I wrong? So setting the DS values without interfering with ECN is > possible only with the DSCP and not with the TOS or FTOS targets ... The ECN Taget will not allow you to set the ECN IPv4 header bits, sorry. This would cause more problem than it's worth. The ECN target will only support a single Operation: Remove the ECN bits from TCP SYN packets. > Taka -- Live long and prosper - Harald Welte / [EMAIL PROTECTED] http://www.gnumonks.org/ ============================================================================ GCS/E/IT d- s-: a-- C+++ UL++++$ P+++ L++++$ E--- W- N++ o? K- w--- O- M+ V-- PS++ PE-- Y++ PGP++ t+ 5-- !X !R tv-- b+++ !DI !D G+ e* h--- r++ y+(*)
