if_index ( index in the kernel datastructure)

Hi, this is with reference to the free BSD networking code for ipv6. Could any body tell why in link local addresses & other link local related things they are storing if_index in the second 16bits of the address. for eg there will be statement like ia->ia_addr.sa_addr.sin_add16[1] = ifp->if_ind

IOCTL's During address formation & deletion.

Hi, I noticed that whenever a new multicast address is formed or in general any address (IPV6) is formed there is a call to ioctl through (*ifp->if_ioctl). Kindly mail me regarding the following doubts. 1) Whether IOCTL's must be done each & every time IP addresses are formed. 2) Whether IOCTL's

Struct ifaddr initialization.

Hi, Please refer if_attach() of . Iam unable to understand why you are constructing a string from interface name & interface unit number ( ifnet structure members ifp->if_unit & ifp->if_name) & storing in the data member of sockaddr_dl structure. Iam also confused why you are storing the hardwa

Struct ifnet initalization.

Hi, I have the following questions. Whether any special initialization fo the ifnet structure is required for ipv6? Some of the structures like ifaddr, sockaddr_dl are initialized in the same way as is done for ipv4. Whether this won't affect the performance? Iam unable to follow the correct co

Struct ifnet initialization.

Hi, Can any body mail me the functions that are called to initialize ifnet structure & who calls them. I know only 2 functions ether_ifattach () ifattach() regards ravi prasad __ Get your own FREE, personal Netscape Webm

Ioctl

Hi, Ihave a doubt regarding if->if_ioctl. My doubt is with reference to ipv6. Whenever a solicited node multicast address is formed it is put the multicast address list by calling in6_addmulti(). The in6_addmulti in turn calls if_addmulti() ( net/if.c ). The if addmulti after putting the addres

mbuf flag

Hi, This is with reference to the mbuf structure widely used in freeBSD networking code. When a packet is received with a multicast MAC address, the ether_input function sets the multicast flag to the mbuf. But i found that it is nowhere used in the netinet6 directory modules. Can any body inf

daemon rtsol

Hi, Sending router solicitations is implemented as a daemon. Can't this daemon be implemented as a part of the kernel itself & called by the timeout() periodically to send the solicitations. Whether such an implementation is possible? Whether such an implementation effects the performance of fre

router advertisement after router solicitation

Hi, This is regarding sending router advertisement on the receipt of router solicitation. When an icmp message of type router solicitation is received in the icmp6_input(). the function calls nd6_rs_input(). This function inturn calls nd6_cach_lladdr(). But no where in the above code rtadv rela

size of data to ip layer from tcp layer

Hi, This is regarding the size of data from TCP/UDP layer to the ip layer. The number of bytes sent from tcp layer to ip layer to send depends on the tcp MSS (maxium segment size).For an ethernet this size is 1460 bytes. Based on the above my understanding is this. a) Since MSS is 1460 bytes

mbuf organization in case 65536 bytes of data.

Hi, My doubt is how data will be organized in buffers in case we want to transmit large amount of data. My doubt is regarding the organization of mbufs in case we want to transmit the maximum ip datagram size. In the normal case data is stored in clusters for data size greater than 208 bytes.

mbuf organization

Hi, My doubt is how data will be organized in buffers in case we want to transmit large amount of data. My doubt is regarding the organization of mbufs in case we want to transmit the maximum ip datagram size. In the normal case data is stored in clusters for data size greater than 208 bytes.

buffers used in case of large amount of data ( eg 65536 bytes).

Hi, My doubt is whether freebsd uses the normal mbuf & clusters in case of large amount of data ( like jumbogram in ipv6 or the maximum ipv4 datagram size of 65536 bytes)? My understanding is that for such a large amount of data, clusters which can hold only 2048 byes are not economical. Hence

Free BSD network buffers

Hi, My doubt is whether mbuf's along with clusters will be used for large data sizes like 65536 bytes? Kindly reply. regards ravi prasad __ Get your own FREE, personal Netscape Webmail account today at http://webmail.netscape.com/

Struct ifaddr usage.

Hi, There is one ifaddr structure whose address & netmask pointers are initial;ized to two sockaddr_dl structures. One of the structures stores a string formed from interface name(ifp->if_name) & interface unit(ifp->if_unit) & the hardware address. the other sockaddr_dl structure stores a mas

rtadvd

Hi, I need the following information on the rtadvd daemon implementation. If no configuration file is created & we have given -s option with rtadvd from where will it construct the rtadvd information ( struct rainfo). regards ravi prasad

rtmsg_input

Hi, This is regarding the rtmsg_input() in the module rtadvd.c of the rtadvd daemon. I think the above function will receive message in the routing socket when a route is added or deleted. My doubt is whether this message will be generated only for addresses other than link local & multicast or

rtmsg_input function of rtadvd daemon

Hi, My understanding of this function is like this. This function will receive input only when a)the routes are added through the route6 command. b) When a prefix is added through the prefix command. c) when a new address is formed from the prefix through the prefix received through router adver

routing advertisement daemon

Hi, Can any body mail me the how a prefix entered through the prefix command is known by the routing advertisement daemon? regards ravi prasad __ Get your own FREE, personal Netscape Webmail account today at http://webmail.netscape

Display of prefixes in aperticular ifnet structure.

Hi, I configured some prefixes through the "prefix" command. I then tried to display them as itis by reading the "ifnet" structure's "if_prefixhead" member. The result is like this. Prefix was set as #prefix wb0 fec0:11::1 the prefix displayed later by reading the ifnet structure is as follows

source address

Hi, My doubt is this. In case a user process is specifying the source & destination address. Whether any check is performed on the source address like, a) address is present on which interface? b) address is already formed? c) Scope of source & destination address are correct? I checked the code

prefixes configured for an interface.

Hi, I gave two prefixes through the prefix command. My other machine formed addresses based on the prefixes. By seeing the addresses i got the following doubt. a)Whether there is any limit on the number prefixes that can be broadcast by a router on a perticular interface? b)Suppose the router

Re: source address

Hi, As you mentioned the source address selection section 1.6 in kame implementation. In this section in the last but one paragraph the following is mentioned. " there are some cases where we dodnot use the above rule. one example is connected TCP session, & we use the address kept in the TCP pr

Re: Display of prefixes in aperticular ifnet structure.

Hi, As per your advice i left the prefix command & gave prefixes in the configuration file /etc/defaults/rc.conf itself. Still my print is the one that that i have mentioned. Kindly reply. regards ravi prasad JINMEI Tatuya / <[EMAIL PROTECTED]> wrote: > > > On Sat, 02 Jun 2001 09:27:30 -04

Re: Display of prefixes in aperticular ifnet structure.

Hi, What i did to print the prefixes on a perticular ifnet structure are as follows. I entered the prefixes a) fec0:11:0:0 in the configuration file for the interface wb0. I wrote a printf in the ip6_input function. the printf is as follows printf(" %x \t", ifp->if_prefixhead.tqe_next->prefix.s6_

transition mechanism 6 to 4

Hi, I just want to check if my understanding of the transition mechanisms in free bsd 4.2 implementation is correct. Assume that a ipv6 packet has to travel through a ipv4 network to an ipv6 destination host. As usual the appropriate routing entry is found for the destination address & nd6_out

Re: Display of prefixes in aperticular ifnet structure.

Hi, Iam furnishin all the information requested by u. 1) My system version( output of uname -a) FReeBSD 4.2-RELEASE FreeBSD 4.2-RELEASE #19: Sat Jun 2 23:43:40 IST 2001 root@:/usr/src/sys/compile/PFXKERNEL i386. 2) Iam not presenly using the prefix command. Iam entering the prefixes in the conf

Automatic tunneling

Hi, My this doubt is regarding automatic tunneling. I know that freeBSd doesnot support it. But i feel u can guide me in this. Suppose iam implementing automatic tunneling then is there a need for me to use any of the gif interfaces. I feel that with the present freeBSD architecture i can impl

interface down.

Hi, This is regarding the formation of link local address in autoconfiguration. the RFc mentions the four cases under which we have to form link local addresses. ( referc section 5.2 of rfc 2462). I have checked the free BSD 4.2 code. I found that the code takes care of 1, 3 & 4th cases & to s

Prefixes for an interface

Hi, Prefixes are given on a perinterface basis. I found that the router itself forms an address with the prefix given. I have the following doubts. a) Can we disable router from forming addresses with prefixes that are given to it? I feel that prefixes are given to an interface by router so th

Route command

Hi, the route command main function opens a routing socket & writes user messages to it through the rtmsg(). My doubt is who is reading from that routing socket? Kindly educate me on this. regards ravi prasad __ Get your own FREE,

keywords.h file included in the route.c file of route command.

Hi, This with reference to the following file /usr/src/sbin/route/route.c & /usr/src/sbin/route/ directory The keywords variable is defined in route.c file as struct keytab { char *kt_cp; int kt_i; } keywords[] = {

routing table updation by the route command.

Hi, My understanding of the routing table updation by the routing command is like this. Any user requests to add or delete a route is stored in the rt_msghdr structure & written to the routing socket. The route_output() of the kernel does the updation by calling appropriate functions. It also

addresses ::1/128 and ff01::/32

Hi, The following lines are from the Kame implementation notes "Each interface joins the solicited multicast address and the link-local all-nodes multicast addresses (e.g. fe80::1:ff01:6317 and ff02::1, respectively, on the link the interface is attached). In addition to a link-local address,

why mbuf?

Hi, My question is why did you choose mbuf? is it because you are writing the code in unix systems? Is it because mbufs allow us to avoid coping as much as possible? I feel that a linear buffer might have saved some of the mbuf modifications. regards ravi prasad ___

prefixes from the rr_prefix structure.

Hi, I gave a prefix . The prefix was fec0::::. Then i tried to print the prefix from the ifprefix structure in the rr_prefix structure. It was some thing like this 00:00:00:00:00:00:fe:c0:11:11:22:22:00:00:00:00. Kindly mail me why zeros are stored at the start? Are there any other

kvm_read

Hi, I found that some of the commands read the virtual memory of the kernel. Is there a way for me to know which address of the memory is currently being accessed? regards ravi prasad __ Your favorite stores, helpful shopping tool

prefixes advertised by routers.

Hi, Assume that two prefix are advertised by router to hosts in the network. After some time the administrator deletes one of the prefixes and the future router advertisement will carry only one prefix. What will hosts do with the deleted prefix? Whether the addresses will be deleted after so

all router address "ff02::02"

Hi, I found that in the rtadvd daemon code there is a function which is called to join to the all router multicast address( ff02::02). My doubt is if rtadvd is disabled for a router then whether it will join the all routers multicast list? one reason i find for such a implementation is that jo

gctimer default value

Hi, When a link address enters the stale state you have added a nd6_gctimer ( garbage collection timer) so that the entry is removed after the address time expires in the stale state. I found that the value of this is one day. Is this the default value. Or this value can changed by the system