Convert the mrtparser to use ibufs in many places.

More is possible but for now this covers most of the message parsers.
OK tb@

Adopt bgpctl code to the ibuf changes done in bgpd.

Adjust code that calls nlri_get_prefix() and friends to work with
an ibuf. show_mrt_update() is mostly converted because of this.
The output functions and the rest of the mrtparser are just
minimally touched to keep the diff reasonable.
OK tb@

KNF

Fix some spelling errors

Thanks Marco D'Itri for spotting them

OK claudio@

Use memset() instead of bzero().
OK tb@ deraadt@

Switch from u_intX_t types to stdint.h uintX_t like it was done in bgpd.
OK florian@ tb@

Adjust handling of RIB_GENERIC_ADDPATH MRT messages. Parse it the way
OpenBGPD and GoBGP dump so that it works with all the MRT implementations
out there supporting this.
While there do some additional minor cleanup.
OK deraadt@

Implemnt the RFC8050 in the MRT parser. With this MRT dumps containing
add-path information will be dumped properly.

There is one bit where this implementation is not in line with the RFC.
The encoding of DUMP_V2_RIB_GENERIC_ADDPATH for non IPv4 or IPv6 prefixes
is implemented the way gobgp did it (which seems to be the only other BGP
implementation that supports generic encoding). The RFC puts the path_id
in the NLRI for generic encoding instead of adding it to RIB entry struct
like in the IPv4 and IPv6 encoding. This is an open discussion point with
the RFC author.

OK benno@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Adopt bgpctl code to the ibuf changes done in bgpd.

Adjust code that calls nlri_get_prefix() and friends to work with
an ibuf. show_mrt_update() is mostly converted because of this.
The output functions and the rest of the mrtparser are just
minimally touched to keep the diff reasonable.
OK tb@

KNF

Fix some spelling errors

Thanks Marco D'Itri for spotting them

OK claudio@

Use memset() instead of bzero().
OK tb@ deraadt@

Switch from u_intX_t types to stdint.h uintX_t like it was done in bgpd.
OK florian@ tb@

Adjust handling of RIB_GENERIC_ADDPATH MRT messages. Parse it the way
OpenBGPD and GoBGP dump so that it works with all the MRT implementations
out there supporting this.
While there do some additional minor cleanup.
OK deraadt@

Implemnt the RFC8050 in the MRT parser. With this MRT dumps containing
add-path information will be dumped properly.

There is one bit where this implementation is not in line with the RFC.
The encoding of DUMP_V2_RIB_GENERIC_ADDPATH for non IPv4 or IPv6 prefixes
is implemented the way gobgp did it (which seems to be the only other BGP
implementation that supports generic encoding). The RFC puts the path_id
in the NLRI for generic encoding instead of adding it to RIB entry struct
like in the IPv4 and IPv6 encoding. This is an open discussion point with
the RFC author.

OK benno@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

KNF

Fix some spelling errors

Thanks Marco D'Itri for spotting them

OK claudio@

Use memset() instead of bzero().
OK tb@ deraadt@

Switch from u_intX_t types to stdint.h uintX_t like it was done in bgpd.
OK florian@ tb@

Adjust handling of RIB_GENERIC_ADDPATH MRT messages. Parse it the way
OpenBGPD and GoBGP dump so that it works with all the MRT implementations
out there supporting this.
While there do some additional minor cleanup.
OK deraadt@

Implemnt the RFC8050 in the MRT parser. With this MRT dumps containing
add-path information will be dumped properly.

There is one bit where this implementation is not in line with the RFC.
The encoding of DUMP_V2_RIB_GENERIC_ADDPATH for non IPv4 or IPv6 prefixes
is implemented the way gobgp did it (which seems to be the only other BGP
implementation that supports generic encoding). The RFC puts the path_id
in the NLRI for generic encoding instead of adding it to RIB entry struct
like in the IPv4 and IPv6 encoding. This is an open discussion point with
the RFC author.

OK benno@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Fix some spelling errors

Thanks Marco D'Itri for spotting them

OK claudio@

Use memset() instead of bzero().
OK tb@ deraadt@

Switch from u_intX_t types to stdint.h uintX_t like it was done in bgpd.
OK florian@ tb@

Adjust handling of RIB_GENERIC_ADDPATH MRT messages. Parse it the way
OpenBGPD and GoBGP dump so that it works with all the MRT implementations
out there supporting this.
While there do some additional minor cleanup.
OK deraadt@

Implemnt the RFC8050 in the MRT parser. With this MRT dumps containing
add-path information will be dumped properly.

There is one bit where this implementation is not in line with the RFC.
The encoding of DUMP_V2_RIB_GENERIC_ADDPATH for non IPv4 or IPv6 prefixes
is implemented the way gobgp did it (which seems to be the only other BGP
implementation that supports generic encoding). The RFC puts the path_id
in the NLRI for generic encoding instead of adding it to RIB entry struct
like in the IPv4 and IPv6 encoding. This is an open discussion point with
the RFC author.

OK benno@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Use memset() instead of bzero().
OK tb@ deraadt@

Switch from u_intX_t types to stdint.h uintX_t like it was done in bgpd.
OK florian@ tb@

Adjust handling of RIB_GENERIC_ADDPATH MRT messages. Parse it the way
OpenBGPD and GoBGP dump so that it works with all the MRT implementations
out there supporting this.
While there do some additional minor cleanup.
OK deraadt@

Implemnt the RFC8050 in the MRT parser. With this MRT dumps containing
add-path information will be dumped properly.

There is one bit where this implementation is not in line with the RFC.
The encoding of DUMP_V2_RIB_GENERIC_ADDPATH for non IPv4 or IPv6 prefixes
is implemented the way gobgp did it (which seems to be the only other BGP
implementation that supports generic encoding). The RFC puts the path_id
in the NLRI for generic encoding instead of adding it to RIB entry struct
like in the IPv4 and IPv6 encoding. This is an open discussion point with
the RFC author.

OK benno@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Switch from u_intX_t types to stdint.h uintX_t like it was done in bgpd.
OK florian@ tb@

Adjust handling of RIB_GENERIC_ADDPATH MRT messages. Parse it the way
OpenBGPD and GoBGP dump so that it works with all the MRT implementations
out there supporting this.
While there do some additional minor cleanup.
OK deraadt@

Implemnt the RFC8050 in the MRT parser. With this MRT dumps containing
add-path information will be dumped properly.

There is one bit where this implementation is not in line with the RFC.
The encoding of DUMP_V2_RIB_GENERIC_ADDPATH for non IPv4 or IPv6 prefixes
is implemented the way gobgp did it (which seems to be the only other BGP
implementation that supports generic encoding). The RFC puts the path_id
in the NLRI for generic encoding instead of adding it to RIB entry struct
like in the IPv4 and IPv6 encoding. This is an open discussion point with
the RFC author.

OK benno@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Adjust handling of RIB_GENERIC_ADDPATH MRT messages. Parse it the way
OpenBGPD and GoBGP dump so that it works with all the MRT implementations
out there supporting this.
While there do some additional minor cleanup.
OK deraadt@

Implemnt the RFC8050 in the MRT parser. With this MRT dumps containing
add-path information will be dumped properly.

There is one bit where this implementation is not in line with the RFC.
The encoding of DUMP_V2_RIB_GENERIC_ADDPATH for non IPv4 or IPv6 prefixes
is implemented the way gobgp did it (which seems to be the only other BGP
implementation that supports generic encoding). The RFC puts the path_id
in the NLRI for generic encoding instead of adding it to RIB entry struct
like in the IPv4 and IPv6 encoding. This is an open discussion point with
the RFC author.

OK benno@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Implemnt the RFC8050 in the MRT parser. With this MRT dumps containing
add-path information will be dumped properly.

There is one bit where this implementation is not in line with the RFC.
The encoding of DUMP_V2_RIB_GENERIC_ADDPATH for non IPv4 or IPv6 prefixes
is implemented the way gobgp did it (which seems to be the only other BGP
implementation that supports generic encoding). The RFC puts the path_id
in the NLRI for generic encoding instead of adding it to RIB entry struct
like in the IPv4 and IPv6 encoding. This is an open discussion point with
the RFC author.

OK benno@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Adjust code to the change of struct bgpd_addr in bgpd.
OK denis@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Add a default case to subtype switch in the table v2 parser.
If a unknown subtype is encountered it is better to fail.

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Implement the bits and pieces that are needed to dump MRT_DUMP_V2_RIB_GENERIC
messages and BGP MPLS VPNs.
OK benno@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Cleanup some spaces here as well. Again no binary change.

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Use struct bgpd_addr instead of a union of sockaddrs to store addresses
in the mrt dump structs. This is more compatible with bgpctl since it uses
struct bgpd_addr in most cases anyway.
OK kn@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Print the timestamp embedded in mrt files for update and status messages.
Also implemented the extended precision format so microseconds are printed
as well when available. The output is relative to the previous message and
follows what kdump does.
OK benno@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@

Remove NULL-checks before free(). I think I've almost scrubbed the
entire tree of these...

Implement the missing bits to parse the other MRT message types.
Printing bgp messages is still missing lots but at least it is a start.
I onced abused tcpdump's bgp protocol handler for this but that is an ugly hack.

move the ignore-microsec-timestamp bit from mrt_parse to mrt_parse_dump_mp
fixes a coredump reachable via show mrt file
From: Benjamin Baier <programmer at netzbasis dot de>, ok claudio

Use reallocarray() throughout to spot multiplicative int overflow.
ok henning benno

fix some leaks in error paths
ok claudio@

Add a function to extract an address instead of copying the same code over
and over again. Also allow the dump attr of struct mrt_parser to be NULL.

Implement a better way to detect the encoding of the MP_REACH_NLRI attr.
Check the addr length value of the RFC 6396 encoding to decide if it is
one or not. While there try to make the comment more clear.

Make it possible to parse MRT table dumps (all 3 formats) and display
them like the show rib / show rib detail output. It is also possible
to filter the output. e.g.
bgpctl show mrt file ./bview.20110914.1600 as 22512 204.209.0.0/16 all
OK sthen@, put it in henning@