| Age | Commit message (Collapse) | Author |
|---|
|
When a new variable used the same name as an existing symbol in an outer
scope, then offset number was defined based on a scope of the existing
symbol ($3) instead of a scope of the new symbol (sym_). That can lead
to two variables sharing the same memory slot.
|
|
Soft scopes are anonymous scopes that most likely do not contain any
symbol, so allocating regular scope is postponed when it is really
needed.
|
|
Direct recursion almost worked, just crashed on function signature check.
Split function parsing such that function signature is saved before
function body is processed. Recursive calls are marked so they can be
avoided during f_same() and similar code walking.
Also, include tower of hanoi solver as a test case.
|
|
Equality comparison is defined on all values, even of different
types, but we still want to do constant promotion if possible.
|
|
Add literal for empty set [], which works both for tree-based sets
and prefix sets by using existing constant promotion mechanism.
Minor changes by committer.
|
|
Use struct f_val as a common argument for as_path_filter(), as suggested
by Alexander Zubkov. That allows to use NULL sets as valid arguments.
|
|
All instructions with a return value (i.e. expressions, ones with
non-zero outval, third argument in INST()) should declare their return
type. Check that automatically by M4 macros.
Set outval of FI_RETURN to 0. The instruction adds one value to stack,
but syntactically it is a statement, not an expression.
Add fake return type declaration to FI_CALL, otherwise the automatic
check would fail builds.
|
|
|
|
Keep list of function parameters in f_line and use it to verify
types of arguments for function calls. Only static type checks
are implemented.
|
|
Pass instructions of function call arguments as vararg arguments to
FI_CALL instruction constructor and move necessary magic from parser
code to interpreter / instruction code.
|
|
Passing protocol to preexport was in fact a historical relic from the
old times when channels weren't a thing. Refactoring that to match
current extensibility needs.
|
|
|
|
This is needed to display every attribute only once with overlay
attribute lists recently introduced.
|
|
|
|
|
|
|
|
|
|
In the multithreaded environment, it is not supposed that anybody
traverses the routing table as the CLI show-route was doing. Now the
routing table traversal is gone and CLI won't hold the table locked
while computing filters.
|
|
|
|
Added an option for export filter to allow for prefiltering based on the
prefix. Routes outside the given prefix are completely ignored. Config
is simple:
export in <net> <filter>;
|
|
|
|
The separate import tables were too memory-greedy, there is no need for
them being stored as full-sized tables.
|
|
|
|
|
|
|
|
|
|
There were quite a lot of conflicts in flowspec validation code which
ultimately led to some code being a bit rewritten, not only adapted from
this or that branch, yet it is still in a limit of a merge.
|
|
Validation is called internally from route table at the same place where
nexthop resolution is done. Also accounting for rte->sender semantics
change (not a channel but the import hook instead).
|
|
When writing flow4 { dst 2001:db8::dead:beef/128; }, BIRD crashed on an
not-well-debuggable segfault as it tried to copy the whole 128-bit
prefix into an IPv4-sized memory.
|
|
The Babel seqno request code keeps track of which seqno requests are
outstanding for a neighbour by putting them onto a per-neighbour list. When
reusing a seqno request, it will try to remove this node, but if the seqno
request in question was a multicast request with no neighbour attached this
will result in a crash because it tries to remove a list node that wasn't
added to any list.
Fix this by making the list remove conditional. Also fix neighbor removal
which were changing seqno requests to multicast ones instead of removing
them.
Fixes: ebd5751cdeb4 ("Babel: Seqno requests are properly decoupled from
neighbors when the underlying interface disappears").
Based on the patch from Toke Høiland-Jørgensen <toke@toke.dk>,
bug reported by Stefan Haller <stefan.haller@stha.de>, thanks.
|
|
|
|
Use timer (configurable as 'gc period') to schedule routing table
GC/pruning to ensure that prune is done on time but not too often.
Randomize GC timers to avoid concentration of GC events from different
tables in one loop cycle.
Fix a bug that caused minimum inter-GC interval be 5 us instead of 5 s.
Make default 'gc period' adaptive based on number of routing tables,
from 10 s for small setups to 600 s for large ones.
In marge multi-table RS setup, the patch improved time of flushing
a downed peer from 20-30 min to <2 min and removed 40s latencies.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
