RFCs recommend to keep TCP connections for a certain time even after they
finished a close handshake, AFAIK, in order to be able to recognize astray
packets when they arrive later. This seems overambitious especially when in
the context of the router where session quota is pretty limited. Therefore,
this commit drops this final timeout and drops closed connections immediately.
Ref #4729
The previous value of 60 seconds was never observed in real-time scenarios and
UDP, for instance always used a timeout of 30 seconds without causing issues.
Note that this applies only to TCP connections in a state other than
ESTABLISHED, i.e., while it is still safe to early-drop the connection.
Ref #4729
The TCP connection state "ESTABLISHED" (in the router "OPEN") is a privileged
one for peers because it lasts very long without any peer interaction (in the
NIC router it's only 10 minutes, but RFCs recommend not less than 2 hours and
4 minutes). Furthermore, TCP connections in this state are normally not
available for early-drop on resource exhaustion. This means that this state
binds resources to a connection potentially for a long time without the option
of regaining them under stress. Therefore, this state should be entered with
care.
Up to now, the router marked a TCP connection with this state as soon as it had
seen one matching packet in both directions, which is rather quick. However,
implementing a very precise tracking of the exact TCP states of both peers and
only marking the connection "ESTABLISHED" when both peers are "ESTABLISHED" is
a difficult task with lots of corner cases.
That said, this commit implements a compromise. The router now has two flags
for each peer of a TCP connection - FIN sent and FIN acked - and sets them
according to the observed TCP flags. The "ESTABLISHED" state is entered only
when FIN acked is set for both peers (without having observed an RST or FIN
flag meanwhile).
Ref #4729
The Reference and Const_reference utility were introduced in order to express
that something is a reference (no null value) but can be changed dynamically
(not possible with built-in C++ references). However, the idea of preventing
every possibility for null pointer faults, with which the router was built
initially, has not prevailed and using pointers instead of the utility saves
logic and makes the code more readable to other C++ developers.
Ref #4729
The deinitialization method of Domain used to rely on Domain::with_dhcp_server
in order to dissolve and destroy a present DHCP server. However, this method
skipped calling its functor argument also when there was a DHCP server but an
invalid one. This commt replaces the with_dhcp_server with a pointer null-check
in order to fix the leak.
Ref #4729
Re-implements an emergency freeing of resources on exhaustion of session quota.
In contrast to the past one, the new algorithm is executed directly where the
exhaustion occurs. Instead of interupting the packet handling and restart it
from the beginning after the freeing action, packet handling is now continued
at the point of exhaustion (if enough resources could be freed). Furthermore,
the new algorithm frees only 100 objects (instead of 1024) at a max as we found
this to better match real-life observations. And finally, the router now drops
ICMP first, then UDP, then TCP - as this better reflects priorities - and
refrains from dropping TCP connections in the ESTABLISHED state. If the router
cannot free a sufficient amount of resources, the packet that caused the
exhaustion is dropped with a warning (verbose_packet_drop="yes").
Ref #4729
Remove the use of C++ exception as much as possible from the router as C++
exception handling can be resource intensive and can make code hard to
understand.
This also removes the garbage collection that the router used to do when a
session ran out of quota. This is motivated by the fact that the garbage
collection was rather simple and removed connection states regardless of their
current state, thereby causing broken connections. The change is part of this
commit as the approach to integrating garbage collection relied strongly on
exception handling.
The user story behind removing garbage collection: The router emergency-dropped
an established TCP connection (with NAPT) and on the next matching packet
re-created it with a different NAPT port, thereby breaking the connection. With
this commit, existing connections are prioritized over new ones during resource
exhaustion and the packets that attempt to create a new connection in such a
state are dropped with a warning in the log (verbose_packet_drop="yes").
Note that the state resolves itself with time as existing connections time out
or are closed by peers.
Ref #4729
Reserved memory regions must be excluded from the corresponding DMA
allocators irrespective of whether the device is in use. Otherwise, an
early allocation of DMA buffers may use the reserved memory regions of a
late acquired device.
Fixes#5232
Add special handling for root bridge access for
12th Gen Framework laptop and Thinkpad X1 Nano Gen2
to avoid bogus MMIO memory accesses as seen in #5234 and #4643.
Fixes#5234
Based on not emulated and therby wrong values, acpica calculates a
physical address which points to somewhere. Remove this hack from core and
try to emulate accesses inside acpica where necessary.
Issue #4643
Issue #5234
This exception got introduced as a mere convenience for implementers of
'Child_policy' but required a special case in the base library.
Following the goal of eliminating exceptions from the base system,
this patch removes it by making a server_id_space mandatory.
Issue #5245
as setup by firmware after boot. The information will/can be used by the
platform driver to re-program bridges after a system suspend/resume cycle.
Issue #5241
This patch allows the use of the 'Id_space' utility without catching
'Unknown_id' exceptions. Instead, the new 'apply' overload takes a
second functor 'missing_fn' as argument, which is called whenever the
lookup fails.
Issue #5244
- Replace 'typedef' by 'using'
- Avoid plain pointers
- Import Genode namespace into Gui::
- Use enum class
- Use Attr struct for passing multiple attributes
- Turn testnit into proper Genode::Component
- Use distinct types for top-level and child views
- Remove dependency from timer
- Use util/geometry.h
- Move header to base/include to make it applicable for base types
like 'Affinity' down the road.
- Represent 'Rect' as typle of point and area, which is the most
common form of initialization, creates in valid 'Rect' by default.
- Turn Point, Area, and Rect into compound types, making x, y, w, h, at,
area accessible without a method call
- 'Rect::Compound' function for constructing a 'Rect' from two points,
replacing a former constructor
- Use result type 'Rect::Cut_remainder' instead of out parameters.
Fixes#5239
This patch turns 'Color' from a class to a POD type by replacing
the constructors by the named create functions 'rgb', 'clamped_rgb',
and 'clamped_rgba'. It thereby enables the initialization of Color
values using { .r = ... } syntax and makes the type usable in const
expressions.
It also narrows the type for color components and alpha values to
uint8_t. So possible integer overflows of computed values are detected
by -Wconversion.
As 'Color::rgb(0, 0, 0)' is repeatedly used as a default value, the
patch adds the 'Color::black()' function.
Fixes#5238
- Move C++ sources from lib/wireguard to app/wireguard, which require
Genode include paths (that conflict with linux)
- Rename lib/wireguard_lx_inc_dirs to lib/wireguard, which builds linux
sources with linux include paths
In '_generic_syncobj_wait' check if sync-objetcs exist, return error
otherwise. Do not wait because the execution model is synchroneous,
meaning no batch buffers are in execution when this function is called
(_drm_mutex).
issue #5224
Use pthread_mutex because the EP can block while the thread executing a
batch buffer and waiting for a completion signal in the VFS-GPU plugin
can still make progress.
Also return to coarser locking until we support Sync-Object Wait
semantics.
issue #5224
Freshly allocated BOs from libdrm have to be zeroed. Since the libdrm/iris
caches VRAM allocations, we don't want to attach/detach parts of dataspaces at
every allocation. Some objects are used long (through caching in the iris
Gallium driver). Therefore, zero object on release where the region might
already be mapped.
issue #5224
This commit adds a weak dummy implementation to a compilation unit
that is referenced by all DDE Linux based driver components to
prevent adding a dummy to every driver.
The upstream implementation is used to allocate order-0 pages in
a batch and users, e.g. page-pool allocator, may rely on that
behaviour and thus it is implemented with this commit.
This commit remedies the somewhat lazy attempt to address unique names
for drivers where the name is directly derived from KBUILD_MODNAME,
e.g. rtlwifi, and also modifies the name accordingly to the rules
of Linux' build-system. The last part becomes necessary as driver
matching tables may rely on that as has already happend with the
ti-sn65dsi86 driver.
This enables the combination of the nic_uplink component with any driver
pkg featuring a runtime file. This is already the case of the pc_nic
driver.
This is a preparatory step for issue #4420.
