The VFS library can be used in single-threaded or multi-threaded
environments and depending on that, signals are handled by the same thread
which uses the VFS library or possibly by a different thread. If a VFS
plugin needs to block to wait for a signal, there is currently no way
which works reliably in both environments.
For this reason, this commit makes the interface of the VFS library
nonblocking, similar to the File_system session interface.
The most important changes are:
- Directories are created and opened with the 'opendir()' function and the
directory entries are read with the recently introduced 'queue_read()'
and 'complete_read()' functions.
- Symbolic links are created and opened with the 'openlink()' function and
the link target is read with the 'queue_read()' and 'complete_read()'
functions and written with the 'write()' function.
- The 'write()' function does not wait for signals anymore. This can have
the effect that data written by a VFS library user has not been
processed by a file system server yet when the library user asks for the
size of the file or closes it (both done with RPC functions at the file
system server). For this reason, a user of the VFS library should
request synchronization before calling 'stat()' or 'close()'. To make
sure that a file system server has processed all write request packets
which a client submitted before the synchronization request,
synchronization is now requested at the file system server with a
synchronization packet instead of an RPC function. Because of this
change, the synchronization interface of the VFS library is now split
into 'queue_sync()' and 'complete_sync()' functions.
Fixes#2399
The calibration of the interpolation parameters was previously only done
periodically every 500 ms. Together with the fact that the parameters
had to be stable for at least 3 calibration steps to enable
interpolation, it took at least 1.5 seconds after establishing a
connection to get microseconds-precise time values.
This is a problem for some drivers that directly start to poll time.
Thus, the timer connection now does a calibration burst as soon as it
switches to the modern mode (the mode with microseconds precision).
During this phase it does several (currently 9) calibration steps
without a delay inbetween. It is assumed that this is fast enough to not
get interrupted by scheduling. Thus, despite being small, the measured
values should be very stable which is why the burst should in most cases
be sufficient to get the interpolation initialized.
Ref #2400
When in modern mode (with local time interpolation), the timer
connection used to maximize the left shifting of its
timestamp-to-microseconds factor. The higher the shift the more precise
is the translation from timestamps to microseconds. If the timestamp
values used for determining the best shift were small - i.e. the delay
between the calibration steps were small - we may got a pretty big
shift. If we then used the shift with bigger timestamp values - i.e.
called curr_time seldom or raised calibration delays - the big shift
value became a problem. The framework had to scale down all measured
timestamps and time values temporarily to stay operative until the next
calibration step.
Thus, we now raise the shift only that much that the resulting factor
fullfills a given minimum. This keeps it as low as possible according
to the precision requirement. Currently, this requirement is set to 8
meaning that the shifted factor shall be at least 2^8 = 256.
Ref #2400
As the timer session now provides a method 'elapsed_us', there is no more need
for doing any internal calculations with values of milliseconds.
Ref #2400
As timer sessions are not expected to be microseconds precise (because
of RPC latency and scheduling), the session interface provided only a
method 'elapsed_ms' although the back end of this method in the timer
driver works with microseconds.
However, in some cases it makes sense to have a method 'elapsed_us'. The
values it returns might be milliseconds away from the "real" time but it
allows you to work with delays smaller than a millisecond without
getting a zero delta value.
This commit is motivated by the need for fast bursts of calibration
steps for the time interpolation in the new timer connection.
Ref #2400
Session_requester inherits from Dynamic_rom_session::Content_producer
which specifies the Buffer_capacity_exceeded exception which is thrown
on insufficient buffer space.
In the timeout framework, we maintain a translation factor value to
translate between time and timestamps. To raise precision we scale-up
the factor when we calculate it and scale-down the result of its
appliance later again. This up and down scaling is achieved through
left and right shifting. Until now, the shift width was statically
choosen. However, some platforms need a big shift width and others a
smaller one. The one static shift width couldn't cover all platforms
which caused overflows or precision problems.
Now, the shift width is choosen optimally for the actual translation
factor each time it gets re-calculated. This way, we can take care that
the shift always renders the best precision level without the risk for
overflows.
Ref #2400
Apparently this construct leads to a compiler errors like
error: second operand to the conditional operator is of type ‘void’, but
the third operand is neither a throw-expression nor of type ‘void’
With the capability-quota mechanism, the terminal-session won't always
be constructed completely on the first try (we may run out of caps in
the middle of the construction). Therefore, all members of the object
must be properly destructable. Furthermore, the patch replaces the
sliced heap by a heap to avoid allocating a new dataspace for each line
of the cell array.
Previously, the Genode::Timer::curr_time always used the
Timer_session::elapsed_ms RPC as back end. Now, Genode::Timer reads
this remote time only in a periodic fashion independently from the calls
to Genode::Timer::curr_time. If now one calls Genode::Timer::curr_time,
the function takes the last read remote time value and adapts it using
the timestamp difference since the remote-time read. The conversion
factor from timestamps to time is estimated on every remote-time read
using the last read remote-time value and the timestamp difference since
the last remote time read.
This commit also re-works the timeout test. The test now has two stages.
In the first stage, it tests fast polling of the
Genode::Timer::curr_time. This stage checks the error between locally
interpolated and timer-driver time as well as wether the locally
interpolated time is monotone and sufficiently homogeneous. In the
second stage several periodic and one-shot timeouts are scheduled at
once. This stage checks if the timeouts trigger sufficiently precise.
This commit adds the new Kernel::time syscall to base-hw. The syscall is
solely used by the Genode::Timer on base-hw as substitute for the
timestamp. This is because on ARM, the timestamp function uses the ARM
performance counter that stops counting when the WFI (wait for
interrupt) instruction is active. This instruction, however is used by
the base-hw idle contexts that get active when no user thread needs to
be scheduled. Thus, the ARM performance counter is not a good choice for
time interpolation and we use the kernel internal time instead.
With this commit, the timeout library becomes a basic library. That means
that it is linked against the LDSO which then provides it to the program it
serves. Furthermore, you can't use the timeout library anymore without the
LDSO because through the kernel-dependent LDSO make-files we can achieve a
kernel-dependent timeout implementation.
This commit introduces a structured Duration type that shall successively
replace the use of Microseconds, Milliseconds, and integer types for duration
values.
Open issues:
* The timeout test fails on Raspberry PI because of precision errors in the
first stage. However, this does not render the framework unusable in general
on the RPI but merely is an issue when speaking of microseconds precision.
* If we run on ARM with another Kernel than HW the timestamp speed may
continuously vary from almost 0 up to CPU speed. The Timer, however,
only uses interpolation if the timestamp speed remained stable (12.5%
tolerance) for at least 3 observation periods. Currently, one period is
100ms, so its 300ms. As long as this is not the case,
Timer_session::elapsed_ms is called instead.
Anyway, it might happen that the CPU load was stable for some time so
interpolation becomes active and now the timestamp speed drops. In the
worst case, we would now have 100ms of slowed down time. The bad thing
about it would be, that this also affects the timeout of the period.
Thus, it might "freeze" the local time for more than 100ms.
On the other hand, if the timestamp speed suddenly raises after some
stable time, interpolated time can get too fast. This would shorten the
period but nonetheless may result in drifting away into the far future.
Now we would have the problem that we can't deliver the real time
anymore until it has caught up because the output of Timer::curr_time
shall be monotone. So, effectively local time might "freeze" again for
more than 100ms.
It would be a solution to not use the Trace::timestamp on ARM w/o HW but
a function whose return value causes the Timer to never use
interpolation because of its stability policy.
Fixes#2400
This patch reduces the number of exception types by facilitating
globally defined exceptions for common usage patterns shared by most
services. In particular, RPC functions that demand a session-resource
upgrade not longer reflect this condition via a session-specific
exception but via the 'Out_of_ram' or 'Out_of_caps' types.
Furthermore, the 'Parent::Service_denied', 'Parent::Unavailable',
'Root::Invalid_args', 'Root::Unavailable', 'Service::Invalid_args',
'Service::Unavailable', and 'Local_service::Factory::Denied' types have
been replaced by the single 'Service_denied' exception type defined in
'session/session.h'.
This consolidation eases the error handling (there are fewer exceptions
to handle), alleviates the need to convert exceptions along the
session-creation call chain, and avoids possible aliasing problems
(catching the wrong type with the same name but living in a different
scope).
This patch mirrors the accounting and trading scheme that Genode employs
for physical memory to the accounting of capability allocations.
Capability quotas must now be explicitly assigned to subsystems by
specifying a 'caps=<amount>' attribute to init's start nodes.
Analogously to RAM quotas, cap quotas can be traded between clients and
servers as part of the session protocol. The capability budget of each
component is maintained by the component's corresponding PD session at
core.
At the current stage, the accounting is applied to RPC capabilities,
signal-context capabilities, and dataspace capabilities. Capabilities
that are dynamically allocated via core's CPU and TRACE service are not
yet covered. Also, the capabilities allocated by resource multiplexers
outside of core (like nitpicker) must be accounted by the respective
servers, which is not covered yet.
If a component runs out of capabilities, core's PD service prints a
warning to the log. To observe the consumption of capabilities per
component in detail, the PD service is equipped with a diagnostic
mode, which can be enabled via the 'diag' attribute in the target
node of init's routing rules. E.g., the following route enables the
diagnostic mode for the PD session of the "timer" component:
<default-route>
<service name="PD" unscoped_label="timer">
<parent diag="yes"/>
</service>
...
</default-route>
For subsystems based on a sub-init instance, init can be configured
to report the capability-quota information of its subsystems by
adding the attribute 'child_caps="yes"' to init's '<report>'
config node. Init's own capability quota can be reported by adding
the attribute 'init_caps="yes"'.
Fixes#2398
This patch reworks the implementation of core's RAM service to make use
of the 'Session_object' and to remove the distinction between the
"metadata" quota and the managed RAM quota. With the new implementation,
the session implicitly allocates its metadata from its own account. So
there is not need to handle 'Out_of_metadata' and 'Quota_exceeded' via
different exceptions. Instead, the new version solely uses the
'Out_of_ram' exception.
Furthermore, the 'Allocator::Out_of_memory' exception has become an alias
for 'Out_of_ram', which simplifies the error handling.
Issue #2398
This patch makes use of the new 'Quota_transfer::Account' by the service
types in base/service.h and uses 'Quota_transfer' objects in
base/child.cc and init/server.cc.
Furthermore, it decouples the notion of an 'Async_service' from
'Child_service'. Init's 'Routed_service' is no longer a 'Child_service'
but is based on the new 'Async_service' instead.
With this patch in place, quota transfers do no longer implicitly use
'Ram_session_client' objects. So transfers can in principle originate
from component-local 'Ram_session_component' objects, e.g., as used by
noux. Therefore, this patch removes a strumbling block for turning noux
into a single threaded component in the future.
Issue #2398
This patch replaces the 'Parent::Quota_exceeded',
'Service::Quota_exceeded', and 'Root::Quota_exceeded' exceptions
by the single 'Insufficient_ram_quota' exception type.
Furthermore, the 'Parent' interface distinguished now between
'Out_of_ram' (the child's RAM is exhausted) from
'Insufficient_ram_quota' (the child's RAM donation does not suffice to
establish the session).
This eliminates ambiguities and removes the need to convert exception
types along the path of the session creation.
Issue #2398
This patch replaces the former use of size_t with the use of the
'Ram_quota' type to improve type safety (in particular to avoid
accidentally mixing up RAM quotas with cap quotas).
Issue #2398
The 'Ram_allocator' interface contains the subset of the RAM session
interface that is needed to satisfy the needs of the 'Heap' and
'Sliced_heap'. Its small size makes it ideal for intercepting memory
allocations as done by the new 'Constrained_ram_allocator' wrapper
class, which is meant to replace the existing 'base/allocator_guard.h'
and 'os/ram_session_guard.h'.
Issue #2398
This commit moves the headers residing in `repos/base/include/spec/*/drivers`
to `repos/base/include/drivers/defs` or repos/base/include/drivers/uart`
respectively. The first one contains definitions about board-specific MMIO
iand RAM addresses, or IRQ lines. While the latter contains device driver
code for UART devices. Those definitions are used by driver implementations
in `repos/base-hw`, `repos/os`, and `repos/dde-linux`, which now need to
include them more explicitely.
This work is a step in the direction of reducing 'SPEC' identifiers overall.
Ref #2403
Since init no longer provides public headers, we have to adjust the
existing users of this headers. The 'init/child_config.h' is used only
by GDB monitor. So the patch moves the header there as an interim fix.
The 'init/child_policy.h' is still used by a few components, so we have
to keep a trimmed-down version of it for now.
This patch enables init to apply changes of any server's <provides>
declarations in a differential way. Servers can in principle be extended
by new services without re-starting them. Of course, changes of the
<provides> declarations may affect clients or would-be clients as this
information is taken into account for the session routing.
The optional 'version' attribute allows for the forced restart of a
child with an otherwise unmodified start node. The specified value is
also reflected in the state report.
This patch covers the resolution of the ROM route for child binaries
via the generic label-rewriting mechanics. Now, the <binary> node has
become merely sytactic sugar for a route like the following:
<start name="test"/>
<route>
<service name="ROM" unscoped_label="test">
<parent label="test-binary-name"/> </service>
...
</route>
...
</start>
A change of the binary name has an effect on the child's ROM route to
the binary and thereby implicitly triggers a child restart due to the
existing re-validation of the routing.
With this check in place, one can safely construct an 'Xml_generator'
even if the report is disabled. This relieves the user of the reporter
from the need to distinguish enabled from disabled reports.
This patch improves the accuracy of init's quota-saturation feature
(handing out all slack quota to a child by specifying an overly high RAM
quota for the child) and makes the RAM preserved by init configurable.
The preservation is specified as follows:
! <config>
! ...
! <resource name="RAM" preserve="1M"/>
! ...
! </config>
If not specified, init has a reasonable default of 160K (on 32 bit) and
320K (on 64 bit).
This patch lets init apply configuration changes to a running scenario
in a differential way. Children are restarted if any of their session
routes change, new children can be added to a running scenario, or
children can deliberately be removed.
Furthermore, the new version of init is able to propagate configuration
changes (modifications of <config> nodes) to its children without
restarting them.