Commit Graph

169 Commits

Author SHA1 Message Date
Piotr Tworek
87e90d640f base-linux: Drop unused member variables
The _root, _gid and _uid private member variables are not used for
anything. GCC does not care, but clang likes to complain about things
like this.

Issue #3938
2020-11-23 12:02:59 +01:00
Piotr Tworek
db71cb8c63 base-linux: Drop ununsed Message::_addr
Not used anywhere in the current codebase. GCC does not care but clang
warns about this.

Issue #3938
2020-11-23 12:02:59 +01:00
Christian Prochaska
ff429a8056 base-linux: build 'lx_iopl()' function on x86 only
Fixes #3934
2020-11-23 12:02:58 +01:00
Christian Prochaska
eafbfb8edf base-linux: socket descriptor caps for RPC (BOARD=pc)
Fixes #3932
2020-11-23 12:02:58 +01:00
Christian Prochaska
b72503e581 base-linux: remove 'lx_ioperm()' function
Fixes #3933
2020-11-23 12:02:58 +01:00
Alexander Boettcher
afab15f1a4 linux: round up dataspace to 4k sizes
Issue #3901
2020-10-09 13:35:55 +02:00
Christian Prochaska
9b6cc75f1c tool/seccomp: allow 'restart_syscall' on x86_64
Fixes #3831
2020-07-30 08:49:23 +02:00
Alexander Boettcher
41380ff769 base: remove Cancelable_lock
- base/cancelable_lock.h becomes base/lock.h
- all members become private within base/lock.h
- solely Mutex and Blockade are friends to use base/lock.h

Fixes #3819
2020-07-30 08:49:23 +02:00
Alexander Boettcher
f3efbe50bb base: remove deprecated cancel_blocking() support
for threads.

Fixes #3806
2020-07-13 11:33:12 +02:00
Alexander Boettcher
ce6f4dffe5 linux: use Mutex
Fixes #3807
2020-07-13 11:33:12 +02:00
Norman Feske
0310c733d5 base-linux: let seccomp permit 'read'
This is needed for using the 'wait_for_continue' debug mechanism.

Fixes #3798
2020-07-03 11:58:31 +02:00
Christian Helmuth
93ab972ddc tool/seccomp: allow sigreturn on x86
Fixes #3799
2020-07-03 11:15:33 +02:00
Stefan Thöni
78497c03ca base-linux: enabled seccomp
base-linux uses seccomp to reduce the available system calls
to the minimum set needed to run base-linux. There are still
some syscalls that allow accessing global state which should
be further reduced.

The combination of seccomp and socket descriptor caps should
provide an intermediate level of security for base-linux
thereby enabling base-linux as a migration path from using
the Linux kernel to the use of microkernel-based Genode.

Fixes #3581
2020-05-27 11:56:46 +02:00
Pirmin Duss
7f6b6d7e13 core-linux: make RAM quota configurable
Add the option to configure the RAM quota for core on base-linux.
The environment variable GENODE_RAM_QUOTA if it exists is interpreted
as number of bytes to provide to the first init.

Issue #3762
2020-05-27 11:56:46 +02:00
Norman Feske
b078224753 Replace Genode::strncpy by Genode::copy_cstring
- Since Genode::strncpy is not 100% compatible with the POSIX
  strncpy function, better use a distinct name.

- Remove bogus return value from the function, easing the potential
  enforcement of mandatory return-value checks later.

Fixes #3752
2020-05-27 11:56:45 +02:00
Christian Helmuth
c11d9b7b5c Remove false API dependencies from libc/posix components
Fixes #3720
2020-05-18 10:16:13 +02:00
Norman Feske
6f6340644b Hide binary_ready_hook_for_platform from public
The hook is meant to be internal to the framework. So better keep its
declaration in the framework-internal globals.h header.

Issue #3581
2020-04-24 14:37:47 +02:00
Norman Feske
b134867f31 Remove Rpc_entrypoint::Native_context
This patch largely reverts the commit "base: lay groundwork for
base-linux caps change" because the use of 'epoll' instead of 'select'
alleviated the need to allocate large FD sets, which motivated the
introduction of the 'Native_context' hook.

Related to issue #3581
2020-04-21 16:50:37 +02:00
Norman Feske
132569d12b base-linux: socket descriptor caps for RPC
On Linux, Genode used to represent each RPC object by a socket
descriptor of the receiving thread (entrypoint) and a globally-unique
value that identifies the object. Because the latter was transferred as
plain message payload, clients had to be trusted to not forge the
values. For this reason, Linux could not be considered as a productive
Genode base platform but remained merely a development vehicle.

This patch changes the RPC mechanism such that each RPC object is
represented by a dedicated socket pair. Entrypoints wait on a set of
the local ends of the socket pairs of all RPC objects managed by the
respective entrypoint. The epoll kernel interface is used as the
underlying mechanism to wait for a set of socket descriptors at the
server side.

When delegating a capability, the remote end of the socket pair is
transferred to the recipient along with a plaintext copy of the
socket-descriptor value of the local end. The latter value serves as a
hint for re-identifiying a capability whenever it is delegated back to
its origin. Note that the client is not trusted to preserve this
information. The integrity of the hint value is protected by comparing
the inode values of incoming and already present capablities at the
originating site (whenever the capability is invoked or presented to the
owner of the RPC object).

The new mechanism effectively equips base-linux with Genode's capablity
model as described in the Chapter 3 of the Genode Foundations book.
That said, the sandboxing of components cannot be assumed at this point
because each component has still direct access to the Linux system-call
interface.

This patch is based on the extensive exploration work conducted by
Stefan Thoeni who strongly motivated the inclusion of this feature into
Genode.

Issue #3581
2020-04-17 12:40:13 +02:00
Norman Feske
7bf47b1982 base-linux: boost max open fds to the hard rlimit
This patch increases the default limit of the maximum number of open
file descriptors to the hard limit of the system. This is needed for
complex scenarios, which require more FDs than the default of 1024
at core.

Related to issue #3581

Fixes #3721
2020-04-17 12:40:12 +02:00
Alexander Boettcher
e87d60ddf7 core: use Mutex/Blockade
Issue #3612
2020-04-17 12:40:12 +02:00
Norman Feske
d8bcaa4fa4 base-linux: disable ASLR for child processes
ASLR collides with the libc's fork mechanism on 32-bit. In particular,
the tool_chain_auto.run scenario would sporadically fail while mirroring
the parent's address space.

Fixes #3710
2020-03-26 11:39:02 +01:00
Stefan Thöni
a7a9855493 base: lay groundwork for base-linux caps change
Include the necessary hooks to introduce file descriptor
based capabilities in base linux.

Issue #3581
2020-02-27 14:46:58 +01:00
Alexander Boettcher
5440cd4b50 base: use blockade in thread
Issue #3612
2020-02-20 12:11:22 +01:00
Alexander Boettcher
00f69bc70d base: add mutex as derivate of lock
The mutex class is more restrictive in usage compared to
Genode::Lock.

- At initialiation time it is ever unlocked.
- No thread is permitted to lock twice. Warn about it
  in case it happens.
- Only the lock onwer is permitted to unlock the mutex.
  Warn about it and don't unlock the mutex in case it happens.

Issue #3612
2020-02-20 12:11:22 +01:00
Christian Helmuth
4299b85cdb base-linux: support invalid dataspace for RM client
This restores behavior introduced in

  commit cb232891bf
  Author: Norman Feske <norman.feske@genode-labs.com>
  Date:   Tue May 10 11:55:25 2016 +0200

    Fix noux.run on base-linux

but removed by issue #2829.

Related to issue #1938
2019-12-19 17:00:47 +01:00
Johannes Kliemann
93639532f0 base-linux: core session support (IO_PORT, IO_MEM, IRQ) 2019-12-19 16:59:02 +01:00
Sebastian Sumpf
2fc6cedcc0 ld: load dynamic linker at static address on Linux
* move stack area to static address at beginning of binary
* set ELF type of linker from shared to executable

fixes #3479
2019-08-21 13:25:25 +02:00
Christian Helmuth
e60b597af5 linux: 32-bit compatibility with GCC versions before 7
GCC version 7 and above generate calls to __divmoddi for 64bit integer
division on 32-bit. Unfortunately, libgcc liberaries of older compilers
lack this symbol and are still in use by Debian/Ubuntu LTS at least.
2019-05-27 14:52:52 +02:00
Martin Stein
181c78d482 timeout: use uint64_t for all plain time values
This enforces the use of unsigned 64-bit values for time in the duration type,
the timeout framework, the timer session, the userland timer-drivers, and the
alarm framework on all platforms. The commit also adapts the code that uses
these tools accross all basic repositories (base, base-*, os. gems, libports,
ports, dde_*) to use unsigned 64-bit values for time as well as far as this
does not imply profound modifications.

Fixes #3208
2019-05-06 16:15:26 +02:00
Alexander Boettcher
ae16edf1d6 trace: support more facets of execution time
- execution time per thread context
- execution time per scheduling context
- quantum and priority

Issue #3192
2019-04-09 12:30:34 +02:00
Norman Feske
5ed5fddb7c base/os: remove deprecated APIs
This commit removes APIs that were previously marked as deprecated. This
change has the following implications:

- The use of the global 'env()' accessor is not possible anymore.
- Boolean accessor methods are no longer prefixed with 'is_'. E.g.,
  instead of 'is_valid()', use 'valid()'.
- The last traces of 'Ram_session' are gone now. The 'Env::ram()'
  accessor returns the 'Ram_allocator' interface, which is a subset of
  the 'Pd_session' interface.
- All connection constructors need the 'Env' as argument.
- The 'Reporter' constructor needs an 'Env' argument now because the
  reporter creates a report connection.
- The old overload 'Child_policy::resolve_session_request' that returned
  a 'Service' does not exist anymore.
- The base/printf.h header has been removed, use base/log.h instead.
- The old notion of 'Signal_dispatcher' is gone. Use 'Signal_handler'.
- Transitional headers like os/server.h, cap_session/,
  volatile_object.h, os/attached*_dataspace.h, signal_rpc_dispatcher.h
  have been removed.
- The distinction between 'Thread_state' and 'Thread_state_base' does
  not exist anymore.
- The header cpu_thread/capability.h along with the type definition of
  'Cpu_thread_capability' has been removed. Use the type
  'Thread_capability' define in cpu_session/cpu_session.h instead.
- Several XML utilities (i.e., at os/include/decorator) could be removed
  because their functionality is nowadays covered by util/xml_node.h.
- The 'os/ram_session_guard.h' has been removed.
  Use 'Constrained_ram_allocator' provided by base/ram_allocator.h instead.

Issue #1987
2019-02-26 14:44:15 +01:00
Norman Feske
b3727a9b46 Add missing override annotations
Issue #3159
2019-02-19 11:12:11 +01:00
Norman Feske
aa66b5d62f base: remove dependency from deprecated APIs
This patch adjusts the implementation of the base library and core such
that the code no longer relies on deprecated APIs except for very few
cases, mainly to keep those deprecated APIs in tact for now.

The most prominent changes are:

- Removing the use of base/printf.h

- Removing of the log backend for printf. The 'Console' with the
  format-string parser is still there along with 'snprintf.h' because
  the latter is still used at a few places, most prominently the
  'Connection' classes.

- Removing the notion of a RAM session, which does not exist in
  Genode anymore. Still the types were preserved (by typedefs to
  PD session) to keep up compatibility. But this transition should
  come to an end now.

- Slight rennovation of core's tracing service, e.g., the use of an
  Attached_dataspace as the Argument_buffer.

- Reducing the reliance on global accessors like deprecated_env() or
  core_env(). Still there is a longish way to go to eliminate all such
  calls. A useful pattern (or at least a stop-gap solution) is to
  pass the 'Env' to the individual compilation units via init functions.

- Avoiding the use of the old 'Child_policy::resolve_session_request'
  interface that returned a 'Service' instead of a 'Route'.

Issue #1987
2019-02-19 11:08:17 +01:00
Norman Feske
bcb24e316c base: fix warnings reported by LLVM analyzer
The warnings were false positives though.
2019-02-12 10:33:14 +01:00
Norman Feske
6b289a1423 base/core: use references instead of pointers
This patch replaces the former prominent use of pointers by references
wherever feasible. This has the following benefits:

* The contract between caller and callee becomes more obvious. When
  passing a reference, the contract says that the argument cannot be
  a null pointer. The caller is responsible to ensure that. Therefore,
  the use of reference eliminates the need to add defensive null-pointer
  checks at the callee site, which sometimes merely exist to be on the
  safe side. The bottom line is that the code becomes easier to follow.

* Reference members must be initialized via an object initializer,
  which promotes a programming style that avoids intermediate object-
  construction states. Within core, there are still a few pointers
  as member variables left though. E.g., caused by the late association
  of 'Platform_thread' objects with their 'Platform_pd' objects.

* If no pointers are present as member variables, we don't need to
  manually provide declarations of a private copy constructor and
  an assignment operator to avoid -Weffc++ errors "class ... has
  pointer data members [-Werror=effc++]".

This patch also changes a few system bindings on NOVA and Fiasco.OC,
e.g., the return value of the global 'cap_map' accessor has become a
reference. Hence, the patch touches a few places outside of core.

Fixes #3135
2019-02-12 10:33:13 +01:00
Norman Feske
bf62d6b896 Move timer from os to base repository
Since the timer and timeout handling is part of the base library (the
dynamic linker), it belongs to the base repository.

Besides moving the timer and its related infrastructure (alarm, timeout
libs, tests) to the base repository, this patch also moves the timer
from the 'drivers' subdirectory directly to 'src' and disamibuates the
timer's build locations for the various kernels. Otherwise the different
timer implementations could interfere with each other when using one
build directory with multiple kernels.

Note that this patch changes the include paths for the former os/timer,
os/alarm.h, os/duration.h, and os/timed_semaphore.h to base/.

Issue #3101
2019-01-14 12:33:57 +01:00
Martin Stein
a79bfad08e base-linux: init trace control 2019-01-07 12:25:46 +01:00
Norman Feske
19d7a488de init: health monitoring of child components
Fixes #3039
2018-11-27 11:36:34 +01:00
Norman Feske
f93294975b base-linux: avoid double define of __always_inline
This patch fixes the following build problem that occurs on recent
GNU/Linux systems:

  /usr/include/x86_64-linux-gnu/sys/cdefs.h:307:0: error: "__always_inline" redefined
  ...
  /usr/include/linux/stddef.h:5:0: note: this is the location of the previous definition

The code in 'cdefs.h' does not check if the '__always_inline' is already
defined. The patch works around the problem by including the 'sys/cdefs.h'
first.
2018-11-16 14:37:20 +01:00
Norman Feske
4b46abf813 base: rm first-class support for static binaries
This patch removes the detection of statically linked executables from
the base framework. It thereby fixes the corner cases encountered with
Sculpt when obtaining the binaries of the runtime from the depot_rom
service that is hosted within the runtime.

Statically linked binaries and hybrid Linux/Genode (lx_hybrid) binaries
can still be started by relabeling the ROM-session route of "ld.lib.so"
to the binary name, pretending that the binary is the dynamic linker.
This can be achieved via init's label rewriting mechanism:

  <route>
    <service name="ROM" unscoped_label="ld.lib.so">
      <parent label="test-platform"/> </service>
  </route>

However, as this is quite cryptic and would need to be applied for all
lx_hybrid components, the patch adds a shortcut to init's configuration.
One can simply add the 'ld="no"' attribute to the <start> node of the
corresponding component:

  <start name="test-platform" ld="no"/>

Fixes #2866
2018-08-02 14:36:38 +02:00
Christian Prochaska
2f7e421eed linux: destroy native CPU client after thread start
The native CPU client holds a capability reference and the reference
counter of the capability can reach its limit when many threads are
successively created and destroyed (destroyed by the Linux kernel).

Fixes #2886
2018-07-03 09:39:34 +02:00
Christian Prochaska
25b47758b7 linux: new slots for non-existent capabilities only
Prevent allocation of new cap-space slots each time a capability is
received by checking if the received capability already exists.

Fixes #2885
2018-07-03 09:39:33 +02:00
Alexander Boettcher
e6d20aba93 base: support to attach RAM dataspaces readonly
Fixes #1633
2018-05-30 13:36:27 +02:00
Norman Feske
eba9c15746 Follow practices suggested by "Effective C++"
The patch adjust the code of the base, base-<kernel>, and os repository.
To adapt existing components to fix violations of the best practices
suggested by "Effective C++" as reported by the -Weffc++ compiler
argument. The changes follow the patterns outlined below:

* A class with virtual functions can no longer publicly inherit base
  classed without a vtable. The inherited object may either be moved
  to a member variable, or inherited privately. The latter would be
  used for classes that inherit 'List::Element' or 'Avl_node'. In order
  to enable the 'List' and 'Avl_tree' to access the meta data, the
  'List' must become a friend.

* Instead of adding a virtual destructor to abstract base classes,
  we inherit the new 'Interface' class, which contains a virtual
  destructor. This way, single-line abstract base classes can stay
  as compact as they are now. The 'Interface' utility resides in
  base/include/util/interface.h.

* With the new warnings enabled, all member variables must be explicitly
  initialized. Basic types may be initialized with '='. All other types
  are initialized with braces '{ ... }' or as class initializers. If
  basic types and non-basic types appear in a row, it is nice to only
  use the brace syntax (also for basic types) and align the braces.

* If a class contains pointers as members, it must now also provide a
  copy constructor and assignment operator. In the most cases, one
  would make them private, effectively disallowing the objects to be
  copied. Unfortunately, this warning cannot be fixed be inheriting
  our existing 'Noncopyable' class (the compiler fails to detect that
  the inheriting class cannot be copied and still gives the error).
  For now, we have to manually add declarations for both the copy
  constructor and assignment operator as private class members. Those
  declarations should be prepended with a comment like this:

        /*
         * Noncopyable
         */
        Thread(Thread const &);
        Thread &operator = (Thread const &);

  In the future, we should revisit these places and try to replace
  the pointers with references. In the presence of at least one
  reference member, the compiler would no longer implicitly generate
  a copy constructor. So we could remove the manual declaration.

Issue #465
2018-01-17 12:14:35 +01:00
Johannes Schlatow
684052df5e base-linux: implement lx_sigmask/lx_sigword explicitly
In newer glibc versions, the '__sigword' and '__sigmask' macros are only
available in sigsetops.h which is not installed. These macros are not
system-specific, hence we implement them explicitly ourselves.

Fixes #2616
2017-12-21 15:01:51 +01:00
Norman Feske
7e3334ca02 core: make region-map verbosity configurable
With this patch, diagnostic messages generated by core's region-map
handling are printed only if the PD session is created with 'diag'
mode enabled.
2017-12-21 15:01:50 +01:00
Alexander Boettcher
15cc6d688f core: add support to export log output as ROM
Issue #2207
2017-12-21 15:01:46 +01:00
Alexander Boettcher
2cb635c3e3 base-*: rename core_log.cc to core_log_out.cc
Issue #2207
2017-12-21 15:01:46 +01:00
Christian Helmuth
ddfd3c0d7e linux: stack-area handling with recent Linux kernels
We moved the stack-area segment 128 MiB behind text and data to comply
with assumptions in the kernel ELF loader.

This commit also reenables static binaries on linux and removes the
unused stack_area.stdlib.ld script.

Fixes #2521
2017-10-05 17:40:00 +02:00