Maxwell Pirtle [Wed, 17 May 2023 11:46:20 +0000 (13:46 +0200)]
Add first "working" version of ODPOR
This commit introduces the first version of
ODPOR which detects a bug using the example
programs found in SimGrid's test suite! There
is still work to be done (e.g. dealing with
transitions that have multiple possible
paths of execution such as MCRandom() will
be non-trivial to resolve).
The key adjustments in this commit are
to the WakeupTree structure itself. Gone
are the days when each node in the tree
maintained a sequence of transitions to
execute; instead, each node contains only
a single transition, and its the paths between
a node and the root that describe the partial
executions that ODPOR needs to explore.
This resulted from a misconception I had about
the structure of wakeup trees. When an insertion
is made, potentially a *chain* of leaf nodes
are added into the tree. This ensures we always
have a single-process branch to pick (which
is what we expected) and greatly improves the
readability of some of the methods found in
State.cpp related to ODPOR
Maxwell Pirtle [Tue, 16 May 2023 13:07:31 +0000 (15:07 +0200)]
Add tests before changes to WakeupTree structure
Maxwell Pirtle [Tue, 16 May 2023 08:55:29 +0000 (10:55 +0200)]
Add first round of extensive docs to ODPOR methods
Maxwell Pirtle [Tue, 16 May 2023 06:48:13 +0000 (08:48 +0200)]
Add test files for WakeupTree
Maxwell Pirtle [Mon, 15 May 2023 13:31:31 +0000 (15:31 +0200)]
Add explicit ODPOR clean-up phase to DFSExplorer
ODPOR requires that we remove elements from the
wakeup tree and add elements to the sleep sets
at state in the execution only AFTER we've
completed fully the recursive exploration of the
space pointed towards by the wakeup trees; i.e.,
we only remove subtrees and add to the sleep sets
when we're walking back UP the state stack looking
for a point to recontinue the execution.
Maxwell Pirtle [Mon, 15 May 2023 08:50:25 +0000 (10:50 +0200)]
Add ODPOR "backtracking" logic
Maxwell Pirtle [Fri, 12 May 2023 11:19:50 +0000 (13:19 +0200)]
Add ODPOR extension computation (lines 4-6)
ODPOR asks us to compute, for each reversible
race in the current maximal execution `E`,
whether some subsequence `v` of events
extending a prefix `E'` of `E` should be
inserted into a wakeup tree.
There's a few subtle details that should
be noted in this computation. ODPOR asks
us to compute `notdep(e, E)`, which means
possibly looking at events that occur
*after* `e'` and THEN adding `proc(e')`.
The subtlety lies in the fact that the
actual transition associated with `e'`
must be added AFTER all of the others.
We got lucky with SDPOR since next_E_p
was already in the last position and
thus required no special treatment. With
ODPOR, we have to be more careful.
Even more subtle is the observation that
any events in `notdep(e, E)` can neither
affect the enabledness of `e'` nor can
`e'` affect the enabledness of any of
those events, or else if they do
affect the enabledness of `e'` they
will necessarily be contained "between"
`e` and `e'`
Maxwell Pirtle [Fri, 12 May 2023 08:11:33 +0000 (10:11 +0200)]
Use `std::shared_ptr<Transition>` for Execution
Prior to this commit, the `Execution` class
maintained raw pointers to the transitions
maintained by the stack used in DFSExplorer.
Instead, we opt for the execution to contain
a collection of shared pointers to facilitate
the creation of partial executions
Maxwell Pirtle [Fri, 12 May 2023 07:55:10 +0000 (09:55 +0200)]
Add ODPOR race detection phase rough-draft
ODPOR is decoposed into two parts: the "passive"
observer phase, where ODPOR selects search points
based on the work it did before to fill in its
wakeup trees, and the active race-detection phase,
where ODPOR looks at all reversible races in an
execution and performs the work to determine if
new traces need be searched. This commit adds
in an outline for the latter phase. Note that
there is still much work to be done with respect
to the race detection -- we still need to implement
the partial execution function for example.
Likewise, handling the issue of multiple execution
possibilities for a given process will be particularly
tricky.
Maxwell Pirtle [Thu, 11 May 2023 13:36:28 +0000 (15:36 +0200)]
Add boolean check for wakeup tree initialization
While not great, one way to prevent the wakeup
tree from continuing to add elements after it
has already done so when empty is to add a
boolean flag that's checked to prevent the
addition. For now, we'll live with it
Maxwell Pirtle [Thu, 11 May 2023 12:18:56 +0000 (14:18 +0200)]
Automatically remove nodes from parents
After removing a node from a WakeupTree,
we need to be careful to also have any references
to the node removed, viz. that held onto the
parent of the node in the list of children.
This commit adds a back reference to the
parent node of a given WakeupTreeNode to
allow removal from the parent when destroyed.
Additionally, some bugs were fixed for creating
subtrees and removing subtrees (we forgot to remove
the root for example)
Maxwell Pirtle [Thu, 11 May 2023 10:38:46 +0000 (12:38 +0200)]
Add tree pruning/subtree methods to State
ODPOR requires that we can remove a
subtree from a wakeup tree as we continue
the exploration. Further, we must also be
able to create a subtree from a given
wakeup tree. The functionality for those
methods has already been added; this commit
instead introduces the management of wakeup
trees specifically in the context of ODPOR
exploration.
Maxwell Pirtle [Thu, 11 May 2023 09:04:23 +0000 (11:04 +0200)]
Add skeleton for expansion phase of ODPOR
Maxwell Pirtle [Wed, 10 May 2023 13:55:18 +0000 (15:55 +0200)]
Add logic for subtree node removal
ODPOR also mandates that subtrees be removed
from a wakeup tree after having finished searching
"in the direction" of the subtree. This commit
adds a method to WakeupTree which, given a handle
to a node in the tree, removes the subtree rooted at
that node.
Maxwell Pirtle [Wed, 10 May 2023 13:03:08 +0000 (15:03 +0200)]
Finish post-order travesal with WakeupTreeIterator
The WakeupTreeIterator is now full implemented:
we now properly add nodes to the top of the
stack in reverse order, resulting in a post-order
traversal as desired.
One subtlety arose with the implementation: the
root node is not contained in the list of any
other node in the tree. To handle this issue,
a "fake" list managed by the iterator was added
into which the root is placed at construction-time.
This prevents us from needing to change any of the
logic, and we can simply treat the root as any other
node in the traversal
Maxwell Pirtle [Wed, 10 May 2023 12:34:02 +0000 (14:34 +0200)]
Add logic for subtree extraction from wakeup trees
Extracting a subtree from a wakeup tree is required
for ODPOR to continue its search and to pass information
down to the other paths that ODPOR decides to visit.
Extracting a copy of a subtree is difficult as it
requires a deep copy of the tree, making sure
that references to any nodes in the tree that is copied
are copied appropriately. One additional subtlety is that
the subtree can only be taken with respect to single-process
branches and, further, that that process is removed from
each descendant of the node that contains the single process.
Maxwell Pirtle [Wed, 10 May 2023 09:38:07 +0000 (11:38 +0200)]
Add complicated computation of v ~_E w to Execution
The relation `v ~_E w` is used when traversing
a WakeupTree to determine is an "equivalent"
execution is already noted in the tree. This commit
adds the first "working" implementation of
several components to this process, viz.
1. Assuming that iteration has been implemented
(which is to come in future commits), the WakeupTree
now implements the `insert_E(v, B)` operation, where
`B` refers to the tree instance itself.
2. The `Execution` class can now determine what
is needed to complete the `insert()` operation,
specifically determining first if a) two sequences
`v` and `w` are related by `v ~_E w` and b)
what the smallest sequence `w'` is such that
`w [= v.w'`. The latter computation uses the neat
observation that (w / v) is the smallest such
sequence (viz. removing anything in `w` that's in
`v` "one at a time") and computes this value as
it is in the process of determining whether the
`~_E` relation noted holds
Maxwell Pirtle [Wed, 10 May 2023 07:08:27 +0000 (09:08 +0200)]
Add skeleton of implementation for tree insertion
The `insert` method on a Wakeup tree is an
essential component of the ODPOR algorithm.
Tree insertion guides the ODPOR algorithm
towards the unique, unsearched traces of the
state space. Insertion is based heavily on the
correctness of computing the `~` relation over
a given execution. Subsequent commits will fill
in this method as well as several others that
will be needed to implement that functionality
Maxwell Pirtle [Tue, 9 May 2023 14:34:04 +0000 (16:34 +0200)]
Add WakeupTreeIterator and WakeupTree skeletons
Maxwell Pirtle [Tue, 16 May 2023 07:39:15 +0000 (09:39 +0200)]
Add more docmentation for get_first_sdpor_initial()
A very important function at the center of SDPOR
was missing significant context that would help
a reader understand what exactly was going on
with the function. What makes the situation tricky
is the fact that the pseudocode's implementation
is blended into several lines and computed "on the
fly" rather than fully.
Maxwell Pirtle [Mon, 15 May 2023 07:37:22 +0000 (09:37 +0200)]
Add more documentation to ClockVector/Execution
Maxwell Pirtle [Mon, 15 May 2023 07:04:11 +0000 (09:04 +0200)]
Add unit tests for ClockVector
Maxwell Pirtle [Tue, 9 May 2023 11:57:07 +0000 (13:57 +0200)]
Fix MANIFEST.in
We should note that the Execution that's a part of
both ODPOR and SDPOR is now contained in DFSExplorer
even in stateless execution. This will likely eventually
be true also for ODPOR; hence, the ODPOR sources
have been added to the SRC_MC_STATELESS portion as
needed
Maxwell Pirtle [Tue, 9 May 2023 08:36:13 +0000 (10:36 +0200)]
Integrate SDPOR into `model-check/reduction` flag
SDPOR and ODPOR can be used as part of the
DFSExplorer (ODPOR should integrate with some
minor details in the DFS implementation, but that's
for a later PR). This commit replaces the simple
`with_dpor` boolean flag used to control the
reduction with the more precise `ReductionMode`
that is newly extracted from within `DFSExplorer`
and is instead made "public"
Maxwell Pirtle [Fri, 5 May 2023 12:47:07 +0000 (14:47 +0200)]
Add more documentation to essential SDPOR methods
Maxwell Pirtle [Tue, 9 May 2023 07:47:27 +0000 (09:47 +0200)]
Modify tesh suite expectations with changes to dependencies
Maxwell Pirtle [Fri, 5 May 2023 12:21:17 +0000 (14:21 +0200)]
Fix ObjectAccess dependency check
The ObjectAccess transition should defer
dependency computation to the ActorJoin action.
Previously, it only decided dependency by
simply checking if the `other` transition
were simply another object access on the
same object
Maxwell Pirtle [Fri, 5 May 2023 09:16:07 +0000 (11:16 +0200)]
Fix subtlety with SDPOR/ODPOR initials
Computing initials for SDPOR/ODPOR requires
extending a search with respect to
a prefix of an execution. Previously, events
added to the extension were referred to by
their location within the larger execution
from which the prefix was taken. The problem
is that the new events need to be referred
to by their position relative to the
*extension* of the prefix; otherwise,
asking about "happens-before" in the extension
while using their handles in the full execution
will give incorrect results
Maxwell Pirtle [Fri, 5 May 2023 08:29:02 +0000 (10:29 +0200)]
Fix two off-by-one errors with clock vectors
Two errors were identified with managing clock
vectors in the Execution class. The index
assigned to the new transition was previously
`size() + 1` when it should have read `size()`.
Furthermore, the loop for checking "happens-between"
events excluded event 0. Both issues are now resolved
and a test with a race at the beginning was added
as verification
Maxwell Pirtle [Fri, 5 May 2023 06:53:47 +0000 (08:53 +0200)]
Mark `ActorJoin` dependent with `target_`
The `ActorJoin` transition was previously
labeled as independent with all other
transitions, including those of the actor
upon which the join acted. However, this
assumption was faulty: the join action is
instead dependent with ANY transition executed
by the actor joined upon, as the action
BECOMES enabled only as soon as the last action
run by the actor is executed.
Maxwell Pirtle [Fri, 5 May 2023 06:46:09 +0000 (08:46 +0200)]
Prevent adding outgoing transition for top-most state
The outgoing transition for the top-most
state of the state stack in the DFSExplorer refers
to that which was taken as part of the last
trace explored by the algorithm. Thus, only the
sequence of transitions leading up to, but
not including, the last state must be included
when reconstructing the Exploration for SDPOR.
Maxwell Pirtle [Thu, 4 May 2023 12:29:44 +0000 (14:29 +0200)]
Add workaround for subtlety with state regeneration
Regeneration of the Execution inside DFSExplorer
should be as simple as choosing the prefix relative
to the appropriate backtracking point. However,
it did not appear to function so simple. After
a good amount of debugging, it appears that
the stack contents can change unexpectedly so
(e.g. a transition "in the middle" of the stack
seemingly switches arbitrarily). Until we can
pinpoint the true cause here, we simple resort
to rebuilding the execution based off the new
stack each time we decide to backtrack. The
downside is that all clock vectors have to be
recomputed after each backtrack. For now,
this is something that's OK to live with,
especially considering that this certainly
won't be a bottle neck in performance
Maxwell Pirtle [Thu, 4 May 2023 07:33:25 +0000 (09:33 +0200)]
Add first tests for "happens-before"
The happens-before relation is used
extensively by SDPOR and ODPOR during
their computations for determining what
executions should be searched next. This
commit introduces the first round of tests
for the happens-before relation on
artificially constructed executions
Maxwell Pirtle [Wed, 3 May 2023 09:35:01 +0000 (11:35 +0200)]
Move SDPOR core computation into a method
The core component of SDPOR involves selecting
an "initial" for a hypothetical computation
extending from a prefix of the current execution
that results in the reversal of the reversible
race SDPOR detected. This comomit moves that
computation as much as possible into the
`Execution` class itself to make the SDPOR
implementation easier to read
Maxwell Pirtle [Wed, 3 May 2023 07:34:12 +0000 (09:34 +0200)]
Add tentatively-working SDPOR implementation
This commit introduces SDPOR into SimGrid
(well, mostly). The current implementation
will be improved upon by cleaning up some
of the code that was added to improve
its testability
Maxwell Pirtle [Tue, 2 May 2023 10:38:43 +0000 (12:38 +0200)]
Add initial outline of SDPOR implementation
The SDPOR algorithm is nearing an implementation
into SimGrid (sans a few technical details and
code clean ups that will be nice to add). The
algorithm largely builds off the existing
infrastructure of DFSExplorer fortunately.
Maxwell Pirtle [Tue, 2 May 2023 06:58:21 +0000 (08:58 +0200)]
Add Execution concept without ExecutionViews
The Execution is the key concept in ODPOR
and SDPOR for generating new traces. Executions
for the basis off of which new search paths are
created
Maxwell Pirtle [Thu, 27 Apr 2023 08:43:37 +0000 (10:43 +0200)]
Remove sdpor folder in favor of odpor only
Maxwell Pirtle [Wed, 26 Apr 2023 13:57:28 +0000 (15:57 +0200)]
Add `Execution` to represent series of transitions
The Execution class represents a series of transitions
taken by a process. An execution importantly also encodes
the happens-before relation among the series of steps
represented. The eventual implementation will leverage
clock vectors to compute the relation between events.
A more refined happens-before relation is left for
as a future (complicated) exercise...
Maxwell Pirtle [Mon, 24 Apr 2023 12:55:12 +0000 (14:55 +0200)]
Add class for eventual "happens-before" computations
Computing the "happens-before" relation over
a sequence of executions is an important step
in determining whether or not DPOR (and SDPOR
alike) needs to add a backtracking point.
This commit introduces the "ClockVector"
class which acts effectively as a light
wrapping around an `unordered_map` mapping
actor ids to integer values.
The essential component of the ClockVector
class is that its contents are "implicit"
in the sense that actors for which no
value is explicitly contained in the map
are assigned to the value `0` by default.
This allows clock vectors to be flexible
enough to support the creation of new
actors and the enabling/disabling of old
actors as they come and go.
Fred Suter [Thu, 11 May 2023 14:29:20 +0000 (10:29 -0400)]
add two more utility functions to the Operation plugin
Martin Quinson [Wed, 10 May 2023 21:35:25 +0000 (23:35 +0200)]
Fix make distcheck
Martin Quinson [Wed, 10 May 2023 21:27:34 +0000 (23:27 +0200)]
MC: Kill the now useless code State::get_recipe
Martin Quinson [Wed, 10 May 2023 21:22:28 +0000 (23:22 +0200)]
Merge branch 'master' of framagit.org:simgrid/simgrid
Martin Quinson [Wed, 10 May 2023 21:21:10 +0000 (23:21 +0200)]
DFS MC: Restore from system states even if we have to run some transitions out of it
This makes the model-check/checkpoint a bit more efficient by
increasing the amount of cases where we can reuse taken checkpoints.
It is still awfully slow because checkpoints are. The next step is to
implement model-check/checkpoint with forks.
Martin Quinson [Wed, 10 May 2023 11:08:09 +0000 (11:08 +0000)]
Merge branch 'dag-lab' into 'master'
Add dag scheduling lab to docs
See merge request simgrid/simgrid!150
Martin Quinson [Wed, 10 May 2023 11:07:36 +0000 (11:07 +0000)]
Merge branch 'changelog-operation-plugin' into 'master'
Add plugins operation and battery to changelog
See merge request simgrid/simgrid!149
Arnaud Giersch [Tue, 9 May 2023 14:46:08 +0000 (16:46 +0200)]
Use forwarding references with std::forward (sonar).
(hope I got it right this time)
Arnaud Giersch [Tue, 9 May 2023 14:27:35 +0000 (16:27 +0200)]
Use a std::vector as an underlying container for backtrack_points, so that maximal_subsets_iterator can have a noexcept move constructor by default.
Arnaud Giersch [Tue, 9 May 2023 13:57:50 +0000 (15:57 +0200)]
Use a static variable for empty_set, and allow default noexcept move constructor.
Adrien Gougeon [Tue, 9 May 2023 12:45:25 +0000 (14:45 +0200)]
add dag scheduling lab
Arnaud Giersch [Mon, 8 May 2023 19:27:14 +0000 (21:27 +0200)]
Plug memory leak.
Arnaud Giersch [Sat, 6 May 2023 11:48:33 +0000 (13:48 +0200)]
Use pointer-to-const for parameter (sonar).
Arnaud Giersch [Fri, 5 May 2023 08:38:39 +0000 (10:38 +0200)]
Fix sonar bug: attribute access on a value that can be 'None'.
Arnaud Giersch [Thu, 4 May 2023 13:13:45 +0000 (15:13 +0200)]
Misc Sonar issues.
* declare functions "const"
* use pointer-to-const or reference-to-const
* use structured bindings
* extract assignment from expression
* mark constructors "explicit"
* pass large object parameters by reference-to-const
* use default implementation constructor
* explicitly capture required variables in lambdas
Fred Suter [Fri, 5 May 2023 01:20:01 +0000 (21:20 -0400)]
mess up with operations
+ chainable setters
+ getters
+ boost instrusive pointers
Martin Quinson [Thu, 4 May 2023 14:14:54 +0000 (16:14 +0200)]
Rename Link::get_usage() to Link::get_load() for consistency with Host::
Adrien Gougeon [Thu, 4 May 2023 09:22:11 +0000 (11:22 +0200)]
add plugins operation and battery
Fred Suter [Wed, 3 May 2023 14:39:56 +0000 (10:39 -0400)]
further simplify this example
Martin Quinson [Fri, 28 Apr 2023 00:01:20 +0000 (02:01 +0200)]
One use less of get_recipe() that will soon die
Martin Quinson [Thu, 27 Apr 2023 23:25:04 +0000 (01:25 +0200)]
Factorize more code between DFSExplo and LivenessExplo, and UDPOR
Arnaud Giersch [Tue, 2 May 2023 14:32:11 +0000 (16:32 +0200)]
Kill useless empty function.
Arnaud Giersch [Tue, 2 May 2023 14:22:02 +0000 (16:22 +0200)]
Missing "override".
Arnaud Giersch [Tue, 2 May 2023 14:14:43 +0000 (16:14 +0200)]
Simplify expression.
Arnaud Giersch [Tue, 2 May 2023 14:12:18 +0000 (16:12 +0200)]
Fix use-after-free observed with the s4u-operation examples.
Arnaud Giersch [Tue, 2 May 2023 14:24:29 +0000 (16:24 +0200)]
Cosmetics (sonar).
Arnaud Giersch [Tue, 11 Apr 2023 09:13:59 +0000 (11:13 +0200)]
Define each identifier in a dedicated statement (sonar).
Arnaud Giersch [Sat, 29 Apr 2023 08:03:02 +0000 (10:03 +0200)]
Revert "Run mc-*-liveness tests serial, and hope to pass on CI."
This reverts commit
881507dc9fbeba0c63112dd0058d9c591b67419e.
In fact, it was just the CI infrastructure that was unusually slow these days.
Arnaud Giersch [Tue, 18 Apr 2023 09:39:48 +0000 (11:39 +0200)]
Whitespace cleanup (codefactor.io).
Martin Quinson [Thu, 27 Apr 2023 13:41:03 +0000 (15:41 +0200)]
Lazily compute the recipe of a state, on need only
This will be useful when leveraging intermediate states during
backtracks. In that case, we need to compute partial recipes, up to
the advanced restaure point.
Martin Quinson [Thu, 27 Apr 2023 12:09:03 +0000 (14:09 +0200)]
More automatic memory mgmt in MC
Martin Quinson [Thu, 27 Apr 2023 11:48:19 +0000 (13:48 +0200)]
MC: give each state an incoming transition too
Using the parent's outgoing transition is not reliable anymore now
that we don't always explore the tree in order, but rather following
the exploration strategies.
Martin Quinson [Tue, 18 Apr 2023 07:01:51 +0000 (09:01 +0200)]
Fix comment [no-ci]
Arnaud Giersch [Thu, 27 Apr 2023 12:57:19 +0000 (14:57 +0200)]
Run mc-*-liveness tests serial, and hope to pass on CI.
Arnaud Giersch [Thu, 27 Apr 2023 09:28:08 +0000 (11:28 +0200)]
Fix comment.
Arnaud Giersch [Thu, 27 Apr 2023 09:27:02 +0000 (11:27 +0200)]
There is no need to declare the default constructors here.
Arnaud Giersch [Thu, 27 Apr 2023 09:26:20 +0000 (11:26 +0200)]
Fix build error: exception specification of explicitly
defaulted move constructor does not match the calculated one.
Martin Quinson [Wed, 26 Apr 2023 12:45:40 +0000 (12:45 +0000)]
Merge branch 'operation-plugin' into 'master'
Operation plugin
See merge request simgrid/simgrid!145
Martin Quinson [Tue, 18 Apr 2023 07:08:17 +0000 (07:08 +0000)]
Merge branch 'udpor-phase7' into 'master'
Phase 7 of UDPOR Integration: Add specialized `ex(C)` computation for `Comm*` transitions
See merge request simgrid/simgrid!144
Martin Quinson [Mon, 17 Apr 2023 08:01:01 +0000 (10:01 +0200)]
sonar fixes
- Undefined return value (that would happen on a platform without any host)
- Local variable shadowing
- Automatic memory management
- const, explicit, etc.
Martin Quinson [Sun, 16 Apr 2023 16:45:59 +0000 (18:45 +0200)]
Let resource profiles finish their initialization by themselves and kill presolve
Martin Quinson [Sun, 16 Apr 2023 16:31:13 +0000 (18:31 +0200)]
No need to presolve the models to finish their initialization nowadays
I think that this was a leftover of the good old C times.
Martin Quinson [Sun, 16 Apr 2023 10:23:18 +0000 (12:23 +0200)]
Only fire on_platform_created after creating the platform, not after loading the deployment
Martin Quinson [Sun, 16 Apr 2023 09:03:31 +0000 (11:03 +0200)]
Add a target to recompile all ns3 tests (and only them)
Adrien Gougeon [Sun, 16 Apr 2023 08:41:16 +0000 (10:41 +0200)]
rename create to init. add init with name only. add setters
Martin Quinson [Sat, 15 Apr 2023 22:38:08 +0000 (00:38 +0200)]
Make sure that we never have a 0 transition at the end of the stack
Martin Quinson [Sat, 15 Apr 2023 21:50:48 +0000 (23:50 +0200)]
Also send the disabled transitions over: UDPOR needs it
Martin Quinson [Sat, 15 Apr 2023 21:40:11 +0000 (23:40 +0200)]
More informative message on system error
Martin Quinson [Fri, 14 Apr 2023 12:21:47 +0000 (14:21 +0200)]
More defensive programing style
Adrien Gougeon [Fri, 14 Apr 2023 10:37:41 +0000 (12:37 +0200)]
add missing '&' causing temporary value to be modified
Adrien Gougeon [Fri, 14 Apr 2023 09:24:00 +0000 (11:24 +0200)]
rename on_start/on_end functions call to on_this_start/on_this_end. add on_start_cb and on_end_cb
Martin Quinson [Thu, 13 Apr 2023 21:51:41 +0000 (23:51 +0200)]
Trust the config system to check that only valid values are used
Arnaud Giersch [Wed, 12 Apr 2023 20:02:34 +0000 (22:02 +0200)]
Disable coverage on circleci.
It seems that it is not used and it slows down the tests.
[ci-skip]
Martin Quinson [Wed, 12 Apr 2023 20:46:30 +0000 (22:46 +0200)]
Make this example valgrind-clean
Martin Quinson [Wed, 12 Apr 2023 19:37:05 +0000 (21:37 +0200)]
Further cosmetics in that example, adding a helper function to s4u on the way
Fred Suter [Wed, 12 Apr 2023 18:19:28 +0000 (14:19 -0400)]
simplify the example now that host can be set to serial
Martin Quinson [Wed, 12 Apr 2023 15:58:18 +0000 (17:58 +0200)]
Use host->set_concurrency_limit(1) in an example where it makes sense
Martin Quinson [Wed, 12 Apr 2023 15:49:30 +0000 (17:49 +0200)]
Document a bit the MC changes
Martin Quinson [Wed, 12 Apr 2023 15:43:10 +0000 (17:43 +0200)]
Allow to set a concurrency limit on disks and hosts
Arnaud Giersch [Wed, 12 Apr 2023 12:25:33 +0000 (14:25 +0200)]
Inline Exploration::system_error.