Commit Graph

392 Commits

Author SHA1 Message Date
hayodea 2be78401b5 Center logo 2025-09-28 15:20:09 -04:00
hayodea eca3f47884 Add 512 and 256p logos 2025-09-28 15:15:40 -04:00
hayodea f3909fc000 New salmanoff logo 2025-09-28 15:09:23 -04:00
hayodea c0798d1bdb DevMgr: detachAll only detaches from attachedDeviceRoles
We no longer try to detach from the collection of specs. We
detach from the collection of attachedDeviceRoles. This means
our cleanup sequence no longer tries to clean up things that were
never set up to begin with.
2025-09-28 13:04:07 -04:00
hayodea e45a9ee5d1 DAP.yy: use cmdlineDASpecs; DevMgr: add attachAllUnattachedDevicesFromCmdlineReq
This method wraps around attachAllUnattachedDevicesFromReq and supplies
it with a sh_ptr<> collection of all DASpecs parsed by the DAP parser
from the cmdline.

The initialization sequence now correctly initializes all DAP specs
given on the cmdline again.
2025-09-28 12:52:59 -04:00
hayodea b43ffcb677 DevMgr: attachAllUnattachedDevsFrom: now takes sh_ptr<vector<Spec>>
This method now accepts a sh_ptr<vector<DeviceAttachmentSpec>> to
tell it specifically which specs to attempt to attach.

This enables us to implement different frontends that supply it
with collections of devices from different sources (GUI, cmdline,
previously failed-to-attach/hot-removed devices, etc).

SMO temporarily initializes none of the devices from the cmdline
during this commit as we transition to implementing the cmdline
collection frontend.
2025-09-28 12:39:45 -04:00
hayodea 2c60248127 DevMgr: Rename at/detachAll*Req():
We've renamed these now to better reflect what they do.
* attachAllSenseDevicesFromSpecsReq=>attachAllUnattachedDevicesFromReq
* detachAllSenseDevicesReq=>detachAllAttachedDeviceRoles

This is also the first step in changing
attachAllUnattachedDevicesFrom to accept a sh_ptr<> to a collection
of DeviceAttachmentSpecs. This will enable us to unify the underlying
spec attachment logic and just create several front-ends for attaching
specs from multiple sources.
2025-09-28 12:19:56 -04:00
hayodea 993bf568fc DevMgr: implement removeDeviceAttachmentSpecReq
This reverts all state changes made by newDeviceAttachmentSpecInd.
2025-09-28 11:41:20 -04:00
hayodea 27e707a22d DevMgr:newDevAttSpecInd: take ref & not sh_ptr to DASpec
Because NewDevAttSpecInd should be the function that creates the
persistent sh_ptr state within DevMgr.
2025-09-28 11:14:55 -04:00
hayodea 7ab6e7b2c3 Formatting 2025-09-28 11:05:04 -04:00
hayodea 5b2354bfe0 DevMgr:newDevAttSpecInd: creates Device in frontend
There's no reason why we have to only create the Device at the
end if everything succeeds. The Device isn't the same thing as the
DeviceRole.
2025-09-28 10:57:43 -04:00
hayodea 42ab935da6 Reorder operations 2025-09-28 09:19:09 -04:00
hayodea 572a8612ed DevMgr,Dev: Add Qutexes
Add Qutex locks to both DeviceManager and Device. These will be
properly used in the upcoming patches.
2025-09-28 01:27:32 -04:00
hayodea 51b70b179c DevMgr: call newDevAttSpecInd & not attDevReq in body:initReq
This performs a more complete device initialization and attachment
sequence. We'll do the corresponding teardown in the shutdown
sequence later.

We might probably do it as deviceRoleGoneAwayInd()
2025-09-28 01:15:36 -04:00
hayodea 1a56e2a107 DevMgr: Add DeviceRoles; attachedDevices unrelated to device state now
We've decided to add a separate notion of a DeviceRole to track attached
device roles now. We no longer use the collection of deviceSpecs to
track which roles have been attached. Rather, this list will simply
collate all known deviceAttachment specs which are expected to be
maintained in an attached state.

SMO can periodically scan through these and cross-reference this
collection with the collection of attachedDeviceRoles. Then it can
re-try to attach those which aren't currently attached at any given
moment. This will give resilience against device attachment failures
or device resets/malfunctions, at runtime.
2025-09-28 00:50:05 -04:00
hayodea e6b8d3e85d DevMgr: Move at/detachSenseDevs[FromSpecs] into DeviceMgr::
This is logically cleaner and it begins preparing our next set
of restructuring changes. To wit: we're revamping the device
manager to distinguish between devices and their roles.
2025-09-27 23:16:46 -04:00
hayodea 52567406ca Lockset: check for registration b4 unregistering 2025-09-27 22:22:35 -04:00
hayodea 2f18ade4ab Locking: pre-emptively detect deadlocks in Lockvoker constructor
We can now check for deadlocks both pre-emptively and reactively.
We can eventually add function vaddrs and even function names to
our tracing code.
2025-09-27 21:27:04 -04:00
hayodea f8bf8083af Locking: Add contin tracing to detect deadlocks
We added the code to trace all the contins linked to a particular
Lockvoker, into SerializedAsyncContinuation. This basically
ensures that we'll almost never deal with a deadlock. So cool.
2025-09-27 20:51:20 -04:00
hayodea 782bcd4567 Async: add sh_ptr<ContinuationChainLink> to Callback<>
This change enables us to finally implement the tracing of
continuations backward from the point of acquisition for deadlock
debugging.
2025-09-27 18:30:09 -04:00
hayodea 2212aec080 VScode stuff that didn't work to make it use /b and not /build 2025-09-22 21:32:17 -04:00
hayodea d0303becd7 Async: Add a new Callback class for accessing caller's contin
We previously passed a sh_ptr to the caller's contin as arguments
to the std::bind() callable. In order for us to be able to trace
deadlocks, we need to be able to access them explicitly.

So here's that change.
2025-09-22 21:30:33 -04:00
hayodea 27bebeb702 Squash: Into debug locks commit 2025-09-22 21:30:14 -04:00
hayodea 09a0041f20 Squash: into debug locks intro commit 2025-09-22 21:29:43 -04:00
hayodea c6c3d6c9e8 Todo: Update 2025-09-22 20:49:00 -04:00
hayodea 092a0954a0 Locking: Add basic reactive deadlock detection foundation
We added a timestamp to each Lockvoker so that we can detect when
a lockvoker has been in a qutex for "too long", where "too long"
is defined arbitrarily as 500ms.

Next we're going to change the way we create callbacks to enable
us to more explicitly access the sh_ptr<AsyncContin> via
the callback object.
2025-09-22 20:45:36 -04:00
hayodea d2ed525106 Debug:Qutex: Add deadlock detection based on elapsed time
We now detect that a deadlock is likely when
CONFIG_DEBUG_QUTEX_DEADLOCK_TIMEOUT_MS has elapsed. This is the
preliminary work required to do a backtrace through the call
stack and figure out if a deadlock has really occured.

To do this, we'd have to go through the async call chain and
search for a previous caller which acquired the same qutex as
the one that first failed during this Lockvoker LockSet acquisition
attempt.
2025-09-21 15:11:28 -04:00
hayodea dbc9569775 Async: Add exception bubbling
CONT_SET_EXC: Set exception on the continuation, to be rethrown
	by the caller.
CONT_SET_EXC_AND_RET: Convenience which returns immediately
	after setting the exception.
2025-09-21 14:21:55 -04:00
hayodea 1e2cc5ef16 scripts: Add sloc 2025-09-20 19:57:13 -04:00
hayodea 31cadb2ee4 Test: Add googletest; Add qutex_tests.cpp
Well, we now have this.
2025-09-20 19:41:57 -04:00
hayodea 329d57a16d Formatting 2025-09-20 18:23:03 -04:00
hayodea 32179eee5e Qutexes: Implement them and supporting classes
Implements: LockSet, SerializedAsynchronousContinuation,
	LockerAndInvoker, LockerAndInvokerBase, Qutex.

Very big leap in functionality here. See qutexes.md for
an explanation of what we've done.
2025-09-20 18:20:52 -04:00
hayodea f05c465d61 Docs:Qutexes Update design but it's not reflective of impl 2025-09-20 18:14:17 -04:00
hayodea 79c50ff191 Docs: Qutex.md: update 2025-09-19 21:36:59 -04:00
hayodea d10217f3f5 Docs: update qutex algo plan 2025-09-19 18:52:27 -04:00
hayodea 4b0e832e27 Asnc: LockSet is now a member of serializedAsyncCont; not base class
We needed to pass in a ref to the parent SerializedAsyncContin, so
we had to make it a member var instead of a base class.
2025-09-19 01:32:52 -04:00
hayodea dd9ce63cb8 Locking:Qutex: Skeleton for Qutex class 2025-09-19 01:19:09 -04:00
hayodea b9322c5e89 Docs: Qutex updated notes -- still WIP 2025-09-19 01:18:32 -04:00
hayodea 2dfa615eb7 Rename: lockSpec.h=>lockSet.h 2025-09-18 23:05:44 -04:00
hayodea 596bc1fbd2 SpinLock: Add acquire/release() and spinPause()
We're very careful to ensure that we add a cpu_relax hint.
2025-09-18 23:01:49 -04:00
hayodea ec50526804 Rename: LockSpec=>LockSet 2025-09-18 22:21:06 -04:00
hayodea eb3366cfd2 Async: Move callOriginalCb out of AsyncCont base, into NonPostedAC
The reasoning here is to prevent silent invocations of
callOriginalCb from a base class pointer/ref.
2025-09-18 22:10:17 -04:00
hayodea b49e281010 Docs: New qutexes design 2025-09-18 20:29:37 -04:00
hayodea 9a23dbbe95 Whitespace 2025-09-17 18:18:09 -04:00
hayodea 5a3c0699f7 Locking: Add LockerAndInvoker to provide spinQing behaviour
We haven't implemented or tested this yet but it looks promising.
Also, it's surprisingly clean!
2025-09-17 16:39:11 -04:00
hayodea 816a047920 Async: new hierachy; manages reply posting and unlocking
Async: Use new [Non]PostedAsyncCont and callOriginalCb

This new hierarchy of classes gives us a central mechanism for
managing both reply-posting and lockSpec unlocking.

* callOriginalCb: Now uses a modern C++ variadic template design
  enabling it to handle both direct calling and std::bind()
  re-binding of an arbitrary number of arguments from the caller.

This enables us to mostly eliminate the repeated, bespoke
definitions of callOriginalCb littered throughout the codebase.

We've also propagated these changes throughout the codebase in
this patch.
2025-09-17 16:38:48 -04:00
hayodea 33c006b178 Locking: Add LockSpec class which manages sets of spinlocks
This class allows us to list a series of spinlocks that are all
acquired and released together. It has simple, primitive detection
for deadlocks and will throw if it detects one.
2025-09-17 16:29:58 -04:00
hayodea 0733bb9a68 Locking: add SpinLock class
Nothing much to add: add a spinlock which has a tryAcquire method.
This will be used as a primitive for building our spinQing locking
system.
2025-09-17 16:28:07 -04:00
hayodea eeaa4ed2df CMake: Add preliminary CPack support
We'll tune it later.
2025-09-16 22:19:48 -04:00
hayodea 0788bbd799 Cmake,DAPS: Add support for DAPS preprocessing
We add a new extension, .dapss for preprocessable DAPS spec files.
We now run the C preprocessor on the .dapss files to produce
.daps files.
2025-09-16 21:30:50 -04:00