Also fix two other minor issues: Make Move constructor available for FIntVirtualizedBulkData, and make Register public so UObjects can notify the registry that they own a BulkData.
#rb Paul.Chipchase
#rnx
#ROBOMERGE-AUTHOR: matt.peters
#ROBOMERGE-SOURCE: CL 18431598 in //UE5/Release-5.0/... via CL 18435352
#ROBOMERGE-BOT: STARSHIP (Release-Engine-Staging -> Release-Engine-Test) (v897-18405271)
[CL 18435594 by matt peters in ue5-release-engine-test branch]
#ROBOMERGE-AUTHOR: marc.audy
#ROBOMERGE-SOURCE: CL 18427792 in //UE5/Release-5.0/... via CL 18427811
#ROBOMERGE-BOT: STARSHIP (Release-Engine-Staging -> Release-Engine-Test) (v897-18405271)
[CL 18427827 by marc audy in ue5-release-engine-test branch]
#rb PJ.Kack
#rnx
#preflight 61a8d6d8a7179bfa550218af
- This is an experimental feature being used during testing and development, it is not intended to be used by a shippable project.
-- The code is all wrapped by a single define, UE_ENABLE_VIRTUALIZATION_TOGGLE, making it easy to disable and remove
-- The actual method called on bulkdata, ::SetVirtualizationOptOut is being submitted already deprecated to prevent it's accidental use.
- The use case is to allow projects to opt out of virtualizing static meshes as they are more fragile (i.e. hard crashes) when their payloads cannot be accessed.
- Even when ::SetVirtualizationOptOut is called the feature will not work unless [Core.System.Experimental]AllowVirtualizationOptOut=True is set in the config file. This allows projects to opt into this.
- When a bulkdata is opting out, we simply send the payload to be stored at the end of the package like we used to, rather than send it to the package trailer. This way we do not need to store any additional state in the trailer to track if a payload is allowed to be virtualized or not. Since this feature is throw away I did not want to make any data format changes.
#ROBOMERGE-AUTHOR: paul.chipchase
#ROBOMERGE-SOURCE: CL 18352158 in //UE5/Release-5.0/... via CL 18352163
#ROBOMERGE-BOT: STARSHIP (Release-Engine-Staging -> Release-Engine-Test) (v895-18170469)
[CL 18352180 by paul chipchase in ue5-release-engine-test branch]
FArchiveReplaceObjectRefBase no longer tracks replaced references by default. GetReferencedResources will ensure if archiver was not run with TrackReplacedReferences.
#rb Michael.Noland
#preflight 61a7c89c1a368fd6038f4b79
#ROBOMERGE-AUTHOR: marc.audy
#ROBOMERGE-SOURCE: CL 18343619 in //UE5/Release-5.0/... via CL 18343639
#ROBOMERGE-BOT: STARSHIP (Release-Engine-Staging -> Release-Engine-Test) (v895-18170469)
[CL 18343679 by marc audy in ue5-release-engine-test branch]
#rb PJ.Kack
#rnx
#preflight 61a4b235405273b2c3daa7c7
### Virtualization
- The idea is to move from our current set up, where the virtualization happens when a package is saved and the payloads are saved to a local virtualization storage system with the final push to persistent storage occurring when the package is submitted to a system where we save payloads into the package file, but move them to persistent virtualization storage when the package file is submitted.
- The main advantage is that if someone submits a package file to source control without virtualizing it, we don't have to worry that others might not be able to load the package, the worst case scenario is that data sizes get bigger until the package is virtualized again.
- This is only the first pass, in the future we can do further optimizations, like not storing payloads locally if we know that they are already in the persistent storage system etc.
- In order to keep the virtualization process simple we want to be able to do it without needing to re-save the package, so to this end we will storage the payloads at the end of the package file as before, but instead of storing them inside of the package file format we will use a new container, the FPackageTrailer that is appended to the package file instead.
- In theory this will make future sidecar work easier as the trailer can just be moved to the sidecar file as as long as we know where to find the trailer things will just work (tm)
- For now VBD will continue to serialize it's offset/size to disk and use those values to read the payload directly when not virtualizing. Ideally we'd use the trailer but this will help reduce the risk.
### Current Issues
- The system does not work with the editor domain and has minor issues with text based assets
- The trailer system is disabled by default via the config system [Core]UsePackageTrailer=False and can then be enabled on select test projects until fully functional.
### PackageTrailer
- The trailer is split into two parts FPackageTrailerBuilder and FPackageTrailer
- FPackageTrailerBuilder is used when building the trailer during package save and FPackageTrailer is the structure we load and actually use.
- When saving the trailer we try to avoid using containers and such so that we can try to keep the specification for the file format very clear so that licensees can build their own tools for the virtualization process if they so desire.
- the header file contains a breakdown of the new format.
### VirtualizedBulkData
- FPayloadToc is now only used by the sidecar experimental feature. In a future update the sidecar will be changed to basically be the package trailer but stored in it's own file.
- FindPayloadsInPackageFile has been moved to PackageTrailer.h as it has a closer association with that code.
- Added a new method, ::LoadFromPackageTrailer, to load payloads from the package trailer without needing to know where the payload is on disk. Using this is opt in with the cvar "Serialization.LoadFromTrailer" and is only provided for debugging purposes. Although once this system has matured it will likely become the preferred way to access payloads.
- We no longer allow virtualization of bulkdata payloads on save, but I am not sure if we might want to allow this as an option in the future. So for now the code branch is disabled by the global constant bAllowVirtualizationOnSave.
- Added utility functions ::GetLinkerLoadFromOwner and ::GetTrailerFromOwner for easy access to the LinkerLoad/Trailer from the owning object.
- Added a utility ::UpdateArchiveData for seeking back to a known position in an archive, writing over it, then seeking back, which is fairly common in the package saving code paths. This might be worth moving to Archive.h for general use.
- ::LoadFromPackageFile has been refactored to prefer returning error values early over nested ifs.
- We no longer serialize the EFlags::IsVirtualized flag to disk. Instead it is applied when the package is loaded if we detect that a payload is virtualized and then removed before saved to disk. This allows the virtualization process to occur without resaving the package.
- We currently support the old serialize to disk path which is enabled when the FPackageTrailer system is disabled and the newer system when it is enabled.
- We do however continue to serialize the offset in the package file to disk when saved. Strictly speaking this is not needed as we can look this up from the package trailer when the package is loaded but continuing to serialize it out to disk helps keep the old and newer data format paths working together without adding too much special case code.
-- Once the FPackageTrailer feature is no longer optional we can consider changing this.
### SavePackage/SavePackage2
- Removed the functions that would create the FPayloadToc
- Added new functions to help create the trailer.
- Moved the end of package tag out of the if to it's own scope (still won't run if a package writer exists), the reasoning is to make it clearer where the package format ends and where we can start writing the payload trailer.
- We have checks to make sure that the trailer does not try to write to disk for both text based assets and if the editor domain is enabled. Support for these features will be added later, but since the package trailer system is disabled by default this shouldn't cause problems.
### LinkerLoad
- After we parse the FPackageFileSummary we now also parse the header of the package trailer. In the workspace domain this will incur additional seek costs (which may or not actually seek and invalidate the internal file cache depending on how large the package's exports are) as the trailer is found after the package
-- In the future, the package trailer header can be stored right after the FPackageFileSummary in the editor domain which will remove this overhead.
- By parsing the header of the trailer at this point, virtualized bulkdata objects in the package can use the look up info to determine if their payloads are virtualized or not, and if not then where they reside inside the package file.
### LinkerSave
- We no longer collect a list of all virtualized bulkdata in a package while saving as we no longer generate the FPayloadToc so BulkDataInPackage has been removed.
- We now gather data to be appended to the package trailer via TrailerBuilder.
### DumpPayloadToc
- This command now dumps info based on the package trailer if one can be found.
#ROBOMERGE-AUTHOR: paul.chipchase
#ROBOMERGE-SOURCE: CL 18307893 in //UE5/Release-5.0/... via CL 18307905
#ROBOMERGE-BOT: STARSHIP (Release-Engine-Staging -> Release-Engine-Test) (v895-18170469)
[CL 18307913 by paul chipchase in ue5-release-engine-test branch]
This represents UE4/Main @18073326, Release-5.0 @18081140 and Dev-PerfTest @18045971
[CL 18081471 by aurel cordonnier in ue5-release-engine-test branch]
This represents UE4/Main @17911760, Release-5.0 @17915875 and Dev-PerfTest @17914035
[CL 17918595 by aurel cordonnier in ue5-release-engine-test branch]
This represents UE4/Main @17774255, Release-5.0 @17791557 and Dev-PerfTest @17789485
[CL 17794212 by aurel cordonnier in ue5-release-engine-test branch]
#rb Danny Couture
#rnx
### Problem
- The crash would occur when the texture being since contains a reference to another texture (via composite) which is also in need of syncing on the users machine.
- After the files on disk have been updated the current packages are reloaded but the original version of the texture will still be in memory when the composite texture notifies that it has been reloaded causing a texture compilation job with the older now invalid texture.
-- VirtualizedBulkData relied on out of date packages not trying to access their payloads on disk once the file has been changed, but with the above scenario it was possible. There could also be other paths causing this (even if the access of out of date package payloads at this point is a waste of time)
### Fix
- The virtualized bulkdata objects now reference themselves with the FLinkerLoader if the payload is on disk, just as the old bulkdata system did. When file changing operations are invoked, the FLinkerLoader will inform it's referenced bulkdata to load the payloads off disk and hold them in memory.
-- Note that virtualized bulkdata was trying to avoid this memory bloat by design, but if we cannot be certain that the bulkdata will not try to access the payload after the file changes (as was thought) then we must take the payload into memory.
- Added a log message to VBD if we detect that the loaded payload does not match the VBD members expectations. This log message can either be an error or a fatal error depending on the define 'VBD_CORRUPTED_PAYLOAD_IS_FATAL' (defaulted to on)
- The registration to FLinkerLoader can be easily disabled by setting the define 'VBD_ALLOW_LINKERLOADER_ATTACHMENT' to 0
### Outstanding Issues
- Cloned bulkdata objects (via assignment) do not currently register themselves to the FLinkerLoader (the old bulkdata system did not) if we were to add this then we need to make sure that the FLinkerLoader registration is thread safe. This should be done as a separate work item so that we can unblock the current workflows that this changelist addresses.
-- Note that if we did not update cloned bulkdata to work with this system then we could drop the registration and just iterate over the package's bulkdata and force them to load into memory when required. However that would leave the cloned virtual bulkdata in an uncertain state and is not desirable.
- When saving a package, the virtualized bulkdata objects that it contains will change their members so that future payload access can come from the newly saved files, where as the old bulkdata system would force the payload to remain in memory for the rest of the editor session. If we want to keep this then we will need to try and make sure that the newly updated bulkdata can be registered so that it receives future notifications about the file being changed.
#ROBOMERGE-AUTHOR: paul.chipchase
#ROBOMERGE-SOURCE: CL 17648118 via CL 17648147 via CL 17648161 via CL 17648163 via CL 17648175
#ROBOMERGE-BOT: STARSHIP (Release-Engine-Staging -> Release-Engine-Test) (v875-17642767)
#ROBOMERGE[STARSHIP]: UE5-Main
[CL 17648184 by paul chipchase in ue5-release-engine-test branch]
* Decouple container concept from IoDispatcher
* Decoiuple PackageStore implementation from AsyncLoading2
* Restore ucas unmount fix that got kist when merrging from UE4
* Fix packages being left in the PackageStiore even after unmounting contaiiners
#rnx
#rb pj.kack, per.larsson
#preflight 61520cc52afc2d0001146ce7
#ROBOMERGE-AUTHOR: carlmagnus.nordin
#ROBOMERGE-SOURCE: CL 17641845 in //UE5/Main/...
#ROBOMERGE-BOT: STARSHIP (Main -> Release-Engine-Test) (v874-17637634)
[CL 17642353 by carlmagnus nordin in ue5-release-engine-test branch]
Serialization and the FBuildDefinition type of derived data have not been implemented yet. This version is functional in the editor when used with buffers or unstructured cache keys.
#rb Zousar.Shaker
#preflight 614cdb35c2ef060001f6425b
#ROBOMERGE-AUTHOR: devin.doucette
#ROBOMERGE-SOURCE: CL 17614269 in //UE5/Main/...
#ROBOMERGE-BOT: STARSHIP (Main -> Release-Engine-Test) (v871-17566257)
[CL 17614332 by devin doucette in ue5-release-engine-test branch]
#rb Per.Larsson
#rnx
#preflight 612e03f6677f0e000144b652
- Updated the comments to make the methods intent very clear (unload current payloads and prevent future payloads from being loaded from disk)
- Fix an inconsistency with BD2 running from IoStore, where calling ::RemoveBulkData was not preventing future loads in all cases. We now invalidate the PackageID when running from the IoStore in addition to unregistering the filepath token when running from pak files.
- No longer clear the flags when calling ::RemoveBulkData to be consistent with BD1.
- We might want to consider deprecating this method name and using something like ::Reset instead, but that would imply that we would reset all members and I do not see the value in making this much larger change if we do not need to do so.
#ROBOMERGE-SOURCE: CL 17366830 in //UE5/Main/...
#ROBOMERGE-BOT: STARSHIP (Main -> Release-Engine-Test) (v865-17346139)
[CL 17366843 by paul chipchase in ue5-release-engine-test branch]
#rb PJ.Kack
#rnx
#jira UE-122149
#preflight 612790e345848f000187730e
- PVS is claiming that delete is being called on a void* which is clearly incorrect (I have confirmed that the correct destructor is being called etc)
- Changing the type in the for loop to auto seems to clear up the confusion, so submitting this as a work around while we reach out to PVS about the false positive.
#ROBOMERGE-SOURCE: CL 17349166 in //UE5/Main/...
#ROBOMERGE-BOT: STARSHIP (Main -> Release-Engine-Test) (v865-17346139)
[CL 17349175 by paul chipchase in ue5-release-engine-test branch]
Remove limit on pending package size since request buffers are no longer allocated up front.
Rely on the iodispatcher to sort requests instead of using the global load order for each package.
No longer store export bundle size and load order in the package store.
#jira FROST-4677
#rb per.larsson
#rnx
#ROBOMERGE-SOURCE: CL 17267090 in //UE5/Main/...
#ROBOMERGE-BOT: STARSHIP (Main -> Release-Engine-Test) (v858-17259218)
[CL 17267107 by carlmagnus nordin in ue5-release-engine-test branch]
Add more flexibility to the redirects/localization functionality by moving redirect information away from the package summaries and import tables
Changed import references from global import hashes to package local export hashes
#rb pj.kack
#rnx
#jira none
#ROBOMERGE-SOURCE: CL 17014898 in //UE5/Main/...
#ROBOMERGE-BOT: STARSHIP (Main -> Release-Engine-Test) (v839-17012307)
#ROBOMERGE[bot1]: dev-enginemerge-test
[CL 17014905 by carlmagnus nordin in ue5-release-engine-test branch]
Updating FVirtualizedUntypedBulkData and textures to use the BulkDataRegistry.
BulkDataRegistry: Add get/put accessors for the cached BulkDataList of packages.
EditorDomain: Move ClassDigests into a global variable that can be shared with BulkDataRegistry.
EditorDomain: Improve performance of GetFileSize by fetching metadata only.
Tickable Cook Objects, for systems used by the cooker that need to be ticked.
Implementation of the the BulkDataRegistry that uses the DDC cache for persistent storage of the BulkDataList.
#rb Devin.Doucette, Paul.Chipchase, Zousar.Shaker
#ROBOMERGE-SOURCE: CL 16768772 in //UE5/Main/...
#ROBOMERGE-BOT: STARSHIP (Main -> Release-Engine-Test) (v835-16672529)
[CL 16768778 by matt peters in ue5-release-engine-test branch]
