#rb Per.Larsson
#jira UE-169626
#rnx
#preflight 639c4112012902cb8db43e13
- This allows us to provide the user with more ways to customize the virtualization and return more detailed info about it if the calling code wishes to log additional info. In both cases we can extend the options and the data returned without changing the API.
- Previously if we virtualized a package that was not checked out in revision control we would warn the user and then skip updating the package file on disk. This means the payloads would be uploaded but the user would be left with no local changes. Since sometimes we know we don't need to check out any package (virtualizing the packages in a change list for example) we don't want to always incur the cost of polling reivision control to see which packages do need checking out. This is why we now allow the caller to request package files be checked out via the new options enum EVirtualizationOptions.
-- If the EVirtualizationOptions::Checkout flag is provided we will poll the revision control status of all package files and then check out those which need it.
-- We still check if packages can be modified and warn the user if they can't, as package files could be locked in other ways.
- Added a new utility function to SourceControlUtilties to make it easier to check out packages. There is similar functionality elsewhere in the code base but the virtualization module is too low level to make use of it.
- Updated existing code that calls ::TryVirtualizePackages and add cases of ''using namespace UE::Virtualization' where required to improve readability.
- The UnrealVirtualizationTool now supports a new cmdline option "-checkout" that can be used when virtualizing packages. This will checkout any package that was actually virtualized so the result can be saved back out to the workspace domain. This means we no longer require the caller to have checked out the packages before running the tool.
[CL 23536832 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#jira UE-171492
#rnx
#preflight 6391f02d67018b14b581c0b0
- It is possible that a project might want to have persistent storage enabled but no caching storage, it is probably a bad idea but it should work. However before this fix doing so would cause an error to be logged during the virtualization of a package and fail the process.
- During the virtualization process we now check if pushing is enabled for cache storage before trying to push there, if not then we log that the phase is being skipped.
- Even though the above fixes the problem we would still get odd results when pushing payloads to cached storage without any backends enabled.
-- We now check if the pushing process is disabled before processing requests, so we can fail faster. There is now also a proper error value for this that can be stored in each request (FPushRequest) where as previously they'd all still display "pending"
-- We also check to see if the requested push type has any associated backend before processing requests and have a specific error value for this. Note that we consider
- We also now force the visibility of the slow task progress bar, as otherwise we might not end up displaying the virtualization progress to the progress dialog (this is a work around to a long standing problem with the slow task system due to the time limit associated with updating the dialog via EnterProgressFrame, this needs to be fixed elsewhere)
#ushell-cherrypick of 23445439 by paul.chipchase
[CL 23461405 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#jira UE-170012
#rnx
#preflight 63724fa79e3bea8079f2ed76
- The original for loop was added at a time where FCompressedBuffer did not support serialization directly, at that time the code worked.
- Now it looks like the FCompressedBuffer returned by the call to ::GetPayload is going out of scope early, which means that the FSharedBuffer ends up zero length (and quite possibly unsafe to use)
- However nowadays FCompressedBuffer has both a operator<< and direct ::Load/::Save methods so we can remove the for loop and simplify the code.
[CL 23133173 by paul chipchase in ue5-main branch]
Tested compiling fortnite, unrealeditor, lyra, qagame with non-unity/pch
#preflight 63635997876630122adeab9f
#rb none
[CL 22958990 by henrik karlsson in ue5-main branch]
#rb PJ.Kack
#jira UE-167578
#rnx
#preflight 635bdddd5b54febd38b7efd0
- Clean up the source control backend by removing the older code path that was ifdefed out and clean up the new path.
- FPerforceConnection no longer takes a TOptional<FSharedBuffer> but takes an optional pointer to a TArray of FSharedBuffer. When gathering files this should return one FSharedBuffer per file requested.
-- It might make more sense to attach the FSharedBuffer to the FP4Record but that would require some bigger changes and would be better suited to being some follow on work. All the code is private in the module and not exposed so it is fairly safe and easy to do these refactors later.
- Replaced the TArray buffer for gathering files with a purpose built class FP4DepotFile.
- When issuing a p4 print command with multiple files we will get one call to ClientUser::OutputStat followed by calls to either ClientUser::OutputText or ClientUser::OutputBinary until all of the file has been delivered, then we will get another call to ClientUser::OutputStat for the next file and so on.
-- Note with binary files we seem to get one final call to ClientUser::OutputBinary with zero size, which I assume is supposed to signal the end of the transfer, text files do not get this, they seem to just stop.
[CL 22888671 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#jira UE-168452
#rnx
#preflight 635a4773f1493f97864fb660
- IVirtualizationBackend can now have its operations disabled after it is created.
-- This means that a backend that only supported pushing and not pulling would end up with no valid operations if marked as read only. This will not assert because we do this after the backend is created and we only check on creation.
[CL 22810918 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#jira UE-168057
#rnx
#preflight 635656e6e6096564af1bc117
- Work on UE-165308 broke the following commandlines as they were accidently converted to cmdline switches rather than values
-- VABackendGraph
-- VAMissBackends
-- VAMissChance
- ::IsCmdLineSet has now been split into two utility wrappers, one for params (bools) and one for values.
[CL 22792790 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#jira UE-163093
#rnx
#preflight 63526abcae33b04ec1ee4a65
- The caller can now request more than one payload to be pulled froim the system at a time. In theory this will allow backends to pull data quicker.
-- The file system backend works with the new system but makes no attempt to take advantage as the backend is not intended for serious production use.
-- The DDC backend should be taking full advantage of the new batching API
-- Note that although the source control API does attempt to take advantage of batching, the internal source control API implementation is not doing so. This will be addressed in a future submit.
- This does make the pulling logic a bit more complicated in FVirtualizationManager as we need to deal with some payloads being found in the first backend, some in the second and some in the third etc.
-- To help deal with this a new class FPullRequestCollection has been added which abstracts a lot of complexity away.
[CL 22707955 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#jira UE-165308
#rnx
#preflight 634ff17e8449f85a47390fbe
- It was pointed out that a command line option containing '-' characters could confuse parsers and it is not worth the risk to keep them.
- Now we accept versions of each command line without the '-' character which will be the version that we document for users in 5.1
[CL 22650940 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#rnx
#preflight 634805f7f93be0f634b8ed5b
- Added a new FArchive 'FFixedBufferWriterArchive' that will let me serialize to an already allocated, fixed sized memory buffer.
- Moved the code for opening and reading packages to shared utility functions so that it is easier to keep the error messages consistent.
- Moved common code to a new function ::TryRehydrateBuilder so that both rehydration methods can share it.
- Note that technically we are changing the API since now we accept a TConstArrayView for the package file paths for both the virtualization and rehydration paths, but existing code will be compatible with this change.
[CL 22511406 by paul chipchase in ue5-main branch]
#rb trivial
#jira UE-160942
#rnx
#preflight 634557f5c272487ed4ab2014
- In addition to adding the hit counter (which can be very useful when debugging) the stats logging code has been factored out to a single lambda tht is called for each set of stats to make it easier to update in future.
[CL 22457126 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#rnx
#preflight 633ad936c250f60502b2d8cb
- Everything was working but when a cvar is set via a device profile we expect the cvar to be registered sooner, hence the warning.
- Added a new define UE_USE_GLOBAL_CVAR to FVirtualizationManager.cpp, when set we use a static global cvar for 'VA.AllowPkgVirtualization' rather then creating and registering one when the system is mounted.
- Since the cvar can now outlive the manager we also needed to add a way to remove the bound lambda from the cvar when the manager is destroyed.
[CL 22322164 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#jira UE-160942
#rnx
#preflight 6336f2ab5c2225fe5f6c08a5
- Removed the status enum from FPushRequest and replaced it with a class that represents the push result instead.
-- This lets us return more context about problems. At the moment it is only being used to return the filter reason if a payload is filtered out. In the future we could return per payload error messages etc.
- Now that we return more detailed info about the payload push we no longer report payloads that were already in the backend as being pushed which could cause some very misleading stats to be shown.
- Changed the backends to use the new results system.
-- Not all paths were correctly setting the state in the file system backend.
-- The source control backend does not set the result in all paths, which will leave the results in the "pending" state. Which will still mark the payload as not uploaded and so is safe to do.
- Removed the pushing of a single method from IVirtualizationBackend as it was only used in one place internally, so we might as well just change that code to call the batch push overload.
[CL 22279127 by paul chipchase in ue5-main branch]
Before:
Total CPU Time: 53783.640625 s
Total time in Parallel executor: 558.66 seconds
After:
Total CPU Time: 47886.140625 s
Total time in Parallel executor: 498.81 seconds
#jira
[CL 22173145 by bryan sefcik in ue5-main branch]
#rb Per.Larsson
#jira UE-163834
#rnx
#preflight 632d98e4b4515b7e221654ec
### Virtualization
- At the moment there is no way in the editor UX to change the source control settings for the source control backend, so what ever settings are found in the global environment would be applied when the editor is first run and then saved to a config file under the <saved> directory. From this point onwards those settings would always be used when setting up the source control backend. If the settings were wrong, the first time that the editor was started, then we'd continue to use the incorrect settings even if the global enviroment was fixed. Making the backend work at that point would require that the user know about the local ini file and change the settings there.
- We cannot fix this by just ignoring the ini file as it has been requested that users can add their settings there in case they want to customize things, so we just need to avoid saving to it.
- Added documentation to the source control backends header file on how to set up the ini file if needed.
### Perforce
- FSourceControlInitSettings now accepts EConfigBehavior which describes how we want the source control provider to deal with its settings with regards to the ini file.
- Note that EConfigBehavior is not a bitfield as we do not wish to support only writing to the ini file, there is no point if it cannot be read from.
- When the source control provider is creatred we pass on the into from EConfigBehavior to the FPerforceSourceControlSettings to enable/disable saving and loading.
[CL 22163769 by paul chipchase in ue5-main branch]
#rb Per.Larsson
#jira UE-164278
#rnx
#preflight 63284a248c3def91aa62bd83
- This replaces the old cvar 'VA.DisableVirtualization' which is being removed.
- We will log each time the value is switched to aid in future debugging.
- The name needed to be changed as we needed to invert how it worked, also "DiableVirtualization" was a bit misleading and made people think it turned the entire system off, not just the process of virtualizing packages on submit.
-- The member bEnablePayloadVirtualization was also renamed to bAllowPackageVirtualization to make its purpose clearer.
- The cmdline version was changed from 'VA-DisableVirtualization' to 'VA-SkipPkgVirtualization' for similar reasons.
[CL 22146910 by paul chipchase in ue5-main branch]
