Error running script: AnalyzePs3Binary.java

[Expoverse]

So I have an EBOOT.ELF and when running the AnalyzePs3Binary.java I get this error:

AnalyzePs3Binary.java> Running...
e_shnum: 0x1E
e_shentsize: 0x40
AnalyzePs3Binary.java> segment_0.1
AnalyzePs3Binary.java> .init
AnalyzePs3Binary.java> .text
AnalyzePs3Binary.java> .fini
AnalyzePs3Binary.java> .sceStub.text
AnalyzePs3Binary.java> .rodata.sceResident
AnalyzePs3Binary.java> .rodata.sceFNID
AnalyzePs3Binary.java> .lib.ent.top
AnalyzePs3Binary.java> .lib.ent.btm
AnalyzePs3Binary.java> .lib.stub.top
AnalyzePs3Binary.java> .lib.stub
AnalyzePs3Binary.java> .lib.stub.btm
AnalyzePs3Binary.java> .rodata
AnalyzePs3Binary.java> .spu_image
AnalyzePs3Binary.java> .sys_proc_param
AnalyzePs3Binary.java> .sys_proc_prx_param
AnalyzePs3Binary.java> .ctors
AnalyzePs3Binary.java> .dtors
AnalyzePs3Binary.java> .jcr
AnalyzePs3Binary.java> .data.rel.ro
AnalyzePs3Binary.java> .data.sceFStub
AnalyzePs3Binary.java> .opd
AnalyzePs3Binary.java> .got
AnalyzePs3Binary.java> .tdata
AnalyzePs3Binary.java> .tbss
AnalyzePs3Binary.java> .data
AnalyzePs3Binary.java> hashes
AnalyzePs3Binary.java> .bss
AnalyzePs3Binary.java> .sceversion
AnalyzePs3Binary.java> .shstrtab
AnalyzePs3Binary.java> _elfSectionHeaders
section: addr=0x0 size=0x0
section: addr=0x10200 size=0x2C
section: addr=0x10230 size=0x1706B78
section: addr=0x1716DA8 size=0x24
section: addr=0x1716DCC size=0x3368
section: addr=0x171A134 size=0xF4
section: addr=0x171A228 size=0x3AC
section: addr=0x171A5D4 size=0x4
section: addr=0x171A5D8 size=0x4
section: addr=0x171A5DC size=0x4
section: addr=0x171A5E0 size=0x2C0
section: addr=0x171A8A0 size=0x4
section: addr=0x171A8B0 size=0xA0738
section: addr=0x17BB000 size=0x19BA80
section: addr=0x1956A80 size=0x20
section: addr=0x1956AA0 size=0x40
section: addr=0x1960000 size=0x248
section: addr=0x1960248 size=0x148
section: addr=0x1960390 size=0x4
section: addr=0x1960398 size=0x19D84
section: addr=0x197A11C size=0x3AC
section: addr=0x197A4C8 size=0x4A8B0
section: addr=0x19C4D78 size=0x9994
section: addr=0x19CE70C size=0x4
section: addr=0x19CE710 size=0x2A4
section: addr=0x19CEA00 size=0x12E12C
section: addr=0x1AFCB40 size=0x240
section: addr=0x1AFCD80 size=0x183998
section: addr=0x0 size=0x2684
section: addr=0x0 size=0x12A
AnalyzePs3Binary.java> Processing EXEC
e_entry: 19B1748
OPD: 197A4C8
> Error running script: AnalyzePs3Binary.java
ghidra.program.model.util.CodeUnitInsertionException: Data conflict at address 019b1748
	at ghidra.program.flatapi.FlatProgramAPI.createData(FlatProgramAPI.java:1715)
	at AnalyzePs3Binary.handleExec(AnalyzePs3Binary.java:75)
	at AnalyzePs3Binary.process(AnalyzePs3Binary.java:99)
	at AnalyzePs3Binary.run(AnalyzePs3Binary.java:31)
	at ghidra.app.script.GhidraScript.executeNormal(GhidraScript.java:405)
	at ghidra.app.script.GhidraScript.doExecute(GhidraScript.java:260)
	at ghidra.app.script.GhidraScript.execute(GhidraScript.java:238)
	at ghidra.app.plugin.core.script.RunScriptTask.run(RunScriptTask.java:47)
	at ghidra.util.task.Task.monitoredRun(Task.java:134)
	at ghidra.util.task.TaskRunner.lambda$startTaskThread$0(TaskRunner.java:106)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642)
	at java.base/java.lang.Thread.run(Thread.java:1575)

What is causing this or the error? It's a ps3 eboot.bin that was converted to an elf. Is it the plugin, gildra, or the elf?

The extension version is -> ghidra_11.1.2_PUBLIC_20240713_Ps3GhidraScripts.zip

[clienthax]

Guessing you missed this?

[Expoverse]

So what I have:

Works
eboot.elf - game 1

Does not work (Stops at TOC in Console Scripting printout)
eboot.elf - game 2
eboot.symbols.elf - game 2

No, I imported the elf into Gildra multiple times to verify and I clicked "no" on analyze now. So elf was has not been analyzed. I have two eboots that it does not work on. It seems to get stuck on the TOC part, I compared this to the first eboot I have to see what came after OPD.

My first eboot works great, which I've been using the repo extension on without issues for the last two years.

Something I noticed is the first eboot that works The Program Tree shows "sections" while these eboots have "names"

Import file:

Click no:

Run script:

[Expoverse]

The eboot that works "sections"

vs eboot that doesn't work "sections"

[clienthax]

I have a feeling that you might just need to use the R2 script instead since it's not stripped, can you upload the misbehaving ones?

[Expoverse]

I can but not here. So the reason I am wanting to use Ghidra is because the eboots I have are compiled with Unreal Engine 3. I have the source code for the game and all the related tools/services for compiling, so hence compiled, but I figured if I use Ghidra to compare a fully symbol to a non symbol (different game), I can match up the functions since they are built on the same UDK version and PS3 edge is pretty much the same except incremental improvements, I wouldn't have much left to figure out since I have the source that compiled these and the knowns would have now been eliminated. So I started with the partial stripped first.

So I spent the last few days trying to get the AnalyzePs3Binary.java to work. Ghidra seems to already pick up on the TOC or something when it is imported, so it is already set. I don't know all the terms so forgive me.

This is rough but I modified AnalyzePs3Binary.java to check if there was already a reference set, which there was, and skip over it. It should check if they are a match to be correct type &&& value, but that needs changed.

// final Address elfEntryPtrAddr = currentAddress.getNewAddress(elfEntryPtr);
// createData(elfEntryPtrAddr, Pointer32DataType.dataType);

//// Custom code!
// Calculate the address for the ELF entry pointer
final Address elfEntryPtrAddr = currentAddress.getNewAddress(elfEntryPtr);

// Check if data already exists at the address
Data existingData = getDataAt(elfEntryPtrAddr);

if (existingData != null) {
	// If data already exists, log the details
	printf("Data already exists at %s: %s\n", elfEntryPtrAddr, existingData.toString());

	// Optionally, check if the existing data matches what you are trying to create
	if (existingData.getDataType().isEquivalent(Pointer32DataType.dataType)) {
		printf("The existing data is of the expected type: Pointer32DataType.\n");
	} else {
		printf("Warning: Existing data is not of the expected type! It is: %s\n",
				existingData.getDataType().getName());
	}
} else {
	// If no data exists, create the new data
	createData(elfEntryPtrAddr, Pointer32DataType.dataType);
	printf("Created new data at %s of type: Pointer32DataType.\n", elfEntryPtrAddr);
}

The AnalyzePs3Binary.java then seemed to be happy and went on to completing the mission.

But here is the catch. The AnalyzePs3Binary.java ran for 14 hours straight and only completed about 8% The Ghidra GUI also became molasses. So okay, lets try to see if we can eliminate a hardware and GUI problem at the same time.

I then moved the project over to my 32 core, 96GB ram server. Installed Kali Linux and I ran the script instead using Ghildra headless mode. analyzeHeadlessREADME

It ran quick without the GUI, but didn't exit the script and hung at the end. I'm also not sure if it even did anything, because when I opened the project up in Ghidra, nothing had changed from the original import, like other fully stripped games do when you run the script in the GUI. So I don't think the script works quite well in headless unless I am using it wrong. But headless is Fast for other scripts and the Ghildra analysis since no calls need to be made to a GUI.

So to sum things up a partial stripped eboot takes enormous amount of time for the prescript. I can't image the time Ghildra GUI analyze takes, the second part. When we aren't even talking about the Symbol version yet, which I hope isn't the case. It's over 100MB in size, so 3 times the partial, and the fully striped (normal game).

There are a lot of assumptions here and wrong statements based on what I experienced. I do believe if the fully symbol eboot can be disassembled&decompiled in Ghidra, then further improvements can be made to decompiling other PS3 games with these repo scripts. Or even new post scripts to fix some Ghildra output during auto analyze.

[eepycats]

same issue
Data conflict at address 018b4604
ghidra.program.model.util.CodeUnitInsertionException: Data conflict at address 018b4604
at ghidra.program.flatapi.FlatProgramAPI.createData(FlatProgramAPI.java:1749)
at AnalyzePs3Binary.handleExec(AnalyzePs3Binary.java:72)
at AnalyzePs3Binary.process(AnalyzePs3Binary.java:99)
at AnalyzePs3Binary.run(AnalyzePs3Binary.java:31)
at ghidra.app.script.GhidraScript.executeNormal(GhidraScript.java:460)
at ghidra.app.script.GhidraScript.doExecute(GhidraScript.java:314)
at ghidra.app.script.GhidraScript.execute(GhidraScript.java:290)
at ghidra.app.plugin.core.script.RunScriptTask.run(RunScriptTask.java:46)
at ghidra.util.task.Task.monitoredRun(Task.java:134)
at ghidra.util.task.TaskRunner.lambda$startTaskThread$0(TaskRunner.java:106)
at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1090)
at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:614)
at java.base/java.lang.Thread.run(Thread.java:1474)

---

Build Date: 2025-Dec-05 1355 EST
Ghidra Version: 12.0
Java Home: C:\Program Files\Eclipse Adoptium\jdk-25.0.1.8-hotspot
JVM Version: Eclipse Adoptium 25.0.1
OS: Windows 11 10.0 amd64
Workstation: DESKTOP-39KHB26

file: https://s3.badcoder.dev/eepycats/fish/TF.elf