[BUG]: Random infinite black screens in Hot Shots Golf Fore! (SCUS-97401)

[Samiversal]

Describe the Bug

The game has a tendency to generate an infinite black screen upon loading a course, forcing to reset the system. What causes this bug is unknown to me, and the bug seems to happen at random, although still somewhat frequently (I've seen up to 3 times in a row)

I do know it is a bug related to emulation since this never happens on the physical copy, and my ISO's hash corresponds to the one in your database. I am also aware that I am not the only person having this issue.

Here are some of my observations (unconfirmed):

• The bug seems to occur more often after force reseting a match (Holding R1+R2+L1+L2+SELECT+START)
• The bug doesn't seem to occur on default 3:4 aspect ratio, but can occurs on 16:9 aspect ratio.

Attatched to this report I have linked an emulog.txt, which the Discord support team of PCSX2 has already went through and said nothing was out of the ordinary.

emulog.txt

Here's a video of what the crash looks like: https://youtu.be/3HyQiLMiHyU

Reproduction Steps

• Go in any gamemode
• Select a course
• Select a character
• Start up the match
• Game is stuck on a black screen (Sometimes)

Expected Behavior

No response

PCSX2 Revision

v2.3.168

Operating System

Windows 11

If Linux - Specify Distro

No response

CPU

i5-12400F

GPU

NVIDIA RTX 3060

GS Settings

No response

Emulation Settings

No response

GS Window Screenshots

No response

Logs & Dumps

emulog.txt

[Goatman13]

I did a little research on that, and it seems to always hang or crash when trying to open one of those files (maybe some others too).

openBinder("¥CRS¥GLY¥GLY0106.XB;1") // hang
openBinder("¥CRS¥GLY¥GLY0118.XB;1") // crash or hang

There is a moment in the code when you can specify which gly file will be opened, and setting it to one of them always produces the same output.

To reproduce:
Set execution breakpoint at 0x234BF4, with condition a1 == 0x006BC1D0
When hit, edit a2 to have value 1 (should be already set), and a3 to have one of problematic values (0x06, 0x12).

Real issue seems to hit little bit later in 0054357C (exactly inside this jal there). Still no idea what real issue is, but with those steps it seems to be 100% reproductible.