Using a not-equal (!=) termination condition in a loop behaves differently then using a less (<) termination behavior. This is not the case on other gpus.
Attached is a screenshot from the Sascha Williams Triangle (https://github.com/SaschaWillems/Vulkan) example with a modified `triangle.frag` shader.
CPU: i7-11370H
Modified triangle.frag shader:
```
#version 450
layout (location = 0) in vec3 inColor;
layout (location = 0) out vec4 outFragColor;
float computeValueA()
{
float value = 0;
for(int i = 0; i != 10; ++i) {
value += 0.1;
}
return value;
}
float computeValueB()
{
float value = 0;
for(int i = 0; i < 10; ++i) {
value += 0.1;
}
return value;
}
void main()
{
float value = gl_FragCoord.x < 640 ? computeValueA() : computeValueB();
outFragColor = vec4(value, value, value, 1.0);
}
```
Note that computeValueA and computeValueB are expected to return the same value, however as you can see in the attached screenshot they do not.
CPU Info:
```
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 39 bits physical, 48 bits virtual
Byte Order: Little Endian
CPU(s): 8
On-line CPU(s) list: 0-7
Vendor ID: GenuineIntel
Model name: 11th Gen Intel(R) Core(TM) i7-11370H @ 3.30GHz
CPU family: 6
Model: 140
Thread(s) per core: 2
Core(s) per socket: 4
Socket(s): 1
Stepping: 1
CPU max MHz: 4800.0000
CPU min MHz: 400.0000
BogoMIPS: 6604.80
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l2 invpcid_single cdp_l2 ssbd ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves split_lock_detect dtherm ida arat pln pts hwp hwp_notify hwp_act_window hwp_epp hwp_pkg_req avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b fsrm avx512_vp2intersect md_clear ibt flush_l1d arch_capabilities
Virtualization: VT-x
L1d cache: 192 KiB (4 instances)
L1i cache: 128 KiB (4 instances)
L2 cache: 5 MiB (4 instances)
L3 cache: 12 MiB (1 instance)
NUMA node(s): 1
NUMA node0 CPU(s): 0-7
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Enhanced IBRS, IBPB conditional, RSB filling, PBRSB-eIBRS SW sequence
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected
```
Link Copied
For completeness sake, here's a screenshot of the same shader on the Nvidia gpu on the same laptop:
Hii @BastianBlokland thank you for reaching out Indie Game Developer Support!
I just did some quick experiments on Windows and I'm not able to see the issue (picture for reference) with the latest driver v.4669 on Iris (Xe) in Windows 11
Our next step: Debug it on a similar configuration to yours using a Linux distro and verify the issue on our end
Stay tuned!
Karen
Hey @BastianBlokland ty for the quick response
Indeed something was wrong with the system I was working with. Now, as I don't have a laptop available, we set up a TGL desktop machine where I tested the code.
After a reinstall with Ubuntu I could successfully create the examples replacing the triangle hsl and gsl shader files with the ones you shared.
My setup has a UHD 750 Graphics card and in the picture you can see the kernel version I have as well.
The shader is loading as in the Windows example I shared using Iris Xe Graphics and also in the SS of the NVIDIA card you tested on. Mesa utils were also recently updated so we have the same Graphics driver also.
Summary: I'm unable to reproduce the issue
I'm thinking it could be related to the combo Iris Xe + Ubuntu which would be too specific.
It may take a while for me to perform that test on a laptop with Ubuntu so I'll appreciate your patience in advance.
In the meantime, I wonder if you could test in a system that has only Intel or Nvidia GPU, just to discard driver conflict in a dual system.
Please let me know your findings!
Karen
