Karthik Karanth

I created a minimal 1-face mesh that reproduces the problem: [crash_test.zip](https://github.com/libigl/libigl/files/6058176/crash_test.zip)

> readSTL shouldn't check the file extension I agree, I meant that it should check the header to determine if its a STL file. But I just realized that that...

This still happens. Could you possibly include some way to easily copy the examples?

I also had to remove the array index. `get_python_lib()` is returning a single string: ```diff diff --git a/setup.py b/setup.py index 3be4a43..dfe69d2 100644 --- a/setup.py +++ b/setup.py @@ -19,9 +19,9 @@...

> Does the application work for you from the command line yes, it works fine > Maybe the simplest thing to try would be to replace the existing glm folder...

some more info: the hid_enumerate API just seems to not list the JoyCon

I was looking into the same problem. The kernel exposes a `highest_perf` number: ``` $ head -n +1 /sys/devices/system/cpu/cpu*/acpi_cppc/highest_perf ==> /sys/devices/system/cpu/cpu0/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu1/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu2/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu3/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu4/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu5/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu6/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu7/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu8/acpi_cppc/highest_perf /sys/devices/system/cpu/cpu9/acpi_cppc/highest_perf...

> Without SSAO and depth pre-pass I was getting +1000 FPS on my 1070, depth pre-pass lowered to ~850, but SSAO brings it down to ~170. How much FPS did...

I've tested the changes on raspbian 10

Thanks @AldaronLau ! yes `std::time::Duration` makes more sense here! Regarding #39 , with this PR (#40) and the WIP PR #38, I am able to use my joycons' accelerometer. I'm...