OSDev.org

EDIT: Solved. I had to learn some Cairo before testing it

You say you ported those applications but appear to be trying to run it in kernel space. You're also writing directly to a physical address (instead of using the LFB information Grub/VBE gave you) when you should be mapping that to virtual memory. You're not checking the return value of the cairo functions (I'm assuming they return NULL or something else upon failure).

goku420 wrote:You say you ported those applications but appear to be trying to run it in kernel space. You're also writing directly to a physical address (instead of using the LFB information Grub/VBE gave you) when you should be mapping that to virtual memory. You're not checking the return value of the cairo functions (I'm assuming they return NULL or something else upon failure).

It was for a simple test, not for real gui.
I'm working with QEMU, so I assume framebuffer is 0xFD000000.

I fixed the problem, looks I should learn the basics of Cairo. Now everything looks good.

Agola wrote:It is for a simple test, not for real gui.
I'm running my os in QEMU, so I assume framebuffer is 0xFD000000.

I fixed the problem, looks I should learn the basics of Cairo. Now everything looks good.

For the benefit of future readers, what was the problem/solution?

goku420 wrote:

Agola wrote:It is for a simple test, not for real gui.
I'm running my os in QEMU, so I assume framebuffer is 0xFD000000.

I fixed the problem, looks I should learn the basics of Cairo. Now everything looks good.

For the benefit of future readers, what was the problem/solution?

Cairo wasn't working.

cairo_format_stride_for_width expects format and width as arguments, but I passed height.
cairo_format_stride_for_width(CAIRO_FORMAT_ARGB32, 480) had to be cairo_format_stride_for_width(CAIRO_FORMAT_ARGB32, 640)

no92 wrote:There is a reason editing posts is generally frowned upon.

Sorry, but I thought it was such a rookie mistake and nobody (except me) can do a mistake like that.