Donald Haase

> semantic versioning So, in a general sense yes (though I'm not familiar with any sort of universal standard, and we don't purport to follow any). But this specifically is...

Linked some other PRs to this where I'd removed other deprecated things. From my review it seems there's only the one left that predates v2.1.0, and that should be removed...

> Linked some other PRs to this where I'd removed other deprecated things. From my review it seems there's only the one left that predates v2.1.0, and that should be...

This is caused by this: https://github.com/KallistiOS/KallistiOS/blob/99308c534d03c585d4b4aa14ad278e072081300b/include/sys/_types.h#L201-L202 Looks like it was added intentionally in '04 during the initial switch/cleanup to newlib in order to preserve the old old KOS behavior. I'd...

Moved this up to v2.3.0 as substantial work was done to reduce it's impact so far that will be in v2.2.0. I'll aim to remove that `#include ` after the...

After piles of other changes to get things prepped for it, I believe that #1208 would have addressed the most common causes of this. @Kazade any chance you could retry...

My biggest thing would be for a new tool to be built around the DC-side being able to receive and jump into an update from the PC side automatically. You...

Assigned to myself to see about harmonizing some of the overlapping changes here and in #619 and #739. All great additions but would need some changes and don't want to...

> > Assigned to myself to see about harmonizing some of the overlapping changes here and in #619 and #739. All great additions but would need some changes and don't...

I think this solution: ``` void g1_dma_irq_handler(uint32 code, void *data) { if(g1_ata_irq_hdl) g1_ata_irq_hdl(code, data); else if (cdrom_irq_hdl) cdrom_irq_hdl(code, data); } ``` Should work just fine, and `hardware_init()` does seem like...