sophia.git/kc/core/memory/memory.c, branch kjarna [no description] https://git.adachristine.is/sophia.git/atom/kc/core/memory/memory.c?h=kjarna 2026-05-26T21:32:27Z we're bringing kjarna back and not doing the crazy stuff with trying to have task management during efi. that was a bit extra. 2026-05-26T21:32:27Z Ada Christine adachristine18@gmail.com 2026-05-26T21:32:27Z urn:sha1:a10aabfcd52f702057316018cd7847ab2bfe4aa1 kjarna initial impl 2024-02-11T23:48:04Z Ada Christine adachristine18@gmail.com 2024-02-11T23:48:04Z urn:sha1:3f9278b48ff4bc0d2cf1604034b3bba548572f04 - don't typedef away function pointers 2024-02-02T01:01:44Z Ada Christine adachristine18@gmail.com 2024-02-02T01:01:44Z urn:sha1:2fa4e64d36e14e77d9e8ed547d2cc96b6c2ee7de - cleanups handle reserved memory regions correctly by ignoring them 2023-05-20T17:42:55Z Ada Christine adachristine18@gmail.com 2023-05-20T17:42:55Z urn:sha1:e9cee5052da1298347c362ee6ae89b7fdaa84d04 preparations for user mode 2023-02-19T23:31:10Z Ada Christine adachristine18@gmail.com 2023-02-19T23:31:10Z urn:sha1:838f39324a1f305562addce6d807400522d9c469 headers moved around 2023-01-30T00:05:53Z Ada Christine adachristine18@gmail.com 2023-01-30T00:05:53Z urn:sha1:798b1dc391f424868f5f4bac8937c191ed4716b8 pulled out kprintf() and friends into lib/ 2022-03-17T14:03:12Z Ada Christine adachristine18@gmail.com 2022-03-17T14:03:12Z urn:sha1:65a111a543f49c645ec21ef44f289c29ccb27e8d some updates to page handling 2022-03-14T20:59:18Z Ada Christine adachristine18@gmail.com 2022-03-14T14:20:34Z urn:sha1:73e5768d3a159b2e9b554fa5396a759f3d283c6c whoops squash me context struct for ISRs rather than using param registers; second-stage handlers should deal with that i think 2022-02-24T16:48:53Z Ada Christine adachristine18@gmail.com 2022-02-24T16:48:53Z urn:sha1:9fcd505ab3106e9fbf586e6d16e53e3e44936215 many changes 2022-02-24T01:58:07Z Ada Christine adachristine18@gmail.com 2022-02-24T01:58:07Z urn:sha1:1923e47fd296399180eaf19b990d5e951d71ccd2 - boot time memory allocation has changed. the fractal page map has been replaced by a single self-mapped page giving 2MiB of ready virtual space to play with without needing a fully-initialized memory manager. this scheme may be repeated for other purposes - the page stack has now been totally overhauled to make use of sparse allocation and all of the ugly init_*() procedures are gone. long live the new interface. also reference counting is a thing now. sort of. - found a bug in vmt_init_node where insertion did not use the root node as the parent in the case of no predecessor node. whoops.