490e1b2777
WIP: Track grant ownership.
2023-07-25 10:52:18 +02:00
8dfa73db35
Post-refactor fixes.
2023-07-25 10:52:18 +02:00
34b4512dbd
Refactor Grants.
...
Page alignment is now verified at compile time, for example.
2023-07-25 10:52:13 +02:00
d4b5899c53
Remove allocator_owned hack.
2023-07-25 10:49:55 +02:00
56f88e80c2
Usercopy migration
2023-07-06 13:03:21 +00:00
58c0c5d040
Remove one level of indirection for Context::name.
2023-05-27 14:27:34 +02:00
4621cd674d
Fix kernel DoS via sigreturn from non-signal context
2023-05-04 15:58:05 +04:00
8e0f54cb31
Clippy fixes
2022-11-11 13:19:14 -07:00
5d55d4eb87
Fix warnings
2022-07-29 18:33:54 -06:00
e99cbcf5d2
Fixes for compiling aarch64
2022-07-29 18:06:53 -06:00
0c80643077
Make graphical_debug arch independent and fix lots of warnings
2022-07-29 15:56:44 -06:00
f3faf33e3f
Add a workaround to treat bootstrap mem as owned.
2022-07-28 14:17:02 +02:00
5bbfdcda6b
Add mmap-min-addr and support unmapping parent PTs.
2022-07-27 10:52:45 +02:00
308c2cc711
Support modifying processes' sigactions.
...
This is, other than vfork, the last piece of functionality that the
previous clone() offered (CLONE_SIGHAND) which previously was not
implemented.
2022-07-27 10:52:29 +02:00
c912d9e0db
Partial: migrate syscall handling code to RMM.
2022-07-27 10:52:23 +02:00
8970ce1fe7
Benefit from addrspace abstraction in switch.
2022-07-27 10:52:22 +02:00
1cdd462244
Move the initfs scheme to userspace.
2022-07-27 10:52:22 +02:00
cb40eb3792
Support reading all grants and transferring grants.
2022-07-27 10:52:10 +02:00
283ada82a0
WIP: Remove SYS_CLONE (to be done in userspace).
2022-07-27 10:52:03 +02:00
563121596d
Fix running on multi_core.
...
Turns out the problem all along was that the ActivePageTable was never
dropped in usermode_bootstrap. So as soon as any other hardware thread
tried to do page table business, it deadlocked!
2022-07-27 10:51:45 +02:00
6e5015dcab
WIP: Add necessary interfaces for setuid/setgid.
2022-07-27 10:51:45 +02:00
37f9b292f1
Add kfmap to fix properly reobtaining grants.
2022-07-27 10:51:45 +02:00
23f49414bd
Fix phys offset, lock grants correctly.
2022-07-27 10:51:45 +02:00
15b029de36
Fix everything all the way to booting to desktop.
2022-07-27 10:51:45 +02:00
e6e1348072
Implement exec, and change UserGrant allocator.
2022-07-27 10:51:44 +02:00
846318e716
WIP: Attempt implementing fexec in userspace.
2022-07-27 10:51:43 +02:00
4d7da495f5
Fix "clone grant using fmap"
...
This does the same as the previous MR, but fixes the issue where the
parent process got the mapping (and at the wrong address) and not the
child process.
2022-04-11 21:31:36 +02:00
e72fd5a0e4
Fix a kernel deadlock in empty().
...
For more information, see https://gitlab.redox-os.org/4lDO2/kernel/-/commit/b3b5d1b864eec030d53ccd2ab907c68958f140aa
2022-04-09 13:38:59 +02:00
87b3bef06c
Revert "Merge branch 'clone_grant_using_fmap' into 'master'"
...
This reverts merge request !190
2022-03-30 14:54:49 +00:00
a3356c3fdd
Clone grant using fmap
2021-12-30 11:27:44 +01:00
cd3dcf153e
Do not print interpreter
2021-12-01 08:40:11 -07:00
4f259e3589
Workaround for thread race conditions
2021-11-30 18:03:34 -07:00
515cab03eb
Fix a number of warnings
2021-11-29 20:01:27 -07:00
5f4978a6dc
Return ENOMEM rather than panicking for PML4 alloc.
2021-08-11 17:46:40 +02:00
67cc6799bd
Fix possible UB by checking for null allocating FX.
...
Namely, the global allocator API in Rust, actually only returns a null
pointer on failure, rather than wrapping it in a Result, which AllocRef
does. Since Box::from_raw(null) is direct UB, this can in theory lead to
very strange behavior.
2021-08-11 17:46:40 +02:00
465c461b60
WIP: Stop using recursive mapping.
...
Note that this is very preliminary, and I merely got my already freezing
kernel branch not to triple fault, but I would probably apply this patch
to upstream.
What is changed here, is that rather than relying on recursive mapping
for accessing page table frames, it now uses linear translation
(virt=phys+KERNEL_OFFSET). The only problem is that the paging code now
makes assumptions that the entire physical address space remains mapped,
which is not necessarily the case on x86_64 architecturally, even though
systems with RAM more than a PML4 are very rare. We'd probably lazily
(but linearly) map physical address space using huge pages.
2021-08-11 17:46:33 +02:00
87241a9d3c
Pass AT_PHDR via a grant rather than the stack.
2021-08-01 14:57:41 +02:00
3eedbeb14d
WIP: Let userspace manage fsbase/gsbase and TLS.
2021-08-01 12:09:22 +02:00
0968e4f87e
Support fsgsbase at compile time.
2021-07-31 10:12:59 +02:00
d3cf4db47b
Prerequisites for risv64 support
2021-05-12 20:23:53 -06:00
558109a9cb
Put the KPCRs in high memory, in their own PML4.
...
This also removes the need to do another semi-expensive remap when
cloning processes, since the KPCRs (for kernel TLS) are no longer stored
in the user PML4.
2021-05-06 21:21:37 +02:00
3eacbdda2a
Remove unnecessary exit() diff.
2021-05-06 19:49:58 +02:00
fd97fa80bb
Move pid back to inner scope in exit() handler.
2021-05-06 19:49:58 +02:00
9b4ce0d0cc
WIP: Fix userspace ACPI shutdown.
2021-05-06 19:49:58 +02:00
8d61c79b23
Use RMM TableKind and fix x86_64 compilation
2021-05-03 21:15:46 -06:00
2aa4d8caf5
Merge remote-tracking branch 'origin/aarch64-rebase' into riscv64
2021-05-03 20:52:59 -06:00
17c261553b
Fixes for building x86_64
2021-05-03 20:43:18 -06:00
b9448274fc
Switch to using RMM PageFlags
2021-05-03 17:28:08 -06:00
dd0616cc8f
Use RMM for TLB flushing
2021-05-03 12:42:16 -06:00
d331f72f2a
Use UTF-8 for all paths
2021-04-28 20:06:07 -06:00
4lDO2