Commit Graph

12 Commits

Author SHA1 Message Date
Lakr
5388e0c9c5 Squash merge startup-hang-fix into main
Prefix research patch comparison doc and normalize root markdown names

Rename research root markdown files to scoped topic names
2026-03-06 02:42:12 +08:00
Lakr
ba52c2274f skip C22 2026-03-05 02:08:19 +08:00
Lakr
f16a1eed37 squash: merge work-jb-next into main 2026-03-05 01:58:35 +08:00
Lakr
c9fd521659 Fix C24: patch_kcall10 sysent table base, chained fixup encoding, PAC signing
Three bugs caused NOT_BOOT (timeout):

1. Wrong sysent table base: first _nosys match is entry 428, not entry 0.
   Entry 0 is the indirect syscall handler. Fixed with backward scan.

2. Raw VA written to chained fixup pointer slot: struct.pack("<Q", cave_va)
   corrupts the fixup chain from sysent[439] onward. Fixed with proper
   auth rebase encoding (_encode_chained_auth_ptr).

3. Missing PAC parameters: dispatch uses BLRAA X8, X17 with X17=0xBCAD.
   Chained fixup must encode diversity=0xBCAD, key=IA, addrDiv=0.
   Chain 'next' field preserved from original entry.

Boot-tested OK via testing ramdisk.
2026-03-04 21:41:44 +08:00
Lakr
894c2d1551 Fix C23: vnode_getattr string anchor resolved to wrong function (AppleImage4)
Root cause: find_string("vnode_getattr") matched "%s: vnode_getattr: %d"
format string inside an AppleImage4 function. The old code then took that
function as vnode_getattr itself, causing BL to call into AppleImage4 with
wrong arguments → PAC failure on indirect branch at a2+48.

Fix: _find_vnode_getattr_via_string() now scans backward from the string
ref for a BL instruction and extracts its target — the real vnode_getattr
(sub_FFFFFE0007CCD1B4 at foff 0xCC91B4).

Bisection confirmed: variants A (stack frame) and B (+ tpidr_el1) boot OK,
variant C (+ BL vnode_getattr) panics with old resolution, boots OK with fix.

Boot-tested: full C23 patch with corrected vnode_getattr — BOOTS OK.
2026-03-04 21:21:23 +08:00
Lakr
3879c703e1 Update patch_hook_cred_label_update_execve.md 2026-03-04 21:00:23 +08:00
Lakr
218d3e0db6 Update patch_hook_cred_label_update_execve.md 2026-03-04 20:59:33 +08:00
Lakr
254a70d210 docs: C24 kcall10 analysis — chained fixup pointer mismatch identified
Sysent table uses arm64e chained fixup pointers but patcher writes raw VAs.
Also missing _munge_wwwwwwww and potential PAC issues with BLR X16.
2026-03-04 20:59:27 +08:00
Lakr
5655fd78d7 Fix B8 convert_port_to_map: was patching PAC check instead of kernel_map guard
Root cause: backward branch search found PAC validation B.EQ (target fell
within error region range) instead of the kernel_map guard B.NE. New approach
walks backward from panic string ADRP to find CMP+B.cond where the branch
goes forward past the panic, then makes it unconditional. Also enables B6
and B8 in find_all(). Boot-tested: PASS.
2026-03-04 20:54:16 +08:00
Lakr
4cfeca3a7e Fix A2: rewrite AMFI execve kill path patch to target shared epilogue
Old approach patched vnode-type assertion BLs (CBZ→panic). New approach
scans backward from function end for the shared MOV W0,#1 kill return
before the LDP x29,x30 epilogue and changes it to MOV W0,#0. Single
instruction converts all 5+ kill paths to success. Boot-tested OK.
2026-03-04 20:43:36 +08:00
Lakr
a3756e29a2 Fix B6 proc_security_policy: was stubbing copyio instead of real target
Root cause: "most-called BL target" heuristic in _proc_info picked copyio
(4 calls, 0x28C bytes) over the real _proc_security_policy (2 calls,
0x134 bytes). Lowered size filter threshold from 0x300 to 0x200 to
correctly exclude utility functions like copyio. Boot-tested: PASS.
2026-03-04 20:42:58 +08:00
Lakr
352d9f8d74 docs: add per-patch kernel JB analysis notes 2026-03-04 20:05:23 +08:00