aboutsummaryrefslogtreecommitdiff
path: root/test (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Add transform-dis-cases-arm64.S and fix a serious bug thus discovered.comex2015-02-232-0/+48
| | | | Commit transform-dis-cases-x86_64.S, which I forgot to earlier.
* Ban calls within transform regions in threadsafe mode.comex2015-02-233-4/+25
|
* Fix vm_remap, of trouble with which the previous diagnosis was completely ↵comex2015-02-231-0/+11
| | | | | | | | | | | incorrect. It's actually the fact that no vm_map_server actually exists on the kernel side - even though the otherwise useless function vm_remap exists in the kernel source, and mig generates server code for it, apparently that server code isn't built or linked into the kobject table. Bizarre. Anyway, switch to mach_vm_remap, which is what vm_remap in userspace calls into.
* various fixescomex2015-02-181-1/+1
|
* Fix hook-function:comex2015-02-182-12/+21
| | | | | | | | | | - Thread stoppage is now complemented by sigaction to catch injected threads (sigaction is not used exclusively because the rest of the program could be trying to use sigaction itself in the meantime - this is a real thing, ask Dolphin) - mprotect is no longer used due to max_protection possibly getting in the way; instead, a copy is created and mapped onto the original.
* fix test casescomex2015-02-141-4/+6
|
* Trampoline fixes.comex2015-02-142-11/+25
| | | | | | | | | | | | | The transformed code was incorrect because it assumed the pointer it was writing to was where the code would execute, but it was actually 'rewritten_temp'. Changed transform_dis_main to take a pc_trampoline pointer, which also helps the test harness. However, this means that it has to be called after the trampoline has been allocated, while before the trampoline allocation depended on the generated size; this change doesn't bother to use two passes or anything, but just allocates a new code buffer if the maximum possible size isn't available - not the end of the world, since trampoline_ptr will still only be increased by the actual size before the next hook in the series (if any).
* test harnesscomex2015-02-092-15/+151
|
* morecomex2015-02-092-3/+3
|
* Fix other architectures compiling - haven't tried *running* anything..comex2015-02-081-5/+5
|
* Refactor disassembly so x86 works, and add x86 transform-dis.comex2015-02-081-27/+14
| | | | | | | | | | This patch is a monolithic mess, because I was too lazy to do the refactor first (that would require some stash fun, since I wasn't actually sure before doing x86 transform-dis what would be needed). Anyway, the resulting code should be cleaner - less duplication. This breaks ARM/ARM64.
* starting x86 supportcomex2015-02-081-26/+61
|
* fix ARM64 too, theoreticallycomex2015-02-031-1/+1
|
* fix my utter failure to handle branches/conditionals correctly (on ARM)comex2015-02-013-8/+22
|
* remove GNU-isms from cbitcomex2015-01-301-0/+16
|
* get rid of the unnecessary CoreFoundation dependencycomex2015-01-292-6/+26
|
* port some old codecomex2015-01-291-0/+49
|
* sortacomex2015-01-281-0/+9
|
* right, fix cleanup. (compared to the old system, this avoids busywaiting, ↵comex2015-01-252-3/+36
| | | | and properly frees the thread resources)
* this is dumbcomex2015-01-251-2/+20
|
* fix substrate-compat; check in testcomex2015-01-251-0/+62
|
* it compiles...comex2015-01-242-2/+2
|
* Add function to deal with mprotecting RW and back. A bit more complex than ↵comex2015-01-242-0/+38
| | | | | | the minimum would be... (and minor build fixes)
* stopping other threads.comex2015-01-211-0/+34
|
* minor cleanup - stop caring about thumb outside of arch-specific codecomex2015-01-212-2/+12
| | | | | (and fix makefile bug where files whose names contained 'arm' would always be ldid'd)
* Some cleanup.comex2015-01-211-1/+1
|
* woo, inject seems to work on all platformscomex2015-01-211-3/+3
|
* ...comex2015-01-201-0/+17
|
* progress on injectioncomex2015-01-201-4/+3
|
* more reorganization - move OS X/iOS specific stuff into its own directorycomex2015-01-191-1/+1
|
* ldrd/strd fix - we definitely need to know that in ldrd r0, r1, [pc], we ↵comex2015-01-171-1/+1
| | | | can't use r1 as scratch
* improve test; thumb2 fixescomex2015-01-174-4/+54
|
* IT, BX LR - untestedcomex2015-01-171-0/+1
|
* reorganizationcomex2015-01-171-1/+1
|
* interpose works; add testcomex2015-01-171-0/+35
|
* add an assertcomex2015-01-171-0/+2
|
* fixes and test - both tests work on all archscomex2015-01-171-0/+46
|
* imp forwarding workscomex2015-01-171-9/+34
|
* this is all wrong.comex2015-01-161-3/+12
| | | | | | | since we don't know whether it's stret, we don't mark the block stret. thus imp_implementationWithBlock doesn't do the stret/block swap, even though it would for a real block, so we can and must always assume the block is first
* and now for something completely different: assembly maybestret-IMPLcomex2015-01-161-0/+26
| | | | forwarding functions for atomicity
* jump dis - seemingly working(!)comex2015-01-164-25/+480
|
* add a test assembly file, not used yetcomex2015-01-131-0/+48
|
* these functions are actually pointless, since it's easier to change the ↵comex2015-01-121-3/+11
| | | | | | *offset* for actual pc-rel loads (and the rest are accounted for, albeit implicitly in the case of add pc) i might end up using the stack :/
* minor improvementscomex2015-01-121-0/+1
|
* improve testcomex2015-01-121-4/+4
|
* that was easycomex2015-01-111-0/+1
|
* thumb2 and stuffcomex2015-01-111-7/+13
|
* thumb2 progresscomex2015-01-111-5/+11
|
* this produces good output (finally) but takes 3.25s to compile 24K of code, lolcomex2015-01-111-18/+23
|
* Revert "C++ is just a giant pain in the ass"comex2015-01-112-9/+7
| | | | This reverts commit 4743027a22ae4e622d7e78f878eff307d02ac373.
Copyright (C) 2019-2021 Reiko Asakura. All Rights Reserved.