You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Tree:
a4d056eddf
3096/hle/bufferc
3096/qcom/clamp
3096/spirv/warp
3096/vk/drain_pending_build
3096/vk/pipeline_compilation
Kernel
android/fix-gpu-driver-fetcher
android6
atomicops-mxwell
civa
descriptor
descriptor_pool_opt
descriptor_set
disable-vap
discfix
display-modified-settings-first
dmnt2
docsupdate1backckc
dynarm7345
dynarmic-coproc
dynarmic-docs-move-inline
dynarmic-ppc64
eden-orbis-ps4
eds-true-adreno-fixes
feature/dmnt
ffmpeg-cross-compile
fix-fibers-2
fix/discord-rpc
fix/no-duplicate-drivers
flatopsfixes23485
freebsd-cubeb
fs_external_dlcupdates
fuck-farenheit-units
install-vulkan-ps1-fix-windows-on-arm
interval-zero
lanobu
liz-coalesce-ptr-to-variant
liz-crash-dumps-solaris
liz-dynarmic-backport-waitpkg
liz-dynarmic-macos-fbsd-port
liz-get-rid-of-mcl-intrusive-list
liz-heaptrack-fix
liz-no-rtti-allowance
lock-term-1
macos-sqbuild
macroify-surface-stuffs
master
maxwell_dma
memsetopsyscallavoid
mike22-carousel-patch-set
mmap-fixews
mutliplayer-filter-better1
n64
nce-strx
nce_cpp
netgate1
netusejthreadstuff
no-d24
pagetable-clustering
pintocputhing
pipelinederivative
pt-fix-attempt
qcom-weird-vk-ftz
quick-fix
refactoreds2
release/0.0.3
release/0.0.4
rem-dup-applet-launch
remove-unused-fastmem-fallback
remredundaant-android-setts
reorder-menu-game-per-config
revert-2695
sampleshaderfract
scmfix-worktree
script/android-icons
selfhost0
showcase
showcase2
sjkdbsdfjkbsdf-2834
smartqueryreset
static-linux
static-vector-thread-list
stuffmadeforfun
sured-revert
test
test-donotmerge
test-revert-gpu-optim
test2
true-eds
true-eds-graphics
update-cpmutil
update-icons-android-11
video-core-improve-dma-logic
vk-fix-oom-force-maller-buffers
vk-surface-andpc
vulkan-thingy
woa-turnip-expr
xbzk-mci-bare-minimum-boot-fix
xbzk-saf-recursive-write-with-permission-request
0.0.0
0.0.1-pre-alpha
0.0.2-pre-alpha
test-tag1
test-tag2
v0.0.3
v0.0.3-rc1
v0.0.3-rc2
v0.0.3-rc3
v0.0.3.git
v0.0.4
v0.0.4-rc1
v0.0.4-rc2
v0.0.4-rc2.test
v0.0.4-rc2.test2
v0.0.4-rc3
v0.0.4-rc3.test1
v0.0.4-rc3.test2
v0.0.4.test
${ noResults }
* gdbstub: fix IsMemoryBreak() returning false while connected to client As a result, the only existing codepath for a memory watchpoint hit to break into GDB (InterpeterMainLoop, GDB_BP_CHECK, ARMul_State::RecordBreak) is finally taken, which exposes incorrect logic* in both RecordBreak and ServeBreak. * a blank BreakpointAddress structure is passed, which sets r15 (PC) to NULL * gdbstub: DynCom: default-initialize two members/vars used in conditionals * gdbstub: DynCom: don't record memory watchpoint hits via RecordBreak() For now, instead check for GDBStub::IsMemoryBreak() in InterpreterMainLoop and ServeBreak. Fixes PC being set to a stale/unhit breakpoint address (often zero) when a memory watchpoint (rwatch, watch, awatch) is handled in ServeBreak() and generates a GDB trap. Reasons for removing a call to RecordBreak() for memory watchpoints: * The``breakpoint_data`` we pass is typed Execute or None. It describes the predicted next code breakpoint hit relative to PC; * GDBStub::IsMemoryBreak() returns true if a recent Read/Write operation hit a watchpoint. It doesn't specify which in return, nor does it trace it anywhere. Thus, the only data we could give RecordBreak() is a placeholder BreakpointAddress at offset NULL and type Access. I found the idea silly, compared to simply relying on GDBStub::IsMemoryBreak(). There is currently no measure in the code that remembers the addresses (and types) of any watchpoints that were hit by an instruction, in order to send them to GDB as "extended stop information." I'm considering an implementation for this. * gdbstub: Change an ASSERT to DEBUG_ASSERT I have never seen the (Reg[15] == last_bkpt.address) assert fail in practice, even after several weeks of (locally) developping various branches around GDB. Only leave it inside Debug builds. |
7 years ago | |
|---|---|---|
| .. | ||
| gdbstub.cpp | gdbstub: Fix some bugs in IsMemoryBreak() and ServeBreak. Add workaround to let watchpoints break into GDB. (#4651) | 7 years ago |
| gdbstub.h | gdbstub: Replace PAddr alias with VAddr | 8 years ago |