Commit Graph

755 Commits

Author SHA1 Message Date
Patric Stout
b0d7cfaa3d Change: [Win32] don't allocate 192KiB of memory on the stack on crash (#11240)
Heap is out of the question, as it might be corrupted.
Allocating this much on stack is silly.

So instead, allocate virtual pages to write the information in.
2023-08-28 12:14:11 +02:00
Jonathan G Rennison
40f9b61b4c Crash log: Windows: Fix screenshot filename in crash dialog box 2023-08-27 22:07:28 +01:00
Patric Stout
d4312c59a4 Codechange: [MacOS] use backtrace() instead of our custom variant (#11233)
As mentioned in the comment, we only did it ourselves as we once
were compatible with versions before 10.5. But that time has long
gone. So let's update the code to a bit more modern approach.
2023-08-27 22:32:11 +02:00
Patric Stout
b00e483b0f Change: recover when possible from crashes during a crash (#11238) 2023-08-27 21:54:05 +02:00
Patric Stout
29a37c2e0b Change: [Win32] unify the crashlog handler with the other OSes (#11236)
While at it, make the crash text a bit more readable, and sync
this with MacOS.
2023-08-27 17:58:31 +00:00
Patric Stout
45cc3acaf5 Codechange: [Win32] unify AppendDecodedStacktrace and LogStacktrace (#11220)
AppendDecodedStacktrace was already doing the same as LogStacktrace,
just with a different name.
2023-08-20 18:57:01 +00:00
Patric Stout
352ac264a5 Remove: [Win32] module-list from crash.log (#11219)
Only Windows implemented this, and it opens the files to read them
to get a CRC. Doing this in a crash-handler is strange at best.

Lastly, nobody has actually ever used this information to come to
some sort of conclusion. The module-list is used in combination
with the crash.dmp, but this information is already embedded in
there.
2023-08-20 17:09:21 +00:00
Patric Stout
f120d2beb8 Add: use breakpad to create crash.dmp on MacOS / Linux too (#11202)
Normally only the Windows platform could create a crash.dmp, making
analysing crash-reports from MacOS / Linux rather tricky.
2023-08-20 17:16:08 +02:00
Patric Stout
8f6df242c4 Remove: [Win32] (pointer-only) stack trace in crash.log (#11211)
It only contains pointers, which nobody can decipher anyway.
So instead, just report "Not supported", like other targets do
when they can't print a sane stack trace.
2023-08-20 17:08:16 +02:00
Patric Stout
a5aebeb1b5 Remove: [Win32] register values in crash.log (#11212) 2023-08-20 16:59:22 +02:00
Jonathan G Rennison
5a28405ced Merge branch 'master' into jgrpp
# Conflicts:
#	src/company_gui.cpp
#	src/group_gui.cpp
#	src/newgrf.cpp
#	src/newgrf_debug_gui.cpp
#	src/saveload/saveload.cpp
2023-08-19 01:28:36 +01:00
Patric Stout
cd10ddf9ae Remove: last remnants of SunOS (#11210)
We have no evidence SunOS has been tried in the last few years,
and the code is awkward enough to just remove.
2023-08-19 00:52:58 +02:00
Jonathan G Rennison
79509b76a6 Crashlog: Unix: Decode REG_ERR for SIGSEGV 2023-08-14 15:58:00 +01:00
Jonathan G Rennison
cb0c8d5e38 Crashlog: Unix: Decode common si_code values for SIGSEGV 2023-08-14 11:24:05 +01:00
Jonathan G Rennison
e676eb1202 Crashlog: Unix: Include REG_ERR in x86 register dump 2023-08-14 11:24:05 +01:00
Jonathan G Rennison
1ec2ed937f Unix: Fix CMake detection and setup of sigaltstack 2023-08-14 00:20:04 +01:00
Jonathan G Rennison
508f871c5f Use core/format.hpp for fmt includes 2023-07-20 16:34:59 +01:00
Rubidium
eaae0bb5e7 Codechange: automatic adding of _t to (u)int types, and WChar to char32_t
for i in `find src -type f|grep -v 3rdparty/fmt|grep -v 3rdparty/catch2|grep -v 3rdparty/opengl|grep -v stdafx.h`; do sed 's/uint16& /uint16 \&/g;s/int8\([ >*),;[]\)/int8_t\1/g;s/int16\([ >*),;[]\)/int16_t\1/g;s/int32\([ >*),;[]\)/int32_t\1/g;s/int64\([ >*),;[]\)/int64_t\1/g;s/ uint32(/ uint32_t(/g;s/_uint8_t/_uint8/;s/Uint8_t/Uint8/;s/ft_int64_t/ft_int64/g;s/uint64$/uint64_t/;s/WChar/char32_t/g;s/char32_t char32_t/char32_t WChar/' -i $i; done
2023-07-19 19:30:14 +02:00
Jonathan G Rennison
1e5bc2c12b Merge branch 'master' into jgrpp 2023-07-04 00:26:00 +01:00
Jonathan G Rennison
ead18b2af2 Merge branch 'master' into jgrpp
# Conflicts:
#	CMakeLists.txt
#	src/3rdparty/md5/md5.h
#	src/3rdparty/squirrel/squirrel/squtils.h
#	src/animated_tile.cpp
#	src/console_func.h
#	src/core/CMakeLists.txt
#	src/core/container_func.hpp
#	src/core/smallstack_type.hpp
#	src/crashlog.cpp
#	src/crashlog.h
#	src/debug.h
#	src/economy.cpp
#	src/gamelog.cpp
#	src/industry_gui.cpp
#	src/lang/catalan.txt
#	src/misc_gui.cpp
#	src/network/network_content.h
#	src/newgrf.cpp
#	src/newgrf.h
#	src/newgrf_config.cpp
#	src/newgrf_config.h
#	src/newgrf_gui.cpp
#	src/os/unix/font_unix.cpp
#	src/os/windows/crashlog_win.cpp
#	src/rail_cmd.cpp
#	src/saveload/animated_tile_sl.cpp
#	src/script/api/script_tilelist.cpp
#	src/settings.cpp
#	src/settingsgen/settingsgen.cpp
#	src/sl/oldloader_sl.cpp
#	src/station.cpp
#	src/station_cmd.cpp
#	src/stdafx.h
#	src/strgen/strgen.cpp
#	src/strgen/strgen_base.cpp
#	src/table/settings/gui_settings.ini
#	src/train_gui.cpp
#	src/vehicle.cpp
#	src/vehicle_base.h
#	src/vehicle_cmd.cpp
#	src/vehicle_gui_base.h
#	src/viewport_sprite_sorter.h
2023-07-02 12:02:36 +01:00
Jonathan G Rennison
d09b504bc5 Merge branch 'master' into jgrpp
# Conflicts:
#	.github/workflows/ci-build.yml
#	.github/workflows/release-linux.yml
#	.github/workflows/release-macos.yml
#	.github/workflows/release-source.yml
#	.github/workflows/release.yml
#	CMakeLists.txt
#	COMPILING.md
#	src/ai/ai_core.cpp
#	src/ai/ai_gui.cpp
#	src/bridge_gui.cpp
#	src/company_gui.cpp
#	src/console_cmds.cpp
#	src/core/CMakeLists.txt
#	src/core/smallmap_type.hpp
#	src/disaster_vehicle.h
#	src/effectvehicle_base.h
#	src/fontcache.cpp
#	src/game/game_core.cpp
#	src/game/game_gui.cpp
#	src/gamelog.cpp
#	src/gamelog_internal.h
#	src/group_gui.cpp
#	src/linkgraph/linkgraph.h
#	src/misc.cpp
#	src/network/core/config.h
#	src/network/core/udp.cpp
#	src/network/network_chat_gui.cpp
#	src/network/network_content_gui.cpp
#	src/network/network_gui.cpp
#	src/newgrf.cpp
#	src/newgrf_gui.cpp
#	src/newgrf_profiling.cpp
#	src/newgrf_profiling.h
#	src/object_gui.cpp
#	src/openttd.cpp
#	src/openttd.h
#	src/order_gui.cpp
#	src/os/windows/font_win32.cpp
#	src/rail_gui.cpp
#	src/road.cpp
#	src/road_gui.cpp
#	src/saveload/afterload.cpp
#	src/saveload/saveload.h
#	src/script/api/script_controller.cpp
#	src/script/api/script_roadtypelist.cpp
#	src/script/script_config.cpp
#	src/script/script_config.hpp
#	src/script/script_instance.cpp
#	src/script/script_scanner.cpp
#	src/script/squirrel.cpp
#	src/script/squirrel_helper.hpp
#	src/settings_gui.cpp
#	src/settings_internal.h
#	src/settings_type.h
#	src/table/settings/network_private_settings.ini
#	src/timetable_gui.cpp
#	src/vehicle.cpp
#	src/vehicle_base.h
#	src/window_gui.h
2023-07-01 02:42:51 +01:00
Rubidium
d5c75bd655 Codechange: use std::string instead of a temporary buffer for iconv calls 2023-06-30 19:15:21 +02:00
Rubidium
69d5b9d326 Cleanup: unused NO_THREADS #ifdefs 2023-06-28 19:07:10 +02:00
PeterN
8465559488 Codechange: Pass face index as font os_handle for FreeType fonts. (#11073)
This allows fallback font detection to test the specific face within the
font rather instead of only the first.
2023-06-27 12:30:46 +01:00
Rubidium
714ff35e9d Codechange: move to std::string over stredup + free 2023-06-18 22:55:15 +02:00
Jonathan G Rennison
551d581215 Unix crash log: Fix stack trace formatting when using addr2line 2023-06-18 20:17:38 +01:00
Rubidium
09f585b772 Codechange: use std::string over stredup+free for splitting font name 2023-06-18 18:32:20 +02:00
Jonathan G Rennison
a7a8f4c695 Windows crash log: Don't bother writing raw stack if it can be decoded
The raw values are almost always useless
2023-06-18 15:46:13 +01:00
Jonathan G Rennison
bcab44dc98 Windows crash log: Fix stack overflow in crash dialog window setup
Excessive stack use for crash log text conversion buffers
2023-06-18 15:38:33 +01:00
Jonathan G Rennison
d211ef557a Windows crash log: Handle crash log faults (MSVC) 2023-06-18 15:38:26 +01:00
Owen Rudge
1a3d1bca59 Remove: OS/2 port 2023-06-16 14:29:14 +01:00
Jonathan G Rennison
8c507d6611 Crash log: Suppress messages from internal faults 2023-06-13 00:57:05 +01:00
Jonathan G Rennison
ad995cbebc Windows crash log: Use TerminateProcess instead of ExitProcess 2023-06-13 00:57:05 +01:00
Jonathan G Rennison
940656bb6c Windows crash log: Handle crash log faults (MinGW) 2023-06-13 00:57:05 +01:00
Jonathan G Rennison
b26a3fa41e Crash log: Write out crash log incrementally on Windows, MacOS 2023-06-13 00:57:05 +01:00
Jonathan G Rennison
76304c1f76 Crash log: Checkpoint each line of decoded stack trace 2023-06-12 19:03:37 +01:00
Jonathan G Rennison
da07c8ce51 Crash log: Move GDB/LLDB logging to its own section 2023-06-12 18:38:27 +01:00
Jonathan G Rennison
dd57fc6ecf Unix crash log: Generalise crash log fault handling to all sections 2023-06-12 18:34:37 +01:00
Jonathan G Rennison
7a9cf87ff8 Windows: Add fallback for DoOSAbort implementation 2023-06-11 17:04:15 +01:00
Jonathan G Rennison
417abcccff Windows: Handle abort exception in VectoredExceptionHandler
To make sure that it is definitely handled
2023-06-11 16:47:01 +01:00
Jonathan G Rennison
cca7c0ef0a Windows: Don't call abort, use RaiseException directly 2023-06-11 16:46:12 +01:00
Jonathan G Rennison
cb6f6f94e6 Fix string handling error in OTTDStringCompare Windows implementation
See: #542
2023-06-11 15:26:54 +01:00
Patric Stout
2618179664 Fix: Wayland crash on startup due to Pango also using FontConfig (#10916)
Basically, we haven't been a good neighbour. Turns out you shouldn't
actually call FcFini when you are done, as some library might still
want to use FontConfig. And they use a shared instance for their
administration.

The idea is that you call FcInit once, and use FcConfigReference
after that to get an instance, you can release. This entry is
ref-counted, and things happen automatically based on that.

At least, I think.

(cherry picked from commit cd751a598a)
2023-06-10 16:42:22 +01:00
PeterN
8034609368 Fix: Ask FontConfig for the face index when opening fonts. (#10878)
This allows selection of the correct face in truetype fonts containing
multiple faces.

(cherry picked from commit 876871157f)
2023-06-10 16:42:22 +01:00
Jonathan G Rennison
ac2f9a21e8 Move upstream saveload to src/saveload/, move jgrpp saveload to src/sl/
Leave afterload in src/saveload/
2023-06-08 00:01:40 +01:00
Rubidium
b2a8d8aea4 Codechange: use std::string for Windows' ISO code mangling 2023-06-07 08:27:18 +02:00
PeterN
f249715db3 Fix #10964: Missing FontConfig format for language. (#10965) 2023-06-06 22:38:11 +00:00
PeterN
eda3defcb5 Codechange: Pass language for font detection as std::string. (#10964) 2023-06-06 20:55:56 +00:00
PeterN
7a0b6b7ddf Codechange: Remove Freetype bits on Win32/OSX. (#10960)
These platforms have their own specific font rendering.
2023-06-06 19:43:27 +01:00
Patric Stout
cd751a598a Fix: Wayland crash on startup due to Pango also using FontConfig (#10916)
Basically, we haven't been a good neighbour. Turns out you shouldn't
actually call FcFini when you are done, as some library might still
want to use FontConfig. And they use a shared instance for their
administration.

The idea is that you call FcInit once, and use FcConfigReference
after that to get an instance, you can release. This entry is
ref-counted, and things happen automatically based on that.

At least, I think.
2023-06-03 19:09:02 +00:00