Commit graph

4 commits

Author SHA1 Message Date
Joachim Henze 0d16cd41f1 [0.4.7][NTOS][WIN32SS][UDFS] Fix double free in ObCreateObject, fix retvals CORE-14271
Backport the following commits:
0.4.15-dev-6401-g 53b30e3f3b [NTOSKRNL][NTGDI] Formatting addendum, no functional change
0.4.15-dev-5487-g e7bbbf049e [NTOS] Fix double free on allocation failure in ObCreateObject
partially 0.4.13-dev-358-g 38db074491 [WIN32SS][UDFS] Misc addendum to CORE-14271 (#1529)
0.4.12-dev-648-g 17af7f0c27 [UDFS] Correctly check SeSinglePrivilegeCheck() return value (#1324) CORE-14271
0.4.12-dev-552-g 0483a5a380 [NTOS:OB] Correctly check SeSinglePrivilegeCheck() return value (#1323) CORE-14271
partially 0.4.10-dev-528-g 2c909db295 [NTOS:OB] Remove duplicated ObpTypeObjectType from ob.h
0.4.8-dev-954-g 862b82f3fa [WIN32K:NTGDI] Fix more uses of RtlCreateUnicodeString. CORE-14271
0.4.8-dev-949-g 7f687154a0 [WIN32K:NTGDI] Correctly interpret RtlCreateUnicodeString return value. CORE-14271
0.4.8-dev-948-g 6e750d189a [NTOS:CM] Correctly interpret RtlCreateUnicodeString return value. CORE-14271

Just for verifying that the size didn't increase anywhere:

ntoskrnl.exe GCC8.4.0dbg RosBEWin2.2.2 master               5.124.096
ntoskrnl.exe GCC4.7.2dbg RosBEWin2.1.6 0.4.14  4.413.440 -> 4.413.440
ntoskrnl.exe GCC4.7.2dbg RosBEWin2.1.6 0.4.13  4.367.360 -> 4.367.360
ntoskrnl.exe GCC4.7.2dbg RosBEWin2.1.6 0.4.12  4.377.600 -> 4.377.600
ntoskrnl.exe GCC4.7.2dbg RosBEWin2.1.6 0.4.11  4.377.088 -> 4.377.088
ntoskrnl.exe GCC4.7.2dbg RosBEWin2.1.6 0.4.10  4.344.320 -> 4.344.320
ntoskrnl.exe GCC4.7.2dbg RosBEWin2.1.6 0.4. 9  4.311.552 -> 4.311.552
ntoskrnl.exe GCC4.7.2dbg RosBEWin2.1.6 0.4. 8  4.296.704 -> 4.296.704
ntoskrnl.exe GCC4.7.2dbg RosBEWin2.1.6 0.4. 7  4.231.168 -> 4.231.168

udfs.sys     GCC8.4.0dbg RosBEWin2.2.2 master             835.584
udfs.sys     GCC4.7.2dbg RosBEWin2.1.6 0.4.14  745.472 -> 745.472
udfs.sys     GCC4.7.2dbg RosBEWin2.1.6 0.4.13  745.472 -> 745.472
udfs.sys     GCC4.7.2dbg RosBEWin2.1.6 0.4.12  749.568 -> 749.568
udfs.sys     GCC4.7.2dbg RosBEWin2.1.6 0.4.11  749.568 -> 749.568
udfs.sys     GCC4.7.2dbg RosBEWin2.1.6 0.4.10  749.568 -> 749.568
udfs.sys     GCC4.7.2dbg RosBEWin2.1.6 0.4. 9  749.568 -> 749.568
udfs.sys     GCC4.7.2dbg RosBEWin2.1.6 0.4. 8  749.568 -> 749.568
udfs.sys     GCC4.7.2dbg RosBEWin2.1.6 0.4. 7  749.568 -> 749.568

user32.dll   GCC8.4.0dbg RosBEWin2.2.2 master               1.585.152
user32.dll   GCC4.7.2dbg RosBEWin2.1.6 0.4.14  1.448.448 -> 1.448.448
user32.dll   GCC4.7.2dbg RosBEWin2.1.6 0.4.13  1.445.376 -> 1.445.376
user32.dll   GCC4.7.2dbg RosBEWin2.1.6 0.4.12  1.455.616 -> 1.455.616
user32.dll   GCC4.7.2dbg RosBEWin2.1.6 0.4.11  1.453.056 -> 1.453.056
user32.dll   GCC4.7.2dbg RosBEWin2.1.6 0.4.10  1.434.624 -> 1.434.624
user32.dll   GCC4.7.2dbg RosBEWin2.1.6 0.4. 9  1.422.336 -> 1.422.336
user32.dll   GCC4.7.2dbg RosBEWin2.1.6 0.4. 8  1.421.824 -> 1.421.824
user32.dll   GCC4.7.2dbg RosBEWin2.1.6 0.4. 7  1.418.752 -> 1.418.752

win32k.sys   GCC8.4.0dbg RosBEWin2.2.2 master               3.477.504
win32k.sys   GCC4.7.2dbg RosBEWin2.1.6 0.4.14  2.904.064 -> 2.904.064
win32k.sys   GCC4.7.2dbg RosBEWin2.1.6 0.4.13  2.895.872 -> 2.895.872
win32k.sys   GCC4.7.2dbg RosBEWin2.1.6 0.4.12  2.887.680 -> 2.887.680
win32k.sys   GCC4.7.2dbg RosBEWin2.1.6 0.4.11  2.867.200 -> 2.867.200
win32k.sys   GCC4.7.2dbg RosBEWin2.1.6 0.4.10  2.863.104 -> 2.863.104
win32k.sys   GCC4.7.2dbg RosBEWin2.1.6 0.4. 9  2.834.432 -> 2.834.432
win32k.sys   GCC4.7.2dbg RosBEWin2.1.6 0.4. 8  2.830.336 -> 2.830.336
win32k.sys   GCC4.7.2dbg RosBEWin2.1.6 0.4. 7  2.830.336 -> 2.830.336
2023-08-02 01:01:40 +02:00
Joachim Henze b4f5df6f4e [0.4.7][FS][SETUP] During 1st stage, disable any FS we don't support install to
... in order to let RawFS these volumes.
This avoids bloated setup and avoids potential bugs in 3rd party drivers.
For now, this commit doesn't change anything, but once IopParseDevice hack
gets removed, this will make a difference!

These commits were part of CORE-6305 but that is not the reason why I picked them,
but because moving those drivers into the cabs reduces the iso file-size a bit
and they reduce a bit of log-spam in bootcds 1st stage setup.

by picking the 2 commits:
0.4.8-dev-442-g 5a650f6ba5
0.4.8-dev-438-g 806cd16328
2022-02-26 18:45:34 +01:00
Joachim Henze 1c3ae3043a [0.4.7][UDFS] Fix crash on boot in release builds with UDFS removable media inserted (#4061) CORE-17598
Changes to specific files and their effects are as follows:
create.cpp - Allows booting past second stage with UDFS media inserted without BSOD
close.cpp - Allows shutdown without hang
dircntrl.cpp - Allows New Hardware Wizard not to hang on initial third phase install

It began to crash with the introduction of the UDFS driver in SVN r74901 == git 3a104c8f20

Fix picked from 0.4.15-dev-3326-g a91f5e8e4d
2021-11-14 21:23:05 +01:00
Colin Finck c2c66aff7d Git conversion: Make reactos the root directory, move rosapps, rostests, wallpapers into modules, and delete rossubsys. 2017-10-03 07:45:34 +00:00