* Per #2272 and #1877, MinGW has issues when delay loading libraries, but
it is possible to apply a workaround to alleviate them, by redefining
DECLSPEC_IMPORT before including the corresponding headers.
* This is a bit more tricky to accomplish for virtdisk, as MinGW's windows.h
header does include virtdisk.h on its own (rather than expect a formal
include as MSVC does), so we have to prevent the virtdisk.h inclusion
first, by defining a macro, and then apply our workaround.
* Per https://sourceforge.net/p/mingw-w64/mailman/mingw-w64-public/thread/ea87573f-65ea-44a2-b4bb-ca96c0a136ab%40akeo.ie/#msg58793876
we are hoping that this should be a temporary workaround and that the root
cause of the issue will be fixed in binutils.
* Closes#2513.
* As opposed to what we originally asserted, Microsoft did enact a blanket revocation
in SkuSiPolicy.p7b for all post 1703 up to 2305 Windows UEFI bootloaders.
* As a result, unconditionally copying SkuSiPolicy.p7b will result in media as recent
as Windows 11 22H2 (v1) being flagged as revoked, which we don't want to enforce as
long as Microsoft themselves haven't entered the enforcing phase of their Black
Lotus mitigation (currently planned for early 2024).
* Because of this, while we add some revocation detection for post 1703 bootloaders,
we set it to only go as far as 20H1 for now, which means that all post 20H1 Windows
10 media and all Windows 11 media will not yet be flagged by Rufus as revoked and
will still boot in a Secure Boot environment due to lack of an SkuSiPolicy.p7b.
* Ultimately, per #2244 we may look for a BOOTMGRSECURITYVERSIONNUMBER resource to
blanket revoke all post 1703 - pre 2305 Windows UEFI bootloaders.
* Also remove the now unused comdlg32 library from the linker.
* MinGW32's delay loading functionality is not yet up to par with MSVC's and especially, for
some libraries (wininet, virtdisk) attempting to delay load them simply crashes the runtime.
* This results in the MinGW32 version of the app crashing when selecting a Windows ISO, as we
will then try to mount the ISO using virtdisk to poke the build version. Note that this crash
does not happen with the MinGW64 version or with MSVC.
* Closes#2272.
* Also fix a Coverity warning in SaveImageThread() and improve the VHD saving code.
* We are seeing reports of access violation exceptions being generated
when looking for processes, with the App Store version.
* Since this is not critical code, add an SEH handler to ignore those.
* MSG_900+ will be used for Windows Store translation, so add them
and makes sure these get filtered out from embedded.loc.
* Also make sure we don't get a "Translated by:" in the English version
when compiled with VS2022.
* Also add Store screenshots and update listing.csv so that we can
autogenerate and upload a complete translation update to the store.
* Newer Intel and AMD CPUs have SSE extensions for SHA-1 and SHA-256 acceleration.
* Add new cpu.c/cpu.h sources to detect the extensions, and use them in checksum.c
if available.
* Acceleration code is taken from https://github.com/noloader/SHA-Intrinsics.
* Pick the version and build number directly from the install[.wim|.esd] XML index.
This forces us to mount the ISO during scan, but it's the only way to get an accurate Build number...
* Also drop linking to version.dll (along with the whole version.dll delay-loading shenanigans).
* Some "unofficial" Windows ISOs use a custom boot.wim that only includes the Setup
image at index 1, rather than at index 2, after the PE image, for official ISOs.
* Also refactor to add a long needed vhd.h header.
* Also fix a MinGW warning.
* Now that we can delay-load DLLs for both MinGW and MSVC, we can also remove
the direct DLL hook that was added into dwmapi.dll due to side loading and
revert to using a direct API call instead.
* This reverts part of e1d864f755.
* Also attempt to silence that damn Coverity warning.
* Now that we can delay-load DLLs for both MinGW and MSVC, we can remove the
cumbersome direct DLL hooks into wininet.dll (which is vulnerable to side
loading when not delay-loaded) and revert to using direct API calls instead.
* This reverts part of e1d864f755.
* Also attempt to silence a Coverity warning.
* This reverts much of commits f6ac559f4d and 1947266837
so that we call the Windows APIs directly again, while ensuring that, by the time we load the DLLs,
sideloading mitigation has already been applied by the application.
* This is a continuation of #1877, and should help prevent re-introducing side-loading issues when we
link against new libraries, as well as allow us to drop some of the manual DLL hooking we've been
doing to prevent it, to clean up the code.
* Note that this is a bit more complex than what the stackoverflow post suggests, because we need to
create delayloaded libs for both 32-bit and 64-bit, which use a different calling convention and
therefore need to use different .def files. So there's a lot of gymkhana involved, with Makefiles
and whatnot, to get us there.
* Also simplify the use of CM_Get_DevNode_Registry_PropertyA() in dev.c since recent versions of
MinGW now have support for it.
* Also fix 2 small issues in net.c (potential overflow) and format.c (memory leak).
* WinTrust.lib is responsible for the MSASN1.dll sideloading issue described in #1877,
so, since we only use it for WinVerifyTrustEx(), hook into that function manually.
* Closes#1877 for the MinGW side.
* Note that we will probably try to use the method suggested by @assarbad and documented at
https://stackoverflow.com/questions/1851267/mingw-gcc-delay-loaded-dll-equivalent/70416894#70416894
to try to put an end to the problem of DLL side loading.
* ef2ff7179d was supposed to apply delay loading to our DLLs, for all MSVC builds,
thereby preventing sideloading attacks, but the patch actually only set the DelayLoadDLLs
property for Debug builds and not Release builds, with the result that side loading could
still be triggered for the Release executables, as demonstrated in #1877.
* This patch therefore properly sets the DelayLoadDLLs for all builds, which should take care
of the side loading vulnerability at least for MSVC executables.
* A subsequent patch will still be needed for MinGW, since there is no equivalent to DelayLoadDLLs.
* This addresses part of #1877.
* Hypothetically if the user's current directory contains a malicious DLL that DLL
could be loaded instead of the one in System32.
* Whereas the previous patch should have taken care of the one DLL referenced by
Rufus that may be vulnerable to this attack (version.dll), we nonetheless add
delay loading for all the libraries we reference as a precautionary measure.
* One can confirm that this works by using dumpbin.exe /IMPORTS to make sure
a specific DLL is delay loaded. Then putting a breakpoint in the delay load
hook should also confirm that the hook is used.
* Closes#1838
* This is part of #1838, where we need to sort the version.dll sideloading problem for MinGW.
* A subsequent patch will be applied to MSVC, to more generally delay the loading of DLLs.
* Also fix a typo with an assert expression.
* Use sources/compatresources.dll, when available, to try to detect the Windows ISO version and build.
* Also report what facility we use for formatting.
Yes!!! We are finally *much* faster than 7-zip for SHA-256, even though
we are also computing MD5 and SHA-1 in parallel. Here are some averaged
comparative results, against the 5.71 GB Win10_20H2_EnglishInternational_x64.iso
(SHA-256 = 08535b6dd0a4311f562e301c3c344b4aefd2e69a82168426b9971d6f8cab35e1):
* Windows' PowerShell Get-FileHash: 48s
* 7-zip's SHA-256 : 31s
* Rufus (64-bit release version) : 23s
* Instead of x86_32 and x86_64.
* This should aid with our appxbundle creation and if Microsoft want to
be wholly incorrect in their arch designations, who am I to judge?...
* *THIS* is what you need to do to replace Microsoft's broken SetDllDirectory("")
implementation and mitigate DLL sideloading from local directories.
* Also fix some comment typos
