Re: Detecting OS loader lock

If this were that easy, why was "TryEnterCriticalSection" ever programmed?
For the same reason it would be very nice (yet years too late!) to either
get the loader lock's state, or a "try" version of critical APIs like
GetModuleHandle().

Exactly, this would at least make it possible to "fail gracefully", instead
of ending up in a deadlock - which potentially could go unnoticed for quite
a while until your customer wonders why nothing is happening (Kinda "no
entries in the event log, the service runs, everything looks ok -> hmm why
do no orders arrive?" situation)

Best regards,
Christoph


.

Relevant Pages

  • Re: Enumerating 32-bit modules from 32-bit processes in WOW64
    ... I seem to recall the behavior of the loader has always been as described ... for a reason that would take a long time ... > the original IA-64 release? ...
    (microsoft.public.win32.programmer.kernel)
  • Re: Please revert git commit 1ad3dcc0
    ... The only reason was a failed LTP testcase which fills up the FD table and then called exec. ... loader and MISC_FMT_OPEN_BINARY, and the ELF loader _does_ actually do it for the case. ... there wasn't any reason to return EMFILE. ... If the spec did require it, then that would be an argument that the LTP testcase is valid, and for keeping the original patch. ...
    (Linux-Kernel)
  • Re: [ANNOUNCE] Release Digsig 1.5: kernel module for run-timeauthentication of binaries
    ... you don't sign nor need to sign perl or bash scripts. ... Why would a loader ... be written in ELF itself? ... There's absolutely no reason for that. ...
    (Linux-Kernel)
  • Re: PXE Loader register dump
    ... > Building the boot loader arguments ... > Relocating the loader and the BTX ... the BIOS executed a breakpoint for some reason: ... like it is supposed to just keep going when it hits a breakpoint. ...
    (freebsd-current)
  • Re: PXE Loader register dump
    ... > Building the boot loader arguments ... > Relocating the loader and the BTX ... the BIOS executed a breakpoint for some reason: ... like it is supposed to just keep going when it hits a breakpoint. ...
    (freebsd-current)