This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [win32] Fix suspend count handling


Pedro Alves escribió:
Lerele wrote:


I'd also like to ask you a question, concerning a comment from Pedro several messages back that has stayed around in my mind since then.


It's not related with this specific thread title, but since it's gdbserver/win32 related, I haven't found appropriate to open a new thread just for this simple question.

The issue is near the end of:
http://sourceware.org/ml/gdb-patches/2007-11/msg00041.html

It's about the fact of gdb win32-nat.c some time ago having the interrupt functionality similar to the one that has been recently implemented using SuspendThread (versus using DebugBreak kind of functions). Pedro commented back then that win32-nat.c did have sometime in the past a similar implementation [that must have been dropped].
Do you know/remember if it was dropped for a specific reason?



Humm, I guess I mistaked win32-nat.c for the winpdo-nat.c files on Apple's gdb:

See in:
http://www.opensource.apple.com/darwinsource/tarballs/other/gdb-186.1.tar.gz



In src/gdb-next/winpdo-nat.c, you'll see a SuspendThreads mechanism. The file header states 1996 as the latest copyright year, but they may have easilly failed to update it. I'm not sure if win32-nat ever had this mechanism, or if it was only added by Apple. Our cvs history only goes back till 1999.

I did say: "I *think* that if...". :-)


Sorry, I misunderstood "you'll see that once a similar method was used." as being an assertion.
Anyway there seems to be a reason not to SuspendThread, whether there was or wasn't such implementation in the past.


Leo.



Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]