hbirr@svn.reactos.com wrote:
- Set the IOSB on error for async irp's with a sync FO (fix bug #609).
Clearly, the problem is that someone is reading the Iosb instead of reading the FileObject->FinalStatus. Do you know who is doing that? Fixing the caller would fix the bug, instead of adding the hack into the IRP Code.
Thanks for the commit, Best regards, Alex Ionescu
Alex Ionescu wrote:
hbirr@svn.reactos.com wrote:
- Set the IOSB on error for async irp's with a sync FO (fix bug
#609).
Do you know who is doing that? Fixing the caller would fix the bug, instead of adding the hack into the IRP Code.
Most of the places in io/file.c are doing it. I've been slowly fixing some of them, but there are still many reading the IoStatus Event even though they are waiting on the file object (Which is incorrect).
I'm slowly fixing them and then the hack can go away.
Best regards, Alex Ionescu