https://bugs.winehq.org/show_bug.cgi?id=55710
--- Comment #21 from Dmitry Timoshkov dmitry@baikal.ru --- (In reply to Paul Gofman from comment #20)
One guess that actually there is an upstream change made the patch broken. Now looking at commit histroy, I'd try with 25db1c5d49dc339e9b5a25514c198a524bd05484 upstream patch reverted, to see if that fixes the problem with the Staging patch.
No, reverting this commit didn't help. It's probably worth to mention that the problem started with at least wine-staging 8.16, which is earler than that commit date.
Is it maybe possible to bisect upstream with only this Staging patch on top?
It would be necessary to figure out first approximate time frame for the bisection.
Looking at that now it seems like that should've broken the patch for sure but in a different way: it would stop catching syscalls from Win libs now loaded at high addresses, not sure how the opposite can happen.
$ ulimit -s 8192
$ echo $LD_PRELOAD
$ cat /proc/sys/vm/legacy_va_layout 0
I'll see if I could figure out what is at the addresses RIP points to.