In the Linux kernel, the following vulnerability has been resolved: mm: don't spin in addstackrecord when gfp flags don't allow syzbot was able to find the following path: addstackrecordtolist mm/pageowner.c:182 [inline] incstackrecordcount mm/pageowner.c:214 [inline] _setpageowner+0x2c3/0x4a0 mm/pageowner.c:333 setpageowner include/linux/pageowner.h:32 [inline] postallochook+0x240/0x2a0 mm/pagealloc.c:1851 prepnewpage mm/pagealloc.c:1859 [inline] getpagefromfreelist+0x21e4/0x22c0 mm/pagealloc.c:3858 allocpagesnolocknoprof+0x94/0x120 mm/pagealloc.c:7554 Don't spin in addstackrecordtolist() when it is called from *_nolock() context.