On 12/3/20 8:05 AM, Catalin Marinas wrote: > On Thu, Dec 03, 2020 at 07:36:10AM -0700, Jens Axboe wrote: >> On 12/3/20 4:01 AM, Catalin Marinas wrote: >>> On Thu, Dec 03, 2020 at 02:25:30PM +1100, Stephen Rothwell wrote: >>>> diff --cc arch/arm64/include/asm/thread_info.h >>>> index 015beafe58f5,cdcf307764aa..000000000000 >>>> --- a/arch/arm64/include/asm/thread_info.h >>>> +++ b/arch/arm64/include/asm/thread_info.h >>>> @@@ -63,7 -66,9 +63,8 @@@ void arch_release_task_struct(struct ta >>>> #define TIF_NOTIFY_RESUME 2 /* callback before returning to user */ >>>> #define TIF_FOREIGN_FPSTATE 3 /* CPU's FP state is not current's */ >>>> #define TIF_UPROBE 4 /* uprobe breakpoint or singlestep */ >>>> - #define TIF_MTE_ASYNC_FAULT 5 /* MTE Asynchronous Tag Check Fault */ >>>> -#define TIF_FSCHECK 5 /* Check FS is USER_DS on return */ >>>> ++#define TIF_NOTIFY_SIGNAL 5 /* signal notifications exist */ >>>> + #define TIF_MTE_ASYNC_FAULT 6 /* MTE Asynchronous Tag Check Fault */ >>>> -#define TIF_NOTIFY_SIGNAL 7 /* signal notifications exist */ >>>> #define TIF_SYSCALL_TRACE 8 /* syscall trace active */ >>>> #define TIF_SYSCALL_AUDIT 9 /* syscall auditing */ >>>> #define TIF_SYSCALL_TRACEPOINT 10 /* syscall tracepoint for ftrace */ >>>> @@@ -96,7 -103,8 +98,8 @@@ >>>> >>>> #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ >>>> _TIF_NOTIFY_RESUME | _TIF_FOREIGN_FPSTATE | \ >>>> - _TIF_UPROBE | _TIF_MTE_ASYNC_FAULT) >>>> - _TIF_UPROBE | _TIF_FSCHECK | _TIF_MTE_ASYNC_FAULT | \ >>>> ++ _TIF_UPROBE | _TIF_MTE_ASYNC_FAULT | \ >>>> + _TIF_NOTIFY_SIGNAL) >>> >>> Thanks Stephen. It looks alright to me. >> >> Agree - I'll rebase my tree when -rc7 is out so we won't have this issue once >> the 5.11 merge window opens. > > I don't think rebasing on -rc7 will help since the arm64 commit > b5a5a01d8e9a is queued for 5.11 (so not in -rc7). Ah indeed, I saw some changes come in yesterday for mainline and assumed it was those. > It shouldn't matter much, Linus likes the occasional conflict ;). > Anyway, I can wait for your pull request to go in if you'd prefer (and > if it happens in the first week of the merging window). Right, not an issue, it's a trivial resolve anyway. That branch is dependent on an x86/core branch, so I'll push it out when that goes in. But Linus usually pulls those early, so don't think we'll have much of an issue there. -- Jens Axboe
