| Message ID | 20210605014216.446867-3-npiggin@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | shoot lazy tlbs | expand |
Excerpts from Nicholas Piggin's message of June 5, 2021 11:42 am: > Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm > when it is context switched. This can be disabled by architectures that > don't require this refcounting if they clean up lazy tlb mms when the > last refcount is dropped. Currently this is always enabled, which is > what existing code does, so the patch is effectively a no-op. > > Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. > > Signed-off-by: Nicholas Piggin <npiggin@gmail.com> Can I give you a couple of incremental patches for 2/4 and 3/4 to improve the implementation requirement comments a bit for benefit of other archs. Thanks, Nick -- Explain the requirements for lazy tlb mm refcounting in the comment, to help with archs that may want to disable this by some means other than MMU_LAZY_TLB_SHOOTDOWN. Signed-off-by: Nicholas Piggin <npiggin@gmail.com> --- arch/Kconfig | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/Kconfig b/arch/Kconfig index 1cff045cdde6..39d8c7dcf220 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -419,6 +419,16 @@ config ARCH_WANT_IRQS_OFF_ACTIVATE_MM shootdowns should enable this. # Use normal mm refcounting for MMU_LAZY_TLB kernel thread references. +# MMU_LAZY_TLB_REFCOUNT=n can improve the scalability of context switching +# to/from kernel threads when the same mm is running on a lot of CPUs (a large +# multi-threaded application), by reducing contention on the mm refcount. +# +# This can be disabled if the architecture ensures no CPUs are using an mm as a +# "lazy tlb" beyond its final refcount (i.e., by the time __mmdrop frees the mm +# or its kernel page tables). This could be arranged by arch_exit_mmap(), or +# final exit(2) TLB flush, for example. arch code must also ensure the +# _lazy_tlb variants of mmgrab/mmdrop are used when dropping the lazy reference +# to a kthread ->active_mm (non-arch code has been converted already). config MMU_LAZY_TLB_REFCOUNT def_bool y
On 6/4/21 6:42 PM, Nicholas Piggin wrote: > Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm > when it is context switched. This can be disabled by architectures that > don't require this refcounting if they clean up lazy tlb mms when the > last refcount is dropped. Currently this is always enabled, which is > what existing code does, so the patch is effectively a no-op. > > Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. I am in favor of this approach, but I would be a lot more comfortable with the resulting code if task->active_mm were at least better documented and possibly even guarded by ifdefs. x86 bare metal currently does not need the core lazy mm refcounting, and x86 bare metal *also* does not need ->active_mm. Under the x86 scheme, if lazy mm refcounting were configured out, ->active_mm could become a dangling pointer, and this makes me extremely uncomfortable. So I tend to think that, depending on config, the core code should either keep ->active_mm [1] alive or get rid of it entirely. [1] I don't really think it belongs in task_struct at all. It's not a property of the task. It's the *per-cpu* mm that the core code is keeping alive for lazy purposes. How about consolidating it with the copy in rq? I guess the short summary of my opinion is that I like making this configurable, but I do not like the state of the code. --Andy
Excerpts from Andy Lutomirski's message of June 9, 2021 2:20 am: > On 6/4/21 6:42 PM, Nicholas Piggin wrote: >> Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm >> when it is context switched. This can be disabled by architectures that >> don't require this refcounting if they clean up lazy tlb mms when the >> last refcount is dropped. Currently this is always enabled, which is >> what existing code does, so the patch is effectively a no-op. >> >> Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. > > I am in favor of this approach, but I would be a lot more comfortable > with the resulting code if task->active_mm were at least better > documented and possibly even guarded by ifdefs. active_mm is fairly well documented in Documentation/active_mm.rst IMO. I don't think anything has changed in 20 years, I don't know what more is needed, but if you can add to documentation that would be nice. Maybe moving a bit of that into .c and .h files? > x86 bare metal currently does not need the core lazy mm refcounting, and > x86 bare metal *also* does not need ->active_mm. Under the x86 scheme, > if lazy mm refcounting were configured out, ->active_mm could become a > dangling pointer, and this makes me extremely uncomfortable. > > So I tend to think that, depending on config, the core code should > either keep ->active_mm [1] alive or get rid of it entirely. I don't actually know what you mean. core code needs the concept of an "active_mm". This is the mm that your kernel threads are using, even in the unmerged CONFIG_LAZY_TLB=n patch, active_mm still points to init_mm for kernel threads. We could hide that idea behind an active_mm() function that would always return &init_mm if mm==NULL, but you still have the concept of an active mm and a pointer that callers must not access after free (because some cases will be CONFIG_LAZY_TLB=y). > [1] I don't really think it belongs in task_struct at all. It's not a > property of the task. It's the *per-cpu* mm that the core code is > keeping alive for lazy purposes. How about consolidating it with the > copy in rq? I agree it's conceptually a per-cpu property. I don't know why it was done this way, maybe it was just convenient and works well for mm and active_mm to be adjacent. Linus might have a better insight. > I guess the short summary of my opinion is that I like making this > configurable, but I do not like the state of the code. I don't think I'd object to moving active_mm to rq and converting all usages to active_mm() while we're there, it would make things a bit more configurable. But I don't see it making core code fundamentally less complex... if you're referring to the x86 mm switching monstrosity, then that's understandable, but I admit I haven't spent enough time looking at it to make a useful comment. A patch would be enlightening, I have the leftover CONFIG_LAZY_TLB=n patch if you were thinking of building on that I can send it to you. Thanks, Nick
On 6/13/21 5:45 PM, Nicholas Piggin wrote: > Excerpts from Andy Lutomirski's message of June 9, 2021 2:20 am: >> On 6/4/21 6:42 PM, Nicholas Piggin wrote: >>> Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm >>> when it is context switched. This can be disabled by architectures that >>> don't require this refcounting if they clean up lazy tlb mms when the >>> last refcount is dropped. Currently this is always enabled, which is >>> what existing code does, so the patch is effectively a no-op. >>> >>> Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. >> >> I am in favor of this approach, but I would be a lot more comfortable >> with the resulting code if task->active_mm were at least better >> documented and possibly even guarded by ifdefs. > > active_mm is fairly well documented in Documentation/active_mm.rst IMO. > I don't think anything has changed in 20 years, I don't know what more > is needed, but if you can add to documentation that would be nice. Maybe > moving a bit of that into .c and .h files? > Quoting from that file: - however, we obviously need to keep track of which address space we "stole" for such an anonymous user. For that, we have "tsk->active_mm", which shows what the currently active address space is. This isn't even true right now on x86. With your patch applied: To support all that, the "struct mm_struct" now has two counters: a "mm_users" counter that is how many "real address space users" there are, and a "mm_count" counter that is the number of "lazy" users (ie anonymous users) plus one if there are any real users. isn't even true any more. >> x86 bare metal currently does not need the core lazy mm refcounting, and >> x86 bare metal *also* does not need ->active_mm. Under the x86 scheme, >> if lazy mm refcounting were configured out, ->active_mm could become a >> dangling pointer, and this makes me extremely uncomfortable. >> >> So I tend to think that, depending on config, the core code should >> either keep ->active_mm [1] alive or get rid of it entirely. > > I don't actually know what you mean. > > core code needs the concept of an "active_mm". This is the mm that your > kernel threads are using, even in the unmerged CONFIG_LAZY_TLB=n patch, > active_mm still points to init_mm for kernel threads. Core code does *not* need this concept. First, it's wrong on x86 since at least 4.15. Any core code that actually assumes that ->active_mm is "active" for any sensible definition of the word active is wrong. Fortunately there is no such code. I looked through all active_mm references in core code. We have: kernel/sched/core.c: it's all refcounting, although it's a bit tangled with membarrier. kernel/kthread.c: same. refcounting and membarrier stuff. kernel/exit.c: exit_mm() a BUG_ON(). kernel/fork.c: initialization code and a warning. kernel/cpu.c: cpu offline stuff. wouldn't be needed if active_mm went away. fs/exec.c: nothing of interest I didn't go through drivers, but I maintain my point. active_mm is there for refcounting. So please don't just make it even more confusing -- do your performance improvement, but improve the code at the same time: get rid of active_mm, at least on architectures that opt out of the refcounting. > > We could hide that idea behind an active_mm() function that would always > return &init_mm if mm==NULL, but you still have the concept of an active > mm and a pointer that callers must not access after free (because some > cases will be CONFIG_LAZY_TLB=y). > >> [1] I don't really think it belongs in task_struct at all. It's not a >> property of the task. It's the *per-cpu* mm that the core code is >> keeping alive for lazy purposes. How about consolidating it with the >> copy in rq? > > I agree it's conceptually a per-cpu property. I don't know why it was > done this way, maybe it was just convenient and works well for mm and > active_mm to be adjacent. Linus might have a better insight. > >> I guess the short summary of my opinion is that I like making this >> configurable, but I do not like the state of the code. > > I don't think I'd object to moving active_mm to rq and converting all > usages to active_mm() while we're there, it would make things a bit > more configurable. But I don't see it making core code fundamentally > less complex... if you're referring to the x86 mm switching monstrosity, > then that's understandable, but I admit I haven't spent enough time > looking at it to make a useful comment. A patch would be enlightening, > I have the leftover CONFIG_LAZY_TLB=n patch if you were thinking of > building on that I can send it to you. > > Thanks, > Nick >
Excerpts from Andy Lutomirski's message of June 14, 2021 1:52 pm: > On 6/13/21 5:45 PM, Nicholas Piggin wrote: >> Excerpts from Andy Lutomirski's message of June 9, 2021 2:20 am: >>> On 6/4/21 6:42 PM, Nicholas Piggin wrote: >>>> Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm >>>> when it is context switched. This can be disabled by architectures that >>>> don't require this refcounting if they clean up lazy tlb mms when the >>>> last refcount is dropped. Currently this is always enabled, which is >>>> what existing code does, so the patch is effectively a no-op. >>>> >>>> Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. >>> >>> I am in favor of this approach, but I would be a lot more comfortable >>> with the resulting code if task->active_mm were at least better >>> documented and possibly even guarded by ifdefs. >> >> active_mm is fairly well documented in Documentation/active_mm.rst IMO. >> I don't think anything has changed in 20 years, I don't know what more >> is needed, but if you can add to documentation that would be nice. Maybe >> moving a bit of that into .c and .h files? >> > > Quoting from that file: > > - however, we obviously need to keep track of which address space we > "stole" for such an anonymous user. For that, we have "tsk->active_mm", > which shows what the currently active address space is. > > This isn't even true right now on x86. From the perspective of core code, it is. x86 might do something crazy with it, but it has to make it appear this way to non-arch code that uses active_mm. Is x86's scheme documented? > With your patch applied: > > To support all that, the "struct mm_struct" now has two counters: a > "mm_users" counter that is how many "real address space users" there are, > and a "mm_count" counter that is the number of "lazy" users (ie anonymous > users) plus one if there are any real users. > > isn't even true any more. Well yeah but the active_mm concept hasn't changed. The refcounting change is hopefully reasonably documented? > > >>> x86 bare metal currently does not need the core lazy mm refcounting, and >>> x86 bare metal *also* does not need ->active_mm. Under the x86 scheme, >>> if lazy mm refcounting were configured out, ->active_mm could become a >>> dangling pointer, and this makes me extremely uncomfortable. >>> >>> So I tend to think that, depending on config, the core code should >>> either keep ->active_mm [1] alive or get rid of it entirely. >> >> I don't actually know what you mean. >> >> core code needs the concept of an "active_mm". This is the mm that your >> kernel threads are using, even in the unmerged CONFIG_LAZY_TLB=n patch, >> active_mm still points to init_mm for kernel threads. > > Core code does *not* need this concept. First, it's wrong on x86 since > at least 4.15. Any core code that actually assumes that ->active_mm is > "active" for any sensible definition of the word active is wrong. > Fortunately there is no such code. > > I looked through all active_mm references in core code. We have: > > kernel/sched/core.c: it's all refcounting, although it's a bit tangled > with membarrier. > > kernel/kthread.c: same. refcounting and membarrier stuff. > > kernel/exit.c: exit_mm() a BUG_ON(). > > kernel/fork.c: initialization code and a warning. > > kernel/cpu.c: cpu offline stuff. wouldn't be needed if active_mm went away. > > fs/exec.c: nothing of interest I might not have been clear. Core code doesn't need active_mm if active_mm somehow goes away. I'm saying active_mm can't go away because it's needed to support (most) archs that do lazy tlb mm switching. The part I don't understand is when you say it can just go away. How? > I didn't go through drivers, but I maintain my point. active_mm is > there for refcounting. So please don't just make it even more confusing > -- do your performance improvement, but improve the code at the same > time: get rid of active_mm, at least on architectures that opt out of > the refcounting. powerpc opts out of the refcounting and can not "get rid of active_mm". Not even in theory. Thanks, Nick
Excerpts from Nicholas Piggin's message of June 14, 2021 2:14 pm: > Excerpts from Andy Lutomirski's message of June 14, 2021 1:52 pm: >> On 6/13/21 5:45 PM, Nicholas Piggin wrote: >>> Excerpts from Andy Lutomirski's message of June 9, 2021 2:20 am: >>>> On 6/4/21 6:42 PM, Nicholas Piggin wrote: >>>>> Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm >>>>> when it is context switched. This can be disabled by architectures that >>>>> don't require this refcounting if they clean up lazy tlb mms when the >>>>> last refcount is dropped. Currently this is always enabled, which is >>>>> what existing code does, so the patch is effectively a no-op. >>>>> >>>>> Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. >>>> >>>> I am in favor of this approach, but I would be a lot more comfortable >>>> with the resulting code if task->active_mm were at least better >>>> documented and possibly even guarded by ifdefs. >>> >>> active_mm is fairly well documented in Documentation/active_mm.rst IMO. >>> I don't think anything has changed in 20 years, I don't know what more >>> is needed, but if you can add to documentation that would be nice. Maybe >>> moving a bit of that into .c and .h files? >>> >> >> Quoting from that file: >> >> - however, we obviously need to keep track of which address space we >> "stole" for such an anonymous user. For that, we have "tsk->active_mm", >> which shows what the currently active address space is. >> >> This isn't even true right now on x86. > > From the perspective of core code, it is. x86 might do something crazy > with it, but it has to make it appear this way to non-arch code that > uses active_mm. > > Is x86's scheme documented? > >> With your patch applied: >> >> To support all that, the "struct mm_struct" now has two counters: a >> "mm_users" counter that is how many "real address space users" there are, >> and a "mm_count" counter that is the number of "lazy" users (ie anonymous >> users) plus one if there are any real users. >> >> isn't even true any more. > > Well yeah but the active_mm concept hasn't changed. The refcounting > change is hopefully reasonably documented? > >> >> >>>> x86 bare metal currently does not need the core lazy mm refcounting, and >>>> x86 bare metal *also* does not need ->active_mm. Under the x86 scheme, >>>> if lazy mm refcounting were configured out, ->active_mm could become a >>>> dangling pointer, and this makes me extremely uncomfortable. >>>> >>>> So I tend to think that, depending on config, the core code should >>>> either keep ->active_mm [1] alive or get rid of it entirely. >>> >>> I don't actually know what you mean. >>> >>> core code needs the concept of an "active_mm". This is the mm that your >>> kernel threads are using, even in the unmerged CONFIG_LAZY_TLB=n patch, >>> active_mm still points to init_mm for kernel threads. >> >> Core code does *not* need this concept. First, it's wrong on x86 since >> at least 4.15. Any core code that actually assumes that ->active_mm is >> "active" for any sensible definition of the word active is wrong. >> Fortunately there is no such code. >> >> I looked through all active_mm references in core code. We have: >> >> kernel/sched/core.c: it's all refcounting, although it's a bit tangled >> with membarrier. >> >> kernel/kthread.c: same. refcounting and membarrier stuff. >> >> kernel/exit.c: exit_mm() a BUG_ON(). >> >> kernel/fork.c: initialization code and a warning. >> >> kernel/cpu.c: cpu offline stuff. wouldn't be needed if active_mm went away. >> >> fs/exec.c: nothing of interest > > I might not have been clear. Core code doesn't need active_mm if > active_mm somehow goes away. I'm saying active_mm can't go away because > it's needed to support (most) archs that do lazy tlb mm switching. > > The part I don't understand is when you say it can just go away. How? > >> I didn't go through drivers, but I maintain my point. active_mm is >> there for refcounting. So please don't just make it even more confusing >> -- do your performance improvement, but improve the code at the same >> time: get rid of active_mm, at least on architectures that opt out of >> the refcounting. > > powerpc opts out of the refcounting and can not "get rid of active_mm". > Not even in theory. That is to say, it does do a type of reference management that requires active_mm so you can argue it has not entirely opted out of refcounting. But we're not just doing refcounting for the sake of refcounting! That would make no sense. active_mm is required because that's the mm that we have switched to (from core code's perspective), and it is integral to know when to switch to a different mm. See how active_mm is a fundamental concept in core code? It's part of the contract between core code and the arch mm context management calls. reference counting follows from there but it's not the _reason_ for this code. Pretend the reference problem does not exit (whether by refcounting or shootdown or garbage collection or whatever). We still can't remove active_mm! We need it to know how to call into arch functions like switch_mm. I don't know if you just forgot that critical requirement in your above list, or you actually are entirely using x86's mental model for this code which is doing something entirely different that does not need it at all. If that is the case I really don't mind some cleanup or wrapper functions for x86 do entirely do its own thing, but if that's the case you can't criticize core code's use of active_mm due to the current state of x86. It's x86 that needs documentation and cleaning up. Thanks, Nick
Excerpts from Nicholas Piggin's message of June 14, 2021 2:47 pm: > Excerpts from Nicholas Piggin's message of June 14, 2021 2:14 pm: >> Excerpts from Andy Lutomirski's message of June 14, 2021 1:52 pm: >>> On 6/13/21 5:45 PM, Nicholas Piggin wrote: >>>> Excerpts from Andy Lutomirski's message of June 9, 2021 2:20 am: >>>>> On 6/4/21 6:42 PM, Nicholas Piggin wrote: >>>>>> Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm >>>>>> when it is context switched. This can be disabled by architectures that >>>>>> don't require this refcounting if they clean up lazy tlb mms when the >>>>>> last refcount is dropped. Currently this is always enabled, which is >>>>>> what existing code does, so the patch is effectively a no-op. >>>>>> >>>>>> Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. >>>>> >>>>> I am in favor of this approach, but I would be a lot more comfortable >>>>> with the resulting code if task->active_mm were at least better >>>>> documented and possibly even guarded by ifdefs. >>>> >>>> active_mm is fairly well documented in Documentation/active_mm.rst IMO. >>>> I don't think anything has changed in 20 years, I don't know what more >>>> is needed, but if you can add to documentation that would be nice. Maybe >>>> moving a bit of that into .c and .h files? >>>> >>> >>> Quoting from that file: >>> >>> - however, we obviously need to keep track of which address space we >>> "stole" for such an anonymous user. For that, we have "tsk->active_mm", >>> which shows what the currently active address space is. >>> >>> This isn't even true right now on x86. >> >> From the perspective of core code, it is. x86 might do something crazy >> with it, but it has to make it appear this way to non-arch code that >> uses active_mm. >> >> Is x86's scheme documented? >> >>> With your patch applied: >>> >>> To support all that, the "struct mm_struct" now has two counters: a >>> "mm_users" counter that is how many "real address space users" there are, >>> and a "mm_count" counter that is the number of "lazy" users (ie anonymous >>> users) plus one if there are any real users. >>> >>> isn't even true any more. >> >> Well yeah but the active_mm concept hasn't changed. The refcounting >> change is hopefully reasonably documented? >> >>> >>> >>>>> x86 bare metal currently does not need the core lazy mm refcounting, and >>>>> x86 bare metal *also* does not need ->active_mm. Under the x86 scheme, >>>>> if lazy mm refcounting were configured out, ->active_mm could become a >>>>> dangling pointer, and this makes me extremely uncomfortable. >>>>> >>>>> So I tend to think that, depending on config, the core code should >>>>> either keep ->active_mm [1] alive or get rid of it entirely. >>>> >>>> I don't actually know what you mean. >>>> >>>> core code needs the concept of an "active_mm". This is the mm that your >>>> kernel threads are using, even in the unmerged CONFIG_LAZY_TLB=n patch, >>>> active_mm still points to init_mm for kernel threads. >>> >>> Core code does *not* need this concept. First, it's wrong on x86 since >>> at least 4.15. Any core code that actually assumes that ->active_mm is >>> "active" for any sensible definition of the word active is wrong. >>> Fortunately there is no such code. >>> >>> I looked through all active_mm references in core code. We have: >>> >>> kernel/sched/core.c: it's all refcounting, although it's a bit tangled >>> with membarrier. >>> >>> kernel/kthread.c: same. refcounting and membarrier stuff. >>> >>> kernel/exit.c: exit_mm() a BUG_ON(). >>> >>> kernel/fork.c: initialization code and a warning. >>> >>> kernel/cpu.c: cpu offline stuff. wouldn't be needed if active_mm went away. >>> >>> fs/exec.c: nothing of interest >> >> I might not have been clear. Core code doesn't need active_mm if >> active_mm somehow goes away. I'm saying active_mm can't go away because >> it's needed to support (most) archs that do lazy tlb mm switching. >> >> The part I don't understand is when you say it can just go away. How? >> >>> I didn't go through drivers, but I maintain my point. active_mm is >>> there for refcounting. So please don't just make it even more confusing >>> -- do your performance improvement, but improve the code at the same >>> time: get rid of active_mm, at least on architectures that opt out of >>> the refcounting. >> >> powerpc opts out of the refcounting and can not "get rid of active_mm". >> Not even in theory. > > That is to say, it does do a type of reference management that requires > active_mm so you can argue it has not entirely opted out of refcounting. > But we're not just doing refcounting for the sake of refcounting! That > would make no sense. > > active_mm is required because that's the mm that we have switched to > (from core code's perspective), and it is integral to know when to > switch to a different mm. See how active_mm is a fundamental concept > in core code? It's part of the contract between core code and the > arch mm context management calls. reference counting follows from there > but it's not the _reason_ for this code. > > Pretend the reference problem does not exit (whether by refcounting or > shootdown or garbage collection or whatever). We still can't remove > active_mm! We need it to know how to call into arch functions like > switch_mm. > > I don't know if you just forgot that critical requirement in your above > list, or you actually are entirely using x86's mental model for this > code which is doing something entirely different that does not need it > at all. If that is the case I really don't mind some cleanup or wrapper > functions for x86 do entirely do its own thing, but if that's the case > you can't criticize core code's use of active_mm due to the current > state of x86. It's x86 that needs documentation and cleaning up. Ah, that must be where your confusion is coming from: x86's switch_mm doesn't use prev anywhere, and the reference scheme it is using appears to be under-documented, although vague references in changelogs suggest it has not actually "opted out" of active_mm refcounting. That's understandable, but please redirect your objections to the proper place. git blame suggests 3d28ebceaffab. Thanks, Nick
Replying to several emails at once... On 6/13/21 10:21 PM, Nicholas Piggin wrote: > Excerpts from Nicholas Piggin's message of June 14, 2021 2:47 pm: >> Excerpts from Nicholas Piggin's message of June 14, 2021 2:14 pm: >>> Excerpts from Andy Lutomirski's message of June 14, 2021 1:52 pm: >>>> On 6/13/21 5:45 PM, Nicholas Piggin wrote: >>>>> Excerpts from Andy Lutomirski's message of June 9, 2021 2:20 am: >>>>>> On 6/4/21 6:42 PM, Nicholas Piggin wrote: >>>>>>> Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm >>>>>>> when it is context switched. This can be disabled by architectures that >>>>>>> don't require this refcounting if they clean up lazy tlb mms when the >>>>>>> last refcount is dropped. Currently this is always enabled, which is >>>>>>> what existing code does, so the patch is effectively a no-op. >>>>>>> >>>>>>> Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. >>>>>> >>>>>> I am in favor of this approach, but I would be a lot more comfortable >>>>>> with the resulting code if task->active_mm were at least better >>>>>> documented and possibly even guarded by ifdefs. >>>>> >>>>> active_mm is fairly well documented in Documentation/active_mm.rst IMO. >>>>> I don't think anything has changed in 20 years, I don't know what more >>>>> is needed, but if you can add to documentation that would be nice. Maybe >>>>> moving a bit of that into .c and .h files? >>>>> >>>> >>>> Quoting from that file: >>>> >>>> - however, we obviously need to keep track of which address space we >>>> "stole" for such an anonymous user. For that, we have "tsk->active_mm", >>>> which shows what the currently active address space is. >>>> >>>> This isn't even true right now on x86. >>> >>> From the perspective of core code, it is. x86 might do something crazy >>> with it, but it has to make it appear this way to non-arch code that >>> uses active_mm. >>> >>> Is x86's scheme documented? arch/x86/include/asm/tlbflush.h documents it a bit: /* * cpu_tlbstate.loaded_mm should match CR3 whenever interrupts * are on. This means that it may not match current->active_mm, * which will contain the previous user mm when we're in lazy TLB * mode even if we've already switched back to swapper_pg_dir. * * During switch_mm_irqs_off(), loaded_mm will be set to * LOADED_MM_SWITCHING during the brief interrupts-off window * when CR3 and loaded_mm would otherwise be inconsistent. This * is for nmi_uaccess_okay()'s benefit. */ >>> >>>> With your patch applied: >>>> >>>> To support all that, the "struct mm_struct" now has two counters: a >>>> "mm_users" counter that is how many "real address space users" there are, >>>> and a "mm_count" counter that is the number of "lazy" users (ie anonymous >>>> users) plus one if there are any real users. >>>> >>>> isn't even true any more. >>> >>> Well yeah but the active_mm concept hasn't changed. The refcounting >>> change is hopefully reasonably documented? active_mm is *only* refcounting in the core code. See below. >>>> >>>> I looked through all active_mm references in core code. We have: >>>> >>>> kernel/sched/core.c: it's all refcounting, although it's a bit tangled >>>> with membarrier. >>>> >>>> kernel/kthread.c: same. refcounting and membarrier stuff. >>>> >>>> kernel/exit.c: exit_mm() a BUG_ON(). >>>> >>>> kernel/fork.c: initialization code and a warning. >>>> >>>> kernel/cpu.c: cpu offline stuff. wouldn't be needed if active_mm went away. >>>> >>>> fs/exec.c: nothing of interest >>> >>> I might not have been clear. Core code doesn't need active_mm if >>> active_mm somehow goes away. I'm saying active_mm can't go away because >>> it's needed to support (most) archs that do lazy tlb mm switching. >>> >>> The part I don't understand is when you say it can just go away. How? #ifdef CONFIG_MMU_TLB_REFCOUNT struct mm_struct *active_mm; #endif >>> >>>> I didn't go through drivers, but I maintain my point. active_mm is >>>> there for refcounting. So please don't just make it even more confusing >>>> -- do your performance improvement, but improve the code at the same >>>> time: get rid of active_mm, at least on architectures that opt out of >>>> the refcounting. >>> >>> powerpc opts out of the refcounting and can not "get rid of active_mm". >>> Not even in theory. >> >> That is to say, it does do a type of reference management that requires >> active_mm so you can argue it has not entirely opted out of refcounting. >> But we're not just doing refcounting for the sake of refcounting! That >> would make no sense. >> >> active_mm is required because that's the mm that we have switched to >> (from core code's perspective), and it is integral to know when to >> switch to a different mm. See how active_mm is a fundamental concept >> in core code? It's part of the contract between core code and the >> arch mm context management calls. reference counting follows from there >> but it's not the _reason_ for this code. I don't understand what contract you're talking about. The core code maintains an active_mm counter and keeps that mm_struct from disappearing. That's *it*. The core code does not care that active_mm is active, and x86 provides evidence of that -- on x86, current->active_mm may well be completely unused. >> >> Pretend the reference problem does not exit (whether by refcounting or >> shootdown or garbage collection or whatever). We still can't remove >> active_mm! We need it to know how to call into arch functions like >> switch_mm. static inline void do_switch_mm(struct task_struct *prev_task, ...) { #ifdef CONFIG_MMU_TLB_REFCOUNT switch_mm(...); #else switch_mm(fewer parameters); /* or pass NULL or whatever. */ #endif } >> >> I don't know if you just forgot that critical requirement in your above >> list, or you actually are entirely using x86's mental model for this >> code which is doing something entirely different that does not need it >> at all. If that is the case I really don't mind some cleanup or wrapper >> functions for x86 do entirely do its own thing, but if that's the case >> you can't criticize core code's use of active_mm due to the current >> state of x86. It's x86 that needs documentation and cleaning up. > > Ah, that must be where your confusion is coming from: x86's switch_mm > doesn't use prev anywhere, and the reference scheme it is using appears > to be under-documented, although vague references in changelogs suggest > it has not actually "opted out" of active_mm refcounting. All of this is true, except I don't believe I'm confused. > > That's understandable, but please redirect your objections to the proper > place. git blame suggests 3d28ebceaffab. Thanks for the snark. Here's the situation. Before that patch, x86 more or less fully used the core scheme. With that patch, x86 has the property that loaded_mm (which is used) either points to init_mm (which is permanently live) or to active_mm (which the core code keeps alive, which is the whole point). x86 still uses the core active_mm refcounting scheme. The result is a bit overcomplicated, but it works, and it enabled massive improvements to the x86 arch code. You are proposing a whole new simplification in which an arch can opt in to telling the core code that it doesn't need to keep active_mm alive. Great! But now it's possible to get quite confused -- either an elaborate dance is needed or current->active_mm could point to freed memory. This is poor design. I'm entirely in favor of allowing arches to opt out of active_mm refcounting. As you've noticed, it's quite expensive on large systems. x86 would opt out, too, if given the opportunity [1]. But I think you should do it right. If an arch opts out of active_mm refcounting, then keeping a lazy mm (if any!) alive becomes entirely the arch code's responsibility. Once that happens, task->active_mm is not just a waste of 4-8 bytes of memory per task, it's actively harmful -- some code, somewhere in the kernel, might dereference it and access freed memory! So please do your patches right. By all means add a new config option, but make that config option make active_mm go away entirely. Then it can't be misused. NAK to the current set. --Andy [1] Hi Rik! I think you had benchmarks that made mm refcounting look quite bad. If x86 can opt out of the core scheme, we can just rejigger the exit_mm path to force-IPI everyone instead of allowing the paravirt code to possibly optimize that path. Or we can use a hazard pointer scheme like my WIP patch.
Excerpts from Andy Lutomirski's message of June 15, 2021 2:20 am: > Replying to several emails at once... > > > On 6/13/21 10:21 PM, Nicholas Piggin wrote: >> Excerpts from Nicholas Piggin's message of June 14, 2021 2:47 pm: >>> Excerpts from Nicholas Piggin's message of June 14, 2021 2:14 pm: >>>> Excerpts from Andy Lutomirski's message of June 14, 2021 1:52 pm: >>>>> On 6/13/21 5:45 PM, Nicholas Piggin wrote: >>>>>> Excerpts from Andy Lutomirski's message of June 9, 2021 2:20 am: >>>>>>> On 6/4/21 6:42 PM, Nicholas Piggin wrote: >>>>>>>> Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm >>>>>>>> when it is context switched. This can be disabled by architectures that >>>>>>>> don't require this refcounting if they clean up lazy tlb mms when the >>>>>>>> last refcount is dropped. Currently this is always enabled, which is >>>>>>>> what existing code does, so the patch is effectively a no-op. >>>>>>>> >>>>>>>> Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. >>>>>>> >>>>>>> I am in favor of this approach, but I would be a lot more comfortable >>>>>>> with the resulting code if task->active_mm were at least better >>>>>>> documented and possibly even guarded by ifdefs. >>>>>> >>>>>> active_mm is fairly well documented in Documentation/active_mm.rst IMO. >>>>>> I don't think anything has changed in 20 years, I don't know what more >>>>>> is needed, but if you can add to documentation that would be nice. Maybe >>>>>> moving a bit of that into .c and .h files? >>>>>> >>>>> >>>>> Quoting from that file: >>>>> >>>>> - however, we obviously need to keep track of which address space we >>>>> "stole" for such an anonymous user. For that, we have "tsk->active_mm", >>>>> which shows what the currently active address space is. >>>>> >>>>> This isn't even true right now on x86. >>>> >>>> From the perspective of core code, it is. x86 might do something crazy >>>> with it, but it has to make it appear this way to non-arch code that >>>> uses active_mm. >>>> >>>> Is x86's scheme documented? > > arch/x86/include/asm/tlbflush.h documents it a bit: > > /* > * cpu_tlbstate.loaded_mm should match CR3 whenever interrupts > * are on. This means that it may not match current->active_mm, > * which will contain the previous user mm when we're in lazy TLB > * mode even if we've already switched back to swapper_pg_dir. > * > * During switch_mm_irqs_off(), loaded_mm will be set to > * LOADED_MM_SWITCHING during the brief interrupts-off window > * when CR3 and loaded_mm would otherwise be inconsistent. This > * is for nmi_uaccess_okay()'s benefit. > */ So the only documentation relating to the current active_mm value or refcounting is that it may not match what the x86 specific code is doing? All this complexity you accuse me of adding is entirely in x86 code. On other architectures, it's very simple and understandable, and documented. I don't know how else to explain this. >>>> >>>>> With your patch applied: >>>>> >>>>> To support all that, the "struct mm_struct" now has two counters: a >>>>> "mm_users" counter that is how many "real address space users" there are, >>>>> and a "mm_count" counter that is the number of "lazy" users (ie anonymous >>>>> users) plus one if there are any real users. >>>>> >>>>> isn't even true any more. >>>> >>>> Well yeah but the active_mm concept hasn't changed. The refcounting >>>> change is hopefully reasonably documented? > > active_mm is *only* refcounting in the core code. See below. It's just not. It's passed in to switch_mm. Most architectures except for x86 require this. >>>>> >>>>> I looked through all active_mm references in core code. We have: >>>>> >>>>> kernel/sched/core.c: it's all refcounting, although it's a bit tangled >>>>> with membarrier. >>>>> >>>>> kernel/kthread.c: same. refcounting and membarrier stuff. >>>>> >>>>> kernel/exit.c: exit_mm() a BUG_ON(). >>>>> >>>>> kernel/fork.c: initialization code and a warning. >>>>> >>>>> kernel/cpu.c: cpu offline stuff. wouldn't be needed if active_mm went away. >>>>> >>>>> fs/exec.c: nothing of interest >>>> >>>> I might not have been clear. Core code doesn't need active_mm if >>>> active_mm somehow goes away. I'm saying active_mm can't go away because >>>> it's needed to support (most) archs that do lazy tlb mm switching. >>>> >>>> The part I don't understand is when you say it can just go away. How? > > #ifdef CONFIG_MMU_TLB_REFCOUNT > struct mm_struct *active_mm; > #endif Thanks for returning the snark. >>>> >>>>> I didn't go through drivers, but I maintain my point. active_mm is >>>>> there for refcounting. So please don't just make it even more confusing >>>>> -- do your performance improvement, but improve the code at the same >>>>> time: get rid of active_mm, at least on architectures that opt out of >>>>> the refcounting. >>>> >>>> powerpc opts out of the refcounting and can not "get rid of active_mm". >>>> Not even in theory. >>> >>> That is to say, it does do a type of reference management that requires >>> active_mm so you can argue it has not entirely opted out of refcounting. >>> But we're not just doing refcounting for the sake of refcounting! That >>> would make no sense. >>> >>> active_mm is required because that's the mm that we have switched to >>> (from core code's perspective), and it is integral to know when to >>> switch to a different mm. See how active_mm is a fundamental concept >>> in core code? It's part of the contract between core code and the >>> arch mm context management calls. reference counting follows from there >>> but it's not the _reason_ for this code. > > I don't understand what contract you're talking about. The core code > maintains an active_mm counter and keeps that mm_struct from > disappearing. That's *it*. The core code does not care that active_mm > is active, and x86 provides evidence of that -- on x86, > current->active_mm may well be completely unused. I already acknowledged archs can do their own thing under the covers if they want. > >>> >>> Pretend the reference problem does not exit (whether by refcounting or >>> shootdown or garbage collection or whatever). We still can't remove >>> active_mm! We need it to know how to call into arch functions like >>> switch_mm. > > static inline void do_switch_mm(struct task_struct *prev_task, ...) > { > #ifdef CONFIG_MMU_TLB_REFCOUNT > switch_mm(...); > #else > switch_mm(fewer parameters); > /* or pass NULL or whatever. */ > #endif > } And prev_task comes from active_mm, ergo core code requires the concept of active_mm. I already said I'm quite happy for wrappers to be added so those can compile out of an arch like x86. That's not doing anything for complexity of core code, because it still needs to maintain the active mm wrappers. >>> I don't know if you just forgot that critical requirement in your above >>> list, or you actually are entirely using x86's mental model for this >>> code which is doing something entirely different that does not need it >>> at all. If that is the case I really don't mind some cleanup or wrapper >>> functions for x86 do entirely do its own thing, but if that's the case >>> you can't criticize core code's use of active_mm due to the current >>> state of x86. It's x86 that needs documentation and cleaning up. >> >> Ah, that must be where your confusion is coming from: x86's switch_mm >> doesn't use prev anywhere, and the reference scheme it is using appears >> to be under-documented, although vague references in changelogs suggest >> it has not actually "opted out" of active_mm refcounting. > > All of this is true, except I don't believe I'm confused. Well someone is. My patches don't change any of the fundamental complexity of core code's maintaining of active_mm, nor alleviate the requirement to keep active_mm around under the new config introduced, and yet you are saying: I am in favor of this approach, but I would be a lot more comfortable with the resulting code if task->active_mm were at least better documented and possibly even guarded by ifdefs. It doesn't make sense. It can't be guarded by ifdefs even if we took away the shootdown IPI's use of it as part of reference / lifetime management. And that you don't like the state of the code but I'm trying to wrangle specifics out of you on that and it's difficult. >> >> That's understandable, but please redirect your objections to the proper >> place. git blame suggests 3d28ebceaffab. > > Thanks for the snark. Is it not true? I don't mean that one patch causing all the x86 complexity or even making the situation worse itself. But you seem to be asking my series to do things that really apply to the x86 changes over the past few years that got us here. > > Here's the situation. > > Before that patch, x86 more or less fully used the core scheme. With > that patch, x86 has the property that loaded_mm (which is used) either > points to init_mm (which is permanently live) or to active_mm (which the > core code keeps alive, which is the whole point). x86 still uses the > core active_mm refcounting scheme. The result is a bit overcomplicated, > but it works, and it enabled massive improvements to the x86 arch code. > > You are proposing a whole new simplification in which an arch can opt in > to telling the core code that it doesn't need to keep active_mm alive. > Great! But now it's possible to get quite confused -- either an > elaborate dance is needed or current->active_mm could point to freed > memory. This is poor design. powerpc still needs active_mm, there's zero point to configuring it away and maintaining exactly the same thing in a per-cpu variable in the arch code because the code has to be there in the core for all other archs anyway so that would be stupid. I would be more than happy to try help review a patch that helps the x86 situation, but my patch is not intended to do what x86 code wants. > I'm entirely in favor of allowing arches to opt out of active_mm > refcounting. As you've noticed, it's quite expensive on large systems. > x86 would opt out, too, if given the opportunity [1]. But I think you > should do it right. If an arch opts out of active_mm refcounting, then > keeping a lazy mm (if any!) alive becomes entirely the arch code's > responsibility. > > Once that happens, task->active_mm is not just a waste > of 4-8 bytes of memory per task, it's actively harmful -- some code, > somewhere in the kernel, might dereference it and access freed memory! That's not an accurate description of my patches. powerpc still requires active_mm exactly the same, and the "elaborate dance" amounts to having CPUs stop using the mm as a lazy tlb when the last user goes away. > > So please do your patches right. By all means add a new config option, > but make that config option make active_mm go away entirely. Then it > can't be misused. I already have a patch that's halfway there I pulled out of the series for the last submission. As I said, feel free to build on it. > > NAK to the current set. I propose instead we do take the series, and write some patches on top of that which helps x86. Here's 1/n 2/n is to make wrappers for active_mm maintenance in core code. 3/n is adjust context_switch_nolazy() to just use init_mm (or NULL) directly and put ifdefs around active_mm. Is that what x86 wants? Thanks, Nick commit 99f90520821b9717d4447872354c2933894a1100 Author: Nicholas Piggin <npiggin@gmail.com> Date: Thu Jul 9 15:01:25 2020 +1000 lazy tlb: allow lazy tlb mm switching to be configurable Add CONFIG_MMU_LAZY_TLB which can be configured out to disable the lazy tlb mechanism entirely, and switches to init_mm when switching to a kernel thread. NOMMU systems could easily go without this and save a bit of code and the refcount atomics, because their mm switch is a no-op. They have not been switched over by default because the arch code needs to be audited and tested for lazy tlb mm refcounting and converted to _lazy_tlb refcounting if necessary. CONFIG_MMU_LAZY_TLB_REFCOUNT is also added, but it must always be enabled if CONFIG_MMU_LAZY_TLB is enabled until the next patch which provides an alternate scheme. Signed-off-by: Nicholas Piggin <npiggin@gmail.com> diff --git a/arch/Kconfig b/arch/Kconfig index cf468c9777d8..3b80042bd0c2 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -418,6 +418,26 @@ config ARCH_WANT_IRQS_OFF_ACTIVATE_MM irqs disabled over activate_mm. Architectures that do IPI based TLB shootdowns should enable this. +# Enable "lazy TLB", which means a user->kernel thread context switch does not +# switch the mm to init_mm and the kernel thread takes a reference to the user +# mm to provide its kernel mapping. This is how Linux has traditionally worked +# (see Documentation/vm/active_mm.rst), for performance. Switching to and from +# idle thread is a performance-critical case. +# +# If mm context switches are inexpensive or free (in the case of NOMMU) then +# this could be disabled. +# +# It would make sense to have this depend on MMU, but need to audit and test +# the NOMMU architectures for lazy mm refcounting first. +config MMU_LAZY_TLB + def_bool y + depends on !NO_MMU_LAZY_TLB + +# This allows archs to disable MMU_LAZY_TLB. mmgrab/mmdrop in arch/ code has +# to be audited and switched to _lazy_tlb postfix as necessary. +config NO_MMU_LAZY_TLB + def_bool n + # Use normal mm refcounting for MMU_LAZY_TLB kernel thread references. # MMU_LAZY_TLB_REFCOUNT=n can improve the scalability of context switching # to/from kernel threads when the same mm is running on a lot of CPUs (a large @@ -431,6 +451,7 @@ config ARCH_WANT_IRQS_OFF_ACTIVATE_MM # to a kthread ->active_mm (non-arch code has been converted already). config MMU_LAZY_TLB_REFCOUNT def_bool y + depends on MMU_LAZY_TLB depends on !MMU_LAZY_TLB_SHOOTDOWN # This option allows MMU_LAZY_TLB_REFCOUNT=n. It ensures no CPUs are using an @@ -445,6 +466,7 @@ config MMU_LAZY_TLB_REFCOUNT # MMU_LAZY_TLB_REFCOUNT=n (see above). config MMU_LAZY_TLB_SHOOTDOWN bool + depends on MMU_LAZY_TLB config ARCH_HAVE_NMI_SAFE_CMPXCHG bool diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 5e10cb712be3..5cb039c686a6 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -4285,22 +4285,10 @@ asmlinkage __visible void schedule_tail(struct task_struct *prev) calculate_sigpending(); } -/* - * context_switch - switch to the new MM and the new thread's register state. - */ -static __always_inline struct rq * -context_switch(struct rq *rq, struct task_struct *prev, - struct task_struct *next, struct rq_flags *rf) +static __always_inline void +context_switch_mm(struct rq *rq, struct task_struct *prev, + struct task_struct *next) { - prepare_task_switch(rq, prev, next); - - /* - * For paravirt, this is coupled with an exit in switch_to to - * combine the page table reload and the switch backend into - * one hypercall. - */ - arch_start_context_switch(prev); - /* * kernel -> kernel lazy + transfer active * user -> kernel lazy + mmgrab_lazy_tlb() active @@ -4345,6 +4333,40 @@ context_switch(struct rq *rq, struct task_struct *prev, #endif } } +} + +static __always_inline void +context_switch_mm_nolazy(struct rq *rq, struct task_struct *prev, + struct task_struct *next) +{ + if (!next->mm) + next->active_mm = &init_mm; + membarrier_switch_mm(rq, prev->active_mm, next->active_mm); + switch_mm_irqs_off(prev->active_mm, next->active_mm, next); + if (!prev->mm) + prev->active_mm = NULL; +} + +/* + * context_switch - switch to the new MM and the new thread's register state. + */ +static __always_inline struct rq * +context_switch(struct rq *rq, struct task_struct *prev, + struct task_struct *next, struct rq_flags *rf) +{ + prepare_task_switch(rq, prev, next); + + /* + * For paravirt, this is coupled with an exit in switch_to to + * combine the page table reload and the switch backend into + * one hypercall. + */ + arch_start_context_switch(prev); + + if (IS_ENABLED(CONFIG_MMU_LAZY_TLB)) + context_switch_mm(rq, prev, next); + else + context_switch_mm_nolazy(rq, prev, next); rq->clock_update_flags &= ~(RQCF_ACT_SKIP|RQCF_REQ_SKIP);
On 6/14/21 5:55 PM, Nicholas Piggin wrote: > Excerpts from Andy Lutomirski's message of June 15, 2021 2:20 am: >> Replying to several emails at once... >> > > So the only documentation relating to the current active_mm value or > refcounting is that it may not match what the x86 specific code is > doing? > > All this complexity you accuse me of adding is entirely in x86 code. > On other architectures, it's very simple and understandable, and > documented. I don't know how else to explain this. And the docs you referred me to will be *wrong* with your patches applied. They are your patches, and they break the semantics. > >>>>> >>>>>> With your patch applied: >>>>>> >>>>>> To support all that, the "struct mm_struct" now has two counters: a >>>>>> "mm_users" counter that is how many "real address space users" there are, >>>>>> and a "mm_count" counter that is the number of "lazy" users (ie anonymous >>>>>> users) plus one if there are any real users. >>>>>> >>>>>> isn't even true any more. >>>>> >>>>> Well yeah but the active_mm concept hasn't changed. The refcounting >>>>> change is hopefully reasonably documented? >> >> active_mm is *only* refcounting in the core code. See below. > > It's just not. It's passed in to switch_mm. Most architectures except > for x86 require this. > Sorry, I was obviously blatantly wrong. Let me say it differently. active_mm does two things: 1. It keeps an mm alive via a refcounting scheme. 2. It passes a parameter to switch_mm() to remind the arch code what the most recently switch-to mm was. #2 is basically useless. An architecture can handle *that* with a percpu variable and two lines of code. If you are getting rid of functionality #1 in the core code via a new arch opt-out, please get rid of #2 as well. *Especially* because, when the arch asks the core code to stop refcounting active_mm, there is absolutely nothing guaranteeing that the parameter that the core code will pass to switch_mm() points to memory that hasn't been freed and reused for another purpose. >>>>> I might not have been clear. Core code doesn't need active_mm if >>>>> active_mm somehow goes away. I'm saying active_mm can't go away because >>>>> it's needed to support (most) archs that do lazy tlb mm switching. >>>>> >>>>> The part I don't understand is when you say it can just go away. How? >> >> #ifdef CONFIG_MMU_TLB_REFCOUNT >> struct mm_struct *active_mm; >> #endif > > Thanks for returning the snark. That wasn't intended to be snark. It was a literal suggestion, and, in fact, it's *exactly* what I'm asking you to do to fix your patches. >> I don't understand what contract you're talking about. The core code >> maintains an active_mm counter and keeps that mm_struct from >> disappearing. That's *it*. The core code does not care that active_mm >> is active, and x86 provides evidence of that -- on x86, >> current->active_mm may well be completely unused. > > I already acknowledged archs can do their own thing under the covers if > they want. No. I am *not* going to write x86 patches that use your feature in a way that will cause the core code to pass around a complete garbage pointer to an mm_struct that is completely unreferenced and may well be deleted. Doing something private in arch code is one thing. Doing something that causes basic common sense to be violated in core code is another thing entirely. >> >> static inline void do_switch_mm(struct task_struct *prev_task, ...) >> { >> #ifdef CONFIG_MMU_TLB_REFCOUNT >> switch_mm(...); >> #else >> switch_mm(fewer parameters); >> /* or pass NULL or whatever. */ >> #endif >> } > > And prev_task comes from active_mm, ergo core code requires the concept > of active_mm. I don't see why this concept is hard. We are literally quibbling about this single line of core code in kernel/sched/core.c: switch_mm_irqs_off(prev->active_mm, next->mm, next); This is not rocket science. There are any number of ways to fix it. For example: #ifdef CONFIG_MMU_TLB_REFCOUNT switch_mm_irqs_off(prev->active_mm, next->mm, next); #else switch_mm_irqs_off(NULL, next->mm, next); #endif If you don't like the NULL, then make the signature depend on the config option. What you may not do is what your patch actually does: switch_mm_irqs_off(random invalid pointer, next->mm, next); Now maybe it works because powerpc's lifecycle rules happen to keep active_mm alive, but I haven't verified it. x86's lifecycle rules *do not*. >>> >>> That's understandable, but please redirect your objections to the proper >>> place. git blame suggests 3d28ebceaffab. >> >> Thanks for the snark. > > Is it not true? I don't mean that one patch causing all the x86 > complexity or even making the situation worse itself. But you seem to be > asking my series to do things that really apply to the x86 changes over > the past few years that got us here. With just my patch from 4.15 applied, task->active_mm points to an actual mm_struct, and that mm_struct will not be freed early. If I opt x86 into your patch's new behavior, then task->active_mm may be freed. akpm, please drop this series until it's fixed. It's a core change to better support arch usecases, but it's unnecessarily fragile, and there is already an arch maintainer pointing out that it's inadequate to robustly support arch usecases. There is no reason to merge it in its present state.
Excerpts from Andy Lutomirski's message of June 16, 2021 10:14 am: > On 6/14/21 5:55 PM, Nicholas Piggin wrote: >> Excerpts from Andy Lutomirski's message of June 15, 2021 2:20 am: >>> Replying to several emails at once... >>> > >> >> So the only documentation relating to the current active_mm value or >> refcounting is that it may not match what the x86 specific code is >> doing? >> >> All this complexity you accuse me of adding is entirely in x86 code. >> On other architectures, it's very simple and understandable, and >> documented. I don't know how else to explain this. > > And the docs you referred me to will be *wrong* with your patches > applied. They are your patches, and they break the semantics. No they aren't wrong, I've documented the shootdown refcounting scheme and Linus' historical email is still right about the active_mm concept, and the refcounting details still match the CONFIG_MMU_TLB_REFCOUNT=y case. If you have some particular place you would like to see more documentation added, please tell me where I'll add something before the series gets upstreamed. >>>>>> >>>>>>> With your patch applied: >>>>>>> >>>>>>> To support all that, the "struct mm_struct" now has two counters: a >>>>>>> "mm_users" counter that is how many "real address space users" there are, >>>>>>> and a "mm_count" counter that is the number of "lazy" users (ie anonymous >>>>>>> users) plus one if there are any real users. >>>>>>> >>>>>>> isn't even true any more. >>>>>> >>>>>> Well yeah but the active_mm concept hasn't changed. The refcounting >>>>>> change is hopefully reasonably documented? >>> >>> active_mm is *only* refcounting in the core code. See below. >> >> It's just not. It's passed in to switch_mm. Most architectures except >> for x86 require this. >> > > Sorry, I was obviously blatantly wrong. Let me say it differently. > active_mm does two things: > > 1. It keeps an mm alive via a refcounting scheme. > > 2. It passes a parameter to switch_mm() to remind the arch code what the > most recently switch-to mm was. > > #2 is basically useless. An architecture can handle *that* with a > percpu variable and two lines of code. > > If you are getting rid of functionality #1 in the core code via a new > arch opt-out, please get rid of #2 as well. *Especially* because, when > the arch asks the core code to stop refcounting active_mm, there is > absolutely nothing guaranteeing that the parameter that the core code > will pass to switch_mm() points to memory that hasn't been freed and > reused for another purpose. You're confused about the patch. It's not opting out of lazy tlb mm or opting out of active_mm, it is changing how the refcounting is done. You were just before trying to tell me it would make the code simpler to remove it, but it clearly wouldn't if powerpc just had to reimplement it in arch code anyway, would it? powerpc is not going to add code to reimplement the exact same thing as active_mm, because it uses active_mm and so does everyone else. For the last time I'm not going to change that. That is something x86 wants. This series is not for x86. It doesn't change anything that x86 does. I have kindly tried to give you suggestions and patches about what you might do with x86, and it can easily be changed, but that is not the purpose of my patch. Please don't reply again telling me to get rid of active_mm. We'll draw a line under this and move on. >>> I don't understand what contract you're talking about. The core code >>> maintains an active_mm counter and keeps that mm_struct from >>> disappearing. That's *it*. The core code does not care that active_mm >>> is active, and x86 provides evidence of that -- on x86, >>> current->active_mm may well be completely unused. >> >> I already acknowledged archs can do their own thing under the covers if >> they want. > > No. > > I am *not* going to write x86 patches that use your feature in a way > that will cause the core code to pass around a complete garbage pointer > to an mm_struct that is completely unreferenced and may well be deleted. > Doing something private in arch code is one thing. Doing something > that causes basic common sense to be violated in core code is another > thing entirely. I'm talking about the relationship between core code's idea of active_mm and the arch's idea as it stands now. I'm not talking about what you might do with x86. The whole point of this was that you have been operating under the misconception that active_mm is not a core kernel concept, because x86 has gone and done its own thing and doesn't use it. That does not mean it is not a core kernel concept! > >>> >>> static inline void do_switch_mm(struct task_struct *prev_task, ...) >>> { >>> #ifdef CONFIG_MMU_TLB_REFCOUNT >>> switch_mm(...); >>> #else >>> switch_mm(fewer parameters); >>> /* or pass NULL or whatever. */ >>> #endif >>> } >> >> And prev_task comes from active_mm, ergo core code requires the concept >> of active_mm. > > I don't see why this concept is hard. We are literally quibbling about > this single line of core code in kernel/sched/core.c: > > switch_mm_irqs_off(prev->active_mm, next->mm, next); > > This is not rocket science. There are any number of ways to fix it. > For example: > > #ifdef CONFIG_MMU_TLB_REFCOUNT > switch_mm_irqs_off(prev->active_mm, next->mm, next); > #else > switch_mm_irqs_off(NULL, next->mm, next); > #endif > > If you don't like the NULL, then make the signature depend on the config > option. > > What you may not do is what your patch actually does: > > switch_mm_irqs_off(random invalid pointer, next->mm, next); You're totally confused about the code, and my patch series. That's not what it does at all. Unless you have spotted a bug, in which case point it out. > > Now maybe it works because powerpc's lifecycle rules happen to keep > active_mm alive, but I haven't verified it. x86's lifecycle rules *do not*. > >>>> >>>> That's understandable, but please redirect your objections to the proper >>>> place. git blame suggests 3d28ebceaffab. >>> >>> Thanks for the snark. >> >> Is it not true? I don't mean that one patch causing all the x86 >> complexity or even making the situation worse itself. But you seem to be >> asking my series to do things that really apply to the x86 changes over >> the past few years that got us here. > > With just my patch from 4.15 applied, task->active_mm points to an > actual mm_struct, and that mm_struct will not be freed early. If I opt > x86 into your patch's new behavior, then task->active_mm may be freed. > > akpm, please drop this series until it's fixed. It's a core change to > better support arch usecases, but it's unnecessarily fragile, and there > is already an arch maintainer pointing out that it's inadequate to > robustly support arch usecases. There is no reason to merge it in its > present state. > No, you've been sniping these for nearly a year now, and I've been pretty accommodating not pushing it upstream, and trying to get anything out of you, a patch or an actual description of the problem, but it's like getting blood from a stone. I've tried to ask what help x86 needs, but nothing. You didn't even reply to my (actual working) patch or questions about it in the last comment! Doing that then throwing out NAK like it's a magic word is pretty rude really. The series is fine, it's documented, it works well, it solves a problem the core code changes are small and reusable, and nobody has pointed out a bug. So at this point, just get over it. Do some patches for x86 on top of it later when you have some time. Or Rik might. None of the patches in my series prevents that. I'll be happy to review your changes or adjust some of the code added in this series so it's usable by x86 if you need. The x86 work could have bee done already with the time bickering about this. Thanks, Nick
Excerpts from Nicholas Piggin's message of June 16, 2021 11:02 am: > Excerpts from Andy Lutomirski's message of June 16, 2021 10:14 am: >> akpm, please drop this series until it's fixed. It's a core change to >> better support arch usecases, but it's unnecessarily fragile, and there >> is already an arch maintainer pointing out that it's inadequate to >> robustly support arch usecases. There is no reason to merge it in its >> present state. Just to make sure I'm not doing anything stupid or fragile for other archs, I had a closer look at a few. sparc32 is the only one I have a SMP capable qemu and initramfs at hand for, took about 5 minutes to convert after fixing 2 other sparc32/mm bugs (patches on linux-sparc), one of them found by the DEBUG_VM code my series added. It seems to work fine, with what little stressing my qemu setup can muster. Simple. Robust. Pretty mechanical conversion follows the documented reciple. Re-uses every single line of code I added outside arch/powerpc/. Requires no elaborate dances. alpha and arm64 are both 4-liners by the looks, sparc64 might reqiure a bit of actual code but doesn't look too hard. So I'm satisfied the code added outside arch/powerpc/ is not some fragile powerpc specific hack. I don't know if other archs will use it, but they easily can use it[*]. And we can make changes to help x86 whenever its needed -- I already posted patch 1/n for configuring out lazy tlb and active_mm from core code rebased on top of mmotm so the series is not preventing such changes. Hopefully this allays some concerns. [*] I do think mmgrab_lazy_tlb is a nice change that self-documents the active_mm refcounting, so I will try to get all the arch code converted to use it over the next few releases, even if they never switch to use lazy tlb shootdown. Thanks, Nick --- arch/sparc/Kconfig | 1 + arch/sparc/kernel/leon_smp.c | 2 +- arch/sparc/kernel/smp_64.c | 2 +- arch/sparc/kernel/sun4d_smp.c | 2 +- arch/sparc/kernel/sun4m_smp.c | 2 +- arch/sparc/kernel/traps_32.c | 2 +- arch/sparc/kernel/traps_64.c | 2 +- 7 files changed, 7 insertions(+), 6 deletions(-) diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig index 164a5254c91c..db9954af57a2 100644 --- a/arch/sparc/Kconfig +++ b/arch/sparc/Kconfig @@ -58,6 +58,7 @@ config SPARC32 select GENERIC_ATOMIC64 select CLZ_TAB select HAVE_UID16 + select MMU_LAZY_TLB_SHOOTDOWN select OLD_SIGACTION config SPARC64 diff --git a/arch/sparc/kernel/leon_smp.c b/arch/sparc/kernel/leon_smp.c index 1eed26d423fb..d00460788048 100644 --- a/arch/sparc/kernel/leon_smp.c +++ b/arch/sparc/kernel/leon_smp.c @@ -92,7 +92,7 @@ void leon_cpu_pre_online(void *arg) : "memory" /* paranoid */); /* Attach to the address space of init_task. */ - mmgrab(&init_mm); + mmgrab_lazy_tlb(&init_mm); current->active_mm = &init_mm; while (!cpumask_test_cpu(cpuid, &smp_commenced_mask)) diff --git a/arch/sparc/kernel/smp_64.c b/arch/sparc/kernel/smp_64.c index e38d8bf454e8..19aa12991f2b 100644 --- a/arch/sparc/kernel/smp_64.c +++ b/arch/sparc/kernel/smp_64.c @@ -127,7 +127,7 @@ void smp_callin(void) current_thread_info()->new_child = 0; /* Attach to the address space of init_task. */ - mmgrab(&init_mm); + mmgrab_lazy_tlb(&init_mm); current->active_mm = &init_mm; /* inform the notifiers about the new cpu */ diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c index ff30f03beb7c..a6f392dcfeaf 100644 --- a/arch/sparc/kernel/sun4d_smp.c +++ b/arch/sparc/kernel/sun4d_smp.c @@ -94,7 +94,7 @@ void sun4d_cpu_pre_online(void *arg) show_leds(cpuid); /* Attach to the address space of init_task. */ - mmgrab(&init_mm); + mmgrab_lazy_tlb(&init_mm); current->active_mm = &init_mm; local_ops->cache_all(); diff --git a/arch/sparc/kernel/sun4m_smp.c b/arch/sparc/kernel/sun4m_smp.c index 228a6527082d..0ee77f066c9e 100644 --- a/arch/sparc/kernel/sun4m_smp.c +++ b/arch/sparc/kernel/sun4m_smp.c @@ -60,7 +60,7 @@ void sun4m_cpu_pre_online(void *arg) : "memory" /* paranoid */); /* Attach to the address space of init_task. */ - mmgrab(&init_mm); + mmgrab_lazy_tlb(&init_mm); current->active_mm = &init_mm; while (!cpumask_test_cpu(cpuid, &smp_commenced_mask)) diff --git a/arch/sparc/kernel/traps_32.c b/arch/sparc/kernel/traps_32.c index 247a0d9683b2..a3186bb30109 100644 --- a/arch/sparc/kernel/traps_32.c +++ b/arch/sparc/kernel/traps_32.c @@ -387,7 +387,7 @@ void trap_init(void) thread_info_offsets_are_bolixed_pete(); /* Attach to the address space of init_task. */ - mmgrab(&init_mm); + mmgrab_lazy_tlb(&init_mm); current->active_mm = &init_mm; /* NOTE: Other cpus have this done as they are started diff --git a/arch/sparc/kernel/traps_64.c b/arch/sparc/kernel/traps_64.c index a850dccd78ea..b6e46732fa69 100644 --- a/arch/sparc/kernel/traps_64.c +++ b/arch/sparc/kernel/traps_64.c @@ -2929,6 +2929,6 @@ void __init trap_init(void) /* Attach to the address space of init_task. On SMP we * do this in smp.c:smp_callin for other cpus. */ - mmgrab(&init_mm); + mmgrab_lazy_tlb(&init_mm); current->active_mm = &init_mm; }
diff --git a/arch/Kconfig b/arch/Kconfig index c45b770d3579..1cff045cdde6 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -418,6 +418,10 @@ config ARCH_WANT_IRQS_OFF_ACTIVATE_MM irqs disabled over activate_mm. Architectures that do IPI based TLB shootdowns should enable this. +# Use normal mm refcounting for MMU_LAZY_TLB kernel thread references. +config MMU_LAZY_TLB_REFCOUNT + def_bool y + config ARCH_HAVE_NMI_SAFE_CMPXCHG bool diff --git a/include/linux/sched/mm.h b/include/linux/sched/mm.h index bfd1baca5266..29e4638ad124 100644 --- a/include/linux/sched/mm.h +++ b/include/linux/sched/mm.h @@ -52,12 +52,21 @@ static inline void mmdrop(struct mm_struct *mm) /* Helpers for lazy TLB mm refcounting */ static inline void mmgrab_lazy_tlb(struct mm_struct *mm) { - mmgrab(mm); + if (IS_ENABLED(CONFIG_MMU_LAZY_TLB_REFCOUNT)) + mmgrab(mm); } static inline void mmdrop_lazy_tlb(struct mm_struct *mm) { - mmdrop(mm); + if (IS_ENABLED(CONFIG_MMU_LAZY_TLB_REFCOUNT)) { + mmdrop(mm); + } else { + /* + * mmdrop_lazy_tlb must provide a full memory barrier, see the + * membarrier comment finish_task_switch which relies on this. + */ + smp_mb(); + } } /** diff --git a/kernel/sched/core.c b/kernel/sched/core.c index e359c76ea2e2..5e10cb712be3 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -4171,7 +4171,7 @@ static struct rq *finish_task_switch(struct task_struct *prev) __releases(rq->lock) { struct rq *rq = this_rq(); - struct mm_struct *mm = rq->prev_mm; + struct mm_struct *mm = NULL; long prev_state; /* @@ -4190,7 +4190,10 @@ static struct rq *finish_task_switch(struct task_struct *prev) current->comm, current->pid, preempt_count())) preempt_count_set(FORK_PREEMPT_COUNT); - rq->prev_mm = NULL; +#ifdef CONFIG_MMU_LAZY_TLB_REFCOUNT + mm = rq->prev_lazy_mm; + rq->prev_lazy_mm = NULL; +#endif /* * A task struct has one reference for the use as "current". @@ -4326,9 +4329,20 @@ context_switch(struct rq *rq, struct task_struct *prev, switch_mm_irqs_off(prev->active_mm, next->mm, next); if (!prev->mm) { // from kernel - /* will mmdrop_lazy_tlb() in finish_task_switch(). */ - rq->prev_mm = prev->active_mm; +#ifdef CONFIG_MMU_LAZY_TLB_REFCOUNT + /* Will mmdrop_lazy_tlb() in finish_task_switch(). */ + rq->prev_lazy_mm = prev->active_mm; prev->active_mm = NULL; +#else + /* + * Without MMU_LAZY_TLB_REFCOUNT there is no lazy + * tracking (because no rq->prev_lazy_mm) in + * finish_task_switch, so no mmdrop_lazy_tlb(), so no + * memory barrier for membarrier (see the membarrier + * comment in finish_task_switch()). Do it here. + */ + smp_mb(); +#endif } } diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index a189bec13729..0729cf19a987 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -961,7 +961,9 @@ struct rq { struct task_struct *idle; struct task_struct *stop; unsigned long next_balance; - struct mm_struct *prev_mm; +#ifdef CONFIG_MMU_LAZY_TLB_REFCOUNT + struct mm_struct *prev_lazy_mm; +#endif unsigned int clock_update_flags; u64 clock;
Add CONFIG_MMU_TLB_REFCOUNT which enables refcounting of the lazy tlb mm when it is context switched. This can be disabled by architectures that don't require this refcounting if they clean up lazy tlb mms when the last refcount is dropped. Currently this is always enabled, which is what existing code does, so the patch is effectively a no-op. Rename rq->prev_mm to rq->prev_lazy_mm, because that's what it is. Signed-off-by: Nicholas Piggin <npiggin@gmail.com> --- arch/Kconfig | 4 ++++ include/linux/sched/mm.h | 13 +++++++++++-- kernel/sched/core.c | 22 ++++++++++++++++++---- kernel/sched/sched.h | 4 +++- 4 files changed, 36 insertions(+), 7 deletions(-)