introduce memory hinting API for external process
Subsystems (20)
v6 cover introduce memory hinting API for external process
Now, we have MADV_PAGEOUT and MADV_COLD as madvise hinting API. With that, application could give hints to kernel what memory range are preferred to be reclaimed. However, in some platform(e.g., Android), the information required to make the hinting decision is not known to the app.
Instead, it is known to a centralized userspace daemon(e.g., ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.
To solve the concern, this patch introduces new syscall - process_madvise(2). Bascially, it's same with madvise(2) syscall but it has some differences.
1. It needs pidfd of target process to provide the hint
2. It supports only MADV_{COLD|PAGEOUT|MERGEABLE|UNMEREABLE} at this moment. Other hints in madvise will be opened when there are explicit requests from community to prevent unexpected bugs we couldn't support.
3. Only privileged processes can do something for other process's address space.
For more detail of the new API, please see "mm: introduce external memory hinting API" description in this patchset.
* from v5 - https://lore.kernel.org/linux-mm/[email protected]/
* use null task and requestor's mm for io_madvise - Jann and Jens
* use right commit description for moving pidfd_get_pid - Christoph
* from v4 - https://lore.kernel.org/linux-mm/[email protected]/
* pass mm down to functions, not accessing task->mm - Jann
* clean up - Alexander
* add Reviewed-by - Alexander, SeongJae
* patch reordering
* from v3 - https://lore.kernel.org/linux-mm/[email protected]/
* verify task->mm aftere access_mm - Oleg
* split some patches for easy review - Alexander
* clean up fatal signal checking - Suren
* from v2 - https://lore.kernel.org/linux-mm/[email protected]/
* check signal callee and caller to bail out - Kirill Tkhai
* put more clarification for justification of new API
* from v1 - https://lore.kernel.org/linux-mm/[email protected]/
* fix syscall number - SeongJae
* use get_pid_task - Kirill Tkhai
* extend API to support pid as well as pidfd - Kirill Tkhai
Minchan Kim (5):
mm: pass task and mm to do_madvise
mm: introduce external memory hinting API
mm: check fatal signal pending of target process
pid: move pidfd_get_pid function to pid.c
mm: support both pid and pidfd for process_madvise
Oleksandr Natalenko (2):
mm/madvise: employ mmget_still_valid for write lock mm/madvise: allow KSM hints for remote API
arch/alpha/kernel/syscalls/syscall.tbl | 1 + arch/arm/tools/syscall.tbl | 1 + arch/arm64/include/asm/unistd.h | 2 +- arch/arm64/include/asm/unistd32.h | 2 + arch/ia64/kernel/syscalls/syscall.tbl | 1 + arch/m68k/kernel/syscalls/syscall.tbl | 1 + arch/microblaze/kernel/syscalls/syscall.tbl | 1 + arch/mips/kernel/syscalls/syscall_n32.tbl | 1 + arch/mips/kernel/syscalls/syscall_n64.tbl | 1 + arch/parisc/kernel/syscalls/syscall.tbl | 1 + arch/powerpc/kernel/syscalls/syscall.tbl | 1 + arch/s390/kernel/syscalls/syscall.tbl | 1 + arch/sh/kernel/syscalls/syscall.tbl | 1 + arch/sparc/kernel/syscalls/syscall.tbl | 1 + arch/x86/entry/syscalls/syscall_32.tbl | 1 + arch/x86/entry/syscalls/syscall_64.tbl | 1 + arch/xtensa/kernel/syscalls/syscall.tbl | 1 + fs/io_uring.c | 2 +- include/linux/mm.h | 3 +- include/linux/pid.h | 1 + include/linux/syscalls.h | 3 + include/uapi/asm-generic/unistd.h | 4 +- kernel/exit.c | 17 --- kernel/pid.c | 17 +++ kernel/sys_ni.c | 1 + mm/madvise.c | 144 ++++++++++++++++---- 26 files changed, 167 insertions(+), 44 deletions(-)
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
PATCH v6 1/7 mm: pass task and mm to do_madvise
In upcoming patches, do_madvise will be called from external process context so we shouldn't asssume "current" is always hinted process's task_struct. Furthermore, we couldn't access mm_struct via task->mm once it's verified by access_mm which will be introduced in next patch[1]. And let's pass *current* and current->mm as arguments of do_madvise so it shouldn't change existing behavior but prepare next patch to make review easy.
Note: io_madvise pass NULL as target_task argument of do_madvise because it couldn't know who is target.
[1] http://lore.kernel.org/r/CAG48ez27=pwm5m_N_988xT1huO7g7h6arTQL44zev6TD-h-7Tg@mail.gmail.com
--- fs/io_uring.c | 2 +- include/linux/mm.h | 3 ++- mm/madvise.c | 34 +++++++++++++++++++--------------- 3 files changed, 22 insertions(+), 17 deletions(-)
fs/io_uring.c +1 -1
@@ -2736,7 +2736,7 @@ static int io_madvise(struct io_kiocb *req, struct io_kiocb **nxt,if (force_nonblock)return -EAGAIN ; - ret =do_madvise (ma->addr, ma->len, ma->advice); + ret =do_madvise (NULL , req->work.mm, ma->addr, ma->len, ma->advice);if (ret <0 )req_set_fail_links (req);io_cqring_add_event (req, ret);
include/linux/mm.h +2 -1
@@ -2323,7 +2323,8 @@ extern int __do_munmap(struct mm_struct *, unsigned long, size_t, struct list_head *uf, bool downgrade);extern int do_munmap (struct mm_struct *,unsigned long ,size_t ,struct list_head *uf); -extern int do_madvise (unsigned long start,size_t len_in,int behavior); +extern int do_madvise (struct task_struct *target_task,struct mm_struct *mm, +unsigned long start,size_t len_in,int behavior); static inline unsigned long do_mmap_pgoff(struct file *file, unsigned long addr,
mm/madvise.c +19 -15
@@ -254,6 +254,7 @@ static long madvise_willneed(struct vm_area_struct *vma,struct vm_area_struct **prev, unsigned long start, unsigned long end) { + struct mm_struct *mm = vma->vm_mm;struct file *file = vma->vm_file;loff_t offset;
@@ -288,12 +289,12 @@ static long madvise_willneed(struct vm_area_struct *vma,*/ *prev =NULL ;/* tell sys_madvise we drop mmap_sem */ get_file (file); -up_read (¤t->mm->mmap_sem); +up_read (&mm->mmap_sem); offset = (loff_t)(start - vma->vm_start) + ((loff_t )vma->vm_pgoff <<PAGE_SHIFT );vfs_fadvise (file, offset, end - start,POSIX_FADV_WILLNEED );fput (file); -down_read (¤t->mm->mmap_sem); +down_read (&mm->mmap_sem);return 0 ; }
@@ -676,7 +677,6 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,if (nr_swap) {if (current->mm == mm)sync_mm_rss (mm); -add_mm_counter (mm,MM_SWAPENTS , nr_swap); }arch_leave_lazy_mmu_mode ();
@@ -756,6 +756,8 @@ static long madvise_dontneed_free(struct vm_area_struct *vma, unsigned long start, unsigned long end, int behavior) { + struct mm_struct *mm = vma->vm_mm; + *prev = vma;if (!can_madv_lru_vma (vma))return -EINVAL ;
@@ -763,8 +765,8 @@ static long madvise_dontneed_free(struct vm_area_struct *vma,if (!userfaultfd_remove (vma, start, end)) { *prev =NULL ;/* mmap_sem has been dropped, prev is stale */ -down_read (¤t->mm->mmap_sem); - vma =find_vma (current->mm, start); +down_read (&mm->mmap_sem); + vma =find_vma (mm, start);if (!vma)return -ENOMEM ;if (start < vma->vm_start) {
@@ -818,6 +820,7 @@ static long madvise_remove(struct vm_area_struct *vma,loff_t offset;int error;struct file *f; +struct mm_struct *mm = vma->vm_mm; *prev =NULL ;/* tell sys_madvise we drop mmap_sem */
@@ -845,13 +848,13 @@ static long madvise_remove(struct vm_area_struct *vma,get_file (f);if (userfaultfd_remove (vma, start, end)) {/* mmap_sem was not released by userfaultfd_remove() */ -up_read (¤t->mm->mmap_sem); +up_read (&mm->mmap_sem); } error =vfs_fallocate (f,FALLOC_FL_PUNCH_HOLE |FALLOC_FL_KEEP_SIZE , offset, end - start);fput (f); -down_read (¤t->mm->mmap_sem); +down_read (&mm->mmap_sem);return error; }
@@ -1044,7 +1047,8 @@ madvise_behavior_valid(int behavior)* -EBADF - map exists, but area maps something that isn't a file. * -EAGAIN - a kernel resource was temporarily unavailable. */ -int do_madvise (unsigned long start,size_t len_in,int behavior) +int do_madvise (struct task_struct *target_task,struct mm_struct *mm, +unsigned long start,size_t len_in,int behavior) {unsigned long end, tmp;struct vm_area_struct *vma, *prev;
@@ -1082,10 +1086,10 @@ int do_madvise(unsigned long start, size_t len_in, int behavior) write =madvise_need_mmap_write (behavior);if (write) { -if (down_write_killable (¤t->mm->mmap_sem)) +if (down_write_killable (&mm->mmap_sem))return -EINTR ; }else { -down_read (¤t->mm->mmap_sem); +down_read (&mm->mmap_sem); }/*
@@ -1093,7 +1097,7 @@ int do_madvise(unsigned long start, size_t len_in, int behavior)* ranges, just ignore them, but return -ENOMEM at the end. * - different from the way of handling in mlock etc. */ - vma =find_vma_prev (current->mm, start, &prev); + vma =find_vma_prev (mm, start, &prev);if (vma && start > vma->vm_start) prev = vma;
@@ -1130,19 +1134,19 @@ int do_madvise(unsigned long start, size_t len_in, int behavior)if (prev) vma = prev->vm_next;else /* madvise_remove dropped mmap_sem */ - vma =find_vma (current->mm, start); + vma =find_vma (mm, start); } out:blk_finish_plug (&plug);if (write) -up_write (¤t->mm->mmap_sem); +up_write (&mm->mmap_sem);else -up_read (¤t->mm->mmap_sem); +up_read (&mm->mmap_sem);return error; } SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior) { -return do_madvise (start, len_in, behavior); +return do_madvise (current, current->mm, start, len_in, behavior); }
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 1/7] mm: pass task and mm to do_madvise
Quoted 186 lines · Minchan Kim
In upcoming patches,
do_madvisewill be called from external process context so we shouldn't asssume "current" is always hinted process'stask_struct. Furthermore, we couldn't accessmm_structviatask->mmonce it's verified byaccess_mmwhich will be introduced in next patch[1]. And let's pass *current* andcurrent->mmas arguments ofdo_madviseso it shouldn't change existing behavior but prepare next patch to make review easy.Note:
io_madvisepass NULL astarget_taskargument ofdo_madvisebecause it couldn't know who is target.[1] http://lore.kernel.org/r/CAG48ez27=pwm5m_N_988xT1huO7g7h6arTQL44zev6TD-h-7Tg@mail.gmail.com
Cc Jens Axboe <[email protected]>Cc Jann Horn <[email protected]>Signed-off-by Minchan Kim <[email protected]>--- fs/io_uring.c | 2 +- include/linux/mm.h | 3 ++- mm/madvise.c | 34 +++++++++++++++++++--------------- 3 files changed, 22 insertions(+), 17 deletions(-)fs/io_uring.c +1
-1@@ -2736,7 +2736,7 @@ static int io_madvise(struct io_kiocb *req, struct io_kiocb **nxt,if (force_nonblock)return -EAGAIN ; - ret =do_madvise (ma->addr, ma->len, ma->advice); + ret =do_madvise (NULL , req->work.mm, ma->addr, ma->len, ma->advice);if (ret <0 )req_set_fail_links (req);io_cqring_add_event (req, ret);include/linux/mm.h +2
-1@@ -2323,7 +2323,8 @@ extern int __do_munmap(struct mm_struct *, unsigned long, size_t, struct list_head *uf, bool downgrade);extern int do_munmap (struct mm_struct *,unsigned long ,size_t ,struct list_head *uf); -extern int do_madvise (unsigned long start,size_t len_in,int behavior); +extern int do_madvise (struct task_struct *target_task,struct mm_struct *mm, +unsigned long start,size_t len_in,int behavior); static inline unsigned long do_mmap_pgoff(struct file *file, unsigned long addr,mm/madvise.c +19
-16@@ -254,6 +254,7 @@ static long madvise_willneed(struct vm_area_struct *vma,struct vm_area_struct **prev, unsigned long start, unsigned long end) { + struct mm_struct *mm = vma->vm_mm;struct file *file = vma->vm_file;loff_t offset;@@ -288,12 +289,12 @@ static long madvise_willneed(struct vm_area_struct *vma,*/ *prev =NULL ;/* tell sys_madvise we drop mmap_sem */ get_file (file); -up_read (¤t->mm->mmap_sem); +up_read (&mm->mmap_sem); offset = (loff_t)(start - vma->vm_start) + ((loff_t )vma->vm_pgoff <<PAGE_SHIFT );vfs_fadvise (file, offset, end - start,POSIX_FADV_WILLNEED );fput (file); -down_read (¤t->mm->mmap_sem); +down_read (&mm->mmap_sem);return 0 ; }@@ -676,7 +677,6 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned long addr,if (nr_swap) {if (current->mm == mm)sync_mm_rss (mm); -add_mm_counter (mm,MM_SWAPENTS , nr_swap); }arch_leave_lazy_mmu_mode ();@@ -756,6 +756,8 @@ static long madvise_dontneed_free(struct vm_area_struct *vma, unsigned long start, unsigned long end, int behavior) { + struct mm_struct *mm = vma->vm_mm; + *prev = vma;if (!can_madv_lru_vma (vma))return -EINVAL ;@@ -763,8 +765,8 @@ static long madvise_dontneed_free(struct vm_area_struct *vma,if (!userfaultfd_remove (vma, start, end)) { *prev =NULL ;/* mmap_sem has been dropped, prev is stale */ -down_read (¤t->mm->mmap_sem); - vma =find_vma (current->mm, start); +down_read (&mm->mmap_sem); + vma =find_vma (mm, start);if (!vma)return -ENOMEM ;if (start < vma->vm_start) {@@ -818,6 +820,7 @@ static long madvise_remove(struct vm_area_struct *vma,loff_t offset;int error;struct file *f; +struct mm_struct *mm = vma->vm_mm; *prev =NULL ;/* tell sys_madvise we drop mmap_sem */ @@ -845,13 +848,13 @@ static long madvise_remove(struct vm_area_struct *vma,get_file (f);if (userfaultfd_remove (vma, start, end)) {/* mmap_sem was not released by userfaultfd_remove() */ -up_read (¤t->mm->mmap_sem); +up_read (&mm->mmap_sem); } error =vfs_fallocate (f,FALLOC_FL_PUNCH_HOLE |FALLOC_FL_KEEP_SIZE , offset, end - start);fput (f); -down_read (¤t->mm->mmap_sem); +down_read (&mm->mmap_sem);return error; }@@ -1044,7 +1047,8 @@ madvise_behavior_valid(int behavior)* -EBADF - map exists, but area maps something that isn't a file. * -EAGAIN - a kernel resource was temporarily unavailable. */ -int do_madvise (unsigned long start,size_t len_in,int behavior) +int do_madvise (struct task_struct *target_task,struct mm_struct *mm, +unsigned long start,size_t len_in,int behavior) {unsigned long end, tmp;struct vm_area_struct *vma, *prev;@@ -1082,10 +1086,10 @@ int do_madvise(unsigned long start, size_t len_in, int behavior) write =madvise_need_mmap_write (behavior);if (write) { -if (down_write_killable (¤t->mm->mmap_sem)) +if (down_write_killable (&mm->mmap_sem))return -EINTR ; }else { -down_read (¤t->mm->mmap_sem); +down_read (&mm->mmap_sem); }/* @@ -1093,7 +1097,7 @@ int do_madvise(unsigned long start, size_t len_in, int behavior)* ranges, just ignore them, but return -ENOMEM at the end. * - different from the way of handling in mlock etc. */ - vma =find_vma_prev (current->mm, start, &prev); + vma =find_vma_prev (mm, start, &prev);if (vma && start > vma->vm_start) prev = vma;@@ -1130,19 +1134,19 @@ int do_madvise(unsigned long start, size_t len_in, int behavior)if (prev) vma = prev->vm_next;else /* madvise_remove dropped mmap_sem */ - vma =find_vma (current->mm, start); + vma =find_vma (mm, start); } out:blk_finish_plug (&plug);if (write) -up_write (¤t->mm->mmap_sem); +up_write (&mm->mmap_sem);else -up_read (¤t->mm->mmap_sem); +up_read (&mm->mmap_sem);return error; } SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior) { -return do_madvise (start, len_in, behavior); +return do_madvise (current, current->mm, start, len_in, behavior); } --
Quoted 2 lines
2.25.0.265.gbab2e86ba0-goog
PATCH v6 2/7 mm: introduce external memory hinting API
In reply to Minchan KimThere is usecase that System Management Software(SMS) want to give a memory hint like MADV_[COLD|PAGEEOUT] to other processes and in the case of Android, it is the ActivityManagerService.
It's similar in spirit to madvise(MADV_WONTNEED), but the information required to make the reclaim decision is not known to the app. Instead, it is known to the centralized userspace daemon(ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.
To solve the issue, this patch introduces a new syscall process_madvise(2). It uses pidfd of an external process to give the hint.
int process_madvise(int pidfd, void *addr, size_t length, int advise,
unsigned long flag);
Since it could affect other process's address range, only privileged process(CAP_SYS_PTRACE) or something else(e.g., being the same UID) gives it the right to ptrace the process could use it successfully. The flag argument is reserved for future use if we need to extend the API.
I think supporting all hints madvise has/will supported/support to process_madvise is rather risky. Because we are not sure all hints make sense from external process and implementation for the hint may rely on the caller being in the current context so it could be error-prone. Thus, I just limited hints as MADV_[COLD|PAGEOUT] in this patch.
If someone want to add other hints, we could hear hear the usecase and review it for each hint. It's safer for maintenance rather than introducing a buggy syscall but hard to fix it later.
Q.1 - Why does any external entity have better knowledge?
Quote from Sandeep
"For Android, every application (including the special SystemServer) are forked from Zygote. The reason of course is to share as many libraries and classes between the two as possible to benefit from the preloading during boot.
After applications start, (almost) all of the APIs end up calling into this SystemServer process over IPC (binder) and back to the application.
In a fully running system, the SystemServer monitors every single process periodically to calculate their PSS / RSS and also decides which process is "important" to the user for interactivity.
So, because of how these processes start _and_ the fact that the SystemServer is looping to monitor each process, it does tend to *know* which address range of the application is not used / useful.
Besides, we can never rely on applications to clean things up themselves. We've had the "hey app1, the system is low on memory, please trim your memory usage down" notifications for a long time[1]. They rely on applications honoring the broadcasts and very few do.
So, if we want to avoid the inevitable killing of the application and restarting it, some way to be able to tell the OS about unimportant memory in these applications will be useful.
- ssp
Q.2 - How to guarantee the race(i.e., object validation) between when giving a hint from an external process and get the hint from the target process?
process_madvise operates on the target process's address space as it exists at the instant that process_madvise is called. If the space target process can run between the time the process_madvise process inspects the target process address space and the time that process_madvise is actually called, process_madvise may operate on memory regions that the calling process does not expect. It's the responsibility of the process calling process_madvise to close this race condition. For example, the calling process can suspend the target process with ptrace, SIGSTOP, or the freezer cgroup so that it doesn't have an opportunity to change its own address space before process_madvise is called. Another option is to operate on memory regions that the caller knows a priori will be unchanged in the target process. Yet another option is to accept the race for certain process_madvise calls after reasoning that mistargeting will do no harm. The suggested API itself does not provide synchronization. It also apply other APIs like move_pages, process_vm_write.
The race isn't really a problem though. Why is it so wrong to require that callers do their own synchronization in some manner? Nobody objects to write(2) merely because it's possible for two processes to open the same file and clobber each other's writes --- instead, we tell people to use flock or something. Think about mmap. It never guarantees newly allocated address space is still valid when the user tries to access it because other threads could unmap the memory right before. That's where we need synchronization by using other API or design from userside. It shouldn't be part of API itself. If someone needs more fine-grained synchronization rather than process level, there were two ideas suggested - cookie[2] and anon-fd[3]. Both are applicable via using last reserved argument of the API but I don't think it's necessary right now since we have already ways to prevent the race so don't want to add additional complexity with more fine-grained optimization model.
To make the API extend, it reserved an unsigned long as last argument so we could support it in future if someone really needs it.
Q.3 - Why doesn't ptrace work?
Injecting an madvise in the target process using ptrace would not work for us because such injected madvise would have to be executed by the target process, which means that process would have to be runnable and that creates the risk of the abovementioned race and hinting a wrong VMA. Furthermore, we want to act the hint in caller's context, not calle because calle is usually limited in cpuset/cgroups or even freezed state so they can't act by themselves quick enough, which causes more thrashing/kill. It doesn't work if the target process are ptraced(e.g., strace, debugger, minidump) because a process can have at most one ptracer.
[1] https://developer.android.com/topic/performance/memory"
[2] process_getinfo for getting the cookie which is updated whenever
vma of process address layout are changed - Daniel Colascione
- https://lore.kernel.org/lkml/[email protected]/T/#m7694416fd179b2066a2c62b5b139b14e3894e224
[3] anonymous fd which is used for the object(i.e., address range)
validation - Michal Hocko
- https://lore.kernel.org/lkml/[email protected]/
--- arch/alpha/kernel/syscalls/syscall.tbl | 1 + arch/arm/tools/syscall.tbl | 1 + arch/arm64/include/asm/unistd.h | 2 +- arch/arm64/include/asm/unistd32.h | 2 + arch/ia64/kernel/syscalls/syscall.tbl | 1 + arch/m68k/kernel/syscalls/syscall.tbl | 1 + arch/microblaze/kernel/syscalls/syscall.tbl | 1 + arch/mips/kernel/syscalls/syscall_n32.tbl | 1 + arch/mips/kernel/syscalls/syscall_n64.tbl | 1 + arch/parisc/kernel/syscalls/syscall.tbl | 1 + arch/powerpc/kernel/syscalls/syscall.tbl | 1 + arch/s390/kernel/syscalls/syscall.tbl | 1 + arch/sh/kernel/syscalls/syscall.tbl | 1 + arch/sparc/kernel/syscalls/syscall.tbl | 1 + arch/x86/entry/syscalls/syscall_32.tbl | 1 + arch/x86/entry/syscalls/syscall_64.tbl | 1 + arch/xtensa/kernel/syscalls/syscall.tbl | 1 + include/linux/syscalls.h | 2 + include/uapi/asm-generic/unistd.h | 4 +- kernel/sys_ni.c | 1 + mm/madvise.c | 64 +++++++++++++++++++++ 21 files changed, 88 insertions(+), 2 deletions(-)
arch/alpha/kernel/syscalls/syscall.tbl +1 -0
@@ -477,3 +477,4 @@ # 545 reserved for clone3 547 common openat2 sys_openat2 548 common pidfd_getfd sys_pidfd_getfd +549 common process_madvise sys_process_madvise
arch/arm/tools/syscall.tbl +1 -0
@@ -451,3 +451,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/arm64/include/asm/unistd.h +1 -1
@@ -38,7 +38,7 @@#define __ARM_NR_compat_set_tls (__ARM_NR_COMPAT_BASE + 5)#define __ARM_NR_COMPAT_END (__ARM_NR_COMPAT_BASE + 0x800) -#define __NR_compat_syscalls 439 +#define __NR_compat_syscalls 440#endif #define __ARCH_WANT_SYS_CLONE
arch/arm64/include/asm/unistd32.h +2 -0
@@ -883,6 +883,8 @@ __SYSCALL(__NR_clone3, sys_clone3) __SYSCALL(__NR_openat2, sys_openat2)#define __NR_pidfd_getfd 438 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd) +#define __NR_process_madvise 439 +__SYSCALL (__NR_process_madvise, process_madvise)/* * Please add new compat syscalls above this comment and update
arch/ia64/kernel/syscalls/syscall.tbl +1 -0
@@ -358,3 +358,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/m68k/kernel/syscalls/syscall.tbl +1 -0
@@ -437,3 +437,4 @@ 435 common clone3 __sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/microblaze/kernel/syscalls/syscall.tbl +1 -0
@@ -443,3 +443,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/mips/kernel/syscalls/syscall_n32.tbl +1 -0
@@ -376,3 +376,4 @@ 435 n32 clone3 __sys_clone3 437 n32 openat2 sys_openat2 438 n32 pidfd_getfd sys_pidfd_getfd +439 n32 process_madvise sys_process_madvise
arch/mips/kernel/syscalls/syscall_n64.tbl +1 -0
@@ -352,3 +352,4 @@ 435 n64 clone3 __sys_clone3 437 n64 openat2 sys_openat2 438 n64 pidfd_getfd sys_pidfd_getfd +439 n64 process_madvise sys_process_madvise
arch/parisc/kernel/syscalls/syscall.tbl +1 -0
@@ -435,3 +435,4 @@ 435 common clone3 sys_clone3_wrapper 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/powerpc/kernel/syscalls/syscall.tbl +1 -0
@@ -519,3 +519,4 @@ 435 nospu clone3 ppc_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/s390/kernel/syscalls/syscall.tbl +1 -0
@@ -440,3 +440,4 @@ 435 common clone3 sys_clone3 sys_clone3 437 common openat2 sys_openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise sys_process_madvise
arch/sh/kernel/syscalls/syscall.tbl +1 -0
@@ -440,3 +440,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/sparc/kernel/syscalls/syscall.tbl +1 -0
@@ -483,3 +483,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/x86/entry/syscalls/syscall_32.tbl +1 -0
@@ -442,3 +442,4 @@ 435 i386 clone3 sys_clone3 __ia32_sys_clone3 437 i386 openat2 sys_openat2 __ia32_sys_openat2 438 i386 pidfd_getfd sys_pidfd_getfd __ia32_sys_pidfd_getfd +439 i386 process_madvise sys_process_madvise __ia32_sys_process_madvise
arch/x86/entry/syscalls/syscall_64.tbl +1 -0
@@ -359,6 +359,7 @@ 435 common clone3 __x64_sys_clone3/ptregs 437 common openat2 __x64_sys_openat2 438 common pidfd_getfd __x64_sys_pidfd_getfd +439 common process_madvise __x64_sys_process_madvise # # x32-specific system call numbers start at 512 to avoid cache impact
arch/xtensa/kernel/syscalls/syscall.tbl +1 -0
@@ -408,3 +408,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
include/linux/syscalls.h +2 -0
@@ -876,6 +876,8 @@ asmlinkage long sys_munlockall(void);asmlinkage long sys_mincore (unsigned long start,size_t len,unsigned char __user * vec);asmlinkage long sys_madvise (unsigned long start,size_t len,int behavior); +asmlinkage long sys_process_madvise (int pidfd,unsigned long start, +size_t len,int behavior,unsigned long flags);asmlinkage long sys_remap_file_pages (unsigned long start,unsigned long size,unsigned long prot,unsigned long pgoff,unsigned long flags);
include/uapi/asm-generic/unistd.h +3 -1
@@ -855,9 +855,11 @@ __SYSCALL(__NR_clone3, sys_clone3) __SYSCALL(__NR_openat2, sys_openat2)#define __NR_pidfd_getfd 438 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd) +#define __NR_pidfd_getfd 439 +__SYSCALL(__NR_process_madvise, sys_process_madvise)#undef __NR_syscalls -#define __NR_syscalls 439 +#define __NR_syscalls 440/* * 32 bit systems traditionally used different
kernel/sys_ni.c +1 -0
@@ -280,6 +280,7 @@ COND_SYSCALL(mlockall);COND_SYSCALL (munlockall);COND_SYSCALL (mincore);COND_SYSCALL (madvise); +COND_SYSCALL (process_madvise);COND_SYSCALL (remap_file_pages);COND_SYSCALL (mbind);COND_SYSCALL_COMPAT (mbind);
mm/madvise.c +64 -0
85 lines not on this page · raw message
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
kbuild test robot
build botRe: [PATCH v6 2/7] mm: introduce external memory hinting API
I love your patch! Perhaps something to improve:
Hi Minchan,
I love your patch! Perhaps something to improve:
[auto build test WARNING on m68k/for-next]
[also build test WARNING on powerpc/next s390/features linus/master v5.6-rc2 next-20200220] [cannot apply to arm64/for-next/core tip/x86/asm arm/for-next hp-parisc/for-next] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system. BTW, we also suggest to use '--base' option to specify the base tree in git format-patch, please see https://stackoverflow.com/a/37406982]
url: https://github.com/0day-ci/linux/commits/Minchan-Kim/introduce-memory-hinting-API-for-external-process/20200220-225155 base: https://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k.git for-next config: nds32-randconfig-a001-20200220 (attached as .config) compiler: nds32le-linux-gcc (GCC) 9.2.0
reproduce:
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
GCC_VERSION=9.2.0 make.cross ARCH=nds32
If you fix the issue, kindly add following tag
All warnings (new ones prefixed by >>):
In file included from arch/nds32/include/uapi/asm/unistd.h:10,
from arch/nds32/include/asm/unistd.h:6,
from arch/nds32/kernel/vdso/sigreturn.S:6:
>> include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined
858 | #define __NR_pidfd_getfd 439
|
include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition
856 | #define __NR_pidfd_getfd 438
|
--
In file included from arch/nds32/include/uapi/asm/unistd.h:10,
from arch/nds32/include/asm/unistd.h:6,
from <stdin>:2:
>> include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined
858 | #define __NR_pidfd_getfd 439
|
include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition
856 | #define __NR_pidfd_getfd 438
|
<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
>> <stdin>:1520:2: warning: #warning syscall process_madvise not implemented [-Wcpp]
--
In file included from arch/nds32/include/uapi/asm/unistd.h:10,
from arch/nds32/include/asm/unistd.h:6,
from <stdin>:2:
>> include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined
858 | #define __NR_pidfd_getfd 439
|
include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition
856 | #define __NR_pidfd_getfd 438
|
<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
>> <stdin>:1520:2: warning: #warning syscall process_madvise not implemented [-Wcpp]
16 real 2 user 4 sys 39.01% cpu make modules_prepare
--
In file included from arch/nds32/include/uapi/asm/unistd.h:10,
from arch/nds32/include/asm/unistd.h:6,
from <stdin>:2:
>> include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined
858 | #define __NR_pidfd_getfd 439
|
include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition
856 | #define __NR_pidfd_getfd 438
|
<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
>> <stdin>:1520:2: warning: #warning syscall process_madvise not implemented [-Wcpp]
In file included from arch/nds32/include/uapi/asm/unistd.h:10,
from arch/nds32/include/asm/unistd.h:6,
from arch/nds32/kernel/vdso/sigreturn.S:6:
>> include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined
858 | #define __NR_pidfd_getfd 439
|
include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition
856 | #define __NR_pidfd_getfd 438
|
In file included from arch/nds32/include/uapi/asm/unistd.h:10,
from arch/nds32/include/asm/unistd.h:6,
from arch/nds32/kernel/vdso/gettimeofday.c:11:
>> include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined
858 | #define __NR_pidfd_getfd 439
|
include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition
856 | #define __NR_pidfd_getfd 438
| 17 real 4 user 6 sys 59.28% cpu make prepare
vim +/__NR_pidfd_getfd +858 include/uapi/asm-generic/unistd.h
853
854 #define __NR_openat2 437
855 __SYSCALL(__NR_openat2, sys_openat2)
856 #define __NR_pidfd_getfd 438
857 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd)
> 858 #define __NR_pidfd_getfd 439
859 __SYSCALL(__NR_process_madvise, sys_process_madvise)
860
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
Re: [PATCH v6 2/7] mm: introduce external memory hinting API
Quoted 99 lines · kbuild test robot
Hi Minchan,
I love your patch! Perhaps something to improve:
[auto build test WARNING on m68k/for-next]
[also build test WARNING on powerpc/next s390/features linus/master v5.6-rc2 next-20200220] [cannot apply to arm64/for-next/core tip/x86/asm arm/for-next hp-parisc/for-next] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system. BTW, we also suggest to use '--base' option to specify the base tree in git format-patch, please see https://stackoverflow.com/a/37406982]url: https://github.com/0day-ci/linux/commits/Minchan-Kim/introduce-memory-hinting-API-for-external-process/20200220-225155 base: https://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k.git for-next config: nds32-randconfig-a001-20200220 (attached as .config) compiler: nds32le-linux-gcc (GCC) 9.2.0reproduce:
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
GCC_VERSION=9.2.0 make.cross ARCH=nds32If you fix the issue, kindly add following tag
Reported-by kbuild test robot <[email protected]>All warnings (new ones prefixed by >>):
In file included from
arch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/sigreturn.S:6:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 | -- In file included from arch/nds32/include/uapi/asm/unistd.h:10, from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]--
In file included fromarch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]16 real 2 user 4 sys 39.01% cpu make
modules_prepare
--
In file included fromarch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]In file included from
arch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/sigreturn.S:6:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 | In file included from arch/nds32/include/uapi/asm/unistd.h:10, from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/gettimeofday.c:11:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |17 real 4 user 6 sys 59.28% cpu make prepare
vim +/__NR_pidfd_getfd +858
include/uapi/asm-generic/unistd.h853 854 #define __NR_openat2 437 855 __SYSCALL(__NR_openat2, sys_openat2) 856 #define __NR_pidfd_getfd 438 857 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd)
Quoted 7 lines
Quoted 1 line
858 #define
__NR_pidfd_getfd439859
__SYSCALL(__NR_process_madvise, sys_process_madvise)
860---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
Hi 0-Day,
Thanks for catching this. Here the fix goes.
Quoted 1 line
From
ff74a830277d880716fa0f69c80e9ec9337303d5Mon Sep 17 00:00:00 2001
From: Minchan Kim <[email protected]>
Date: Fri, 14 Feb 2020 07:42:03 -0800
Subject: [PATCH v6 2/8] mm: introduce external memory hinting API
There is usecase that System Management Software(SMS) want to give a memory hint like MADV_[COLD|PAGEEOUT] to other processes and in the case of Android, it is the ActivityManagerService.
It's similar in spirit to madvise(MADV_WONTNEED), but the information required to make the reclaim decision is not known to the app. Instead, it is known to the centralized userspace daemon(ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.
To solve the issue, this patch introduces a new syscall process_madvise(2). It uses pidfd of an external process to give the hint.
int process_madvise(int pidfd, void *addr, size_t length, int advise,
unsigned long flag);
Since it could affect other process's address range, only privileged process(CAP_SYS_PTRACE) or something else(e.g., being the same UID) gives it the right to ptrace the process could use it successfully. The flag argument is reserved for future use if we need to extend the API.
I think supporting all hints madvise has/will supported/support to process_madvise is rather risky. Because we are not sure all hints make sense from external process and implementation for the hint may rely on the caller being in the current context so it could be error-prone. Thus, I just limited hints as MADV_[COLD|PAGEOUT] in this patch.
If someone want to add other hints, we could hear hear the usecase and review it for each hint. It's safer for maintenance rather than introducing a buggy syscall but hard to fix it later.
Q.1 - Why does any external entity have better knowledge?
Quote from Sandeep
"For Android, every application (including the special SystemServer) are forked from Zygote. The reason of course is to share as many libraries and classes between the two as possible to benefit from the preloading during boot.
After applications start, (almost) all of the APIs end up calling into this SystemServer process over IPC (binder) and back to the application.
In a fully running system, the SystemServer monitors every single process periodically to calculate their PSS / RSS and also decides which process is "important" to the user for interactivity.
So, because of how these processes start _and_ the fact that the SystemServer is looping to monitor each process, it does tend to *know* which address range of the application is not used / useful.
Besides, we can never rely on applications to clean things up themselves. We've had the "hey app1, the system is low on memory, please trim your memory usage down" notifications for a long time[1]. They rely on applications honoring the broadcasts and very few do.
So, if we want to avoid the inevitable killing of the application and restarting it, some way to be able to tell the OS about unimportant memory in these applications will be useful.
- ssp
Q.2 - How to guarantee the race(i.e., object validation) between when giving a hint from an external process and get the hint from the target process?
process_madvise operates on the target process's address space as it exists at the instant that process_madvise is called. If the space target process can run between the time the process_madvise process inspects the target process address space and the time that process_madvise is actually called, process_madvise may operate on memory regions that the calling process does not expect. It's the responsibility of the process calling process_madvise to close this race condition. For example, the calling process can suspend the target process with ptrace, SIGSTOP, or the freezer cgroup so that it doesn't have an opportunity to change its own address space before process_madvise is called. Another option is to operate on memory regions that the caller knows a priori will be unchanged in the target process. Yet another option is to accept the race for certain process_madvise calls after reasoning that mistargeting will do no harm. The suggested API itself does not provide synchronization. It also apply other APIs like move_pages, process_vm_write.
The race isn't really a problem though. Why is it so wrong to require that callers do their own synchronization in some manner? Nobody objects to write(2) merely because it's possible for two processes to open the same file and clobber each other's writes --- instead, we tell people to use flock or something. Think about mmap. It never guarantees newly allocated address space is still valid when the user tries to access it because other threads could unmap the memory right before. That's where we need synchronization by using other API or design from userside. It shouldn't be part of API itself. If someone needs more fine-grained synchronization rather than process level, there were two ideas suggested - cookie[2] and anon-fd[3]. Both are applicable via using last reserved argument of the API but I don't think it's necessary right now since we have already ways to prevent the race so don't want to add additional complexity with more fine-grained optimization model.
To make the API extend, it reserved an unsigned long as last argument so we could support it in future if someone really needs it.
Q.3 - Why doesn't ptrace work?
Injecting an madvise in the target process using ptrace would not work for us because such injected madvise would have to be executed by the target process, which means that process would have to be runnable and that creates the risk of the abovementioned race and hinting a wrong VMA. Furthermore, we want to act the hint in caller's context, not calle because calle is usually limited in cpuset/cgroups or even freezed state so they can't act by themselves quick enough, which causes more thrashing/kill. It doesn't work if the target process are ptraced(e.g., strace, debugger, minidump) because a process can have at most one ptracer.
[1] https://developer.android.com/topic/performance/memory"
[2] process_getinfo for getting the cookie which is updated whenever
vma of process address layout are changed - Daniel Colascione
- https://lore.kernel.org/lkml/20190520035254.57579-1-minchan(a)kernel.org/T/#m7694416fd179b2066a2c62b5b139b14e3894e224
[3] anonymous fd which is used for the object(i.e., address range)
validation - Michal Hocko
- https://lore.kernel.org/lkml/20200120112722.GY18451(a)dhcp22.suse.cz/
--- arch/alpha/kernel/syscalls/syscall.tbl | 1 + arch/arm/tools/syscall.tbl | 1 + arch/arm64/include/asm/unistd.h | 2 +- arch/arm64/include/asm/unistd32.h | 2 + arch/ia64/kernel/syscalls/syscall.tbl | 1 + arch/m68k/kernel/syscalls/syscall.tbl | 1 + arch/microblaze/kernel/syscalls/syscall.tbl | 1 + arch/mips/kernel/syscalls/syscall_n32.tbl | 1 + arch/mips/kernel/syscalls/syscall_n64.tbl | 1 + arch/parisc/kernel/syscalls/syscall.tbl | 1 + arch/powerpc/kernel/syscalls/syscall.tbl | 1 + arch/s390/kernel/syscalls/syscall.tbl | 1 + arch/sh/kernel/syscalls/syscall.tbl | 1 + arch/sparc/kernel/syscalls/syscall.tbl | 1 + arch/x86/entry/syscalls/syscall_32.tbl | 1 + arch/x86/entry/syscalls/syscall_64.tbl | 1 + arch/xtensa/kernel/syscalls/syscall.tbl | 1 + include/linux/syscalls.h | 2 + include/uapi/asm-generic/unistd.h | 4 +- kernel/sys_ni.c | 1 + mm/madvise.c | 64 +++++++++++++++++++++ 21 files changed, 88 insertions(+), 2 deletions(-)
arch/alpha/kernel/syscalls/syscall.tbl +1 -0
@@ -477,3 +477,4 @@ # 545 reserved for clone3 547 common openat2 sys_openat2 548 common pidfd_getfd sys_pidfd_getfd +549 common process_madvise sys_process_madvise
arch/arm/tools/syscall.tbl +1 -0
@@ -451,3 +451,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/arm64/include/asm/unistd.h +1 -1
@@ -38,7 +38,7 @@#define __ARM_NR_compat_set_tls (__ARM_NR_COMPAT_BASE + 5)#define __ARM_NR_COMPAT_END (__ARM_NR_COMPAT_BASE + 0x800) -#define __NR_compat_syscalls 439 +#define __NR_compat_syscalls 440#endif #define __ARCH_WANT_SYS_CLONE
arch/arm64/include/asm/unistd32.h +2 -0
@@ -883,6 +883,8 @@ __SYSCALL(__NR_clone3, sys_clone3) __SYSCALL(__NR_openat2, sys_openat2)#define __NR_pidfd_getfd 438 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd) +#define __NR_process_madvise 439 +__SYSCALL (__NR_process_madvise, process_madvise)/* * Please add new compat syscalls above this comment and update
arch/ia64/kernel/syscalls/syscall.tbl +1 -0
@@ -358,3 +358,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/m68k/kernel/syscalls/syscall.tbl +1 -0
@@ -437,3 +437,4 @@ 435 common clone3 __sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/microblaze/kernel/syscalls/syscall.tbl +1 -0
@@ -443,3 +443,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/mips/kernel/syscalls/syscall_n32.tbl +1 -0
@@ -376,3 +376,4 @@ 435 n32 clone3 __sys_clone3 437 n32 openat2 sys_openat2 438 n32 pidfd_getfd sys_pidfd_getfd +439 n32 process_madvise sys_process_madvise
arch/mips/kernel/syscalls/syscall_n64.tbl +1 -0
@@ -352,3 +352,4 @@ 435 n64 clone3 __sys_clone3 437 n64 openat2 sys_openat2 438 n64 pidfd_getfd sys_pidfd_getfd +439 n64 process_madvise sys_process_madvise
arch/parisc/kernel/syscalls/syscall.tbl +1 -0
@@ -435,3 +435,4 @@ 435 common clone3 sys_clone3_wrapper 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/powerpc/kernel/syscalls/syscall.tbl +1 -0
@@ -519,3 +519,4 @@ 435 nospu clone3 ppc_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/s390/kernel/syscalls/syscall.tbl +1 -0
@@ -440,3 +440,4 @@ 435 common clone3 sys_clone3 sys_clone3 437 common openat2 sys_openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise sys_process_madvise
arch/sh/kernel/syscalls/syscall.tbl +1 -0
@@ -440,3 +440,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/sparc/kernel/syscalls/syscall.tbl +1 -0
@@ -483,3 +483,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
arch/x86/entry/syscalls/syscall_32.tbl +1 -0
@@ -442,3 +442,4 @@ 435 i386 clone3 sys_clone3 __ia32_sys_clone3 437 i386 openat2 sys_openat2 __ia32_sys_openat2 438 i386 pidfd_getfd sys_pidfd_getfd __ia32_sys_pidfd_getfd +439 i386 process_madvise sys_process_madvise __ia32_sys_process_madvise
arch/x86/entry/syscalls/syscall_64.tbl +1 -0
@@ -359,6 +359,7 @@ 435 common clone3 __x64_sys_clone3/ptregs 437 common openat2 __x64_sys_openat2 438 common pidfd_getfd __x64_sys_pidfd_getfd +439 common process_madvise __x64_sys_process_madvise # # x32-specific system call numbers start at 512 to avoid cache impact
arch/xtensa/kernel/syscalls/syscall.tbl +1 -0
@@ -408,3 +408,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise
include/linux/syscalls.h +2 -0
@@ -876,6 +876,8 @@ asmlinkage long sys_munlockall(void);asmlinkage long sys_mincore (unsigned long start,size_t len,unsigned char __user * vec);asmlinkage long sys_madvise (unsigned long start,size_t len,int behavior); +asmlinkage long sys_process_madvise (int pidfd,unsigned long start, +size_t len,int behavior,unsigned long flags);asmlinkage long sys_remap_file_pages (unsigned long start,unsigned long size,unsigned long prot,unsigned long pgoff,unsigned long flags);
include/uapi/asm-generic/unistd.h +3 -1
@@ -855,9 +855,11 @@ __SYSCALL(__NR_clone3, sys_clone3) __SYSCALL(__NR_openat2, sys_openat2)#define __NR_pidfd_getfd 438 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd) +#define __NR_process_madvise 439 +__SYSCALL(__NR_process_madvise, sys_process_madvise)#undef __NR_syscalls -#define __NR_syscalls 439 +#define __NR_syscalls 440/* * 32 bit systems traditionally used different
kernel/sys_ni.c +1 -0
@@ -280,6 +280,7 @@ COND_SYSCALL(mlockall);COND_SYSCALL (munlockall);COND_SYSCALL (mincore);COND_SYSCALL (madvise); +COND_SYSCALL (process_madvise);COND_SYSCALL (remap_file_pages);COND_SYSCALL (mbind);COND_SYSCALL_COMPAT (mbind);
mm/madvise.c +64 -0
85 lines not on this page · raw message
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 2/7] mm: introduce external memory hinting API
Hi Andrew,
I submit the fix with inlining here thread but if you prefer submitting new v7 revision with more inputs, please tell me.
I am happy to resend whole patchset.
Thanks.
Quoted 563 lines · Minchan Kim
Quoted 99 lines · kbuild test robot
Hi Minchan,
I love your patch! Perhaps something to improve:
[auto build test WARNING on m68k/for-next]
[also build test WARNING on powerpc/next s390/features linus/master v5.6-rc2 next-20200220] [cannot apply to arm64/for-next/core tip/x86/asm arm/for-next hp-parisc/for-next] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system. BTW, we also suggest to use '--base' option to specify the base tree in git format-patch, please see https://stackoverflow.com/a/37406982]url: https://github.com/0day-ci/linux/commits/Minchan-Kim/introduce-memory-hinting-API-for-external-process/20200220-225155 base: https://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k.git for-next config: nds32-randconfig-a001-20200220 (attached as .config) compiler: nds32le-linux-gcc (GCC) 9.2.0reproduce:
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
GCC_VERSION=9.2.0 make.cross ARCH=nds32If you fix the issue, kindly add following tag
Reported-by kbuild test robot <[email protected]>All warnings (new ones prefixed by >>):
In file included from
arch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/sigreturn.S:6:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 | -- In file included from arch/nds32/include/uapi/asm/unistd.h:10, from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]--
In file included fromarch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]16 real 2 user 4 sys 39.01% cpu make
modules_prepare
--
In file included fromarch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]In file included from
arch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/sigreturn.S:6:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 | In file included from arch/nds32/include/uapi/asm/unistd.h:10, from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/gettimeofday.c:11:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |17 real 4 user 6 sys 59.28% cpu make prepare
vim +/__NR_pidfd_getfd +858
include/uapi/asm-generic/unistd.h853 854 #define __NR_openat2 437 855 __SYSCALL(__NR_openat2, sys_openat2) 856 #define __NR_pidfd_getfd 438 857 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd)Quoted 7 lines
Quoted 1 line
858 #define
__NR_pidfd_getfd439859
__SYSCALL(__NR_process_madvise, sys_process_madvise)
860---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.orgHi 0-Day,
Thanks for catching this. Here the fix goes.
From
ff74a830277d880716fa0f69c80e9ec9337303d5Mon Sep 17 00:00:00 2001
From: Minchan Kim <[email protected]>
Date: Fri, 14 Feb 2020 07:42:03 -0800
Subject: [PATCH v6 2/8] mm: introduce external memory hinting APIThere is usecase that System Management Software(SMS) want to give a memory hint like MADV_[COLD|PAGEEOUT] to other processes and in the case of Android, it is the ActivityManagerService.
It's similar in spirit to
madvise(MADV_WONTNEED), but the information required to make the reclaim decision is not known to the app. Instead, it is known to the centralized userspace daemon(ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.To solve the issue, this patch introduces a new syscall
process_madvise(2). It uses pidfd of an external process to give the hint.int process_madvise(int pidfd, void *addr,
size_tlength, int advise,unsigned long flag);Since it could affect other process's address range, only privileged
process(CAP_SYS_PTRACE)or somethingelse(e.g., being the same UID)gives it the right to ptrace the process could use it successfully. The flag argument is reserved for future use if we need to extend the API.I think supporting all hints madvise has/will supported/support to
process_madviseis rather risky. Because we are not sure all hints make sense from external process and implementation for the hint may rely on the caller being in the current context so it could be error-prone. Thus, I just limited hints as MADV_[COLD|PAGEOUT] in this patch.If someone want to add other hints, we could hear hear the usecase and review it for each hint. It's safer for maintenance rather than introducing a buggy syscall but hard to fix it later.
Q.1 - Why does any external entity have better knowledge?
Quote from Sandeep
"For Android, every application (including the special SystemServer) are forked from Zygote. The reason of course is to share as many libraries and classes between the two as possible to benefit from the preloading during boot.After applications start, (almost) all of the APIs end up calling into this SystemServer process over IPC (binder) and back to the application.
In a fully running system, the SystemServer monitors every single process periodically to calculate their PSS / RSS and also decides which process is "important" to the user for interactivity.
So, because of how these processes start _and_ the fact that the SystemServer is looping to monitor each process, it does tend to *know* which address range of the application is not used / useful.
Besides, we can never rely on applications to clean things up themselves. We've had the "hey app1, the system is low on memory, please trim your memory usage down" notifications for a long time[1]. They rely on applications honoring the broadcasts and very few do.
So, if we want to avoid the inevitable killing of the application and restarting it, some way to be able to tell the OS about unimportant memory in these applications will be useful.
- ssp
Q.2 - How to guarantee the
race(i.e., object validation)between when giving a hint from an external process and get the hint from the target process?
process_madviseoperates on the target process's address space as it exists at the instant thatprocess_madviseis called. If the space target process can run between the time theprocess_madviseprocess inspects the target process address space and the time thatprocess_madviseis actually called,process_madvisemay operate on memory regions that the calling process does not expect. It's the responsibility of the process callingprocess_madviseto close this race condition. For example, the calling process can suspend the target process with ptrace, SIGSTOP, or the freezer cgroup so that it doesn't have an opportunity to change its own address space beforeprocess_madviseis called. Another option is to operate on memory regions that the caller knows a priori will be unchanged in the target process. Yet another option is to accept the race for certainprocess_madvisecalls after reasoning that mistargeting will do no harm. The suggested API itself does not provide synchronization. It also apply other APIs likemove_pages,process_vm_write.The race isn't really a problem though. Why is it so wrong to require that callers do their own synchronization in some manner? Nobody objects to write(2) merely because it's possible for two processes to open the same file and clobber each other's writes --- instead, we tell people to use flock or something. Think about mmap. It never guarantees newly allocated address space is still valid when the user tries to access it because other threads could unmap the memory right before. That's where we need synchronization by using other API or design from userside. It shouldn't be part of API itself. If someone needs more fine-grained synchronization rather than process level, there were two ideas suggested - cookie[2] and anon-fd[3]. Both are applicable via using last reserved argument of the API but I don't think it's necessary right now since we have already ways to prevent the race so don't want to add additional complexity with more fine-grained optimization model.
To make the API extend, it reserved an unsigned long as last argument so we could support it in future if someone really needs it.
Q.3 - Why doesn't ptrace work?
Injecting an madvise in the target process using ptrace would not work for us because such injected madvise would have to be executed by the target process, which means that process would have to be runnable and that creates the risk of the abovementioned race and hinting a wrong VMA. Furthermore, we want to act the hint in caller's context, not calle because calle is usually limited in cpuset/cgroups or even freezed state so they can't act by themselves quick enough, which causes more thrashing/kill. It doesn't work if the target process are
ptraced(e.g., strace, debugger, minidump)because a process can have at most one ptracer.[1] https://developer.android.com/topic/performance/memory"
[2]process_getinfofor getting the cookie which is updated whenever
vma of process address layout are changed - Daniel Colascione
- https://lore.kernel.org/lkml/20190520035254.57579-1-minchan(a)kernel.org/T/#m7694416fd179b2066a2c62b5b139b14e3894e224
[3] anonymous fd which is used for theobject(i.e., address range)
validation - Michal Hocko
- https://lore.kernel.org/lkml/20200120112722.GY18451(a)dhcp22.suse.cz/Signed-off-by Minchan Kim <[email protected]>--- arch/alpha/kernel/syscalls/syscall.tbl | 1 + arch/arm/tools/syscall.tbl | 1 + arch/arm64/include/asm/unistd.h | 2 +- arch/arm64/include/asm/unistd32.h | 2 + arch/ia64/kernel/syscalls/syscall.tbl | 1 + arch/m68k/kernel/syscalls/syscall.tbl | 1 + arch/microblaze/kernel/syscalls/syscall.tbl | 1 + arch/mips/kernel/syscalls/syscall_n32.tbl | 1 + arch/mips/kernel/syscalls/syscall_n64.tbl | 1 + arch/parisc/kernel/syscalls/syscall.tbl | 1 + arch/powerpc/kernel/syscalls/syscall.tbl | 1 + arch/s390/kernel/syscalls/syscall.tbl | 1 + arch/sh/kernel/syscalls/syscall.tbl | 1 + arch/sparc/kernel/syscalls/syscall.tbl | 1 + arch/x86/entry/syscalls/syscall_32.tbl | 1 + arch/x86/entry/syscalls/syscall_64.tbl | 1 + arch/xtensa/kernel/syscalls/syscall.tbl | 1 + include/linux/syscalls.h | 2 + include/uapi/asm-generic/unistd.h | 4 +- kernel/sys_ni.c | 1 + mm/madvise.c | 64 +++++++++++++++++++++ 21 files changed, 88 insertions(+), 2 deletions(-)arch/alpha/kernel/syscalls/syscall.tbl +1
-0@@ -477,3 +477,4 @@ # 545 reserved for clone3 547 common openat2 sys_openat2 548 common pidfd_getfd sys_pidfd_getfd +549 common process_madvise sys_process_madvisearch/arm/tools/syscall.tbl +1
-0@@ -451,3 +451,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/arm64/include/asm/unistd.h +1
-1@@ -38,7 +38,7 @@#define __ARM_NR_compat_set_tls (__ARM_NR_COMPAT_BASE + 5)#define __ARM_NR_COMPAT_END (__ARM_NR_COMPAT_BASE + 0x800) -#define __NR_compat_syscalls 439 +#define __NR_compat_syscalls 440#endif #define __ARCH_WANT_SYS_CLONEarch/arm64/include/asm/unistd32.h +2
-0@@ -883,6 +883,8 @@ __SYSCALL(__NR_clone3, sys_clone3) __SYSCALL(__NR_openat2, sys_openat2)#define __NR_pidfd_getfd 438 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd) +#define __NR_process_madvise 439 +__SYSCALL (__NR_process_madvise, process_madvise)/* * Please add new compat syscalls above this comment and update arch/ia64/kernel/syscalls/syscall.tbl +1
-0@@ -358,3 +358,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/m68k/kernel/syscalls/syscall.tbl +1
-0@@ -437,3 +437,4 @@ 435 common clone3 __sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/microblaze/kernel/syscalls/syscall.tbl +1
-0@@ -443,3 +443,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/mips/kernel/syscalls/syscall_n32.tbl +1
-0@@ -376,3 +376,4 @@ 435 n32 clone3 __sys_clone3 437 n32 openat2 sys_openat2 438 n32 pidfd_getfd sys_pidfd_getfd +439 n32 process_madvise sys_process_madvisearch/mips/kernel/syscalls/syscall_n64.tbl +1
-0@@ -352,3 +352,4 @@ 435 n64 clone3 __sys_clone3 437 n64 openat2 sys_openat2 438 n64 pidfd_getfd sys_pidfd_getfd +439 n64 process_madvise sys_process_madvisearch/parisc/kernel/syscalls/syscall.tbl +1
-0@@ -435,3 +435,4 @@ 435 common clone3 sys_clone3_wrapper 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/powerpc/kernel/syscalls/syscall.tbl +1
-0@@ -519,3 +519,4 @@ 435 nospu clone3 ppc_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/s390/kernel/syscalls/syscall.tbl +1
-0@@ -440,3 +440,4 @@ 435 common clone3 sys_clone3 sys_clone3 437 common openat2 sys_openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise sys_process_madvisearch/sh/kernel/syscalls/syscall.tbl +1
-0@@ -440,3 +440,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/sparc/kernel/syscalls/syscall.tbl +1
-0@@ -483,3 +483,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/x86/entry/syscalls/syscall_32.tbl +1
-0@@ -442,3 +442,4 @@ 435 i386 clone3 sys_clone3 __ia32_sys_clone3 437 i386 openat2 sys_openat2 __ia32_sys_openat2 438 i386 pidfd_getfd sys_pidfd_getfd __ia32_sys_pidfd_getfd +439 i386 process_madvise sys_process_madvise __ia32_sys_process_madvisearch/x86/entry/syscalls/syscall_64.tbl +1
-0@@ -359,6 +359,7 @@ 435 common clone3 __x64_sys_clone3/ptregs 437 common openat2 __x64_sys_openat2 438 common pidfd_getfd __x64_sys_pidfd_getfd +439 common process_madvise __x64_sys_process_madvise # # x32-specific system call numbers start at 512 to avoid cache impactarch/xtensa/kernel/syscalls/syscall.tbl +1
-0@@ -408,3 +408,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madviseinclude/linux/syscalls.h +2
-0@@ -876,6 +876,8 @@ asmlinkage long sys_munlockall(void);asmlinkage long sys_mincore (unsigned long start,size_t len,unsigned char __user * vec);asmlinkage long sys_madvise (unsigned long start,size_t len,int behavior); +asmlinkage long sys_process_madvise (int pidfd,unsigned long start, +size_t len,int behavior,unsigned long flags);asmlinkage long sys_remap_file_pages (unsigned long start,unsigned long size,unsigned long prot,unsigned long pgoff,unsigned long flags);include/uapi/asm-generic/unistd.h +3
-1@@ -855,9 +855,11 @@ __SYSCALL(__NR_clone3, sys_clone3) __SYSCALL(__NR_openat2, sys_openat2)#define __NR_pidfd_getfd 438 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd) +#define __NR_process_madvise 439 +__SYSCALL(__NR_process_madvise, sys_process_madvise)#undef __NR_syscalls -#define __NR_syscalls 439 +#define __NR_syscalls 440/* * 32 bit systems traditionally used different kernel/sys_ni.c +1
-0@@ -280,6 +280,7 @@ COND_SYSCALL(mlockall);COND_SYSCALL (munlockall);COND_SYSCALL (mincore);COND_SYSCALL (madvise); +COND_SYSCALL (process_madvise);COND_SYSCALL (remap_file_pages);COND_SYSCALL (mbind);COND_SYSCALL_COMPAT (mbind);mm/madvise.c +64
-0@@ -17,6 +17,7 @@#include <linux/falloc.h> #include <linux/fadvise.h> #include <linux/sched.h> +#include <linux/sched/mm.h> #include <linux/ksm.h> #include <linux/fs.h> #include <linux/file.h> @@ -986,6 +987,18 @@ madvise_behavior_valid(int behavior) } } +static bool +process_madvise_behavior_valid (int behavior) +{ +switch (behavior) { +case MADV_COLD : +case MADV_PAGEOUT : +return true; +default : +return false; + } +} +/* * The madvise(2) system call. * @@ -1033,6 +1046,11 @@ madvise_behavior_valid(int behavior)* MADV_DONTDUMP - the application wants to prevent pages in the given range * from being included in its core dump. * MADV_DODUMP - cancel MADV_DONTDUMP: no longer exclude from core dump. +* MADV_COLD - the application uses the memory less so the kernel can +* deactivate the memory to evict them quickly when the memory +* pressure happen. +* MADV_PAGEOUT - the application uses the memroy very rarely so kernel can +* page out the memory instantly. * * return values: * zero - success @@ -1150,3 +1168,49 @@ SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior) {return do_madvise (current, current->mm, start, len_in, behavior); } + +SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start, + size_t, len_in, int, behavior, unsigned long, flags) +{ +int ret; +struct fd f; +struct pid *pid; +struct task_struct *task; +struct mm_struct *mm; + +if (flags !=0 ) +return -EINVAL ; + +if (!process_madvise_behavior_valid (behavior)) +return -EINVAL ; + + f =fdget (pidfd); +if (!f.file) +return -EBADF ; + + pid =pidfd_pid (f.file); +if (IS_ERR (pid)) { + ret =PTR_ERR (pid); + goto fdput; + } + + task =get_pid_task (pid,PIDTYPE_PID ); +if (!task) { + ret = -ESRCH ; + goto fdput; + } + + mm =mm_access (task,PTRACE_MODE_ATTACH_FSCREDS ); +if (IS_ERR_OR_NULL (mm)) { + ret =IS_ERR (mm) ?PTR_ERR (mm) : -ESRCH ; + goto release_task; + } + + ret =do_madvise (task, mm, start, len_in, behavior); +mmput (mm); +release_task: +put_task_struct (task); +fdput: +fdput (f); +return ret; +}Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 2/7] mm: introduce external memory hinting API
Quoted 505 lines · Minchan Kim
Hi Andrew,
I submit the fix with inlining here thread but if you prefer submitting new v7 revision with more inputs, please tell me.
I am happy to resend whole patchset.Thanks.
Quoted 495 lines · Minchan Kim
Quoted 99 lines · kbuild test robot
Hi Minchan,
I love your patch! Perhaps something to improve:
[auto build test WARNING on m68k/for-next]
[also build test WARNING on powerpc/next s390/features linus/master v5.6-rc2 next-20200220] [cannot apply to arm64/for-next/core tip/x86/asm arm/for-next hp-parisc/for-next] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system. BTW, we also suggest to use '--base' option to specify the base tree in git format-patch, please see https://stackoverflow.com/a/37406982]url: https://github.com/0day-ci/linux/commits/Minchan-Kim/introduce-memory-hinting-API-for-external-process/20200220-225155 base: https://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k.git for-next config: nds32-randconfig-a001-20200220 (attached as .config) compiler: nds32le-linux-gcc (GCC) 9.2.0reproduce:
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
GCC_VERSION=9.2.0 make.cross ARCH=nds32If you fix the issue, kindly add following tag
Reported-by kbuild test robot <[email protected]>All warnings (new ones prefixed by >>):
In file included from
arch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/sigreturn.S:6:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 | -- In file included from arch/nds32/include/uapi/asm/unistd.h:10, from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]--
In file included fromarch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]16 real 2 user 4 sys 39.01% cpu make
modules_prepare
--
In file included fromarch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from <stdin>:2:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |<stdin>:1511:2: warning: #warning syscall clone3 not implemented [-Wcpp]
Quoted 1 line
Quoted 1 line
<stdin>:1520:2: warning: #warning syscall
process_madvisenot implemented [-Wcpp]In file included from
arch/nds32/include/uapi/asm/unistd.h:10,from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/sigreturn.S:6:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 | In file included from arch/nds32/include/uapi/asm/unistd.h:10, from arch/nds32/include/asm/unistd.h:6, from arch/nds32/kernel/vdso/gettimeofday.c:11:Quoted 1 line
Quoted 1 line
include/uapi/asm-generic/unistd.h:858: warning: "__NR_pidfd_getfd" redefined858 | #define __NR_pidfd_getfd 439 | include/uapi/asm-generic/unistd.h:856: note: this is the location of the previous definition 856 | #define __NR_pidfd_getfd 438 |17 real 4 user 6 sys 59.28% cpu make prepare
vim +/__NR_pidfd_getfd +858
include/uapi/asm-generic/unistd.h853 854 #define __NR_openat2 437 855 __SYSCALL(__NR_openat2, sys_openat2) 856 #define __NR_pidfd_getfd 438 857 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd)Quoted 7 lines
Quoted 1 line
858 #define
__NR_pidfd_getfd439859
__SYSCALL(__NR_process_madvise, sys_process_madvise)
860---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.orgHi 0-Day,
Thanks for catching this. Here the fix goes.
From
ff74a830277d880716fa0f69c80e9ec9337303d5Mon Sep 17 00:00:00 2001
From: Minchan Kim <[email protected]>
Date: Fri, 14 Feb 2020 07:42:03 -0800
Subject: [PATCH v6 2/8] mm: introduce external memory hinting APIThere is usecase that System Management Software(SMS) want to give a memory hint like MADV_[COLD|PAGEEOUT] to other processes and in the case of Android, it is the ActivityManagerService.
It's similar in spirit to
madvise(MADV_WONTNEED), but the information required to make the reclaim decision is not known to the app. Instead, it is known to the centralized userspace daemon(ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.To solve the issue, this patch introduces a new syscall
process_madvise(2). It uses pidfd of an external process to give the hint.int process_madvise(int pidfd, void *addr,
size_tlength, int advise,unsigned long flag);Since it could affect other process's address range, only privileged
process(CAP_SYS_PTRACE)or somethingelse(e.g., being the same UID)gives it the right to ptrace the process could use it successfully. The flag argument is reserved for future use if we need to extend the API.I think supporting all hints madvise has/will supported/support to
process_madviseis rather risky. Because we are not sure all hints make sense from external process and implementation for the hint may rely on the caller being in the current context so it could be error-prone. Thus, I just limited hints as MADV_[COLD|PAGEOUT] in this patch.If someone want to add other hints, we could hear hear the usecase and review it for each hint. It's safer for maintenance rather than introducing a buggy syscall but hard to fix it later.
Q.1 - Why does any external entity have better knowledge?
Quote from Sandeep
"For Android, every application (including the special SystemServer) are forked from Zygote. The reason of course is to share as many libraries and classes between the two as possible to benefit from the preloading during boot.After applications start, (almost) all of the APIs end up calling into this SystemServer process over IPC (binder) and back to the application.
In a fully running system, the SystemServer monitors every single process periodically to calculate their PSS / RSS and also decides which process is "important" to the user for interactivity.
So, because of how these processes start _and_ the fact that the SystemServer is looping to monitor each process, it does tend to *know* which address range of the application is not used / useful.
Besides, we can never rely on applications to clean things up themselves. We've had the "hey app1, the system is low on memory, please trim your memory usage down" notifications for a long time[1]. They rely on applications honoring the broadcasts and very few do.
So, if we want to avoid the inevitable killing of the application and restarting it, some way to be able to tell the OS about unimportant memory in these applications will be useful.
- ssp
Q.2 - How to guarantee the
race(i.e., object validation)between when giving a hint from an external process and get the hint from the target process?
process_madviseoperates on the target process's address space as it exists at the instant thatprocess_madviseis called. If the space target process can run between the time theprocess_madviseprocess inspects the target process address space and the time thatprocess_madviseis actually called,process_madvisemay operate on memory regions that the calling process does not expect. It's the responsibility of the process callingprocess_madviseto close this race condition. For example, the calling process can suspend the target process with ptrace, SIGSTOP, or the freezer cgroup so that it doesn't have an opportunity to change its own address space beforeprocess_madviseis called. Another option is to operate on memory regions that the caller knows a priori will be unchanged in the target process. Yet another option is to accept the race for certainprocess_madvisecalls after reasoning that mistargeting will do no harm. The suggested API itself does not provide synchronization. It also apply other APIs likemove_pages,process_vm_write.The race isn't really a problem though. Why is it so wrong to require that callers do their own synchronization in some manner? Nobody objects to write(2) merely because it's possible for two processes to open the same file and clobber each other's writes --- instead, we tell people to use flock or something. Think about mmap. It never guarantees newly allocated address space is still valid when the user tries to access it because other threads could unmap the memory right before. That's where we need synchronization by using other API or design from userside. It shouldn't be part of API itself. If someone needs more fine-grained synchronization rather than process level, there were two ideas suggested - cookie[2] and anon-fd[3]. Both are applicable via using last reserved argument of the API but I don't think it's necessary right now since we have already ways to prevent the race so don't want to add additional complexity with more fine-grained optimization model.
To make the API extend, it reserved an unsigned long as last argument so we could support it in future if someone really needs it.
Q.3 - Why doesn't ptrace work?
Injecting an madvise in the target process using ptrace would not work for us because such injected madvise would have to be executed by the target process, which means that process would have to be runnable and that creates the risk of the abovementioned race and hinting a wrong VMA. Furthermore, we want to act the hint in caller's context, not calle because calle is usually limited in cpuset/cgroups or even freezed state so they can't act by themselves quick enough, which causes more thrashing/kill. It doesn't work if the target process are
ptraced(e.g., strace, debugger, minidump)because a process can have at most one ptracer.[1] https://developer.android.com/topic/performance/memory"
[2]process_getinfofor getting the cookie which is updated whenever
vma of process address layout are changed - Daniel Colascione
- https://lore.kernel.org/lkml/20190520035254.57579-1-minchan(a)kernel.org/T/#m7694416fd179b2066a2c62b5b139b14e3894e224
[3] anonymous fd which is used for theobject(i.e., address range)
validation - Michal Hocko
- https://lore.kernel.org/lkml/20200120112722.GY18451(a)dhcp22.suse.cz/Signed-off-by Minchan Kim <[email protected]>--- arch/alpha/kernel/syscalls/syscall.tbl | 1 + arch/arm/tools/syscall.tbl | 1 + arch/arm64/include/asm/unistd.h | 2 +- arch/arm64/include/asm/unistd32.h | 2 + arch/ia64/kernel/syscalls/syscall.tbl | 1 + arch/m68k/kernel/syscalls/syscall.tbl | 1 + arch/microblaze/kernel/syscalls/syscall.tbl | 1 + arch/mips/kernel/syscalls/syscall_n32.tbl | 1 + arch/mips/kernel/syscalls/syscall_n64.tbl | 1 + arch/parisc/kernel/syscalls/syscall.tbl | 1 + arch/powerpc/kernel/syscalls/syscall.tbl | 1 + arch/s390/kernel/syscalls/syscall.tbl | 1 + arch/sh/kernel/syscalls/syscall.tbl | 1 + arch/sparc/kernel/syscalls/syscall.tbl | 1 + arch/x86/entry/syscalls/syscall_32.tbl | 1 + arch/x86/entry/syscalls/syscall_64.tbl | 1 + arch/xtensa/kernel/syscalls/syscall.tbl | 1 + include/linux/syscalls.h | 2 + include/uapi/asm-generic/unistd.h | 4 +- kernel/sys_ni.c | 1 + mm/madvise.c | 64 +++++++++++++++++++++ 21 files changed, 88 insertions(+), 2 deletions(-)arch/alpha/kernel/syscalls/syscall.tbl +1
-0@@ -477,3 +477,4 @@ # 545 reserved for clone3 547 common openat2 sys_openat2 548 common pidfd_getfd sys_pidfd_getfd +549 common process_madvise sys_process_madvisearch/arm/tools/syscall.tbl +1
-0@@ -451,3 +451,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/arm64/include/asm/unistd.h +1
-1@@ -38,7 +38,7 @@#define __ARM_NR_compat_set_tls (__ARM_NR_COMPAT_BASE + 5)#define __ARM_NR_COMPAT_END (__ARM_NR_COMPAT_BASE + 0x800) -#define __NR_compat_syscalls 439 +#define __NR_compat_syscalls 440#endif #define __ARCH_WANT_SYS_CLONEarch/arm64/include/asm/unistd32.h +2
-0@@ -883,6 +883,8 @@ __SYSCALL(__NR_clone3, sys_clone3) __SYSCALL(__NR_openat2, sys_openat2)#define __NR_pidfd_getfd 438 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd) +#define __NR_process_madvise 439 +__SYSCALL (__NR_process_madvise, process_madvise)/* * Please add new compat syscalls above this comment and update arch/ia64/kernel/syscalls/syscall.tbl +1
-0@@ -358,3 +358,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/m68k/kernel/syscalls/syscall.tbl +1
-0@@ -437,3 +437,4 @@ 435 common clone3 __sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/microblaze/kernel/syscalls/syscall.tbl +1
-0@@ -443,3 +443,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/mips/kernel/syscalls/syscall_n32.tbl +1
-0@@ -376,3 +376,4 @@ 435 n32 clone3 __sys_clone3 437 n32 openat2 sys_openat2 438 n32 pidfd_getfd sys_pidfd_getfd +439 n32 process_madvise sys_process_madvisearch/mips/kernel/syscalls/syscall_n64.tbl +1
-0@@ -352,3 +352,4 @@ 435 n64 clone3 __sys_clone3 437 n64 openat2 sys_openat2 438 n64 pidfd_getfd sys_pidfd_getfd +439 n64 process_madvise sys_process_madvisearch/parisc/kernel/syscalls/syscall.tbl +1
-0@@ -435,3 +435,4 @@ 435 common clone3 sys_clone3_wrapper 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/powerpc/kernel/syscalls/syscall.tbl +1
-0@@ -519,3 +519,4 @@ 435 nospu clone3 ppc_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/s390/kernel/syscalls/syscall.tbl +1
-0@@ -440,3 +440,4 @@ 435 common clone3 sys_clone3 sys_clone3 437 common openat2 sys_openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvise sys_process_madvisearch/sh/kernel/syscalls/syscall.tbl +1
-0@@ -440,3 +440,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/sparc/kernel/syscalls/syscall.tbl +1
-0@@ -483,3 +483,4 @@ # 435 reserved for clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madvisearch/x86/entry/syscalls/syscall_32.tbl +1
-0@@ -442,3 +442,4 @@ 435 i386 clone3 sys_clone3 __ia32_sys_clone3 437 i386 openat2 sys_openat2 __ia32_sys_openat2 438 i386 pidfd_getfd sys_pidfd_getfd __ia32_sys_pidfd_getfd +439 i386 process_madvise sys_process_madvise __ia32_sys_process_madvisearch/x86/entry/syscalls/syscall_64.tbl +1
-0@@ -359,6 +359,7 @@ 435 common clone3 __x64_sys_clone3/ptregs 437 common openat2 __x64_sys_openat2 438 common pidfd_getfd __x64_sys_pidfd_getfd +439 common process_madvise __x64_sys_process_madvise # # x32-specific system call numbers start at 512 to avoid cache impactarch/xtensa/kernel/syscalls/syscall.tbl +1
-0@@ -408,3 +408,4 @@ 435 common clone3 sys_clone3 437 common openat2 sys_openat2 438 common pidfd_getfd sys_pidfd_getfd +439 common process_madvise sys_process_madviseinclude/linux/syscalls.h +2
-0@@ -876,6 +876,8 @@ asmlinkage long sys_munlockall(void);asmlinkage long sys_mincore (unsigned long start,size_t len,unsigned char __user * vec);asmlinkage long sys_madvise (unsigned long start,size_t len,int behavior); +asmlinkage long sys_process_madvise (int pidfd,unsigned long start, +size_t len,int behavior,unsigned long flags);asmlinkage long sys_remap_file_pages (unsigned long start,unsigned long size,unsigned long prot,unsigned long pgoff,unsigned long flags);include/uapi/asm-generic/unistd.h +3
-1@@ -855,9 +855,11 @@ __SYSCALL(__NR_clone3, sys_clone3) __SYSCALL(__NR_openat2, sys_openat2)#define __NR_pidfd_getfd 438 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd) +#define __NR_process_madvise 439 +__SYSCALL(__NR_process_madvise, sys_process_madvise)#undef __NR_syscalls -#define __NR_syscalls 439 +#define __NR_syscalls 440/* * 32 bit systems traditionally used different kernel/sys_ni.c +1
-0@@ -280,6 +280,7 @@ COND_SYSCALL(mlockall);COND_SYSCALL (munlockall);COND_SYSCALL (mincore);COND_SYSCALL (madvise); +COND_SYSCALL (process_madvise);COND_SYSCALL (remap_file_pages);COND_SYSCALL (mbind);COND_SYSCALL_COMPAT (mbind);mm/madvise.c +13
-0@@ -17,6 +17,7 @@#include <linux/falloc.h> #include <linux/fadvise.h> #include <linux/sched.h> +#include <linux/sched/mm.h> #include <linux/ksm.h> #include <linux/fs.h> #include <linux/file.h> @@ -986,6 +987,18 @@ madvise_behavior_valid(int behavior) } } +static bool +process_madvise_behavior_valid (int behavior) +{ +switch (behavior) { +case MADV_COLD : +case MADV_PAGEOUT : +return true; +default : +return false; + } +} +
Quoted 8 lines
mm/madvise.c madvise_behavior_valid(int behavior) · line 1049
Quoted 8 lines
mm/madvise.c +1
-0/*/ * The madvise(2) system call. * @@ -1033,6 +1046,11 @@ madvise_behavior_valid(int behavior)* MADV_DONTDUMP - the application wants to prevent pages in the given range * from being included in its core dump. * MADV_DODUMP - cancel MADV_DONTDUMP: no longer exclude from core dump. +* MADV_COLD - the application uses the memory less so the kernel can
"kernel can" implies that kernel might not deactivate the pages, which IIUC is not the case. Maybe rephrase as "MADV_COLD - the application is not expected to use this memory soon, deactivate pages in this range so that they can be reclaimed easily if memory pressure happens.""
Quoted 3 lines
mm/madvise.c madvise_behavior_valid(int behavior) · line 1052
Quoted 3 lines
mm/madvise.c +3
-0+* deactivate the memory to evict them quickly when the memory +* pressure happen. +* MADV_PAGEOUT - the application uses the memroy very rarely so kernel can
s/memroy/memory
Quoted 1 line
mm/madvise.c madvise_behavior_valid(int behavior) · line 1053
Quoted 1 line
mm/madvise.c +1
-0+* page out the memory instantly.
same nit about the usage of "kernel can". Maybe rephrase as "MADV_PAGEOUT - the application is not expected to use this memory soon, page out the pages in this range immediately.""
Quoted 48 lines
Quoted 48 lines
mm/madvise.c +41
-0* * return values: * zero - success @@ -1150,3 +1168,49 @@ SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior) {return do_madvise (current, current->mm, start, len_in, behavior); } + +SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start, + size_t, len_in, int, behavior, unsigned long, flags) +{ +int ret; +struct fd f; +struct pid *pid; +struct task_struct *task; +struct mm_struct *mm; + +if (flags !=0 ) +return -EINVAL ; + +if (!process_madvise_behavior_valid (behavior)) +return -EINVAL ; + + f =fdget (pidfd); +if (!f.file) +return -EBADF ; + + pid =pidfd_pid (f.file); +if (IS_ERR (pid)) { + ret =PTR_ERR (pid); + goto fdput; + } + + task =get_pid_task (pid,PIDTYPE_PID ); +if (!task) { + ret = -ESRCH ; + goto fdput; + } + + mm =mm_access (task,PTRACE_MODE_ATTACH_FSCREDS ); +if (IS_ERR_OR_NULL (mm)) { + ret =IS_ERR (mm) ?PTR_ERR (mm) : -ESRCH ; + goto release_task; + } + + ret =do_madvise (task, mm, start, len_in, behavior); +mmput (mm); +release_task:
Quoted 9 lines
Quoted 8 lines
mm/madvise.c +5
-1+put_task_struct (task); +fdput: +fdput (f); +return ret; +} --2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 2/7] mm: introduce external memory hinting API
On Fri, Feb 28, 2020 at 02:14:56PM -0800, Suren Baghdasaryan wrote:
< snip >
Quoted 34 lines
Quoted 34 lines
Quoted 34 lines
mm/madvise.c +13
-0@@ -17,6 +17,7 @@#include <linux/falloc.h> #include <linux/fadvise.h> #include <linux/sched.h> +#include <linux/sched/mm.h> #include <linux/ksm.h> #include <linux/fs.h> #include <linux/file.h> @@ -986,6 +987,18 @@ madvise_behavior_valid(int behavior) } } +static bool +process_madvise_behavior_valid (int behavior) +{ +switch (behavior) { +case MADV_COLD : +case MADV_PAGEOUT : +return true; +default : +return false; + } +} +/*/ * The madvise(2) system call. * @@ -1033,6 +1046,11 @@ madvise_behavior_valid(int behavior)* MADV_DONTDUMP - the application wants to prevent pages in the given range * from being included in its core dump.
Quoted 8 lines
Quoted 2 lines
mm/madvise.c madvise_behavior_valid(int behavior) · line 1049
Quoted 2 lines
mm/madvise.c +1
-0* MADV_DODUMP - cancel MADV_DONTDUMP: no longer exclude from core dump. +* MADV_COLD - the application uses the memory less so the kernel can "kernel can" implies that kernel might not deactivate the pages, which IIUC is not the case. Maybe rephrase as "
MADV_COLD- the application is not expected to use this memory soon, deactivate pages in this range so that they can be reclaimed easily if memory pressure happens.""
That is much better.
Quoted 6 lines
Quoted 4 lines
mm/madvise.c madvise_behavior_valid(int behavior) · line 1052
Quoted 4 lines
mm/madvise.c +3
-0+* deactivate the memory to evict them quickly when the memory +* pressure happen. +* MADV_PAGEOUT - the application uses the memroy very rarely so kernel can s/memroy/memory
Fixed.
Quoted 6 lines
Quoted 2 lines
mm/madvise.c madvise_behavior_valid(int behavior) · line 1053
Quoted 2 lines
mm/madvise.c +1
-0+* page out the memory instantly. same nit about the usage of "kernel can". Maybe rephrase as "
MADV_PAGEOUT- the application is not expected to use this memory soon, page out the pages in this range immediately.""
Yub.
Quoted 59 lines
Quoted 49 lines
Quoted 49 lines
mm/madvise.c +41
-0* * return values: * zero - success @@ -1150,3 +1168,49 @@ SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior) {return do_madvise (current, current->mm, start, len_in, behavior); } + +SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start, + size_t, len_in, int, behavior, unsigned long, flags) +{ +int ret; +struct fd f; +struct pid *pid; +struct task_struct *task; +struct mm_struct *mm; + +if (flags !=0 ) +return -EINVAL ; + +if (!process_madvise_behavior_valid (behavior)) +return -EINVAL ; + + f =fdget (pidfd); +if (!f.file) +return -EBADF ; + + pid =pidfd_pid (f.file); +if (IS_ERR (pid)) { + ret =PTR_ERR (pid); + goto fdput; + } + + task =get_pid_task (pid,PIDTYPE_PID ); +if (!task) { + ret = -ESRCH ; + goto fdput; + } + + mm =mm_access (task,PTRACE_MODE_ATTACH_FSCREDS ); +if (IS_ERR_OR_NULL (mm)) { + ret =IS_ERR (mm) ?PTR_ERR (mm) : -ESRCH ; + goto release_task; + } + + ret =do_madvise (task, mm, start, len_in, behavior); +mmput (mm); +release_task:Quoted 9 lines
Quoted 8 lines
mm/madvise.c +5
-1+put_task_struct (task); +fdput: +fdput (f); +return ret; +} --2.25.0.265.gbab2e86ba0-goog
Quoted 1 line
Reviewed-by Suren Baghdasaryan <[email protected]>
Thanks, Suren!
kbuild test robot
build botRe: [PATCH v6 2/7] mm: introduce external memory hinting API
In reply to Minchan KimI love your patch! Yet something to improve:
Hi Minchan,
I love your patch! Yet something to improve:
[auto build test ERROR on m68k/for-next]
[also build test ERROR on powerpc/next s390/features linus/master v5.6-rc2 next-20200220] [cannot apply to arm64/for-next/core tip/x86/asm arm/for-next hp-parisc/for-next] [if your patch is applied to the wrong git tree, please drop us a note to help improve the system. BTW, we also suggest to use '--base' option to specify the base tree in git format-patch, please see https://stackoverflow.com/a/37406982]
url: https://github.com/0day-ci/linux/commits/Minchan-Kim/introduce-memory-hinting-API-for-external-process/20200220-225155 base: https://git.kernel.org/pub/scm/linux/kernel/git/geert/linux-m68k.git for-next config: arm64-randconfig-a001-20200220 (attached as .config) compiler: aarch64-linux-gcc (GCC) 7.5.0
reproduce:
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# save the attached .config to linux build tree
GCC_VERSION=7.5.0 make.cross ARCH=arm64
If you fix the issue, kindly add following tag
All errors (new ones prefixed by >>):
In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from include/uapi/linux/unistd.h:8,
from include/linux/syscalls.h:81,
from arch/arm64/kernel/sys.c:16:
include/uapi/asm-generic/unistd.h:858:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 439 include/uapi/asm-generic/unistd.h:856:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 438
In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64/kernel/sys.c:53:
include/uapi/asm-generic/unistd.h:856:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 438 In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from include/uapi/linux/unistd.h:8,
from include/linux/syscalls.h:81,
from arch/arm64/kernel/sys.c:16:
include/uapi/asm-generic/unistd.h:858:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 439 In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64/kernel/sys.c:53:
include/uapi/asm-generic/unistd.h:858:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 439 include/uapi/asm-generic/unistd.h:856:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 438
In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64/kernel/sys.c:60:
include/uapi/asm-generic/unistd.h:856:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 438 In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64/kernel/sys.c:53:
include/uapi/asm-generic/unistd.h:858:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 439 In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64/kernel/sys.c:60:
include/uapi/asm-generic/unistd.h:858:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 439 include/uapi/asm-generic/unistd.h:856:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 438
>> include/uapi/asm-generic/unistd.h:859:11: error: '__NR_process_madvise' undeclared here (not in a function); did you mean '__NR_process_vm_writev'?
__SYSCALL(__NR_process_madvise, sys_process_madvise)
^
arch/arm64/kernel/sys.c:56:29: note: in definition of macro '__SYSCALL'
#define __SYSCALL(nr, sym) [nr] = __arm64_##sym,
^~
include/uapi/asm-generic/unistd.h:859:11: error: array index in initializer not of integer type
__SYSCALL(__NR_process_madvise, sys_process_madvise)
^
arch/arm64/kernel/sys.c:56:29: note: in definition of macro '__SYSCALL'
#define __SYSCALL(nr, sym) [nr] = __arm64_##sym,
^~
include/uapi/asm-generic/unistd.h:859:11: note: (near initialization for 'sys_call_table')
__SYSCALL(__NR_process_madvise, sys_process_madvise)
^
arch/arm64/kernel/sys.c:56:29: note: in definition of macro '__SYSCALL'
#define __SYSCALL(nr, sym) [nr] = __arm64_##sym,
^~
--
In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from include/uapi/linux/unistd.h:8,
from include/linux/syscalls.h:81,
from arch/arm64//kernel/sys.c:16:
include/uapi/asm-generic/unistd.h:858:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 439 include/uapi/asm-generic/unistd.h:856:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 438
In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64//kernel/sys.c:53:
include/uapi/asm-generic/unistd.h:856:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 438 In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from include/uapi/linux/unistd.h:8,
from include/linux/syscalls.h:81,
from arch/arm64//kernel/sys.c:16:
include/uapi/asm-generic/unistd.h:858:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 439 In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64//kernel/sys.c:53:
include/uapi/asm-generic/unistd.h:858:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 439 include/uapi/asm-generic/unistd.h:856:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 438
In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64//kernel/sys.c:60:
include/uapi/asm-generic/unistd.h:856:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 438 In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64//kernel/sys.c:53:
include/uapi/asm-generic/unistd.h:858:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 439 In file included from arch/arm64/include/uapi/asm/unistd.h:24:0,
from arch/arm64/include/asm/unistd.h:47,
from arch/arm64//kernel/sys.c:60:
include/uapi/asm-generic/unistd.h:858:0: warning: "__NR_pidfd_getfd" redefined
#define __NR_pidfd_getfd 439 include/uapi/asm-generic/unistd.h:856:0: note: this is the location of the previous definition
#define __NR_pidfd_getfd 438
>> include/uapi/asm-generic/unistd.h:859:11: error: '__NR_process_madvise' undeclared here (not in a function); did you mean '__NR_process_vm_writev'?
__SYSCALL(__NR_process_madvise, sys_process_madvise)
^
arch/arm64//kernel/sys.c:56:29: note: in definition of macro '__SYSCALL'
#define __SYSCALL(nr, sym) [nr] = __arm64_##sym,
^~
include/uapi/asm-generic/unistd.h:859:11: error: array index in initializer not of integer type
__SYSCALL(__NR_process_madvise, sys_process_madvise)
^
arch/arm64//kernel/sys.c:56:29: note: in definition of macro '__SYSCALL'
#define __SYSCALL(nr, sym) [nr] = __arm64_##sym,
^~
include/uapi/asm-generic/unistd.h:859:11: note: (near initialization for 'sys_call_table')
__SYSCALL(__NR_process_madvise, sys_process_madvise)
^
arch/arm64//kernel/sys.c:56:29: note: in definition of macro '__SYSCALL'
#define __SYSCALL(nr, sym) [nr] = __arm64_##sym,
^~vim +859 include/uapi/asm-generic/unistd.h
853
854 #define __NR_openat2 437
855 __SYSCALL(__NR_openat2, sys_openat2)
856 #define __NR_pidfd_getfd 438
857 __SYSCALL(__NR_pidfd_getfd, sys_pidfd_getfd)
858 #define __NR_pidfd_getfd 439
> 859 __SYSCALL(__NR_process_madvise, sys_process_madvise)
860
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
PATCH v6 3/7 mm: check fatal signal pending of target process
In reply to Minchan KimBail out to prevent unnecessary CPU overhead if target process has pending fatal signal during (MADV_COLD|MADV_PAGEOUT) operation.
--- mm/madvise.c | 29 +++++++++++++++++++++-------- 1 file changed, 21 insertions(+), 8 deletions(-)
mm/madvise.c +21 -8
@@ -36,6 +36,7 @@struct madvise_walk_private {struct mmu_gather *tlb;bool pageout; +struct task_struct *target_task; };/*
@@ -316,6 +317,10 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,if (fatal_signal_pending (current))return -EINTR ; +if (private->target_task && +fatal_signal_pending (private->target_task)) +return -EINTR ; +#ifdef CONFIG_TRANSPARENT_HUGEPAGE if (pmd_trans_huge (*pmd)) {pmd_t orig_pmd;
@@ -471,12 +476,14 @@ static const struct mm_walk_ops cold_walk_ops = { };static void madvise_cold_page_range (struct mmu_gather *tlb, +struct task_struct *task,struct vm_area_struct *vma,unsigned long addr,unsigned long end) {struct madvise_walk_private walk_private = { .pageout = false, .tlb = tlb, + .target_task = task, };tlb_start_vma (tlb, vma);
@@ -484,7 +491,8 @@ static void madvise_cold_page_range(struct mmu_gather *tlb,tlb_end_vma (tlb, vma); } -static long madvise_cold (struct vm_area_struct *vma, +static long madvise_cold (struct task_struct *task, +struct vm_area_struct *vma,struct vm_area_struct **prev,unsigned long start_addr,unsigned long end_addr) {
@@ -497,19 +505,21 @@ static long madvise_cold(struct vm_area_struct *vma,lru_add_drain ();tlb_gather_mmu (&tlb, mm, start_addr, end_addr); -madvise_cold_page_range (&tlb, vma, start_addr, end_addr); +madvise_cold_page_range (&tlb, task, vma, start_addr, end_addr);tlb_finish_mmu (&tlb, start_addr, end_addr);return 0 ; }static void madvise_pageout_page_range (struct mmu_gather *tlb, +struct task_struct *task,struct vm_area_struct *vma,unsigned long addr,unsigned long end) {struct madvise_walk_private walk_private = { .pageout = true, .tlb = tlb, + .target_task = task, };tlb_start_vma (tlb, vma);
@@ -533,7 +543,8 @@ static inline bool can_do_pageout(struct vm_area_struct *vma)inode_permission (file_inode (vma->vm_file),MAY_WRITE ) ==0 ; } -static long madvise_pageout (struct vm_area_struct *vma, +static long madvise_pageout (struct task_struct *task, +struct vm_area_struct *vma,struct vm_area_struct **prev,unsigned long start_addr,unsigned long end_addr) {
@@ -549,7 +560,7 @@ static long madvise_pageout(struct vm_area_struct *vma,lru_add_drain ();tlb_gather_mmu (&tlb, mm, start_addr, end_addr); -madvise_pageout_page_range (&tlb, vma, start_addr, end_addr); +madvise_pageout_page_range (&tlb, task, vma, start_addr, end_addr);tlb_finish_mmu (&tlb, start_addr, end_addr);return 0 ;
@@ -929,7 +940,8 @@ static int madvise_inject_error(int behavior,#endif static long -madvise_vma (struct vm_area_struct *vma,struct vm_area_struct **prev, +madvise_vma (struct task_struct *task,struct vm_area_struct *vma, +struct vm_area_struct **prev,unsigned long start,unsigned long end,int behavior) {switch (behavior) {
@@ -938,9 +950,9 @@ madvise_vma(struct vm_area_struct *vma, struct vm_area_struct **prev,case MADV_WILLNEED :return madvise_willneed (vma, prev, start, end);case MADV_COLD : -return madvise_cold (vma, prev, start, end); +return madvise_cold (task, vma, prev, start, end);case MADV_PAGEOUT : -return madvise_pageout (vma, prev, start, end); +return madvise_pageout (task, vma, prev, start, end);case MADV_FREE :case MADV_DONTNEED :return madvise_dontneed_free (vma, prev, start, end, behavior);
@@ -1140,7 +1152,8 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm, tmp = end;/* Here vma->vm_start <= start < tmp <= (end|vma->vm_end). */ - error =madvise_vma (vma, &prev, start, tmp, behavior); + error =madvise_vma (target_task, vma, &prev, + start, tmp, behavior);if (error) goto out; start = tmp;
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 3/7] mm: check fatal signal pending of target process
Quoted 132 lines · Minchan Kim
Bail out to prevent unnecessary CPU overhead if target process has pending fatal signal during (MADV_COLD|MADV_PAGEOUT) operation.
Signed-off-by Minchan Kim <[email protected]>--- mm/madvise.c | 29 +++++++++++++++++++++-------- 1 file changed, 21 insertions(+), 8 deletions(-)mm/madvise.c +21
-9@@ -36,6 +36,7 @@struct madvise_walk_private {struct mmu_gather *tlb;bool pageout; +struct task_struct *target_task; };/* @@ -316,6 +317,10 @@ static int madvise_cold_or_pageout_pte_range(pmd_t *pmd,if (fatal_signal_pending (current))return -EINTR ; +if (private->target_task && +fatal_signal_pending (private->target_task)) +return -EINTR ; +#ifdef CONFIG_TRANSPARENT_HUGEPAGE if (pmd_trans_huge (*pmd)) {pmd_t orig_pmd;@@ -471,12 +476,14 @@ static const struct mm_walk_ops cold_walk_ops = { };static void madvise_cold_page_range (struct mmu_gather *tlb, +struct task_struct *task,struct vm_area_struct *vma,unsigned long addr,unsigned long end) {struct madvise_walk_private walk_private = { .pageout = false, .tlb = tlb, + .target_task = task, };tlb_start_vma (tlb, vma);@@ -484,7 +491,8 @@ static void madvise_cold_page_range(struct mmu_gather *tlb,tlb_end_vma (tlb, vma); } -static long madvise_cold (struct vm_area_struct *vma, +static long madvise_cold (struct task_struct *task, +struct vm_area_struct *vma,struct vm_area_struct **prev,unsigned long start_addr,unsigned long end_addr) {@@ -497,19 +505,21 @@ static long madvise_cold(struct vm_area_struct *vma,lru_add_drain ();tlb_gather_mmu (&tlb, mm, start_addr, end_addr); -madvise_cold_page_range (&tlb, vma, start_addr, end_addr); +madvise_cold_page_range (&tlb, task, vma, start_addr, end_addr);tlb_finish_mmu (&tlb, start_addr, end_addr);return 0 ; }static void madvise_pageout_page_range (struct mmu_gather *tlb, +struct task_struct *task,struct vm_area_struct *vma,unsigned long addr,unsigned long end) {struct madvise_walk_private walk_private = { .pageout = true, .tlb = tlb, + .target_task = task, };tlb_start_vma (tlb, vma);@@ -533,7 +543,8 @@ static inline bool can_do_pageout(struct vm_area_struct *vma)inode_permission (file_inode (vma->vm_file),MAY_WRITE ) ==0 ; } -static long madvise_pageout (struct vm_area_struct *vma, +static long madvise_pageout (struct task_struct *task, +struct vm_area_struct *vma,struct vm_area_struct **prev,unsigned long start_addr,unsigned long end_addr) {@@ -549,7 +560,7 @@ static long madvise_pageout(struct vm_area_struct *vma,lru_add_drain ();tlb_gather_mmu (&tlb, mm, start_addr, end_addr); -madvise_pageout_page_range (&tlb, vma, start_addr, end_addr); +madvise_pageout_page_range (&tlb, task, vma, start_addr, end_addr);tlb_finish_mmu (&tlb, start_addr, end_addr);return 0 ;@@ -929,7 +940,8 @@ static int madvise_inject_error(int behavior,#endif static long -madvise_vma (struct vm_area_struct *vma,struct vm_area_struct **prev, +madvise_vma (struct task_struct *task,struct vm_area_struct *vma, +struct vm_area_struct **prev,unsigned long start,unsigned long end,int behavior) {switch (behavior) {@@ -938,9 +950,9 @@ madvise_vma(struct vm_area_struct *vma, struct vm_area_struct **prev,case MADV_WILLNEED :return madvise_willneed (vma, prev, start, end);case MADV_COLD : -return madvise_cold (vma, prev, start, end); +return madvise_cold (task, vma, prev, start, end);case MADV_PAGEOUT : -return madvise_pageout (vma, prev, start, end); +return madvise_pageout (task, vma, prev, start, end);case MADV_FREE :case MADV_DONTNEED :return madvise_dontneed_free (vma, prev, start, end, behavior);@@ -1140,7 +1152,8 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm, tmp = end;/* Here vma->vm_start <= start < tmp <= (end|vma->vm_end). */ - error =madvise_vma (vma, &prev, start, tmp, behavior); + error =madvise_vma (target_task, vma, &prev, + start, tmp, behavior);if (error) goto out; start = tmp; --
Quoted 2 lines
2.25.0.265.gbab2e86ba0-goog
PATCH v6 4/7 pid: move pidfd_get_pid function to pid.c
In reply to Minchan Kimprocess_madvise syscall needs pidfd_get_pid function to translate pidfd to pid so this patch move the function to kernel/pid.c.
--- include/linux/pid.h | 1 + kernel/exit.c | 17 ----------------- kernel/pid.c | 17 +++++++++++++++++ 3 files changed, 18 insertions(+), 17 deletions(-)
include/linux/pid.h +1 -0
@@ -75,6 +75,7 @@ extern const struct file_operations pidfd_fops;struct file ;extern struct pid *pidfd_pid (const struct file *file); +struct pid *pidfd_get_pid (unsigned int fd);static inline struct pid *get_pid (struct pid *pid) {
kernel/exit.c +0 -17
@@ -1470,23 +1470,6 @@ static long do_wait(struct wait_opts *wo)return retval; } -static struct pid *pidfd_get_pid (unsigned int fd) -{ -struct fd f; -struct pid *pid; - - f =fdget (fd); -if (!f.file) -return ERR_PTR (-EBADF ); - - pid =pidfd_pid (f.file); -if (!IS_ERR (pid)) -get_pid (pid); - -fdput (f); -return pid; -} -static long kernel_waitid (int which,pid_t upid,struct waitid_info *infop,int options,struct rusage *ru) {
kernel/pid.c +17 -0
@@ -496,6 +496,23 @@ struct pid *find_ge_pid(int nr, struct pid_namespace *ns)return idr_get_next (&ns->idr, &nr); } +struct pid *pidfd_get_pid (unsigned int fd) +{ +struct fd f; +struct pid *pid; + + f =fdget (fd); +if (!f.file) +return ERR_PTR (-EBADF ); + + pid =pidfd_pid (f.file); +if (!IS_ERR (pid)) +get_pid (pid); + +fdput (f); +return pid; +} +/** * pidfd_create() - Create a new pid file descriptor. *
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 4/7] pid: move pidfd_get_pid function to pid.c
Quoted 84 lines · Minchan Kim
process_madvisesyscall needspidfd_get_pidfunction to translate pidfd to pid so this patch move the function tokernel/pid.c.Cc Christoph Hellwig <[email protected]>Cc Christian Brauner <[email protected]>Suggested-by Alexander Duyck <[email protected]>Reviewed-by Alexander Duyck <[email protected]>Signed-off-by Minchan Kim <[email protected]>--- include/linux/pid.h | 1 + kernel/exit.c | 17 ----------------- kernel/pid.c | 17 +++++++++++++++++ 3 files changed, 18 insertions(+), 17 deletions(-)include/linux/pid.h +1
-0@@ -75,6 +75,7 @@ extern const struct file_operations pidfd_fops;struct file ;extern struct pid *pidfd_pid (const struct file *file); +struct pid *pidfd_get_pid (unsigned int fd);static inline struct pid *get_pid (struct pid *pid) {kernel/exit.c +0
-17@@ -1470,23 +1470,6 @@ static long do_wait(struct wait_opts *wo)return retval; } -static struct pid *pidfd_get_pid (unsigned int fd) -{ -struct fd f; -struct pid *pid; - - f =fdget (fd); -if (!f.file) -return ERR_PTR (-EBADF ); - - pid =pidfd_pid (f.file); -if (!IS_ERR (pid)) -get_pid (pid); - -fdput (f); -return pid; -} -static long kernel_waitid (int which,pid_t upid,struct waitid_info *infop,int options,struct rusage *ru) {kernel/pid.c +17
-1@@ -496,6 +496,23 @@ struct pid *find_ge_pid(int nr, struct pid_namespace *ns)return idr_get_next (&ns->idr, &nr); } +struct pid *pidfd_get_pid (unsigned int fd) +{ +struct fd f; +struct pid *pid; + + f =fdget (fd); +if (!f.file) +return ERR_PTR (-EBADF ); + + pid =pidfd_pid (f.file); +if (!IS_ERR (pid)) +get_pid (pid); + +fdput (f); +return pid; +} +/** * pidfd_create() - Create a new pid file descriptor. * --
Quoted 2 lines
2.25.0.265.gbab2e86ba0-goog
PATCH v6 5/7 mm: support both pid and pidfd for process_madvise
In reply to Minchan KimThere is a demand[1] to support pid as well pidfd for process_madvise to reduce unnecessary syscall to get pidfd if the user has control of the target process(ie, they could guarantee the process is not gone or pid is not reused. Or, it might be okay to give a hint to wrong process).
This patch aims for supporting both options like waitid(2). So, the syscall is currently,
int process_madvise(int which, pid_t pid, void *addr, size_t length, int advise, unsigned long flag);
@which is actually idtype_t for userspace libray and currently, it supports P_PID and P_PIDFD.
[1] https://lore.kernel.org/linux-mm/[email protected]/
--- include/linux/syscalls.h | 3 ++- mm/madvise.c | 34 ++++++++++++++++++++++------------ 2 files changed, 24 insertions(+), 13 deletions(-)
include/linux/syscalls.h +2 -1
@@ -876,7 +876,8 @@ asmlinkage long sys_munlockall(void);asmlinkage long sys_mincore (unsigned long start,size_t len,unsigned char __user * vec);asmlinkage long sys_madvise (unsigned long start,size_t len,int behavior); -asmlinkage long sys_process_madvise (int pidfd,unsigned long start, + +asmlinkage long sys_process_madvise (int which,pid_t pid,unsigned long start,size_t len,int behavior,unsigned long flags); asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size, unsigned long prot, unsigned long pgoff,
mm/madvise.c +22 -12
@@ -1182,11 +1182,10 @@ SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior)return do_madvise (current, current->mm, start, len_in, behavior); } -SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start, +SYSCALL_DEFINE6(process_madvise, int, which, pid_t, upid, unsigned long, start, size_t, len_in, int, behavior, unsigned long, flags) {int ret; -struct fd f;struct pid *pid;struct task_struct *task;struct mm_struct *mm;
@@ -1197,20 +1196,31 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start,if (!process_madvise_behavior_valid (behavior))return -EINVAL ; - f =fdget (pidfd); -if (!f.file) -return -EBADF ; +switch (which) { +case P_PID : +if (upid <=0 ) +return -EINVAL ; + + pid =find_get_pid (upid); +if (!pid) +return -ESRCH ; +break ; +case P_PIDFD : +if (upid <0 ) +return -EINVAL ; - pid =pidfd_pid (f.file); -if (IS_ERR (pid)) { - ret =PTR_ERR (pid); - goto fdput; + pid =pidfd_get_pid (upid); +if (IS_ERR (pid)) +return PTR_ERR (pid); +break ; +default : +return -EINVAL ; } task =get_pid_task (pid,PIDTYPE_PID );if (!task) { ret = -ESRCH ; - goto fdput; + goto put_pid; } mm =mm_access (task,PTRACE_MODE_ATTACH_FSCREDS );
@@ -1223,7 +1233,7 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start,mmput (mm); release_task:put_task_struct (task); -fdput: -fdput (f); +put_pid: +put_pid (pid);return ret; }
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 5/7] mm: support both pid and pidfd for process_madvise
Quoted 6 lines · Minchan Kim
There is a demand[1] to support pid as well pidfd for
process_madviseto reduce unnecessary syscall to get pidfd if the user has control of the target process(ie, they could guarantee the process is not gone or pid is not reused. Or, it might be okay to give a hint to wrong process).
nit: When would "give a hint to wrong process" be ok? I would just remove this part.
Quoted 102 lines
This patch aims for supporting both options like waitid(2). So, the syscall is currently,
int process_madvise(int which,
pid_tpid, void *addr,
size_tlength, int advise, unsigned long flag);@which is actually
idtype_tfor userspace libray and currently, it supportsP_PIDandP_PIDFD.[1] https://lore.kernel.org/linux-mm/[email protected]/
Cc Christian Brauner <[email protected]>Suggested-by Kirill Tkhai <[email protected]>Signed-off-by Minchan Kim <[email protected]>--- include/linux/syscalls.h | 3 ++- mm/madvise.c | 34 ++++++++++++++++++++++------------ 2 files changed, 24 insertions(+), 13 deletions(-)include/linux/syscalls.h +2
-1@@ -876,7 +876,8 @@ asmlinkage long sys_munlockall(void);asmlinkage long sys_mincore (unsigned long start,size_t len,unsigned char __user * vec);asmlinkage long sys_madvise (unsigned long start,size_t len,int behavior); -asmlinkage long sys_process_madvise (int pidfd,unsigned long start, + +asmlinkage long sys_process_madvise (int which,pid_t pid,unsigned long start,size_t len,int behavior,unsigned long flags); asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size, unsigned long prot, unsigned long pgoff,mm/madvise.c +22
-13@@ -1182,11 +1182,10 @@ SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior)return do_madvise (current, current->mm, start, len_in, behavior); } -SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start, +SYSCALL_DEFINE6(process_madvise, int, which, pid_t, upid, unsigned long, start, size_t, len_in, int, behavior, unsigned long, flags) {int ret; -struct fd f;struct pid *pid;struct task_struct *task;struct mm_struct *mm;@@ -1197,20 +1196,31 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start,if (!process_madvise_behavior_valid (behavior))return -EINVAL ; - f =fdget (pidfd); -if (!f.file) -return -EBADF ; +switch (which) { +case P_PID : +if (upid <=0 ) +return -EINVAL ; + + pid =find_get_pid (upid); +if (!pid) +return -ESRCH ; +break ; +case P_PIDFD : +if (upid <0 ) +return -EINVAL ; - pid =pidfd_pid (f.file); -if (IS_ERR (pid)) { - ret =PTR_ERR (pid); - goto fdput; + pid =pidfd_get_pid (upid); +if (IS_ERR (pid)) +return PTR_ERR (pid); +break ; +default : +return -EINVAL ; } task =get_pid_task (pid,PIDTYPE_PID );if (!task) { ret = -ESRCH ; - goto fdput; + goto put_pid; } mm =mm_access (task,PTRACE_MODE_ATTACH_FSCREDS );@@ -1223,7 +1233,7 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start,mmput (mm); release_task:put_task_struct (task); -fdput: -fdput (f); +put_pid: +put_pid (pid);return ret; } --
Quoted 2 lines
2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 5/7] mm: support both pid and pidfd for process_madvise
Quoted 10 lines · Suren Baghdasaryan
Quoted 6 lines · Minchan Kim
There is a demand[1] to support pid as well pidfd for
process_madviseto reduce unnecessary syscall to get pidfd if the user has control of the target process(ie, they could guarantee the process is not gone or pid is not reused. Or, it might be okay to give a hint to wrong process).nit: When would "give a hint to wrong process" be ok? I would just remove this part.
I wanted to say non destructive hints. It's already true for other some hints because they are just best effort so it's not critical to be failed. If you mind it, I will remove the phrase.
Thanks.
Quoted 107 lines
Quoted 103 lines
This patch aims for supporting both options like waitid(2). So, the syscall is currently,
int process_madvise(int which,
pid_tpid, void *addr,
size_tlength, int advise, unsigned long flag);@which is actually
idtype_tfor userspace libray and currently, it supportsP_PIDandP_PIDFD.[1] https://lore.kernel.org/linux-mm/[email protected]/
Cc Christian Brauner <[email protected]>Suggested-by Kirill Tkhai <[email protected]>Signed-off-by Minchan Kim <[email protected]>--- include/linux/syscalls.h | 3 ++- mm/madvise.c | 34 ++++++++++++++++++++++------------ 2 files changed, 24 insertions(+), 13 deletions(-)include/linux/syscalls.h +2
-1@@ -876,7 +876,8 @@ asmlinkage long sys_munlockall(void);asmlinkage long sys_mincore (unsigned long start,size_t len,unsigned char __user * vec);asmlinkage long sys_madvise (unsigned long start,size_t len,int behavior); -asmlinkage long sys_process_madvise (int pidfd,unsigned long start, + +asmlinkage long sys_process_madvise (int which,pid_t pid,unsigned long start,size_t len,int behavior,unsigned long flags); asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size, unsigned long prot, unsigned long pgoff,mm/madvise.c +22
-13@@ -1182,11 +1182,10 @@ SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior)return do_madvise (current, current->mm, start, len_in, behavior); } -SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start, +SYSCALL_DEFINE6(process_madvise, int, which, pid_t, upid, unsigned long, start, size_t, len_in, int, behavior, unsigned long, flags) {int ret; -struct fd f;struct pid *pid;struct task_struct *task;struct mm_struct *mm;@@ -1197,20 +1196,31 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start,if (!process_madvise_behavior_valid (behavior))return -EINVAL ; - f =fdget (pidfd); -if (!f.file) -return -EBADF ; +switch (which) { +case P_PID : +if (upid <=0 ) +return -EINVAL ; + + pid =find_get_pid (upid); +if (!pid) +return -ESRCH ; +break ; +case P_PIDFD : +if (upid <0 ) +return -EINVAL ; - pid =pidfd_pid (f.file); -if (IS_ERR (pid)) { - ret =PTR_ERR (pid); - goto fdput; + pid =pidfd_get_pid (upid); +if (IS_ERR (pid)) +return PTR_ERR (pid); +break ; +default : +return -EINVAL ; } task =get_pid_task (pid,PIDTYPE_PID );if (!task) { ret = -ESRCH ; - goto fdput; + goto put_pid; } mm =mm_access (task,PTRACE_MODE_ATTACH_FSCREDS );@@ -1223,7 +1233,7 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start,mmput (mm); release_task:put_task_struct (task); -fdput: -fdput (f); +put_pid: +put_pid (pid);return ret; } --Quoted 2 lines
2.25.0.265.gbab2e86ba0-goog
Reviewed-by Suren Baghdasaryan <[email protected]>
Suren Baghdasaryan
reviewerRe: [PATCH v6 5/7] mm: support both pid and pidfd for process_madvise
Quoted 10 lines · Minchan Kim
Quoted 8 lines · Suren Baghdasaryan
Quoted 6 lines · Minchan Kim
There is a demand[1] to support pid as well pidfd for
process_madviseto reduce unnecessary syscall to get pidfd if the user has control of the target process(ie, they could guarantee the process is not gone or pid is not reused. Or, it might be okay to give a hint to wrong process).
Quoted 6 lines
Quoted 2 lines
nit: When would "give a hint to wrong process" be ok? I would just remove this part.
I wanted to say non destructive hints. It's already true for other some hints because they are just best effort so it's not critical to be failed. If you mind it, I will remove the phrase.
Up to you, or maybe call it a "non-fatal" error? Saying that it's ok to hint a wrong process sounds wrong to me.
Quoted 110 lines
Thanks.
Quoted 107 lines
Quoted 103 lines
This patch aims for supporting both options like waitid(2). So, the syscall is currently,
int process_madvise(int which,
pid_tpid, void *addr,
size_tlength, int advise, unsigned long flag);@which is actually
idtype_tfor userspace libray and currently, it supportsP_PIDandP_PIDFD.[1] https://lore.kernel.org/linux-mm/[email protected]/
Cc Christian Brauner <[email protected]>Suggested-by Kirill Tkhai <[email protected]>Signed-off-by Minchan Kim <[email protected]>--- include/linux/syscalls.h | 3 ++- mm/madvise.c | 34 ++++++++++++++++++++++------------ 2 files changed, 24 insertions(+), 13 deletions(-)include/linux/syscalls.h +2
-1@@ -876,7 +876,8 @@ asmlinkage long sys_munlockall(void);asmlinkage long sys_mincore (unsigned long start,size_t len,unsigned char __user * vec);asmlinkage long sys_madvise (unsigned long start,size_t len,int behavior); -asmlinkage long sys_process_madvise (int pidfd,unsigned long start, + +asmlinkage long sys_process_madvise (int which,pid_t pid,unsigned long start,size_t len,int behavior,unsigned long flags); asmlinkage long sys_remap_file_pages(unsigned long start, unsigned long size, unsigned long prot, unsigned long pgoff,mm/madvise.c +22
-13@@ -1182,11 +1182,10 @@ SYSCALL_DEFINE3(madvise, unsigned long, start, size_t, len_in, int, behavior)return do_madvise (current, current->mm, start, len_in, behavior); } -SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start, +SYSCALL_DEFINE6(process_madvise, int, which, pid_t, upid, unsigned long, start, size_t, len_in, int, behavior, unsigned long, flags) {int ret; -struct fd f;struct pid *pid;struct task_struct *task;struct mm_struct *mm;@@ -1197,20 +1196,31 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start,if (!process_madvise_behavior_valid (behavior))return -EINVAL ; - f =fdget (pidfd); -if (!f.file) -return -EBADF ; +switch (which) { +case P_PID : +if (upid <=0 ) +return -EINVAL ; + + pid =find_get_pid (upid); +if (!pid) +return -ESRCH ; +break ; +case P_PIDFD : +if (upid <0 ) +return -EINVAL ; - pid =pidfd_pid (f.file); -if (IS_ERR (pid)) { - ret =PTR_ERR (pid); - goto fdput; + pid =pidfd_get_pid (upid); +if (IS_ERR (pid)) +return PTR_ERR (pid); +break ; +default : +return -EINVAL ; } task =get_pid_task (pid,PIDTYPE_PID );if (!task) { ret = -ESRCH ; - goto fdput; + goto put_pid; } mm =mm_access (task,PTRACE_MODE_ATTACH_FSCREDS );@@ -1223,7 +1233,7 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, unsigned long, start,mmput (mm); release_task:put_task_struct (task); -fdput: -fdput (f); +put_pid: +put_pid (pid);return ret; } --Quoted 2 lines
2.25.0.265.gbab2e86ba0-goog
Reviewed-by Suren Baghdasaryan <[email protected]>
PATCH v6 6/7 mm/madvise: employ mmget_still_valid for write lock
In reply to Minchan KimFrom: Oleksandr Natalenko <[email protected]>
Do the very same trick as we already do since 04f5866e41fb. KSM hints will require locking mmap_sem for write since they modify vm_flags, so for remote KSM hinting this additional check is needed.
--- mm/madvise.c | 3 +++ 1 file changed, 3 insertions(+)
mm/madvise.c +3 -0
@@ -1118,6 +1118,8 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm,if (write) {if (down_write_killable (&mm->mmap_sem))return -EINTR ; +if (current->mm != mm && !mmget_still_valid (mm)) + goto skip_mm; }else {down_read (&mm->mmap_sem); }
@@ -1169,6 +1171,7 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm, } out:blk_finish_plug (&plug); +skip_mm:if (write)up_write (&mm->mmap_sem);else
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
Suren Baghdasaryan
reviewerRe: [PATCH v6 6/7] mm/madvise: employ mmget_still_valid for write lock
Quoted 18 lines · Minchan Kim
From: Oleksandr Natalenko <[email protected]>
Do the very same trick as we already do since
04f5866e41fb. KSM hints will require lockingmmap_semfor write since they modifyvm_flags, so for remote KSM hinting this additional check is needed.Signed-off-by Oleksandr Natalenko <[email protected]>Signed-off-by Minchan Kim <[email protected]>--- mm/madvise.c | 3 +++ 1 file changed, 3 insertions(+)mm/madvise.c +0
-0
mm/madvise.c int do_madvise(struct task_struct *target_task, struct mm_struct *mm, · line 1121
Quoted 4 lines
mm/madvise.c +1
-0@@ -1118,6 +1118,8 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm,if (write) {if (down_write_killable (&mm->mmap_sem))return -EINTR ; +if (current->mm != mm && !mmget_still_valid (mm))
mmget_still_valid() seems pretty light-weight, so why not just use that without checking that the mm belongs to the current process first?
Quoted 15 lines
mm/madvise.c +2
-1+ goto skip_mm; }else {down_read (&mm->mmap_sem); }@@ -1169,6 +1171,7 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm, } out:blk_finish_plug (&plug); +skip_mm:if (write)up_write (&mm->mmap_sem);else --2.25.0.265.gbab2e86ba0-goog
Re: [PATCH v6 6/7] mm/madvise: employ mmget_still_valid for write lock
Hello.
Quoted 14 lines · Suren Baghdasaryan
Quoted 13 lines · Minchan Kim
From: Oleksandr Natalenko <[email protected]>
Do the very same trick as we already do since
04f5866e41fb. KSM hints will require lockingmmap_semfor write since they modifyvm_flags, so for remote KSM hinting this additional check is needed.Signed-off-by Oleksandr Natalenko <[email protected]>Signed-off-by Minchan Kim <[email protected]>--- mm/madvise.c | 3 +++ 1 file changed, 3 insertions(+)
Quoted 13 lines
mm/madvise.c int do_madvise(struct task_struct *target_task, struct mm_struct *mm, · line 1121
Quoted 9 lines
mm/madvise.c +1
-0@@ -1118,6 +1118,8 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm,if (write) {if (down_write_killable (&mm->mmap_sem))return -EINTR ; +if (current->mm != mm && !mmget_still_valid (mm))
mmget_still_valid()seems pretty light-weight, so why not just use that without checking that the mm belongs to the current process first?
I'd keep the checks separate to a) do not functionally change current->mm == mm case; b) clearly separate the intention to call mmget_still_valid() only for remote access (using mmget_still_valid() for current->mm == mm does not make any sense here, IMO, since there's no possibility of expecting a core dump at this point); c) ease the job for reviewer once mmget_still_valid() is scheduled to be removed (I hope it eventually goes away indeed).
Quoted 17 lines
Quoted 16 lines
mm/madvise.c +2
-1+ goto skip_mm; }else {down_read (&mm->mmap_sem); }@@ -1169,6 +1171,7 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm, } out:blk_finish_plug (&plug); +skip_mm:if (write)up_write (&mm->mmap_sem);else --2.25.0.265.gbab2e86ba0-goog
Signature · 3 lines
Best regards,
Oleksandr Natalenko (post-factum)
Principal Software Maintenance EngineerRe: [PATCH v6 6/7] mm/madvise: employ mmget_still_valid for write lock
Quoted 32 lines · Oleksandr Natalenko
Hello.
Quoted 14 lines · Suren Baghdasaryan
Quoted 13 lines · Minchan Kim
From: Oleksandr Natalenko <[email protected]>
Do the very same trick as we already do since
04f5866e41fb. KSM hints will require lockingmmap_semfor write since they modifyvm_flags, so for remote KSM hinting this additional check is needed.Signed-off-by Oleksandr Natalenko <[email protected]>Signed-off-by Minchan Kim <[email protected]>--- mm/madvise.c | 3 +++ 1 file changed, 3 insertions(+)Quoted 13 lines
mm/madvise.c int do_madvise(struct task_struct *target_task, struct mm_struct *mm, · line 1121
Quoted 9 lines
mm/madvise.c +1
-0@@ -1118,6 +1118,8 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm,if (write) {if (down_write_killable (&mm->mmap_sem))return -EINTR ; +if (current->mm != mm && !mmget_still_valid (mm))
mmget_still_valid()seems pretty light-weight, so why not just use that without checking that the mm belongs to the current process first?
Quoted 8 lines
I'd keep the checks separate to a) do not functionally change
current->mm== mm case; b) clearly separate the intention to callmmget_still_valid()only for remote access (usingmmget_still_valid()forcurrent->mm== mm does not make any sense here, IMO, since there's no possibility of expecting a core dump at this point); c) ease the job for reviewer oncemmget_still_valid()is scheduled to be removed (I hope it eventually goes away indeed).
Makes sense. Thanks!
Quoted 18 lines
Quoted 17 lines
Quoted 14 lines
mm/madvise.c +2
-1+ goto skip_mm; }else {down_read (&mm->mmap_sem); }@@ -1169,6 +1171,7 @@ int do_madvise(struct task_struct *target_task, struct mm_struct *mm, } out:blk_finish_plug (&plug); +skip_mm:if (write)up_write (&mm->mmap_sem);else --Quoted 2 lines
2.25.0.265.gbab2e86ba0-goog
Quoted 5 lines
--
Best regards,
Oleksandr Natalenko (post-factum)
Principal Software Maintenance Engineer
PATCH v6 7/7 mm/madvise: allow KSM hints for remote API
In reply to Minchan KimFrom: Oleksandr Natalenko <[email protected]>
It all began with the fact that KSM works only on memory that is marked by madvise(). And the only way to get around that is to either:
* use LD_PRELOAD; or
* patch the kernel with something like UKSM or PKSM.
(i skip ptrace can of worms here intentionally)
To overcome this restriction, lets employ a new remote madvise API. This can be used by some small userspace helper daemon that will do auto-KSM job for us.
I think of two major consumers of remote KSM hints:
* hosts, that run containers, especially similar ones and especially in
a trusted environment, sharing the same runtime like Node.js;
* heavy applications, that can be run in multiple instances, not limited to opensource ones like Firefox, but also those that cannot be modified since they are binary-only and, maybe, statically linked.
Speaking of statistics, more numbers can be found in the very first submission, that is related to this one [1]. For my current setup with two Firefox instances I get 100 to 200 MiB saved for the second instance depending on the amount of tabs.
1 FF instance with 15 tabs:
$ echo "$(cat /sys/kernel/mm/ksm/pages_sharing) * 4 / 1024" | bc 410
2 FF instances, second one has 12 tabs (all the tabs are different):
$ echo "$(cat /sys/kernel/mm/ksm/pages_sharing) * 4 / 1024" | bc 592
At the very moment I do not have specific numbers for containerised workload, but those should be comparable in case the containers share similar/same runtime.
[1] https://lore.kernel.org/patchwork/patch/1012142/
--- mm/madvise.c | 4 ++++ 1 file changed, 4 insertions(+)
mm/madvise.c +4 -0
@@ -1005,6 +1005,10 @@ process_madvise_behavior_valid(int behavior)switch (behavior) {case MADV_COLD :case MADV_PAGEOUT : +#ifdef CONFIG_KSM +case MADV_MERGEABLE : +case MADV_UNMERGEABLE : +#endif return true;default :return false;
Signature · 1 line
2.25.0.265.gbab2e86ba0-goog
Andrew Morton
maintainerQuoted 8 lines · On Tue, 18 Feb 2020 17:44:26 -0800 Minchan Kim <[email protected]> wrote:
Now, we have
MADV_PAGEOUTandMADV_COLDas madvise hinting API. With that, application could give hints to kernel what memory range are preferred to be reclaimed. However, in someplatform(e.g., Android), the information required to make the hinting decision is not known to the app.
Instead, it is known to a centralized userspacedaemon(e.g., ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.
This patchset doesn't seem to be getting a lot of interest from other potential users? It seems very specialized. Are there or will there ever be any users of this apart from one Android daemon?
Also, it doesn't terribly hard for ActivityManagerService to tell another process "now run madvise with these arguments". Please explain why this is not practical in ActivityManagerService and also within other potential users of this syscall.
Suren Baghdasaryan
reviewerQuoted 12 lines · Andrew Morton
Quoted 8 lines · On Tue, 18 Feb 2020 17:44:26 -0800 Minchan Kim <[email protected]> wrote:
Now, we have
MADV_PAGEOUTandMADV_COLDas madvise hinting API. With that, application could give hints to kernel what memory range are preferred to be reclaimed. However, in someplatform(e.g., Android), the information required to make the hinting decision is not known to the app.
Instead, it is known to a centralized userspacedaemon(e.g., ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.
Quoted 3 lines
This patchset doesn't seem to be getting a lot of interest from other potential users? It seems very specialized. Are there or will there ever be any users of this apart from one Android daemon?
Don't know if this can be considered another user since it's still in the Android realm of things.
I'm interested in extending process_madvise() to support MADV_DONTNEED to expedite memory ripping of a process killed by Android Low Memory Killer. But for that I need process_madvise() to be accepted first. IIRC Crome team was interested in these madv hints as well at some point...
Quoted 7 lines
Also, it doesn't terribly hard for ActivityManagerService to tell another process "now run madvise with these arguments". Please explain why this is not practical in ActivityManagerService and also within other potential users of this syscall.
Hi Andrew,
Quoted 11 lines · Andrew Morton
Quoted 8 lines · On Tue, 18 Feb 2020 17:44:26 -0800 Minchan Kim <[email protected]> wrote:
Now, we have
MADV_PAGEOUTandMADV_COLDas madvise hinting API. With that, application could give hints to kernel what memory range are preferred to be reclaimed. However, in someplatform(e.g., Android), the information required to make the hinting decision is not known to the app.
Instead, it is known to a centralized userspacedaemon(e.g., ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.
Quoted 3 lines
This patchset doesn't seem to be getting a lot of interest from other potential users? It seems very specialized. Are there or will there ever be any users of this apart from one Android daemon?
Quote from http://lkml.kernel.org/r/[email protected]
"
Brian Geffon in ChromeOS team had an experiment with process_madvise(2) Quote form him:
"What I found is that by using process_madvise after a tab has been back grounded for more than 45 seconds reduced the average tab switch times by 25%! This is a huge result and very obvious validation that process_madvise hints works well for the ChromeOS use case."
"
Quoted 5 lines
Also, it doesn't terribly hard for ActivityManagerService to tell another process "now run madvise with these arguments". Please explain why this is not practical in ActivityManagerService and also within other potential users of this syscall.
I think that's the almost a same question why ptrace doesn't work so I summarizes the part in [2/7]:
* makes target task runnable creates memory layout change window so hiniting a wrong vma
* target task(e.g., background task) could live in little core with cpuset/group limited environment so we couldn't react quick enough, which causes more killing.
Thanks.
To expand on how ChromeOS benefits from this, we've advanced far beyond the experimentation phase we've deployed an older version of this idea that was procfs based on several ChromeOS kernels. These are now rolled out to ChromeOS stable channel where we've been testing and the results have been amazing. To elaborate on the setup, Chrome is a multi process architecture where each tab is a separate process and sometimes a single tab can even represent multiple processes. The primary Chrome process has a lot of visibility into the amount of time a user has been spending interacting with a tab (process) and using this knowledge these hints provided to the kernel allow it to make much better swap decisions and amortize the cost of swap over different memory pressure levels meaning that we were better able to reclaim memory which allow us to avoid having to discard tabs or even worse oom.
I'd be happy to expand even more if anyone is interested.
Brian
Quoted 49 lines · Minchan Kim
Hi Andrew,
Quoted 11 lines · Andrew Morton
Quoted 8 lines · On Tue, 18 Feb 2020 17:44:26 -0800 Minchan Kim <[email protected]> wrote:
Now, we have
MADV_PAGEOUTandMADV_COLDas madvise hinting API. With that, application could give hints to kernel what memory range are preferred to be reclaimed. However, in someplatform(e.g., Android), the information required to make the hinting decision is not known to the app.
Instead, it is known to a centralized userspacedaemon(e.g., ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.Quoted 3 lines
This patchset doesn't seem to be getting a lot of interest from other potential users? It seems very specialized. Are there or will there ever be any users of this apart from one Android daemon?
Quote from http://lkml.kernel.org/r/[email protected]
"
Brian Geffon in ChromeOS team had an experiment withprocess_madvise(2)Quote form him:
"What I found is that by usingprocess_madviseafter a tab has been back grounded for more than 45 seconds reduced the average tab switch times by 25%! This is a huge result and very obvious validation thatprocess_madvisehints works well for the ChromeOS use case."
"Quoted 5 lines
Also, it doesn't terribly hard for ActivityManagerService to tell another process "now run madvise with these arguments". Please explain why this is not practical in ActivityManagerService and also within other potential users of this syscall.
I think that's the almost a same question why ptrace doesn't work so I summarizes the part in [2/7]:
* makes target task runnable creates memory layout change window so hiniting a wrong vma
* target
task(e.g., background task)could live in little core with cpuset/group limited environment so we couldn't react quick enough, which causes more killing.Thanks.
Quoted 11 lines · On Wed, 19 Feb 2020 12:01:23 -0800 Andrew Morton <[email protected]>…
Quoted 8 lines · On Tue, 18 Feb 2020 17:44:26 -0800 Minchan Kim <[email protected]> wrote:
Now, we have
MADV_PAGEOUTandMADV_COLDas madvise hinting API. With that, application could give hints to kernel what memory range are preferred to be reclaimed. However, in someplatform(e.g., Android), the information required to make the hinting decision is not known to the app.
Instead, it is known to a centralized userspacedaemon(e.g., ActivityManagerService), and that daemon must be able to initiate reclaim on its own without any app involvement.
Quoted 8 lines
This patchset doesn't seem to be getting a lot of interest from other potential users? It seems very specialized. Are there or will there ever be any users of this apart from one Android daemon?
Also, it doesn't terribly hard for ActivityManagerService to tell another process "now run madvise with these arguments". Please explain why this is not practical in ActivityManagerService and also within other potential users of this syscall.
I personally have interest in and hope successful development/merge of this patchset.
The interested usecases of 'madvise_process()' for me is optimizations of general memory-intensive workloads having dynamic data access patterns on hierarchical memory systems (e.g., multi-tier memory or fast storage based swap devices). In more detail, I'm already using a part of this patchset for my RFC patchset implementing Data Access Monitoring-based Operation Schemes[1]. For my specific case, I don't need new system call but only target task argument, though.
Once in a past, before joining my current company, I tried using 'madvise()' to optimize some scientific HPC programs. The improvement results were clear, but optimizing each of the workloads was challenging and time-consuming. I believe this new systemcall to be very helpful for such cases, either.
[1] https://lore.kernel.org/linux-mm/[email protected]/
Thanks,
SeongJae Park