xfs: reduce ilock contention on buffered randrw workload
Subsystems (4)
PATCH 1/1 xfs: reduce ilock contention on buffered randrw workload
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling generic_file_read_iter() twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.
Since this is a POC patch it also includes a separate fix to the copy_page_to_iter() helper when called with discard iterator. There is no other caller in the kernel to this method with a discard iterator as far as I could see.
---
Hi Folks,
With this experimenital patch I was able to bring performance of random rw workload benchmark on xfs much closer to ext4.
Ext4, as most Linux filesystems doesn't take the shared inode lock on buffered reads and does not provide atomic buffered reads w.r.t buffered writes.
Following are the numbers I got with filebench randomrw workload [1] on a VM with 4 CPUs and spindles.
Note that this improvement is unrelated to the rw_semaphore starvation issue that was observed when running the same benchmark on fast SDD drive [2].
=== random read/write - cold page cache === --- EXT4 ---
filebench randomrw (8 read threads, 8 write threads) kernel 5.1.0-rc2, ext4
rand-write1 862304ops 14235ops/s 111.2mb/s 0.5ms/op
rand-read1 22065ops 364ops/s 2.8mb/s 21.5ms/op
--- XFS ---
filebench randomrw (8 read threads, 8 write threads) kernel 5.1.0-rc2, xfs
rand-write1 39451ops 657ops/s 5.1mb/s 12.0ms/op rand-read1 2035ops 34ops/s 0.3mb/s 232.7ms/op
--- XFS+ ---
filebench randomrw (8 read threads, 8 write threads) kernel 5.1.0-rc2+ (xfs page cache warmup patch)
rand-write1 935597ops 15592ops/s 121.8mb/s 0.5ms/op
rand-read1 4446ops 74ops/s 0.6mb/s 107.6ms/op
To measure the effects of two passes of generic_file_read_iter(), I ran a random read [2] benchmark on 5GB file with warm and cold page cache.
=== random read - cold page cache ===
--- EXT4 ---
filebench randomread (8 read threads) - cold page cache kernel 5.1.0-rc2
rand-read1 23589ops 393ops/s 3.1mb/s 20.3ms/op
--- XFS ---
filebench randomread (8 read threads) - cold page cache kernel 5.1.0-rc2
rand-read1 20578ops 343ops/s 2.7mb/s 23.3ms/op
--- XFS+ ---
filebench randomread (8 read threads) - cold page cache kernel 5.1.0-rc2+ (xfs page cache warmup patch)
rand-read1 20476ops 341ops/s 2.7mb/s 23.4ms/op
=== random read - warm page cache ===
--- EXT4 ---
filebench randomread (8 read threads) - warm page cache kernel 5.1.0-rc2
rand-read1 58168696ops 969410ops/s 7573.5mb/s 0.0ms/op
--- XFS ---
filebench randomread (8 read threads) - warm page cache kernel 5.1.0-rc2
rand-read1 52748818ops 878951ops/s 6866.8mb/s 0.0ms/op
--- XFS+ ---
filebench randomread (8 read threads) - warm page cache kernel 5.1.0-rc2+ (xfs page cache warmup patch)
rand-read1 52770537ops 879445ops/s 6870.7mb/s 0.0ms/op
The numbers of this benchmark do not show and measurable difference for readers only workload with either cold or warm page cache.
If needed I can provide more measurments with fio or with different workloads and different drives.
Fire away!
Thanks,
Amir.
[1] https://github.com/amir73il/filebench/blob/overlayfs-devel/workloads/randomrw.f
[2] https://marc.info/?l=linux-xfs&m=155347265016053&w=2
[3] https://github.com/amir73il/filebench/blob/overlayfs-devel/workloads/randomread.f
fs/xfs/xfs_file.c | 14 ++++++++++++++ lib/iov_iter.c | 5 +++-- 2 files changed, 17 insertions(+), 2 deletions(-)
fs/xfs/xfs_file.c +14 -0
@@ -240,6 +240,20 @@ xfs_file_buffered_aio_read(if (!xfs_ilock_nowait (ip,XFS_IOLOCK_SHARED ))return -EAGAIN ; }else { +/* +* Warm up page cache to minimize time spent under +* shared ilock. +*/ +struct iov_iter iter; +loff_t pos = iocb->ki_pos; + +iov_iter_discard (&iter,READ ,iov_iter_count (to)); + ret =generic_file_read_iter (iocb, &iter); +if (ret <=0 ) +return ret; + + iocb->ki_pos = pos; +xfs_ilock (ip,XFS_IOLOCK_SHARED ); } ret =generic_file_read_iter (iocb, to);
lib/iov_iter.c +3 -2
@@ -893,9 +893,10 @@ size_t copy_page_to_iter(struct page *page, size_t offset, size_t bytes,size_t wanted =copy_to_iter (kaddr + offset, bytes, i);kunmap_atomic (kaddr);return wanted; - }else if (unlikely (iov_iter_is_discard (i))) + }else if (unlikely (iov_iter_is_discard (i))) { + i->count -= bytes;return bytes; -else if (likely (!iov_iter_is_pipe (i))) + }else if (likely (!iov_iter_is_pipe (i)))return copy_page_to_iter_iovec (page, offset, bytes, i);else return copy_page_to_iter_pipe (page, offset, bytes, i);
Signature · 1 line
2.7.4
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
FYI, I'm working on a range lock implementation that should both solve the performance issue and the reader starvation issue at the same time by allowing concurrent buffered reads and writes to different file ranges.
IO range locks will allow proper exclusion for other extent manipulation operations like fallocate and truncate, and eventually even allow truncate, hole punch, file extension, etc to run concurrently with other non-overlapping IO. They solve more than just the performance issue you are seeing....
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.
Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
This statement leaves me wondering, if ext4 does not takes i_rwsem on generic_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case is i_mmap_sem, which is read locked in the page fault handler.
And xfs does the same type of synchronization with MMAPLOCK, so while my patch may not be safe, I cannot follow why from your explanation, so please explain if I am missing something.
One thing that Darrick mentioned earlier was that IOLOCK is also used by xfs to synchronization pNFS leases (probably listed under 'etc' in your explanation). I consent that my patch does not look safe w.r.t pNFS leases, but that can be sorted out with a hammer #ifndef CONFIG_EXPORTFS_BLOCK_OPS
or with finer instruments.
Quoted 11 lines
FYI, I'm working on a range lock implementation that should both solve the performance issue and the reader starvation issue at the same time by allowing concurrent buffered reads and writes to different file ranges.
IO range locks will allow proper exclusion for other extent manipulation operations like fallocate and truncate, and eventually even allow truncate, hole punch, file extension, etc to run concurrently with other non-overlapping IO. They solve more than just the performance issue you are seeing....
I'm glad to hear that. IO range locks are definitely a more wholesome solution to the problem looking forward.
However, I am still interested to continue the discussion on my POC patch. One reason is that I am guessing it would be much easier for distros to backport and pick up to solve performance issues.
Even if my patch doesn't get applied upstream nor picked by distros, I would still like to understand its flaws and limitations. I know... if I break it, I get to keep the pieces, but the information that you provide helps me make my risk assessments.
Thanks,
Amir.
Quoted 19 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
Quoted 6 lines
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.
Nope, the i_mmap_sem is for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to the read() path.
Quoted 3 lines
And xfs does the same type of synchronization with MMAPLOCK, so while my patch may not be safe, I cannot follow why from your explanation, so please explain if I am missing something.
mmap_sem inversions require independent locks for IO path and page faults - the MMAPLOCK does not protect anything in the read()/write() IO path.
Quoted 3 lines
One thing that Darrick mentioned earlier was that IOLOCK is also used by xfs to synchronization pNFS leases (probably listed under 'etc' in your explanation).
PNFS leases are separate to the IO locking. All the IO locking does is serialise new IO submission against the process of breaking leases so that extent maps that have been shared under the lease are invalidated correctly. i.e. we can't start IO until the lease has been recalled and the external client has guaranteed it won't read/write data from the stale extent map.
If you do IO outside the IOLOCK, then you break those serialisation guarantees and risk data corruption and/or stale data exposure...
Quoted 4 lines
I consent that my patch does not look safe
w.r.t pNFS leases, but that can be sorted out with a hammer #ifndefCONFIG_EXPORTFS_BLOCK_OPS
or with finer instruments.
All you see is this:
truncate: read()
IOLOCK_EXCL
flush relevant cached data
truncate page cache
pre-read page cache between new eof and old eof IOLOCK_SHARED <blocks>
start transaction
ILOCK_EXCL
update isize
remove extents
....
commit xactn
IOLOCK unlock
<gets lock>
sees beyond EOF, returns 0
So you see the read() doing the right thing (detect EOF, returning short read). Great.
But what I see is uptodate pages containing stale data being left in the page cache beyond EOF. That is th eproblem here - truncate must not leave stale pages beyond EOF behind - it's the landmine that causes future things to go wrong.
e.g. now the app does post-eof preallocation so the range those pages are cached over are allocated as unwritten - the filesystem will do this without even looking at the page cache because it's beyond EOF. Now we extend the file past those cached pages, and iomap_zero() sees the range as unwritten and so does not write zeros to the blocks between the old EOF and the new EOF. Now the app reads from that range (say it does a sub-page write, triggering a page cache RMW cycle). the read goes to instantiate the page cache page, finds a page already in the cache that is uptodate, and uses it without zeroing or reading from disk.
And now we have stale data exposure and/or data corruption.
If can come up with quite a few scenarios where this particular "populate cache after invalidation" race can cause similar problems for XFS. Hole punch and most of the other fallocate extent manipulations have the same serialisation requirements - no IO over the range of the operation can be *initiated* between the /start/ of the page cache invalidation and the end of the specific extent manipulation operation.
So how does ext4 avoid this problem on truncate?
History lesson: truncate in Linux (and hence ext4) has traditionally been serialised by the hacky post-page-lock checks that are strewn all through the page cache and mm/ subsystem. i.e. every time you look up and lock a page cache page, you have to check the page->mapping and page->index to ensure that the lookup-and-lock hasn't raced with truncate. This only works because truncate requires the inode size to be updated before invalidating the page cache - that's the "hacky" part of it.
IOWs, the burden of detecting truncate races is strewn throughout the mm/ subsystem, rather than being the responisibility of the filesystem. This is made worse by the fact this mechanism simply doesn't work for hole punching because there is no file size change to indicate that the page lookup is racing with an in-progress invalidation.
That means the mm/ and page cache code is unable to detect hole punch races, and so the serialisation of invalidation vs page cache instantiation has to be done in the filesystem. And no Linux native filesystem had the infrastructure for such serialisation because they never had to implement anything to ensure truncate was serialised against new and in-progress IO.
The result of this is that, AFAICT, ext4 does not protect against read() vs hole punch races - it's hole punching code it does:
Hole Punch: read():
inode_lock()inode_dio_wait(inode);down_write(i_mmap_sem)truncate_pagecache_range()
ext4_file_iter_read() ext4_map_blocks() down_read(i_data_sem) <gets mapping> <populates page cache over hole> <reads stale data into cache> .....
down_write(i_data_sem)
remove extents
IOWs, ext4 is safe against truncate because of the change-inode-size-before-invalidation hacks, but the lack of serialise buffered reads means that hole punch and other similar fallocate based extent manipulations can race against reads....
Quoted 3 lines
However, I am still interested to continue the discussion on my POC patch. One reason is that I am guessing it would be much easier for distros to backport and pick up to solve performance issues.
Upstream first, please. If it's not fit for upstream, then it most definitely is not fit for backporting to distro kernels.
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Quoted 28 lines · Dave Chinner
Quoted 19 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
Quoted 6 lines
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.
Quoted 4 lines
Nope, the
i_mmap_semis for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to theread()path.
I'm at lost here. Why are page faults completely irrelevant to read() path? Aren't full pages supposed to be faulted in on read() after truncate_pagecache_range()? And aren't partial pages supposed to be partially zeroed and uptodate after truncate_pagecache_range()?
Quoted 8 lines
Quoted 3 lines
And xfs does the same type of synchronization with MMAPLOCK, so while my patch may not be safe, I cannot follow why from your explanation, so please explain if I am missing something.
mmap_seminversions require independent locks for IO path and page faults - the MMAPLOCK does not protect anything in the read()/write() IO path.
[...]
Quoted 93 lines
All you see is this:
truncate:
read()
IOLOCK_EXCL
flush relevant cached data
truncate page cache
pre-read page cache between new eof and old eofIOLOCK_SHARED<blocks>
start transaction
ILOCK_EXCL
update isize
remove extents
....
commit xactn
IOLOCK unlock
<gets lock>
sees beyond EOF, returns 0So you see the
read()doing the right thing (detect EOF, returning short read). Great.But what I see is uptodate pages containing stale data being left in the page cache beyond EOF. That is th eproblem here - truncate must not leave stale pages beyond EOF behind - it's the landmine that causes future things to go wrong.
e.g. now the app does post-eof preallocation so the range those pages are cached over are allocated as unwritten - the filesystem will do this without even looking at the page cache because it's beyond EOF. Now we extend the file past those cached pages, and
iomap_zero()sees the range as unwritten and so does not write zeros to the blocks between the old EOF and the new EOF. Now the app reads from that range (say it does a sub-page write, triggering a page cache RMW cycle). the read goes to instantiate the page cache page, finds a page already in the cache that is uptodate, and uses it without zeroing or reading from disk.And now we have stale data exposure and/or data corruption.
If can come up with quite a few scenarios where this particular "populate cache after invalidation" race can cause similar problems for XFS. Hole punch and most of the other fallocate extent manipulations have the same serialisation requirements - no IO over the range of the operation can be *initiated* between the /start/ of the page cache invalidation and the end of the specific extent manipulation operation.
So how does ext4 avoid this problem on truncate?
History lesson: truncate in Linux (and hence ext4) has traditionally been serialised by the hacky post-page-lock checks that are strewn all through the page cache and
mm/subsystem. i.e. every time you look up and lock a page cache page, you have to check thepage->mappingandpage->indexto ensure that the lookup-and-lock hasn't raced with truncate. This only works because truncate requires the inode size to be updated before invalidating the page cache - that's the "hacky" part of it.IOWs, the burden of detecting truncate races is strewn throughout the
mm/subsystem, rather than being the responisibility of the filesystem. This is made worse by the fact this mechanism simply doesn't work for hole punching because there is no file size change to indicate that the page lookup is racing with an in-progress invalidation.That means the
mm/and page cache code is unable to detect hole punch races, and so the serialisation of invalidation vs page cache instantiation has to be done in the filesystem. And no Linux native filesystem had the infrastructure for such serialisation because they never had to implement anything to ensure truncate was serialised against new and in-progress IO.The result of this is that, AFAICT, ext4 does not protect against
read()vs hole punch races - it's hole punching code it does:Hole Punch:
read():
inode_lock()inode_dio_wait(inode);down_write(i_mmap_sem)truncate_pagecache_range()ext4_file_iter_read() ext4_map_blocks() down_read(i_data_sem) <gets mapping> <populates page cache over hole> <reads stale data into cache> .....
Quoted 8 lines
down_write(i_data_sem)
remove extentsIOWs, ext4 is safe against truncate because of the change-inode-size-before-invalidation hacks, but the lack of serialise buffered reads means that hole punch and other similar fallocate based extent manipulations can race against reads....
Adding some ext4 folks to comment on the above.
Could it be that those races were already addressed by Lukas' work:
https://lore.kernel.org/patchwork/cover/371861/
Thanks,
Amir.
Jan Kara
reviewerQuoted 29 lines · Amir Goldstein
Quoted 28 lines · Dave Chinner
Quoted 25 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.
Quoted 8 lines
Quoted 4 lines
Nope, the
i_mmap_semis for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to theread()path.I'm at lost here. Why are page faults completely irrelevant to
read()path? Aren't full pages supposed to be faulted in onread()aftertruncate_pagecache_range()?
During read(2), pages are not "faulted in". Just look at
what generic_file_buffered_read() does. It uses completely separate code to add page to page cache, trigger readahead, and possibly call ->readpage() to fill the page with data. "fault" path (handled by filemap_fault()) applies only to accesses from userspace to mmaps.
Quoted 2 lines
And aren't partial pages supposed to be partially zeroed and uptodate after
truncate_pagecache_range()?
truncate_pagecache_range() does take care of page zeroing, that is correct (if those pages are in page cache in the first place). I'm not sure how this relates to the above discussion though.
Honza
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Quoted 40 lines · Jan Kara
Quoted 29 lines · Amir Goldstein
Quoted 28 lines · Dave Chinner
Quoted 25 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.Quoted 8 lines
Quoted 4 lines
Nope, the
i_mmap_semis for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to theread()path.I'm at lost here. Why are page faults completely irrelevant to
read()path? Aren't full pages supposed to be faulted in onread()aftertruncate_pagecache_range()?
Quoted 6 lines
During read(2), pages are not "faulted in". Just look at
whatgeneric_file_buffered_read()does. It uses completely separate code to add page to page cache, trigger readahead, and possibly call ->readpage() to fill the page with data. "fault" path (handled byfilemap_fault()) applies only to accesses from userspace to mmaps.
Oh! thanks for fixing my blind spot.
So if you agree with Dave that ext4, and who knows what other fs, are vulnerable to populating page cache with stale "uptodate" data, then it seems to me that also xfs is vulnerable via readahead(2) and posix_fadvise().
Mind you, I recently added an fadvise f_op, so it could be used by xfs to synchronize with IOLOCK.
Perhaps a better solution would be for truncate_pagecache_range() to leave zeroed or Unwritten (i.e. lazy zeroed by read) pages in page cache. When we have shared pages for files, these pages could be deduped.
Thanks,
Amir.
Jan Kara
reviewerQuoted 48 lines · Amir Goldstein
Quoted 40 lines · Jan Kara
Quoted 29 lines · Amir Goldstein
Quoted 28 lines · Dave Chinner
Quoted 25 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.Quoted 8 lines
Quoted 4 lines
Nope, the
i_mmap_semis for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to theread()path.I'm at lost here. Why are page faults completely irrelevant to
read()path? Aren't full pages supposed to be faulted in onread()aftertruncate_pagecache_range()?Quoted 6 lines
During read(2), pages are not "faulted in". Just look at
whatgeneric_file_buffered_read()does. It uses completely separate code to add page to page cache, trigger readahead, and possibly call ->readpage() to fill the page with data. "fault" path (handled byfilemap_fault()) applies only to accesses from userspace to mmaps.
Quoted 3 lines
Oh! thanks for fixing my blind spot.
So if you agree with Dave that ext4, and who knows what other fs, are vulnerable to populating page cache with stale "uptodate" data,
Not that many filesystems support punching holes but you're right.
Quoted 2 lines
then it seems to me that also xfs is vulnerable via readahead(2) and
posix_fadvise().
Yes, this is correct AFAICT.
Quoted 2 lines
Mind you, I recently added an fadvise
f_op, so it could be used by xfs to synchronize with IOLOCK.
And yes, this should work.
Quoted 4 lines
Perhaps a better solution would be for
truncate_pagecache_range()to leave zeroed or Unwritten (i.e. lazy zeroed by read) pages in page cache. When we have shared pages for files, these pages could be deduped.
No, I wouldn't really mess with sharing pages due to this. It would be hard to make that scale resonably and would be rather complex. We really need a proper and reasonably simple synchronization mechanism between operations removing blocks from inode and operations filling in page cache of the inode. Page lock was supposed to provide this but doesn't quite work because hole punching first remove pagecache pages and then go removing all blocks.
So I agree with Dave that going for range lock is really the cleanest way forward here without causing big regressions for mixed rw workloads. I'm just thinking how to best do that without introducing lot of boilerplate code into each filesystem.
Honza
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Quoted 79 lines · Jan Kara
Quoted 48 lines · Amir Goldstein
Quoted 40 lines · Jan Kara
Quoted 29 lines · Amir Goldstein
Quoted 28 lines · Dave Chinner
Quoted 25 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.Quoted 8 lines
Quoted 4 lines
Nope, the
i_mmap_semis for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to theread()path.I'm at lost here. Why are page faults completely irrelevant to
read()path? Aren't full pages supposed to be faulted in onread()aftertruncate_pagecache_range()?Quoted 6 lines
During read(2), pages are not "faulted in". Just look at
whatgeneric_file_buffered_read()does. It uses completely separate code to add page to page cache, trigger readahead, and possibly call ->readpage() to fill the page with data. "fault" path (handled byfilemap_fault()) applies only to accesses from userspace to mmaps.Quoted 3 lines
Oh! thanks for fixing my blind spot.
So if you agree with Dave that ext4, and who knows what other fs, are vulnerable to populating page cache with stale "uptodate" data,Not that many filesystems support punching holes but you're right.
Quoted 2 lines
then it seems to me that also xfs is vulnerable via readahead(2) and
posix_fadvise().Yes, this is correct AFAICT.
Quoted 2 lines
Mind you, I recently added an fadvise
f_op, so it could be used by xfs to synchronize with IOLOCK.And yes, this should work.
Quoted 4 lines
Perhaps a better solution would be for
truncate_pagecache_range()to leave zeroed or Unwritten (i.e. lazy zeroed by read) pages in page cache. When we have shared pages for files, these pages could be deduped.No, I wouldn't really mess with sharing pages due to this. It would be hard to make that scale resonably and would be rather complex. We really need a proper and reasonably simple synchronization mechanism between operations removing blocks from inode and operations filling in page cache of the inode. Page lock was supposed to provide this but doesn't quite work because hole punching first remove pagecache pages and then go removing all blocks.
Quoted 4 lines
So I agree with Dave that going for range lock is really the cleanest way forward here without causing big regressions for mixed rw workloads. I'm just thinking how to best do that without introducing lot of boilerplate code into each filesystem.
Hi Jan, Dave,
Trying to circle back to this after 3 years!
Seeing that there is no progress with range locks and
that the mixed rw workloads performance issue still very much exists.
Is the situation now different than 3 years ago with invalidate_lock? Would my approach of pre-warm page cache before taking IOLOCK be safe if page cache is pre-warmed with invalidate_lock held?
For the pNFS leases issue, as I wrote back in pre-COVID era, I intend to opt-out of this optimization with #ifndef CONFIG_EXPORTFS_BLOCK_OPS
Thanks,
Amir.
Jan Kara
reviewerQuoted 85 lines · Amir Goldstein
Quoted 79 lines · Jan Kara
Quoted 48 lines · Amir Goldstein
Quoted 40 lines · Jan Kara
Quoted 29 lines · Amir Goldstein
Quoted 28 lines · Dave Chinner
Quoted 25 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.Quoted 8 lines
Quoted 4 lines
Nope, the
i_mmap_semis for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to theread()path.I'm at lost here. Why are page faults completely irrelevant to
read()path? Aren't full pages supposed to be faulted in onread()aftertruncate_pagecache_range()?Quoted 6 lines
During read(2), pages are not "faulted in". Just look at
whatgeneric_file_buffered_read()does. It uses completely separate code to add page to page cache, trigger readahead, and possibly call ->readpage() to fill the page with data. "fault" path (handled byfilemap_fault()) applies only to accesses from userspace to mmaps.Quoted 3 lines
Oh! thanks for fixing my blind spot.
So if you agree with Dave that ext4, and who knows what other fs, are vulnerable to populating page cache with stale "uptodate" data,Not that many filesystems support punching holes but you're right.
Quoted 2 lines
then it seems to me that also xfs is vulnerable via readahead(2) and
posix_fadvise().Yes, this is correct AFAICT.
Quoted 2 lines
Mind you, I recently added an fadvise
f_op, so it could be used by xfs to synchronize with IOLOCK.And yes, this should work.
Quoted 4 lines
Perhaps a better solution would be for
truncate_pagecache_range()to leave zeroed or Unwritten (i.e. lazy zeroed by read) pages in page cache. When we have shared pages for files, these pages could be deduped.No, I wouldn't really mess with sharing pages due to this. It would be hard to make that scale resonably and would be rather complex. We really need a proper and reasonably simple synchronization mechanism between operations removing blocks from inode and operations filling in page cache of the inode. Page lock was supposed to provide this but doesn't quite work because hole punching first remove pagecache pages and then go removing all blocks.
Quoted 4 lines
So I agree with Dave that going for range lock is really the cleanest way forward here without causing big regressions for mixed rw workloads. I'm just thinking how to best do that without introducing lot of boilerplate code into each filesystem.
Quoted 7 lines
Hi Jan, Dave,
Trying to circle back to this after 3 years!
Seeing that there is no progress with range locks and
that the mixed rw workloads performance issue still very much exists.Is the situation now different than 3 years ago with
invalidate_lock?
Yes, I've implemented invalidate_lock exactly to fix the issues you've pointed out without regressing the mixed rw workloads (because invalidate_lock is taken in shared mode only for reads and usually not at all for writes).
Quoted 2 lines
Would my approach of pre-warm page cache before taking IOLOCK be safe if page cache is pre-warmed with
invalidate_lockheld?
Why would it be needed? But yes, with invalidate_lock you could presumably make that idea safe...
Honza
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Quoted 100 lines · Jan Kara
Quoted 85 lines · Amir Goldstein
Quoted 79 lines · Jan Kara
Quoted 48 lines · Amir Goldstein
Quoted 40 lines · Jan Kara
Quoted 29 lines · Amir Goldstein
Quoted 28 lines · Dave Chinner
Quoted 25 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.Quoted 8 lines
Quoted 4 lines
Nope, the
i_mmap_semis for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to theread()path.I'm at lost here. Why are page faults completely irrelevant to
read()path? Aren't full pages supposed to be faulted in onread()aftertruncate_pagecache_range()?Quoted 6 lines
During read(2), pages are not "faulted in". Just look at
whatgeneric_file_buffered_read()does. It uses completely separate code to add page to page cache, trigger readahead, and possibly call ->readpage() to fill the page with data. "fault" path (handled byfilemap_fault()) applies only to accesses from userspace to mmaps.Quoted 3 lines
Oh! thanks for fixing my blind spot.
So if you agree with Dave that ext4, and who knows what other fs, are vulnerable to populating page cache with stale "uptodate" data,Not that many filesystems support punching holes but you're right.
Quoted 2 lines
then it seems to me that also xfs is vulnerable via readahead(2) and
posix_fadvise().Yes, this is correct AFAICT.
Quoted 2 lines
Mind you, I recently added an fadvise
f_op, so it could be used by xfs to synchronize with IOLOCK.And yes, this should work.
Quoted 4 lines
Perhaps a better solution would be for
truncate_pagecache_range()to leave zeroed or Unwritten (i.e. lazy zeroed by read) pages in page cache. When we have shared pages for files, these pages could be deduped.No, I wouldn't really mess with sharing pages due to this. It would be hard to make that scale resonably and would be rather complex. We really need a proper and reasonably simple synchronization mechanism between operations removing blocks from inode and operations filling in page cache of the inode. Page lock was supposed to provide this but doesn't quite work because hole punching first remove pagecache pages and then go removing all blocks.
Quoted 4 lines
So I agree with Dave that going for range lock is really the cleanest way forward here without causing big regressions for mixed rw workloads. I'm just thinking how to best do that without introducing lot of boilerplate code into each filesystem.
Quoted 7 lines
Hi Jan, Dave,
Trying to circle back to this after 3 years!
Seeing that there is no progress with range locks and
that the mixed rw workloads performance issue still very much exists.Is the situation now different than 3 years ago with
invalidate_lock?Yes, I've implemented
invalidate_lockexactly to fix the issues you've pointed out without regressing the mixed rw workloads (becauseinvalidate_lockis taken in shared mode only for reads and usually not at all for writes).
Quoted 5 lines
Quoted 2 lines
Would my approach of pre-warm page cache before taking IOLOCK be safe if page cache is pre-warmed with
invalidate_lockheld?Why would it be needed? But yes, with
invalidate_lockyou could presumably make that idea safe...
To remind you, the context in which I pointed you to the punch hole race issue in "other file systems" was a discussion about trying to relax the "atomic write"
POSIX semantics [1] of xfs.
There was a lot of discussions around range locks and changing the fairness of rwsem readers and writer, but none of this changes the fact that as long as the lock is file wide (and it does not look like that is going to change in the near
future), it is better for lock contention to perform the serialization on page cache
read/write and not on disk read/write.
Therefore, *if* it is acceptable to pre-warn page cache for buffered read under invalidate_lock, that is a simple way to bring the xfs performance with random rw mix workload on par with ext4 performance without losing the atomic write POSIX semantics. So everyone can be happy?
In addition to Dave's concerns about stale page cache races with hole punch, I found in the original discussion these concern from Darrick:
Quoted 5 lines
Reads and writes are not the only thing xfs uses
i_rwsemto synchronise. Reflink remap uses it to make sure everything's flushed to disk and that page cache contents remain clean while the remap is ongoing. I'm pretty sure pnfs uses it for similar reasons when granting and committing write leases.
To reiterate, pNFS leases are not the common case. To address this issue, I intend to opt-out of pre-warm optimization when pNFS leases are present, either globally, or per file, whatever xfs developers tell me to do.
From my understanding of the code, xfs_reflink_remap_prep() takes care of taking invalidate_lock(s), so populating page cache under invalidate_lock should be safe also w.r.t reflink/dedupe.
Darrick, am I missing anything?
Thanks,
Amir.
[1] https://lore.kernel.org/all/CAOQ4uxgSc7hK1=GuUajzG1Z+ks6gzFFX+EtuBMULOk0s85zi3A@mail.gmail.com/
[2] https://lore.kernel.org/linux-xfs/20190325154731.GT1183@magnolia/
Jan Kara
reviewerQuoted 104 lines · Amir Goldstein
Quoted 103 lines · Jan Kara
Quoted 85 lines · Amir Goldstein
Quoted 79 lines · Jan Kara
Quoted 48 lines · Amir Goldstein
Quoted 40 lines · Jan Kara
Quoted 29 lines · Amir Goldstein
Quoted 28 lines · Dave Chinner
Quoted 25 lines · Amir Goldstein
Quoted 10 lines · Dave Chinner
Quoted 7 lines · Amir Goldstein
This patch improves performance of mixed random rw workload on xfs without relaxing the atomic buffered read/write guaranty that xfs has always provided.
We achieve that by calling
generic_file_read_iter()twice. Once with a discard iterator to warm up page cache before taking the shared ilock and once again under shared ilock.Quoted 7 lines
This will race with thing like truncate, hole punching, etc that serialise IO and invalidate the page cache for data integrity reasons under the IOLOCK. These rely on there being no IO to the inode in progress at all to work correctly, which this patch violates. IOWs, while this is fast, it is not safe and so not a viable approach to solving the problem.
This statement leaves me wondering, if ext4 does not takes
i_rwsemongeneric_file_read_iter(), how does ext4 (or any other fs for that matter) guaranty buffered read synchronization with truncate, hole punching etc?
The answer in ext4 case isi_mmap_sem, which is read locked in the page fault handler.Quoted 8 lines
Quoted 4 lines
Nope, the
i_mmap_semis for serialisation of /page faults/ against truncate, holepunching, etc. Completely irrelevant to theread()path.I'm at lost here. Why are page faults completely irrelevant to
read()path? Aren't full pages supposed to be faulted in onread()aftertruncate_pagecache_range()?Quoted 6 lines
During read(2), pages are not "faulted in". Just look at
whatgeneric_file_buffered_read()does. It uses completely separate code to add page to page cache, trigger readahead, and possibly call ->readpage() to fill the page with data. "fault" path (handled byfilemap_fault()) applies only to accesses from userspace to mmaps.Quoted 3 lines
Oh! thanks for fixing my blind spot.
So if you agree with Dave that ext4, and who knows what other fs, are vulnerable to populating page cache with stale "uptodate" data,Not that many filesystems support punching holes but you're right.
Quoted 2 lines
then it seems to me that also xfs is vulnerable via readahead(2) and
posix_fadvise().Yes, this is correct AFAICT.
Quoted 2 lines
Mind you, I recently added an fadvise
f_op, so it could be used by xfs to synchronize with IOLOCK.And yes, this should work.
Quoted 4 lines
Perhaps a better solution would be for
truncate_pagecache_range()to leave zeroed or Unwritten (i.e. lazy zeroed by read) pages in page cache. When we have shared pages for files, these pages could be deduped.No, I wouldn't really mess with sharing pages due to this. It would be hard to make that scale resonably and would be rather complex. We really need a proper and reasonably simple synchronization mechanism between operations removing blocks from inode and operations filling in page cache of the inode. Page lock was supposed to provide this but doesn't quite work because hole punching first remove pagecache pages and then go removing all blocks.
Quoted 4 lines
So I agree with Dave that going for range lock is really the cleanest way forward here without causing big regressions for mixed rw workloads. I'm just thinking how to best do that without introducing lot of boilerplate code into each filesystem.
Quoted 7 lines
Hi Jan, Dave,
Trying to circle back to this after 3 years!
Seeing that there is no progress with range locks and
that the mixed rw workloads performance issue still very much exists.Is the situation now different than 3 years ago with
invalidate_lock?Yes, I've implemented
invalidate_lockexactly to fix the issues you've pointed out without regressing the mixed rw workloads (becauseinvalidate_lockis taken in shared mode only for reads and usually not at all for writes).Quoted 2 lines
Would my approach of pre-warm page cache before taking IOLOCK be safe if page cache is pre-warmed with
invalidate_lockheld?
Quoted 6 lines
Quoted 2 lines
Why would it be needed? But yes, with
invalidate_lockyou could presumably make that idea safe...To remind you, the context in which I pointed you to the punch hole race issue in "other file systems" was a discussion about trying to relax the "atomic write" POSIX semantics [1] of xfs.
Ah, I see. Sorry, I already forgot :-|
Quoted 11 lines
There was a lot of discussions around range locks and changing the fairness of rwsem readers and writer, but none of this changes the fact that as long as the lock is file wide (and it does not look like that is going to change in the near future), it is better for lock contention to perform the serialization on page cache read/write and not on disk read/write.
Therefore, *if* it is acceptable to pre-warn page cache for buffered read under
invalidate_lock, that is a simple way to bring the xfs performance with random rw mix workload on par with ext4 performance without losing the atomic write POSIX semantics. So everyone can be happy?
So to spell out your proposal so that we are on the same page: you want to use invalidate_lock + page locks to achieve "writes are atomic wrt reads" property XFS currently has without holding i_rwsem in shared mode during reads. Am I getting it correct?
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
Honza
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Quoted 16 lines
Quoted 16 lines
Quoted 16 lines
Quoted 7 lines
Hi Jan, Dave,
Trying to circle back to this after 3 years!
Seeing that there is no progress with range locks and
that the mixed rw workloads performance issue still very much exists.Is the situation now different than 3 years ago with
invalidate_lock?Yes, I've implemented
invalidate_lockexactly to fix the issues you've pointed out without regressing the mixed rw workloads (becauseinvalidate_lockis taken in shared mode only for reads and usually not at all for writes).Quoted 2 lines
Would my approach of pre-warm page cache before taking IOLOCK be safe if page cache is pre-warmed with
invalidate_lockheld?
Quoted 8 lines
Quoted 6 lines
Quoted 2 lines
Why would it be needed? But yes, with
invalidate_lockyou could presumably make that idea safe...To remind you, the context in which I pointed you to the punch hole race issue in "other file systems" was a discussion about trying to relax the "atomic write" POSIX semantics [1] of xfs.
Ah, I see. Sorry, I already forgot :-|
Understandable. It has been 3 years ;-)
Quoted 13 lines
Quoted 8 lines
There was a lot of discussions around range locks and changing the fairness of rwsem readers and writer, but none of this changes the fact that as long as the lock is file wide (and it does not look like that is going to change in the near future), it is better for lock contention to perform the serialization on page cache read/write and not on disk read/write.
Quoted 4 lines
Therefore, *if* it is acceptable to pre-warn page cache for buffered read under
invalidate_lock, that is a simple way to bring the xfs performance with random rw mix workload on par with ext4 performance without losing the atomic write POSIX semantics. So everyone can be happy?
Quoted 4 lines
So to spell out your proposal so that we are on the same page: you want to use
invalidate_lock+ page locks to achieve "writes are atomic wrt reads" property XFS currently has without holdingi_rwsemin shared mode during reads. Am I getting it correct?
Not exactly.
Quoted 5 lines
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
I imagine that the atomic r/w synchronisation will remain *exactly* as it is today by taking XFS_IOLOCK_SHARED around generic_file_read_iter(), when reading data into user buffer, but before that, I would like to issue and wait for read of the pages in the range to reduce the probability of doing the read I/O under XFS_IOLOCK_SHARED.
The pre-warm of page cache does not need to abide to the atomic read semantics and it is also tolerable if some pages are evicted in between pre-warn and read to user buffer - in the worst case this will result in I/O amplification, but for the common case, it will be a big win for the mixed random r/w performance on xfs.
To reduce risk of page cache thrashing we can limit this optimization to a maximum number of page cache pre-warm.
The questions are:
1. Does this plan sound reasonable?
2. Is there a ready helper (force_page_cache_readahead?) that I can use which takes the required page/invalidate locks?
Thanks,
Amir.
Jan Kara
reviewerQuoted 17 lines · Amir Goldstein
Quoted 4 lines
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
I imagine that the atomic r/w synchronisation will remain *exactly* as it is today by taking
XFS_IOLOCK_SHAREDaroundgeneric_file_read_iter(), when reading data into user buffer, but before that, I would like to issue and wait for read of the pages in the range to reduce the probability of doing the read I/O underXFS_IOLOCK_SHARED.The pre-warm of page cache does not need to abide to the atomic read semantics and it is also tolerable if some pages are evicted in between pre-warn and read to user buffer - in the worst case this will result in I/O amplification, but for the common case, it will be a big win for the mixed random r/w performance on xfs.
Quoted 5 lines
To reduce risk of page cache thrashing we can limit this optimization to a maximum number of page cache pre-warm.
The questions are:
1. Does this plan sound reasonable?
Ah, I see now. So essentially the idea is to pull the readahead (which is currently happening from filemap_read() -> filemap_get_pages()) out from under the i_rwsem. It looks like a fine idea to me.
Quoted 2 lines
2. Is there a ready helper (
force_page_cache_readahead?) that I can use which takes the required page/invalidate locks?
page_cache_sync_readahead() should be the function you need. It does take care to lock invalidate_lock internally when creating & reading pages. I just cannot comment on whether calling this without i_rwsem does not break some internal XFS expectations for stuff like reflink etc.
Honza
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Quoted 22 lines · Jan Kara
Quoted 20 lines · Amir Goldstein
Quoted 4 lines
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
I imagine that the atomic r/w synchronisation will remain *exactly* as it is today by taking
XFS_IOLOCK_SHAREDaroundgeneric_file_read_iter(), when reading data into user buffer, but before that, I would like to issue and wait for read of the pages in the range to reduce the probability of doing the read I/O underXFS_IOLOCK_SHARED.The pre-warm of page cache does not need to abide to the atomic read semantics and it is also tolerable if some pages are evicted in between pre-warn and read to user buffer - in the worst case this will result in I/O amplification, but for the common case, it will be a big win for the mixed random r/w performance on xfs.
To reduce risk of page cache thrashing we can limit this optimization to a maximum number of page cache pre-warm.
Quoted 6 lines
Quoted 2 lines
The questions are:
1. Does this plan sound reasonable?Ah, I see now. So essentially the idea is to pull the readahead (which is currently happening from
filemap_read()->filemap_get_pages()) out from under thei_rwsem. It looks like a fine idea to me.
Great!
Anyone doesn't like the idea or has another suggestion?
Quoted 6 lines
Quoted 3 lines
2. Is there a ready helper (
force_page_cache_readahead?) that I can use which takes the required page/invalidate locks?
page_cache_sync_readahead()should be the function you need. It does take care to lockinvalidate_lockinternally when creating & reading pages. I
Thanks, I'll try that.
Quoted 2 lines
just cannot comment on whether calling this without
i_rwsemdoes not break some internal XFS expectations for stuff like reflink etc.
relink is done under xfs_ilock2_io_mmap => filemap_invalidate_lock_two so it should not be a problem.
pNFS leases I need to look into.
Thanks,
Amir.
Quoted 26 lines · Amir Goldstein
Quoted 25 lines · Jan Kara
Quoted 22 lines · Amir Goldstein
Quoted 4 lines
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
I imagine that the atomic r/w synchronisation will remain *exactly* as it is today by taking
XFS_IOLOCK_SHAREDaroundgeneric_file_read_iter(), when reading data into user buffer, but before that, I would like to issue and wait for read of the pages in the range to reduce the probability of doing the read I/O underXFS_IOLOCK_SHARED.The pre-warm of page cache does not need to abide to the atomic read semantics and it is also tolerable if some pages are evicted in between pre-warn and read to user buffer - in the worst case this will result in I/O amplification, but for the common case, it will be a big win for the mixed random r/w performance on xfs.
To reduce risk of page cache thrashing we can limit this optimization to a maximum number of page cache pre-warm.
The questions are:
1. Does this plan sound reasonable?
Quoted 6 lines
Quoted 3 lines
Ah, I see now. So essentially the idea is to pull the readahead (which is currently happening from
filemap_read()->filemap_get_pages()) out from under thei_rwsem. It looks like a fine idea to me.Great!
Anyone doesn't like the idea or has another suggestion?
I guess I'm still confused.
The problem was the the XFS IOLOCK was being held while we waited for readahead to complete. To fix this, you're planning on waiting for readahead to complete with the invalidate lock held? I don't see the benefit.
I see the invalidate_lock as being roughly equivalent to the IOLOCK, just pulled up to the VFS. Is that incorrect?
Quoted 37 lines · Matthew Wilcox
Quoted 26 lines · Amir Goldstein
Quoted 25 lines · Jan Kara
Quoted 22 lines · Amir Goldstein
Quoted 4 lines
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
I imagine that the atomic r/w synchronisation will remain *exactly* as it is today by taking
XFS_IOLOCK_SHAREDaroundgeneric_file_read_iter(), when reading data into user buffer, but before that, I would like to issue and wait for read of the pages in the range to reduce the probability of doing the read I/O underXFS_IOLOCK_SHARED.The pre-warm of page cache does not need to abide to the atomic read semantics and it is also tolerable if some pages are evicted in between pre-warn and read to user buffer - in the worst case this will result in I/O amplification, but for the common case, it will be a big win for the mixed random r/w performance on xfs.
To reduce risk of page cache thrashing we can limit this optimization to a maximum number of page cache pre-warm.
The questions are:
1. Does this plan sound reasonable?Quoted 6 lines
Quoted 3 lines
Ah, I see now. So essentially the idea is to pull the readahead (which is currently happening from
filemap_read()->filemap_get_pages()) out from under thei_rwsem. It looks like a fine idea to me.Great!
Anyone doesn't like the idea or has another suggestion?I guess I'm still confused.
Quoted 8 lines
The problem was the the XFS IOLOCK was being held while we waited for readahead to complete. To fix this, you're planning on waiting for readahead to complete with the invalidate lock held? I don't see the benefit.
I see the
invalidate_lockas being roughly equivalent to the IOLOCK, just pulled up to the VFS. Is that incorrect?
This question coming from you really shakes my confidence.
This entire story started from the observation that xfs performance of concurrent mixed rw workload is two orders of magnitude worse than ext4 on slow disk.
The reason for the difference was that xfs was taking the IOLOCK shared on reads and ext4 did not.
That had two very different reasons:
1. POSIX atomic read/write semantics unique to xfs
2. Correctness w.r.t. races with punch hole etc, which lead to the conclusion that all non-xfs filesystems are buggy in that respect
The solution of pulling IOLOCK to vfs would have solved the bug but at the cost of severely regressing the mix rw workload on all fs.
The point of Jan's work on invalidate_lock was to fix the bug and avoid the regression. I hope that worked out, but I did not test the mixed rw workload on ext4 after invalidate_lock.
IIUC, ideally, invalidate_lock was supposed to be taken only for adding pages to page cache and locking them, but not during IO in order to synchronize against truncating pages (punch hole). But from this comment in filemap_create_folio() I just learned that that is not exactly the case:
"...Note that we could release invalidate_lock after inserting the
folio into the page cache as the locked folio would then be enough to synchronize with hole punching. But..."
Even so, because invalidate_lock is not taken by writers and reader that work on existing pages and because invalidate_lock is not held for the entire read/write operation, statistically it should be less contended than IOLOCK for some workloads, but I am afraid that for the workload I tested (bs=8K and mostly cold page cache) it will be contended with current vfs code.
I am going to go find a machine with slow disk to test the random rw workload again on both xfs and ext4 pre and post invalidate_lock and to try out the pre-warm page cache solution.
The results could be:
a) ext4 random rw performance has been degraded by invalidate_lock
b) pre-warm page cache before taking IOLOCK is going to improve xfs random rw performance
c) A little bit of both
It would be great if you and Jan could agree on the facts and confirm that my observations about invalidate_lock are correct until I get the test results.
Thanks,
Amir.
Jan Kara
reviewerQuoted 61 lines · Amir Goldstein
Quoted 37 lines · Matthew Wilcox
Quoted 26 lines · Amir Goldstein
Quoted 25 lines · Jan Kara
Quoted 22 lines · Amir Goldstein
Quoted 4 lines
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
I imagine that the atomic r/w synchronisation will remain *exactly* as it is today by taking
XFS_IOLOCK_SHAREDaroundgeneric_file_read_iter(), when reading data into user buffer, but before that, I would like to issue and wait for read of the pages in the range to reduce the probability of doing the read I/O underXFS_IOLOCK_SHARED.The pre-warm of page cache does not need to abide to the atomic read semantics and it is also tolerable if some pages are evicted in between pre-warn and read to user buffer - in the worst case this will result in I/O amplification, but for the common case, it will be a big win for the mixed random r/w performance on xfs.
To reduce risk of page cache thrashing we can limit this optimization to a maximum number of page cache pre-warm.
The questions are:
1. Does this plan sound reasonable?Quoted 6 lines
Quoted 3 lines
Ah, I see now. So essentially the idea is to pull the readahead (which is currently happening from
filemap_read()->filemap_get_pages()) out from under thei_rwsem. It looks like a fine idea to me.Great!
Anyone doesn't like the idea or has another suggestion?I guess I'm still confused.
Quoted 8 lines
The problem was the the XFS IOLOCK was being held while we waited for readahead to complete. To fix this, you're planning on waiting for readahead to complete with the invalidate lock held? I don't see the benefit.
I see the
invalidate_lockas being roughly equivalent to the IOLOCK, just pulled up to the VFS. Is that incorrect?This question coming from you really shakes my confidence.
This entire story started from the observation that xfs performance of concurrent mixed rw workload is two orders of magnitude worse than ext4 on slow disk.
The reason for the difference was that xfs was taking the IOLOCK shared on reads and ext4 did not.
That had two very different reasons:
1. POSIX atomic read/write semantics unique to xfs
2. Correctness w.r.t. races with punch hole etc, which lead to the conclusion that all non-xfs filesystems are buggy in that respect
Quoted 6 lines
The solution of pulling IOLOCK to vfs would have solved the bug but at the cost of severely regressing the mix rw workload on all fs.
The point of Jan's work on
invalidate_lockwas to fix the bug and avoid the regression. I hope that worked out, but I did not test the mixed rw workload on ext4 afterinvalidate_lock.
Yes, it did work out :)
Quoted 9 lines
IIUC, ideally,
invalidate_lockwas supposed to be taken only for adding pages to page cache and locking them, but not during IO in order to synchronize against truncating pages (punch hole). But from this comment infilemap_create_folio()I just learned that that is not exactly the case:
"...Note that we could releaseinvalidate_lockafter inserting the
folio into the page cache as the locked folio would then be enough to synchronize with hole punching. But..."
Quoted 6 lines
Even so, because
invalidate_lockis not taken by writers and reader that work on existing pages and becauseinvalidate_lockis not held for the entire read/write operation, statistically it should be less contended than IOLOCK for some workloads, but I am afraid that for the workload I tested (bs=8K and mostly cold page cache) it will be contended with current vfs code.
Well, the rules are: When you are adding pages to the page cache you need to either hold i_rwsem or invalidate_lock at least in shared mode. Places removing underlying storage from the page cache pages are responsible for holding both i_rwsem and invalidate_lock in exclusive mode. Writes generally hold i_rwsem exclusive (or shared for overwriting direct IO), reads hold invalidate_lock shared. So there will not be contention on invalidate_lock as such for a mixed rw workload, except for the internal contention of the shared invalidate_lock holders on the cacheline holding the invalidate_lock (which is non-negligible as well in heavily parallel loads but that's a separate story).
Quoted 9 lines
I am going to go find a machine with slow disk to test the random rw workload again on both xfs and ext4 pre and post
invalidate_lockand to try out the pre-warm page cache solution.The results could be:
a) ext4 random rw performance has been degraded byinvalidate_lock
b) pre-warm page cache before taking IOLOCK is going to improve xfs random rw performance
c) A little bit of both
Well, numbers always beat the theory so I'm all for measuring it but let me say our kernel performance testing within SUSE didn't show significant hit being introduced by invalidate_lock for any major filesystem.
I hope this clears out things a bit.
Honza
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Quoted 9 lines
Quoted 9 lines
I am going to go find a machine with slow disk to test the random rw workload again on both xfs and ext4 pre and post
invalidate_lockand to try out the pre-warm page cache solution.The results could be:
a) ext4 random rw performance has been degraded byinvalidate_lock
b) pre-warm page cache before taking IOLOCK is going to improve xfs random rw performance
c) A little bit of both
The correct answer is b. :)
Quoted 5 lines
Well, numbers always beat the theory so I'm all for measuring it but let me say our kernel performance testing within SUSE didn't show significant hit being introduced by
invalidate_lockfor any major filesystem.
Here are the numbers produced on v5.10.109, on v5.19-rc3 and on v5.19-rc3+ which includes the pre-warn test patch [1].
The numbers are produced by a filebench workload [2] that runs 8 random reader threads and 8 random writer threads for 60 seconds on a cold cache preallocated 5GB file.
Note that the machine I tested with has much faster storage than the one that was used 3 years ago, but the performance impact of IOLOCK is still very clear, even larger in this test.
If there are no other objections to the pre-warm concept, I will go on to write and test a proper patch.
Thanks,
Amir.
[1] https://github.com/amir73il/linux/commit/70e94f3471739c442b1110ee46e8b59e5d5f5042
[2] https://github.com/amir73il/filebench/blob/overlayfs-devel/workloads/randomrw.f
--- EXT4 5.10 ---
filebench randomrw (8 read threads, 8 write threads) kernel 5.10.109, ext4
Test #1:
rand-write1 3002127ops 50020ops/s 390.8mb/s 0.156ms/op
[0.002ms - 213.755ms]
rand-read1 31749234ops 528988ops/s 4132.7mb/s
0.010ms/op [0.001ms - 68.884ms]
Test #2:
rand-write1 3083679ops 51381ops/s 401.4mb/s 0.152ms/op
[0.002ms - 181.368ms]
rand-read1 32182118ops 536228ops/s 4189.3mb/s
0.010ms/op [0.001ms - 61.158ms]
--- EXT4 5.19 ---
filebench randomrw (8 read threads, 8 write threads) kernel 5.19-rc3, ext4
Test #1:
rand-write1 2829917ops 47159ops/s 368.4mb/s 0.160ms/op
[0.002ms - 4709.167ms]
rand-read1 36997540ops 616542ops/s 4816.7mb/s
0.009ms/op [0.001ms - 4704.105ms]
Test #2:
rand-write1 2764486ops 46067ops/s 359.9mb/s 0.170ms/op
[0.002ms - 5042.597ms]
rand-read1 38893279ops 648118ops/s 5063.4mb/s
0.008ms/op [0.001ms - 5004.069ms]
--- XFS 5.10 ---
filebench randomrw (8 read threads, 8 write threads) kernel 5.10.109, xfs
Test #1:
rand-write1 1049278ops 17485ops/s 136.6mb/s 0.456ms/op
[0.002ms - 224.062ms]
rand-read1 33325ops 555ops/s 4.3mb/s 14.392ms/op
[0.007ms - 224.833ms]
Test #2:
rand-write1 1127497ops 18788ops/s 146.8mb/s 0.424ms/op
[0.003ms - 445.810ms]
rand-read1 35341ops 589ops/s 4.6mb/s 13.566ms/op
[0.005ms - 445.529ms]
--- XFS 5.19 ---
filebench randomrw (8 read threads, 8 write threads) kernel 5.19-rc3, xfs
Test #1:
rand-write1 3295934ops 54920ops/s 429.1mb/s 0.144ms/op
[0.003ms - 109.703ms]
rand-read1 86768ops 1446ops/s 11.3mb/s 5.520ms/op
[0.003ms - 372.000ms]
Test #2:
rand-write1 3246935ops 54106ops/s 422.7mb/s 0.146ms/op
[0.002ms - 103.505ms]
rand-read1 167018ops 2783ops/s 21.7mb/s 2.867ms/op
[0.003ms - 101.105ms]
--- XFS+ 5.19 ---
filebench randomrw (8 read threads, 8 write threads) kernel 5.19-rc3+ (xfs page cache warmup patch)
Test #1:
rand-write1 3054567ops 50899ops/s 397.6mb/s 0.154ms/op
[0.002ms - 201.531ms]
rand-read1 38107333ops 634990ops/s 4960.9mb/s
0.008ms/op [0.001ms - 60.027ms]
Test #2:
rand-write1 2704416ops 45053ops/s 352.0mb/s 0.174ms/op
[0.002ms - 287.079ms]
rand-read1 38589737ops 642874ops/s 5022.4mb/s
0.008ms/op [0.001ms - 60.741ms]
Quoted 24 lines · Amir Goldstein
Quoted 22 lines · Jan Kara
Quoted 20 lines · Amir Goldstein
Quoted 4 lines
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
I imagine that the atomic r/w synchronisation will remain *exactly* as it is today by taking
XFS_IOLOCK_SHAREDaroundgeneric_file_read_iter(), when reading data into user buffer, but before that, I would like to issue and wait for read of the pages in the range to reduce the probability of doing the read I/O underXFS_IOLOCK_SHARED.The pre-warm of page cache does not need to abide to the atomic read semantics and it is also tolerable if some pages are evicted in between pre-warn and read to user buffer - in the worst case this will result in I/O amplification, but for the common case, it will be a big win for the mixed random r/w performance on xfs.
To reduce risk of page cache thrashing we can limit this optimization to a maximum number of page cache pre-warm.
Quoted 7 lines
Quoted 6 lines
Quoted 2 lines
The questions are:
1. Does this plan sound reasonable?Ah, I see now. So essentially the idea is to pull the readahead (which is currently happening from
filemap_read()->filemap_get_pages()) out from under thei_rwsem. It looks like a fine idea to me.
Although I was able to demonstrate performance improvement with page cache pre-warming on low latency disks, when testing on a common standard system [*], page cache pre-warming did not yield any improvement to the mixed rw workload.
[*] I ran the following fio workload on e2-standard-8 GCE machine:
[global]
filename=/mnt/xfs/testfile.fio
norandommap
randrepeat=0
size=5G
bs=8K
ioengine=psync
numjobs=8
group_reporting=1
direct=0
fallocate=1
end_fsync=0
runtime=60
[xfs-read]
readwrite=randread
[xfs-write]
readwrite=randwrite
The difference between ext4 and xfs with this machine/workload was two orders of magnitude:
root@xfstests:~# fio ./ext4.fio ... Run status group 0 (all jobs): READ: bw=826MiB/s (866MB/s), 826MiB/s-826MiB/s (866MB/s-866MB/s), io=40.0GiB (42.9GB), run=49585-49585msec WRITE: bw=309MiB/s (324MB/s), 309MiB/s-309MiB/s (324MB/s-324MB/s), io=18.1GiB (19.5GB), run=60003-60003msec
root@xfstests:~# fio ./xfs.fio ... Run status group 0 (all jobs): READ: bw=7053KiB/s (7223kB/s), 7053KiB/s-7053KiB/s (7223kB/s-7223kB/s), io=413MiB (433MB), run=60007-60007msec WRITE: bw=155MiB/s (163MB/s), 155MiB/s-155MiB/s (163MB/s-163MB/s), io=9324MiB (9777MB), run=60006-60006msec
I verified that without XFS_IOLOCK_SHARED xfs fio results are on par with ext4 results for this workload.
Quoted 9 lines
Quoted 3 lines
just cannot comment on whether calling this without
i_rwsemdoes not break some internal XFS expectations for stuff like reflink etc.relink is done under
xfs_ilock2_io_mmap=>filemap_invalidate_lock_twoso it should not be a problem.pNFS leases I need to look into.
I wonder if xfs_fs_map_blocks() and xfs_fs_commit_blocks() should not be taking the invalidate lock before calling invalidate_inode_pages2() like the xfs callers of truncate_pagecache_range() do?
If we do that, then I don't see a problem with buffered read without XFS_IOLOCK_SHARED w.r.t. correctness of layout leases.
Dave, Christoph,
I know that you said that changing the atomic buffered read semantics is out of the question and that you also objected to a mount option (which nobody will know how to use) and I accept that.
Given that a performant range locks implementation is not something trivial to accomplish (Dave please correct me if I am wrong), and given the massive performance impact of XFS_IOLOCK_SHARED on this workload,
what do you think about POSIX_FADV_TORN_RW that a specific application can use to opt-out of atomic buffer read semantics?
The specific application that I want to modify to use this hint is Samba. Samba uses IO threads by default to issue pread/pwrite on the server for IO requested by the SMB client. The IO size is normally larger than xfs block size and the range may not be block aligned.
The SMB protocol has explicit byte range locks and the server implements them, so it is pretty safe to assume that a client that did not request range locks does not need xfs to do the implicit range locking for it.
For this reason and because of the huge performance win, I would like to implement POSIX_FADV_TORN_RW in xfs and have Samba try to set this hint when supported.
It is very much possible that NFSv4 servers (user and kennel) would also want to set this hint for very similar reasons.
Thoughts?
Thanks,
Amir.
Quoted 122 lines · Amir Goldstein
Quoted 24 lines · Amir Goldstein
Quoted 22 lines · Jan Kara
Quoted 20 lines · Amir Goldstein
Quoted 4 lines
How exactly do you imagine the synchronization of buffered read against buffered write would work? Lock all pages for the read range in the page cache? You'd need to be careful to not bring the machine OOM when someone asks to read a huge range...
I imagine that the atomic r/w synchronisation will remain *exactly* as it is today by taking
XFS_IOLOCK_SHAREDaroundgeneric_file_read_iter(), when reading data into user buffer, but before that, I would like to issue and wait for read of the pages in the range to reduce the probability of doing the read I/O underXFS_IOLOCK_SHARED.The pre-warm of page cache does not need to abide to the atomic read semantics and it is also tolerable if some pages are evicted in between pre-warn and read to user buffer - in the worst case this will result in I/O amplification, but for the common case, it will be a big win for the mixed random r/w performance on xfs.
To reduce risk of page cache thrashing we can limit this optimization to a maximum number of page cache pre-warm.
Quoted 7 lines
Quoted 6 lines
Quoted 2 lines
The questions are:
1. Does this plan sound reasonable?Ah, I see now. So essentially the idea is to pull the readahead (which is currently happening from
filemap_read()->filemap_get_pages()) out from under thei_rwsem. It looks like a fine idea to me.Although I was able to demonstrate performance improvement with page cache pre-warming on low latency disks, when testing on a common standard system [*], page cache pre-warming did not yield any improvement to the mixed rw workload.
[*] I ran the following fio workload on e2-standard-8 GCE machine:
[global]
filename=/mnt/xfs/testfile.fio
norandommap
randrepeat=0
size=5G
bs=8K
ioengine=psync
numjobs=8
group_reporting=1
direct=0
fallocate=1
end_fsync=0
runtime=60[xfs-read]
readwrite=randread[xfs-write]
readwrite=randwriteThe difference between ext4 and xfs with this machine/workload was two orders of magnitude:
root@xfstests:~# fio ./ext4.fio ... Run status group 0 (all jobs): READ: bw=826MiB/s (866MB/s), 826MiB/s-826MiB/s (866MB/s-866MB/s), io=40.0GiB (42.9GB), run=49585-49585msec WRITE: bw=309MiB/s (324MB/s), 309MiB/s-309MiB/s (324MB/s-324MB/s), io=18.1GiB (19.5GB), run=60003-60003msecroot@xfstests:~# fio ./xfs.fio ... Run status group 0 (all jobs): READ: bw=7053KiB/s (7223kB/s), 7053KiB/s-7053KiB/s (7223kB/s-7223kB/s), io=413MiB (433MB), run=60007-60007msec WRITE: bw=155MiB/s (163MB/s), 155MiB/s-155MiB/s (163MB/s-163MB/s), io=9324MiB (9777MB), run=60006-60006msecI verified that without
XFS_IOLOCK_SHAREDxfs fio results are on par with ext4 results for this workload.Quoted 9 lines
Quoted 3 lines
just cannot comment on whether calling this without
i_rwsemdoes not break some internal XFS expectations for stuff like reflink etc.relink is done under
xfs_ilock2_io_mmap=>filemap_invalidate_lock_twoso it should not be a problem.pNFS leases I need to look into.
I wonder if
xfs_fs_map_blocks()andxfs_fs_commit_blocks()should not be taking the invalidate lock before callinginvalidate_inode_pages2()like the xfs callers oftruncate_pagecache_range()do?If we do that, then I don't see a problem with buffered read without
XFS_IOLOCK_SHAREDw.r.t. correctness of layout leases.Dave, Christoph,
I know that you said that changing the atomic buffered read semantics is out of the question and that you also objected to a mount option (which nobody will know how to use) and I accept that.
Given that a performant range locks implementation is not something trivial to accomplish (Dave please correct me if I am wrong), and given the massive performance impact of
XFS_IOLOCK_SHAREDon this workload,
what do you think aboutPOSIX_FADV_TORN_RWthat a specific application can use to opt-out of atomic buffer read semantics?The specific application that I want to modify to use this hint is Samba. Samba uses IO threads by default to issue pread/pwrite on the server for IO requested by the SMB client. The IO size is normally larger than xfs block size and the range may not be block aligned.
The SMB protocol has explicit byte range locks and the server implements them, so it is pretty safe to assume that a client that did not request range locks does not need xfs to do the implicit range locking for it.
Quoted 8 lines
For this reason and because of the huge performance win, I would like to implement
POSIX_FADV_TORN_RWin xfs and have Samba try to set this hint when supported.It is very much possible that NFSv4 servers (user and kennel) would also want to set this hint for very similar reasons.
Thoughts?
How about range locks for i_rwsem and invalidate_lock? That could reduce contention on VM farms, though I can only assume that, given that I don't have a reference implementation to play with...
--D
Quoted 2 lines
Thanks,
Amir.
Quoted 26 lines
Quoted 26 lines
Dave, Christoph,
I know that you said that changing the atomic buffered read semantics is out of the question and that you also objected to a mount option (which nobody will know how to use) and I accept that.
Given that a performant range locks implementation is not something trivial to accomplish (Dave please correct me if I am wrong), and given the massive performance impact of
XFS_IOLOCK_SHAREDon this workload,
what do you think aboutPOSIX_FADV_TORN_RWthat a specific application can use to opt-out of atomic buffer read semantics?The specific application that I want to modify to use this hint is Samba. Samba uses IO threads by default to issue pread/pwrite on the server for IO requested by the SMB client. The IO size is normally larger than xfs block size and the range may not be block aligned.
The SMB protocol has explicit byte range locks and the server implements them, so it is pretty safe to assume that a client that did not request range locks does not need xfs to do the implicit range locking for it.
For this reason and because of the huge performance win, I would like to implement
POSIX_FADV_TORN_RWin xfs and have Samba try to set this hint when supported.
Quoted 9 lines
Quoted 4 lines
It is very much possible that NFSv4 servers (user and kennel) would also want to set this hint for very similar reasons.
Thoughts?
How about range locks for
i_rwsemandinvalidate_lock? That could reduce contention on VM farms, though I can only assume that, given that I don't have a reference implementation to play with...
If you are asking if I have the bandwidth to work on range lock then the answer is that I do not.
IIRC, Dave had a WIP and ran some benchmarks with range locks, but I do not know at which state that work is.
The question is, if application developers know (or believe) that their application does not care about torn reads, are we insisting not to allow them to opt out of atomic buffered reads (which they do not need) because noone has the time to work on range locks?
If that is the final decision then if customers come to me to complain about this workload, my response will be:
If this workload is important for your application, either
- contribute developer resource to work on range locks
- carry a patch in your kernel
or
- switch to another filesystem for this workload
Thanks,
Amir.
Quoted 36 lines · Amir Goldstein
Quoted 26 lines
Quoted 26 lines
Dave, Christoph,
I know that you said that changing the atomic buffered read semantics is out of the question and that you also objected to a mount option (which nobody will know how to use) and I accept that.
Given that a performant range locks implementation is not something trivial to accomplish (Dave please correct me if I am wrong), and given the massive performance impact of
XFS_IOLOCK_SHAREDon this workload,
what do you think aboutPOSIX_FADV_TORN_RWthat a specific application can use to opt-out of atomic buffer read semantics?The specific application that I want to modify to use this hint is Samba. Samba uses IO threads by default to issue pread/pwrite on the server for IO requested by the SMB client. The IO size is normally larger than xfs block size and the range may not be block aligned.
The SMB protocol has explicit byte range locks and the server implements them, so it is pretty safe to assume that a client that did not request range locks does not need xfs to do the implicit range locking for it.
For this reason and because of the huge performance win, I would like to implement
POSIX_FADV_TORN_RWin xfs and have Samba try to set this hint when supported.Quoted 9 lines
Quoted 4 lines
It is very much possible that NFSv4 servers (user and kennel) would also want to set this hint for very similar reasons.
Thoughts?
How about range locks for
i_rwsemandinvalidate_lock? That could reduce contention on VM farms, though I can only assume that, given that I don't have a reference implementation to play with...
Quoted 5 lines
If you are asking if I have the bandwidth to work on range lock then the answer is that I do not.
IIRC, Dave had a WIP and ran some benchmarks with range locks, but I do not know at which state that work is.
Yeah, that's what I was getting at -- I really wish Dave would post that as an RFC. The last time I talked to him about it, he was worried that the extra complexity of the range lock structure would lead to more memory traffic and overhead.
I /know/ there are a lot of cloud vendors that would appreciate the speedup that range locking might provide. I'm also fairly sure there are also people who want maximum single threaded iops and will /not/ like range locks, but I think we ought to let kernel distributors choose which one they want.
Recently I've been playing around with static keys, because certain parts of xfs online fsck need to hook into libxfs. The hooks have some overhead, so I'd want to reduce the cost of that to making the instruction prefetcher skip over a nop sled when fsck isn't running. I sorta suspect this is a way out -- the distributor selects a default locking implementation at kbuild time, and we allow a kernel command line parameter to switch (if desired) during early boot. That only works if the compiler supports asm goto (iirc) but that's not /so/ uncommon.
I'll try to prod Dave about this later today, maybe we can find someone to work on it if he'd post the prototype.
--D
Quoted 17 lines
The question is, if application developers know (or believe) that their application does not care about torn reads, are we insisting not to allow them to opt out of atomic buffered reads (which they do not need) because noone has the time to work on range locks?
If that is the final decision then if customers come to me to complain about this workload, my response will be:
If this workload is important for your application, either
- contribute developer resource to work on range locks
- carry a patch in your kernel
or
- switch to another filesystem for this workloadThanks,
Amir.
Quoted 40 lines · Darrick J. Wong
Quoted 39 lines · Amir Goldstein
Quoted 26 lines
Quoted 26 lines
Dave, Christoph,
I know that you said that changing the atomic buffered read semantics is out of the question and that you also objected to a mount option (which nobody will know how to use) and I accept that.
Given that a performant range locks implementation is not something trivial to accomplish (Dave please correct me if I am wrong), and given the massive performance impact of
XFS_IOLOCK_SHAREDon this workload,
what do you think aboutPOSIX_FADV_TORN_RWthat a specific application can use to opt-out of atomic buffer read semantics?The specific application that I want to modify to use this hint is Samba. Samba uses IO threads by default to issue pread/pwrite on the server for IO requested by the SMB client. The IO size is normally larger than xfs block size and the range may not be block aligned.
The SMB protocol has explicit byte range locks and the server implements them, so it is pretty safe to assume that a client that did not request range locks does not need xfs to do the implicit range locking for it.
For this reason and because of the huge performance win, I would like to implement
POSIX_FADV_TORN_RWin xfs and have Samba try to set this hint when supported.Quoted 9 lines
Quoted 4 lines
It is very much possible that NFSv4 servers (user and kennel) would also want to set this hint for very similar reasons.
Thoughts?
How about range locks for
i_rwsemandinvalidate_lock? That could reduce contention on VM farms, though I can only assume that, given that I don't have a reference implementation to play with...If you are asking if I have the bandwidth to work on range lock then the answer is that I do not.
Quoted 7 lines
Quoted 2 lines
IIRC, Dave had a WIP and ran some benchmarks with range locks, but I do not know at which state that work is.
Yeah, that's what I was getting at -- I really wish Dave would post that as an RFC. The last time I talked to him about it, he was worried that the extra complexity of the range lock structure would lead to more memory traffic and overhead.
The reason I haven't posted it is that I don't think range locks can ever be made to perform and scale as we need for the IO path.
The problem with range locks is that the structure that tracks the locked ranges needs locking itself, and that means taking an IO lock is no longer just a single atomic operation - it's at least two atomic ops (lock, unlock on a spin lock) and then a bunch of cache misses while searching up the structure containing the range locks looking for overlaps.
Hence a { range_lock(); range_unlock(); } pair is at minimum twice as costly { down_read(); up_read(); } and that shows up dramatically with direct IO. My test system (2s, 32p) handles about 3 million atomic ops for a single cacheline across many CPUs before it breaks down into cacheline contention and goes really slow.
That means I can run 1.6 million read/write DIO iops to a single file on the test machine with shared rwsem locking (~3.2 million atomic ops a second) but with range locks (assuming just atomic op overhead) that drops to ~800k r/w DIO ops. The hardware IO capacity is just over 1.7MIOPS...
In reality, this contended cacheline is not the limiting factor for range locks - the limiting factor is the time it takes to run the critical section inside that lock. This was found with the mmap_sem when it was converted to range locks - the cache misses doing rb-tree pointer chasing with the spinlock held meant that the actual range lock rate topped out at about 180k lock,unlock pairs per second. i.e. an order of magnitude slower than a rwsem on this machine.
Read this thread again:
https://lore.kernel.org/linux-xfs/[email protected]/
That's really the elephant in the range locking room: a range lock with a locked search and update aglorithm can't scale beyond a single CPU in it's critical section. It's a hard limit no matter how many CPUs you have and how much concurrency the workload has - the range lock has a single threaded critical section that results in a hard computational limit on range lock operations.
Hence I was looking at using a novel OLC btree algorithm for storing the range locks. The RCU-based OLC btree is largely lockless, allowing conconcurrent search, insert and delete operations on range based index. I've largely got the OLC btree to work, but that simply exposed a further problem that range locks need to handle.
That is, the biggest problem for scaling range lock performance is that locking is mostly singleton operation - very few workloads actually use concurrent access to a single file and hence need multiple range locks held at once. As a result, the typical concurrent IO range lock workload results in a single node btree, and so all lookups, inserts and remove hammer the seqlocks on a single node and we end up contending on a single cache line again. Comared to a rwsem, we consume a lot more CPU overhead before we detect a change has occurred and we need to go around and try again.
That said, it's better than previous range lock implementations in that it gets up to about 400-450k mixed DIO and buffered iops, but it is still way, way down on using a shared rwsem or serialising at a page granularity via page locks.
Yes, I know there are many advantages to range locking. Because we can exclude specific ranges, operations like truncate, fallocate, buffered writes, etc can all run concurrently with reads. DIO can run perfectly coherently with buffered IO (mmap is still a problem!). We can extend files without having to serialise against other IO within the existing EOF. We can pass lock contexts with AIO so that the inode can be unlocked at completion and we can get rid of the nasty inode_dio_wait() stuff we have for synchronisation with DIO. And so on.
IOWs, while there are many upsides to range locking the reality is that single file IO performance will not scale to storage hardware capability any more. I have few thoughts on how range locking could be further optimised to avoid such overheads in the cases where range locking is not necessary, but I really don't think that the scalability of a range lock will ever be sufficient to allow us to track every IO we have in flight.
Quoted 5 lines
I /know/ there are a lot of cloud vendors that would appreciate the speedup that range locking might provide. I'm also fairly sure there are also people who want maximum single threaded iops and will /not/ like range locks, but I think we ought to let kernel distributors choose which one they want.
Speedup for what operations? Not single file DIO, and only for mixed buffered read/write. Perhaps for mixed fallocate/DIO workloads might benefit, but I reluctantly came to the conclusion that there really aren't many workloads that even a highly optimised rangelock would actually end up improving performance for...
Quoted 9 lines
Recently I've been playing around with static keys, because certain parts of xfs online fsck need to hook into libxfs. The hooks have some overhead, so I'd want to reduce the cost of that to making the instruction prefetcher skip over a nop sled when fsck isn't running. I sorta suspect this is a way out -- the distributor selects a default locking implementation at kbuild time, and we allow a kernel command line parameter to switch (if desired) during early boot. That only works if the compiler supports asm goto (iirc) but that's not /so/ uncommon.
I think it's a lot harder than that. The range lock requires a signification on-stack structure to be declared in the context that the lock is being taken. i.e.:
+#define RANGE_LOCK_FULL (LLONG_MAX)
+
+struct range_lock {
+ uint64_t start; + uint64_t end; + struct list_head wait_list;
+#ifdef __KERNEL__
+ struct task_struct *task;
+#else
+ pthread_cond_t task;
+#endif
+};
+
+#define __RANGE_LOCK_INITIALIZER(__name, __start, __end) { \
+ .start = (__start) \
+ ,.end = (__end) \
+ ,.wait_list = LIST_HEAD_INIT((__name).wait_list) \+ }
+
+#define DEFINE_RANGE_LOCK(name, start, end) \
+ struct range_lock name = __RANGE_LOCK_INITIALIZER((name), (start), (end))
+
+#define DEFINE_RANGE_LOCK_FULL(name) \
+ struct range_lock name = __RANGE_LOCK_INITIALIZER((name), 0, RANGE_LOCK_FULL)
And code that uses it looks like:
+static inline void
+xfs_iolock_range_init(
+ struct xfs_inode *ip, + struct range_lock *rlock, + uint64_t start, + uint64_t count)
+{
+ int rounding;
+
+ rounding = max_t(int, i_blocksize(VFS_I(ip)), PAGE_SIZE);
+ range_lock_init(rlock, round_down(start, rounding),
+ round_up(start + count, rounding));
+}
+
STATIC ssize_t
xfs_file_dio_read(
struct kiocb *iocb,
struct iov_iter *to)
{
struct xfs_inode *ip = XFS_I(file_inode(iocb->ki_filp));
+ size_t count = iov_iter_count(to);
ssize_t ret;
+ struct range_lock rlock;trace_xfs_file_direct_read(iocb, to);
- if (!iov_iter_count(to))
+ if (!count)
return 0; /* skip atime */
file_accessed(iocb->ki_filp);
- ret = xfs_ilock_iocb(iocb, XFS_VFSLOCK_SHARED); + xfs_iolock_range_init(ip, &rlock, iocb->ki_pos, count); + ret = xfs_ilock_iocb(iocb, &rlock, XFS_VFSLOCK_SHARED);
Hence I don't think this is as simple as using static keys to switch code paths as there's a whole lot more information that needs to be set up for range locks compared to just using rwsems....
-Dave.
Signature · 2 lines
Dave Chinner [email protected]
Quoted 15 lines · Dave Chinner
Quoted 14 lines · Darrick J. Wong
Quoted 13 lines · Amir Goldstein
Quoted 13 lines
Quoted 13 lines
Dave, Christoph,
I know that you said that changing the atomic buffered read semantics is out of the question and that you also objected to a mount option (which nobody will know how to use) and I accept that.
Given that a performant range locks implementation is not something trivial to accomplish (Dave please correct me if I am wrong), and given the massive performance impact of
XFS_IOLOCK_SHAREDon this workload,
what do you think aboutPOSIX_FADV_TORN_RWthat a specific application can use to opt-out of atomic buffer read semantics?
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
The specific application that I want to modify to use this hint is Samba. Samba uses IO threads by default to issue pread/pwrite on the server for IO requested by the SMB client. The IO size is normally larger than xfs block size and the range may not be block aligned.
The SMB protocol has explicit byte range locks and the server implements them, so it is pretty safe to assume that a client that did not request range locks does not need xfs to do the implicit range locking for it.
That doesn't cover concurrent local (server side) access to the file. It's not uncommon to have the same filesystems exported by both Samba and NFS at the same time, and the only point of co-ordination between the two is the underlying local filesystem....
IOWs, when we are talking about local filesystem behaviour, what a network protocol does above the filesystem is largely irrelevant to the synchronisation required within the filesystem implementation....
Quoted 20 lines
Quoted 20 lines
Quoted 19 lines
Quoted 4 lines
Quoted 4 lines
For this reason and because of the huge performance win, I would like to implement
POSIX_FADV_TORN_RWin xfs and have Samba try to set this hint when supported.Quoted 9 lines
Quoted 4 lines
It is very much possible that NFSv4 servers (user and kennel) would also want to set this hint for very similar reasons.
Thoughts?
How about range locks for
i_rwsemandinvalidate_lock? That could reduce contention on VM farms, though I can only assume that, given that I don't have a reference implementation to play with...If you are asking if I have the bandwidth to work on range lock then the answer is that I do not.
IIRC, Dave had a WIP and ran some benchmarks with range locks, but I do not know at which state that work is.
Quoted 7 lines
Quoted 4 lines
Yeah, that's what I was getting at -- I really wish Dave would post that as an RFC. The last time I talked to him about it, he was worried that the extra complexity of the range lock structure would lead to more memory traffic and overhead.
The reason I haven't posted it is that I don't think range locks can ever be made to perform and scale as we need for the IO path.
[snip range lock scalability and perf issues]
As I just discussed on #xfs with Darrick, there are other options we can persue here.
The first question we need to ask ourselves is this: what are we protecting against with exclusive buffered write behaviour?
The answer is that we know there are custom enterprise database applications out there that assume that 8-16kB buffered writes are atomic. I wish I could say these are legacy applications these days, but they aren't - they are still in production use, and the applications build on those custom database engines are still under active development and use.
AFAIK, the 8kB atomic write behaviour is historical and came from applications originally designed for Solaris and hardware that had an 8kB page size. Hence buffered 8kB writes were assumed to be the largest atomic write size that concurrent reads would not see write tearing. These applications are now run on x86-64 boxes with 4kB page size, but they still assume that 8kB writes are atomic and can't tear.
So, really, these days the atomic write behaviour of XFS is catering for these small random read/write IO applications, not to provide atomic writes for bulk data moving applications writing 2GB of data per write() syscall. Hence we can fairly safely say that we really only need "exclusive" buffered write locking for relatively small multipage IOs, not huge IOs.
We can do single page shared buffered writes immediately - we guarantee that while the folio is locked, a buffered read cannot access the data until the folio is unlocked. So that could be the first step to relaxing the exclusive locking requirement for buffered writes.
Next we need to consider that we now have large folio support in the page cache, which means we can treat contiguous file ranges larger than a single page a single atomic unit if they are covered by a multi-page folio. As such, if we have a single multi-page folio that spans the entire write() range already in cache, we can run that write atomically under a shared IO lock the same as we can do with single page folios.
However, what happens if the folio is smaller than the range we need to write? Well, in that case, we have to abort the shared lock write and upgrade to an exclusive lock before trying again.
Of course, we can only determine if the write can go ahead once we have the folio locked. That means we need a new non-blocking write condition to be handled by the iomap code. We already have several of them because of IOCB_NOWAIT semantics that io_uring requires for buffered writes, so we are already well down the path of needing to support fully non-blocking writes through iomap.
Further, the recent concurrent write data corruption that we uncovered requires a new hook in the iomap write path to allow writes to be aborted for remapping because the cached iomap has become stale. This validity check can only be done once the folio has locked - if the cached iomap is stale once we have the page locked, then we have to back out and remap the write range and re-run the write.
IOWs, we are going to have to add write retries to the iomap write path for data integrity purposes. These checks must be done only after the folio has been locked, so we really end up getting the "can't do atomic write" retry infrastructure for free with the data corruption fixes...
With this in place, it becomes trivial to support atomic writes with shared locking all the way up to PMD sizes (or whatever the maximum multipage folio size the arch supports is) with a minimal amount of extra code.
At this point, we have a buffered write path that tries to do shared locking first, and only falls back to exclusive locking if the page cache doesn't contain a folio large enough to soak up the entire write.
In future, Darrick suggested we might be able to do a "trygetlock a bunch of folios" operation that locks a range of folios within the current iomap in one go, and then we write into all of them in a batch before unlocking them all. This would give us multi-folio atomic writes with shared locking - this is much more complex, and it's unclear that multi-folio write batching will gain us anything over the single folio check described above...
Finally, for anything that is concurrently reading and writing lots of data in chunks larger than PMD sizes, the application should really be using DIO with AIO or io_uring. So falling back to exclusive locking for such large single buffered write IOs doesn't seem like a huge issue right now....
Thoughts?
-Dave.
Signature · 2 lines
Dave Chinner [email protected]
Quoted 31 lines · Dave Chinner
Quoted 15 lines · Dave Chinner
Quoted 14 lines · Darrick J. Wong
Quoted 13 lines · Amir Goldstein
Quoted 13 lines
Quoted 13 lines
Dave, Christoph,
I know that you said that changing the atomic buffered read semantics is out of the question and that you also objected to a mount option (which nobody will know how to use) and I accept that.
Given that a performant range locks implementation is not something trivial to accomplish (Dave please correct me if I am wrong), and given the massive performance impact of
XFS_IOLOCK_SHAREDon this workload,
what do you think aboutPOSIX_FADV_TORN_RWthat a specific application can use to opt-out of atomic buffer read semantics?Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
The specific application that I want to modify to use this hint is Samba. Samba uses IO threads by default to issue pread/pwrite on the server for IO requested by the SMB client. The IO size is normally larger than xfs block size and the range may not be block aligned.
The SMB protocol has explicit byte range locks and the server implements them, so it is pretty safe to assume that a client that did not request range locks does not need xfs to do the implicit range locking for it.
That doesn't cover concurrent local (server side) access to the file. It's not uncommon to have the same filesystems exported by both Samba and NFS at the same time, and the only point of co-ordination between the two is the underlying local filesystem....
Quoted 5 lines
IOWs, when we are talking about local filesystem behaviour, what a network protocol does above the filesystem is largely irrelevant to the synchronisation required within the filesystem implementation....
Perhaps I did not explain my proposal well.
Maybe I should have named it POSIX_FADV_TORN_READ.
The fadvise() from nfs/smb server would affect only the behavior of buffered reads on the open fd of that server, it will not affect any buffered IO on that inode not associated with the application that opted-in to this behavior.
Also, the nfs/smb server buffered writes are still under the same exclusive IO lock as the rest of the callers.
Quoted 43 lines
Quoted 20 lines
Quoted 20 lines
Quoted 19 lines
Quoted 4 lines
Quoted 4 lines
For this reason and because of the huge performance win, I would like to implement
POSIX_FADV_TORN_RWin xfs and have Samba try to set this hint when supported.Quoted 9 lines
Quoted 4 lines
It is very much possible that NFSv4 servers (user and kennel) would also want to set this hint for very similar reasons.
Thoughts?
How about range locks for
i_rwsemandinvalidate_lock? That could reduce contention on VM farms, though I can only assume that, given that I don't have a reference implementation to play with...If you are asking if I have the bandwidth to work on range lock then the answer is that I do not.
IIRC, Dave had a WIP and ran some benchmarks with range locks, but I do not know at which state that work is.
Quoted 7 lines
Quoted 4 lines
Yeah, that's what I was getting at -- I really wish Dave would post that as an RFC. The last time I talked to him about it, he was worried that the extra complexity of the range lock structure would lead to more memory traffic and overhead.
The reason I haven't posted it is that I don't think range locks can ever be made to perform and scale as we need for the IO path.
[snip range lock scalability and perf issues]
As I just discussed on #xfs with Darrick, there are other options we can persue here.
The first question we need to ask ourselves is this: what are we protecting against with exclusive buffered write behaviour?
The answer is that we know there are custom enterprise database applications out there that assume that 8-16kB buffered writes are atomic. I wish I could say these are legacy applications these days, but they aren't - they are still in production use, and the applications build on those custom database engines are still under active development and use.
Quoted 8 lines
AFAIK, the 8kB atomic write behaviour is historical and came from applications originally designed for Solaris and hardware that had an 8kB page size. Hence buffered 8kB writes were assumed to be the largest atomic write size that concurrent reads would not see write tearing. These applications are now run on x86-64 boxes with 4kB page size, but they still assume that 8kB writes are atomic and can't tear.
Interesting. I did not know which applications needed that behavior. The customer benchmark that started the complaint uses 8k buffered IO size (as do the benchmarks that I posted in this thread), so far as I am concerned, fixing small buffered IO will solve the problem.
Quoted 64 lines
So, really, these days the atomic write behaviour of XFS is catering for these small random read/write IO applications, not to provide atomic writes for bulk data moving applications writing 2GB of data per
write()syscall. Hence we can fairly safely say that we really only need "exclusive" buffered write locking for relatively small multipage IOs, not huge IOs.We can do single page shared buffered writes immediately - we guarantee that while the folio is locked, a buffered read cannot access the data until the folio is unlocked. So that could be the first step to relaxing the exclusive locking requirement for buffered writes.
Next we need to consider that we now have large folio support in the page cache, which means we can treat contiguous file ranges larger than a single page a single atomic unit if they are covered by a multi-page folio. As such, if we have a single multi-page folio that spans the entire
write()range already in cache, we can run that write atomically under a shared IO lock the same as we can do with single page folios.However, what happens if the folio is smaller than the range we need to write? Well, in that case, we have to abort the shared lock write and upgrade to an exclusive lock before trying again.
Of course, we can only determine if the write can go ahead once we have the folio locked. That means we need a new non-blocking write condition to be handled by the iomap code. We already have several of them because of
IOCB_NOWAITsemantics thatio_uringrequires for buffered writes, so we are already well down the path of needing to support fully non-blocking writes through iomap.Further, the recent concurrent write data corruption that we uncovered requires a new hook in the iomap write path to allow writes to be aborted for remapping because the cached iomap has become stale. This validity check can only be done once the folio has locked - if the cached iomap is stale once we have the page locked, then we have to back out and remap the write range and re-run the write.
IOWs, we are going to have to add write retries to the iomap write path for data integrity purposes. These checks must be done only after the folio has been locked, so we really end up getting the "can't do atomic write" retry infrastructure for free with the data corruption fixes...
With this in place, it becomes trivial to support atomic writes with shared locking all the way up to PMD sizes (or whatever the maximum multipage folio size the arch supports is) with a minimal amount of extra code.
At this point, we have a buffered write path that tries to do shared locking first, and only falls back to exclusive locking if the page cache doesn't contain a folio large enough to soak up the entire write.
In future, Darrick suggested we might be able to do a "trygetlock a bunch of folios" operation that locks a range of folios within the current iomap in one go, and then we write into all of them in a batch before unlocking them all. This would give us multi-folio atomic writes with shared locking - this is much more complex, and it's unclear that multi-folio write batching will gain us anything over the single folio check described above...
Quoted 7 lines
Finally, for anything that is concurrently reading and writing lots of data in chunks larger than PMD sizes, the application should really be using DIO with AIO or
io_uring. So falling back to exclusive locking for such large single buffered write IOs doesn't seem like a huge issue right now....Thoughts?
That sounds like a great plan.
I especially liked the "get it for free" part ;) Is there already WIP for the data integrity issue fix?
If there is anything I can do to assist, run the benchmark or anything please let me know.
In the meanwhile, I will run the benchmark with XFS_IOLOCK_SHARED on the write() path.
Thanks!
Amir.
Quoted 1 line · Amir Goldstein
[...]
Quoted 22 lines
Quoted 13 lines
As I just discussed on #xfs with Darrick, there are other options we can persue here.
The first question we need to ask ourselves is this: what are we protecting against with exclusive buffered write behaviour?
The answer is that we know there are custom enterprise database applications out there that assume that 8-16kB buffered writes are atomic. I wish I could say these are legacy applications these days, but they aren't - they are still in production use, and the applications build on those custom database engines are still under active development and use.
Quoted 8 lines
AFAIK, the 8kB atomic write behaviour is historical and came from applications originally designed for Solaris and hardware that had an 8kB page size. Hence buffered 8kB writes were assumed to be the largest atomic write size that concurrent reads would not see write tearing. These applications are now run on x86-64 boxes with 4kB page size, but they still assume that 8kB writes are atomic and can't tear.
Quoted 30 lines
Interesting. I did not know which applications needed that behavior. The customer benchmark that started the complaint uses 8k buffered IO size (as do the benchmarks that I posted in this thread), so far as I am concerned, fixing small buffered IO will solve the problem.
Quoted 25 lines
So, really, these days the atomic write behaviour of XFS is catering for these small random read/write IO applications, not to provide atomic writes for bulk data moving applications writing 2GB of data per
write()syscall. Hence we can fairly safely say that we really only need "exclusive" buffered write locking for relatively small multipage IOs, not huge IOs.We can do single page shared buffered writes immediately - we guarantee that while the folio is locked, a buffered read cannot access the data until the folio is unlocked. So that could be the first step to relaxing the exclusive locking requirement for buffered writes.
Next we need to consider that we now have large folio support in the page cache, which means we can treat contiguous file ranges larger than a single page a single atomic unit if they are covered by a multi-page folio. As such, if we have a single multi-page folio that spans the entire
write()range already in cache, we can run that write atomically under a shared IO lock the same as we can do with single page folios.However, what happens if the folio is smaller than the range we need to write? Well, in that case, we have to abort the shared lock write and upgrade to an exclusive lock before trying again.
Please correct me if I am wrong, but with current upstream, the only way that multi page folios are created for 4K block / 4K page setup are during readahead.
We *could* allocate multi page folios on write to an allocated block range that maps inside a single extent, but there is no code for that today.
It seems that without this code, any write to a region of page cache not pre-populated using readahead, would get exclusive iolock for 8K buffered writes until that single page folio cache entry is evicted.
Am I reading the iomap code correctly?
Quoted 45 lines
Quoted 45 lines
Of course, we can only determine if the write can go ahead once we have the folio locked. That means we need a new non-blocking write condition to be handled by the iomap code. We already have several of them because of
IOCB_NOWAITsemantics thatio_uringrequires for buffered writes, so we are already well down the path of needing to support fully non-blocking writes through iomap.Further, the recent concurrent write data corruption that we uncovered requires a new hook in the iomap write path to allow writes to be aborted for remapping because the cached iomap has become stale. This validity check can only be done once the folio has locked - if the cached iomap is stale once we have the page locked, then we have to back out and remap the write range and re-run the write.
IOWs, we are going to have to add write retries to the iomap write path for data integrity purposes. These checks must be done only after the folio has been locked, so we really end up getting the "can't do atomic write" retry infrastructure for free with the data corruption fixes...
With this in place, it becomes trivial to support atomic writes with shared locking all the way up to PMD sizes (or whatever the maximum multipage folio size the arch supports is) with a minimal amount of extra code.
At this point, we have a buffered write path that tries to do shared locking first, and only falls back to exclusive locking if the page cache doesn't contain a folio large enough to soak up the entire write.
In future, Darrick suggested we might be able to do a "trygetlock a bunch of folios" operation that locks a range of folios within the current iomap in one go, and then we write into all of them in a batch before unlocking them all. This would give us multi-folio atomic writes with shared locking - this is much more complex, and it's unclear that multi-folio write batching will gain us anything over the single folio check described above...
Finally, for anything that is concurrently reading and writing lots of data in chunks larger than PMD sizes, the application should really be using DIO with AIO or
io_uring. So falling back to exclusive locking for such large single buffered write IOs doesn't seem like a huge issue right now....
Quoted 6 lines
Quoted 1 line
Thoughts?
That sounds like a great plan.
I especially liked the "get it for free" part ;) Is there already WIP for the data integrity issue fix?
OK. I see your patch set.
Quoted 6 lines
If there is anything I can do to assist, run the benchmark or anything please let me know.
In the meanwhile, I will run the benchmark with
XFS_IOLOCK_SHAREDon thewrite()path.
As expected, results without exclusive iolock look very good [*].
Thanks,
Amir.
[*] I ran the following fio workload on e2-standard-8 GCE machine:
[global]
filename=/mnt/xfs/testfile.fio
norandommap
randrepeat=0
size=5G
bs=8K
ioengine=psync
numjobs=8
group_reporting=1
direct=0
fallocate=1
end_fsync=0
runtime=60
[xfs-read]
readwrite=randread
[xfs-write]
readwrite=randwrite
========================= v6.0-rc4 (BAD) ========= Run #1:
READ: bw=7053KiB/s (7223kB/s)
WRITE: bw=155MiB/s (163MB/s)
Run #2:
READ: bw=4672KiB/s (4784kB/s)
WRITE: bw=355MiB/s (372MB/s)
Run #3:
READ: bw=5887KiB/s (6028kB/s)
WRITE: bw=137MiB/s (144MB/s)
========================= v6.0-rc4 (read no iolock like ext4 - GOOD) =========
READ: bw=742MiB/s (778MB/s)
WRITE: bw=345MiB/s (361MB/s)
========================= v6.0-rc4 (write shared iolock - BETTER) ========= Run #1:
READ: bw=762MiB/s (799MB/s)
WRITE: bw=926MiB/s (971MB/s)
Run #2:
READ: bw=170MiB/s (178MB/s)
WRITE: bw=982MiB/s (1029MB/s)
Run #3:
READ: bw=755MiB/s (792MB/s)
WRITE: bw=933MiB/s (978MB/s)
Jan Kara
reviewerQuoted 16 lines · Dave Chinner
The result of this is that, AFAICT, ext4 does not protect against
read()vs hole punch races - it's hole punching code it does:Hole Punch:
read():
inode_lock()inode_dio_wait(inode);down_write(i_mmap_sem)truncate_pagecache_range()ext4_file_iter_read() ext4_map_blocks() down_read(i_data_sem) <gets mapping> <populates page cache over hole> <reads stale data into cache> .....
Quoted 7 lines
down_write(i_data_sem)
remove extentsIOWs, ext4 is safe against truncate because of the change-inode-size-before-invalidation hacks, but the lack of serialise buffered reads means that hole punch and other similar fallocate based extent manipulations can race against reads....
Hum, you are right. Ext4 is buggy in this regard. I've fixed the race for page fault in ea3d7209ca01 "ext4: fix races between page faults and hole punching" but didn't realize the problem is there for buffered reads as well. I'll think how we can fix this. Thanks for noticing this!
Honza
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
On 08/04/19 14:03, Jan Kara wrote:
<>
Quoted 5 lines
Hum, you are right. Ext4 is buggy in this regard. I've fixed the race for page fault in
ea3d7209ca01"ext4: fix races between page faults and hole punching" but didn't realize the problem is there for buffered reads as well. I'll think how we can fix this. Thanks for noticing this!
This is very interesting. Please CC me if you have devised a test for this?
For a long time I want to enhance fsstress --verify that each writer always writes the long address it writes to as data (ie. file looks like an increasing long counter) And readers want to see only this pattern or zero.
Quoted 2 lines
Honza
Thanks
Boaz
Jan Kara
reviewerQuoted 4 lines · Dave Chinner
FYI, I'm working on a range lock implementation that should both solve the performance issue and the reader starvation issue at the same time by allowing concurrent buffered reads and writes to different file ranges.
Are you aware of range locks Davidlohr has implemented [1]? It didn't get merged because he had no in-tree user at the time (he was more aiming at converting mmap_sem which is rather difficult). But the generic lock implementation should be well usable.
Added Davidlohr to CC.
Honza
[1] https://lkml.org/lkml/2017/3/7/22
Signature · 2 lines
Jan Kara <[email protected]> SUSE Labs, CR
Quoted 11 lines · Jan Kara
Quoted 4 lines · Dave Chinner
FYI, I'm working on a range lock implementation that should both solve the performance issue and the reader starvation issue at the same time by allowing concurrent buffered reads and writes to different file ranges.
Are you aware of range locks Davidlohr has implemented [1]? It didn't get merged because he had no in-tree user at the time (he was more aiming at converting
mmap_semwhich is rather difficult). But the generic lock implementation should be well usable.
Quoted 5 lines
Added Davidlohr to CC.
Honza
fyi this was the latest version (had some naming updates per peterz).
Quoted 14 lines · Davidlohr Bueso
Quoted 13 lines · Jan Kara
Quoted 4 lines · Dave Chinner
FYI, I'm working on a range lock implementation that should both solve the performance issue and the reader starvation issue at the same time by allowing concurrent buffered reads and writes to different file ranges.
Are you aware of range locks Davidlohr has implemented [1]? It didn't get merged because he had no in-tree user at the time (he was more aiming at converting
mmap_semwhich is rather difficult). But the generic lock implementation should be well usable.Added Davidlohr to CC.
Quoted 7 lines
Quoted 3 lines
Honza
fyi this was the latest version (had some naming updates per peterz).
No, I wasn't aware of these because they haven't ever been posted to a list I subscribe to and they haven't been merged. I'll go have a look at them over the next few days.
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Quoted 19 lines · Dave Chinner
Quoted 18 lines · Davidlohr Bueso
Quoted 16 lines · Jan Kara
Quoted 4 lines · Dave Chinner
FYI, I'm working on a range lock implementation that should both solve the performance issue and the reader starvation issue at the same time by allowing concurrent buffered reads and writes to different file ranges.
Are you aware of range locks Davidlohr has implemented [1]? It didn't get merged because he had no in-tree user at the time (he was more aiming at converting
mmap_semwhich is rather difficult). But the generic lock implementation should be well usable.Added Davidlohr to CC.
Honza
Quoted 7 lines
Quoted 3 lines
fyi this was the latest version (had some naming updates per peterz).
No, I wasn't aware of these because they haven't ever been posted to a list I subscribe to and they haven't been merged. I'll go have a look at them over the next few days.
Rather disappointing, to tell the truth.
The implementation doesn't scale nearly as a rwsem for concurrent IO using shared access (i.e. XFS direct IO). That's the baseline we need to get close to for range locks to be a viable prospect.
Fio randrw numbers on a single file on a pmem device on a 16p machine using 4kB AIO-DIO iodepth 128 w/ fio on 5.1.0-rc3:
IOPS read/write (direct IO)
fio processes rwsem rangelock
1 78k / 78k 75k / 75k
2 131k / 131k 123k / 123k
4 267k / 267k 183k / 183k
8 372k / 372k 177k / 177k
16 315k / 315k 135k / 135k
So uncontended, non-overlapping range performance is not really even in the ballpark right now, unfortunately.
To indicate that range locking is actually working, let's do buffered read/write, which takes the rwsem exclusive for writes and so will be bound by that:
IOPS read/write (buffered IO)
fio processes rwsem rangelock
1 57k / 57k 64k / 64k
2 61k / 61k 111k / 111k
4 61k / 61k 228k / 228k
8 55k / 55k 195k / 195k
16 15k / 15k 40k / 40k
So the win for mixed buffered IO is huge but it's at the cost of direct IO performance. We can't make that tradeoff, so if this implementation cannot be substantially improved we really can'ti use it.
FUndamentally, the problem is that the interval tree work is all done under a spinlock, so by the time we get to 16p, 70% of the 16p that is being burnt is on the spinlock, and only 30% of the CPU time is actually doing IO work:
+ 70.78% 2.50% [kernel] [k] do_raw_spin_lock + 69.72% 0.07% [kernel] [k] _raw_spin_lock_irqsave - 68.27% 68.10% [kernel] [k] __pv_queued_spin_lock_slowpath ..... + 15.61% 0.07% [kernel] [k] range_write_unlock + 15.39% 0.06% [kernel] [k] range_read_unlock + 15.11% 0.12% [kernel] [k] range_read_lock + 15.11% 0.13% [kernel] [k] range_write_lock
.....
+ 12.92% 0.11% [kernel] [k] range_is_locked
FWIW, I'm not convinced about the scalability of the rb/interval tree, to tell you the truth. We got rid of the rbtree in XFS for cache indexing because the multi-level pointer chasing was just too expensive to do under a spinlock - it's just not a cache efficient structure for random index object storage.
FWIW, I have basic hack to replace the i_rwsem in XFS with a full range read or write lock with my XFS range lock implementation so it just behaves like a rwsem at this point. It is not in any way optimised at this point. Numbers for same AIO-DIO test are:
IOPS read/write (direct IO)
processes rwsem DB rangelock XFS rangelock
1 78k / 78k 75k / 75k 74k / 74k
2 131k / 131k 123k / 123k 134k / 134k
4 267k / 267k 183k / 183k 246k / 246k
8 372k / 372k 177k / 177k 306k / 306k
16 315k / 315k 135k / 135k 240k / 240k
Performance is much closer to rwsems, but the single lock critical section still causes serious amounts of cacheline contention on pure shared-read lock workloads like this.
FWIW, the XFS rangelock numbers match what I've been seeing with userspace perf tests using unoptimised pthread mutexes - ~600,000 lock/unlock cycles a second over ~100 concurrent non-overlapping ranges seem to be the limitation before futex contention burns down the house
AS it is, I've done all the work to make XFS use proper range locks on top of your patch - I'll go back tomorrow and modify the XFS range locks to use the same API as your lock implementation. I'm interested to see what falls out when they are used as real range locks rather than a glorified rwsem....
Cheers,
Dave.
PS: there's a double lock bug in range_is_locked().....
Signature · 2 lines
Dave Chinner [email protected]
Quoted 15 lines · Dave Chinner
Quoted 14 lines · Dave Chinner
Quoted 13 lines · Davidlohr Bueso
Quoted 12 lines · Jan Kara
Quoted 4 lines · Dave Chinner
FYI, I'm working on a range lock implementation that should both solve the performance issue and the reader starvation issue at the same time by allowing concurrent buffered reads and writes to different file ranges.
Are you aware of range locks Davidlohr has implemented [1]? It didn't get merged because he had no in-tree user at the time (he was more aiming at converting
mmap_semwhich is rather difficult). But the generic lock implementation should be well usable.Added Davidlohr to CC.
.....
Quoted 10 lines
Fio randrw numbers on a single file on a pmem device on a 16p machine using 4kB AIO-DIO iodepth 128 w/ fio on 5.1.0-rc3:
IOPS read/write (direct IO)
fio processes rwsem rangelock
1 78k / 78k 75k / 75k
2 131k / 131k 123k / 123k
4 267k / 267k 183k / 183k
8 372k / 372k 177k / 177k
16 315k / 315k 135k / 135k
....
Quoted 5 lines
FWIW, I'm not convinced about the scalability of the rb/interval tree, to tell you the truth. We got rid of the rbtree in XFS for cache indexing because the multi-level pointer chasing was just too expensive to do under a spinlock - it's just not a cache efficient structure for random index object storage.
Yeah, definitely not convinced an rbtree is the right structure here. Locking of the tree is the limitation....
Quoted 4 lines
FWIW, I have basic hack to replace the
i_rwsemin XFS with a full range read or write lock with my XFS range lock implementation so it just behaves like a rwsem at this point. It is not in any way optimised at this point. Numbers for same AIO-DIO test are:
Now the stuff I've been working on has the same interface as Davidlohr's patch, so I can swap and change them without thinking about it. It's still completely unoptimised, but:
IOPS read/write (direct IO)
processes rwsem DB rangelock XFS rangelock
1 78k / 78k 75k / 75k 72k / 72k
2 131k / 131k 123k / 123k 133k / 133k
4 267k / 267k 183k / 183k 237k / 237k
8 372k / 372k 177k / 177k 265k / 265k
16 315k / 315k 135k / 135k 228k / 228k
It's still substantially faster than the interval tree code.
BTW, if I take away the rwsem serialisation altogether, this test tops out at just under 500k/500k at 8 threads, and at 16 threads has started dropping off (~440k/440k). So the rwsem is a scalability limitation at just 8 threads....
/me goes off and thinks more about adding optimistic lock coupling to the XFS iext btree to get rid of the need for tree-wide locking altogether
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Quoted 13 lines · Dave Chinner
Now the stuff I've been working on has the same interface as Davidlohr's patch, so I can swap and change them without thinking about it. It's still completely unoptimised, but:
IOPS read/write (direct IO)
processes rwsem DB rangelock XFS
rangelock
1 78k / 78k 75k / 75k 72k / 72k
2 131k / 131k 123k / 123k 133k / 133k
4 267k / 267k 183k / 183k 237k / 237k
8 372k / 372k 177k / 177k 265k / 265k
16 315k / 315k 135k / 135k 228k / 228k
Quoted 1 line
It's still substantially faster than the interval tree code.
In general another big difference between both rangelock vs rwsems (when comparing them with full ranges) is that the latter will do writer optimistic spinning, so saving a context switch under the right scenarios provides mayor wins for rwsems -- I'm not sure if this applies to your fio tests, though. And pretty soon readers will also do this, hence rwsem will become a try-hard-not-to-sleep lock.
One of the reasons why I was hesitant with Btrees was the fact that insertion requires memory allocation, something I wanted to avoid... per your numbers, sacrificing tree depth was the wrong choice. Thanks for sharing these numbers.
Quoted 9 lines
BTW, if I take away the rwsem serialisation altogether, this test tops out at just under 500k/500k at 8 threads, and at 16 threads has started dropping off (~440k/440k). So the rwsem is a scalability limitation at just 8 threads....
/me goes off and thinks more about adding optimistic lock coupling to the XFS iext btree to get rid of the need for tree-wide locking altogether
I was not aware of this code.
Thanks,
Davidlohr
Quoted 14 lines · Davidlohr Bueso
Quoted 13 lines · Dave Chinner
Now the stuff I've been working on has the same interface as Davidlohr's patch, so I can swap and change them without thinking about it. It's still completely unoptimised, but:
IOPS read/write (direct IO)
processes rwsem DB rangelock XFS
rangelock
1 78k / 78k 75k / 75k 72k / 72k
2 131k / 131k 123k / 123k 133k / 133k
4 267k / 267k 183k / 183k 237k / 237k
8 372k / 372k 177k / 177k 265k / 265k
16 315k / 315k 135k / 135k 228k / 228k
Quoted 8 lines
Quoted 1 line
It's still substantially faster than the interval tree code.
In general another big difference between both rangelock vs rwsems (when comparing them with full ranges) is that the latter will do writer optimistic spinning, so saving a context switch under the right scenarios provides mayor wins for rwsems -- I'm not sure if this applies to your fio tests, though. And pretty soon readers will also do this, hence rwsem will become a try-hard-not-to-sleep lock.
Right, the optimistic spin-on-owner behaviour basically causes both rwsems and mutexes to behave as spin locks under these fio workloads until all CPUs are spinning at 100% CPU usage before they even start to behave like a sleeping lock. Essentailly, when I'm using the XFS range locks as "full range only" rwsem equivalent locks, they end up behaving like rwsems in terms of spinning on the tree mutex as the single read-lock range is shared between all concurrent readers and so has relatively low lookup and modification cost.
If I optmised the record update, it would be even closer to the rwsem performance, but full range read locks are going to be rare. Hell, even conflicting lock ranges are going to be rare in the IO path, because the application is doing something very wrong if it is doing overlapping concurrent IOs....
Quoted 2 lines
One of the reasons why I was hesitant with Btrees was the fact that insertion requires memory allocation, something I wanted to avoid...
Yup, that's an issue, and one of the reasons why I've used a mutex for the tree lock rather than a spin lock. But in the end, there's no difference to CPU usage on contention as they both act as spin locks in these cases...
If it becomes an issue, I'll add a mempool for the tree nodes and make sure that it is pre-populated sufficiently before starting a multi-level tree modification.
Quoted 2 lines
per your numbers, sacrificing tree depth was the wrong choice. Thanks for sharing these numbers.
Yeah, anything that pointer chases through cachelines and dirties multiple cachelines on rebalance really doesn't scale for a fast-lookup structure. For btrees multi-level split/merge hurt, but for the lookup and simple insert/remove case the cacheline miss cost is substantially amortised by packing multiple records per cacheline.
Quoted 10 lines
Quoted 8 lines
BTW, if I take away the rwsem serialisation altogether, this test tops out at just under 500k/500k at 8 threads, and at 16 threads has started dropping off (~440k/440k). So the rwsem is a scalability limitation at just 8 threads....
/me goes off and thinks more about adding optimistic lock coupling to the XFS iext btree to get rid of the need for tree-wide locking altogether
I was not aware of this code.
It's relatively new, and directly tailored to the needs of caching the XFS extent tree - it's not really a generic btree in that it's record store format is the XFS on-disk extent record. i.e. it only stores 54 bits of start offset and 21 bits of length in it's 16 byte records, and the rest of the space is for the record data.
As such, I'm not really implementing a generic range lock - it is highly tailored to the requirements and constraints of XFS filesystem IO. e.g. using max(block size, PAGE_SIZE) granularity range indexing means we can support maximum IO sizes (4GB) with 21 bits of length in the record. And RANGE_LOCK_FULL is implemented as length = MAXEXTLEN + a hidden state bit so it supports exclusive locking from (start, RANGE_LOCK_FULL) and things like truncate, EOF zeroing use this to exclude IO from the range they are working on correctly....
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Quoted 16 lines · Dave Chinner
Quoted 15 lines · Davidlohr Bueso
Quoted 14 lines · Dave Chinner
Now the stuff I've been working on has the same interface as Davidlohr's patch, so I can swap and change them without thinking about it. It's still completely unoptimised, but:
IOPS read/write (direct IO)
processes rwsem DB rangelock XFS
rangelock
1 78k / 78k 75k / 75k 72k / 72k
2 131k / 131k 123k / 123k 133k / 133k
4 267k / 267k 183k / 183k 237k / 237k
8 372k / 372k 177k / 177k 265k / 265k
16 315k / 315k 135k / 135k 228k / 228kIt's still substantially faster than the interval tree code.
....
Quoted 11 lines
Quoted 5 lines
Quoted 3 lines
/me goes off and thinks more about adding optimistic lock coupling to the XFS iext btree to get rid of the need for tree-wide locking altogether
I was not aware of this code.
It's relatively new, and directly tailored to the needs of caching the XFS extent tree - it's not really a generic btree in that it's record store format is the XFS on-disk extent record. i.e. it only stores 54 bits of start offset and 21 bits of length in it's 16 byte records, and the rest of the space is for the record data.
SO now I have a mostly working OLC btree based on this tree which is plumbed into xfsprogs userspace and some testing code. I think I can say now that the code will actually work, and it /should/ scale better than a rwsem.
The userspace test harness that I have ran a "thread profile" to indicated scalability. Basically it ran each thread in a different offset range and locked a hundred ranges and then unlocked them, and then looped over this. The btree is a single level for the first 14 locks, 2-level for up to 210 locks, and 3-level for up to 3150 locks. Hence most of this testing results in the btree being 2-3 levels and so largely removes the global root node lock as a point of contention. It's "best case" for concurrency for an OLC btree.
On a 16p machine:
Range lock/unlock ops/s
threads mutex btree OLC btree
1 5239442 949487
2 1014466 1398539
4 985940 2405275
8 733195 3288435
16 653429 2809225
When looking at these numbers, remember that the mutex btree kernel range lock performed a lot better than the interval tree range lock, and they were only ~30% down on an rwsem. The mutex btree code shows cache residency effects for the single threaded load, hence it looks much faster than it is for occasional and multithreaded access.
However, at 2 threads (where hot CPU caches don't affect the performance), the OLC btree is 40% faster, and at 8 threads it is 4.5x faster than the mutex btree. The OLC btree starts slowing down at 16 threads, largely because the tree itself doesn't have enough depth to provide the interior nodes to scale to higher concurrency levels without contention, but it's still running at 4.5x faster than the mutex btree....
The best part is when I run worse case threaded workloads on the OLC btree. If I run the same 100-lock loops, but this time change the offsets of each thread so they interleave into adjacent records in the btree (i.e. every thread touches every leaf), then the performance is still pretty damn good:
Range lock/unlock ops/s
threads Worst Case Best Case
1 1045991 949487 2 1530212 1398539 4 1147099 2405275 8 1602114 3288435
16 1731890 2809225
IOWs, performance is down and somewhat variable around tree height changes (4 threads straddles the 2-3 level tree height threshold), but it's still a massive improvement on the mutex_btree and it's not going backwards as threads are added.
Concept proven.
Next steps are:
- separate the OLC btree from the XFS iext btree implementation. It will still have a similar interface (i.e. can't manipulate the btree records directly), but there's sufficient difference in structure for them to be separate implementations.
- expand records out to full 64bit extents. The iext tree memory usage constraints no longer apply, so the record size can go up a little bit.
- work out whether RCU read locking and kfree_rcu() will work with the requirement to do memory allocation while holding rcu_read_lock(). Alternative is an internal garbage collector mechanism, kinda like I've hacked up to simulate kfree_rcu() in userspace.
- fix all the little bugs that still exist in the code.
- Think about structural optimisations like parent pointers to avoid costly path walks to find parents for modifications.
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Quoted 3 lines · Dave Chinner
Concept proven.
Next steps are:
....
Quoted 5 lines
- work out whether RCU read locking and
kfree_rcu()will work with the requirement to do memory allocation while holdingrcu_read_lock(). Alternative is an internal garbage collector mechanism, kinda like I've hacked up to simulatekfree_rcu()in userspace.
Internal RCU interactions are now solved. Actually very simple in the end, should be very easy to integrate into the code.
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
kernel test robot
build bot[xfs] fa3fe73bed: aim7.jobs-per-min -10.9% regression
In reply to Amir GoldsteinFYI, we noticed a -10.9% regression of aim7.jobs-per-min due to commit:
Greeting,
FYI, we noticed a -10.9% regression of aim7.jobs-per-min due to commit:
commit: fa3fe73bed17967cb623461b166c46dcffb661a7 ("[POC][PATCH] xfs: reduce ilock contention on buffered randrw workload")
url: https://github.com/0day-ci/linux/commits/Amir-Goldstein/xfs-reduce-ilock-contention-on-buffered-randrw-workload/20190405-100748
in testcase: aim7
on test machine: 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory with following parameters:
disk: 4BRD_12G md: RAID0 fs: xfs test: disk_rr load: 3000 cpufreq_governor: performance
test-description: AIM7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of multiuser system.
test-url: https://sourceforge.net/projects/aimbench/files/aim-suite7/
In addition to that, the commit also has significant impact on the following tests:
+------------------+-----------------------------------------------------------------------+ | testcase: change | vm-scalability: vm-scalability.median -5.2% regression | | test machine | 4 threads Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz with 8G memory | | test parameters | cpufreq_governor=performance | | | runtime=300s | | | test=lru-file-readonce | | | ucode=0x20 | +------------------+-----------------------------------------------------------------------+ | testcase: change | aim7: aim7.jobs-per-min -79.4% regression | | test machine | 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory | | test parameters | cpufreq_governor=performance | | | disk=4BRD_12G | | | fs=xfs | | | load=3000 | | | md=RAID1 | | | test=disk_cp | +------------------+-----------------------------------------------------------------------+
Details are as below:
-------------------------------------------------------------------------------------------------->
To reproduce:
git clone https://github.com/intel/lkp-tests.git
cd lkp-tests
bin/lkp install job.yaml # job file is attached in this email
bin/lkp run job.yaml
=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/load/md/rootfs/tbox_group/test/testcase:
gcc-7/performance/4BRD_12G/xfs/x86_64-rhel-7.6/3000/RAID0/debian-x86_64-2018-04-03.cgz/lkp-ivb-ep01/disk_rr/aim7
commit:
8e22ba96d4 (" RISC-V Patches for 5.1-rc4")
fa3fe73bed ("xfs: reduce ilock contention on buffered randrw workload")
8e22ba96d44c4ad5 fa3fe73bed17967cb623461b166
---------------- ---------------------------
%stddev %change %stddev
\ | \
174264 -10.9% 155325 aim7.jobs-per-min
103.67 +12.2% 116.34 aim7.time.elapsed_time
103.67 +12.2% 116.34 aim7.time.elapsed_time.max
13290 ± 7% +1050.1% 152859 aim7.time.involuntary_context_switches
2760 ± 2% +5.7% 2917 aim7.time.maximum_resident_set_size
148794 ± 2% +11.8% 166280 aim7.time.minor_page_faults
2305 +68.6% 3887 aim7.time.system_time
313.88 -56.9% 135.37 aim7.time.user_time
1087532 -48.9% 555946 aim7.time.voluntary_context_switches
38.60 ± 2% -59.6% 15.61 iostat.cpu.idle
53.90 +51.0% 81.41 iostat.cpu.system
7.49 -60.2% 2.98 iostat.cpu.user
37.49 ± 2% -23.3 14.14 mpstat.cpu.all.idle%
0.00 ± 39% -0.0 0.00 ± 55% mpstat.cpu.all.iowait%
0.03 ± 63% -0.0 0.00 ±130% mpstat.cpu.all.soft%
54.86 +28.0 82.83 mpstat.cpu.all.sys%
7.62 -4.6 3.02 mpstat.cpu.all.usr%
38.00 -60.5% 15.00 vmstat.cpu.id
53.00 +52.8% 81.00 vmstat.cpu.sy
23.75 +53.0% 36.33 vmstat.procs.r
21582 -51.9% 10376 vmstat.system.cs
80340 +2.6% 82446 vmstat.system.in
6656 ± 4% +301.7% 26743 meminfo.Active(file)
1398433 -16.5% 1167608 meminfo.Dirty
1416644 -17.3% 1171397 meminfo.Inactive
1397201 -17.6% 1151246 meminfo.Inactive(file)
29353 ± 2% +42.4% 41789 ± 3% meminfo.Shmem
41450 ± 2% -14.2% 35584 meminfo.max_used_kB
70509709 ± 41% -70.9% 20493630 ± 3% cpuidle.C1.time
689042 ± 44% -70.2% 205612 ± 4% cpuidle.C1.usage
69060244 ± 48% -39.8% 41576976 ± 35% cpuidle.C1E.time
554731 ± 58% -56.4% 241775 ± 33% cpuidle.C1E.usage
1.13e+09 ± 12% -58.3% 4.715e+08 ± 3% cpuidle.C6.time
2020589 ± 9% -65.1% 705482 ± 2% cpuidle.C6.usage
44479 ± 3% -73.1% 11947 ± 3% cpuidle.POLL.time
3370 ± 2% +292.1% 13213 numa-meminfo.node0.Active(file)
695340 -16.3% 581687 numa-meminfo.node0.Dirty
1359044 -10.6% 1215608 numa-meminfo.node0.FilePages
707196 -17.1% 586045 numa-meminfo.node0.Inactive
694427 -17.4% 573469 numa-meminfo.node0.Inactive(file)
3130 ± 4% +334.3% 13594 numa-meminfo.node1.Active(file)
699736 -16.2% 586092 numa-meminfo.node1.Dirty
706171 -17.1% 585519 ± 2% numa-meminfo.node1.Inactive
699495 -17.4% 577906 numa-meminfo.node1.Inactive(file)
2905 ± 10% +19.3% 3465 ± 7% slabinfo.Acpi-Namespace.active_objs
2905 ± 10% +19.3% 3465 ± 7% slabinfo.Acpi-Namespace.num_objs
1005 ± 8% +26.6% 1272 ± 4% slabinfo.blkdev_ioc.active_objs
1005 ± 8% +26.6% 1272 ± 4% slabinfo.blkdev_ioc.num_objs
2764 -12.7% 2413 slabinfo.dmaengine-unmap-16.active_objs
2834 -14.8% 2413 slabinfo.dmaengine-unmap-16.num_objs
2308 ± 6% +17.9% 2722 ± 4% slabinfo.kmalloc-128.active_objs
2308 ± 6% +17.9% 2722 ± 4% slabinfo.kmalloc-128.num_objs
2366 -10.8% 2109 slabinfo.xfs_inode.active_objs
2436 -11.1% 2164 slabinfo.xfs_inode.num_objs
861.75 ± 2% +284.9% 3317 ± 2% numa-vmstat.node0.nr_active_file
174078 -16.8% 144873 numa-vmstat.node0.nr_dirty
340009 -10.8% 303329 numa-vmstat.node0.nr_file_pages
173830 -17.9% 142779 numa-vmstat.node0.nr_inactive_file
862.00 ± 2% +284.5% 3314 ± 2% numa-vmstat.node0.nr_zone_active_file
173830 -17.9% 142774 numa-vmstat.node0.nr_zone_inactive_file
174355 -16.8% 145128 numa-vmstat.node0.nr_zone_write_pending
793.00 ± 3% +322.9% 3353 numa-vmstat.node1.nr_active_file
175039 -16.7% 145725 numa-vmstat.node1.nr_dirty
174980 -17.9% 143680 numa-vmstat.node1.nr_inactive_file
795.00 ± 3% +321.3% 3349 numa-vmstat.node1.nr_zone_active_file
174981 -17.9% 143677 numa-vmstat.node1.nr_zone_inactive_file
175254 -16.7% 145951 numa-vmstat.node1.nr_zone_write_pending
89307 +3.9% 92750 proc-vmstat.nr_active_anon
1631 ± 3% +310.5% 6697 proc-vmstat.nr_active_file
349106 -16.4% 291910 proc-vmstat.nr_dirty
671274 -8.0% 617399 proc-vmstat.nr_file_pages
4860 +3.8% 5044 proc-vmstat.nr_inactive_anon
348815 -17.5% 287737 proc-vmstat.nr_inactive_file
5958 +2.7% 6117 proc-vmstat.nr_mapped
7337 ± 2% +42.4% 10450 ± 3% proc-vmstat.nr_shmem
89307 +3.9% 92750 proc-vmstat.nr_zone_active_anon
1631 ± 3% +310.5% 6697 proc-vmstat.nr_zone_active_file
4860 +3.8% 5044 proc-vmstat.nr_zone_inactive_anon
348814 -17.5% 287736 proc-vmstat.nr_zone_inactive_file
349664 -16.4% 292363 proc-vmstat.nr_zone_write_pending
561.75 ±129% +2428.7% 14205 proc-vmstat.numa_hint_faults
2127 ± 18% +534.8% 13501 ± 5% proc-vmstat.numa_pages_migrated
69610469 +9.7% 76384348 proc-vmstat.pgactivate
444949 +10.3% 490981 proc-vmstat.pgfault
2127 ± 18% +534.8% 13501 ± 5% proc-vmstat.pgmigrate_success
767.75 +234.0% 2564 turbostat.Avg_MHz
63.61 +22.6 86.26 turbostat.Busy%
1206 +146.4% 2973 turbostat.Bzy_MHz
685178 ± 44% -70.7% 200607 ± 4% turbostat.C1
1.64 ± 39% -1.2 0.43 ± 2% turbostat.C1%
554320 ± 58% -56.5% 241296 ± 33% turbostat.C1E
1.61 ± 48% -0.7 0.88 ± 35% turbostat.C1E%
2016855 ± 9% -65.2% 702260 ± 2% turbostat.C6
26.40 ± 11% -16.4 9.97 ± 2% turbostat.C6%
27.23 ± 2% -68.8% 8.49 ± 7% turbostat.CPU%c1
1.55 ± 70% -86.5% 0.21 ± 57% turbostat.CPU%c3
46.47 +210.0% 144.08 turbostat.CorWatt
59.50 +19.3% 71.00 turbostat.CoreTmp
8684289 +13.4% 9851854 turbostat.IRQ
4.56 ± 19% -47.5% 2.39 ± 9% turbostat.Pkg%pc2
59.50 +19.3% 71.00 turbostat.PkgTmp
72.91 +134.8% 171.23 turbostat.PkgWatt
33.43 +1.5% 33.94 turbostat.RAMWatt
8530 +10.7% 9440 turbostat.SMI
6.38 ± 11% -32.0% 4.33 ± 39% sched_debug.cfs_rq:/.load_avg.min
514985 +66.0% 855038 sched_debug.cfs_rq:/.min_vruntime.avg
597546 ± 2% +60.1% 956564 ± 2% sched_debug.cfs_rq:/.min_vruntime.max
496727 +66.9% 829055 sched_debug.cfs_rq:/.min_vruntime.min
0.51 ± 3% +19.5% 0.61 ± 4% sched_debug.cfs_rq:/.nr_running.avg
0.43 ± 5% -31.3% 0.29 ± 5% sched_debug.cfs_rq:/.nr_running.stddev
2.86 ±115% +296.8% 11.36 ± 66% sched_debug.cfs_rq:/.removed.util_avg.avg
75.12 ±100% +160.7% 195.83 ± 20% sched_debug.cfs_rq:/.removed.util_avg.max
14.08 ±105% +184.6% 40.08 ± 36% sched_debug.cfs_rq:/.removed.util_avg.stddev
11.50 +20.3% 13.83 ± 2% sched_debug.cfs_rq:/.runnable_load_avg.avg
31.50 ± 6% +33.9% 42.17 ± 20% sched_debug.cfs_rq:/.runnable_load_avg.max
11118 ± 4% -35.2% 7203 ± 15% sched_debug.cfs_rq:/.runnable_weight.stddev
675.05 +15.4% 778.94 ± 5% sched_debug.cfs_rq:/.util_avg.avg
1251 ± 6% +38.8% 1737 ± 10% sched_debug.cfs_rq:/.util_avg.max
317.25 ± 16% -40.2% 189.67 ± 53% sched_debug.cfs_rq:/.util_avg.min
205.16 ± 8% +68.5% 345.67 ± 8% sched_debug.cfs_rq:/.util_avg.stddev
112.36 ± 6% +80.7% 203.05 ± 6% sched_debug.cfs_rq:/.util_est_enqueued.avg
573.00 ± 18% +45.3% 832.83 ± 23% sched_debug.cfs_rq:/.util_est_enqueued.max
524579 ± 6% -13.6% 453404 ± 3% sched_debug.cpu.avg_idle.avg
244291 ± 4% -9.1% 222083 ± 5% sched_debug.cpu.avg_idle.stddev
11.71 ± 4% -49.3% 5.94 ± 8% sched_debug.cpu.clock.stddev
11.71 ± 4% -49.3% 5.94 ± 8% sched_debug.cpu.clock_task.stddev
11.06 ± 6% +28.9% 14.26 ± 2% sched_debug.cpu.cpu_load[0].avg
29.25 ± 11% +56.1% 45.67 ± 24% sched_debug.cpu.cpu_load[0].max
12.07 ± 8% +27.6% 15.40 sched_debug.cpu.cpu_load[1].avg
13.73 ± 13% +22.1% 16.77 ± 2% sched_debug.cpu.cpu_load[2].avg
1.50 ± 84% +188.9% 4.33 ± 39% sched_debug.cpu.cpu_load[3].min
1142 ± 12% -25.3% 853.96 ± 4% sched_debug.cpu.curr->pid.stddev
14069 ± 7% -15.2% 11926 ± 12% sched_debug.cpu.load.stddev
0.00 ± 4% -16.6% 0.00 ± 3% sched_debug.cpu.next_balance.stddev
0.51 ± 4% +26.0% 0.64 ± 7% sched_debug.cpu.nr_running.avg
16934 -47.6% 8876 sched_debug.cpu.nr_switches.avg
25332 ± 2% -27.0% 18503 ± 6% sched_debug.cpu.nr_switches.max
14912 -52.7% 7050 sched_debug.cpu.nr_switches.min
66.50 ± 4% +95.7% 130.17 ± 15% sched_debug.cpu.nr_uninterruptible.max
9.88 ± 47% -682.3% -57.50 sched_debug.cpu.nr_uninterruptible.min
13.40 ± 8% +185.5% 38.27 ± 5% sched_debug.cpu.nr_uninterruptible.stddev
4.77 -100.0% 0.00 sched_debug.rt_rq:/.rt_runtime.stddev
13.52 ± 12% -60.7% 5.31 ± 11% perf-stat.i.MPKI
3.995e+09 +462.1% 2.245e+10 perf-stat.i.branch-instructions
2.58 ± 9% -1.5 1.09 ± 3% perf-stat.i.branch-miss-rate%
61148500 -10.7% 54581970 ± 2% perf-stat.i.branch-misses
15.27 ± 3% +5.0 20.26 perf-stat.i.cache-miss-rate%
24227403 ± 2% +35.1% 32730873 perf-stat.i.cache-misses
22100 -52.5% 10507 perf-stat.i.context-switches
2.05 ± 6% -32.9% 1.38 ± 3% perf-stat.i.cpi
3.059e+10 +235.1% 1.025e+11 perf-stat.i.cpu-cycles
1453 ± 2% +49.5% 2172 perf-stat.i.cpu-migrations
1212 ± 2% +142.2% 2937 perf-stat.i.cycles-between-cache-misses
2.10 ± 6% -1.2 0.85 ± 6% perf-stat.i.dTLB-load-miss-rate%
1.442e+08 ± 6% +73.1% 2.496e+08 ± 7% perf-stat.i.dTLB-load-misses
6.303e+09 +365.6% 2.935e+10 perf-stat.i.dTLB-loads
0.08 ± 9% -0.0 0.04 ± 12% perf-stat.i.dTLB-store-miss-rate%
3068070 ± 7% +24.7% 3826133 ± 15% perf-stat.i.dTLB-store-misses
4.064e+09 +309.2% 1.663e+10 perf-stat.i.dTLB-stores
8435967 ± 3% -25.8% 6259422 perf-stat.i.iTLB-load-misses
2.021e+10 +428.7% 1.069e+11 perf-stat.i.instructions
2390 ± 5% +560.1% 15782 perf-stat.i.instructions-per-iTLB-miss
0.62 +56.0% 0.97 perf-stat.i.ipc
20.68 +2.3 22.94 perf-stat.i.node-load-miss-rate%
2760936 +29.0% 3562738 perf-stat.i.node-load-misses
12616257 +6.2% 13392887 perf-stat.i.node-loads
19.20 ± 4% -3.2 16.04 perf-stat.i.node-store-miss-rate%
2286706 +31.6% 3009786 perf-stat.i.node-store-misses
10050267 ± 2% +70.0% 17083826 perf-stat.i.node-stores
8.39 -81.4% 1.56 perf-stat.overall.MPKI
1.53 -1.3 0.24 ± 2% perf-stat.overall.branch-miss-rate%
14.28 +5.3 19.58 perf-stat.overall.cache-miss-rate%
1.51 -36.6% 0.96 perf-stat.overall.cpi
1262 +147.9% 3131 perf-stat.overall.cycles-between-cache-misses
2.24 ± 6% -1.4 0.84 ± 7% perf-stat.overall.dTLB-load-miss-rate%
0.08 ± 6% -0.1 0.02 ± 15% perf-stat.overall.dTLB-store-miss-rate%
2398 ± 3% +612.0% 17076 perf-stat.overall.instructions-per-iTLB-miss
0.66 +57.8% 1.04 perf-stat.overall.ipc
17.95 +3.1 21.01 perf-stat.overall.node-load-miss-rate%
18.54 ± 2% -3.6 14.98 perf-stat.overall.node-store-miss-rate%
3.957e+09 +462.5% 2.226e+10 perf-stat.ps.branch-instructions
60563941 -10.7% 54108124 ± 2% perf-stat.ps.branch-misses
23995092 ± 2% +35.2% 32443881 perf-stat.ps.cache-misses
21887 -52.4% 10415 perf-stat.ps.context-switches
3.029e+10 +235.4% 1.016e+11 perf-stat.ps.cpu-cycles
1439 ± 2% +49.6% 2153 perf-stat.ps.cpu-migrations
1.428e+08 ± 6% +73.3% 2.474e+08 ± 7% perf-stat.ps.dTLB-load-misses
6.242e+09 +366.0% 2.909e+10 perf-stat.ps.dTLB-loads
3038630 ± 7% +24.8% 3792456 ± 15% perf-stat.ps.dTLB-store-misses
4.025e+09 +309.6% 1.649e+10 perf-stat.ps.dTLB-stores
8354826 ± 3% -25.7% 6204340 perf-stat.ps.iTLB-load-misses
2.002e+10 +429.1% 1.059e+11 perf-stat.ps.instructions
2734433 +29.1% 3531311 perf-stat.ps.node-load-misses
12494983 +6.2% 13274613 perf-stat.ps.node-loads
2264742 +31.7% 2983304 perf-stat.ps.node-store-misses
9953749 ± 2% +70.1% 16933523 perf-stat.ps.node-stores
2.12e+12 +484.3% 1.238e+13 perf-stat.total.instructions
212836 +10.4% 234958 interrupts.CPU0.LOC:Local_timer_interrupts
1104 ± 26% +260.6% 3984 ± 5% interrupts.CPU0.RES:Rescheduling_interrupts
212437 +10.6% 234923 interrupts.CPU1.LOC:Local_timer_interrupts
1004 ± 34% +237.6% 3392 ± 4% interrupts.CPU1.RES:Rescheduling_interrupts
212437 +10.5% 234681 interrupts.CPU10.LOC:Local_timer_interrupts
1254 ± 51% +158.9% 3246 interrupts.CPU10.RES:Rescheduling_interrupts
212545 +10.4% 234638 interrupts.CPU11.LOC:Local_timer_interrupts
721.50 ± 10% +454.4% 4000 ± 25% interrupts.CPU11.RES:Rescheduling_interrupts
212604 +10.4% 234633 interrupts.CPU12.LOC:Local_timer_interrupts
786.00 ± 6% +308.9% 3213 ± 2% interrupts.CPU12.RES:Rescheduling_interrupts
725.75 ± 12% +360.3% 3340 ± 2% interrupts.CPU13.RES:Rescheduling_interrupts
773.25 ± 93% +775.2% 6767 ± 28% interrupts.CPU14.NMI:Non-maskable_interrupts
773.25 ± 93% +775.2% 6767 ± 28% interrupts.CPU14.PMI:Performance_monitoring_interrupts
716.75 ± 10% +355.1% 3262 ± 3% interrupts.CPU14.RES:Rescheduling_interrupts
725.00 ± 8% +361.4% 3345 ± 5% interrupts.CPU15.RES:Rescheduling_interrupts
212119 +10.6% 234586 interrupts.CPU16.LOC:Local_timer_interrupts
390.25 ±172% +1285.7% 5407 ± 35% interrupts.CPU16.NMI:Non-maskable_interrupts
390.25 ±172% +1285.7% 5407 ± 35% interrupts.CPU16.PMI:Performance_monitoring_interrupts
692.25 ± 13% +428.8% 3660 ± 14% interrupts.CPU16.RES:Rescheduling_interrupts
212372 +10.5% 234579 interrupts.CPU17.LOC:Local_timer_interrupts
723.00 ± 10% +357.1% 3304 interrupts.CPU17.RES:Rescheduling_interrupts
212228 +10.6% 234681 interrupts.CPU18.LOC:Local_timer_interrupts
790.00 ± 10% +306.5% 3211 ± 3% interrupts.CPU18.RES:Rescheduling_interrupts
795.75 ± 14% +353.3% 3607 ± 3% interrupts.CPU19.RES:Rescheduling_interrupts
212317 +10.2% 233944 interrupts.CPU2.LOC:Local_timer_interrupts
956.00 ± 19% +377.3% 4562 ± 26% interrupts.CPU2.RES:Rescheduling_interrupts
212471 +10.5% 234677 interrupts.CPU20.LOC:Local_timer_interrupts
739.50 ± 11% +341.3% 3263 interrupts.CPU20.RES:Rescheduling_interrupts
751.75 ±170% +802.2% 6782 ± 28% interrupts.CPU21.NMI:Non-maskable_interrupts
751.75 ±170% +802.2% 6782 ± 28% interrupts.CPU21.PMI:Performance_monitoring_interrupts
666.75 ± 8% +401.0% 3340 interrupts.CPU21.RES:Rescheduling_interrupts
742.50 ± 14% +339.0% 3259 ± 2% interrupts.CPU22.RES:Rescheduling_interrupts
212505 +10.4% 234501 interrupts.CPU23.LOC:Local_timer_interrupts
725.00 ± 6% +353.0% 3284 interrupts.CPU23.RES:Rescheduling_interrupts
212540 +10.4% 234711 interrupts.CPU24.LOC:Local_timer_interrupts
796.75 ± 8% +307.1% 3243 interrupts.CPU24.RES:Rescheduling_interrupts
760.25 ± 7% +323.9% 3222 interrupts.CPU25.RES:Rescheduling_interrupts
212521 +10.3% 234433 interrupts.CPU26.LOC:Local_timer_interrupts
731.00 ± 10% +348.1% 3275 ± 2% interrupts.CPU26.RES:Rescheduling_interrupts
1647 ± 65% +391.9% 8102 interrupts.CPU27.NMI:Non-maskable_interrupts
1647 ± 65% +391.9% 8102 interrupts.CPU27.PMI:Performance_monitoring_interrupts
696.75 ± 9% +364.1% 3233 ± 2% interrupts.CPU27.RES:Rescheduling_interrupts
212335 +10.5% 234699 interrupts.CPU28.LOC:Local_timer_interrupts
859.25 ± 10% +274.9% 3221 ± 3% interrupts.CPU28.RES:Rescheduling_interrupts
724.50 ± 10% +344.8% 3222 ± 3% interrupts.CPU29.RES:Rescheduling_interrupts
212478 +10.5% 234859 interrupts.CPU3.LOC:Local_timer_interrupts
725.75 ± 7% +369.2% 3405 ± 2% interrupts.CPU3.RES:Rescheduling_interrupts
212297 +10.4% 234380 interrupts.CPU30.LOC:Local_timer_interrupts
653.00 ± 4% +406.5% 3307 interrupts.CPU30.RES:Rescheduling_interrupts
716.00 ± 12% +356.7% 3270 interrupts.CPU31.RES:Rescheduling_interrupts
212377 +10.4% 234446 interrupts.CPU32.LOC:Local_timer_interrupts
739.25 ± 16% +338.4% 3240 interrupts.CPU32.RES:Rescheduling_interrupts
773.50 ± 5% +318.3% 3235 interrupts.CPU33.RES:Rescheduling_interrupts
212097 +10.5% 234446 interrupts.CPU34.LOC:Local_timer_interrupts
807.25 ± 17% +311.5% 3321 interrupts.CPU34.RES:Rescheduling_interrupts
1874 ± 65% +117.5% 4075 interrupts.CPU35.NMI:Non-maskable_interrupts
1874 ± 65% +117.5% 4075 interrupts.CPU35.PMI:Performance_monitoring_interrupts
704.00 ± 10% +363.1% 3260 ± 2% interrupts.CPU35.RES:Rescheduling_interrupts
765.75 ± 11% +326.7% 3267 interrupts.CPU36.RES:Rescheduling_interrupts
2066 ± 43% +96.7% 4065 interrupts.CPU37.NMI:Non-maskable_interrupts
2066 ± 43% +96.7% 4065 interrupts.CPU37.PMI:Performance_monitoring_interrupts
710.75 ± 14% +368.9% 3332 ± 4% interrupts.CPU37.RES:Rescheduling_interrupts
212185 +10.6% 234607 interrupts.CPU38.LOC:Local_timer_interrupts
1126 ±108% +381.1% 5416 ± 35% interrupts.CPU38.NMI:Non-maskable_interrupts
1126 ±108% +381.1% 5416 ± 35% interrupts.CPU38.PMI:Performance_monitoring_interrupts
690.00 ± 10% +376.4% 3287 ± 2% interrupts.CPU38.RES:Rescheduling_interrupts
212183 +10.5% 234362 interrupts.CPU39.LOC:Local_timer_interrupts
693.00 ± 9% +360.0% 3188 interrupts.CPU39.RES:Rescheduling_interrupts
1215 ± 61% +166.0% 3232 interrupts.CPU4.RES:Rescheduling_interrupts
794.25 ± 10% +308.6% 3245 interrupts.CPU5.RES:Rescheduling_interrupts
212385 +10.3% 234361 interrupts.CPU6.LOC:Local_timer_interrupts
1500 ± 57% +120.6% 3311 interrupts.CPU6.RES:Rescheduling_interrupts
790.50 ± 13% +314.7% 3278 ± 3% interrupts.CPU7.RES:Rescheduling_interrupts
212494 +10.4% 234598 interrupts.CPU8.LOC:Local_timer_interrupts
690.00 ± 12% +359.7% 3171 interrupts.CPU8.RES:Rescheduling_interrupts
651.50 ± 7% +404.6% 3287 interrupts.CPU9.RES:Rescheduling_interrupts
8500355 +10.3% 9379328 interrupts.LOC:Local_timer_interrupts
53220 ± 11% +172.2% 144867 ± 7% interrupts.NMI:Non-maskable_interrupts
53220 ± 11% +172.2% 144867 ± 7% interrupts.PMI:Performance_monitoring_interrupts
32058 ± 3% +319.0% 134339 interrupts.RES:Rescheduling_interrupts
346.75 ± 8% -59.3% 141.00 ± 8% interrupts.TLB:TLB_shootdowns
37.49 -23.8 13.67 perf-profile.calltrace.cycles-pp.write
32.13 ± 2% -20.3 11.82 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.write
31.88 ± 2% -20.2 11.73 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
22.34 ± 6% -19.7 2.63 ± 13% perf-profile.calltrace.cycles-pp.secondary_startup_64
30.70 ± 2% -19.3 11.36 perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
21.87 ± 7% -19.3 2.55 ± 11% perf-profile.calltrace.cycles-pp.cpu_startup_entry.start_secondary.secondary_startup_64
21.87 ± 7% -19.3 2.55 ± 11% perf-profile.calltrace.cycles-pp.start_secondary.secondary_startup_64
21.87 ± 7% -19.3 2.55 ± 11% perf-profile.calltrace.cycles-pp.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
21.43 ± 7% -18.9 2.50 ± 11% perf-profile.calltrace.cycles-pp.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary.secondary_startup_64
29.87 -18.8 11.06 perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
20.88 ± 7% -18.4 2.46 ± 11% perf-profile.calltrace.cycles-pp.intel_idle.cpuidle_enter_state.do_idle.cpu_startup_entry.start_secondary
26.39 -16.7 9.72 perf-profile.calltrace.cycles-pp.new_sync_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
24.86 ± 2% -15.6 9.30 perf-profile.calltrace.cycles-pp.xfs_file_buffered_aio_write.new_sync_write.vfs_write.ksys_write.do_syscall_64
20.87 -12.9 7.93 perf-profile.calltrace.cycles-pp.iomap_file_buffered_write.xfs_file_buffered_aio_write.new_sync_write.vfs_write.ksys_write
20.72 -12.9 7.87 perf-profile.calltrace.cycles-pp.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write.new_sync_write.vfs_write
15.79 -9.8 5.98 perf-profile.calltrace.cycles-pp.iomap_write_actor.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write.new_sync_write
8.44 -5.6 2.86 perf-profile.calltrace.cycles-pp.lseek64
8.52 ± 2% -5.3 3.19 perf-profile.calltrace.cycles-pp.iomap_write_begin.iomap_write_actor.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write
6.07 ± 3% -3.9 2.14 perf-profile.calltrace.cycles-pp.grab_cache_page_write_begin.iomap_write_begin.iomap_write_actor.iomap_apply.iomap_file_buffered_write
5.83 ± 3% -3.8 2.04 ± 2% perf-profile.calltrace.cycles-pp.pagecache_get_page.grab_cache_page_write_begin.iomap_write_begin.iomap_write_actor.iomap_apply
4.17 ± 3% -2.6 1.52 perf-profile.calltrace.cycles-pp.close
4.11 ± 3% -2.6 1.50 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.close
4.11 ± 3% -2.6 1.50 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.close
4.09 ± 3% -2.6 1.50 perf-profile.calltrace.cycles-pp.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe.close
4.08 ± 3% -2.6 1.49 perf-profile.calltrace.cycles-pp.task_work_run.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe.close
4.06 ± 3% -2.6 1.48 perf-profile.calltrace.cycles-pp.__fput.task_work_run.exit_to_usermode_loop.do_syscall_64.entry_SYSCALL_64_after_hwframe
3.88 ± 3% -2.5 1.43 perf-profile.calltrace.cycles-pp.dput.__fput.task_work_run.exit_to_usermode_loop.do_syscall_64
3.87 ± 3% -2.4 1.42 perf-profile.calltrace.cycles-pp.dentry_kill.dput.__fput.task_work_run.exit_to_usermode_loop
3.87 ± 3% -2.4 1.42 perf-profile.calltrace.cycles-pp.__dentry_kill.dentry_kill.dput.__fput.task_work_run
3.64 ± 3% -2.2 1.40 perf-profile.calltrace.cycles-pp.xfs_file_iomap_begin.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write.new_sync_write
3.44 ± 2% -2.2 1.23 ± 3% perf-profile.calltrace.cycles-pp.iomap_write_end.iomap_write_actor.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write
3.11 ± 4% -2.1 1.06 ± 3% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.lseek64
2.89 ± 5% -1.9 0.97 ± 4% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.lseek64
3.09 ± 2% -1.9 1.19 perf-profile.calltrace.cycles-pp.xfs_file_iomap_begin_delay.xfs_file_iomap_begin.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write
2.99 ± 2% -1.9 1.10 perf-profile.calltrace.cycles-pp.evict.__dentry_kill.dentry_kill.dput.__fput
2.97 ± 2% -1.9 1.09 perf-profile.calltrace.cycles-pp.truncate_inode_pages_range.evict.__dentry_kill.dentry_kill.dput
2.81 -1.9 0.94 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.lseek64
2.78 ± 3% -1.8 0.96 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.read
2.78 ± 2% -1.8 0.97 ± 3% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.write
2.39 ± 3% -1.6 0.80 ± 3% perf-profile.calltrace.cycles-pp.add_to_page_cache_lru.pagecache_get_page.grab_cache_page_write_begin.iomap_write_begin.iomap_write_actor
2.32 ± 3% -1.6 0.76 ± 5% perf-profile.calltrace.cycles-pp.xfs_file_aio_write_checks.xfs_file_buffered_aio_write.new_sync_write.vfs_write.ksys_write
2.29 -1.5 0.78 ± 2% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.write
2.28 -1.5 0.78 ± 2% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.read
1.98 ± 3% -1.4 0.56 ± 3% perf-profile.calltrace.cycles-pp.creat
1.96 ± 3% -1.4 0.54 ± 3% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.creat
1.96 ± 3% -1.4 0.54 ± 3% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.creat
1.95 ± 3% -1.4 0.54 ± 3% perf-profile.calltrace.cycles-pp.do_sys_open.do_syscall_64.entry_SYSCALL_64_after_hwframe.creat
1.93 ± 3% -1.4 0.54 ± 4% perf-profile.calltrace.cycles-pp.do_filp_open.do_sys_open.do_syscall_64.entry_SYSCALL_64_after_hwframe.creat
1.92 ± 3% -1.4 0.53 ± 4% perf-profile.calltrace.cycles-pp.path_openat.do_filp_open.do_sys_open.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.09 ± 2% -1.4 0.72 ± 2% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.lseek64
1.81 ± 7% -1.2 0.60 ± 5% perf-profile.calltrace.cycles-pp.ksys_lseek.do_syscall_64.entry_SYSCALL_64_after_hwframe.lseek64
2.60 ± 5% -1.2 1.41 ± 2% perf-profile.calltrace.cycles-pp.copyout.copy_page_to_iter.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter
2.57 ± 5% -1.2 1.39 ± 2% perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyout.copy_page_to_iter.generic_file_read_iter.xfs_file_buffered_aio_read
2.04 -1.2 0.88 perf-profile.calltrace.cycles-pp.iomap_read_page_sync.iomap_write_begin.iomap_write_actor.iomap_apply.iomap_file_buffered_write
1.94 ± 2% -1.1 0.81 perf-profile.calltrace.cycles-pp.iov_iter_copy_from_user_atomic.iomap_write_actor.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write
1.70 ± 3% -1.0 0.67 ± 3% perf-profile.calltrace.cycles-pp.security_file_permission.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.48 ± 3% -1.0 0.52 perf-profile.calltrace.cycles-pp.__alloc_pages_nodemask.pagecache_get_page.grab_cache_page_write_begin.iomap_write_begin.iomap_write_actor
1.62 ± 2% -1.0 0.67 perf-profile.calltrace.cycles-pp.copyin.iov_iter_copy_from_user_atomic.iomap_write_actor.iomap_apply.iomap_file_buffered_write
1.48 ± 2% -0.9 0.53 ± 4% perf-profile.calltrace.cycles-pp.iomap_set_page_dirty.iomap_write_end.iomap_write_actor.iomap_apply.iomap_file_buffered_write
1.55 ± 3% -0.9 0.64 perf-profile.calltrace.cycles-pp.copy_user_enhanced_fast_string.copyin.iov_iter_copy_from_user_atomic.iomap_write_actor.iomap_apply
1.46 ± 5% -0.9 0.56 perf-profile.calltrace.cycles-pp.security_file_permission.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.49 ± 2% -0.8 0.66 perf-profile.calltrace.cycles-pp.memset_erms.iomap_read_page_sync.iomap_write_begin.iomap_write_actor.iomap_apply
0.77 ± 8% -0.2 0.59 perf-profile.calltrace.cycles-pp.touch_atime.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
0.00 +1.0 1.01 perf-profile.calltrace.cycles-pp.rcu_all_qs._cond_resched.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter
0.00 +1.9 1.90 perf-profile.calltrace.cycles-pp._cond_resched.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
3.44 ± 5% +3.1 6.51 perf-profile.calltrace.cycles-pp.copy_page_to_iter.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
1.94 ± 3% +3.4 5.34 perf-profile.calltrace.cycles-pp.mark_page_accessed.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
0.00 +5.5 5.50 ± 3% perf-profile.calltrace.cycles-pp.xas_start.xas_load.find_get_entry.pagecache_get_page.generic_file_read_iter
0.00 +6.2 6.23 perf-profile.calltrace.cycles-pp.___might_sleep.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
0.75 ± 3% +12.5 13.26 perf-profile.calltrace.cycles-pp.xas_load.find_get_entry.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read
1.88 ± 2% +29.5 31.36 perf-profile.calltrace.cycles-pp.find_get_entry.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter
1.98 ± 2% +33.0 34.98 perf-profile.calltrace.cycles-pp.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
22.57 ± 2% +55.3 77.91 perf-profile.calltrace.cycles-pp.read
17.07 +58.9 75.99 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.read
16.81 ± 2% +59.1 75.89 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
15.73 ± 2% +59.8 75.53 perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
14.88 ± 2% +60.4 75.25 perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
11.85 ± 2% +62.3 74.12 perf-profile.calltrace.cycles-pp.new_sync_read.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
11.49 ± 2% +62.5 73.98 perf-profile.calltrace.cycles-pp.xfs_file_read_iter.new_sync_read.vfs_read.ksys_read.do_syscall_64
9.23 ± 2% +62.6 71.83 perf-profile.calltrace.cycles-pp.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read.vfs_read
10.26 ± 2% +63.3 73.60 perf-profile.calltrace.cycles-pp.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read.vfs_read.ksys_read
17585 ± 2% -22.6% 13604 ± 3% softirqs.CPU0.RCU
18027 ± 2% -31.3% 12389 ± 2% softirqs.CPU0.SCHED
17656 ± 3% -26.0% 13065 ± 2% softirqs.CPU1.RCU
14588 ± 2% -37.4% 9138 softirqs.CPU1.SCHED
17689 ± 6% -28.2% 12698 ± 9% softirqs.CPU10.RCU
14546 -43.2% 8264 ± 3% softirqs.CPU10.SCHED
16926 ± 4% -25.2% 12664 ± 12% softirqs.CPU11.RCU
14657 -36.3% 9333 ± 5% softirqs.CPU11.SCHED
19656 ± 10% -26.4% 14468 ± 20% softirqs.CPU12.RCU
15085 ± 3% -39.6% 9112 softirqs.CPU12.SCHED
45423 ± 3% +12.4% 51043 ± 3% softirqs.CPU12.TIMER
17215 ± 7% -23.8% 13114 ± 5% softirqs.CPU13.RCU
14710 -39.1% 8960 ± 3% softirqs.CPU13.SCHED
17320 ± 4% -28.2% 12431 ± 2% softirqs.CPU14.RCU
14956 -37.3% 9380 ± 2% softirqs.CPU14.SCHED
18506 ± 6% -29.2% 13108 ± 4% softirqs.CPU15.RCU
14702 -38.8% 8995 ± 2% softirqs.CPU15.SCHED
17497 ± 6% -31.1% 12060 ± 4% softirqs.CPU16.RCU
14987 -39.0% 9144 ± 4% softirqs.CPU16.SCHED
44555 ± 2% +14.0% 50798 ± 3% softirqs.CPU16.TIMER
17239 ± 6% -26.2% 12720 ± 4% softirqs.CPU17.RCU
14707 -39.0% 8968 ± 2% softirqs.CPU17.SCHED
17035 ± 5% -24.4% 12872 ± 4% softirqs.CPU18.RCU
14862 -39.0% 9065 softirqs.CPU18.SCHED
44056 ± 2% +13.3% 49898 softirqs.CPU18.TIMER
17108 ± 6% -29.4% 12077 ± 4% softirqs.CPU19.RCU
14789 -39.6% 8934 ± 2% softirqs.CPU19.SCHED
19187 ± 8% -22.8% 14803 ± 15% softirqs.CPU2.RCU
15056 ± 6% -37.8% 9368 ± 2% softirqs.CPU2.SCHED
45404 ± 5% +13.9% 51724 ± 4% softirqs.CPU2.TIMER
17685 ± 5% -28.2% 12690 ± 9% softirqs.CPU20.RCU
15162 ± 5% -41.0% 8951 softirqs.CPU20.SCHED
17396 ± 5% -30.1% 12152 ± 11% softirqs.CPU21.RCU
14572 -38.6% 8943 ± 2% softirqs.CPU21.SCHED
17193 ± 5% -28.8% 12242 ± 7% softirqs.CPU22.RCU
15052 -40.9% 8899 ± 3% softirqs.CPU22.SCHED
44869 ± 3% +22.3% 54895 ± 15% softirqs.CPU22.TIMER
17908 ± 9% -30.2% 12494 ± 5% softirqs.CPU23.RCU
14659 -39.0% 8946 ± 3% softirqs.CPU23.SCHED
18382 ± 10% -30.9% 12706 ± 2% softirqs.CPU24.RCU
14776 -38.7% 9061 softirqs.CPU24.SCHED
44675 ± 2% +12.5% 50240 softirqs.CPU24.TIMER
17003 ± 3% -22.1% 13251 ± 11% softirqs.CPU25.RCU
14698 -39.0% 8973 ± 4% softirqs.CPU25.SCHED
18396 ± 10% -36.1% 11747 ± 3% softirqs.CPU26.RCU
14924 ± 2% -40.0% 8949 ± 3% softirqs.CPU26.SCHED
44944 +16.4% 52321 ± 4% softirqs.CPU26.TIMER
17076 ± 5% -26.1% 12614 ± 6% softirqs.CPU27.RCU
14680 -38.8% 8982 ± 3% softirqs.CPU27.SCHED
43977 ± 2% +10.7% 48691 softirqs.CPU27.TIMER
16848 ± 3% -27.6% 12203 ± 5% softirqs.CPU28.RCU
14838 ± 3% -38.3% 9156 ± 2% softirqs.CPU28.SCHED
44188 +12.3% 49639 softirqs.CPU28.TIMER
17635 ± 8% -24.4% 13339 ± 7% softirqs.CPU29.RCU
14668 -38.6% 9004 ± 2% softirqs.CPU29.SCHED
16857 ± 3% -21.4% 13243 ± 7% softirqs.CPU3.RCU
14869 -38.8% 9098 softirqs.CPU3.SCHED
17771 ± 8% -29.0% 12612 ± 10% softirqs.CPU30.RCU
14272 ± 3% -41.1% 8401 ± 4% softirqs.CPU30.SCHED
42106 +11.9% 47104 softirqs.CPU30.TIMER
17493 ± 6% -30.5% 12163 ± 7% softirqs.CPU31.RCU
14723 -38.7% 9022 ± 2% softirqs.CPU31.SCHED
17607 ± 20% -32.7% 11841 ± 26% softirqs.CPU32.RCU
15051 ± 2% -40.2% 8995 softirqs.CPU32.SCHED
13429 ± 3% -28.4% 9620 ± 4% softirqs.CPU33.RCU
14633 -37.3% 9169 ± 2% softirqs.CPU33.SCHED
14829 ± 8% -35.2% 9614 softirqs.CPU34.RCU
14929 -39.3% 9068 softirqs.CPU34.SCHED
45384 ± 4% +9.7% 49801 ± 2% softirqs.CPU34.TIMER
13817 ± 9% -24.8% 10384 ± 8% softirqs.CPU35.RCU
14756 -39.6% 8914 ± 2% softirqs.CPU35.SCHED
13609 ± 2% -22.7% 10525 ± 9% softirqs.CPU36.RCU
14906 ± 2% -40.1% 8927 ± 2% softirqs.CPU36.SCHED
44085 +33.3% 58762 ± 21% softirqs.CPU36.TIMER
14308 ± 14% -26.3% 10550 ± 12% softirqs.CPU37.RCU
14712 -38.6% 9029 softirqs.CPU37.SCHED
15404 ± 8% -30.2% 10753 ± 9% softirqs.CPU38.RCU
14757 -39.3% 8963 softirqs.CPU38.SCHED
13292 ± 4% -22.4% 10309 ± 10% softirqs.CPU39.RCU
14643 ± 2% -39.2% 8901 softirqs.CPU39.SCHED
17513 ± 8% -27.8% 12646 ± 3% softirqs.CPU4.RCU
15099 ± 3% -40.3% 9020 softirqs.CPU4.SCHED
17507 ± 8% -27.9% 12614 ± 3% softirqs.CPU5.RCU
14752 -39.2% 8975 ± 3% softirqs.CPU5.SCHED
17584 ± 5% -31.6% 12025 ± 2% softirqs.CPU6.RCU
15115 ± 2% -42.0% 8760 ± 3% softirqs.CPU6.SCHED
18973 ± 3% -29.0% 13480 ± 3% softirqs.CPU7.RCU
14761 -39.3% 8957 ± 3% softirqs.CPU7.SCHED
17096 ± 4% -28.8% 12176 softirqs.CPU8.RCU
14989 ± 2% -40.5% 8926 softirqs.CPU8.SCHED
17209 ± 7% -30.8% 11908 ± 4% softirqs.CPU9.RCU
14720 -38.3% 9076 softirqs.CPU9.SCHED
680461 ± 4% -27.8% 491610 ± 3% softirqs.RCU
595414 -39.0% 363137 softirqs.SCHED
aim7.jobs-per-min 180000 +-+----------------------------------------------------------------+
| : : : : + O : : : : : : + |
160000 O-O O O O O O : O O O O : O O O O O O O : |
140000 +-+ : :: : : : :: : : : : |
|: : : : : : : : : : : : : : : : : |
120000 +-+: : : : : : : : : : : : : : : : |
100000 +-+: : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
80000 +-+: : : : : : : : : : : : : : : : |
60000 +-+: : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
40000 +-+ : : : :: :: : : : : : : |
20000 +-+ : : : : : : : : : : : |
| : : : : : : : : : : : |
0 +-+----O--------O---O-------------O-------O--------O---------------+ aim7.time.user_time
350 +-+-------------------------------------------------------------------+
| + + + +. .+ + + + +.+.+..+.+ +.+. .+.|
300 +-+ : : : + : : : : : : : +. |
| : : : : : : : : : : : |
250 +-+: :: : : : : : :: : : : |
|: : : : : : : : : : : : : : : : : |
200 +-+: : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
150 O-O :O :O :O:O: : : : : : : : : : : : |
|: : : : : : : O O :O:O:O O: :O: O:O : O O O O: : |
100 +-+ : : : : : : : : : : : : : : : : |
| : : :: :: :: : : :: : : :: |
50 +-+ : : : : : : : : : : : |
| : : : : : : : : : : : |
0 +-+----O--------O----O-------------O--------O---------O---------------+ aim7.time.system_time
4000 +-+---------O-O----------O--O----------O-O----O-O--O---O-------------+
O O O O O O O O O |
3500 +-+ |
3000 +-+ |
| |
2500 +-+ |
| + + + +..+.+ + + + +.+..+.+.+ +.+.+..+.|
2000 +-+ : : : : : : : : : : |
| :: :: :: : : : : : : :: : : : |
1500 +-+: : : : : : : : : : : : : : : : |
1000 +-+ : : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
500 +-+ : : : : :: :: : : :: : : : : |
| : : : : : : : : : : : |
0 +-+----O--------O----O-------------O--------O--------O---------------+
aim7.time.elapsed_time 120 +-+---------O-O-----------O-O-----------O-O----O-O-O----O-------------+
O O O O O O O O O |
100 +-+ + + +.+.+ + + + +.+.+..+.+ +.+.+..+.|
| : : : : : : : : : : : |
| :: :: : : : : : : : : : |
80 +-+: :: :: : : :: :: : : : : : |
|: : : : : : : : : : : : : : : : : |
60 +-+: : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
40 +-+ : : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
|: : : : : :: : : : : :: : : :: |
20 +-+ : : : :: : : : :: : : :: |
| : : : : : : : : : : : |
0 +-+----O--------O----O-------------O--------O---------O---------------+
aim7.time.elapsed_time.max 120 +-+---------O-O-----------O-O-----------O-O----O-O-O----O-------------+
O O O O O O O O O |
100 +-+ + + +.+.+ + + + +.+.+..+.+ +.+.+..+.|
| : : : : : : : : : : : |
| :: :: : : : : : : : : : |
80 +-+: :: :: : : :: :: : : : : : |
|: : : : : : : : : : : : : : : : : |
60 +-+: : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
40 +-+ : : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
|: : : : : :: : : : : :: : : :: |
20 +-+ : : : :: : : : :: : : :: |
| : : : : : : : : : : : |
0 +-+----O--------O----O-------------O--------O---------O---------------+
aim7.time.voluntary_context_switches 1.2e+06 +-+---------------------------------------------------------------+
| + + + +.+.+ + + + +.+.+..+.+ +.+..+.+.|
1e+06 +-+ : : : : : : : : : : |
| : : : : : : : : : : : |
| : : : : : :: : :: : : : |
800000 +-+: : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
600000 O-O:O :O: : : : O: : : O O: :O:O: : O : : |
|: : : :O:O: : O :O:O : : : : :O :O O O: : |
400000 +-+: : : : : : : : : : : : : : : : |
|: : : : : : : : : : : : : : : : : |
| : : : :: : : : : : : : |
200000 +-+ : : : : : : : : : : : |
| : : : : : : : : : : : |
0 +-+----O-------O----O------------O--------O--------O--------------+ aim7.time.involuntary_context_switches
160000 +-+------------------------O---------------------------------------+
O O O O O O O O O O O O O O O O O O |
140000 +-+ |
120000 +-+ |
| |
100000 +-+ |
| |
80000 +-+ |
| |
60000 +-+ |
40000 +-+ |
| |
20000 +-+ |
|.+. .+. .+.. .+.+.+.. .+. .+. .+. .+.+.+.+..+. .+.+..+.+.|
0 +-+----O--------O---O-------------O-------O--------O---------------+ [*] bisect-good sample
[O] bisect-bad sample
***************************************************************************************************
lkp-ivb-d02: 4 threads Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz with 8G memory
=========================================================================================
compiler/cpufreq_governor/kconfig/rootfs/runtime/tbox_group/test/testcase/ucode:
gcc-7/performance/x86_64-rhel-7.6/debian-x86_64-2018-04-03.cgz/300s/lkp-ivb-d02/lru-file-readonce/vm-scalability/0x20
commit:
8e22ba96d4 (" RISC-V Patches for 5.1-rc4")
fa3fe73bed ("xfs: reduce ilock contention on buffered randrw workload")
8e22ba96d44c4ad5 fa3fe73bed17967cb623461b166
---------------- ---------------------------
fail:runs %reproduction fail:runs
| | |
:4 50% 2:4 dmesg.WARNING:at#for_ip_swapgs_restore_regs_and_return_to_usermode/0x
:4 50% 2:4 dmesg.WARNING:stack_recursion
%stddev %change %stddev
\ | \
1213560 -5.2% 1150446 vm-scalability.median
4826960 -5.3% 4569635 vm-scalability.throughput
3737 -2.1% 3658 vm-scalability.time.maximum_resident_set_size
315.50 +1.4% 320.00 vm-scalability.time.percent_of_cpu_this_job_got
854.51 +2.3% 874.06 vm-scalability.time.system_time
103.13 -6.1% 96.80 vm-scalability.time.user_time
1.448e+09 -5.3% 1.371e+09 vm-scalability.workload
97466 ± 3% -19.6% 78394 ± 18% softirqs.CPU3.RCU
735.00 ± 9% -18.2% 601.33 ± 3% slabinfo.kmalloc-rcl-96.active_objs
735.00 ± 9% -18.2% 601.33 ± 3% slabinfo.kmalloc-rcl-96.num_objs
7589 +1.9% 7732 interrupts.CAL:Function_call_interrupts
756810 ± 6% -10.5% 677379 interrupts.CPU2.LOC:Local_timer_interrupts
2172 ± 13% -36.1% 1387 ± 28% interrupts.CPU3.CAL:Function_call_interrupts
701245 ± 6% +20.1% 842502 ± 12% interrupts.CPU3.LOC:Local_timer_interrupts
-17569 -72.7% -4788 sched_debug.cfs_rq:/.spread0.min
807.29 ± 3% +11.4% 899.61 ± 2% sched_debug.cfs_rq:/.util_avg.min
172.38 ± 15% -35.8% 110.68 ± 16% sched_debug.cfs_rq:/.util_avg.stddev
186.29 ± 5% -4.8% 177.39 ± 3% sched_debug.cpu.cpu_load[3].min
3.96 ± 14% -24.2% 3.00 sched_debug.cpu.nr_running.max
1.17 ± 19% -35.4% 0.76 sched_debug.cpu.nr_running.stddev
9.50 ± 39% +175.4% 26.17 ± 44% sched_debug.cpu.nr_uninterruptible.max
-12.37 +125.4% -27.89 sched_debug.cpu.nr_uninterruptible.min
8.35 ± 24% +154.7% 21.27 ± 41% sched_debug.cpu.nr_uninterruptible.stddev
5224 ± 4% -26.6% 3835 ± 13% proc-vmstat.compact_daemon_wake
18209 ± 4% +11.0% 20204 proc-vmstat.kswapd_low_wmark_hit_quickly
3.624e+08 -5.3% 3.431e+08 proc-vmstat.numa_hit
3.624e+08 -5.3% 3.431e+08 proc-vmstat.numa_local
18216 ± 4% +11.0% 20216 proc-vmstat.pageoutrun
1.923e+08 -5.5% 1.818e+08 proc-vmstat.pgalloc_dma32
1.704e+08 -5.2% 1.615e+08 proc-vmstat.pgalloc_normal
3.628e+08 -5.3% 3.435e+08 proc-vmstat.pgfree
3.604e+08 -5.4% 3.411e+08 proc-vmstat.pgscan_kswapd
3.604e+08 -5.4% 3.411e+08 proc-vmstat.pgsteal_kswapd
5729332 -5.9% 5391641 proc-vmstat.slabs_scanned
2593384 -6.0% 2437353 proc-vmstat.workingset_nodereclaim
32.26 ± 7% -4.1 28.21 ± 5% perf-profile.calltrace.cycles-pp.ondemand_readahead.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
32.20 ± 7% -4.0 28.17 ± 5% perf-profile.calltrace.cycles-pp.__do_page_cache_readahead.ondemand_readahead.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter
23.30 ± 8% -3.0 20.35 ± 5% perf-profile.calltrace.cycles-pp.read_pages.__do_page_cache_readahead.ondemand_readahead.generic_file_read_iter.xfs_file_buffered_aio_read
23.22 ± 8% -2.9 20.27 ± 5% perf-profile.calltrace.cycles-pp.iomap_readpages.read_pages.__do_page_cache_readahead.ondemand_readahead.generic_file_read_iter
23.16 ± 8% -2.9 20.23 ± 5% perf-profile.calltrace.cycles-pp.iomap_apply.iomap_readpages.read_pages.__do_page_cache_readahead.ondemand_readahead
22.79 ± 8% -2.9 19.87 ± 5% perf-profile.calltrace.cycles-pp.iomap_readpages_actor.iomap_apply.iomap_readpages.read_pages.__do_page_cache_readahead
10.41 ± 8% -1.4 8.97 ± 2% perf-profile.calltrace.cycles-pp.iomap_readpage_actor.iomap_readpages_actor.iomap_apply.iomap_readpages.read_pages
11.40 ± 8% -1.3 10.08 ± 7% perf-profile.calltrace.cycles-pp.add_to_page_cache_lru.iomap_readpages_actor.iomap_apply.iomap_readpages.read_pages
7.92 ± 8% -1.1 6.82 ± 2% perf-profile.calltrace.cycles-pp.memset_erms.iomap_readpage_actor.iomap_readpages_actor.iomap_apply.iomap_readpages
6.88 ± 7% -0.7 6.15 ± 8% perf-profile.calltrace.cycles-pp.__add_to_page_cache_locked.add_to_page_cache_lru.iomap_readpages_actor.iomap_apply.iomap_readpages
6.16 ± 8% -0.6 5.53 ± 2% perf-profile.calltrace.cycles-pp.__alloc_pages_nodemask.__do_page_cache_readahead.ondemand_readahead.generic_file_read_iter.xfs_file_buffered_aio_read
3.75 ± 8% -0.4 3.32 ± 7% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64.read
2.96 ± 11% -0.4 2.57 ± 5% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.write
2.97 ± 6% -0.4 2.58 ± 8% perf-profile.calltrace.cycles-pp.syscall_return_via_sysret.read
1.49 ± 11% -0.3 1.19 ± 2% perf-profile.calltrace.cycles-pp.iomap_set_range_uptodate.iomap_readpage_actor.iomap_readpages_actor.iomap_apply.iomap_readpages
0.63 ± 4% -0.3 0.36 ± 70% perf-profile.calltrace.cycles-pp.mark_page_accessed.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
1.24 ± 9% -0.3 0.98 ± 4% perf-profile.calltrace.cycles-pp.__list_del_entry_valid.get_page_from_freelist.__alloc_pages_nodemask.__do_page_cache_readahead.ondemand_readahead
0.62 ± 7% -0.2 0.38 ± 71% perf-profile.calltrace.cycles-pp.__fget_light.__fdget_pos.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
1.76 ± 6% -0.2 1.52 ± 11% perf-profile.calltrace.cycles-pp.selinux_file_permission.security_file_permission.vfs_write.ksys_write.do_syscall_64
1.65 ± 4% -0.2 1.42 ± 7% perf-profile.calltrace.cycles-pp.xa_load.__do_page_cache_readahead.ondemand_readahead.generic_file_read_iter.xfs_file_buffered_aio_read
1.46 ± 4% -0.2 1.24 ± 6% perf-profile.calltrace.cycles-pp.xas_load.xa_load.__do_page_cache_readahead.ondemand_readahead.generic_file_read_iter
0.73 ± 9% -0.1 0.63 ± 7% perf-profile.calltrace.cycles-pp._raw_spin_lock_irq.__add_to_page_cache_locked.add_to_page_cache_lru.iomap_readpages_actor.iomap_apply
0.58 ± 9% +0.4 0.97 ± 13% perf-profile.calltrace.cycles-pp.current_time.atime_needs_update.touch_atime.generic_file_read_iter.xfs_file_buffered_aio_read
0.93 ± 6% +0.7 1.63 ± 12% perf-profile.calltrace.cycles-pp.atime_needs_update.touch_atime.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter
2.69 ± 6% +0.8 3.51 ± 7% perf-profile.calltrace.cycles-pp.xas_load.find_get_entry.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read
1.09 ± 6% +0.9 1.94 ± 13% perf-profile.calltrace.cycles-pp.touch_atime.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
4.15 ± 6% +1.2 5.40 ± 7% perf-profile.calltrace.cycles-pp.find_get_entry.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter
4.29 ± 6% +1.4 5.69 ± 8% perf-profile.calltrace.cycles-pp.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
17.08 -9.4% 15.47 ± 2% perf-stat.i.MPKI
1.638e+09 +4.2% 1.706e+09 perf-stat.i.branch-instructions
0.94 ± 2% -0.1 0.86 perf-stat.i.branch-miss-rate%
15189313 ± 3% -4.3% 14541135 perf-stat.i.branch-misses
60.39 -0.4 59.95 perf-stat.i.cache-miss-rate%
84514861 -6.2% 79245867 ± 2% perf-stat.i.cache-misses
1.393e+08 -5.7% 1.314e+08 perf-stat.i.cache-references
1.56 -3.8% 1.50 perf-stat.i.cpi
76.62 -2.8% 74.44 perf-stat.i.cpu-migrations
166.97 +7.7% 179.74 perf-stat.i.cycles-between-cache-misses
1.55 +0.2 1.71 perf-stat.i.dTLB-load-miss-rate%
40318360 +13.9% 45905767 perf-stat.i.dTLB-load-misses
2.554e+09 +3.0% 2.632e+09 perf-stat.i.dTLB-loads
0.07 ± 2% +0.0 0.08 ± 6% perf-stat.i.dTLB-store-miss-rate%
1413447 ± 2% +14.8% 1622498 ± 5% perf-stat.i.dTLB-store-misses
8.311e+09 +3.8% 8.629e+09 perf-stat.i.instructions
0.65 +3.7% 0.67 perf-stat.i.ipc
16.76 -9.2% 15.23 ± 2% perf-stat.overall.MPKI
0.93 ± 3% -0.1 0.85 perf-stat.overall.branch-miss-rate%
60.65 -0.3 60.31 perf-stat.overall.cache-miss-rate%
1.55 -3.7% 1.49 perf-stat.overall.cpi
152.55 +6.7% 162.72 perf-stat.overall.cycles-between-cache-misses
1.55 +0.2 1.71 perf-stat.overall.dTLB-load-miss-rate%
0.07 ± 2% +0.0 0.08 ± 6% perf-stat.overall.dTLB-store-miss-rate%
0.64 +3.8% 0.67 perf-stat.overall.ipc
1744 +9.6% 1911 perf-stat.overall.path-length
1.633e+09 +4.2% 1.702e+09 perf-stat.ps.branch-instructions
15141912 ± 3% -4.2% 14504616 perf-stat.ps.branch-misses
84251986 -6.2% 79053575 perf-stat.ps.cache-misses
1.389e+08 -5.6% 1.311e+08 perf-stat.ps.cache-references
76.39 -2.8% 74.26 perf-stat.ps.cpu-migrations
40194141 +13.9% 45794346 perf-stat.ps.dTLB-load-misses
2.547e+09 +3.1% 2.625e+09 perf-stat.ps.dTLB-loads
1409206 ± 2% +14.9% 1618569 ± 5% perf-stat.ps.dTLB-store-misses
8.286e+09 +3.9% 8.608e+09 perf-stat.ps.instructions
2.525e+12 +3.8% 2.621e+12 perf-stat.total.instructions***************************************************************************************************
lkp-ivb-ep01: 40 threads Intel(R) Xeon(R) CPU E5-2690 v2 @ 3.00GHz with 384G memory
=========================================================================================
compiler/cpufreq_governor/disk/fs/kconfig/load/md/rootfs/tbox_group/test/testcase:
gcc-7/performance/4BRD_12G/xfs/x86_64-rhel-7.6/3000/RAID1/debian-x86_64-2018-04-03.cgz/lkp-ivb-ep01/disk_cp/aim7
commit:
8e22ba96d4 (" RISC-V Patches for 5.1-rc4")
fa3fe73bed ("xfs: reduce ilock contention on buffered randrw workload")
8e22ba96d44c4ad5 fa3fe73bed17967cb623461b166
---------------- ---------------------------
fail:runs %reproduction fail:runs
| | |
1:4 -25% :4 dmesg.WARNING:at#for_ip_interrupt_entry/0x
%stddev %change %stddev
\ | \
230947 -79.4% 47688 aim7.jobs-per-min
78.62 +380.5% 377.76 aim7.time.elapsed_time
78.62 +380.5% 377.76 aim7.time.elapsed_time.max
89921 +604.3% 633289 ± 2% aim7.time.involuntary_context_switches
150384 ± 2% +35.4% 203653 aim7.time.minor_page_faults
2333 +518.7% 14439 aim7.time.system_time
102.60 -16.9% 85.26 aim7.time.user_time
715299 -32.7% 481312 aim7.time.voluntary_context_switches
31.39 +4.6% 32.84 boot-time.boot
24.75 +5.8% 26.17 boot-time.dhcp
1073 +3.8% 1114 boot-time.idle
23.78 ± 3% -19.7 4.03 ± 2% mpstat.cpu.all.idle%
0.03 ± 40% -0.0 0.00 ± 36% mpstat.cpu.all.iowait%
72.86 +22.5 95.34 mpstat.cpu.all.sys%
3.31 -2.7 0.62 mpstat.cpu.all.usr%
37946520 ± 4% +15.8% 43936831 numa-numastat.node0.local_node
37954399 ± 4% +15.8% 43943353 numa-numastat.node0.numa_hit
39497938 ± 3% -13.6% 34121982 numa-numastat.node1.local_node
39503258 ± 3% -13.6% 34128579 numa-numastat.node1.numa_hit
25.67 ± 5% -82.4% 4.53 ± 2% iostat.cpu.idle
71.06 +33.5% 94.85 iostat.cpu.system
3.25 ± 2% -80.9% 0.62 iostat.cpu.user
71.45 ± 13% -71.0% 20.71 ± 8% iostat.md0.w/s
2598 ± 35% -84.7% 397.41 ± 21% iostat.md0.wkB/s
52331183 ± 8% -88.0% 6278911 ± 8% cpuidle.C1.time
469089 ± 6% -85.8% 66489 ± 11% cpuidle.C1.usage
82837567 ± 43% -92.6% 6161783 ± 26% cpuidle.C1E.time
590943 ± 43% -92.7% 43113 ± 29% cpuidle.C1E.usage
5.515e+08 ± 14% -24.1% 4.185e+08 cpuidle.C6.time
15970 ± 4% -54.4% 7275 ± 10% cpuidle.POLL.time
25.50 ± 5% -84.3% 4.00 vmstat.cpu.id
70.50 ± 2% +33.3% 94.00 vmstat.cpu.sy
2597 ± 35% -84.7% 397.50 ± 21% vmstat.io.bo
30.75 +64.2% 50.50 vmstat.procs.r
18635 -76.3% 4421 vmstat.system.cs
82704 -1.5% 81496 vmstat.system.in
1160 +11.5% 1294 meminfo.Active(file)
79489 ± 3% +124.5% 178419 meminfo.AnonHugePages
5776899 ± 14% +21.9% 7039627 meminfo.DirectMap2M
206267 ± 8% +22.6% 252858 ± 14% meminfo.DirectMap4k
16915 ± 6% +15.8% 19586 meminfo.Dirty
64.50 ±135% +1520.9% 1045 ± 66% meminfo.Mlocked
37100 ± 2% +25.7% 46652 ± 4% meminfo.Shmem
34867 ± 4% -79.0% 7314 meminfo.max_used_kB
158732 ± 12% +29.9% 206208 ± 4% numa-meminfo.node0.Active
158381 ± 12% +30.0% 205908 ± 4% numa-meminfo.node0.Active(anon)
47728 ± 57% +141.7% 115362 ± 14% numa-meminfo.node0.AnonHugePages
153983 ± 13% +25.7% 193516 ± 10% numa-meminfo.node0.AnonPages
8151 ± 7% +21.4% 9894 numa-meminfo.node0.Dirty
8140 ± 5% +21.8% 9912 numa-meminfo.node0.Inactive(file)
32023 ± 83% +97.1% 63131 ± 25% numa-meminfo.node1.AnonHugePages
8247 ± 5% +16.4% 9602 numa-meminfo.node1.Dirty
35.00 ±146% +1547.1% 576.50 ± 74% numa-meminfo.node1.Mlocked
39629 ± 12% +29.9% 51479 ± 4% numa-vmstat.node0.nr_active_anon
38530 ± 13% +25.6% 48375 ± 10% numa-vmstat.node0.nr_anon_pages
15978818 ± 6% +33.2% 21290840 numa-vmstat.node0.nr_dirtied
1995 ± 2% +25.1% 2495 numa-vmstat.node0.nr_dirty
2032 +23.1% 2502 numa-vmstat.node0.nr_inactive_file
6.50 ±119% +1700.0% 117.00 ± 56% numa-vmstat.node0.nr_mlock
39629 ± 12% +29.9% 51479 ± 4% numa-vmstat.node0.nr_zone_active_anon
2030 +23.2% 2501 numa-vmstat.node0.nr_zone_inactive_file
1755 ± 4% +35.2% 2374 numa-vmstat.node0.nr_zone_write_pending
16510488 ± 6% +33.3% 22001420 numa-vmstat.node0.numa_hit
16502752 ± 6% +32.6% 21879456 numa-vmstat.node0.numa_local
7743 ± 40% +1475.2% 121965 ± 35% numa-vmstat.node0.numa_other
2061 ± 3% +18.3% 2439 numa-vmstat.node1.nr_dirty
7.75 ±144% +1758.1% 144.00 ± 74% numa-vmstat.node1.nr_mlock
1860 ± 3% +23.5% 2297 numa-vmstat.node1.nr_zone_write_pending
196119 -58.3% 81767 ± 52% numa-vmstat.node1.numa_other
2223 +37.3% 3051 turbostat.Avg_MHz
76.70 +19.4 96.06 turbostat.Busy%
464516 ± 6% -86.6% 62290 ± 10% turbostat.C1
1.61 ± 7% -1.6 0.04 turbostat.C1%
590696 ± 43% -92.8% 42693 ± 29% turbostat.C1E
2.55 ± 42% -2.5 0.04 ± 25% turbostat.C1E%
16.96 ± 12% -14.2 2.75 turbostat.C6%
12.86 ± 3% -84.1% 2.04 ± 3% turbostat.CPU%c1
9.76 ± 17% -84.9% 1.48 ± 4% turbostat.CPU%c6
119.72 +31.4% 157.31 turbostat.CorWatt
6809310 +355.8% 31038009 turbostat.IRQ
6.33 ± 17% -82.5% 1.11 ± 2% turbostat.Pkg%pc2
1.33 ± 47% -86.8% 0.17 ± 2% turbostat.Pkg%pc6
147.27 +27.0% 186.99 turbostat.PkgWatt
37.69 +28.6% 48.47 turbostat.RAMWatt
6460 +191.0% 18800 turbostat.SMI
89233 +7.5% 95956 proc-vmstat.nr_active_anon
287.25 +12.4% 323.00 proc-vmstat.nr_active_file
84940 +5.2% 89315 proc-vmstat.nr_anon_pages
4106 ± 5% +18.3% 4858 proc-vmstat.nr_dirty
327571 +1.0% 330803 proc-vmstat.nr_file_pages
5077 -1.9% 4979 proc-vmstat.nr_inactive_anon
4144 ± 5% +17.0% 4850 proc-vmstat.nr_inactive_file
53995 +3.3% 55798 proc-vmstat.nr_kernel_stack
6170 -2.4% 6019 proc-vmstat.nr_mapped
16.00 ±136% +1531.2% 261.00 ± 66% proc-vmstat.nr_mlock
9279 ± 2% +25.7% 11667 ± 4% proc-vmstat.nr_shmem
21055 +1.8% 21423 proc-vmstat.nr_slab_reclaimable
37405 +2.2% 38234 proc-vmstat.nr_slab_unreclaimable
89233 +7.5% 95956 proc-vmstat.nr_zone_active_anon
287.25 +12.4% 323.00 proc-vmstat.nr_zone_active_file
5077 -1.9% 4979 proc-vmstat.nr_zone_inactive_anon
4144 ± 5% +17.0% 4850 proc-vmstat.nr_zone_inactive_file
3554 ± 4% +29.3% 4594 proc-vmstat.nr_zone_write_pending
719.75 ±149% +8190.4% 59670 ± 4% proc-vmstat.numa_hint_faults
615.75 ±172% +5185.6% 32546 ± 9% proc-vmstat.numa_hint_faults_local
4064 ±116% +652.2% 30570 ± 12% proc-vmstat.numa_pages_migrated
15803 ± 96% +462.5% 88891 ± 4% proc-vmstat.numa_pte_updates
4867 ± 6% +15.7% 5633 ± 9% proc-vmstat.pgactivate
383704 ± 2% +202.9% 1162233 proc-vmstat.pgfault
4064 ±116% +652.2% 30570 ± 12% proc-vmstat.pgmigrate_success
1648 ± 2% +4.7% 1725 proc-vmstat.unevictable_pgs_culled
2583 ± 2% -30.4% 1798 slabinfo.dmaengine-unmap-16.active_objs
2586 ± 2% -30.5% 1798 slabinfo.dmaengine-unmap-16.num_objs
2018 ± 3% +17.4% 2370 slabinfo.ebitmap_node.active_objs
2018 ± 3% +17.4% 2370 slabinfo.ebitmap_node.num_objs
2435 +67.7% 4084 ± 2% slabinfo.kmalloc-128.active_objs
2435 +67.7% 4084 ± 2% slabinfo.kmalloc-128.num_objs
4185 ± 3% -12.1% 3680 slabinfo.kmalloc-192.active_objs
4191 ± 3% -11.5% 3711 ± 2% slabinfo.kmalloc-192.num_objs
5502 +15.7% 6368 slabinfo.kmalloc-512.active_objs
7072 ± 2% +15.4% 8164 slabinfo.kmalloc-96.active_objs
7203 +13.9% 8205 slabinfo.kmalloc-96.num_objs
12686 ± 7% +31.9% 16736 ± 9% slabinfo.lsm_file_cache.active_objs
12686 ± 7% +31.9% 16736 ± 9% slabinfo.lsm_file_cache.num_objs
2524 +13.3% 2860 slabinfo.names_cache.active_objs
2574 +14.0% 2933 slabinfo.names_cache.num_objs
118.75 ± 30% +134.1% 278.00 ± 8% slabinfo.nfs_commit_data.active_objs
118.75 ± 30% +134.1% 278.00 ± 8% slabinfo.nfs_commit_data.num_objs
5332 ± 3% -26.0% 3945 ± 5% slabinfo.pool_workqueue.active_objs
5347 ± 2% -26.2% 3946 ± 5% slabinfo.pool_workqueue.num_objs
1244 +12.3% 1396 slabinfo.xfs_btree_cur.active_objs
1244 +12.3% 1396 slabinfo.xfs_btree_cur.num_objs
1039 +13.1% 1176 slabinfo.xfs_buf_item.active_objs
1039 +13.1% 1176 slabinfo.xfs_buf_item.num_objs
1140 +13.0% 1289 slabinfo.xfs_da_state.active_objs
1140 +13.0% 1289 slabinfo.xfs_da_state.num_objs
1144 ± 3% +18.4% 1355 slabinfo.xfs_efd_item.active_objs
1144 ± 3% +18.4% 1355 slabinfo.xfs_efd_item.num_objs
2140 ± 2% -36.6% 1357 slabinfo.xfs_inode.active_objs
2223 -39.0% 1357 slabinfo.xfs_inode.num_objs
24.48 -19.9 4.58 ± 4% perf-profile.calltrace.cycles-pp.write
21.42 -17.4 4.04 ± 4% perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.write
21.28 -17.3 4.02 ± 4% perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
20.66 -16.7 3.93 ± 4% perf-profile.calltrace.cycles-pp.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
20.19 -16.3 3.84 ± 4% perf-profile.calltrace.cycles-pp.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe.write
18.01 -14.5 3.50 ± 3% perf-profile.calltrace.cycles-pp.new_sync_write.vfs_write.ksys_write.do_syscall_64.entry_SYSCALL_64_after_hwframe
17.11 -13.8 3.36 ± 4% perf-profile.calltrace.cycles-pp.xfs_file_buffered_aio_write.new_sync_write.vfs_write.ksys_write.do_syscall_64
13.64 -11.1 2.55 ± 3% perf-profile.calltrace.cycles-pp.iomap_file_buffered_write.xfs_file_buffered_aio_write.new_sync_write.vfs_write.ksys_write
13.56 -11.0 2.54 ± 3% perf-profile.calltrace.cycles-pp.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write.new_sync_write.vfs_write
11.81 -11.0 0.82 ± 7% perf-profile.calltrace.cycles-pp.touch_atime.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
11.72 -10.9 0.79 ± 8% perf-profile.calltrace.cycles-pp.atime_needs_update.touch_atime.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter
10.27 -8.4 1.91 ± 4% perf-profile.calltrace.cycles-pp.iomap_write_actor.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write.new_sync_write
8.05 -7.5 0.58 ± 5% perf-profile.calltrace.cycles-pp.xfs_ilock.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read.vfs_read
7.99 -7.4 0.58 ± 5% perf-profile.calltrace.cycles-pp.down_read.xfs_ilock.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
7.27 ± 4% -7.3 0.00 perf-profile.calltrace.cycles-pp.security_file_permission.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
6.16 ± 2% -5.9 0.27 ±100% perf-profile.calltrace.cycles-pp.up_read.xfs_iunlock.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
6.21 ± 2% -5.7 0.52 ± 3% perf-profile.calltrace.cycles-pp.xfs_iunlock.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read.vfs_read
5.04 ± 4% -5.0 0.00 perf-profile.calltrace.cycles-pp.selinux_file_permission.security_file_permission.vfs_read.ksys_read.do_syscall_64
5.42 -4.4 1.01 ± 4% perf-profile.calltrace.cycles-pp.iomap_write_begin.iomap_write_actor.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write
4.12 -3.3 0.78 ± 7% perf-profile.calltrace.cycles-pp.grab_cache_page_write_begin.iomap_write_begin.iomap_write_actor.iomap_apply.iomap_file_buffered_write
3.95 -3.2 0.73 ± 6% perf-profile.calltrace.cycles-pp.pagecache_get_page.grab_cache_page_write_begin.iomap_write_begin.iomap_write_actor.iomap_apply
2.40 ± 3% -1.9 0.51 perf-profile.calltrace.cycles-pp.xfs_file_iomap_begin.iomap_apply.iomap_file_buffered_write.xfs_file_buffered_aio_write.new_sync_write
2.39 -1.8 0.58 ± 6% perf-profile.calltrace.cycles-pp.xfs_file_aio_write_checks.xfs_file_buffered_aio_write.new_sync_write.vfs_write.ksys_write
1.64 -0.4 1.25 ± 2% perf-profile.calltrace.cycles-pp.copy_page_to_iter.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
0.56 ± 7% +1.5 2.02 perf-profile.calltrace.cycles-pp.xas_start.xas_load.find_get_entry.pagecache_get_page.generic_file_read_iter
0.00 +1.8 1.77 ± 4% perf-profile.calltrace.cycles-pp.___might_sleep.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
0.75 ± 6% +3.4 4.12 perf-profile.calltrace.cycles-pp.xas_load.find_get_entry.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read
0.00 +6.6 6.58 ± 4% perf-profile.calltrace.cycles-pp.mark_page_accessed.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
66.40 +28.0 94.41 perf-profile.calltrace.cycles-pp.read
62.72 +31.1 93.86 perf-profile.calltrace.cycles-pp.entry_SYSCALL_64_after_hwframe.read
62.59 +31.2 93.83 perf-profile.calltrace.cycles-pp.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
61.99 +31.7 93.72 perf-profile.calltrace.cycles-pp.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
61.50 +32.1 93.65 perf-profile.calltrace.cycles-pp.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe.read
49.59 +43.5 93.05 perf-profile.calltrace.cycles-pp.new_sync_read.vfs_read.ksys_read.do_syscall_64.entry_SYSCALL_64_after_hwframe
2.29 ± 3% +47.1 49.36 perf-profile.calltrace.cycles-pp.find_get_entry.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter
2.34 ± 3% +47.8 50.17 perf-profile.calltrace.cycles-pp.pagecache_get_page.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read
41.75 +51.0 92.78 perf-profile.calltrace.cycles-pp.xfs_file_read_iter.new_sync_read.vfs_read.ksys_read.do_syscall_64
37.36 +55.3 92.64 perf-profile.calltrace.cycles-pp.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read.vfs_read.ksys_read
23.02 +68.2 91.25 perf-profile.calltrace.cycles-pp.generic_file_read_iter.xfs_file_buffered_aio_read.xfs_file_read_iter.new_sync_read.vfs_read
16.69 ± 4% -37.3% 10.47 perf-stat.i.MPKI
4.871e+09 +42.2% 6.925e+09 perf-stat.i.branch-instructions
2.68 ± 8% -2.0 0.66 ± 4% perf-stat.i.branch-miss-rate%
66135811 ± 2% -55.2% 29604797 perf-stat.i.branch-misses
12.56 ± 3% +19.2 31.78 perf-stat.i.cache-miss-rate%
28607414 +253.7% 1.012e+08 perf-stat.i.cache-misses
2.693e+08 +17.3% 3.16e+08 perf-stat.i.cache-references
19130 -76.9% 4419 perf-stat.i.context-switches
4.00 -6.7% 3.73 perf-stat.i.cpi
8.942e+10 +36.0% 1.216e+11 perf-stat.i.cpu-cycles
2280 -66.5% 764.28 perf-stat.i.cpu-migrations
2819 ± 2% -57.8% 1189 ± 2% perf-stat.i.cycles-between-cache-misses
1.91 ± 6% -0.6 1.26 ± 12% perf-stat.i.dTLB-load-miss-rate%
1.636e+08 ± 7% -28.5% 1.17e+08 ± 14% perf-stat.i.dTLB-load-misses
7.678e+09 +17.2% 9.002e+09 perf-stat.i.dTLB-loads
5.022e+09 +2.3% 5.138e+09 perf-stat.i.dTLB-stores
8198957 ± 4% -75.3% 2023647 ± 2% perf-stat.i.iTLB-load-misses
1570759 ± 18% -75.7% 381261 ± 48% perf-stat.i.iTLB-loads
2.463e+10 +33.6% 3.29e+10 perf-stat.i.instructions
2775 ± 3% +479.2% 16074 perf-stat.i.instructions-per-iTLB-miss
4535 -34.2% 2986 perf-stat.i.minor-faults
46.43 +0.7 47.17 perf-stat.i.node-load-miss-rate%
15958314 +262.7% 57878361 ± 3% perf-stat.i.node-load-misses
18031706 +255.5% 64100733 ± 2% perf-stat.i.node-loads
7699378 +268.1% 28341867 perf-stat.i.node-store-misses
9555974 +292.0% 37461477 perf-stat.i.node-stores
4535 -34.2% 2986 perf-stat.i.page-faults
10.93 -12.2% 9.60 perf-stat.overall.MPKI
1.36 -0.9 0.43 perf-stat.overall.branch-miss-rate%
10.62 +21.4 32.02 perf-stat.overall.cache-miss-rate%
3.63 +1.9% 3.70 perf-stat.overall.cpi
3125 -61.5% 1202 perf-stat.overall.cycles-between-cache-misses
2.09 ± 6% -0.8 1.28 ± 12% perf-stat.overall.dTLB-load-miss-rate%
3011 ± 4% +440.1% 16262 perf-stat.overall.instructions-per-iTLB-miss
44.62 -1.5 43.07 perf-stat.overall.node-store-miss-rate%
4.808e+09 +43.6% 6.906e+09 perf-stat.ps.branch-instructions
65286911 ± 2% -54.8% 29529411 perf-stat.ps.branch-misses
28238780 +257.3% 1.009e+08 perf-stat.ps.cache-misses
2.658e+08 +18.5% 3.151e+08 perf-stat.ps.cache-references
18882 -76.7% 4407 perf-stat.ps.context-switches
8.825e+10 +37.5% 1.213e+11 perf-stat.ps.cpu-cycles
2250 -66.1% 762.20 perf-stat.ps.cpu-migrations
1.615e+08 ± 7% -27.7% 1.167e+08 ± 14% perf-stat.ps.dTLB-load-misses
7.579e+09 +18.5% 8.977e+09 perf-stat.ps.dTLB-loads
4.957e+09 +3.4% 5.124e+09 perf-stat.ps.dTLB-stores
8092496 ± 4% -75.1% 2018134 ± 2% perf-stat.ps.iTLB-load-misses
1550345 ± 18% -75.5% 380213 ± 48% perf-stat.ps.iTLB-loads
2.431e+10 +34.9% 3.281e+10 perf-stat.ps.instructions
4485 -33.6% 2979 perf-stat.ps.minor-faults
78985 +1.0% 79787 perf-stat.ps.msec
15750551 +266.5% 57718276 ± 3% perf-stat.ps.node-load-misses
17797012 +259.2% 63923411 ± 2% perf-stat.ps.node-loads
7599430 +271.9% 28263476 perf-stat.ps.node-store-misses
9432838 +296.0% 37357987 perf-stat.ps.node-stores
4485 -33.6% 2979 perf-stat.ps.page-faults
1.944e+12 +539.2% 1.243e+13 perf-stat.total.instructions
11300 ± 4% +37.0% 15483 softirqs.CPU0.RCU
38205 ± 2% +340.7% 168362 softirqs.CPU0.TIMER
10724 ± 6% +63.0% 17483 ± 17% softirqs.CPU1.RCU
33429 ± 3% +448.7% 183422 ± 16% softirqs.CPU1.TIMER
11440 ± 5% +68.7% 19299 ± 7% softirqs.CPU10.RCU
35285 ± 3% +386.2% 171569 ± 5% softirqs.CPU10.TIMER
10439 ± 5% +25.1% 13064 softirqs.CPU11.RCU
34696 ± 3% +352.7% 157067 ± 4% softirqs.CPU11.TIMER
39317 ± 6% +317.6% 164203 ± 2% softirqs.CPU12.TIMER
11453 ± 11% +13.8% 13037 ± 5% softirqs.CPU13.RCU
35175 ± 4% +342.5% 155666 ± 5% softirqs.CPU13.TIMER
10594 ± 3% +38.0% 14624 softirqs.CPU14.RCU
8846 ± 2% +8.5% 9602 ± 4% softirqs.CPU14.SCHED
36407 +347.0% 162753 softirqs.CPU14.TIMER
11327 ± 11% +17.1% 13267 softirqs.CPU15.RCU
35261 ± 4% +356.7% 161037 ± 4% softirqs.CPU15.TIMER
37699 ± 6% +338.9% 165465 ± 2% softirqs.CPU16.TIMER
10542 ± 3% +33.3% 14055 ± 7% softirqs.CPU17.RCU
35568 ± 4% +352.3% 160870 ± 3% softirqs.CPU17.TIMER
10503 ± 9% +45.2% 15246 ± 10% softirqs.CPU18.RCU
8824 ± 3% +9.3% 9641 ± 4% softirqs.CPU18.SCHED
36329 +345.0% 161670 softirqs.CPU18.TIMER
10560 ± 9% +20.9% 12770 ± 2% softirqs.CPU19.RCU
35217 ± 4% +348.3% 157892 ± 4% softirqs.CPU19.TIMER
11935 ± 6% +25.0% 14913 ± 2% softirqs.CPU2.RCU
40085 ± 21% +308.5% 163727 softirqs.CPU2.TIMER
10580 ± 6% +33.3% 14108 ± 2% softirqs.CPU20.RCU
35607 ± 3% +367.2% 166363 softirqs.CPU20.TIMER
10779 ± 5% +18.8% 12808 softirqs.CPU21.RCU
34607 +390.2% 169655 ± 8% softirqs.CPU21.TIMER
10214 ± 3% +42.6% 14566 softirqs.CPU22.RCU
35941 +362.4% 166191 softirqs.CPU22.TIMER
10428 ± 4% +22.4% 12766 ± 2% softirqs.CPU23.RCU
36565 ± 6% +337.8% 160071 ± 4% softirqs.CPU23.TIMER
11727 ± 10% +31.8% 15459 ± 7% softirqs.CPU24.RCU
37658 ± 8% +342.1% 166490 softirqs.CPU24.TIMER
10509 ± 3% +23.4% 12967 ± 3% softirqs.CPU25.RCU
35179 ± 2% +359.4% 161626 ± 4% softirqs.CPU25.TIMER
10700 ± 2% +35.0% 14446 ± 3% softirqs.CPU26.RCU
36295 ± 3% +353.5% 164615 softirqs.CPU26.TIMER
11108 ± 5% +17.9% 13096 softirqs.CPU27.RCU
36245 ± 6% +344.5% 161123 ± 3% softirqs.CPU27.TIMER
10762 ± 5% +37.0% 14746 ± 2% softirqs.CPU28.RCU
35529 ± 2% +359.1% 163130 softirqs.CPU28.TIMER
34944 ± 3% +348.8% 156849 ± 4% softirqs.CPU29.TIMER
10712 ± 7% +27.0% 13602 softirqs.CPU3.RCU
38235 ± 15% +318.8% 160146 ± 4% softirqs.CPU3.TIMER
11682 ± 9% +48.6% 17357 ± 5% softirqs.CPU30.RCU
39787 ± 26% +358.7% 182513 ± 11% softirqs.CPU30.TIMER
11089 ± 13% +23.1% 13652 ± 6% softirqs.CPU31.RCU
35040 ± 5% +346.6% 156483 ± 5% softirqs.CPU31.TIMER
37864 ± 5% +333.5% 164148 ± 2% softirqs.CPU32.TIMER
34838 ± 4% +346.1% 155416 ± 5% softirqs.CPU33.TIMER
9059 ± 11% +28.6% 11654 ± 5% softirqs.CPU34.RCU
36135 +352.8% 163637 softirqs.CPU34.TIMER
9929 ± 13% +20.5% 11964 ± 6% softirqs.CPU35.RCU
35253 ± 3% +355.4% 160552 ± 4% softirqs.CPU35.TIMER
9812 ± 7% +28.1% 12569 ± 10% softirqs.CPU36.RCU
36233 +356.0% 165213 ± 2% softirqs.CPU36.TIMER
8617 ± 6% +19.9% 10331 softirqs.CPU37.RCU
34947 ± 3% +360.5% 160950 ± 2% softirqs.CPU37.TIMER
9614 ± 14% +33.1% 12801 ± 8% softirqs.CPU38.RCU
36368 +344.5% 161670 softirqs.CPU38.TIMER
35076 ± 4% +349.6% 157695 ± 4% softirqs.CPU39.TIMER
11551 ± 11% +28.1% 14798 softirqs.CPU4.RCU
36610 +355.5% 166772 softirqs.CPU4.TIMER
35479 ± 2% +355.4% 161573 ± 4% softirqs.CPU5.TIMER
10648 ± 2% +39.9% 14894 softirqs.CPU6.RCU
36168 ± 2% +354.4% 164361 softirqs.CPU6.TIMER
12361 ± 8% +18.1% 14604 ± 5% softirqs.CPU7.RCU
41756 ± 27% +283.5% 160130 ± 3% softirqs.CPU7.TIMER
11070 ± 8% +42.1% 15727 ± 4% softirqs.CPU8.RCU
35520 ± 2% +359.7% 163300 softirqs.CPU8.TIMER
10999 ± 8% +19.7% 13168 softirqs.CPU9.RCU
35178 ± 3% +346.3% 157017 ± 5% softirqs.CPU9.TIMER
431390 ± 2% +28.1% 552492 softirqs.RCU
1451755 ± 2% +349.9% 6531407 ± 2% softirqs.TIMER
17335 ± 8% +50.0% 25998 ± 2% sched_debug.cfs_rq:/.load.avg
49802 ± 5% +31.5% 65479 ± 9% sched_debug.cfs_rq:/.load.max
109.88 ± 25% -50.3% 54.58 ± 11% sched_debug.cfs_rq:/.load_avg.avg
787480 +761.6% 6784621 sched_debug.cfs_rq:/.min_vruntime.avg
880219 ± 4% +703.3% 7070715 sched_debug.cfs_rq:/.min_vruntime.max
770885 +747.1% 6530040 sched_debug.cfs_rq:/.min_vruntime.min
18301 ± 28% +830.8% 170356 ± 10% sched_debug.cfs_rq:/.min_vruntime.stddev
0.58 ± 8% +54.4% 0.89 sched_debug.cfs_rq:/.nr_running.avg
0.37 ± 7% -74.9% 0.09 ± 38% sched_debug.cfs_rq:/.nr_running.stddev
41.05 ± 33% -82.2% 7.29 sched_debug.cfs_rq:/.removed.load_avg.avg
510.88 -71.4% 146.29 sched_debug.cfs_rq:/.removed.load_avg.max
135.76 ± 15% -76.6% 31.78 sched_debug.cfs_rq:/.removed.load_avg.stddev
1893 ± 33% -82.2% 337.03 sched_debug.cfs_rq:/.removed.runnable_sum.avg
23609 -71.3% 6787 sched_debug.cfs_rq:/.removed.runnable_sum.max
6261 ± 15% -76.5% 1469 sched_debug.cfs_rq:/.removed.runnable_sum.stddev
10.72 ± 57% -76.9% 2.47 ± 34% sched_debug.cfs_rq:/.removed.util_avg.avg
174.75 ± 33% -68.1% 55.71 ± 32% sched_debug.cfs_rq:/.removed.util_avg.max
38.33 ± 43% -71.6% 10.87 ± 34% sched_debug.cfs_rq:/.removed.util_avg.stddev
13.12 ± 11% +59.8% 20.96 sched_debug.cfs_rq:/.runnable_load_avg.avg
39.25 ± 7% +56.1% 61.29 ± 3% sched_debug.cfs_rq:/.runnable_load_avg.max
9.49 ± 6% +22.5% 11.63 ± 3% sched_debug.cfs_rq:/.runnable_load_avg.stddev
15014 ± 7% +39.4% 20929 sched_debug.cfs_rq:/.runnable_weight.avg
36981 ± 7% +32.4% 48981 sched_debug.cfs_rq:/.runnable_weight.max
-655.34 -22775.6% 148603 ± 3% sched_debug.cfs_rq:/.spread0.avg
92077 ± 49% +372.2% 434756 ± 7% sched_debug.cfs_rq:/.spread0.max
-17235 +513.2% -105683 sched_debug.cfs_rq:/.spread0.min
18312 ± 28% +830.1% 170323 ± 10% sched_debug.cfs_rq:/.spread0.stddev
773.86 ± 4% +23.2% 953.22 sched_debug.cfs_rq:/.util_avg.avg
1596 ± 9% +40.5% 2243 ± 7% sched_debug.cfs_rq:/.util_avg.max
309.13 ± 11% +30.8% 404.46 ± 10% sched_debug.cfs_rq:/.util_avg.stddev
168.92 ± 16% +251.7% 594.11 ± 4% sched_debug.cfs_rq:/.util_est_enqueued.avg
712.88 ± 22% +143.8% 1738 ± 5% sched_debug.cfs_rq:/.util_est_enqueued.max
165.72 ± 17% +108.2% 345.00 sched_debug.cfs_rq:/.util_est_enqueued.stddev
415111 ± 5% +58.1% 656270 sched_debug.cpu.avg_idle.avg
62646 ± 20% +415.9% 323170 ± 13% sched_debug.cpu.avg_idle.min
225351 ± 8% -21.7% 176447 sched_debug.cpu.avg_idle.stddev
69939 +213.6% 219304 sched_debug.cpu.clock.avg
69948 +213.5% 219322 sched_debug.cpu.clock.max
69929 +213.6% 219284 sched_debug.cpu.clock.min
5.54 ± 15% +100.7% 11.12 sched_debug.cpu.clock.stddev
69939 +213.6% 219304 sched_debug.cpu.clock_task.avg
69948 +213.5% 219322 sched_debug.cpu.clock_task.max
69929 +213.6% 219284 sched_debug.cpu.clock_task.min
5.54 ± 15% +100.7% 11.12 sched_debug.cpu.clock_task.stddev
13.70 ± 13% +67.5% 22.95 ± 6% sched_debug.cpu.cpu_load[0].avg
40.25 ± 12% +246.2% 139.36 ± 57% sched_debug.cpu.cpu_load[0].max
9.51 ± 5% +144.4% 23.23 ± 50% sched_debug.cpu.cpu_load[0].stddev
14.31 ± 11% +61.5% 23.10 ± 5% sched_debug.cpu.cpu_load[1].avg
33.38 ± 7% +192.8% 97.71 ± 41% sched_debug.cpu.cpu_load[1].max
1.12 ±148% +534.9% 7.14 sched_debug.cpu.cpu_load[1].min
7.73 ± 10% +123.5% 17.28 ± 36% sched_debug.cpu.cpu_load[1].stddev
15.17 ± 9% +51.7% 23.01 ± 2% sched_debug.cpu.cpu_load[2].avg
37.62 ± 16% +106.7% 77.79 ± 21% sched_debug.cpu.cpu_load[2].max
1.88 ±103% +300.0% 7.50 ± 2% sched_debug.cpu.cpu_load[2].min
7.43 ± 11% +91.4% 14.21 ± 21% sched_debug.cpu.cpu_load[2].stddev
16.76 ± 7% +37.2% 22.98 sched_debug.cpu.cpu_load[3].avg
3.38 ± 36% +139.2% 8.07 sched_debug.cpu.cpu_load[3].min
11.42 ± 4% +8.8% 12.42 ± 6% sched_debug.cpu.cpu_load[3].stddev
19.03 ± 3% +21.8% 23.19 sched_debug.cpu.cpu_load[4].avg
173.25 ± 3% -48.9% 88.50 ± 3% sched_debug.cpu.cpu_load[4].max
4.62 ± 4% +85.3% 8.57 sched_debug.cpu.cpu_load[4].min
26.77 ± 2% -44.3% 14.92 ± 3% sched_debug.cpu.cpu_load[4].stddev
1725 ± 13% +62.4% 2803 ± 5% sched_debug.cpu.curr->pid.avg
3860 ± 9% +98.5% 7662 ± 2% sched_debug.cpu.curr->pid.max
17235 ± 9% +50.1% 25869 sched_debug.cpu.load.avg
49790 ± 5% +31.5% 65479 ± 9% sched_debug.cpu.load.max
0.00 ± 6% +65.8% 0.00 ± 29% sched_debug.cpu.next_balance.stddev
40051 +376.3% 190749 sched_debug.cpu.nr_load_updates.avg
48118 ± 4% +315.5% 199922 sched_debug.cpu.nr_load_updates.max
37637 +401.8% 188874 sched_debug.cpu.nr_load_updates.min
0.61 ± 11% +83.4% 1.12 ± 7% sched_debug.cpu.nr_running.avg
1.62 ± 13% +62.6% 2.64 ± 8% sched_debug.cpu.nr_running.max
0.43 ± 8% +17.0% 0.51 ± 8% sched_debug.cpu.nr_running.stddev
15323 +38.3% 21185 sched_debug.cpu.nr_switches.avg
23666 ± 3% +33.5% 31603 ± 2% sched_debug.cpu.nr_switches.max
13256 +27.2% 16862 sched_debug.cpu.nr_switches.min
2125 ± 4% +45.4% 3091 ± 4% sched_debug.cpu.nr_switches.stddev
36.53 +57.0% 57.34 sched_debug.cpu.nr_uninterruptible.avg
121.62 ± 10% +53.4% 186.57 ± 6% sched_debug.cpu.nr_uninterruptible.max
35.08 ± 12% +51.7% 53.23 ± 7% sched_debug.cpu.nr_uninterruptible.stddev
69929 +213.6% 219281 sched_debug.cpu_clk
66357 +224.9% 215625 sched_debug.ktime
70373 +212.3% 219744 sched_debug.sched_clk
187.50 ± 32% +270.7% 695.00 interrupts.37:IR-PCI-MSI.524289-edge.eth0-TxRx-0
87.75 ± 49% +149.6% 219.00 interrupts.39:IR-PCI-MSI.524291-edge.eth0-TxRx-2
133.00 ± 60% +129.3% 305.00 ± 4% interrupts.40:IR-PCI-MSI.524292-edge.eth0-TxRx-3
157.75 ± 95% +204.0% 479.50 ± 32% interrupts.41:IR-PCI-MSI.524293-edge.eth0-TxRx-4
57.75 ± 6% +409.1% 294.00 ± 25% interrupts.42:IR-PCI-MSI.524294-edge.eth0-TxRx-5
93.50 ± 75% +749.7% 794.50 ± 31% interrupts.43:IR-PCI-MSI.524295-edge.eth0-TxRx-6
55.75 ± 12% +979.8% 602.00 interrupts.44:IR-PCI-MSI.524296-edge.eth0-TxRx-7
39.75 ± 2% +366.7% 185.50 interrupts.94:IR-PCI-MSI.512000-edge.ahci[0000:00:1f.2]
41912 ± 9% +213.8% 131529 interrupts.CAL:Function_call_interrupts
1078 ± 10% +217.8% 3426 ± 6% interrupts.CPU0.CAL:Function_call_interrupts
160883 +371.4% 758361 interrupts.CPU0.LOC:Local_timer_interrupts
3951 ± 70% +128.9% 9046 interrupts.CPU0.NMI:Non-maskable_interrupts
3951 ± 70% +128.9% 9046 interrupts.CPU0.PMI:Performance_monitoring_interrupts
2756 ± 14% +212.4% 8612 ± 3% interrupts.CPU0.RES:Rescheduling_interrupts
1052 ± 7% +209.7% 3260 ± 7% interrupts.CPU1.CAL:Function_call_interrupts
160927 +371.3% 758448 interrupts.CPU1.LOC:Local_timer_interrupts
2268 ± 16% +241.3% 7743 ± 9% interrupts.CPU1.RES:Rescheduling_interrupts
949.25 ± 18% +238.6% 3214 interrupts.CPU10.CAL:Function_call_interrupts
160048 ± 2% +373.5% 757791 interrupts.CPU10.LOC:Local_timer_interrupts
2401 ± 26% +269.4% 8870 ± 6% interrupts.CPU10.RES:Rescheduling_interrupts
1078 ± 9% +208.4% 3325 ± 4% interrupts.CPU11.CAL:Function_call_interrupts
160525 +372.2% 757955 interrupts.CPU11.LOC:Local_timer_interrupts
2224 ± 5% +223.7% 7202 ± 6% interrupts.CPU11.RES:Rescheduling_interrupts
1084 ± 10% +186.2% 3103 interrupts.CPU12.CAL:Function_call_interrupts
160874 +371.2% 758018 interrupts.CPU12.LOC:Local_timer_interrupts
2084 ± 2% +290.9% 8147 ± 4% interrupts.CPU12.RES:Rescheduling_interrupts
1081 ± 11% +194.8% 3186 ± 10% interrupts.CPU13.CAL:Function_call_interrupts
160936 +371.0% 757976 interrupts.CPU13.LOC:Local_timer_interrupts
2226 ± 2% +191.5% 6490 interrupts.CPU13.RES:Rescheduling_interrupts
1091 ± 9% +223.0% 3526 ± 2% interrupts.CPU14.CAL:Function_call_interrupts
160591 +372.2% 758257 interrupts.CPU14.LOC:Local_timer_interrupts
2239 ± 11% +291.6% 8771 ± 11% interrupts.CPU14.RES:Rescheduling_interrupts
1079 ± 11% +208.8% 3331 ± 5% interrupts.CPU15.CAL:Function_call_interrupts
160602 +372.1% 758250 interrupts.CPU15.LOC:Local_timer_interrupts
2150 ± 3% +209.2% 6647 interrupts.CPU15.RES:Rescheduling_interrupts
1064 ± 12% +228.0% 3490 interrupts.CPU16.CAL:Function_call_interrupts
160603 +371.8% 757703 interrupts.CPU16.LOC:Local_timer_interrupts
2102 ± 2% +288.3% 8162 ± 3% interrupts.CPU16.RES:Rescheduling_interrupts
997.75 ± 9% +227.0% 3262 ± 7% interrupts.CPU17.CAL:Function_call_interrupts
160890 +371.2% 758088 interrupts.CPU17.LOC:Local_timer_interrupts
2114 ± 2% +216.2% 6683 interrupts.CPU17.RES:Rescheduling_interrupts
1042 ± 7% +187.8% 3001 ± 4% interrupts.CPU18.CAL:Function_call_interrupts
160581 +372.1% 758185 interrupts.CPU18.LOC:Local_timer_interrupts
2195 ± 9% +319.5% 9210 ± 7% interrupts.CPU18.RES:Rescheduling_interrupts
1033 ± 10% +215.9% 3263 ± 5% interrupts.CPU19.CAL:Function_call_interrupts
160518 +372.4% 758272 interrupts.CPU19.LOC:Local_timer_interrupts
2198 ± 4% +307.0% 8945 interrupts.CPU19.RES:Rescheduling_interrupts
55.75 ± 12% +979.8% 602.00 interrupts.CPU2.44:IR-PCI-MSI.524296-edge.eth0-TxRx-7
1077 ± 9% +208.1% 3319 ± 4% interrupts.CPU2.CAL:Function_call_interrupts
160815 +371.5% 758168 interrupts.CPU2.LOC:Local_timer_interrupts
2070 ± 4% +383.7% 10015 interrupts.CPU2.RES:Rescheduling_interrupts
1050 ± 6% +209.2% 3247 ± 7% interrupts.CPU20.CAL:Function_call_interrupts
160541 +372.1% 757904 interrupts.CPU20.LOC:Local_timer_interrupts
2020 ± 3% +301.7% 8117 ± 2% interrupts.CPU20.RES:Rescheduling_interrupts
1030 ± 10% +211.6% 3210 ± 3% interrupts.CPU21.CAL:Function_call_interrupts
160769 +371.6% 758128 interrupts.CPU21.LOC:Local_timer_interrupts
1957 ±173% +228.8% 6435 ± 44% interrupts.CPU21.NMI:Non-maskable_interrupts
1957 ±173% +228.8% 6435 ± 44% interrupts.CPU21.PMI:Performance_monitoring_interrupts
2120 ± 3% +205.5% 6477 interrupts.CPU21.RES:Rescheduling_interrupts
1026 ± 10% +211.6% 3198 ± 12% interrupts.CPU22.CAL:Function_call_interrupts
160736 +371.6% 757984 interrupts.CPU22.LOC:Local_timer_interrupts
2042 ± 2% +301.8% 8205 interrupts.CPU22.RES:Rescheduling_interrupts
1035 ± 8% +195.1% 3056 interrupts.CPU23.CAL:Function_call_interrupts
160511 +372.5% 758362 interrupts.CPU23.LOC:Local_timer_interrupts
2116 ± 2% +234.4% 7076 ± 5% interrupts.CPU23.RES:Rescheduling_interrupts
1062 ± 9% +201.8% 3205 ± 10% interrupts.CPU24.CAL:Function_call_interrupts
160679 +372.0% 758454 interrupts.CPU24.LOC:Local_timer_interrupts
2075 +288.5% 8063 interrupts.CPU24.RES:Rescheduling_interrupts
1038 ± 8% +243.2% 3563 interrupts.CPU25.CAL:Function_call_interrupts
160868 +371.3% 758094 interrupts.CPU25.LOC:Local_timer_interrupts
2108 +210.2% 6540 ± 3% interrupts.CPU25.RES:Rescheduling_interrupts
187.50 ± 32% +270.7% 695.00 interrupts.CPU26.37:IR-PCI-MSI.524289-edge.eth0-TxRx-0
1007 ± 7% +223.5% 3260 ± 3% interrupts.CPU26.CAL:Function_call_interrupts
160779 +371.6% 758284 interrupts.CPU26.LOC:Local_timer_interrupts
2098 ± 2% +291.7% 8219 interrupts.CPU26.RES:Rescheduling_interrupts
1048 ± 9% +193.2% 3073 interrupts.CPU27.CAL:Function_call_interrupts
160687 +371.9% 758313 interrupts.CPU27.LOC:Local_timer_interrupts
2109 ± 4% +245.1% 7278 ± 6% interrupts.CPU27.RES:Rescheduling_interrupts
1036 ± 12% +242.1% 3544 interrupts.CPU28.CAL:Function_call_interrupts
160492 +372.3% 757965 interrupts.CPU28.LOC:Local_timer_interrupts
991.00 ±172% +580.5% 6744 ± 30% interrupts.CPU28.NMI:Non-maskable_interrupts
991.00 ±172% +580.5% 6744 ± 30% interrupts.CPU28.PMI:Performance_monitoring_interrupts
2149 ± 6% +333.0% 9305 interrupts.CPU28.RES:Rescheduling_interrupts
998.00 ± 5% +224.9% 3243 ± 8% interrupts.CPU29.CAL:Function_call_interrupts
160614 +372.2% 758375 interrupts.CPU29.LOC:Local_timer_interrupts
1277 ±127% +260.2% 4601 interrupts.CPU29.NMI:Non-maskable_interrupts
1277 ±127% +260.2% 4601 interrupts.CPU29.PMI:Performance_monitoring_interrupts
2169 +212.4% 6776 interrupts.CPU29.RES:Rescheduling_interrupts
1084 ± 11% +224.5% 3519 interrupts.CPU3.CAL:Function_call_interrupts
160908 +371.1% 757989 interrupts.CPU3.LOC:Local_timer_interrupts
2284 ± 12% +192.3% 6675 interrupts.CPU3.RES:Rescheduling_interrupts
87.75 ± 49% +149.6% 219.00 interrupts.CPU30.39:IR-PCI-MSI.524291-edge.eth0-TxRx-2
1085 ± 11% +185.3% 3097 ± 15% interrupts.CPU30.CAL:Function_call_interrupts
160639 +372.1% 758299 interrupts.CPU30.LOC:Local_timer_interrupts
2114 ± 5% +285.7% 8157 ± 2% interrupts.CPU30.RES:Rescheduling_interrupts
1058 ± 9% +216.8% 3354 ± 6% interrupts.CPU31.CAL:Function_call_interrupts
160640 +371.9% 757999 interrupts.CPU31.LOC:Local_timer_interrupts
2115 ± 5% +234.4% 7073 ± 5% interrupts.CPU31.RES:Rescheduling_interrupts
133.00 ± 60% +129.3% 305.00 ± 4% interrupts.CPU32.40:IR-PCI-MSI.524292-edge.eth0-TxRx-3
1028 ± 9% +249.1% 3590 interrupts.CPU32.CAL:Function_call_interrupts
160441 +372.6% 758242 interrupts.CPU32.LOC:Local_timer_interrupts
1292 ±123% +600.6% 9056 interrupts.CPU32.NMI:Non-maskable_interrupts
1292 ±123% +600.6% 9056 interrupts.CPU32.PMI:Performance_monitoring_interrupts
2161 ± 3% +269.5% 7984 ± 2% interrupts.CPU32.RES:Rescheduling_interrupts
1052 ± 7% +197.5% 3129 ± 7% interrupts.CPU33.CAL:Function_call_interrupts
160555 +372.2% 758200 interrupts.CPU33.LOC:Local_timer_interrupts
2129 ± 2% +212.0% 6644 interrupts.CPU33.RES:Rescheduling_interrupts
157.75 ± 95% +204.0% 479.50 ± 32% interrupts.CPU34.41:IR-PCI-MSI.524293-edge.eth0-TxRx-4
1038 ± 11% +199.1% 3106 interrupts.CPU34.CAL:Function_call_interrupts
160579 +372.0% 757860 interrupts.CPU34.LOC:Local_timer_interrupts
2041 ± 3% +361.8% 9427 ± 9% interrupts.CPU34.RES:Rescheduling_interrupts
1013 ± 7% +232.2% 3366 ± 7% interrupts.CPU35.CAL:Function_call_interrupts
160263 +373.1% 758135 interrupts.CPU35.LOC:Local_timer_interrupts
2087 ± 3% +230.3% 6893 ± 6% interrupts.CPU35.RES:Rescheduling_interrupts
57.75 ± 6% +409.1% 294.00 ± 25% interrupts.CPU36.42:IR-PCI-MSI.524294-edge.eth0-TxRx-5
1043 ± 11% +193.5% 3062 ± 3% interrupts.CPU36.CAL:Function_call_interrupts
160457 +372.5% 758097 interrupts.CPU36.LOC:Local_timer_interrupts
2142 ± 2% +291.2% 8381 ± 4% interrupts.CPU36.RES:Rescheduling_interrupts
1042 ± 11% +240.9% 3554 interrupts.CPU37.CAL:Function_call_interrupts
159944 +374.0% 758180 interrupts.CPU37.LOC:Local_timer_interrupts
2029 ± 93% +234.1% 6781 ± 32% interrupts.CPU37.NMI:Non-maskable_interrupts
2029 ± 93% +234.1% 6781 ± 32% interrupts.CPU37.PMI:Performance_monitoring_interrupts
2109 ± 4% +215.2% 6647 ± 2% interrupts.CPU37.RES:Rescheduling_interrupts
93.50 ± 75% +749.7% 794.50 ± 31% interrupts.CPU38.43:IR-PCI-MSI.524295-edge.eth0-TxRx-6
1083 ± 12% +230.4% 3578 interrupts.CPU38.CAL:Function_call_interrupts
160648 +371.5% 757502 interrupts.CPU38.LOC:Local_timer_interrupts
2128 +306.2% 8646 ± 9% interrupts.CPU38.RES:Rescheduling_interrupts
1047 ± 4% +212.3% 3272 ± 9% interrupts.CPU39.CAL:Function_call_interrupts
160454 +372.5% 758119 interrupts.CPU39.LOC:Local_timer_interrupts
2095 +214.2% 6583 interrupts.CPU39.RES:Rescheduling_interrupts
1070 ± 11% +211.3% 3333 ± 8% interrupts.CPU4.CAL:Function_call_interrupts
160548 +372.3% 758250 interrupts.CPU4.LOC:Local_timer_interrupts
2130 ± 4% +285.3% 8209 interrupts.CPU4.RES:Rescheduling_interrupts
963.25 ± 17% +225.8% 3138 interrupts.CPU5.CAL:Function_call_interrupts
160994 +371.0% 758309 interrupts.CPU5.LOC:Local_timer_interrupts
2117 ± 2% +230.2% 6992 ± 6% interrupts.CPU5.RES:Rescheduling_interrupts
39.75 ± 2% +366.7% 185.50 interrupts.CPU6.94:IR-PCI-MSI.512000-edge.ahci[0000:00:1f.2]
1048 ± 13% +223.4% 3389 ± 6% interrupts.CPU6.CAL:Function_call_interrupts
160695 +372.0% 758502 interrupts.CPU6.LOC:Local_timer_interrupts
2143 ± 3% +280.7% 8158 interrupts.CPU6.RES:Rescheduling_interrupts
1073 ± 9% +226.0% 3500 interrupts.CPU7.CAL:Function_call_interrupts
160777 +371.3% 757691 interrupts.CPU7.LOC:Local_timer_interrupts
3248 ± 36% +111.7% 6876 ± 34% interrupts.CPU7.NMI:Non-maskable_interrupts
3248 ± 36% +111.7% 6876 ± 34% interrupts.CPU7.PMI:Performance_monitoring_interrupts
2175 +205.1% 6636 ± 2% interrupts.CPU7.RES:Rescheduling_interrupts
1045 ± 12% +196.7% 3101 interrupts.CPU8.CAL:Function_call_interrupts
160575 +372.3% 758457 interrupts.CPU8.LOC:Local_timer_interrupts
2188 ± 5% +277.1% 8252 ± 4% interrupts.CPU8.RES:Rescheduling_interrupts
1091 ± 9% +186.0% 3122 interrupts.CPU9.CAL:Function_call_interrupts
160626 +372.2% 758451 interrupts.CPU9.LOC:Local_timer_interrupts
5576 ± 40% -75.7% 1354 ± 99% interrupts.CPU9.NMI:Non-maskable_interrupts
5576 ± 40% -75.7% 1354 ± 99% interrupts.CPU9.PMI:Performance_monitoring_interrupts
2159 ± 2% +212.9% 6755 interrupts.CPU9.RES:Rescheduling_interrupts
6425227 +372.0% 30325627 interrupts.LOC:Local_timer_interrupts
86362 +258.6% 309674 interrupts.RES:Rescheduling_interrupts
198.25 ± 3% -33.4% 132.00 interrupts.TLB:TLB_shootdownsDisclaimer:
Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance.
Thanks,
Rong Chen
#
# Automatically generated file; DO NOT EDIT.
# Linux/x86_64 5.1.0-rc3 Kernel Configuration
#
#
# Compiler: gcc-7 (Debian 7.3.0-1) 7.3.0
#CONFIG_CC_IS_GCC=yCONFIG_GCC_VERSION=70300CONFIG_CLANG_VERSION=0CONFIG_CC_HAS_ASM_GOTO=yCONFIG_CC_HAS_WARN_MAYBE_UNINITIALIZED=yCONFIG_IRQ_WORK=yCONFIG_BUILDTIME_EXTABLE_SORT=yCONFIG_THREAD_INFO_IN_TASK=y
#
# General setup
#CONFIG_INIT_ENV_ARG_LIMIT=32
# CONFIG_COMPILE_TEST is not setCONFIG_LOCALVERSION=""CONFIG_LOCALVERSION_AUTO=yCONFIG_BUILD_SALT=""CONFIG_HAVE_KERNEL_GZIP=yCONFIG_HAVE_KERNEL_BZIP2=yCONFIG_HAVE_KERNEL_LZMA=yCONFIG_HAVE_KERNEL_XZ=yCONFIG_HAVE_KERNEL_LZO=yCONFIG_HAVE_KERNEL_LZ4=yCONFIG_KERNEL_GZIP=y
# CONFIG_KERNEL_BZIP2 is not set
# CONFIG_KERNEL_LZMA is not set
# CONFIG_KERNEL_XZ is not set
# CONFIG_KERNEL_LZO is not set
# CONFIG_KERNEL_LZ4 is not setCONFIG_DEFAULT_HOSTNAME="(none)"CONFIG_SWAP=yCONFIG_SYSVIPC=yCONFIG_SYSVIPC_SYSCTL=yCONFIG_POSIX_MQUEUE=yCONFIG_POSIX_MQUEUE_SYSCTL=yCONFIG_CROSS_MEMORY_ATTACH=yCONFIG_USELIB=yCONFIG_AUDIT=yCONFIG_HAVE_ARCH_AUDITSYSCALL=yCONFIG_AUDITSYSCALL=y
#
# IRQ subsystem
#CONFIG_GENERIC_IRQ_PROBE=yCONFIG_GENERIC_IRQ_SHOW=yCONFIG_GENERIC_IRQ_EFFECTIVE_AFF_MASK=yCONFIG_GENERIC_PENDING_IRQ=yCONFIG_GENERIC_IRQ_MIGRATION=yCONFIG_IRQ_DOMAIN=yCONFIG_IRQ_SIM=yCONFIG_IRQ_DOMAIN_HIERARCHY=yCONFIG_GENERIC_MSI_IRQ=yCONFIG_GENERIC_MSI_IRQ_DOMAIN=yCONFIG_GENERIC_IRQ_MATRIX_ALLOCATOR=yCONFIG_GENERIC_IRQ_RESERVATION_MODE=yCONFIG_IRQ_FORCED_THREADING=yCONFIG_SPARSE_IRQ=y
# CONFIG_GENERIC_IRQ_DEBUGFS is not setCONFIG_CLOCKSOURCE_WATCHDOG=yCONFIG_ARCH_CLOCKSOURCE_DATA=yCONFIG_ARCH_CLOCKSOURCE_INIT=yCONFIG_CLOCKSOURCE_VALIDATE_LAST_CYCLE=yCONFIG_GENERIC_TIME_VSYSCALL=yCONFIG_GENERIC_CLOCKEVENTS=yCONFIG_GENERIC_CLOCKEVENTS_BROADCAST=yCONFIG_GENERIC_CLOCKEVENTS_MIN_ADJUST=yCONFIG_GENERIC_CMOS_UPDATE=y
#
# Timers subsystem
#CONFIG_TICK_ONESHOT=yCONFIG_NO_HZ_COMMON=y
# CONFIG_HZ_PERIODIC is not set
# CONFIG_NO_HZ_IDLE is not setCONFIG_NO_HZ_FULL=yCONFIG_CONTEXT_TRACKING=y
# CONFIG_CONTEXT_TRACKING_FORCE is not setCONFIG_NO_HZ=yCONFIG_HIGH_RES_TIMERS=y
# CONFIG_PREEMPT_NONE is not setCONFIG_PREEMPT_VOLUNTARY=y
# CONFIG_PREEMPT is not setCONFIG_PREEMPT_COUNT=y
#
# CPU/Task time and stats accounting
#CONFIG_VIRT_CPU_ACCOUNTING=yCONFIG_VIRT_CPU_ACCOUNTING_GEN=y
# CONFIG_IRQ_TIME_ACCOUNTING is not setCONFIG_HAVE_SCHED_AVG_IRQ=yCONFIG_BSD_PROCESS_ACCT=yCONFIG_BSD_PROCESS_ACCT_V3=yCONFIG_TASKSTATS=yCONFIG_TASK_DELAY_ACCT=yCONFIG_TASK_XACCT=yCONFIG_TASK_IO_ACCOUNTING=y
# CONFIG_PSI is not setCONFIG_CPU_ISOLATION=y
#
# RCU Subsystem
#CONFIG_TREE_RCU=y
# CONFIG_RCU_EXPERT is not setCONFIG_SRCU=yCONFIG_TREE_SRCU=yCONFIG_TASKS_RCU=yCONFIG_RCU_STALL_COMMON=yCONFIG_RCU_NEED_SEGCBLIST=yCONFIG_RCU_NOCB_CPU=yCONFIG_BUILD_BIN2C=yCONFIG_IKCONFIG=yCONFIG_IKCONFIG_PROC=yCONFIG_LOG_BUF_SHIFT=20CONFIG_LOG_CPU_MAX_BUF_SHIFT=12CONFIG_PRINTK_SAFE_LOG_BUF_SHIFT=13CONFIG_HAVE_UNSTABLE_SCHED_CLOCK=yCONFIG_ARCH_SUPPORTS_NUMA_BALANCING=yCONFIG_ARCH_WANT_BATCHED_UNMAP_TLB_FLUSH=yCONFIG_ARCH_SUPPORTS_INT128=yCONFIG_NUMA_BALANCING=yCONFIG_NUMA_BALANCING_DEFAULT_ENABLED=yCONFIG_CGROUPS=yCONFIG_PAGE_COUNTER=yCONFIG_MEMCG=yCONFIG_MEMCG_SWAP=yCONFIG_MEMCG_SWAP_ENABLED=yCONFIG_MEMCG_KMEM=yCONFIG_BLK_CGROUP=y
# CONFIG_DEBUG_BLK_CGROUP is not setCONFIG_CGROUP_WRITEBACK=yCONFIG_CGROUP_SCHED=yCONFIG_FAIR_GROUP_SCHED=yCONFIG_CFS_BANDWIDTH=yCONFIG_RT_GROUP_SCHED=yCONFIG_CGROUP_PIDS=yCONFIG_CGROUP_RDMA=yCONFIG_CGROUP_FREEZER=yCONFIG_CGROUP_HUGETLB=yCONFIG_CPUSETS=yCONFIG_PROC_PID_CPUSET=yCONFIG_CGROUP_DEVICE=yCONFIG_CGROUP_CPUACCT=yCONFIG_CGROUP_PERF=yCONFIG_CGROUP_BPF=y
# CONFIG_CGROUP_DEBUG is not setCONFIG_SOCK_CGROUP_DATA=yCONFIG_NAMESPACES=yCONFIG_UTS_NS=yCONFIG_IPC_NS=yCONFIG_USER_NS=yCONFIG_PID_NS=yCONFIG_NET_NS=yCONFIG_CHECKPOINT_RESTORE=yCONFIG_SCHED_AUTOGROUP=y
# CONFIG_SYSFS_DEPRECATED is not setCONFIG_RELAY=yCONFIG_BLK_DEV_INITRD=yCONFIG_INITRAMFS_SOURCE=""CONFIG_RD_GZIP=yCONFIG_RD_BZIP2=yCONFIG_RD_LZMA=yCONFIG_RD_XZ=yCONFIG_RD_LZO=yCONFIG_RD_LZ4=yCONFIG_CC_OPTIMIZE_FOR_PERFORMANCE=y
# CONFIG_CC_OPTIMIZE_FOR_SIZE is not setCONFIG_SYSCTL=yCONFIG_ANON_INODES=yCONFIG_HAVE_UID16=yCONFIG_SYSCTL_EXCEPTION_TRACE=yCONFIG_HAVE_PCSPKR_PLATFORM=yCONFIG_BPF=yCONFIG_EXPERT=yCONFIG_UID16=yCONFIG_MULTIUSER=yCONFIG_SGETMASK_SYSCALL=yCONFIG_SYSFS_SYSCALL=y
# CONFIG_SYSCTL_SYSCALL is not setCONFIG_FHANDLE=yCONFIG_POSIX_TIMERS=yCONFIG_PRINTK=yCONFIG_PRINTK_NMI=yCONFIG_BUG=yCONFIG_ELF_CORE=yCONFIG_PCSPKR_PLATFORM=yCONFIG_BASE_FULL=yCONFIG_FUTEX=yCONFIG_FUTEX_PI=yCONFIG_EPOLL=yCONFIG_SIGNALFD=yCONFIG_TIMERFD=yCONFIG_EVENTFD=yCONFIG_SHMEM=yCONFIG_AIO=yCONFIG_IO_URING=yCONFIG_ADVISE_SYSCALLS=yCONFIG_MEMBARRIER=yCONFIG_KALLSYMS=yCONFIG_KALLSYMS_ALL=yCONFIG_KALLSYMS_ABSOLUTE_PERCPU=yCONFIG_KALLSYMS_BASE_RELATIVE=yCONFIG_BPF_SYSCALL=yCONFIG_BPF_JIT_ALWAYS_ON=yCONFIG_USERFAULTFD=yCONFIG_ARCH_HAS_MEMBARRIER_SYNC_CORE=yCONFIG_RSEQ=y
# CONFIG_DEBUG_RSEQ is not setCONFIG_EMBEDDED=yCONFIG_HAVE_PERF_EVENTS=y
# CONFIG_PC104 is not set
#
# Kernel Performance Events And Counters
#CONFIG_PERF_EVENTS=y
# CONFIG_DEBUG_PERF_USE_VMALLOC is not set CONFIG_VM_EVENT_COUNTERS=yCONFIG_SLUB_DEBUG=y
# CONFIG_SLUB_MEMCG_SYSFS_ON is not set
# CONFIG_COMPAT_BRK is not set
# CONFIG_SLAB is not setCONFIG_SLUB=y
# CONFIG_SLOB is not set CONFIG_SLAB_MERGE_DEFAULT=y
# CONFIG_SLAB_FREELIST_RANDOM is not set
# CONFIG_SLAB_FREELIST_HARDENED is not set CONFIG_SLUB_CPU_PARTIAL=y CONFIG_SYSTEM_DATA_VERIFICATION=y CONFIG_PROFILING=yCONFIG_TRACEPOINTS=yCONFIG_64BIT=yCONFIG_X86_64=yCONFIG_X86=yCONFIG_INSTRUCTION_DECODER=y CONFIG_OUTPUT_FORMAT="elf64-x86-64" CONFIG_ARCH_DEFCONFIG="arch/x86/configs/x86_64_defconfig" CONFIG_LOCKDEP_SUPPORT=y CONFIG_STACKTRACE_SUPPORT=yCONFIG_MMU=yCONFIG_ARCH_MMAP_RND_BITS_MIN=28 CONFIG_ARCH_MMAP_RND_BITS_MAX=32 CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MIN=8 CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MAX=16 CONFIG_GENERIC_ISA_DMA=yCONFIG_GENERIC_BUG=y CONFIG_GENERIC_BUG_RELATIVE_POINTERS=y CONFIG_GENERIC_HWEIGHT=y CONFIG_ARCH_MAY_HAVE_PC_FDC=y CONFIG_RWSEM_XCHGADD_ALGORITHM=y CONFIG_GENERIC_CALIBRATE_DELAY=y CONFIG_ARCH_HAS_CPU_RELAX=y CONFIG_ARCH_HAS_CACHE_LINE_SIZE=y CONFIG_ARCH_HAS_FILTER_PGPROT=y CONFIG_HAVE_SETUP_PER_CPU_AREA=y CONFIG_NEED_PER_CPU_EMBED_FIRST_CHUNK=y CONFIG_NEED_PER_CPU_PAGE_FIRST_CHUNK=y CONFIG_ARCH_HIBERNATION_POSSIBLE=y CONFIG_ARCH_SUSPEND_POSSIBLE=y CONFIG_ARCH_WANT_HUGE_PMD_SHARE=y CONFIG_ARCH_WANT_GENERAL_HUGETLB=y CONFIG_ZONE_DMA32=yCONFIG_AUDIT_ARCH=y CONFIG_ARCH_SUPPORTS_OPTIMIZED_INLINING=y CONFIG_ARCH_SUPPORTS_DEBUG_PAGEALLOC=y CONFIG_HAVE_INTEL_TXT=yCONFIG_X86_64_SMP=y CONFIG_ARCH_SUPPORTS_UPROBES=y CONFIG_FIX_EARLYCON_MEM=y CONFIG_DYNAMIC_PHYSICAL_MASK=y CONFIG_PGTABLE_LEVELS=5 CONFIG_CC_HAS_SANE_STACKPROTECTOR=y
#
# Processor type and features
#CONFIG_ZONE_DMA=yCONFIG_SMP=yCONFIG_X86_FEATURE_NAMES=yCONFIG_X86_X2APIC=yCONFIG_X86_MPPARSE=y
# CONFIG_GOLDFISH is not setCONFIG_RETPOLINE=y
# CONFIG_X86_CPU_RESCTRL is not setCONFIG_X86_EXTENDED_PLATFORM=y
# CONFIG_X86_NUMACHIP is not set
# CONFIG_X86_VSMP is not setCONFIG_X86_UV=y
# CONFIG_X86_GOLDFISH is not set
# CONFIG_X86_INTEL_MID is not setCONFIG_X86_INTEL_LPSS=yCONFIG_X86_AMD_PLATFORM_DEVICE=yCONFIG_IOSF_MBI=y
# CONFIG_IOSF_MBI_DEBUG is not setCONFIG_X86_SUPPORTS_MEMORY_FAILURE=y
# CONFIG_SCHED_OMIT_FRAME_POINTER is not setCONFIG_HYPERVISOR_GUEST=yCONFIG_PARAVIRT=yCONFIG_PARAVIRT_XXL=y
# CONFIG_PARAVIRT_DEBUG is not setCONFIG_PARAVIRT_SPINLOCKS=y
# CONFIG_QUEUED_LOCK_STAT is not setCONFIG_XEN=yCONFIG_XEN_PV=yCONFIG_XEN_PV_SMP=y
# CONFIG_XEN_DOM0 is not setCONFIG_XEN_PVHVM=yCONFIG_XEN_PVHVM_SMP=yCONFIG_XEN_512GB=yCONFIG_XEN_SAVE_RESTORE=y
# CONFIG_XEN_DEBUG_FS is not set
# CONFIG_XEN_PVH is not setCONFIG_KVM_GUEST=y
# CONFIG_PVH is not set
# CONFIG_KVM_DEBUG_FS is not setCONFIG_PARAVIRT_TIME_ACCOUNTING=yCONFIG_PARAVIRT_CLOCK=y
# CONFIG_JAILHOUSE_GUEST is not set
# CONFIG_MK8 is not set
# CONFIG_MPSC is not set
# CONFIG_MCORE2 is not set
# CONFIG_MATOM is not setCONFIG_GENERIC_CPU=yCONFIG_X86_INTERNODE_CACHE_SHIFT=6CONFIG_X86_L1_CACHE_SHIFT=6CONFIG_X86_TSC=yCONFIG_X86_CMPXCHG64=yCONFIG_X86_CMOV=yCONFIG_X86_MINIMUM_CPU_FAMILY=64CONFIG_X86_DEBUGCTLMSR=y
# CONFIG_PROCESSOR_SELECT is not setCONFIG_CPU_SUP_INTEL=yCONFIG_CPU_SUP_AMD=yCONFIG_CPU_SUP_HYGON=yCONFIG_CPU_SUP_CENTAUR=yCONFIG_HPET_TIMER=yCONFIG_HPET_EMULATE_RTC=yCONFIG_DMI=yCONFIG_GART_IOMMU=y
# CONFIG_CALGARY_IOMMU is not setCONFIG_MAXSMP=yCONFIG_NR_CPUS_RANGE_BEGIN=8192CONFIG_NR_CPUS_RANGE_END=8192CONFIG_NR_CPUS_DEFAULT=8192CONFIG_NR_CPUS=8192CONFIG_SCHED_SMT=yCONFIG_SCHED_MC=yCONFIG_SCHED_MC_PRIO=yCONFIG_X86_LOCAL_APIC=yCONFIG_X86_IO_APIC=yCONFIG_X86_REROUTE_FOR_BROKEN_BOOT_IRQS=yCONFIG_X86_MCE=yCONFIG_X86_MCELOG_LEGACY=yCONFIG_X86_MCE_INTEL=yCONFIG_X86_MCE_AMD=yCONFIG_X86_MCE_THRESHOLD=yCONFIG_X86_MCE_INJECT=mCONFIG_X86_THERMAL_VECTOR=y
#
# Performance monitoring
#CONFIG_PERF_EVENTS_INTEL_UNCORE=y CONFIG_PERF_EVENTS_INTEL_RAPL=y CONFIG_PERF_EVENTS_INTEL_CSTATE=y
# CONFIG_PERF_EVENTS_AMD_POWER is not set CONFIG_X86_16BIT=yCONFIG_X86_ESPFIX64=y CONFIG_X86_VSYSCALL_EMULATION=yCONFIG_I8K=mCONFIG_MICROCODE=yCONFIG_MICROCODE_INTEL=yCONFIG_MICROCODE_AMD=y CONFIG_MICROCODE_OLD_INTERFACE=y CONFIG_X86_MSR=yCONFIG_X86_CPUID=yCONFIG_X86_5LEVEL=yCONFIG_X86_DIRECT_GBPAGES=y
# CONFIG_X86_CPA_STATISTICS is not set CONFIG_ARCH_HAS_MEM_ENCRYPT=y CONFIG_AMD_MEM_ENCRYPT=y
# CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT is not set CONFIG_NUMA=yCONFIG_AMD_NUMA=yCONFIG_X86_64_ACPI_NUMA=y CONFIG_NODES_SPAN_OTHER_NODES=y
# CONFIG_NUMA_EMU is not set CONFIG_NODES_SHIFT=10 CONFIG_ARCH_SPARSEMEM_ENABLE=y CONFIG_ARCH_SPARSEMEM_DEFAULT=y CONFIG_ARCH_SELECT_MEMORY_MODEL=y CONFIG_ARCH_MEMORY_PROBE=y CONFIG_ARCH_PROC_KCORE_TEXT=y CONFIG_ILLEGAL_POINTER_VALUE=0xdead000000000000 CONFIG_X86_PMEM_LEGACY_DEVICE=y CONFIG_X86_PMEM_LEGACY=m CONFIG_X86_CHECK_BIOS_CORRUPTION=y
# CONFIG_X86_BOOTPARAM_MEMORY_CORRUPTION_CHECK is not set CONFIG_X86_RESERVE_LOW=64CONFIG_MTRR=yCONFIG_MTRR_SANITIZER=y CONFIG_MTRR_SANITIZER_ENABLE_DEFAULT=1 CONFIG_MTRR_SANITIZER_SPARE_REG_NR_DEFAULT=1 CONFIG_X86_PAT=yCONFIG_ARCH_USES_PG_UNCACHED=y CONFIG_ARCH_RANDOM=yCONFIG_X86_SMAP=yCONFIG_X86_INTEL_UMIP=yCONFIG_X86_INTEL_MPX=y CONFIG_X86_INTEL_MEMORY_PROTECTION_KEYS=y CONFIG_EFI=yCONFIG_EFI_STUB=yCONFIG_EFI_MIXED=yCONFIG_SECCOMP=y
# CONFIG_HZ_100 is not set
# CONFIG_HZ_250 is not set
# CONFIG_HZ_300 is not setCONFIG_HZ_1000=yCONFIG_HZ=1000CONFIG_SCHED_HRTICK=yCONFIG_KEXEC=yCONFIG_KEXEC_FILE=y CONFIG_ARCH_HAS_KEXEC_PURGATORY=y CONFIG_KEXEC_VERIFY_SIG=y CONFIG_KEXEC_BZIMAGE_VERIFY_SIG=y CONFIG_CRASH_DUMP=yCONFIG_KEXEC_JUMP=y CONFIG_PHYSICAL_START=0x1000000 CONFIG_RELOCATABLE=yCONFIG_RANDOMIZE_BASE=yCONFIG_X86_NEED_RELOCS=y CONFIG_PHYSICAL_ALIGN=0x200000 CONFIG_DYNAMIC_MEMORY_LAYOUT=y CONFIG_RANDOMIZE_MEMORY=y CONFIG_RANDOMIZE_MEMORY_PHYSICAL_PADDING=0xa CONFIG_HOTPLUG_CPU=y CONFIG_BOOTPARAM_HOTPLUG_CPU0=y
# CONFIG_DEBUG_HOTPLUG_CPU0 is not set
# CONFIG_COMPAT_VDSO is not set CONFIG_LEGACY_VSYSCALL_EMULATE=y
# CONFIG_LEGACY_VSYSCALL_NONE is not set
# CONFIG_CMDLINE_BOOL is not set CONFIG_MODIFY_LDT_SYSCALL=y CONFIG_HAVE_LIVEPATCH=yCONFIG_LIVEPATCH=yCONFIG_ARCH_HAS_ADD_PAGES=y CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y CONFIG_ARCH_ENABLE_MEMORY_HOTREMOVE=y CONFIG_USE_PERCPU_NUMA_NODE_ID=y CONFIG_ARCH_ENABLE_SPLIT_PMD_PTLOCK=y CONFIG_ARCH_ENABLE_HUGEPAGE_MIGRATION=y CONFIG_ARCH_ENABLE_THP_MIGRATION=y
#
# Power management and ACPI options
#CONFIG_ARCH_HIBERNATION_HEADER=yCONFIG_SUSPEND=yCONFIG_SUSPEND_FREEZER=y
# CONFIG_SUSPEND_SKIP_SYNC is not setCONFIG_HIBERNATE_CALLBACKS=yCONFIG_HIBERNATION=yCONFIG_PM_STD_PARTITION=""CONFIG_PM_SLEEP=yCONFIG_PM_SLEEP_SMP=y
# CONFIG_PM_AUTOSLEEP is not set
# CONFIG_PM_WAKELOCKS is not setCONFIG_PM=yCONFIG_PM_DEBUG=yCONFIG_PM_ADVANCED_DEBUG=y
# CONFIG_PM_TEST_SUSPEND is not setCONFIG_PM_SLEEP_DEBUG=y
# CONFIG_DPM_WATCHDOG is not setCONFIG_PM_TRACE=yCONFIG_PM_TRACE_RTC=yCONFIG_PM_CLK=y
# CONFIG_WQ_POWER_EFFICIENT_DEFAULT is not set
# CONFIG_ENERGY_MODEL is not setCONFIG_ARCH_SUPPORTS_ACPI=yCONFIG_ACPI=yCONFIG_ACPI_LEGACY_TABLES_LOOKUP=yCONFIG_ARCH_MIGHT_HAVE_ACPI_PDC=yCONFIG_ACPI_SYSTEM_POWER_STATES_SUPPORT=y
# CONFIG_ACPI_DEBUGGER is not setCONFIG_ACPI_SPCR_TABLE=yCONFIG_ACPI_LPIT=yCONFIG_ACPI_SLEEP=y
# CONFIG_ACPI_PROCFS_POWER is not setCONFIG_ACPI_REV_OVERRIDE_POSSIBLE=yCONFIG_ACPI_EC_DEBUGFS=mCONFIG_ACPI_AC=yCONFIG_ACPI_BATTERY=yCONFIG_ACPI_BUTTON=yCONFIG_ACPI_VIDEO=mCONFIG_ACPI_FAN=y
# CONFIG_ACPI_TAD is not setCONFIG_ACPI_DOCK=yCONFIG_ACPI_CPU_FREQ_PSS=yCONFIG_ACPI_PROCESSOR_CSTATE=yCONFIG_ACPI_PROCESSOR_IDLE=yCONFIG_ACPI_CPPC_LIB=yCONFIG_ACPI_PROCESSOR=yCONFIG_ACPI_IPMI=mCONFIG_ACPI_HOTPLUG_CPU=yCONFIG_ACPI_PROCESSOR_AGGREGATOR=mCONFIG_ACPI_THERMAL=yCONFIG_ACPI_NUMA=yCONFIG_ARCH_HAS_ACPI_TABLE_UPGRADE=yCONFIG_ACPI_TABLE_UPGRADE=y
# CONFIG_ACPI_DEBUG is not setCONFIG_ACPI_PCI_SLOT=yCONFIG_ACPI_CONTAINER=yCONFIG_ACPI_HOTPLUG_MEMORY=yCONFIG_ACPI_HOTPLUG_IOAPIC=yCONFIG_ACPI_SBS=mCONFIG_ACPI_HED=yCONFIG_ACPI_CUSTOM_METHOD=mCONFIG_ACPI_BGRT=y
# CONFIG_ACPI_REDUCED_HARDWARE_ONLY is not setCONFIG_ACPI_NFIT=m
# CONFIG_NFIT_SECURITY_DEBUG is not setCONFIG_HAVE_ACPI_APEI=yCONFIG_HAVE_ACPI_APEI_NMI=yCONFIG_ACPI_APEI=yCONFIG_ACPI_APEI_GHES=yCONFIG_ACPI_APEI_PCIEAER=yCONFIG_ACPI_APEI_MEMORY_FAILURE=yCONFIG_ACPI_APEI_EINJ=mCONFIG_ACPI_APEI_ERST_DEBUG=y
# CONFIG_DPTF_POWER is not setCONFIG_ACPI_WATCHDOG=yCONFIG_ACPI_EXTLOG=mCONFIG_ACPI_ADXL=y
# CONFIG_PMIC_OPREGION is not set
# CONFIG_ACPI_CONFIGFS is not setCONFIG_X86_PM_TIMER=yCONFIG_SFI=y
#
# CPU Frequency scaling
#CONFIG_CPU_FREQ=y CONFIG_CPU_FREQ_GOV_ATTR_SET=y CONFIG_CPU_FREQ_GOV_COMMON=y
# CONFIG_CPU_FREQ_STAT is not set
# CONFIG_CPU_FREQ_DEFAULT_GOV_PERFORMANCE is not set # CONFIG_CPU_FREQ_DEFAULT_GOV_POWERSAVE is not set # CONFIG_CPU_FREQ_DEFAULT_GOV_USERSPACE is not set CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
# CONFIG_CPU_FREQ_DEFAULT_GOV_CONSERVATIVE is not set # CONFIG_CPU_FREQ_DEFAULT_GOV_SCHEDUTIL is not set CONFIG_CPU_FREQ_GOV_PERFORMANCE=y CONFIG_CPU_FREQ_GOV_POWERSAVE=y CONFIG_CPU_FREQ_GOV_USERSPACE=y CONFIG_CPU_FREQ_GOV_ONDEMAND=y CONFIG_CPU_FREQ_GOV_CONSERVATIVE=y
# CONFIG_CPU_FREQ_GOV_SCHEDUTIL is not set
#
# CPU frequency scaling drivers
#CONFIG_X86_INTEL_PSTATE=yCONFIG_X86_PCC_CPUFREQ=mCONFIG_X86_ACPI_CPUFREQ=mCONFIG_X86_ACPI_CPUFREQ_CPB=yCONFIG_X86_POWERNOW_K8=mCONFIG_X86_AMD_FREQ_SENSITIVITY=m
# CONFIG_X86_SPEEDSTEP_CENTRINO is not setCONFIG_X86_P4_CLOCKMOD=m
#
# shared options # CONFIG_X86_SPEEDSTEP_LIB=m
#
# CPU Idle
#CONFIG_CPU_IDLE=y
# CONFIG_CPU_IDLE_GOV_LADDER is not setCONFIG_CPU_IDLE_GOV_MENU=y
# CONFIG_CPU_IDLE_GOV_TEO is not setCONFIG_INTEL_IDLE=y
#
# Bus options (PCI etc.)
#CONFIG_PCI_DIRECT=yCONFIG_PCI_MMCONFIG=yCONFIG_PCI_XEN=yCONFIG_MMCONF_FAM10H=y
# CONFIG_PCI_CNB20LE_QUIRK is not set
# CONFIG_ISA_BUS is not setCONFIG_ISA_DMA_API=yCONFIG_AMD_NB=y
# CONFIG_X86_SYSFB is not set
#
# Binary Emulations
#CONFIG_IA32_EMULATION=y
# CONFIG_X86_X32 is not setCONFIG_COMPAT_32=yCONFIG_COMPAT=yCONFIG_COMPAT_FOR_U64_ALIGNMENT=yCONFIG_SYSVIPC_COMPAT=yCONFIG_X86_DEV_DMA_OPS=yCONFIG_HAVE_GENERIC_GUP=y
#
# Firmware Drivers
#CONFIG_EDD=m
# CONFIG_EDD_OFF is not setCONFIG_FIRMWARE_MEMMAP=yCONFIG_DMIID=yCONFIG_DMI_SYSFS=yCONFIG_DMI_SCAN_MACHINE_NON_EFI_FALLBACK=yCONFIG_ISCSI_IBFT_FIND=yCONFIG_ISCSI_IBFT=mCONFIG_FW_CFG_SYSFS=y
# CONFIG_FW_CFG_SYSFS_CMDLINE is not set
# CONFIG_GOOGLE_FIRMWARE is not set
#
# EFI (Extensible Firmware Interface) Support
#CONFIG_EFI_VARS=yCONFIG_EFI_ESRT=yCONFIG_EFI_VARS_PSTORE=yCONFIG_EFI_VARS_PSTORE_DEFAULT_DISABLE=yCONFIG_EFI_RUNTIME_MAP=y
# CONFIG_EFI_FAKE_MEMMAP is not setCONFIG_EFI_RUNTIME_WRAPPERS=y
# CONFIG_EFI_BOOTLOADER_CONTROL is not set
# CONFIG_EFI_CAPSULE_LOADER is not set
# CONFIG_EFI_TEST is not setCONFIG_APPLE_PROPERTIES=y
# CONFIG_RESET_ATTACK_MITIGATION is not setCONFIG_UEFI_CPER=yCONFIG_UEFI_CPER_X86=yCONFIG_EFI_DEV_PATH_PARSER=yCONFIG_EFI_EARLYCON=y
#
# Tegra firmware driver
#CONFIG_HAVE_KVM=yCONFIG_HAVE_KVM_IRQCHIP=yCONFIG_HAVE_KVM_IRQFD=yCONFIG_HAVE_KVM_IRQ_ROUTING=yCONFIG_HAVE_KVM_EVENTFD=yCONFIG_KVM_MMIO=yCONFIG_KVM_ASYNC_PF=yCONFIG_HAVE_KVM_MSI=yCONFIG_HAVE_KVM_CPU_RELAX_INTERCEPT=yCONFIG_KVM_VFIO=yCONFIG_KVM_GENERIC_DIRTYLOG_READ_PROTECT=yCONFIG_KVM_COMPAT=yCONFIG_HAVE_KVM_IRQ_BYPASS=yCONFIG_VIRTUALIZATION=yCONFIG_KVM=mCONFIG_KVM_INTEL=mCONFIG_KVM_AMD=mCONFIG_KVM_AMD_SEV=yCONFIG_KVM_MMU_AUDIT=yCONFIG_VHOST_NET=m
# CONFIG_VHOST_SCSI is not setCONFIG_VHOST_VSOCK=mCONFIG_VHOST=m
# CONFIG_VHOST_CROSS_ENDIAN_LEGACY is not set
#
# General architecture-dependent options
#CONFIG_CRASH_CORE=yCONFIG_KEXEC_CORE=yCONFIG_HOTPLUG_SMT=yCONFIG_OPROFILE=mCONFIG_OPROFILE_EVENT_MULTIPLEX=yCONFIG_HAVE_OPROFILE=yCONFIG_OPROFILE_NMI_TIMER=yCONFIG_KPROBES=yCONFIG_JUMP_LABEL=y
# CONFIG_STATIC_KEYS_SELFTEST is not setCONFIG_OPTPROBES=yCONFIG_KPROBES_ON_FTRACE=yCONFIG_UPROBES=yCONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS=yCONFIG_ARCH_USE_BUILTIN_BSWAP=yCONFIG_KRETPROBES=yCONFIG_USER_RETURN_NOTIFIER=yCONFIG_HAVE_IOREMAP_PROT=yCONFIG_HAVE_KPROBES=yCONFIG_HAVE_KRETPROBES=yCONFIG_HAVE_OPTPROBES=yCONFIG_HAVE_KPROBES_ON_FTRACE=yCONFIG_HAVE_FUNCTION_ERROR_INJECTION=yCONFIG_HAVE_NMI=yCONFIG_HAVE_ARCH_TRACEHOOK=yCONFIG_HAVE_DMA_CONTIGUOUS=yCONFIG_GENERIC_SMP_IDLE_THREAD=yCONFIG_ARCH_HAS_FORTIFY_SOURCE=yCONFIG_ARCH_HAS_SET_MEMORY=yCONFIG_HAVE_ARCH_THREAD_STRUCT_WHITELIST=yCONFIG_ARCH_WANTS_DYNAMIC_TASK_STRUCT=yCONFIG_HAVE_REGS_AND_STACK_ACCESS_API=yCONFIG_HAVE_RSEQ=yCONFIG_HAVE_FUNCTION_ARG_ACCESS_API=yCONFIG_HAVE_CLK=yCONFIG_HAVE_HW_BREAKPOINT=yCONFIG_HAVE_MIXED_BREAKPOINTS_REGS=yCONFIG_HAVE_USER_RETURN_NOTIFIER=yCONFIG_HAVE_PERF_EVENTS_NMI=yCONFIG_HAVE_HARDLOCKUP_DETECTOR_PERF=yCONFIG_HAVE_PERF_REGS=yCONFIG_HAVE_PERF_USER_STACK_DUMP=yCONFIG_HAVE_ARCH_JUMP_LABEL=yCONFIG_HAVE_ARCH_JUMP_LABEL_RELATIVE=yCONFIG_HAVE_RCU_TABLE_FREE=yCONFIG_HAVE_RCU_TABLE_INVALIDATE=yCONFIG_ARCH_HAVE_NMI_SAFE_CMPXCHG=yCONFIG_HAVE_ALIGNED_STRUCT_PAGE=yCONFIG_HAVE_CMPXCHG_LOCAL=yCONFIG_HAVE_CMPXCHG_DOUBLE=yCONFIG_ARCH_WANT_COMPAT_IPC_PARSE_VERSION=yCONFIG_ARCH_WANT_OLD_COMPAT_IPC=yCONFIG_HAVE_ARCH_SECCOMP_FILTER=yCONFIG_SECCOMP_FILTER=yCONFIG_HAVE_ARCH_STACKLEAK=yCONFIG_HAVE_STACKPROTECTOR=yCONFIG_CC_HAS_STACKPROTECTOR_NONE=yCONFIG_STACKPROTECTOR=yCONFIG_STACKPROTECTOR_STRONG=yCONFIG_HAVE_ARCH_WITHIN_STACK_FRAMES=yCONFIG_HAVE_CONTEXT_TRACKING=yCONFIG_HAVE_VIRT_CPU_ACCOUNTING_GEN=yCONFIG_HAVE_IRQ_TIME_ACCOUNTING=yCONFIG_HAVE_MOVE_PMD=yCONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE=yCONFIG_HAVE_ARCH_TRANSPARENT_HUGEPAGE_PUD=yCONFIG_HAVE_ARCH_HUGE_VMAP=yCONFIG_HAVE_ARCH_SOFT_DIRTY=yCONFIG_HAVE_MOD_ARCH_SPECIFIC=yCONFIG_MODULES_USE_ELF_RELA=yCONFIG_HAVE_IRQ_EXIT_ON_IRQ_STACK=yCONFIG_ARCH_HAS_ELF_RANDOMIZE=yCONFIG_HAVE_ARCH_MMAP_RND_BITS=yCONFIG_HAVE_EXIT_THREAD=yCONFIG_ARCH_MMAP_RND_BITS=28CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS=yCONFIG_ARCH_MMAP_RND_COMPAT_BITS=8CONFIG_HAVE_ARCH_COMPAT_MMAP_BASES=yCONFIG_HAVE_COPY_THREAD_TLS=yCONFIG_HAVE_STACK_VALIDATION=yCONFIG_HAVE_RELIABLE_STACKTRACE=yCONFIG_OLD_SIGSUSPEND3=yCONFIG_COMPAT_OLD_SIGACTION=yCONFIG_COMPAT_32BIT_TIME=yCONFIG_HAVE_ARCH_VMAP_STACK=yCONFIG_VMAP_STACK=yCONFIG_ARCH_HAS_STRICT_KERNEL_RWX=yCONFIG_STRICT_KERNEL_RWX=yCONFIG_ARCH_HAS_STRICT_MODULE_RWX=yCONFIG_STRICT_MODULE_RWX=yCONFIG_ARCH_HAS_REFCOUNT=y
# CONFIG_REFCOUNT_FULL is not setCONFIG_HAVE_ARCH_PREL32_RELOCATIONS=yCONFIG_ARCH_USE_MEMREMAP_PROT=y
#
# GCOV-based kernel profiling
#
# CONFIG_GCOV_KERNEL is not setCONFIG_ARCH_HAS_GCOV_PROFILE_ALL=yCONFIG_PLUGIN_HOSTCC="g++"CONFIG_HAVE_GCC_PLUGINS=y
# CONFIG_GCC_PLUGINS is not setCONFIG_RT_MUTEXES=yCONFIG_BASE_SMALL=0CONFIG_MODULES=yCONFIG_MODULE_FORCE_LOAD=yCONFIG_MODULE_UNLOAD=y
# CONFIG_MODULE_FORCE_UNLOAD is not set
# CONFIG_MODVERSIONS is not set
# CONFIG_MODULE_SRCVERSION_ALL is not setCONFIG_MODULE_SIG=y
# CONFIG_MODULE_SIG_FORCE is not setCONFIG_MODULE_SIG_ALL=y
# CONFIG_MODULE_SIG_SHA1 is not set
# CONFIG_MODULE_SIG_SHA224 is not setCONFIG_MODULE_SIG_SHA256=y
# CONFIG_MODULE_SIG_SHA384 is not set
# CONFIG_MODULE_SIG_SHA512 is not setCONFIG_MODULE_SIG_HASH="sha256"
# CONFIG_MODULE_COMPRESS is not set
# CONFIG_TRIM_UNUSED_KSYMS is not setCONFIG_MODULES_TREE_LOOKUP=yCONFIG_BLOCK=yCONFIG_BLK_SCSI_REQUEST=yCONFIG_BLK_DEV_BSG=yCONFIG_BLK_DEV_BSGLIB=yCONFIG_BLK_DEV_INTEGRITY=yCONFIG_BLK_DEV_ZONED=yCONFIG_BLK_DEV_THROTTLING=y
# CONFIG_BLK_DEV_THROTTLING_LOW is not set
# CONFIG_BLK_CMDLINE_PARSER is not set
# CONFIG_BLK_WBT is not set
# CONFIG_BLK_CGROUP_IOLATENCY is not setCONFIG_BLK_DEBUG_FS=yCONFIG_BLK_DEBUG_FS_ZONED=y
# CONFIG_BLK_SED_OPAL is not set
#
# Partition Types
#CONFIG_PARTITION_ADVANCED=y
# CONFIG_ACORN_PARTITION is not set
# CONFIG_AIX_PARTITION is not setCONFIG_OSF_PARTITION=yCONFIG_AMIGA_PARTITION=y
# CONFIG_ATARI_PARTITION is not setCONFIG_MAC_PARTITION=yCONFIG_MSDOS_PARTITION=yCONFIG_BSD_DISKLABEL=yCONFIG_MINIX_SUBPARTITION=yCONFIG_SOLARIS_X86_PARTITION=yCONFIG_UNIXWARE_DISKLABEL=y
# CONFIG_LDM_PARTITION is not setCONFIG_SGI_PARTITION=y
# CONFIG_ULTRIX_PARTITION is not setCONFIG_SUN_PARTITION=yCONFIG_KARMA_PARTITION=yCONFIG_EFI_PARTITION=y
# CONFIG_SYSV68_PARTITION is not set
# CONFIG_CMDLINE_PARTITION is not setCONFIG_BLOCK_COMPAT=yCONFIG_BLK_MQ_PCI=yCONFIG_BLK_MQ_VIRTIO=yCONFIG_BLK_MQ_RDMA=yCONFIG_BLK_PM=y
#
# IO Schedulers
#CONFIG_MQ_IOSCHED_DEADLINE=yCONFIG_MQ_IOSCHED_KYBER=y
# CONFIG_IOSCHED_BFQ is not setCONFIG_PREEMPT_NOTIFIERS=yCONFIG_PADATA=yCONFIG_ASN1=yCONFIG_INLINE_SPIN_UNLOCK_IRQ=yCONFIG_INLINE_READ_UNLOCK=yCONFIG_INLINE_READ_UNLOCK_IRQ=yCONFIG_INLINE_WRITE_UNLOCK=yCONFIG_INLINE_WRITE_UNLOCK_IRQ=yCONFIG_ARCH_SUPPORTS_ATOMIC_RMW=yCONFIG_MUTEX_SPIN_ON_OWNER=yCONFIG_RWSEM_SPIN_ON_OWNER=yCONFIG_LOCK_SPIN_ON_OWNER=yCONFIG_ARCH_USE_QUEUED_SPINLOCKS=yCONFIG_QUEUED_SPINLOCKS=yCONFIG_ARCH_USE_QUEUED_RWLOCKS=yCONFIG_QUEUED_RWLOCKS=yCONFIG_ARCH_HAS_SYNC_CORE_BEFORE_USERMODE=yCONFIG_ARCH_HAS_SYSCALL_WRAPPER=yCONFIG_FREEZER=y
#
# Executable file formats
#CONFIG_BINFMT_ELF=yCONFIG_COMPAT_BINFMT_ELF=yCONFIG_ELFCORE=yCONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS=yCONFIG_BINFMT_SCRIPT=yCONFIG_BINFMT_MISC=mCONFIG_COREDUMP=y
#
# Memory Management options
#CONFIG_SELECT_MEMORY_MODEL=yCONFIG_SPARSEMEM_MANUAL=yCONFIG_SPARSEMEM=yCONFIG_NEED_MULTIPLE_NODES=yCONFIG_HAVE_MEMORY_PRESENT=yCONFIG_SPARSEMEM_EXTREME=yCONFIG_SPARSEMEM_VMEMMAP_ENABLE=yCONFIG_SPARSEMEM_VMEMMAP=yCONFIG_HAVE_MEMBLOCK_NODE_MAP=yCONFIG_ARCH_DISCARD_MEMBLOCK=yCONFIG_MEMORY_ISOLATION=yCONFIG_HAVE_BOOTMEM_INFO_NODE=yCONFIG_MEMORY_HOTPLUG=yCONFIG_MEMORY_HOTPLUG_SPARSE=y
# CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE is not setCONFIG_MEMORY_HOTREMOVE=yCONFIG_SPLIT_PTLOCK_CPUS=4CONFIG_MEMORY_BALLOON=yCONFIG_BALLOON_COMPACTION=yCONFIG_COMPACTION=yCONFIG_MIGRATION=yCONFIG_PHYS_ADDR_T_64BIT=yCONFIG_BOUNCE=yCONFIG_VIRT_TO_BUS=yCONFIG_MMU_NOTIFIER=yCONFIG_KSM=yCONFIG_DEFAULT_MMAP_MIN_ADDR=4096CONFIG_ARCH_SUPPORTS_MEMORY_FAILURE=yCONFIG_MEMORY_FAILURE=yCONFIG_HWPOISON_INJECT=mCONFIG_TRANSPARENT_HUGEPAGE=yCONFIG_TRANSPARENT_HUGEPAGE_ALWAYS=y
# CONFIG_TRANSPARENT_HUGEPAGE_MADVISE is not setCONFIG_ARCH_WANTS_THP_SWAP=yCONFIG_THP_SWAP=yCONFIG_TRANSPARENT_HUGE_PAGECACHE=yCONFIG_CLEANCACHE=yCONFIG_FRONTSWAP=yCONFIG_CMA=y
# CONFIG_CMA_DEBUG is not set
# CONFIG_CMA_DEBUGFS is not setCONFIG_CMA_AREAS=7CONFIG_MEM_SOFT_DIRTY=yCONFIG_ZSWAP=yCONFIG_ZPOOL=yCONFIG_ZBUD=y
# CONFIG_Z3FOLD is not setCONFIG_ZSMALLOC=y
# CONFIG_PGTABLE_MAPPING is not set
# CONFIG_ZSMALLOC_STAT is not setCONFIG_GENERIC_EARLY_IOREMAP=yCONFIG_DEFERRED_STRUCT_PAGE_INIT=yCONFIG_IDLE_PAGE_TRACKING=yCONFIG_ARCH_HAS_ZONE_DEVICE=yCONFIG_ZONE_DEVICE=yCONFIG_ARCH_HAS_HMM=yCONFIG_MIGRATE_VMA_HELPER=yCONFIG_DEV_PAGEMAP_OPS=yCONFIG_HMM=yCONFIG_HMM_MIRROR=y
# CONFIG_DEVICE_PRIVATE is not set
# CONFIG_DEVICE_PUBLIC is not setCONFIG_FRAME_VECTOR=yCONFIG_ARCH_USES_HIGH_VMA_FLAGS=yCONFIG_ARCH_HAS_PKEYS=y
# CONFIG_PERCPU_STATS is not set
# CONFIG_GUP_BENCHMARK is not setCONFIG_ARCH_HAS_PTE_SPECIAL=yCONFIG_NET=yCONFIG_COMPAT_NETLINK_MESSAGES=yCONFIG_NET_INGRESS=yCONFIG_NET_EGRESS=yCONFIG_SKB_EXTENSIONS=y
#
# Networking options
#CONFIG_PACKET=yCONFIG_PACKET_DIAG=mCONFIG_UNIX=yCONFIG_UNIX_SCM=yCONFIG_UNIX_DIAG=m
# CONFIG_TLS is not setCONFIG_XFRM=yCONFIG_XFRM_ALGO=yCONFIG_XFRM_USER=y
# CONFIG_XFRM_INTERFACE is not setCONFIG_XFRM_SUB_POLICY=yCONFIG_XFRM_MIGRATE=yCONFIG_XFRM_STATISTICS=yCONFIG_XFRM_IPCOMP=mCONFIG_NET_KEY=mCONFIG_NET_KEY_MIGRATE=y
# CONFIG_SMC is not set
# CONFIG_XDP_SOCKETS is not setCONFIG_INET=yCONFIG_IP_MULTICAST=yCONFIG_IP_ADVANCED_ROUTER=yCONFIG_IP_FIB_TRIE_STATS=yCONFIG_IP_MULTIPLE_TABLES=yCONFIG_IP_ROUTE_MULTIPATH=yCONFIG_IP_ROUTE_VERBOSE=yCONFIG_IP_ROUTE_CLASSID=yCONFIG_IP_PNP=yCONFIG_IP_PNP_DHCP=y
# CONFIG_IP_PNP_BOOTP is not set
# CONFIG_IP_PNP_RARP is not setCONFIG_NET_IPIP=mCONFIG_NET_IPGRE_DEMUX=mCONFIG_NET_IP_TUNNEL=mCONFIG_NET_IPGRE=mCONFIG_NET_IPGRE_BROADCAST=yCONFIG_IP_MROUTE_COMMON=yCONFIG_IP_MROUTE=yCONFIG_IP_MROUTE_MULTIPLE_TABLES=yCONFIG_IP_PIMSM_V1=yCONFIG_IP_PIMSM_V2=yCONFIG_SYN_COOKIES=yCONFIG_NET_IPVTI=mCONFIG_NET_UDP_TUNNEL=mCONFIG_NET_FOU=mCONFIG_NET_FOU_IP_TUNNELS=yCONFIG_INET_AH=mCONFIG_INET_ESP=m
# CONFIG_INET_ESP_OFFLOAD is not setCONFIG_INET_IPCOMP=mCONFIG_INET_XFRM_TUNNEL=mCONFIG_INET_TUNNEL=mCONFIG_INET_XFRM_MODE_TRANSPORT=mCONFIG_INET_XFRM_MODE_TUNNEL=mCONFIG_INET_XFRM_MODE_BEET=mCONFIG_INET_DIAG=mCONFIG_INET_TCP_DIAG=mCONFIG_INET_UDP_DIAG=m
# CONFIG_INET_RAW_DIAG is not set
# CONFIG_INET_DIAG_DESTROY is not setCONFIG_TCP_CONG_ADVANCED=yCONFIG_TCP_CONG_BIC=mCONFIG_TCP_CONG_CUBIC=yCONFIG_TCP_CONG_WESTWOOD=mCONFIG_TCP_CONG_HTCP=mCONFIG_TCP_CONG_HSTCP=mCONFIG_TCP_CONG_HYBLA=mCONFIG_TCP_CONG_VEGAS=m
# CONFIG_TCP_CONG_NV is not setCONFIG_TCP_CONG_SCALABLE=mCONFIG_TCP_CONG_LP=mCONFIG_TCP_CONG_VENO=mCONFIG_TCP_CONG_YEAH=mCONFIG_TCP_CONG_ILLINOIS=mCONFIG_TCP_CONG_DCTCP=m
# CONFIG_TCP_CONG_CDG is not set
# CONFIG_TCP_CONG_BBR is not setCONFIG_DEFAULT_CUBIC=y
# CONFIG_DEFAULT_RENO is not setCONFIG_DEFAULT_TCP_CONG="cubic"CONFIG_TCP_MD5SIG=yCONFIG_IPV6=yCONFIG_IPV6_ROUTER_PREF=yCONFIG_IPV6_ROUTE_INFO=yCONFIG_IPV6_OPTIMISTIC_DAD=yCONFIG_INET6_AH=mCONFIG_INET6_ESP=m
# CONFIG_INET6_ESP_OFFLOAD is not setCONFIG_INET6_IPCOMP=mCONFIG_IPV6_MIP6=m
# CONFIG_IPV6_ILA is not setCONFIG_INET6_XFRM_TUNNEL=mCONFIG_INET6_TUNNEL=mCONFIG_INET6_XFRM_MODE_TRANSPORT=mCONFIG_INET6_XFRM_MODE_TUNNEL=mCONFIG_INET6_XFRM_MODE_BEET=mCONFIG_INET6_XFRM_MODE_ROUTEOPTIMIZATION=mCONFIG_IPV6_VTI=mCONFIG_IPV6_SIT=mCONFIG_IPV6_SIT_6RD=yCONFIG_IPV6_NDISC_NODETYPE=yCONFIG_IPV6_TUNNEL=mCONFIG_IPV6_GRE=mCONFIG_IPV6_FOU=mCONFIG_IPV6_FOU_TUNNEL=mCONFIG_IPV6_MULTIPLE_TABLES=y
# CONFIG_IPV6_SUBTREES is not setCONFIG_IPV6_MROUTE=yCONFIG_IPV6_MROUTE_MULTIPLE_TABLES=yCONFIG_IPV6_PIMSM_V2=yCONFIG_IPV6_SEG6_LWTUNNEL=y
# CONFIG_IPV6_SEG6_HMAC is not setCONFIG_IPV6_SEG6_BPF=yCONFIG_NETLABEL=yCONFIG_NETWORK_SECMARK=yCONFIG_NET_PTP_CLASSIFY=yCONFIG_NETWORK_PHY_TIMESTAMPING=yCONFIG_NETFILTER=yCONFIG_NETFILTER_ADVANCED=yCONFIG_BRIDGE_NETFILTER=m
#
# Core Netfilter Configuration
#CONFIG_NETFILTER_INGRESS=yCONFIG_NETFILTER_NETLINK=mCONFIG_NETFILTER_FAMILY_BRIDGE=yCONFIG_NETFILTER_FAMILY_ARP=yCONFIG_NETFILTER_NETLINK_ACCT=mCONFIG_NETFILTER_NETLINK_QUEUE=mCONFIG_NETFILTER_NETLINK_LOG=mCONFIG_NETFILTER_NETLINK_OSF=mCONFIG_NF_CONNTRACK=mCONFIG_NF_LOG_COMMON=m
# CONFIG_NF_LOG_NETDEV is not setCONFIG_NETFILTER_CONNCOUNT=mCONFIG_NF_CONNTRACK_MARK=yCONFIG_NF_CONNTRACK_SECMARK=yCONFIG_NF_CONNTRACK_ZONES=yCONFIG_NF_CONNTRACK_PROCFS=yCONFIG_NF_CONNTRACK_EVENTS=yCONFIG_NF_CONNTRACK_TIMEOUT=yCONFIG_NF_CONNTRACK_TIMESTAMP=yCONFIG_NF_CONNTRACK_LABELS=yCONFIG_NF_CT_PROTO_DCCP=yCONFIG_NF_CT_PROTO_GRE=yCONFIG_NF_CT_PROTO_SCTP=yCONFIG_NF_CT_PROTO_UDPLITE=yCONFIG_NF_CONNTRACK_AMANDA=mCONFIG_NF_CONNTRACK_FTP=mCONFIG_NF_CONNTRACK_H323=mCONFIG_NF_CONNTRACK_IRC=mCONFIG_NF_CONNTRACK_BROADCAST=mCONFIG_NF_CONNTRACK_NETBIOS_NS=mCONFIG_NF_CONNTRACK_SNMP=mCONFIG_NF_CONNTRACK_PPTP=mCONFIG_NF_CONNTRACK_SANE=mCONFIG_NF_CONNTRACK_SIP=mCONFIG_NF_CONNTRACK_TFTP=mCONFIG_NF_CT_NETLINK=mCONFIG_NF_CT_NETLINK_TIMEOUT=m
# CONFIG_NETFILTER_NETLINK_GLUE_CT is not setCONFIG_NF_NAT=mCONFIG_NF_NAT_NEEDED=yCONFIG_NF_NAT_AMANDA=mCONFIG_NF_NAT_FTP=mCONFIG_NF_NAT_IRC=mCONFIG_NF_NAT_SIP=mCONFIG_NF_NAT_TFTP=mCONFIG_NF_NAT_REDIRECT=yCONFIG_NF_NAT_MASQUERADE=yCONFIG_NETFILTER_SYNPROXY=mCONFIG_NF_TABLES=m
# CONFIG_NF_TABLES_SET is not set
# CONFIG_NF_TABLES_INET is not set
# CONFIG_NF_TABLES_NETDEV is not set
# CONFIG_NFT_NUMGEN is not setCONFIG_NFT_CT=mCONFIG_NFT_COUNTER=m
# CONFIG_NFT_CONNLIMIT is not setCONFIG_NFT_LOG=mCONFIG_NFT_LIMIT=mCONFIG_NFT_MASQ=mCONFIG_NFT_REDIR=m
# CONFIG_NFT_TUNNEL is not set
# CONFIG_NFT_OBJREF is not setCONFIG_NFT_QUEUE=m
# CONFIG_NFT_QUOTA is not setCONFIG_NFT_REJECT=mCONFIG_NFT_COMPAT=mCONFIG_NFT_HASH=m
# CONFIG_NFT_XFRM is not set
# CONFIG_NFT_SOCKET is not set
# CONFIG_NFT_OSF is not set
# CONFIG_NFT_TPROXY is not set
# CONFIG_NF_FLOW_TABLE is not setCONFIG_NETFILTER_XTABLES=y
#
# Xtables combined modules
#CONFIG_NETFILTER_XT_MARK=mCONFIG_NETFILTER_XT_CONNMARK=mCONFIG_NETFILTER_XT_SET=m
#
# Xtables targets
#CONFIG_NETFILTER_XT_TARGET_AUDIT=mCONFIG_NETFILTER_XT_TARGET_CHECKSUM=mCONFIG_NETFILTER_XT_TARGET_CLASSIFY=mCONFIG_NETFILTER_XT_TARGET_CONNMARK=mCONFIG_NETFILTER_XT_TARGET_CONNSECMARK=mCONFIG_NETFILTER_XT_TARGET_CT=mCONFIG_NETFILTER_XT_TARGET_DSCP=mCONFIG_NETFILTER_XT_TARGET_HL=mCONFIG_NETFILTER_XT_TARGET_HMARK=mCONFIG_NETFILTER_XT_TARGET_IDLETIMER=mCONFIG_NETFILTER_XT_TARGET_LED=mCONFIG_NETFILTER_XT_TARGET_LOG=mCONFIG_NETFILTER_XT_TARGET_MARK=mCONFIG_NETFILTER_XT_NAT=mCONFIG_NETFILTER_XT_TARGET_NETMAP=mCONFIG_NETFILTER_XT_TARGET_NFLOG=mCONFIG_NETFILTER_XT_TARGET_NFQUEUE=mCONFIG_NETFILTER_XT_TARGET_NOTRACK=mCONFIG_NETFILTER_XT_TARGET_RATEEST=mCONFIG_NETFILTER_XT_TARGET_REDIRECT=mCONFIG_NETFILTER_XT_TARGET_TEE=mCONFIG_NETFILTER_XT_TARGET_TPROXY=mCONFIG_NETFILTER_XT_TARGET_TRACE=mCONFIG_NETFILTER_XT_TARGET_SECMARK=mCONFIG_NETFILTER_XT_TARGET_TCPMSS=mCONFIG_NETFILTER_XT_TARGET_TCPOPTSTRIP=m
#
# Xtables matches
#CONFIG_NETFILTER_XT_MATCH_ADDRTYPE=mCONFIG_NETFILTER_XT_MATCH_BPF=mCONFIG_NETFILTER_XT_MATCH_CGROUP=mCONFIG_NETFILTER_XT_MATCH_CLUSTER=mCONFIG_NETFILTER_XT_MATCH_COMMENT=mCONFIG_NETFILTER_XT_MATCH_CONNBYTES=mCONFIG_NETFILTER_XT_MATCH_CONNLABEL=mCONFIG_NETFILTER_XT_MATCH_CONNLIMIT=mCONFIG_NETFILTER_XT_MATCH_CONNMARK=mCONFIG_NETFILTER_XT_MATCH_CONNTRACK=mCONFIG_NETFILTER_XT_MATCH_CPU=mCONFIG_NETFILTER_XT_MATCH_DCCP=mCONFIG_NETFILTER_XT_MATCH_DEVGROUP=mCONFIG_NETFILTER_XT_MATCH_DSCP=mCONFIG_NETFILTER_XT_MATCH_ECN=mCONFIG_NETFILTER_XT_MATCH_ESP=mCONFIG_NETFILTER_XT_MATCH_HASHLIMIT=mCONFIG_NETFILTER_XT_MATCH_HELPER=mCONFIG_NETFILTER_XT_MATCH_HL=m
# CONFIG_NETFILTER_XT_MATCH_IPCOMP is not setCONFIG_NETFILTER_XT_MATCH_IPRANGE=mCONFIG_NETFILTER_XT_MATCH_IPVS=mCONFIG_NETFILTER_XT_MATCH_L2TP=mCONFIG_NETFILTER_XT_MATCH_LENGTH=mCONFIG_NETFILTER_XT_MATCH_LIMIT=mCONFIG_NETFILTER_XT_MATCH_MAC=mCONFIG_NETFILTER_XT_MATCH_MARK=mCONFIG_NETFILTER_XT_MATCH_MULTIPORT=mCONFIG_NETFILTER_XT_MATCH_NFACCT=mCONFIG_NETFILTER_XT_MATCH_OSF=mCONFIG_NETFILTER_XT_MATCH_OWNER=mCONFIG_NETFILTER_XT_MATCH_POLICY=mCONFIG_NETFILTER_XT_MATCH_PHYSDEV=mCONFIG_NETFILTER_XT_MATCH_PKTTYPE=mCONFIG_NETFILTER_XT_MATCH_QUOTA=mCONFIG_NETFILTER_XT_MATCH_RATEEST=mCONFIG_NETFILTER_XT_MATCH_REALM=mCONFIG_NETFILTER_XT_MATCH_RECENT=mCONFIG_NETFILTER_XT_MATCH_SCTP=mCONFIG_NETFILTER_XT_MATCH_SOCKET=mCONFIG_NETFILTER_XT_MATCH_STATE=mCONFIG_NETFILTER_XT_MATCH_STATISTIC=mCONFIG_NETFILTER_XT_MATCH_STRING=mCONFIG_NETFILTER_XT_MATCH_TCPMSS=mCONFIG_NETFILTER_XT_MATCH_TIME=mCONFIG_NETFILTER_XT_MATCH_U32=mCONFIG_IP_SET=mCONFIG_IP_SET_MAX=256CONFIG_IP_SET_BITMAP_IP=mCONFIG_IP_SET_BITMAP_IPMAC=mCONFIG_IP_SET_BITMAP_PORT=mCONFIG_IP_SET_HASH_IP=mCONFIG_IP_SET_HASH_IPMARK=mCONFIG_IP_SET_HASH_IPPORT=mCONFIG_IP_SET_HASH_IPPORTIP=mCONFIG_IP_SET_HASH_IPPORTNET=mCONFIG_IP_SET_HASH_IPMAC=mCONFIG_IP_SET_HASH_MAC=mCONFIG_IP_SET_HASH_NETPORTNET=mCONFIG_IP_SET_HASH_NET=mCONFIG_IP_SET_HASH_NETNET=mCONFIG_IP_SET_HASH_NETPORT=mCONFIG_IP_SET_HASH_NETIFACE=mCONFIG_IP_SET_LIST_SET=mCONFIG_IP_VS=mCONFIG_IP_VS_IPV6=y
# CONFIG_IP_VS_DEBUG is not setCONFIG_IP_VS_TAB_BITS=12
#
# IPVS transport protocol load balancing support
#CONFIG_IP_VS_PROTO_TCP=yCONFIG_IP_VS_PROTO_UDP=yCONFIG_IP_VS_PROTO_AH_ESP=yCONFIG_IP_VS_PROTO_ESP=yCONFIG_IP_VS_PROTO_AH=yCONFIG_IP_VS_PROTO_SCTP=y
#
# IPVS scheduler
#CONFIG_IP_VS_RR=mCONFIG_IP_VS_WRR=mCONFIG_IP_VS_LC=mCONFIG_IP_VS_WLC=m
# CONFIG_IP_VS_FO is not set
# CONFIG_IP_VS_OVF is not setCONFIG_IP_VS_LBLC=mCONFIG_IP_VS_LBLCR=mCONFIG_IP_VS_DH=mCONFIG_IP_VS_SH=m
# CONFIG_IP_VS_MH is not setCONFIG_IP_VS_SED=mCONFIG_IP_VS_NQ=m
#
# IPVS SH scheduler
#CONFIG_IP_VS_SH_TAB_BITS=8
#
# IPVS MH scheduler
#CONFIG_IP_VS_MH_TAB_INDEX=12
#
# IPVS application helper
#CONFIG_IP_VS_FTP=mCONFIG_IP_VS_NFCT=yCONFIG_IP_VS_PE_SIP=m
#
# IP: Netfilter Configuration
#CONFIG_NF_DEFRAG_IPV4=mCONFIG_NF_SOCKET_IPV4=mCONFIG_NF_TPROXY_IPV4=m
# CONFIG_NF_TABLES_IPV4 is not set
# CONFIG_NF_TABLES_ARP is not setCONFIG_NF_DUP_IPV4=m
# CONFIG_NF_LOG_ARP is not setCONFIG_NF_LOG_IPV4=mCONFIG_NF_REJECT_IPV4=mCONFIG_NF_NAT_SNMP_BASIC=mCONFIG_NF_NAT_PPTP=mCONFIG_NF_NAT_H323=mCONFIG_IP_NF_IPTABLES=mCONFIG_IP_NF_MATCH_AH=mCONFIG_IP_NF_MATCH_ECN=mCONFIG_IP_NF_MATCH_RPFILTER=mCONFIG_IP_NF_MATCH_TTL=mCONFIG_IP_NF_FILTER=mCONFIG_IP_NF_TARGET_REJECT=mCONFIG_IP_NF_TARGET_SYNPROXY=mCONFIG_IP_NF_NAT=mCONFIG_IP_NF_TARGET_MASQUERADE=mCONFIG_IP_NF_TARGET_NETMAP=mCONFIG_IP_NF_TARGET_REDIRECT=mCONFIG_IP_NF_MANGLE=mCONFIG_IP_NF_TARGET_CLUSTERIP=mCONFIG_IP_NF_TARGET_ECN=mCONFIG_IP_NF_TARGET_TTL=mCONFIG_IP_NF_RAW=mCONFIG_IP_NF_SECURITY=mCONFIG_IP_NF_ARPTABLES=mCONFIG_IP_NF_ARPFILTER=mCONFIG_IP_NF_ARP_MANGLE=m
#
# IPv6: Netfilter Configuration
#CONFIG_NF_SOCKET_IPV6=mCONFIG_NF_TPROXY_IPV6=m
# CONFIG_NF_TABLES_IPV6 is not setCONFIG_NF_DUP_IPV6=mCONFIG_NF_REJECT_IPV6=mCONFIG_NF_LOG_IPV6=mCONFIG_IP6_NF_IPTABLES=mCONFIG_IP6_NF_MATCH_AH=mCONFIG_IP6_NF_MATCH_EUI64=mCONFIG_IP6_NF_MATCH_FRAG=mCONFIG_IP6_NF_MATCH_OPTS=mCONFIG_IP6_NF_MATCH_HL=mCONFIG_IP6_NF_MATCH_IPV6HEADER=mCONFIG_IP6_NF_MATCH_MH=mCONFIG_IP6_NF_MATCH_RPFILTER=mCONFIG_IP6_NF_MATCH_RT=m
# CONFIG_IP6_NF_MATCH_SRH is not setCONFIG_IP6_NF_TARGET_HL=mCONFIG_IP6_NF_FILTER=mCONFIG_IP6_NF_TARGET_REJECT=mCONFIG_IP6_NF_TARGET_SYNPROXY=mCONFIG_IP6_NF_MANGLE=mCONFIG_IP6_NF_RAW=mCONFIG_IP6_NF_SECURITY=mCONFIG_IP6_NF_NAT=mCONFIG_IP6_NF_TARGET_MASQUERADE=mCONFIG_IP6_NF_TARGET_NPT=mCONFIG_NF_DEFRAG_IPV6=m
# CONFIG_NF_TABLES_BRIDGE is not setCONFIG_BRIDGE_NF_EBTABLES=mCONFIG_BRIDGE_EBT_BROUTE=mCONFIG_BRIDGE_EBT_T_FILTER=mCONFIG_BRIDGE_EBT_T_NAT=mCONFIG_BRIDGE_EBT_802_3=mCONFIG_BRIDGE_EBT_AMONG=mCONFIG_BRIDGE_EBT_ARP=mCONFIG_BRIDGE_EBT_IP=mCONFIG_BRIDGE_EBT_IP6=mCONFIG_BRIDGE_EBT_LIMIT=mCONFIG_BRIDGE_EBT_MARK=mCONFIG_BRIDGE_EBT_PKTTYPE=mCONFIG_BRIDGE_EBT_STP=mCONFIG_BRIDGE_EBT_VLAN=mCONFIG_BRIDGE_EBT_ARPREPLY=mCONFIG_BRIDGE_EBT_DNAT=mCONFIG_BRIDGE_EBT_MARK_T=mCONFIG_BRIDGE_EBT_REDIRECT=mCONFIG_BRIDGE_EBT_SNAT=mCONFIG_BRIDGE_EBT_LOG=mCONFIG_BRIDGE_EBT_NFLOG=m
# CONFIG_BPFILTER is not setCONFIG_IP_DCCP=mCONFIG_INET_DCCP_DIAG=m
#
# DCCP CCIDs Configuration
#
# CONFIG_IP_DCCP_CCID2_DEBUG is not setCONFIG_IP_DCCP_CCID3=y
# CONFIG_IP_DCCP_CCID3_DEBUG is not setCONFIG_IP_DCCP_TFRC_LIB=y
#
# DCCP Kernel Hacking
#
# CONFIG_IP_DCCP_DEBUG is not set CONFIG_IP_SCTP=m
# CONFIG_SCTP_DBG_OBJCNT is not set
# CONFIG_SCTP_DEFAULT_COOKIE_HMAC_MD5 is not set CONFIG_SCTP_DEFAULT_COOKIE_HMAC_SHA1=y
# CONFIG_SCTP_DEFAULT_COOKIE_HMAC_NONE is not set CONFIG_SCTP_COOKIE_HMAC_MD5=y CONFIG_SCTP_COOKIE_HMAC_SHA1=y CONFIG_INET_SCTP_DIAG=m
# CONFIG_RDS is not set
# CONFIG_TIPC is not setCONFIG_ATM=mCONFIG_ATM_CLIP=m
# CONFIG_ATM_CLIP_NO_ICMP is not set CONFIG_ATM_LANE=m
# CONFIG_ATM_MPOA is not set CONFIG_ATM_BR2684=m
# CONFIG_ATM_BR2684_IPFILTER is not set CONFIG_L2TP=mCONFIG_L2TP_DEBUGFS=mCONFIG_L2TP_V3=yCONFIG_L2TP_IP=mCONFIG_L2TP_ETH=mCONFIG_STP=mCONFIG_GARP=mCONFIG_MRP=mCONFIG_BRIDGE=m CONFIG_BRIDGE_IGMP_SNOOPING=y CONFIG_BRIDGE_VLAN_FILTERING=y CONFIG_HAVE_NET_DSA=y
# CONFIG_NET_DSA is not set CONFIG_VLAN_8021Q=mCONFIG_VLAN_8021Q_GVRP=y CONFIG_VLAN_8021Q_MVRP=y
# CONFIG_DECNET is not setCONFIG_LLC=m
# CONFIG_LLC2 is not set
# CONFIG_ATALK is not set
# CONFIG_X25 is not set
# CONFIG_LAPB is not set
# CONFIG_PHONET is not setCONFIG_6LOWPAN=m
# CONFIG_6LOWPAN_DEBUGFS is not set CONFIG_6LOWPAN_NHC=m CONFIG_6LOWPAN_NHC_DEST=m CONFIG_6LOWPAN_NHC_FRAGMENT=m CONFIG_6LOWPAN_NHC_HOP=m CONFIG_6LOWPAN_NHC_IPV6=m CONFIG_6LOWPAN_NHC_MOBILITY=m CONFIG_6LOWPAN_NHC_ROUTING=m CONFIG_6LOWPAN_NHC_UDP=m
# CONFIG_6LOWPAN_GHC_EXT_HDR_HOP is not set # CONFIG_6LOWPAN_GHC_UDP is not set
# CONFIG_6LOWPAN_GHC_ICMPV6 is not set
# CONFIG_6LOWPAN_GHC_EXT_HDR_DEST is not set # CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG is not set # CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE is not set CONFIG_IEEE802154=m
# CONFIG_IEEE802154_NL802154_EXPERIMENTAL is not set CONFIG_IEEE802154_SOCKET=m CONFIG_IEEE802154_6LOWPAN=m CONFIG_MAC802154=mCONFIG_NET_SCHED=y
#
# Queueing/Scheduling
#CONFIG_NET_SCH_CBQ=mCONFIG_NET_SCH_HTB=mCONFIG_NET_SCH_HFSC=mCONFIG_NET_SCH_ATM=mCONFIG_NET_SCH_PRIO=mCONFIG_NET_SCH_MULTIQ=mCONFIG_NET_SCH_RED=mCONFIG_NET_SCH_SFB=mCONFIG_NET_SCH_SFQ=mCONFIG_NET_SCH_TEQL=mCONFIG_NET_SCH_TBF=m
# CONFIG_NET_SCH_CBS is not set
# CONFIG_NET_SCH_ETF is not set
# CONFIG_NET_SCH_TAPRIO is not setCONFIG_NET_SCH_GRED=mCONFIG_NET_SCH_DSMARK=mCONFIG_NET_SCH_NETEM=mCONFIG_NET_SCH_DRR=mCONFIG_NET_SCH_MQPRIO=m
# CONFIG_NET_SCH_SKBPRIO is not setCONFIG_NET_SCH_CHOKE=mCONFIG_NET_SCH_QFQ=mCONFIG_NET_SCH_CODEL=mCONFIG_NET_SCH_FQ_CODEL=m
# CONFIG_NET_SCH_CAKE is not setCONFIG_NET_SCH_FQ=m
# CONFIG_NET_SCH_HHF is not set
# CONFIG_NET_SCH_PIE is not setCONFIG_NET_SCH_INGRESS=mCONFIG_NET_SCH_PLUG=m
# CONFIG_NET_SCH_DEFAULT is not set
#
# Classification
#CONFIG_NET_CLS=yCONFIG_NET_CLS_BASIC=mCONFIG_NET_CLS_TCINDEX=mCONFIG_NET_CLS_ROUTE4=mCONFIG_NET_CLS_FW=mCONFIG_NET_CLS_U32=mCONFIG_CLS_U32_PERF=yCONFIG_CLS_U32_MARK=yCONFIG_NET_CLS_RSVP=mCONFIG_NET_CLS_RSVP6=mCONFIG_NET_CLS_FLOW=mCONFIG_NET_CLS_CGROUP=yCONFIG_NET_CLS_BPF=mCONFIG_NET_CLS_FLOWER=mCONFIG_NET_CLS_MATCHALL=mCONFIG_NET_EMATCH=yCONFIG_NET_EMATCH_STACK=32CONFIG_NET_EMATCH_CMP=mCONFIG_NET_EMATCH_NBYTE=mCONFIG_NET_EMATCH_U32=mCONFIG_NET_EMATCH_META=mCONFIG_NET_EMATCH_TEXT=m
# CONFIG_NET_EMATCH_CANID is not setCONFIG_NET_EMATCH_IPSET=m
# CONFIG_NET_EMATCH_IPT is not setCONFIG_NET_CLS_ACT=yCONFIG_NET_ACT_POLICE=mCONFIG_NET_ACT_GACT=mCONFIG_GACT_PROB=yCONFIG_NET_ACT_MIRRED=mCONFIG_NET_ACT_SAMPLE=mCONFIG_NET_ACT_IPT=mCONFIG_NET_ACT_NAT=mCONFIG_NET_ACT_PEDIT=mCONFIG_NET_ACT_SIMP=mCONFIG_NET_ACT_SKBEDIT=mCONFIG_NET_ACT_CSUM=mCONFIG_NET_ACT_VLAN=m
# CONFIG_NET_ACT_BPF is not setCONFIG_NET_ACT_CONNMARK=mCONFIG_NET_ACT_SKBMOD=m
# CONFIG_NET_ACT_IFE is not setCONFIG_NET_ACT_TUNNEL_KEY=mCONFIG_NET_CLS_IND=yCONFIG_NET_SCH_FIFO=yCONFIG_DCB=yCONFIG_DNS_RESOLVER=m
# CONFIG_BATMAN_ADV is not setCONFIG_OPENVSWITCH=mCONFIG_OPENVSWITCH_GRE=mCONFIG_OPENVSWITCH_VXLAN=mCONFIG_OPENVSWITCH_GENEVE=mCONFIG_VSOCKETS=mCONFIG_VSOCKETS_DIAG=mCONFIG_VMWARE_VMCI_VSOCKETS=mCONFIG_VIRTIO_VSOCKETS=mCONFIG_VIRTIO_VSOCKETS_COMMON=mCONFIG_HYPERV_VSOCKETS=mCONFIG_NETLINK_DIAG=mCONFIG_MPLS=yCONFIG_NET_MPLS_GSO=y
# CONFIG_MPLS_ROUTING is not setCONFIG_NET_NSH=m
# CONFIG_HSR is not setCONFIG_NET_SWITCHDEV=yCONFIG_NET_L3_MASTER_DEV=y
# CONFIG_NET_NCSI is not setCONFIG_RPS=yCONFIG_RFS_ACCEL=yCONFIG_XPS=y
# CONFIG_CGROUP_NET_PRIO is not setCONFIG_CGROUP_NET_CLASSID=yCONFIG_NET_RX_BUSY_POLL=yCONFIG_BQL=yCONFIG_BPF_JIT=yCONFIG_BPF_STREAM_PARSER=yCONFIG_NET_FLOW_LIMIT=y
#
# Network testing
#CONFIG_NET_PKTGEN=mCONFIG_NET_DROP_MONITOR=y
# CONFIG_HAMRADIO is not setCONFIG_CAN=mCONFIG_CAN_RAW=mCONFIG_CAN_BCM=mCONFIG_CAN_GW=m
#
# CAN Device Drivers
#CONFIG_CAN_VCAN=m
# CONFIG_CAN_VXCAN is not setCONFIG_CAN_SLCAN=mCONFIG_CAN_DEV=mCONFIG_CAN_CALC_BITTIMING=yCONFIG_CAN_C_CAN=mCONFIG_CAN_C_CAN_PLATFORM=mCONFIG_CAN_C_CAN_PCI=mCONFIG_CAN_CC770=m
# CONFIG_CAN_CC770_ISA is not setCONFIG_CAN_CC770_PLATFORM=m
# CONFIG_CAN_IFI_CANFD is not set
# CONFIG_CAN_M_CAN is not set
# CONFIG_CAN_PEAK_PCIEFD is not setCONFIG_CAN_SJA1000=m
# CONFIG_CAN_SJA1000_ISA is not setCONFIG_CAN_SJA1000_PLATFORM=mCONFIG_CAN_EMS_PCI=mCONFIG_CAN_PEAK_PCI=mCONFIG_CAN_PEAK_PCIEC=yCONFIG_CAN_KVASER_PCI=mCONFIG_CAN_PLX_PCI=mCONFIG_CAN_SOFTING=m
#
# CAN SPI interfaces
#
# CONFIG_CAN_HI311X is not set
# CONFIG_CAN_MCP251X is not set
#
# CAN USB interfaces
#CONFIG_CAN_8DEV_USB=mCONFIG_CAN_EMS_USB=mCONFIG_CAN_ESD_USB2=m
# CONFIG_CAN_GS_USB is not setCONFIG_CAN_KVASER_USB=m
# CONFIG_CAN_MCBA_USB is not setCONFIG_CAN_PEAK_USB=m
# CONFIG_CAN_UCAN is not set
# CONFIG_CAN_DEBUG_DEVICES is not setCONFIG_BT=mCONFIG_BT_BREDR=yCONFIG_BT_RFCOMM=mCONFIG_BT_RFCOMM_TTY=yCONFIG_BT_BNEP=mCONFIG_BT_BNEP_MC_FILTER=yCONFIG_BT_BNEP_PROTO_FILTER=yCONFIG_BT_CMTP=mCONFIG_BT_HIDP=mCONFIG_BT_HS=yCONFIG_BT_LE=y
# CONFIG_BT_6LOWPAN is not set
# CONFIG_BT_LEDS is not set
# CONFIG_BT_SELFTEST is not setCONFIG_BT_DEBUGFS=y
#
# Bluetooth device drivers
#CONFIG_BT_INTEL=mCONFIG_BT_BCM=mCONFIG_BT_RTL=mCONFIG_BT_HCIBTUSB=m
# CONFIG_BT_HCIBTUSB_AUTOSUSPEND is not setCONFIG_BT_HCIBTUSB_BCM=yCONFIG_BT_HCIBTUSB_RTL=yCONFIG_BT_HCIBTSDIO=mCONFIG_BT_HCIUART=mCONFIG_BT_HCIUART_H4=yCONFIG_BT_HCIUART_BCSP=yCONFIG_BT_HCIUART_ATH3K=y
# CONFIG_BT_HCIUART_INTEL is not set
# CONFIG_BT_HCIUART_AG6XX is not set
# CONFIG_BT_HCIUART_MRVL is not setCONFIG_BT_HCIBCM203X=mCONFIG_BT_HCIBPA10X=mCONFIG_BT_HCIBFUSB=mCONFIG_BT_HCIVHCI=mCONFIG_BT_MRVL=mCONFIG_BT_MRVL_SDIO=mCONFIG_BT_ATH3K=m
# CONFIG_AF_RXRPC is not set
# CONFIG_AF_KCM is not setCONFIG_STREAM_PARSER=yCONFIG_FIB_RULES=yCONFIG_WIRELESS=yCONFIG_WIRELESS_EXT=yCONFIG_WEXT_CORE=yCONFIG_WEXT_PROC=yCONFIG_WEXT_PRIV=yCONFIG_CFG80211=m
# CONFIG_NL80211_TESTMODE is not set
# CONFIG_CFG80211_DEVELOPER_WARNINGS is not set
# CONFIG_CFG80211_CERTIFICATION_ONUS is not setCONFIG_CFG80211_REQUIRE_SIGNED_REGDB=yCONFIG_CFG80211_USE_KERNEL_REGDB_KEYS=yCONFIG_CFG80211_DEFAULT_PS=y
# CONFIG_CFG80211_DEBUGFS is not setCONFIG_CFG80211_CRDA_SUPPORT=yCONFIG_CFG80211_WEXT=yCONFIG_LIB80211=m
# CONFIG_LIB80211_DEBUG is not setCONFIG_MAC80211=mCONFIG_MAC80211_HAS_RC=yCONFIG_MAC80211_RC_MINSTREL=yCONFIG_MAC80211_RC_DEFAULT_MINSTREL=yCONFIG_MAC80211_RC_DEFAULT="minstrel_ht"CONFIG_MAC80211_MESH=yCONFIG_MAC80211_LEDS=yCONFIG_MAC80211_DEBUGFS=y
# CONFIG_MAC80211_MESSAGE_TRACING is not set
# CONFIG_MAC80211_DEBUG_MENU is not setCONFIG_MAC80211_STA_HASH_MAX_SIZE=0
# CONFIG_WIMAX is not setCONFIG_RFKILL=mCONFIG_RFKILL_LEDS=yCONFIG_RFKILL_INPUT=y
# CONFIG_RFKILL_GPIO is not setCONFIG_NET_9P=yCONFIG_NET_9P_VIRTIO=y
# CONFIG_NET_9P_XEN is not set
# CONFIG_NET_9P_RDMA is not set
# CONFIG_NET_9P_DEBUG is not set
# CONFIG_CAIF is not setCONFIG_CEPH_LIB=m
# CONFIG_CEPH_LIB_PRETTYDEBUG is not setCONFIG_CEPH_LIB_USE_DNS_RESOLVER=y
# CONFIG_NFC is not setCONFIG_PSAMPLE=m
# CONFIG_NET_IFE is not setCONFIG_LWTUNNEL=yCONFIG_LWTUNNEL_BPF=yCONFIG_DST_CACHE=yCONFIG_GRO_CELLS=yCONFIG_NET_SOCK_MSG=y
# CONFIG_NET_DEVLINK is not setCONFIG_PAGE_POOL=yCONFIG_FAILOVER=mCONFIG_HAVE_EBPF_JIT=y
#
# Device Drivers
#CONFIG_HAVE_EISA=y
# CONFIG_EISA is not setCONFIG_HAVE_PCI=yCONFIG_PCI=yCONFIG_PCI_DOMAINS=yCONFIG_PCIEPORTBUS=yCONFIG_HOTPLUG_PCI_PCIE=yCONFIG_PCIEAER=yCONFIG_PCIEAER_INJECT=mCONFIG_PCIE_ECRC=yCONFIG_PCIEASPM=y
# CONFIG_PCIEASPM_DEBUG is not setCONFIG_PCIEASPM_DEFAULT=y
# CONFIG_PCIEASPM_POWERSAVE is not set
# CONFIG_PCIEASPM_POWER_SUPERSAVE is not set
# CONFIG_PCIEASPM_PERFORMANCE is not setCONFIG_PCIE_PME=y
# CONFIG_PCIE_DPC is not set
# CONFIG_PCIE_PTM is not setCONFIG_PCI_MSI=yCONFIG_PCI_MSI_IRQ_DOMAIN=yCONFIG_PCI_QUIRKS=y
# CONFIG_PCI_DEBUG is not set
# CONFIG_PCI_REALLOC_ENABLE_AUTO is not setCONFIG_PCI_STUB=y
# CONFIG_PCI_PF_STUB is not set
# CONFIG_XEN_PCIDEV_FRONTEND is not setCONFIG_PCI_ATS=yCONFIG_PCI_LOCKLESS_CONFIG=yCONFIG_PCI_IOV=yCONFIG_PCI_PRI=yCONFIG_PCI_PASID=y
# CONFIG_PCI_P2PDMA is not setCONFIG_PCI_LABEL=yCONFIG_PCI_HYPERV=mCONFIG_HOTPLUG_PCI=yCONFIG_HOTPLUG_PCI_ACPI=yCONFIG_HOTPLUG_PCI_ACPI_IBM=m
# CONFIG_HOTPLUG_PCI_CPCI is not setCONFIG_HOTPLUG_PCI_SHPC=y
#
# PCI controller drivers
#
#
# Cadence PCIe controllers support
#CONFIG_VMD=y
#
# DesignWare PCI Core Support
#
# CONFIG_PCIE_DW_PLAT_HOST is not set
# CONFIG_PCI_MESON is not set
#
# PCI Endpoint
#
# CONFIG_PCI_ENDPOINT is not set
#
# PCI switch controller drivers
#
# CONFIG_PCI_SW_SWITCHTEC is not setCONFIG_PCCARD=y
# CONFIG_PCMCIA is not setCONFIG_CARDBUS=y
#
# PC-card bridges
#CONFIG_YENTA=mCONFIG_YENTA_O2=yCONFIG_YENTA_RICOH=yCONFIG_YENTA_TI=yCONFIG_YENTA_ENE_TUNE=yCONFIG_YENTA_TOSHIBA=y
# CONFIG_RAPIDIO is not set
#
# Generic Driver Options
#CONFIG_UEVENT_HELPER=yCONFIG_UEVENT_HELPER_PATH=""CONFIG_DEVTMPFS=yCONFIG_DEVTMPFS_MOUNT=yCONFIG_STANDALONE=yCONFIG_PREVENT_FIRMWARE_BUILD=y
#
# Firmware loader
#CONFIG_FW_LOADER=yCONFIG_EXTRA_FIRMWARE=""CONFIG_FW_LOADER_USER_HELPER=yCONFIG_FW_LOADER_USER_HELPER_FALLBACK=yCONFIG_WANT_DEV_COREDUMP=yCONFIG_ALLOW_DEV_COREDUMP=yCONFIG_DEV_COREDUMP=y
# CONFIG_DEBUG_DRIVER is not set
# CONFIG_DEBUG_DEVRES is not set
# CONFIG_DEBUG_TEST_DRIVER_REMOVE is not set
# CONFIG_TEST_ASYNC_DRIVER_PROBE is not setCONFIG_SYS_HYPERVISOR=yCONFIG_GENERIC_CPU_AUTOPROBE=yCONFIG_GENERIC_CPU_VULNERABILITIES=yCONFIG_REGMAP=yCONFIG_REGMAP_I2C=yCONFIG_REGMAP_SPI=yCONFIG_REGMAP_IRQ=yCONFIG_DMA_SHARED_BUFFER=y
# CONFIG_DMA_FENCE_TRACE is not set
#
# Bus devices
#CONFIG_CONNECTOR=yCONFIG_PROC_EVENTS=y
# CONFIG_GNSS is not setCONFIG_MTD=m
# CONFIG_MTD_TESTS is not set
# CONFIG_MTD_CMDLINE_PARTS is not set
# CONFIG_MTD_AR7_PARTS is not set
#
# Partition parsers
#
# CONFIG_MTD_REDBOOT_PARTS is not set
#
# User Modules And Translation Layers
#CONFIG_MTD_BLKDEVS=mCONFIG_MTD_BLOCK=m
# CONFIG_MTD_BLOCK_RO is not set
# CONFIG_FTL is not set
# CONFIG_NFTL is not set
# CONFIG_INFTL is not set
# CONFIG_RFD_FTL is not set
# CONFIG_SSFDC is not set
# CONFIG_SM_FTL is not set
# CONFIG_MTD_OOPS is not set
# CONFIG_MTD_SWAP is not set
# CONFIG_MTD_PARTITIONED_MASTER is not set
#
# RAM/ROM/Flash chip drivers
#
# CONFIG_MTD_CFI is not set
# CONFIG_MTD_JEDECPROBE is not setCONFIG_MTD_MAP_BANK_WIDTH_1=yCONFIG_MTD_MAP_BANK_WIDTH_2=yCONFIG_MTD_MAP_BANK_WIDTH_4=yCONFIG_MTD_CFI_I1=yCONFIG_MTD_CFI_I2=y
# CONFIG_MTD_RAM is not set
# CONFIG_MTD_ROM is not set
# CONFIG_MTD_ABSENT is not set
#
# Mapping drivers for chip access
#
# CONFIG_MTD_COMPLEX_MAPPINGS is not set
# CONFIG_MTD_INTEL_VR_NOR is not set
# CONFIG_MTD_PLATRAM is not set
#
# Self-contained MTD device drivers
#
# CONFIG_MTD_PMC551 is not set
# CONFIG_MTD_DATAFLASH is not set
# CONFIG_MTD_MCHP23K256 is not set
# CONFIG_MTD_SST25L is not set
# CONFIG_MTD_SLRAM is not set
# CONFIG_MTD_PHRAM is not set
# CONFIG_MTD_MTDRAM is not set
# CONFIG_MTD_BLOCK2MTD is not set
#
# Disk-On-Chip Device Drivers
#
# CONFIG_MTD_DOCG3 is not set
# CONFIG_MTD_ONENAND is not set
# CONFIG_MTD_NAND is not set
# CONFIG_MTD_SPI_NAND is not set
#
# LPDDR & LPDDR2 PCM memory drivers
#
# CONFIG_MTD_LPDDR is not set
# CONFIG_MTD_SPI_NOR is not setCONFIG_MTD_UBI=mCONFIG_MTD_UBI_WL_THRESHOLD=4096CONFIG_MTD_UBI_BEB_LIMIT=20
# CONFIG_MTD_UBI_FASTMAP is not set
# CONFIG_MTD_UBI_GLUEBI is not set
# CONFIG_MTD_UBI_BLOCK is not set
# CONFIG_OF is not setCONFIG_ARCH_MIGHT_HAVE_PC_PARPORT=yCONFIG_PARPORT=mCONFIG_PARPORT_PC=mCONFIG_PARPORT_SERIAL=m
# CONFIG_PARPORT_PC_FIFO is not set
# CONFIG_PARPORT_PC_SUPERIO is not set
# CONFIG_PARPORT_AX88796 is not setCONFIG_PARPORT_1284=yCONFIG_PARPORT_NOT_PC=yCONFIG_PNP=y
# CONFIG_PNP_DEBUG_MESSAGES is not set
#
# Protocols
#CONFIG_PNPACPI=yCONFIG_BLK_DEV=yCONFIG_BLK_DEV_NULL_BLK=mCONFIG_BLK_DEV_NULL_BLK_FAULT_INJECTION=yCONFIG_BLK_DEV_FD=mCONFIG_CDROM=m
# CONFIG_PARIDE is not setCONFIG_BLK_DEV_PCIESSD_MTIP32XX=m
# CONFIG_ZRAM is not set
# CONFIG_BLK_DEV_UMEM is not setCONFIG_BLK_DEV_LOOP=mCONFIG_BLK_DEV_LOOP_MIN_COUNT=0
# CONFIG_BLK_DEV_CRYPTOLOOP is not set
# CONFIG_BLK_DEV_DRBD is not setCONFIG_BLK_DEV_NBD=m
# CONFIG_BLK_DEV_SKD is not setCONFIG_BLK_DEV_SX8=mCONFIG_BLK_DEV_RAM=mCONFIG_BLK_DEV_RAM_COUNT=16CONFIG_BLK_DEV_RAM_SIZE=16384CONFIG_CDROM_PKTCDVD=mCONFIG_CDROM_PKTCDVD_BUFFERS=8
# CONFIG_CDROM_PKTCDVD_WCACHE is not setCONFIG_ATA_OVER_ETH=mCONFIG_XEN_BLKDEV_FRONTEND=mCONFIG_VIRTIO_BLK=y
# CONFIG_VIRTIO_BLK_SCSI is not setCONFIG_BLK_DEV_RBD=m
# CONFIG_BLK_DEV_RSXX is not set
#
# NVME Support
#CONFIG_NVME_CORE=mCONFIG_BLK_DEV_NVME=mCONFIG_NVME_MULTIPATH=yCONFIG_NVME_FABRICS=mCONFIG_NVME_RDMA=mCONFIG_NVME_FC=m
# CONFIG_NVME_TCP is not setCONFIG_NVME_TARGET=mCONFIG_NVME_TARGET_LOOP=mCONFIG_NVME_TARGET_RDMA=mCONFIG_NVME_TARGET_FC=mCONFIG_NVME_TARGET_FCLOOP=m
# CONFIG_NVME_TARGET_TCP is not set
#
# Misc devices
#CONFIG_SENSORS_LIS3LV02D=m
# CONFIG_AD525X_DPOT is not set
# CONFIG_DUMMY_IRQ is not set
# CONFIG_IBM_ASM is not set
# CONFIG_PHANTOM is not setCONFIG_SGI_IOC4=mCONFIG_TIFM_CORE=mCONFIG_TIFM_7XX1=m
# CONFIG_ICS932S401 is not setCONFIG_ENCLOSURE_SERVICES=mCONFIG_SGI_XP=mCONFIG_HP_ILO=mCONFIG_SGI_GRU=m
# CONFIG_SGI_GRU_DEBUG is not setCONFIG_APDS9802ALS=mCONFIG_ISL29003=mCONFIG_ISL29020=mCONFIG_SENSORS_TSL2550=mCONFIG_SENSORS_BH1770=mCONFIG_SENSORS_APDS990X=m
# CONFIG_HMC6352 is not set
# CONFIG_DS1682 is not setCONFIG_VMWARE_BALLOON=m
# CONFIG_USB_SWITCH_FSA9480 is not set
# CONFIG_LATTICE_ECP3_CONFIG is not set
# CONFIG_SRAM is not set
# CONFIG_PCI_ENDPOINT_TEST is not setCONFIG_PVPANIC=y
# CONFIG_C2PORT is not set
#
# EEPROM support
#CONFIG_EEPROM_AT24=m
# CONFIG_EEPROM_AT25 is not setCONFIG_EEPROM_LEGACY=mCONFIG_EEPROM_MAX6875=mCONFIG_EEPROM_93CX6=m
# CONFIG_EEPROM_93XX46 is not set
# CONFIG_EEPROM_IDT_89HPESX is not set
# CONFIG_EEPROM_EE1004 is not setCONFIG_CB710_CORE=m
# CONFIG_CB710_DEBUG is not setCONFIG_CB710_DEBUG_ASSUMPTIONS=y
#
# Texas Instruments shared transport line discipline #
# CONFIG_TI_ST is not set CONFIG_SENSORS_LIS3_I2C=m CONFIG_ALTERA_STAPL=mCONFIG_INTEL_MEI=mCONFIG_INTEL_MEI_ME=m
# CONFIG_INTEL_MEI_TXE is not set
# CONFIG_INTEL_MEI_HDCP is not set CONFIG_VMWARE_VMCI=m
#
# Intel MIC & related support
#
#
# Intel MIC Bus Driver
#
# CONFIG_INTEL_MIC_BUS is not set
#
# SCIF Bus Driver
#
# CONFIG_SCIF_BUS is not set
#
# VOP Bus Driver
#
# CONFIG_VOP_BUS is not set
#
# Intel MIC Host Driver
#
#
# Intel MIC Card Driver
#
#
# SCIF Driver
#
#
# Intel MIC Coprocessor State Management (COSM) Drivers #
#
# VOP Driver
#
# CONFIG_GENWQE is not set
# CONFIG_ECHO is not set
# CONFIG_MISC_ALCOR_PCI is not set
# CONFIG_MISC_RTSX_PCI is not set
# CONFIG_MISC_RTSX_USB is not set
# CONFIG_HABANA_AI is not setCONFIG_HAVE_IDE=y
# CONFIG_IDE is not set
#
# SCSI device support
#CONFIG_SCSI_MOD=yCONFIG_RAID_ATTRS=mCONFIG_SCSI=yCONFIG_SCSI_DMA=yCONFIG_SCSI_NETLINK=yCONFIG_SCSI_PROC_FS=y
#
# SCSI support type (disk, tape, CD-ROM)
#CONFIG_BLK_DEV_SD=mCONFIG_CHR_DEV_ST=mCONFIG_CHR_DEV_OSST=mCONFIG_BLK_DEV_SR=mCONFIG_BLK_DEV_SR_VENDOR=yCONFIG_CHR_DEV_SG=mCONFIG_CHR_DEV_SCH=mCONFIG_SCSI_ENCLOSURE=mCONFIG_SCSI_CONSTANTS=yCONFIG_SCSI_LOGGING=yCONFIG_SCSI_SCAN_ASYNC=y
#
# SCSI Transports
#CONFIG_SCSI_SPI_ATTRS=mCONFIG_SCSI_FC_ATTRS=mCONFIG_SCSI_ISCSI_ATTRS=mCONFIG_SCSI_SAS_ATTRS=mCONFIG_SCSI_SAS_LIBSAS=mCONFIG_SCSI_SAS_ATA=yCONFIG_SCSI_SAS_HOST_SMP=yCONFIG_SCSI_SRP_ATTRS=mCONFIG_SCSI_LOWLEVEL=yCONFIG_ISCSI_TCP=mCONFIG_ISCSI_BOOT_SYSFS=mCONFIG_SCSI_CXGB3_ISCSI=mCONFIG_SCSI_CXGB4_ISCSI=mCONFIG_SCSI_BNX2_ISCSI=mCONFIG_SCSI_BNX2X_FCOE=mCONFIG_BE2ISCSI=m
# CONFIG_BLK_DEV_3W_XXXX_RAID is not setCONFIG_SCSI_HPSA=mCONFIG_SCSI_3W_9XXX=mCONFIG_SCSI_3W_SAS=m
# CONFIG_SCSI_ACARD is not setCONFIG_SCSI_AACRAID=m
# CONFIG_SCSI_AIC7XXX is not setCONFIG_SCSI_AIC79XX=mCONFIG_AIC79XX_CMDS_PER_DEVICE=4CONFIG_AIC79XX_RESET_DELAY_MS=15000
# CONFIG_AIC79XX_DEBUG_ENABLE is not setCONFIG_AIC79XX_DEBUG_MASK=0
# CONFIG_AIC79XX_REG_PRETTY_PRINT is not set
# CONFIG_SCSI_AIC94XX is not setCONFIG_SCSI_MVSAS=m
# CONFIG_SCSI_MVSAS_DEBUG is not setCONFIG_SCSI_MVSAS_TASKLET=yCONFIG_SCSI_MVUMI=m
# CONFIG_SCSI_DPT_I2O is not set
# CONFIG_SCSI_ADVANSYS is not setCONFIG_SCSI_ARCMSR=m
# CONFIG_SCSI_ESAS2R is not set
# CONFIG_MEGARAID_NEWGEN is not set
# CONFIG_MEGARAID_LEGACY is not setCONFIG_MEGARAID_SAS=mCONFIG_SCSI_MPT3SAS=mCONFIG_SCSI_MPT2SAS_MAX_SGE=128CONFIG_SCSI_MPT3SAS_MAX_SGE=128CONFIG_SCSI_MPT2SAS=mCONFIG_SCSI_SMARTPQI=mCONFIG_SCSI_UFSHCD=mCONFIG_SCSI_UFSHCD_PCI=m
# CONFIG_SCSI_UFS_DWC_TC_PCI is not set
# CONFIG_SCSI_UFSHCD_PLATFORM is not set
# CONFIG_SCSI_UFS_BSG is not setCONFIG_SCSI_HPTIOP=m
# CONFIG_SCSI_BUSLOGIC is not set
# CONFIG_SCSI_MYRB is not set
# CONFIG_SCSI_MYRS is not setCONFIG_VMWARE_PVSCSI=m
# CONFIG_XEN_SCSI_FRONTEND is not setCONFIG_HYPERV_STORAGE=mCONFIG_LIBFC=mCONFIG_LIBFCOE=mCONFIG_FCOE=mCONFIG_FCOE_FNIC=m
# CONFIG_SCSI_SNIC is not set
# CONFIG_SCSI_DMX3191D is not set
# CONFIG_SCSI_GDTH is not setCONFIG_SCSI_ISCI=m
# CONFIG_SCSI_IPS is not setCONFIG_SCSI_INITIO=m
# CONFIG_SCSI_INIA100 is not set
# CONFIG_SCSI_PPA is not set
# CONFIG_SCSI_IMM is not setCONFIG_SCSI_STEX=m
# CONFIG_SCSI_SYM53C8XX_2 is not set
# CONFIG_SCSI_IPR is not set
# CONFIG_SCSI_QLOGIC_1280 is not setCONFIG_SCSI_QLA_FC=mCONFIG_TCM_QLA2XXX=m
# CONFIG_TCM_QLA2XXX_DEBUG is not setCONFIG_SCSI_QLA_ISCSI=mCONFIG_QEDI=mCONFIG_QEDF=m
# CONFIG_SCSI_LPFC is not set
# CONFIG_SCSI_DC395x is not set
# CONFIG_SCSI_AM53C974 is not set
# CONFIG_SCSI_WD719X is not setCONFIG_SCSI_DEBUG=mCONFIG_SCSI_PMCRAID=mCONFIG_SCSI_PM8001=mCONFIG_SCSI_BFA_FC=mCONFIG_SCSI_VIRTIO=mCONFIG_SCSI_CHELSIO_FCOE=mCONFIG_SCSI_DH=yCONFIG_SCSI_DH_RDAC=yCONFIG_SCSI_DH_HP_SW=yCONFIG_SCSI_DH_EMC=yCONFIG_SCSI_DH_ALUA=yCONFIG_ATA=mCONFIG_ATA_VERBOSE_ERROR=yCONFIG_ATA_ACPI=y
# CONFIG_SATA_ZPODD is not setCONFIG_SATA_PMP=y
#
# Controllers with non-SFF native interface
#CONFIG_SATA_AHCI=mCONFIG_SATA_MOBILE_LPM_POLICY=0CONFIG_SATA_AHCI_PLATFORM=m
# CONFIG_SATA_INIC162X is not setCONFIG_SATA_ACARD_AHCI=mCONFIG_SATA_SIL24=mCONFIG_ATA_SFF=y
#
# SFF controllers with custom DMA interface
#CONFIG_PDC_ADMA=mCONFIG_SATA_QSTOR=mCONFIG_SATA_SX4=mCONFIG_ATA_BMDMA=y
#
# SATA SFF controllers with BMDMA
#CONFIG_ATA_PIIX=m
# CONFIG_SATA_DWC is not setCONFIG_SATA_MV=mCONFIG_SATA_NV=mCONFIG_SATA_PROMISE=mCONFIG_SATA_SIL=mCONFIG_SATA_SIS=mCONFIG_SATA_SVW=mCONFIG_SATA_ULI=mCONFIG_SATA_VIA=mCONFIG_SATA_VITESSE=m
#
# PATA SFF controllers with BMDMA
#CONFIG_PATA_ALI=mCONFIG_PATA_AMD=mCONFIG_PATA_ARTOP=mCONFIG_PATA_ATIIXP=mCONFIG_PATA_ATP867X=mCONFIG_PATA_CMD64X=m
# CONFIG_PATA_CYPRESS is not set
# CONFIG_PATA_EFAR is not setCONFIG_PATA_HPT366=mCONFIG_PATA_HPT37X=mCONFIG_PATA_HPT3X2N=mCONFIG_PATA_HPT3X3=m
# CONFIG_PATA_HPT3X3_DMA is not setCONFIG_PATA_IT8213=mCONFIG_PATA_IT821X=mCONFIG_PATA_JMICRON=mCONFIG_PATA_MARVELL=mCONFIG_PATA_NETCELL=mCONFIG_PATA_NINJA32=m
# CONFIG_PATA_NS87415 is not setCONFIG_PATA_OLDPIIX=m
# CONFIG_PATA_OPTIDMA is not setCONFIG_PATA_PDC2027X=mCONFIG_PATA_PDC_OLD=m
# CONFIG_PATA_RADISYS is not setCONFIG_PATA_RDC=mCONFIG_PATA_SCH=mCONFIG_PATA_SERVERWORKS=mCONFIG_PATA_SIL680=mCONFIG_PATA_SIS=mCONFIG_PATA_TOSHIBA=m
# CONFIG_PATA_TRIFLEX is not setCONFIG_PATA_VIA=m
# CONFIG_PATA_WINBOND is not set
#
# PIO-only SFF controllers
#
# CONFIG_PATA_CMD640_PCI is not set
# CONFIG_PATA_MPIIX is not set
# CONFIG_PATA_NS87410 is not set
# CONFIG_PATA_OPTI is not set
# CONFIG_PATA_PLATFORM is not set
# CONFIG_PATA_RZ1000 is not set
#
# Generic fallback / legacy drivers
#CONFIG_PATA_ACPI=mCONFIG_ATA_GENERIC=m
# CONFIG_PATA_LEGACY is not set CONFIG_MD=yCONFIG_BLK_DEV_MD=yCONFIG_MD_AUTODETECT=yCONFIG_MD_LINEAR=mCONFIG_MD_RAID0=mCONFIG_MD_RAID1=mCONFIG_MD_RAID10=mCONFIG_MD_RAID456=mCONFIG_MD_MULTIPATH=mCONFIG_MD_FAULTY=m
# CONFIG_MD_CLUSTER is not set
# CONFIG_BCACHE is not set CONFIG_BLK_DEV_DM_BUILTIN=y CONFIG_BLK_DEV_DM=mCONFIG_DM_DEBUG=yCONFIG_DM_BUFIO=m
# CONFIG_DM_DEBUG_BLOCK_MANAGER_LOCKING is not set CONFIG_DM_BIO_PRISON=m CONFIG_DM_PERSISTENT_DATA=m
# CONFIG_DM_UNSTRIPED is not set CONFIG_DM_CRYPT=mCONFIG_DM_SNAPSHOT=m CONFIG_DM_THIN_PROVISIONING=m CONFIG_DM_CACHE=mCONFIG_DM_CACHE_SMQ=m
# CONFIG_DM_WRITECACHE is not set CONFIG_DM_ERA=mCONFIG_DM_MIRROR=m CONFIG_DM_LOG_USERSPACE=mCONFIG_DM_RAID=mCONFIG_DM_ZERO=mCONFIG_DM_MULTIPATH=m CONFIG_DM_MULTIPATH_QL=m CONFIG_DM_MULTIPATH_ST=mCONFIG_DM_DELAY=mCONFIG_DM_UEVENT=yCONFIG_DM_FLAKEY=mCONFIG_DM_VERITY=m
# CONFIG_DM_VERITY_FEC is not set CONFIG_DM_SWITCH=mCONFIG_DM_LOG_WRITES=m
# CONFIG_DM_INTEGRITY is not set
# CONFIG_DM_ZONED is not set CONFIG_TARGET_CORE=mCONFIG_TCM_IBLOCK=mCONFIG_TCM_FILEIO=mCONFIG_TCM_PSCSI=mCONFIG_TCM_USER2=m CONFIG_LOOPBACK_TARGET=mCONFIG_TCM_FC=mCONFIG_ISCSI_TARGET=m CONFIG_ISCSI_TARGET_CXGB4=m
# CONFIG_SBP_TARGET is not set CONFIG_FUSION=yCONFIG_FUSION_SPI=m
# CONFIG_FUSION_FC is not set CONFIG_FUSION_SAS=m CONFIG_FUSION_MAX_SGE=128 CONFIG_FUSION_CTL=m CONFIG_FUSION_LOGGING=y
#
# IEEE 1394 (FireWire) support
#CONFIG_FIREWIRE=mCONFIG_FIREWIRE_OHCI=mCONFIG_FIREWIRE_SBP2=mCONFIG_FIREWIRE_NET=m
# CONFIG_FIREWIRE_NOSY is not setCONFIG_MACINTOSH_DRIVERS=yCONFIG_MAC_EMUMOUSEBTN=yCONFIG_NETDEVICES=yCONFIG_MII=yCONFIG_NET_CORE=yCONFIG_BONDING=mCONFIG_DUMMY=m
# CONFIG_EQUALIZER is not setCONFIG_NET_FC=yCONFIG_IFB=mCONFIG_NET_TEAM=mCONFIG_NET_TEAM_MODE_BROADCAST=mCONFIG_NET_TEAM_MODE_ROUNDROBIN=mCONFIG_NET_TEAM_MODE_RANDOM=mCONFIG_NET_TEAM_MODE_ACTIVEBACKUP=mCONFIG_NET_TEAM_MODE_LOADBALANCE=mCONFIG_MACVLAN=mCONFIG_MACVTAP=m
# CONFIG_IPVLAN is not setCONFIG_VXLAN=mCONFIG_GENEVE=m
# CONFIG_GTP is not setCONFIG_MACSEC=yCONFIG_NETCONSOLE=mCONFIG_NETCONSOLE_DYNAMIC=yCONFIG_NETPOLL=yCONFIG_NET_POLL_CONTROLLER=yCONFIG_NTB_NETDEV=mCONFIG_TUN=mCONFIG_TAP=m
# CONFIG_TUN_VNET_CROSS_LE is not setCONFIG_VETH=mCONFIG_VIRTIO_NET=mCONFIG_NLMON=mCONFIG_NET_VRF=yCONFIG_VSOCKMON=m
# CONFIG_ARCNET is not set
# CONFIG_ATM_DRIVERS is not set
#
# CAIF transport drivers
#
#
# Distributed Switch Architecture drivers
#CONFIG_ETHERNET=yCONFIG_MDIO=y
# CONFIG_NET_VENDOR_3COM is not set
# CONFIG_NET_VENDOR_ADAPTEC is not setCONFIG_NET_VENDOR_AGERE=y
# CONFIG_ET131X is not setCONFIG_NET_VENDOR_ALACRITECH=y
# CONFIG_SLICOSS is not set
# CONFIG_NET_VENDOR_ALTEON is not set
# CONFIG_ALTERA_TSE is not setCONFIG_NET_VENDOR_AMAZON=yCONFIG_ENA_ETHERNET=mCONFIG_NET_VENDOR_AMD=yCONFIG_AMD8111_ETH=mCONFIG_PCNET32=mCONFIG_AMD_XGBE=m
# CONFIG_AMD_XGBE_DCB is not setCONFIG_AMD_XGBE_HAVE_ECC=yCONFIG_NET_VENDOR_AQUANTIA=yCONFIG_AQTION=mCONFIG_NET_VENDOR_ARC=yCONFIG_NET_VENDOR_ATHEROS=yCONFIG_ATL2=mCONFIG_ATL1=mCONFIG_ATL1E=mCONFIG_ATL1C=mCONFIG_ALX=mCONFIG_NET_VENDOR_AURORA=y
# CONFIG_AURORA_NB8800 is not setCONFIG_NET_VENDOR_BROADCOM=yCONFIG_B44=mCONFIG_B44_PCI_AUTOSELECT=yCONFIG_B44_PCICORE_AUTOSELECT=yCONFIG_B44_PCI=y
# CONFIG_BCMGENET is not setCONFIG_BNX2=mCONFIG_CNIC=mCONFIG_TIGON3=yCONFIG_TIGON3_HWMON=yCONFIG_BNX2X=mCONFIG_BNX2X_SRIOV=y
# CONFIG_SYSTEMPORT is not setCONFIG_BNXT=mCONFIG_BNXT_SRIOV=yCONFIG_BNXT_FLOWER_OFFLOAD=yCONFIG_BNXT_DCB=yCONFIG_BNXT_HWMON=yCONFIG_NET_VENDOR_BROCADE=yCONFIG_BNA=mCONFIG_NET_VENDOR_CADENCE=yCONFIG_MACB=mCONFIG_MACB_USE_HWSTAMP=y
# CONFIG_MACB_PCI is not setCONFIG_NET_VENDOR_CAVIUM=y
# CONFIG_THUNDER_NIC_PF is not set
# CONFIG_THUNDER_NIC_VF is not set
# CONFIG_THUNDER_NIC_BGX is not set
# CONFIG_THUNDER_NIC_RGX is not setCONFIG_CAVIUM_PTP=yCONFIG_LIQUIDIO=mCONFIG_LIQUIDIO_VF=mCONFIG_NET_VENDOR_CHELSIO=y
# CONFIG_CHELSIO_T1 is not setCONFIG_CHELSIO_T3=mCONFIG_CHELSIO_T4=m
# CONFIG_CHELSIO_T4_DCB is not setCONFIG_CHELSIO_T4VF=mCONFIG_CHELSIO_LIB=mCONFIG_NET_VENDOR_CISCO=yCONFIG_ENIC=mCONFIG_NET_VENDOR_CORTINA=y
# CONFIG_CX_ECAT is not setCONFIG_DNET=mCONFIG_NET_VENDOR_DEC=yCONFIG_NET_TULIP=yCONFIG_DE2104X=mCONFIG_DE2104X_DSL=0CONFIG_TULIP=y
# CONFIG_TULIP_MWI is not setCONFIG_TULIP_MMIO=y
# CONFIG_TULIP_NAPI is not setCONFIG_DE4X5=mCONFIG_WINBOND_840=mCONFIG_DM9102=mCONFIG_ULI526X=mCONFIG_PCMCIA_XIRCOM=m
# CONFIG_NET_VENDOR_DLINK is not setCONFIG_NET_VENDOR_EMULEX=yCONFIG_BE2NET=mCONFIG_BE2NET_HWMON=yCONFIG_BE2NET_BE2=yCONFIG_BE2NET_BE3=yCONFIG_BE2NET_LANCER=yCONFIG_BE2NET_SKYHAWK=yCONFIG_NET_VENDOR_EZCHIP=y
# CONFIG_NET_VENDOR_HP is not setCONFIG_NET_VENDOR_HUAWEI=y
# CONFIG_HINIC is not set
# CONFIG_NET_VENDOR_I825XX is not setCONFIG_NET_VENDOR_INTEL=y
# CONFIG_E100 is not setCONFIG_E1000=yCONFIG_E1000E=yCONFIG_E1000E_HWTS=yCONFIG_IGB=yCONFIG_IGB_HWMON=yCONFIG_IGBVF=mCONFIG_IXGB=yCONFIG_IXGBE=yCONFIG_IXGBE_HWMON=yCONFIG_IXGBE_DCB=yCONFIG_IXGBEVF=mCONFIG_I40E=mCONFIG_I40E_DCB=yCONFIG_IAVF=mCONFIG_I40EVF=m
# CONFIG_ICE is not setCONFIG_FM10K=m
# CONFIG_IGC is not setCONFIG_JME=mCONFIG_NET_VENDOR_MARVELL=yCONFIG_MVMDIO=mCONFIG_SKGE=y
# CONFIG_SKGE_DEBUG is not setCONFIG_SKGE_GENESIS=yCONFIG_SKY2=m
# CONFIG_SKY2_DEBUG is not setCONFIG_NET_VENDOR_MELLANOX=yCONFIG_MLX4_EN=mCONFIG_MLX4_EN_DCB=yCONFIG_MLX4_CORE=mCONFIG_MLX4_DEBUG=yCONFIG_MLX4_CORE_GEN2=yCONFIG_MLX5_CORE=m
# CONFIG_MLX5_FPGA is not setCONFIG_MLX5_CORE_EN=yCONFIG_MLX5_EN_ARFS=yCONFIG_MLX5_EN_RXNFC=yCONFIG_MLX5_MPFS=yCONFIG_MLX5_ESWITCH=yCONFIG_MLX5_CORE_EN_DCB=yCONFIG_MLX5_CORE_IPOIB=yCONFIG_MLXSW_CORE=mCONFIG_MLXSW_CORE_HWMON=yCONFIG_MLXSW_CORE_THERMAL=yCONFIG_MLXSW_PCI=mCONFIG_MLXSW_I2C=mCONFIG_MLXSW_SWITCHIB=mCONFIG_MLXSW_SWITCHX2=mCONFIG_MLXSW_SPECTRUM=mCONFIG_MLXSW_SPECTRUM_DCB=yCONFIG_MLXSW_MINIMAL=mCONFIG_MLXFW=m
# CONFIG_NET_VENDOR_MICREL is not set
# CONFIG_NET_VENDOR_MICROCHIP is not setCONFIG_NET_VENDOR_MICROSEMI=y
# CONFIG_MSCC_OCELOT_SWITCH is not setCONFIG_NET_VENDOR_MYRI=yCONFIG_MYRI10GE=mCONFIG_MYRI10GE_DCA=y
# CONFIG_FEALNX is not set
# CONFIG_NET_VENDOR_NATSEMI is not setCONFIG_NET_VENDOR_NETERION=y
# CONFIG_S2IO is not set
# CONFIG_VXGE is not setCONFIG_NET_VENDOR_NETRONOME=yCONFIG_NFP=mCONFIG_NFP_APP_FLOWER=yCONFIG_NFP_APP_ABM_NIC=y
# CONFIG_NFP_DEBUG is not setCONFIG_NET_VENDOR_NI=y
# CONFIG_NI_XGE_MANAGEMENT_ENET is not set
# CONFIG_NET_VENDOR_NVIDIA is not setCONFIG_NET_VENDOR_OKI=yCONFIG_ETHOC=mCONFIG_NET_VENDOR_PACKET_ENGINES=y
# CONFIG_HAMACHI is not setCONFIG_YELLOWFIN=mCONFIG_NET_VENDOR_QLOGIC=yCONFIG_QLA3XXX=mCONFIG_QLCNIC=mCONFIG_QLCNIC_SRIOV=yCONFIG_QLCNIC_DCB=yCONFIG_QLCNIC_HWMON=yCONFIG_QLGE=mCONFIG_NETXEN_NIC=mCONFIG_QED=mCONFIG_QED_LL2=yCONFIG_QED_SRIOV=yCONFIG_QEDE=mCONFIG_QED_RDMA=yCONFIG_QED_ISCSI=yCONFIG_QED_FCOE=yCONFIG_QED_OOO=yCONFIG_NET_VENDOR_QUALCOMM=y
# CONFIG_QCOM_EMAC is not set
# CONFIG_RMNET is not set
# CONFIG_NET_VENDOR_RDC is not setCONFIG_NET_VENDOR_REALTEK=y
# CONFIG_ATP is not setCONFIG_8139CP=yCONFIG_8139TOO=y
# CONFIG_8139TOO_PIO is not set
# CONFIG_8139TOO_TUNE_TWISTER is not setCONFIG_8139TOO_8129=y
# CONFIG_8139_OLD_RX_RESET is not setCONFIG_R8169=yCONFIG_NET_VENDOR_RENESAS=yCONFIG_NET_VENDOR_ROCKER=yCONFIG_ROCKER=mCONFIG_NET_VENDOR_SAMSUNG=y
# CONFIG_SXGBE_ETH is not set
# CONFIG_NET_VENDOR_SEEQ is not setCONFIG_NET_VENDOR_SOLARFLARE=yCONFIG_SFC=mCONFIG_SFC_MTD=yCONFIG_SFC_MCDI_MON=yCONFIG_SFC_SRIOV=yCONFIG_SFC_MCDI_LOGGING=yCONFIG_SFC_FALCON=mCONFIG_SFC_FALCON_MTD=y
# CONFIG_NET_VENDOR_SILAN is not set
# CONFIG_NET_VENDOR_SIS is not setCONFIG_NET_VENDOR_SMSC=yCONFIG_EPIC100=m
# CONFIG_SMSC911X is not setCONFIG_SMSC9420=mCONFIG_NET_VENDOR_SOCIONEXT=y
# CONFIG_NET_VENDOR_STMICRO is not set
# CONFIG_NET_VENDOR_SUN is not setCONFIG_NET_VENDOR_SYNOPSYS=y
# CONFIG_DWC_XLGMAC is not set
# CONFIG_NET_VENDOR_TEHUTI is not setCONFIG_NET_VENDOR_TI=y
# CONFIG_TI_CPSW_PHY_SEL is not set
# CONFIG_TI_CPSW_ALE is not setCONFIG_TLAN=m
# CONFIG_NET_VENDOR_VIA is not set
# CONFIG_NET_VENDOR_WIZNET is not set
# CONFIG_FDDI is not set
# CONFIG_HIPPI is not set
# CONFIG_NET_SB1000 is not setCONFIG_MDIO_DEVICE=yCONFIG_MDIO_BUS=y
# CONFIG_MDIO_BCM_UNIMAC is not setCONFIG_MDIO_BITBANG=m
# CONFIG_MDIO_GPIO is not set
# CONFIG_MDIO_MSCC_MIIM is not set
# CONFIG_MDIO_THUNDER is not setCONFIG_PHYLIB=yCONFIG_SWPHY=y
# CONFIG_LED_TRIGGER_PHY is not set
#
# MII PHY device drivers
#CONFIG_AMD_PHY=m
# CONFIG_AQUANTIA_PHY is not set
# CONFIG_ASIX_PHY is not setCONFIG_AT803X_PHY=m
# CONFIG_BCM7XXX_PHY is not setCONFIG_BCM87XX_PHY=mCONFIG_BCM_NET_PHYLIB=mCONFIG_BROADCOM_PHY=mCONFIG_CICADA_PHY=m
# CONFIG_CORTINA_PHY is not setCONFIG_DAVICOM_PHY=m
# CONFIG_DP83822_PHY is not set
# CONFIG_DP83TC811_PHY is not set
# CONFIG_DP83848_PHY is not set
# CONFIG_DP83867_PHY is not setCONFIG_FIXED_PHY=yCONFIG_ICPLUS_PHY=m
# CONFIG_INTEL_XWAY_PHY is not setCONFIG_LSI_ET1011C_PHY=mCONFIG_LXT_PHY=mCONFIG_MARVELL_PHY=m
# CONFIG_MARVELL_10G_PHY is not setCONFIG_MICREL_PHY=m
# CONFIG_MICROCHIP_PHY is not set
# CONFIG_MICROCHIP_T1_PHY is not set
# CONFIG_MICROSEMI_PHY is not setCONFIG_NATIONAL_PHY=mCONFIG_QSEMI_PHY=mCONFIG_REALTEK_PHY=y
# CONFIG_RENESAS_PHY is not set
# CONFIG_ROCKCHIP_PHY is not setCONFIG_SMSC_PHY=mCONFIG_STE10XP=m
# CONFIG_TERANETICS_PHY is not setCONFIG_VITESSE_PHY=m
# CONFIG_XILINX_GMII2RGMII is not set
# CONFIG_MICREL_KS8995MA is not set
# CONFIG_PLIP is not setCONFIG_PPP=mCONFIG_PPP_BSDCOMP=mCONFIG_PPP_DEFLATE=mCONFIG_PPP_FILTER=yCONFIG_PPP_MPPE=mCONFIG_PPP_MULTILINK=yCONFIG_PPPOATM=mCONFIG_PPPOE=mCONFIG_PPTP=mCONFIG_PPPOL2TP=mCONFIG_PPP_ASYNC=mCONFIG_PPP_SYNC_TTY=mCONFIG_SLIP=mCONFIG_SLHC=mCONFIG_SLIP_COMPRESSED=yCONFIG_SLIP_SMART=y
# CONFIG_SLIP_MODE_SLIP6 is not setCONFIG_USB_NET_DRIVERS=yCONFIG_USB_CATC=yCONFIG_USB_KAWETH=yCONFIG_USB_PEGASUS=yCONFIG_USB_RTL8150=yCONFIG_USB_RTL8152=m
# CONFIG_USB_LAN78XX is not setCONFIG_USB_USBNET=yCONFIG_USB_NET_AX8817X=yCONFIG_USB_NET_AX88179_178A=mCONFIG_USB_NET_CDCETHER=yCONFIG_USB_NET_CDC_EEM=yCONFIG_USB_NET_CDC_NCM=mCONFIG_USB_NET_HUAWEI_CDC_NCM=mCONFIG_USB_NET_CDC_MBIM=mCONFIG_USB_NET_DM9601=y
# CONFIG_USB_NET_SR9700 is not set
# CONFIG_USB_NET_SR9800 is not setCONFIG_USB_NET_SMSC75XX=yCONFIG_USB_NET_SMSC95XX=yCONFIG_USB_NET_GL620A=yCONFIG_USB_NET_NET1080=yCONFIG_USB_NET_PLUSB=yCONFIG_USB_NET_MCS7830=yCONFIG_USB_NET_RNDIS_HOST=yCONFIG_USB_NET_CDC_SUBSET_ENABLE=yCONFIG_USB_NET_CDC_SUBSET=yCONFIG_USB_ALI_M5632=yCONFIG_USB_AN2720=yCONFIG_USB_BELKIN=yCONFIG_USB_ARMLINUX=yCONFIG_USB_EPSON2888=yCONFIG_USB_KC2190=yCONFIG_USB_NET_ZAURUS=yCONFIG_USB_NET_CX82310_ETH=mCONFIG_USB_NET_KALMIA=mCONFIG_USB_NET_QMI_WWAN=mCONFIG_USB_HSO=mCONFIG_USB_NET_INT51X1=yCONFIG_USB_IPHETH=yCONFIG_USB_SIERRA_NET=yCONFIG_USB_VL600=m
# CONFIG_USB_NET_CH9200 is not set
# CONFIG_USB_NET_AQC111 is not setCONFIG_WLAN=y
# CONFIG_WIRELESS_WDS is not setCONFIG_WLAN_VENDOR_ADMTEK=y
# CONFIG_ADM8211 is not setCONFIG_ATH_COMMON=mCONFIG_WLAN_VENDOR_ATH=y
# CONFIG_ATH_DEBUG is not set
# CONFIG_ATH5K is not set
# CONFIG_ATH5K_PCI is not setCONFIG_ATH9K_HW=mCONFIG_ATH9K_COMMON=mCONFIG_ATH9K_COMMON_DEBUG=yCONFIG_ATH9K_BTCOEX_SUPPORT=yCONFIG_ATH9K=mCONFIG_ATH9K_PCI=yCONFIG_ATH9K_AHB=yCONFIG_ATH9K_DEBUGFS=y
# CONFIG_ATH9K_STATION_STATISTICS is not set
# CONFIG_ATH9K_DYNACK is not setCONFIG_ATH9K_WOW=yCONFIG_ATH9K_RFKILL=y
# CONFIG_ATH9K_CHANNEL_CONTEXT is not setCONFIG_ATH9K_PCOEM=yCONFIG_ATH9K_HTC=m
# CONFIG_ATH9K_HTC_DEBUGFS is not set
# CONFIG_ATH9K_HWRNG is not set
# CONFIG_ATH9K_COMMON_SPECTRAL is not setCONFIG_CARL9170=mCONFIG_CARL9170_LEDS=y
# CONFIG_CARL9170_DEBUGFS is not setCONFIG_CARL9170_WPC=y
# CONFIG_CARL9170_HWRNG is not set
# CONFIG_ATH6KL is not set
# CONFIG_AR5523 is not setCONFIG_WIL6210=mCONFIG_WIL6210_ISR_COR=yCONFIG_WIL6210_TRACING=yCONFIG_WIL6210_DEBUGFS=yCONFIG_ATH10K=mCONFIG_ATH10K_CE=yCONFIG_ATH10K_PCI=m
# CONFIG_ATH10K_SDIO is not set
# CONFIG_ATH10K_USB is not set
# CONFIG_ATH10K_DEBUG is not setCONFIG_ATH10K_DEBUGFS=y
# CONFIG_ATH10K_SPECTRAL is not set
# CONFIG_ATH10K_TRACING is not set
# CONFIG_WCN36XX is not setCONFIG_WLAN_VENDOR_ATMEL=y
# CONFIG_ATMEL is not set
# CONFIG_AT76C50X_USB is not setCONFIG_WLAN_VENDOR_BROADCOM=y
# CONFIG_B43 is not set
# CONFIG_B43LEGACY is not setCONFIG_BRCMUTIL=mCONFIG_BRCMSMAC=mCONFIG_BRCMFMAC=mCONFIG_BRCMFMAC_PROTO_BCDC=yCONFIG_BRCMFMAC_PROTO_MSGBUF=yCONFIG_BRCMFMAC_SDIO=yCONFIG_BRCMFMAC_USB=yCONFIG_BRCMFMAC_PCIE=y
# CONFIG_BRCM_TRACING is not set
# CONFIG_BRCMDBG is not setCONFIG_WLAN_VENDOR_CISCO=y
# CONFIG_AIRO is not setCONFIG_WLAN_VENDOR_INTEL=y
# CONFIG_IPW2100 is not set
# CONFIG_IPW2200 is not setCONFIG_IWLEGACY=mCONFIG_IWL4965=mCONFIG_IWL3945=m
#
# iwl3945 / iwl4965 Debugging Options # CONFIG_IWLEGACY_DEBUG=y CONFIG_IWLEGACY_DEBUGFS=y CONFIG_IWLWIFI=m CONFIG_IWLWIFI_LEDS=y CONFIG_IWLDVM=m CONFIG_IWLMVM=m CONFIG_IWLWIFI_OPMODE_MODULAR=y # CONFIG_IWLWIFI_BCAST_FILTERING is not set # CONFIG_IWLWIFI_PCIE_RTPM is not set
#
# Debugging Options
#
# CONFIG_IWLWIFI_DEBUG is not setCONFIG_IWLWIFI_DEBUGFS=y
# CONFIG_IWLWIFI_DEVICE_TRACING is not setCONFIG_WLAN_VENDOR_INTERSIL=y
# CONFIG_HOSTAP is not set
# CONFIG_HERMES is not set
# CONFIG_P54_COMMON is not set
# CONFIG_PRISM54 is not setCONFIG_WLAN_VENDOR_MARVELL=y
# CONFIG_LIBERTAS is not set
# CONFIG_LIBERTAS_THINFIRM is not setCONFIG_MWIFIEX=mCONFIG_MWIFIEX_SDIO=mCONFIG_MWIFIEX_PCIE=mCONFIG_MWIFIEX_USB=mCONFIG_MWL8K=mCONFIG_WLAN_VENDOR_MEDIATEK=y
# CONFIG_MT7601U is not set
# CONFIG_MT76x0U is not set
# CONFIG_MT76x0E is not set
# CONFIG_MT76x2E is not set
# CONFIG_MT76x2U is not set
# CONFIG_MT7603E is not setCONFIG_WLAN_VENDOR_RALINK=yCONFIG_RT2X00=m
# CONFIG_RT2400PCI is not set
# CONFIG_RT2500PCI is not setCONFIG_RT61PCI=mCONFIG_RT2800PCI=mCONFIG_RT2800PCI_RT33XX=yCONFIG_RT2800PCI_RT35XX=yCONFIG_RT2800PCI_RT53XX=yCONFIG_RT2800PCI_RT3290=y
# CONFIG_RT2500USB is not setCONFIG_RT73USB=mCONFIG_RT2800USB=mCONFIG_RT2800USB_RT33XX=yCONFIG_RT2800USB_RT35XX=yCONFIG_RT2800USB_RT3573=yCONFIG_RT2800USB_RT53XX=yCONFIG_RT2800USB_RT55XX=yCONFIG_RT2800USB_UNKNOWN=yCONFIG_RT2800_LIB=mCONFIG_RT2800_LIB_MMIO=mCONFIG_RT2X00_LIB_MMIO=mCONFIG_RT2X00_LIB_PCI=mCONFIG_RT2X00_LIB_USB=mCONFIG_RT2X00_LIB=mCONFIG_RT2X00_LIB_FIRMWARE=yCONFIG_RT2X00_LIB_CRYPTO=yCONFIG_RT2X00_LIB_LEDS=yCONFIG_RT2X00_LIB_DEBUGFS=y
# CONFIG_RT2X00_DEBUG is not setCONFIG_WLAN_VENDOR_REALTEK=y
# CONFIG_RTL8180 is not setCONFIG_RTL8187=mCONFIG_RTL8187_LEDS=yCONFIG_RTL_CARDS=mCONFIG_RTL8192CE=mCONFIG_RTL8192SE=mCONFIG_RTL8192DE=mCONFIG_RTL8723AE=mCONFIG_RTL8723BE=mCONFIG_RTL8188EE=mCONFIG_RTL8192EE=mCONFIG_RTL8821AE=mCONFIG_RTL8192CU=mCONFIG_RTLWIFI=mCONFIG_RTLWIFI_PCI=mCONFIG_RTLWIFI_USB=m
# CONFIG_RTLWIFI_DEBUG is not setCONFIG_RTL8192C_COMMON=mCONFIG_RTL8723_COMMON=mCONFIG_RTLBTCOEXIST=m
# CONFIG_RTL8XXXU is not setCONFIG_WLAN_VENDOR_RSI=y
# CONFIG_RSI_91X is not setCONFIG_WLAN_VENDOR_ST=y
# CONFIG_CW1200 is not setCONFIG_WLAN_VENDOR_TI=y
# CONFIG_WL1251 is not set
# CONFIG_WL12XX is not set
# CONFIG_WL18XX is not set
# CONFIG_WLCORE is not setCONFIG_WLAN_VENDOR_ZYDAS=y
# CONFIG_USB_ZD1201 is not set
# CONFIG_ZD1211RW is not setCONFIG_WLAN_VENDOR_QUANTENNA=y
# CONFIG_QTNFMAC_PCIE is not setCONFIG_MAC80211_HWSIM=m
# CONFIG_USB_NET_RNDIS_WLAN is not set
# CONFIG_VIRT_WIFI is not set
#
# Enable WiMAX (Networking options) to see the WiMAX drivers #CONFIG_WAN=y
# CONFIG_LANMEDIA is not setCONFIG_HDLC=mCONFIG_HDLC_RAW=m
# CONFIG_HDLC_RAW_ETH is not set CONFIG_HDLC_CISCO=mCONFIG_HDLC_FR=mCONFIG_HDLC_PPP=m
#
# X.25/LAPB support is disabled
#
# CONFIG_PCI200SYN is not set
# CONFIG_WANXL is not set
# CONFIG_PC300TOO is not set
# CONFIG_FARSYNC is not set
# CONFIG_DSCC4 is not setCONFIG_DLCI=mCONFIG_DLCI_MAX=8
# CONFIG_SBNI is not setCONFIG_IEEE802154_DRIVERS=mCONFIG_IEEE802154_FAKELB=m
# CONFIG_IEEE802154_AT86RF230 is not set
# CONFIG_IEEE802154_MRF24J40 is not set
# CONFIG_IEEE802154_CC2520 is not set
# CONFIG_IEEE802154_ATUSB is not set
# CONFIG_IEEE802154_ADF7242 is not set
# CONFIG_IEEE802154_CA8210 is not set
# CONFIG_IEEE802154_MCR20A is not set
# CONFIG_IEEE802154_HWSIM is not setCONFIG_XEN_NETDEV_FRONTEND=mCONFIG_VMXNET3=mCONFIG_FUJITSU_ES=mCONFIG_THUNDERBOLT_NET=mCONFIG_HYPERV_NET=mCONFIG_NETDEVSIM=mCONFIG_NET_FAILOVER=mCONFIG_ISDN=yCONFIG_ISDN_I4L=mCONFIG_ISDN_PPP=yCONFIG_ISDN_PPP_VJ=yCONFIG_ISDN_MPP=yCONFIG_IPPP_FILTER=y
# CONFIG_ISDN_PPP_BSDCOMP is not setCONFIG_ISDN_AUDIO=yCONFIG_ISDN_TTY_FAX=y
#
# ISDN feature submodules
#CONFIG_ISDN_DIVERSION=m
#
# ISDN4Linux hardware drivers
#
#
# Passive cards
#CONFIG_ISDN_DRV_HISAX=m
#
# D-channel protocol features
#CONFIG_HISAX_EURO=yCONFIG_DE_AOC=yCONFIG_HISAX_NO_SENDCOMPLETE=yCONFIG_HISAX_NO_LLC=yCONFIG_HISAX_NO_KEYPAD=yCONFIG_HISAX_1TR6=yCONFIG_HISAX_NI1=yCONFIG_HISAX_MAX_CARDS=8
#
# HiSax supported cards
#CONFIG_HISAX_16_3=yCONFIG_HISAX_TELESPCI=yCONFIG_HISAX_S0BOX=yCONFIG_HISAX_FRITZPCI=yCONFIG_HISAX_AVM_A1_PCMCIA=yCONFIG_HISAX_ELSA=yCONFIG_HISAX_DIEHLDIVA=yCONFIG_HISAX_SEDLBAUER=yCONFIG_HISAX_NETJET=yCONFIG_HISAX_NETJET_U=yCONFIG_HISAX_NICCY=yCONFIG_HISAX_BKM_A4T=yCONFIG_HISAX_SCT_QUADRO=yCONFIG_HISAX_GAZEL=yCONFIG_HISAX_HFC_PCI=yCONFIG_HISAX_W6692=yCONFIG_HISAX_HFC_SX=yCONFIG_HISAX_ENTERNOW_PCI=y
# CONFIG_HISAX_DEBUG is not set
#
# HiSax PCMCIA card service modules
#
#
# HiSax sub driver modules
#CONFIG_HISAX_ST5481=m
# CONFIG_HISAX_HFCUSB is not setCONFIG_HISAX_HFC4S8S=mCONFIG_HISAX_FRITZ_PCIPNP=mCONFIG_ISDN_CAPI=m
# CONFIG_CAPI_TRACE is not setCONFIG_ISDN_CAPI_CAPI20=mCONFIG_ISDN_CAPI_MIDDLEWARE=yCONFIG_ISDN_CAPI_CAPIDRV=m
# CONFIG_ISDN_CAPI_CAPIDRV_VERBOSE is not set
#
# CAPI hardware drivers
#CONFIG_CAPI_AVM=yCONFIG_ISDN_DRV_AVMB1_B1PCI=mCONFIG_ISDN_DRV_AVMB1_B1PCIV4=yCONFIG_ISDN_DRV_AVMB1_T1PCI=mCONFIG_ISDN_DRV_AVMB1_C4=mCONFIG_ISDN_DRV_GIGASET=mCONFIG_GIGASET_CAPI=yCONFIG_GIGASET_BASE=mCONFIG_GIGASET_M105=mCONFIG_GIGASET_M101=m
# CONFIG_GIGASET_DEBUG is not setCONFIG_HYSDN=mCONFIG_HYSDN_CAPI=yCONFIG_MISDN=mCONFIG_MISDN_DSP=mCONFIG_MISDN_L1OIP=m
#
# mISDN hardware drivers
#CONFIG_MISDN_HFCPCI=mCONFIG_MISDN_HFCMULTI=mCONFIG_MISDN_HFCUSB=mCONFIG_MISDN_AVMFRITZ=mCONFIG_MISDN_SPEEDFAX=mCONFIG_MISDN_INFINEON=mCONFIG_MISDN_W6692=mCONFIG_MISDN_NETJET=mCONFIG_MISDN_IPAC=mCONFIG_MISDN_ISAR=mCONFIG_ISDN_HDLC=m
# CONFIG_NVM is not set
#
# Input device support
#CONFIG_INPUT=yCONFIG_INPUT_LEDS=yCONFIG_INPUT_FF_MEMLESS=yCONFIG_INPUT_POLLDEV=mCONFIG_INPUT_SPARSEKMAP=m
# CONFIG_INPUT_MATRIXKMAP is not set
#
# Userland interfaces
#CONFIG_INPUT_MOUSEDEV=y
# CONFIG_INPUT_MOUSEDEV_PSAUX is not setCONFIG_INPUT_MOUSEDEV_SCREEN_X=1024CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768CONFIG_INPUT_JOYDEV=mCONFIG_INPUT_EVDEV=y
# CONFIG_INPUT_EVBUG is not set
#
# Input Device Drivers
#CONFIG_INPUT_KEYBOARD=y
# CONFIG_KEYBOARD_ADC is not set
# CONFIG_KEYBOARD_ADP5588 is not set
# CONFIG_KEYBOARD_ADP5589 is not setCONFIG_KEYBOARD_ATKBD=y
# CONFIG_KEYBOARD_QT1070 is not set
# CONFIG_KEYBOARD_QT2160 is not set
# CONFIG_KEYBOARD_DLINK_DIR685 is not set
# CONFIG_KEYBOARD_LKKBD is not set
# CONFIG_KEYBOARD_GPIO is not set
# CONFIG_KEYBOARD_GPIO_POLLED is not set
# CONFIG_KEYBOARD_TCA6416 is not set
# CONFIG_KEYBOARD_TCA8418 is not set
# CONFIG_KEYBOARD_MATRIX is not set
# CONFIG_KEYBOARD_LM8323 is not set
# CONFIG_KEYBOARD_LM8333 is not set
# CONFIG_KEYBOARD_MAX7359 is not set
# CONFIG_KEYBOARD_MCS is not set
# CONFIG_KEYBOARD_MPR121 is not set
# CONFIG_KEYBOARD_NEWTON is not set
# CONFIG_KEYBOARD_OPENCORES is not set
# CONFIG_KEYBOARD_SAMSUNG is not set
# CONFIG_KEYBOARD_STOWAWAY is not set
# CONFIG_KEYBOARD_SUNKBD is not set
# CONFIG_KEYBOARD_TM2_TOUCHKEY is not set
# CONFIG_KEYBOARD_XTKBD is not setCONFIG_INPUT_MOUSE=yCONFIG_MOUSE_PS2=yCONFIG_MOUSE_PS2_ALPS=yCONFIG_MOUSE_PS2_BYD=yCONFIG_MOUSE_PS2_LOGIPS2PP=yCONFIG_MOUSE_PS2_SYNAPTICS=yCONFIG_MOUSE_PS2_SYNAPTICS_SMBUS=yCONFIG_MOUSE_PS2_CYPRESS=yCONFIG_MOUSE_PS2_LIFEBOOK=yCONFIG_MOUSE_PS2_TRACKPOINT=yCONFIG_MOUSE_PS2_ELANTECH=yCONFIG_MOUSE_PS2_ELANTECH_SMBUS=yCONFIG_MOUSE_PS2_SENTELIC=y
# CONFIG_MOUSE_PS2_TOUCHKIT is not setCONFIG_MOUSE_PS2_FOCALTECH=yCONFIG_MOUSE_PS2_VMMOUSE=yCONFIG_MOUSE_PS2_SMBUS=yCONFIG_MOUSE_SERIAL=mCONFIG_MOUSE_APPLETOUCH=mCONFIG_MOUSE_BCM5974=mCONFIG_MOUSE_CYAPA=m
# CONFIG_MOUSE_ELAN_I2C is not setCONFIG_MOUSE_VSXXXAA=m
# CONFIG_MOUSE_GPIO is not setCONFIG_MOUSE_SYNAPTICS_I2C=mCONFIG_MOUSE_SYNAPTICS_USB=m
# CONFIG_INPUT_JOYSTICK is not setCONFIG_INPUT_TABLET=yCONFIG_TABLET_USB_ACECAD=mCONFIG_TABLET_USB_AIPTEK=mCONFIG_TABLET_USB_GTCO=m
# CONFIG_TABLET_USB_HANWANG is not setCONFIG_TABLET_USB_KBTAB=m
# CONFIG_TABLET_USB_PEGASUS is not set
# CONFIG_TABLET_SERIAL_WACOM4 is not setCONFIG_INPUT_TOUCHSCREEN=yCONFIG_TOUCHSCREEN_PROPERTIES=y
# CONFIG_TOUCHSCREEN_ADS7846 is not set
# CONFIG_TOUCHSCREEN_AD7877 is not set
# CONFIG_TOUCHSCREEN_AD7879 is not set
# CONFIG_TOUCHSCREEN_ADC is not set
# CONFIG_TOUCHSCREEN_ATMEL_MXT is not set
# CONFIG_TOUCHSCREEN_AUO_PIXCIR is not set
# CONFIG_TOUCHSCREEN_BU21013 is not set
# CONFIG_TOUCHSCREEN_BU21029 is not set
# CONFIG_TOUCHSCREEN_CHIPONE_ICN8505 is not set
# CONFIG_TOUCHSCREEN_CY8CTMG110 is not set
# CONFIG_TOUCHSCREEN_CYTTSP_CORE is not set
# CONFIG_TOUCHSCREEN_CYTTSP4_CORE is not set
# CONFIG_TOUCHSCREEN_DYNAPRO is not set
# CONFIG_TOUCHSCREEN_HAMPSHIRE is not set
# CONFIG_TOUCHSCREEN_EETI is not set
# CONFIG_TOUCHSCREEN_EGALAX_SERIAL is not set
# CONFIG_TOUCHSCREEN_EXC3000 is not set
# CONFIG_TOUCHSCREEN_FUJITSU is not set
# CONFIG_TOUCHSCREEN_GOODIX is not set
# CONFIG_TOUCHSCREEN_HIDEEP is not set
# CONFIG_TOUCHSCREEN_ILI210X is not set
# CONFIG_TOUCHSCREEN_S6SY761 is not set
# CONFIG_TOUCHSCREEN_GUNZE is not set
# CONFIG_TOUCHSCREEN_EKTF2127 is not set
# CONFIG_TOUCHSCREEN_ELAN is not setCONFIG_TOUCHSCREEN_ELO=mCONFIG_TOUCHSCREEN_WACOM_W8001=mCONFIG_TOUCHSCREEN_WACOM_I2C=m
# CONFIG_TOUCHSCREEN_MAX11801 is not set
# CONFIG_TOUCHSCREEN_MCS5000 is not set
# CONFIG_TOUCHSCREEN_MMS114 is not set
# CONFIG_TOUCHSCREEN_MELFAS_MIP4 is not set
# CONFIG_TOUCHSCREEN_MTOUCH is not set
# CONFIG_TOUCHSCREEN_INEXIO is not set
# CONFIG_TOUCHSCREEN_MK712 is not set
# CONFIG_TOUCHSCREEN_PENMOUNT is not set
# CONFIG_TOUCHSCREEN_EDT_FT5X06 is not set
# CONFIG_TOUCHSCREEN_TOUCHRIGHT is not set
# CONFIG_TOUCHSCREEN_TOUCHWIN is not set
# CONFIG_TOUCHSCREEN_PIXCIR is not set
# CONFIG_TOUCHSCREEN_WDT87XX_I2C is not set
# CONFIG_TOUCHSCREEN_WM97XX is not set
# CONFIG_TOUCHSCREEN_USB_COMPOSITE is not set
# CONFIG_TOUCHSCREEN_TOUCHIT213 is not set
# CONFIG_TOUCHSCREEN_TSC_SERIO is not set
# CONFIG_TOUCHSCREEN_TSC2004 is not set
# CONFIG_TOUCHSCREEN_TSC2005 is not set
# CONFIG_TOUCHSCREEN_TSC2007 is not set
# CONFIG_TOUCHSCREEN_RM_TS is not set
# CONFIG_TOUCHSCREEN_SILEAD is not set
# CONFIG_TOUCHSCREEN_SIS_I2C is not set
# CONFIG_TOUCHSCREEN_ST1232 is not set
# CONFIG_TOUCHSCREEN_STMFTS is not set
# CONFIG_TOUCHSCREEN_SUR40 is not set
# CONFIG_TOUCHSCREEN_SURFACE3_SPI is not set
# CONFIG_TOUCHSCREEN_SX8654 is not set
# CONFIG_TOUCHSCREEN_TPS6507X is not set
# CONFIG_TOUCHSCREEN_ZET6223 is not set
# CONFIG_TOUCHSCREEN_ZFORCE is not set
# CONFIG_TOUCHSCREEN_ROHM_BU21023 is not setCONFIG_INPUT_MISC=y
# CONFIG_INPUT_AD714X is not set
# CONFIG_INPUT_BMA150 is not set
# CONFIG_INPUT_E3X0_BUTTON is not set
# CONFIG_INPUT_MSM_VIBRATOR is not setCONFIG_INPUT_PCSPKR=m
# CONFIG_INPUT_MMA8450 is not setCONFIG_INPUT_APANEL=mCONFIG_INPUT_GP2A=m
# CONFIG_INPUT_GPIO_BEEPER is not set
# CONFIG_INPUT_GPIO_DECODER is not setCONFIG_INPUT_ATLAS_BTNS=mCONFIG_INPUT_ATI_REMOTE2=mCONFIG_INPUT_KEYSPAN_REMOTE=m
# CONFIG_INPUT_KXTJ9 is not setCONFIG_INPUT_POWERMATE=mCONFIG_INPUT_YEALINK=mCONFIG_INPUT_CM109=mCONFIG_INPUT_UINPUT=m
# CONFIG_INPUT_PCF8574 is not set
# CONFIG_INPUT_PWM_BEEPER is not set
# CONFIG_INPUT_PWM_VIBRA is not setCONFIG_INPUT_GPIO_ROTARY_ENCODER=m
# CONFIG_INPUT_ADXL34X is not set
# CONFIG_INPUT_IMS_PCU is not set
# CONFIG_INPUT_CMA3000 is not setCONFIG_INPUT_XEN_KBDDEV_FRONTEND=m
# CONFIG_INPUT_IDEAPAD_SLIDEBAR is not set
# CONFIG_INPUT_DRV260X_HAPTICS is not set
# CONFIG_INPUT_DRV2665_HAPTICS is not set
# CONFIG_INPUT_DRV2667_HAPTICS is not setCONFIG_RMI4_CORE=m
# CONFIG_RMI4_I2C is not set
# CONFIG_RMI4_SPI is not setCONFIG_RMI4_SMB=mCONFIG_RMI4_F03=yCONFIG_RMI4_F03_SERIO=mCONFIG_RMI4_2D_SENSOR=yCONFIG_RMI4_F11=yCONFIG_RMI4_F12=yCONFIG_RMI4_F30=y
# CONFIG_RMI4_F34 is not set
# CONFIG_RMI4_F54 is not set
# CONFIG_RMI4_F55 is not set
#
# Hardware I/O ports
#CONFIG_SERIO=yCONFIG_ARCH_MIGHT_HAVE_PC_SERIO=yCONFIG_SERIO_I8042=yCONFIG_SERIO_SERPORT=y
# CONFIG_SERIO_CT82C710 is not set
# CONFIG_SERIO_PARKBD is not set
# CONFIG_SERIO_PCIPS2 is not setCONFIG_SERIO_LIBPS2=yCONFIG_SERIO_RAW=mCONFIG_SERIO_ALTERA_PS2=m
# CONFIG_SERIO_PS2MULT is not setCONFIG_SERIO_ARC_PS2=m
# CONFIG_SERIO_OLPC_APSP is not setCONFIG_HYPERV_KEYBOARD=m
# CONFIG_SERIO_GPIO_PS2 is not set
# CONFIG_USERIO is not set
# CONFIG_GAMEPORT is not set
#
# Character devices
#CONFIG_TTY=yCONFIG_VT=yCONFIG_CONSOLE_TRANSLATIONS=yCONFIG_VT_CONSOLE=yCONFIG_VT_CONSOLE_SLEEP=yCONFIG_HW_CONSOLE=yCONFIG_VT_HW_CONSOLE_BINDING=yCONFIG_UNIX98_PTYS=y
# CONFIG_LEGACY_PTYS is not setCONFIG_SERIAL_NONSTANDARD=y
# CONFIG_ROCKETPORT is not setCONFIG_CYCLADES=m
# CONFIG_CYZ_INTR is not set
# CONFIG_MOXA_INTELLIO is not set
# CONFIG_MOXA_SMARTIO is not setCONFIG_SYNCLINK=mCONFIG_SYNCLINKMP=mCONFIG_SYNCLINK_GT=mCONFIG_NOZOMI=m
# CONFIG_ISI is not setCONFIG_N_HDLC=mCONFIG_N_GSM=m
# CONFIG_TRACE_SINK is not setCONFIG_LDISC_AUTOLOAD=yCONFIG_DEVMEM=y
# CONFIG_DEVKMEM is not set
#
# Serial drivers
#CONFIG_SERIAL_EARLYCON=y CONFIG_SERIAL_8250=y
# CONFIG_SERIAL_8250_DEPRECATED_OPTIONS is not set CONFIG_SERIAL_8250_PNP=y
# CONFIG_SERIAL_8250_FINTEK is not set CONFIG_SERIAL_8250_CONSOLE=y CONFIG_SERIAL_8250_DMA=y CONFIG_SERIAL_8250_PCI=y CONFIG_SERIAL_8250_EXAR=y CONFIG_SERIAL_8250_NR_UARTS=32 CONFIG_SERIAL_8250_RUNTIME_UARTS=4 CONFIG_SERIAL_8250_EXTENDED=y CONFIG_SERIAL_8250_MANY_PORTS=y CONFIG_SERIAL_8250_SHARE_IRQ=y
# CONFIG_SERIAL_8250_DETECT_IRQ is not set CONFIG_SERIAL_8250_RSA=y CONFIG_SERIAL_8250_DW=y
# CONFIG_SERIAL_8250_RT288X is not set CONFIG_SERIAL_8250_LPSS=y CONFIG_SERIAL_8250_MID=y
# CONFIG_SERIAL_8250_MOXA is not set
#
# Non-8250 serial port support
#
# CONFIG_SERIAL_MAX3100 is not set
# CONFIG_SERIAL_MAX310X is not set
# CONFIG_SERIAL_UARTLITE is not setCONFIG_SERIAL_CORE=yCONFIG_SERIAL_CORE_CONSOLE=yCONFIG_SERIAL_JSM=m
# CONFIG_SERIAL_SCCNXP is not set
# CONFIG_SERIAL_SC16IS7XX is not set
# CONFIG_SERIAL_ALTERA_JTAGUART is not set
# CONFIG_SERIAL_ALTERA_UART is not set
# CONFIG_SERIAL_IFX6X60 is not setCONFIG_SERIAL_ARC=mCONFIG_SERIAL_ARC_NR_PORTS=1
# CONFIG_SERIAL_RP2 is not set
# CONFIG_SERIAL_FSL_LPUART is not set
# CONFIG_SERIAL_DEV_BUS is not set
# CONFIG_TTY_PRINTK is not setCONFIG_PRINTER=m
# CONFIG_LP_CONSOLE is not setCONFIG_PPDEV=mCONFIG_HVC_DRIVER=yCONFIG_HVC_IRQ=yCONFIG_HVC_XEN=yCONFIG_HVC_XEN_FRONTEND=yCONFIG_VIRTIO_CONSOLE=yCONFIG_IPMI_HANDLER=mCONFIG_IPMI_DMI_DECODE=yCONFIG_IPMI_PLAT_DATA=y
# CONFIG_IPMI_PANIC_EVENT is not setCONFIG_IPMI_DEVICE_INTERFACE=mCONFIG_IPMI_SI=mCONFIG_IPMI_SSIF=mCONFIG_IPMI_WATCHDOG=mCONFIG_IPMI_POWEROFF=mCONFIG_HW_RANDOM=yCONFIG_HW_RANDOM_TIMERIOMEM=mCONFIG_HW_RANDOM_INTEL=mCONFIG_HW_RANDOM_AMD=mCONFIG_HW_RANDOM_VIA=mCONFIG_HW_RANDOM_VIRTIO=yCONFIG_NVRAM=y
# CONFIG_R3964 is not set
# CONFIG_APPLICOM is not set
# CONFIG_MWAVE is not setCONFIG_RAW_DRIVER=yCONFIG_MAX_RAW_DEVS=8192CONFIG_HPET=yCONFIG_HPET_MMAP=y
# CONFIG_HPET_MMAP_DEFAULT is not setCONFIG_HANGCHECK_TIMER=mCONFIG_UV_MMTIMER=mCONFIG_TCG_TPM=yCONFIG_HW_RANDOM_TPM=yCONFIG_TCG_TIS_CORE=yCONFIG_TCG_TIS=y
# CONFIG_TCG_TIS_SPI is not setCONFIG_TCG_TIS_I2C_ATMEL=mCONFIG_TCG_TIS_I2C_INFINEON=mCONFIG_TCG_TIS_I2C_NUVOTON=mCONFIG_TCG_NSC=mCONFIG_TCG_ATMEL=mCONFIG_TCG_INFINEON=m
# CONFIG_TCG_XEN is not setCONFIG_TCG_CRB=y
# CONFIG_TCG_VTPM_PROXY is not setCONFIG_TCG_TIS_ST33ZP24=mCONFIG_TCG_TIS_ST33ZP24_I2C=m
# CONFIG_TCG_TIS_ST33ZP24_SPI is not setCONFIG_TELCLOCK=mCONFIG_DEVPORT=y
# CONFIG_XILLYBUS is not set
# CONFIG_RANDOM_TRUST_CPU is not set
#
# I2C support
#CONFIG_I2C=yCONFIG_ACPI_I2C_OPREGION=yCONFIG_I2C_BOARDINFO=yCONFIG_I2C_COMPAT=yCONFIG_I2C_CHARDEV=mCONFIG_I2C_MUX=m
#
# Multiplexer I2C Chip support
#
# CONFIG_I2C_MUX_GPIO is not set
# CONFIG_I2C_MUX_LTC4306 is not set
# CONFIG_I2C_MUX_PCA9541 is not set
# CONFIG_I2C_MUX_PCA954x is not set
# CONFIG_I2C_MUX_REG is not set
# CONFIG_I2C_MUX_MLXCPLD is not setCONFIG_I2C_HELPER_AUTO=yCONFIG_I2C_SMBUS=mCONFIG_I2C_ALGOBIT=yCONFIG_I2C_ALGOPCA=m
#
# I2C Hardware Bus support
#
#
# PC SMBus host controller drivers
#
# CONFIG_I2C_ALI1535 is not set
# CONFIG_I2C_ALI1563 is not set
# CONFIG_I2C_ALI15X3 is not setCONFIG_I2C_AMD756=mCONFIG_I2C_AMD756_S4882=mCONFIG_I2C_AMD8111=mCONFIG_I2C_I801=mCONFIG_I2C_ISCH=mCONFIG_I2C_ISMT=mCONFIG_I2C_PIIX4=mCONFIG_I2C_NFORCE2=mCONFIG_I2C_NFORCE2_S4985=m
# CONFIG_I2C_NVIDIA_GPU is not set
# CONFIG_I2C_SIS5595 is not set
# CONFIG_I2C_SIS630 is not setCONFIG_I2C_SIS96X=mCONFIG_I2C_VIA=mCONFIG_I2C_VIAPRO=m
#
# ACPI drivers
#CONFIG_I2C_SCMI=m
#
# I2C system bus drivers (mostly embedded / system-on-chip) #
# CONFIG_I2C_CBUS_GPIO is not set CONFIG_I2C_DESIGNWARE_CORE=m CONFIG_I2C_DESIGNWARE_PLATFORM=m
# CONFIG_I2C_DESIGNWARE_SLAVE is not set
# CONFIG_I2C_DESIGNWARE_PCI is not set
# CONFIG_I2C_DESIGNWARE_BAYTRAIL is not set
# CONFIG_I2C_EMEV2 is not set
# CONFIG_I2C_GPIO is not set
# CONFIG_I2C_OCORES is not set CONFIG_I2C_PCA_PLATFORM=mCONFIG_I2C_SIMTEC=m
# CONFIG_I2C_XILINX is not set
#
# External I2C/SMBus adapter drivers
#CONFIG_I2C_DIOLAN_U2C=mCONFIG_I2C_PARPORT=mCONFIG_I2C_PARPORT_LIGHT=m
# CONFIG_I2C_ROBOTFUZZ_OSIF is not set
# CONFIG_I2C_TAOS_EVM is not setCONFIG_I2C_TINY_USB=mCONFIG_I2C_VIPERBOARD=m
#
# Other I2C/SMBus bus drivers
#
# CONFIG_I2C_MLXCPLD is not setCONFIG_I2C_STUB=m
# CONFIG_I2C_SLAVE is not set
# CONFIG_I2C_DEBUG_CORE is not set
# CONFIG_I2C_DEBUG_ALGO is not set
# CONFIG_I2C_DEBUG_BUS is not set
# CONFIG_I3C is not setCONFIG_SPI=y
# CONFIG_SPI_DEBUG is not setCONFIG_SPI_MASTER=y
# CONFIG_SPI_MEM is not set
#
# SPI Master Controller Drivers
#
# CONFIG_SPI_ALTERA is not set
# CONFIG_SPI_AXI_SPI_ENGINE is not set
# CONFIG_SPI_BITBANG is not set
# CONFIG_SPI_BUTTERFLY is not set
# CONFIG_SPI_CADENCE is not set
# CONFIG_SPI_DESIGNWARE is not set
# CONFIG_SPI_NXP_FLEXSPI is not set
# CONFIG_SPI_GPIO is not set
# CONFIG_SPI_LM70_LLP is not set
# CONFIG_SPI_OC_TINY is not set
# CONFIG_SPI_PXA2XX is not set
# CONFIG_SPI_ROCKCHIP is not set
# CONFIG_SPI_SC18IS602 is not set
# CONFIG_SPI_SIFIVE is not set
# CONFIG_SPI_MXIC is not set
# CONFIG_SPI_XCOMM is not set
# CONFIG_SPI_XILINX is not set
# CONFIG_SPI_ZYNQMP_GQSPI is not set
#
# SPI Protocol Masters
#
# CONFIG_SPI_SPIDEV is not set
# CONFIG_SPI_LOOPBACK_TEST is not set
# CONFIG_SPI_TLE62X0 is not set
# CONFIG_SPI_SLAVE is not set
# CONFIG_SPMI is not set
# CONFIG_HSI is not setCONFIG_PPS=y
# CONFIG_PPS_DEBUG is not set
#
# PPS clients support
#
# CONFIG_PPS_CLIENT_KTIMER is not setCONFIG_PPS_CLIENT_LDISC=mCONFIG_PPS_CLIENT_PARPORT=mCONFIG_PPS_CLIENT_GPIO=m
#
# PPS generators support
#
#
# PTP clock support
#CONFIG_PTP_1588_CLOCK=yCONFIG_DP83640_PHY=mCONFIG_PTP_1588_CLOCK_KVM=mCONFIG_PINCTRL=yCONFIG_PINMUX=yCONFIG_PINCONF=yCONFIG_GENERIC_PINCONF=y
# CONFIG_DEBUG_PINCTRL is not setCONFIG_PINCTRL_AMD=m
# CONFIG_PINCTRL_MCP23S08 is not set
# CONFIG_PINCTRL_SX150X is not setCONFIG_PINCTRL_BAYTRAIL=y
# CONFIG_PINCTRL_CHERRYVIEW is not setCONFIG_PINCTRL_INTEL=m
# CONFIG_PINCTRL_BROXTON is not setCONFIG_PINCTRL_CANNONLAKE=m
# CONFIG_PINCTRL_CEDARFORK is not setCONFIG_PINCTRL_DENVERTON=mCONFIG_PINCTRL_GEMINILAKE=m
# CONFIG_PINCTRL_ICELAKE is not setCONFIG_PINCTRL_LEWISBURG=mCONFIG_PINCTRL_SUNRISEPOINT=mCONFIG_GPIOLIB=yCONFIG_GPIOLIB_FASTPATH_LIMIT=512CONFIG_GPIO_ACPI=yCONFIG_GPIOLIB_IRQCHIP=y
# CONFIG_DEBUG_GPIO is not setCONFIG_GPIO_SYSFS=yCONFIG_GPIO_GENERIC=m
#
# Memory mapped GPIO drivers
#CONFIG_GPIO_AMDPT=m
# CONFIG_GPIO_DWAPB is not set
# CONFIG_GPIO_EXAR is not set
# CONFIG_GPIO_GENERIC_PLATFORM is not setCONFIG_GPIO_ICH=m
# CONFIG_GPIO_LYNXPOINT is not set
# CONFIG_GPIO_MB86S7X is not setCONFIG_GPIO_MOCKUP=y
# CONFIG_GPIO_VX855 is not set
# CONFIG_GPIO_AMD_FCH is not set
#
# Port-mapped I/O GPIO drivers
#
# CONFIG_GPIO_F7188X is not set
# CONFIG_GPIO_IT87 is not set
# CONFIG_GPIO_SCH is not set
# CONFIG_GPIO_SCH311X is not set
# CONFIG_GPIO_WINBOND is not set
# CONFIG_GPIO_WS16C48 is not set
#
# I2C GPIO expanders
#
# CONFIG_GPIO_ADP5588 is not set
# CONFIG_GPIO_MAX7300 is not set
# CONFIG_GPIO_MAX732X is not set
# CONFIG_GPIO_PCA953X is not set
# CONFIG_GPIO_PCF857X is not set
# CONFIG_GPIO_TPIC2810 is not set
#
# MFD GPIO expanders
#
#
# PCI GPIO expanders
#
# CONFIG_GPIO_AMD8111 is not set
# CONFIG_GPIO_ML_IOH is not set
# CONFIG_GPIO_PCI_IDIO_16 is not set
# CONFIG_GPIO_PCIE_IDIO_24 is not set
# CONFIG_GPIO_RDC321X is not set
#
# SPI GPIO expanders
#
# CONFIG_GPIO_MAX3191X is not set
# CONFIG_GPIO_MAX7301 is not set
# CONFIG_GPIO_MC33880 is not set
# CONFIG_GPIO_PISOSR is not set
# CONFIG_GPIO_XRA1403 is not set
#
# USB GPIO expanders
#CONFIG_GPIO_VIPERBOARD=m
# CONFIG_W1 is not set
# CONFIG_POWER_AVS is not setCONFIG_POWER_RESET=y
# CONFIG_POWER_RESET_RESTART is not setCONFIG_POWER_SUPPLY=y
# CONFIG_POWER_SUPPLY_DEBUG is not set
# CONFIG_PDA_POWER is not set
# CONFIG_GENERIC_ADC_BATTERY is not set
# CONFIG_TEST_POWER is not set
# CONFIG_CHARGER_ADP5061 is not set
# CONFIG_BATTERY_DS2780 is not set
# CONFIG_BATTERY_DS2781 is not set
# CONFIG_BATTERY_DS2782 is not set
# CONFIG_BATTERY_SBS is not set
# CONFIG_CHARGER_SBS is not set
# CONFIG_MANAGER_SBS is not set
# CONFIG_BATTERY_BQ27XXX is not set
# CONFIG_BATTERY_MAX17040 is not set
# CONFIG_BATTERY_MAX17042 is not set
# CONFIG_CHARGER_MAX8903 is not set
# CONFIG_CHARGER_LP8727 is not set
# CONFIG_CHARGER_GPIO is not set
# CONFIG_CHARGER_LTC3651 is not set
# CONFIG_CHARGER_BQ2415X is not set
# CONFIG_CHARGER_BQ24257 is not set
# CONFIG_CHARGER_BQ24735 is not set
# CONFIG_CHARGER_BQ25890 is not setCONFIG_CHARGER_SMB347=m
# CONFIG_BATTERY_GAUGE_LTC2941 is not set
# CONFIG_CHARGER_RT9455 is not setCONFIG_HWMON=yCONFIG_HWMON_VID=m
# CONFIG_HWMON_DEBUG_CHIP is not set
#
# Native drivers
#CONFIG_SENSORS_ABITUGURU=mCONFIG_SENSORS_ABITUGURU3=m
# CONFIG_SENSORS_AD7314 is not setCONFIG_SENSORS_AD7414=mCONFIG_SENSORS_AD7418=mCONFIG_SENSORS_ADM1021=mCONFIG_SENSORS_ADM1025=mCONFIG_SENSORS_ADM1026=mCONFIG_SENSORS_ADM1029=mCONFIG_SENSORS_ADM1031=mCONFIG_SENSORS_ADM9240=mCONFIG_SENSORS_ADT7X10=m
# CONFIG_SENSORS_ADT7310 is not setCONFIG_SENSORS_ADT7410=mCONFIG_SENSORS_ADT7411=mCONFIG_SENSORS_ADT7462=mCONFIG_SENSORS_ADT7470=mCONFIG_SENSORS_ADT7475=mCONFIG_SENSORS_ASC7621=mCONFIG_SENSORS_K8TEMP=mCONFIG_SENSORS_K10TEMP=mCONFIG_SENSORS_FAM15H_POWER=mCONFIG_SENSORS_APPLESMC=mCONFIG_SENSORS_ASB100=m
# CONFIG_SENSORS_ASPEED is not setCONFIG_SENSORS_ATXP1=mCONFIG_SENSORS_DS620=mCONFIG_SENSORS_DS1621=mCONFIG_SENSORS_DELL_SMM=mCONFIG_SENSORS_I5K_AMB=mCONFIG_SENSORS_F71805F=mCONFIG_SENSORS_F71882FG=mCONFIG_SENSORS_F75375S=mCONFIG_SENSORS_FSCHMD=m
# CONFIG_SENSORS_FTSTEUTATES is not setCONFIG_SENSORS_GL518SM=mCONFIG_SENSORS_GL520SM=mCONFIG_SENSORS_G760A=m
# CONFIG_SENSORS_G762 is not set
# CONFIG_SENSORS_HIH6130 is not setCONFIG_SENSORS_IBMAEM=mCONFIG_SENSORS_IBMPEX=m
# CONFIG_SENSORS_IIO_HWMON is not set
# CONFIG_SENSORS_I5500 is not setCONFIG_SENSORS_CORETEMP=mCONFIG_SENSORS_IT87=mCONFIG_SENSORS_JC42=m
# CONFIG_SENSORS_POWR1220 is not setCONFIG_SENSORS_LINEAGE=m
# CONFIG_SENSORS_LTC2945 is not set
# CONFIG_SENSORS_LTC2990 is not setCONFIG_SENSORS_LTC4151=mCONFIG_SENSORS_LTC4215=m
# CONFIG_SENSORS_LTC4222 is not setCONFIG_SENSORS_LTC4245=m
# CONFIG_SENSORS_LTC4260 is not setCONFIG_SENSORS_LTC4261=m
# CONFIG_SENSORS_MAX1111 is not setCONFIG_SENSORS_MAX16065=mCONFIG_SENSORS_MAX1619=mCONFIG_SENSORS_MAX1668=mCONFIG_SENSORS_MAX197=m
# CONFIG_SENSORS_MAX31722 is not set
# CONFIG_SENSORS_MAX6621 is not setCONFIG_SENSORS_MAX6639=mCONFIG_SENSORS_MAX6642=mCONFIG_SENSORS_MAX6650=mCONFIG_SENSORS_MAX6697=m
# CONFIG_SENSORS_MAX31790 is not setCONFIG_SENSORS_MCP3021=m
# CONFIG_SENSORS_TC654 is not set
# CONFIG_SENSORS_ADCXX is not setCONFIG_SENSORS_LM63=m
# CONFIG_SENSORS_LM70 is not setCONFIG_SENSORS_LM73=mCONFIG_SENSORS_LM75=mCONFIG_SENSORS_LM77=mCONFIG_SENSORS_LM78=mCONFIG_SENSORS_LM80=mCONFIG_SENSORS_LM83=mCONFIG_SENSORS_LM85=mCONFIG_SENSORS_LM87=mCONFIG_SENSORS_LM90=mCONFIG_SENSORS_LM92=mCONFIG_SENSORS_LM93=mCONFIG_SENSORS_LM95234=mCONFIG_SENSORS_LM95241=mCONFIG_SENSORS_LM95245=mCONFIG_SENSORS_PC87360=mCONFIG_SENSORS_PC87427=mCONFIG_SENSORS_NTC_THERMISTOR=m
# CONFIG_SENSORS_NCT6683 is not setCONFIG_SENSORS_NCT6775=m
# CONFIG_SENSORS_NCT7802 is not set
# CONFIG_SENSORS_NCT7904 is not set
# CONFIG_SENSORS_NPCM7XX is not set
# CONFIG_SENSORS_OCC_P8_I2C is not setCONFIG_SENSORS_PCF8591=mCONFIG_PMBUS=mCONFIG_SENSORS_PMBUS=mCONFIG_SENSORS_ADM1275=m
# CONFIG_SENSORS_IBM_CFFPS is not set
# CONFIG_SENSORS_IR35221 is not setCONFIG_SENSORS_LM25066=mCONFIG_SENSORS_LTC2978=m
# CONFIG_SENSORS_LTC3815 is not setCONFIG_SENSORS_MAX16064=m
# CONFIG_SENSORS_MAX20751 is not set
# CONFIG_SENSORS_MAX31785 is not setCONFIG_SENSORS_MAX34440=mCONFIG_SENSORS_MAX8688=m
# CONFIG_SENSORS_TPS40422 is not set
# CONFIG_SENSORS_TPS53679 is not setCONFIG_SENSORS_UCD9000=mCONFIG_SENSORS_UCD9200=mCONFIG_SENSORS_ZL6100=mCONFIG_SENSORS_SHT15=mCONFIG_SENSORS_SHT21=m
# CONFIG_SENSORS_SHT3x is not set
# CONFIG_SENSORS_SHTC1 is not setCONFIG_SENSORS_SIS5595=mCONFIG_SENSORS_DME1737=mCONFIG_SENSORS_EMC1403=m
# CONFIG_SENSORS_EMC2103 is not setCONFIG_SENSORS_EMC6W201=mCONFIG_SENSORS_SMSC47M1=mCONFIG_SENSORS_SMSC47M192=mCONFIG_SENSORS_SMSC47B397=mCONFIG_SENSORS_SCH56XX_COMMON=mCONFIG_SENSORS_SCH5627=mCONFIG_SENSORS_SCH5636=m
# CONFIG_SENSORS_STTS751 is not set
# CONFIG_SENSORS_SMM665 is not set
# CONFIG_SENSORS_ADC128D818 is not setCONFIG_SENSORS_ADS1015=mCONFIG_SENSORS_ADS7828=m
# CONFIG_SENSORS_ADS7871 is not setCONFIG_SENSORS_AMC6821=mCONFIG_SENSORS_INA209=mCONFIG_SENSORS_INA2XX=m
# CONFIG_SENSORS_INA3221 is not set
# CONFIG_SENSORS_TC74 is not setCONFIG_SENSORS_THMC50=mCONFIG_SENSORS_TMP102=m
# CONFIG_SENSORS_TMP103 is not set
# CONFIG_SENSORS_TMP108 is not setCONFIG_SENSORS_TMP401=mCONFIG_SENSORS_TMP421=mCONFIG_SENSORS_VIA_CPUTEMP=mCONFIG_SENSORS_VIA686A=mCONFIG_SENSORS_VT1211=mCONFIG_SENSORS_VT8231=m
# CONFIG_SENSORS_W83773G is not setCONFIG_SENSORS_W83781D=mCONFIG_SENSORS_W83791D=mCONFIG_SENSORS_W83792D=mCONFIG_SENSORS_W83793=mCONFIG_SENSORS_W83795=m
# CONFIG_SENSORS_W83795_FANCTRL is not setCONFIG_SENSORS_W83L785TS=mCONFIG_SENSORS_W83L786NG=mCONFIG_SENSORS_W83627HF=mCONFIG_SENSORS_W83627EHF=m
# CONFIG_SENSORS_XGENE is not set
#
# ACPI drivers
#CONFIG_SENSORS_ACPI_POWER=m CONFIG_SENSORS_ATK0110=mCONFIG_THERMAL=y
# CONFIG_THERMAL_STATISTICS is not set CONFIG_THERMAL_EMERGENCY_POWEROFF_DELAY_MS=0 CONFIG_THERMAL_HWMON=y CONFIG_THERMAL_WRITABLE_TRIPS=y CONFIG_THERMAL_DEFAULT_GOV_STEP_WISE=y
# CONFIG_THERMAL_DEFAULT_GOV_FAIR_SHARE is not set # CONFIG_THERMAL_DEFAULT_GOV_USER_SPACE is not set # CONFIG_THERMAL_DEFAULT_GOV_POWER_ALLOCATOR is not set CONFIG_THERMAL_GOV_FAIR_SHARE=y CONFIG_THERMAL_GOV_STEP_WISE=y CONFIG_THERMAL_GOV_BANG_BANG=y CONFIG_THERMAL_GOV_USER_SPACE=y
# CONFIG_THERMAL_GOV_POWER_ALLOCATOR is not set # CONFIG_CLOCK_THERMAL is not set
# CONFIG_DEVFREQ_THERMAL is not set
# CONFIG_THERMAL_EMULATION is not set
#
# Intel thermal drivers
#CONFIG_INTEL_POWERCLAMP=mCONFIG_X86_PKG_TEMP_THERMAL=mCONFIG_INTEL_SOC_DTS_IOSF_CORE=m
# CONFIG_INTEL_SOC_DTS_THERMAL is not set
#
# ACPI INT340X thermal drivers
#CONFIG_INT340X_THERMAL=mCONFIG_ACPI_THERMAL_REL=m
# CONFIG_INT3406_THERMAL is not set
# CONFIG_INTEL_PCH_THERMAL is not set
# CONFIG_GENERIC_ADC_THERMAL is not setCONFIG_WATCHDOG=yCONFIG_WATCHDOG_CORE=y
# CONFIG_WATCHDOG_NOWAYOUT is not setCONFIG_WATCHDOG_HANDLE_BOOT_ENABLED=yCONFIG_WATCHDOG_SYSFS=y
#
# Watchdog Device Drivers
#CONFIG_SOFT_WATCHDOG=mCONFIG_WDAT_WDT=m
# CONFIG_XILINX_WATCHDOG is not set
# CONFIG_ZIIRAVE_WATCHDOG is not set
# CONFIG_CADENCE_WATCHDOG is not set
# CONFIG_DW_WATCHDOG is not set
# CONFIG_MAX63XX_WATCHDOG is not set
# CONFIG_ACQUIRE_WDT is not set
# CONFIG_ADVANTECH_WDT is not setCONFIG_ALIM1535_WDT=mCONFIG_ALIM7101_WDT=m
# CONFIG_EBC_C384_WDT is not setCONFIG_F71808E_WDT=mCONFIG_SP5100_TCO=mCONFIG_SBC_FITPC2_WATCHDOG=m
# CONFIG_EUROTECH_WDT is not setCONFIG_IB700_WDT=mCONFIG_IBMASR=m
# CONFIG_WAFER_WDT is not setCONFIG_I6300ESB_WDT=yCONFIG_IE6XX_WDT=mCONFIG_ITCO_WDT=yCONFIG_ITCO_VENDOR_SUPPORT=yCONFIG_IT8712F_WDT=mCONFIG_IT87_WDT=mCONFIG_HP_WATCHDOG=mCONFIG_HPWDT_NMI_DECODING=y
# CONFIG_SC1200_WDT is not set
# CONFIG_PC87413_WDT is not setCONFIG_NV_TCO=m
# CONFIG_60XX_WDT is not set
# CONFIG_CPU5_WDT is not setCONFIG_SMSC_SCH311X_WDT=m
# CONFIG_SMSC37B787_WDT is not set
# CONFIG_TQMX86_WDT is not setCONFIG_VIA_WDT=mCONFIG_W83627HF_WDT=mCONFIG_W83877F_WDT=mCONFIG_W83977F_WDT=mCONFIG_MACHZ_WDT=m
# CONFIG_SBC_EPX_C3_WATCHDOG is not setCONFIG_INTEL_MEI_WDT=m
# CONFIG_NI903X_WDT is not set
# CONFIG_NIC7018_WDT is not set
# CONFIG_MEN_A21_WDT is not setCONFIG_XEN_WDT=m
#
# PCI-based Watchdog Cards
#CONFIG_PCIPCWATCHDOG=mCONFIG_WDTPCI=m
#
# USB-based Watchdog Cards
#CONFIG_USBPCWATCHDOG=m
#
# Watchdog Pretimeout Governors
#
# CONFIG_WATCHDOG_PRETIMEOUT_GOV is not setCONFIG_SSB_POSSIBLE=yCONFIG_SSB=mCONFIG_SSB_SPROM=yCONFIG_SSB_PCIHOST_POSSIBLE=yCONFIG_SSB_PCIHOST=yCONFIG_SSB_SDIOHOST_POSSIBLE=yCONFIG_SSB_SDIOHOST=yCONFIG_SSB_DRIVER_PCICORE_POSSIBLE=yCONFIG_SSB_DRIVER_PCICORE=yCONFIG_SSB_DRIVER_GPIO=yCONFIG_BCMA_POSSIBLE=yCONFIG_BCMA=mCONFIG_BCMA_HOST_PCI_POSSIBLE=yCONFIG_BCMA_HOST_PCI=y
# CONFIG_BCMA_HOST_SOC is not setCONFIG_BCMA_DRIVER_PCI=yCONFIG_BCMA_DRIVER_GMAC_CMN=yCONFIG_BCMA_DRIVER_GPIO=y
# CONFIG_BCMA_DEBUG is not set
#
# Multifunction device drivers
#CONFIG_MFD_CORE=y
# CONFIG_MFD_AS3711 is not set
# CONFIG_PMIC_ADP5520 is not set
# CONFIG_MFD_AAT2870_CORE is not set
# CONFIG_MFD_BCM590XX is not set
# CONFIG_MFD_BD9571MWV is not set
# CONFIG_MFD_AXP20X_I2C is not set
# CONFIG_MFD_CROS_EC is not set
# CONFIG_MFD_MADERA is not set
# CONFIG_PMIC_DA903X is not set
# CONFIG_MFD_DA9052_SPI is not set
# CONFIG_MFD_DA9052_I2C is not set
# CONFIG_MFD_DA9055 is not set
# CONFIG_MFD_DA9062 is not set
# CONFIG_MFD_DA9063 is not set
# CONFIG_MFD_DA9150 is not set
# CONFIG_MFD_DLN2 is not set
# CONFIG_MFD_MC13XXX_SPI is not set
# CONFIG_MFD_MC13XXX_I2C is not set
# CONFIG_HTC_PASIC3 is not set
# CONFIG_HTC_I2CPLD is not set
# CONFIG_MFD_INTEL_QUARK_I2C_GPIO is not setCONFIG_LPC_ICH=mCONFIG_LPC_SCH=m
# CONFIG_INTEL_SOC_PMIC_CHTDC_TI is not setCONFIG_MFD_INTEL_LPSS=yCONFIG_MFD_INTEL_LPSS_ACPI=yCONFIG_MFD_INTEL_LPSS_PCI=y
# CONFIG_MFD_JANZ_CMODIO is not set
# CONFIG_MFD_KEMPLD is not set
# CONFIG_MFD_88PM800 is not set
# CONFIG_MFD_88PM805 is not set
# CONFIG_MFD_88PM860X is not set
# CONFIG_MFD_MAX14577 is not set
# CONFIG_MFD_MAX77693 is not set
# CONFIG_MFD_MAX77843 is not set
# CONFIG_MFD_MAX8907 is not set
# CONFIG_MFD_MAX8925 is not set
# CONFIG_MFD_MAX8997 is not set
# CONFIG_MFD_MAX8998 is not set
# CONFIG_MFD_MT6397 is not set
# CONFIG_MFD_MENF21BMC is not set
# CONFIG_EZX_PCAP is not setCONFIG_MFD_VIPERBOARD=m
# CONFIG_MFD_RETU is not set
# CONFIG_MFD_PCF50633 is not set
# CONFIG_UCB1400_CORE is not set
# CONFIG_MFD_RDC321X is not set
# CONFIG_MFD_RT5033 is not set
# CONFIG_MFD_RC5T583 is not set
# CONFIG_MFD_SEC_CORE is not set
# CONFIG_MFD_SI476X_CORE is not setCONFIG_MFD_SM501=mCONFIG_MFD_SM501_GPIO=y
# CONFIG_MFD_SKY81452 is not set
# CONFIG_MFD_SMSC is not set
# CONFIG_ABX500_CORE is not set
# CONFIG_MFD_SYSCON is not set
# CONFIG_MFD_TI_AM335X_TSCADC is not set
# CONFIG_MFD_LP3943 is not set
# CONFIG_MFD_LP8788 is not set
# CONFIG_MFD_TI_LMU is not set
# CONFIG_MFD_PALMAS is not set
# CONFIG_TPS6105X is not set
# CONFIG_TPS65010 is not set
# CONFIG_TPS6507X is not set
# CONFIG_MFD_TPS65086 is not set
# CONFIG_MFD_TPS65090 is not set
# CONFIG_MFD_TI_LP873X is not set
# CONFIG_MFD_TPS6586X is not set
# CONFIG_MFD_TPS65910 is not set
# CONFIG_MFD_TPS65912_I2C is not set
# CONFIG_MFD_TPS65912_SPI is not set
# CONFIG_MFD_TPS80031 is not set
# CONFIG_TWL4030_CORE is not set
# CONFIG_TWL6040_CORE is not set
# CONFIG_MFD_WL1273_CORE is not set
# CONFIG_MFD_LM3533 is not set
# CONFIG_MFD_TQMX86 is not setCONFIG_MFD_VX855=m
# CONFIG_MFD_ARIZONA_I2C is not set
# CONFIG_MFD_ARIZONA_SPI is not set
# CONFIG_MFD_WM8400 is not set
# CONFIG_MFD_WM831X_I2C is not set
# CONFIG_MFD_WM831X_SPI is not set
# CONFIG_MFD_WM8350_I2C is not set
# CONFIG_MFD_WM8994 is not set
# CONFIG_REGULATOR is not setCONFIG_RC_CORE=mCONFIG_RC_MAP=mCONFIG_LIRC=yCONFIG_RC_DECODERS=yCONFIG_IR_NEC_DECODER=mCONFIG_IR_RC5_DECODER=mCONFIG_IR_RC6_DECODER=mCONFIG_IR_JVC_DECODER=mCONFIG_IR_SONY_DECODER=mCONFIG_IR_SANYO_DECODER=mCONFIG_IR_SHARP_DECODER=mCONFIG_IR_MCE_KBD_DECODER=m
# CONFIG_IR_XMP_DECODER is not set
# CONFIG_IR_IMON_DECODER is not set
# CONFIG_IR_RCMM_DECODER is not setCONFIG_RC_DEVICES=yCONFIG_RC_ATI_REMOTE=mCONFIG_IR_ENE=mCONFIG_IR_IMON=m
# CONFIG_IR_IMON_RAW is not setCONFIG_IR_MCEUSB=mCONFIG_IR_ITE_CIR=mCONFIG_IR_FINTEK=mCONFIG_IR_NUVOTON=mCONFIG_IR_REDRAT3=mCONFIG_IR_STREAMZAP=mCONFIG_IR_WINBOND_CIR=m
# CONFIG_IR_IGORPLUGUSB is not setCONFIG_IR_IGUANA=mCONFIG_IR_TTUSBIR=mCONFIG_RC_LOOPBACK=m
# CONFIG_IR_SERIAL is not set
# CONFIG_IR_SIR is not set
# CONFIG_RC_XBOX_DVD is not setCONFIG_MEDIA_SUPPORT=m
#
# Multimedia core support
#CONFIG_MEDIA_CAMERA_SUPPORT=yCONFIG_MEDIA_ANALOG_TV_SUPPORT=yCONFIG_MEDIA_DIGITAL_TV_SUPPORT=yCONFIG_MEDIA_RADIO_SUPPORT=y
# CONFIG_MEDIA_SDR_SUPPORT is not set
# CONFIG_MEDIA_CEC_SUPPORT is not set
# CONFIG_MEDIA_CONTROLLER is not setCONFIG_VIDEO_DEV=mCONFIG_VIDEO_V4L2=m
# CONFIG_VIDEO_ADV_DEBUG is not set
# CONFIG_VIDEO_FIXED_MINOR_RANGES is not setCONFIG_VIDEO_TUNER=mCONFIG_VIDEOBUF_GEN=mCONFIG_VIDEOBUF_DMA_SG=mCONFIG_VIDEOBUF_VMALLOC=mCONFIG_DVB_CORE=m
# CONFIG_DVB_MMAP is not setCONFIG_DVB_NET=yCONFIG_TTPCI_EEPROM=mCONFIG_DVB_MAX_ADAPTERS=8CONFIG_DVB_DYNAMIC_MINORS=y
# CONFIG_DVB_DEMUX_SECTION_LOSS_LOG is not set
# CONFIG_DVB_ULE_DEBUG is not set
#
# Media drivers
#CONFIG_MEDIA_USB_SUPPORT=y
#
# Webcam devices
#CONFIG_USB_VIDEO_CLASS=mCONFIG_USB_VIDEO_CLASS_INPUT_EVDEV=yCONFIG_USB_GSPCA=mCONFIG_USB_M5602=mCONFIG_USB_STV06XX=mCONFIG_USB_GL860=mCONFIG_USB_GSPCA_BENQ=mCONFIG_USB_GSPCA_CONEX=mCONFIG_USB_GSPCA_CPIA1=m
# CONFIG_USB_GSPCA_DTCS033 is not setCONFIG_USB_GSPCA_ETOMS=mCONFIG_USB_GSPCA_FINEPIX=mCONFIG_USB_GSPCA_JEILINJ=mCONFIG_USB_GSPCA_JL2005BCD=m
# CONFIG_USB_GSPCA_KINECT is not setCONFIG_USB_GSPCA_KONICA=mCONFIG_USB_GSPCA_MARS=mCONFIG_USB_GSPCA_MR97310A=mCONFIG_USB_GSPCA_NW80X=mCONFIG_USB_GSPCA_OV519=mCONFIG_USB_GSPCA_OV534=mCONFIG_USB_GSPCA_OV534_9=mCONFIG_USB_GSPCA_PAC207=mCONFIG_USB_GSPCA_PAC7302=mCONFIG_USB_GSPCA_PAC7311=mCONFIG_USB_GSPCA_SE401=mCONFIG_USB_GSPCA_SN9C2028=mCONFIG_USB_GSPCA_SN9C20X=mCONFIG_USB_GSPCA_SONIXB=mCONFIG_USB_GSPCA_SONIXJ=mCONFIG_USB_GSPCA_SPCA500=mCONFIG_USB_GSPCA_SPCA501=mCONFIG_USB_GSPCA_SPCA505=mCONFIG_USB_GSPCA_SPCA506=mCONFIG_USB_GSPCA_SPCA508=mCONFIG_USB_GSPCA_SPCA561=mCONFIG_USB_GSPCA_SPCA1528=mCONFIG_USB_GSPCA_SQ905=mCONFIG_USB_GSPCA_SQ905C=mCONFIG_USB_GSPCA_SQ930X=mCONFIG_USB_GSPCA_STK014=m
# CONFIG_USB_GSPCA_STK1135 is not setCONFIG_USB_GSPCA_STV0680=mCONFIG_USB_GSPCA_SUNPLUS=mCONFIG_USB_GSPCA_T613=mCONFIG_USB_GSPCA_TOPRO=m
# CONFIG_USB_GSPCA_TOUPTEK is not setCONFIG_USB_GSPCA_TV8532=mCONFIG_USB_GSPCA_VC032X=mCONFIG_USB_GSPCA_VICAM=mCONFIG_USB_GSPCA_XIRLINK_CIT=mCONFIG_USB_GSPCA_ZC3XX=mCONFIG_USB_PWC=m
# CONFIG_USB_PWC_DEBUG is not setCONFIG_USB_PWC_INPUT_EVDEV=y
# CONFIG_VIDEO_CPIA2 is not setCONFIG_USB_ZR364XX=mCONFIG_USB_STKWEBCAM=mCONFIG_USB_S2255=m
# CONFIG_VIDEO_USBTV is not set
#
# Analog TV USB devices
#CONFIG_VIDEO_PVRUSB2=mCONFIG_VIDEO_PVRUSB2_SYSFS=yCONFIG_VIDEO_PVRUSB2_DVB=y
# CONFIG_VIDEO_PVRUSB2_DEBUGIFC is not setCONFIG_VIDEO_HDPVR=mCONFIG_VIDEO_USBVISION=m
# CONFIG_VIDEO_STK1160_COMMON is not set
# CONFIG_VIDEO_GO7007 is not set
#
# Analog/digital TV USB devices
#CONFIG_VIDEO_AU0828=mCONFIG_VIDEO_AU0828_V4L2=y
# CONFIG_VIDEO_AU0828_RC is not setCONFIG_VIDEO_CX231XX=mCONFIG_VIDEO_CX231XX_RC=yCONFIG_VIDEO_CX231XX_ALSA=mCONFIG_VIDEO_CX231XX_DVB=mCONFIG_VIDEO_TM6000=mCONFIG_VIDEO_TM6000_ALSA=mCONFIG_VIDEO_TM6000_DVB=m
#
# Digital TV USB devices
#CONFIG_DVB_USB=m
# CONFIG_DVB_USB_DEBUG is not setCONFIG_DVB_USB_DIB3000MC=mCONFIG_DVB_USB_A800=mCONFIG_DVB_USB_DIBUSB_MB=m
# CONFIG_DVB_USB_DIBUSB_MB_FAULTY is not setCONFIG_DVB_USB_DIBUSB_MC=mCONFIG_DVB_USB_DIB0700=mCONFIG_DVB_USB_UMT_010=mCONFIG_DVB_USB_CXUSB=mCONFIG_DVB_USB_M920X=mCONFIG_DVB_USB_DIGITV=mCONFIG_DVB_USB_VP7045=mCONFIG_DVB_USB_VP702X=mCONFIG_DVB_USB_GP8PSK=mCONFIG_DVB_USB_NOVA_T_USB2=mCONFIG_DVB_USB_TTUSB2=mCONFIG_DVB_USB_DTT200U=mCONFIG_DVB_USB_OPERA1=mCONFIG_DVB_USB_AF9005=mCONFIG_DVB_USB_AF9005_REMOTE=mCONFIG_DVB_USB_PCTV452E=mCONFIG_DVB_USB_DW2102=mCONFIG_DVB_USB_CINERGY_T2=mCONFIG_DVB_USB_DTV5100=mCONFIG_DVB_USB_AZ6027=mCONFIG_DVB_USB_TECHNISAT_USB2=mCONFIG_DVB_USB_V2=mCONFIG_DVB_USB_AF9015=mCONFIG_DVB_USB_AF9035=mCONFIG_DVB_USB_ANYSEE=mCONFIG_DVB_USB_AU6610=mCONFIG_DVB_USB_AZ6007=mCONFIG_DVB_USB_CE6230=mCONFIG_DVB_USB_EC168=mCONFIG_DVB_USB_GL861=mCONFIG_DVB_USB_LME2510=mCONFIG_DVB_USB_MXL111SF=mCONFIG_DVB_USB_RTL28XXU=m
# CONFIG_DVB_USB_DVBSKY is not set
# CONFIG_DVB_USB_ZD1301 is not setCONFIG_DVB_TTUSB_BUDGET=mCONFIG_DVB_TTUSB_DEC=mCONFIG_SMS_USB_DRV=mCONFIG_DVB_B2C2_FLEXCOP_USB=m
# CONFIG_DVB_B2C2_FLEXCOP_USB_DEBUG is not set
# CONFIG_DVB_AS102 is not set
#
# Webcam, TV (analog/digital) USB devices
#CONFIG_VIDEO_EM28XX=m
# CONFIG_VIDEO_EM28XX_V4L2 is not setCONFIG_VIDEO_EM28XX_ALSA=mCONFIG_VIDEO_EM28XX_DVB=mCONFIG_VIDEO_EM28XX_RC=mCONFIG_MEDIA_PCI_SUPPORT=y
#
# Media capture support
#
# CONFIG_VIDEO_MEYE is not set
# CONFIG_VIDEO_SOLO6X10 is not set
# CONFIG_VIDEO_TW5864 is not set
# CONFIG_VIDEO_TW68 is not set
# CONFIG_VIDEO_TW686X is not set
#
# Media capture/analog TV support
#CONFIG_VIDEO_IVTV=m
# CONFIG_VIDEO_IVTV_DEPRECATED_IOCTLS is not set
# CONFIG_VIDEO_IVTV_ALSA is not setCONFIG_VIDEO_FB_IVTV=m
# CONFIG_VIDEO_FB_IVTV_FORCE_PAT is not set
# CONFIG_VIDEO_HEXIUM_GEMINI is not set
# CONFIG_VIDEO_HEXIUM_ORION is not set
# CONFIG_VIDEO_MXB is not set
# CONFIG_VIDEO_DT3155 is not set
#
# Media capture/analog/hybrid TV support
#CONFIG_VIDEO_CX18=mCONFIG_VIDEO_CX18_ALSA=mCONFIG_VIDEO_CX23885=mCONFIG_MEDIA_ALTERA_CI=m
# CONFIG_VIDEO_CX25821 is not setCONFIG_VIDEO_CX88=mCONFIG_VIDEO_CX88_ALSA=mCONFIG_VIDEO_CX88_BLACKBIRD=mCONFIG_VIDEO_CX88_DVB=mCONFIG_VIDEO_CX88_ENABLE_VP3054=yCONFIG_VIDEO_CX88_VP3054=mCONFIG_VIDEO_CX88_MPEG=mCONFIG_VIDEO_BT848=mCONFIG_DVB_BT8XX=mCONFIG_VIDEO_SAA7134=mCONFIG_VIDEO_SAA7134_ALSA=mCONFIG_VIDEO_SAA7134_RC=yCONFIG_VIDEO_SAA7134_DVB=mCONFIG_VIDEO_SAA7164=m
#
# Media digital TV PCI Adapters
#CONFIG_DVB_AV7110_IR=yCONFIG_DVB_AV7110=mCONFIG_DVB_AV7110_OSD=yCONFIG_DVB_BUDGET_CORE=mCONFIG_DVB_BUDGET=mCONFIG_DVB_BUDGET_CI=mCONFIG_DVB_BUDGET_AV=mCONFIG_DVB_BUDGET_PATCH=mCONFIG_DVB_B2C2_FLEXCOP_PCI=m
# CONFIG_DVB_B2C2_FLEXCOP_PCI_DEBUG is not setCONFIG_DVB_PLUTO2=mCONFIG_DVB_DM1105=mCONFIG_DVB_PT1=m
# CONFIG_DVB_PT3 is not setCONFIG_MANTIS_CORE=mCONFIG_DVB_MANTIS=mCONFIG_DVB_HOPPER=mCONFIG_DVB_NGENE=mCONFIG_DVB_DDBRIDGE=m
# CONFIG_DVB_DDBRIDGE_MSIENABLE is not set
# CONFIG_DVB_SMIPCIE is not set
# CONFIG_DVB_NETUP_UNIDVB is not set
# CONFIG_V4L_PLATFORM_DRIVERS is not set
# CONFIG_V4L_MEM2MEM_DRIVERS is not set
# CONFIG_V4L_TEST_DRIVERS is not set
# CONFIG_DVB_PLATFORM_DRIVERS is not set
#
# Supported MMC/SDIO adapters
#CONFIG_SMS_SDIO_DRV=mCONFIG_RADIO_ADAPTERS=yCONFIG_RADIO_TEA575X=m
# CONFIG_RADIO_SI470X is not set
# CONFIG_RADIO_SI4713 is not set
# CONFIG_USB_MR800 is not set
# CONFIG_USB_DSBR is not set
# CONFIG_RADIO_MAXIRADIO is not set
# CONFIG_RADIO_SHARK is not set
# CONFIG_RADIO_SHARK2 is not set
# CONFIG_USB_KEENE is not set
# CONFIG_USB_RAREMONO is not set
# CONFIG_USB_MA901 is not set
# CONFIG_RADIO_TEA5764 is not set
# CONFIG_RADIO_SAA7706H is not set
# CONFIG_RADIO_TEF6862 is not set
# CONFIG_RADIO_WL1273 is not set
#
# Texas Instruments WL128x FM driver (ST based)
#
#
# Supported FireWire (IEEE 1394) Adapters
#CONFIG_DVB_FIREDTV=mCONFIG_DVB_FIREDTV_INPUT=yCONFIG_MEDIA_COMMON_OPTIONS=y
#
# common driver options # CONFIG_VIDEO_CX2341X=m CONFIG_VIDEO_TVEEPROM=m CONFIG_CYPRESS_FIRMWARE=m CONFIG_VIDEOBUF2_CORE=m CONFIG_VIDEOBUF2_V4L2=m CONFIG_VIDEOBUF2_MEMOPS=m CONFIG_VIDEOBUF2_VMALLOC=m CONFIG_VIDEOBUF2_DMA_SG=m CONFIG_VIDEOBUF2_DVB=m CONFIG_DVB_B2C2_FLEXCOP=m CONFIG_VIDEO_SAA7146=m CONFIG_VIDEO_SAA7146_VV=m CONFIG_SMS_SIANO_MDTV=m CONFIG_SMS_SIANO_RC=y # CONFIG_SMS_SIANO_DEBUGFS is not set
#
# Media ancillary drivers (tuners, sensors, i2c, spi, frontends) #CONFIG_MEDIA_SUBDRV_AUTOSELECT=yCONFIG_MEDIA_ATTACH=yCONFIG_VIDEO_IR_I2C=m
#
# Audio decoders, processors and mixers
#CONFIG_VIDEO_TVAUDIO=mCONFIG_VIDEO_TDA7432=mCONFIG_VIDEO_MSP3400=mCONFIG_VIDEO_CS3308=mCONFIG_VIDEO_CS5345=mCONFIG_VIDEO_CS53L32A=mCONFIG_VIDEO_WM8775=mCONFIG_VIDEO_WM8739=mCONFIG_VIDEO_VP27SMPX=m
#
# RDS decoders
#CONFIG_VIDEO_SAA6588=m
#
# Video decoders
#CONFIG_VIDEO_SAA711X=m
#
# Video and audio decoders
#CONFIG_VIDEO_SAA717X=mCONFIG_VIDEO_CX25840=m
#
# Video encoders
#CONFIG_VIDEO_SAA7127=m
#
# Camera sensor devices
#
#
# Flash devices
#
#
# Video improvement chips
#CONFIG_VIDEO_UPD64031A=mCONFIG_VIDEO_UPD64083=m
#
# Audio/Video compression chips
#CONFIG_VIDEO_SAA6752HS=m
#
# SDR tuner chips
#
#
# Miscellaneous helper chips
#CONFIG_VIDEO_M52790=m
#
# Media SPI Adapters
#
# CONFIG_CXD2880_SPI_DRV is not setCONFIG_MEDIA_TUNER=mCONFIG_MEDIA_TUNER_SIMPLE=mCONFIG_MEDIA_TUNER_TDA18250=mCONFIG_MEDIA_TUNER_TDA8290=mCONFIG_MEDIA_TUNER_TDA827X=mCONFIG_MEDIA_TUNER_TDA18271=mCONFIG_MEDIA_TUNER_TDA9887=mCONFIG_MEDIA_TUNER_TEA5761=mCONFIG_MEDIA_TUNER_TEA5767=mCONFIG_MEDIA_TUNER_MT20XX=mCONFIG_MEDIA_TUNER_MT2060=mCONFIG_MEDIA_TUNER_MT2063=mCONFIG_MEDIA_TUNER_MT2266=mCONFIG_MEDIA_TUNER_MT2131=mCONFIG_MEDIA_TUNER_QT1010=mCONFIG_MEDIA_TUNER_XC2028=mCONFIG_MEDIA_TUNER_XC5000=mCONFIG_MEDIA_TUNER_XC4000=mCONFIG_MEDIA_TUNER_MXL5005S=mCONFIG_MEDIA_TUNER_MXL5007T=mCONFIG_MEDIA_TUNER_MC44S803=mCONFIG_MEDIA_TUNER_MAX2165=mCONFIG_MEDIA_TUNER_TDA18218=mCONFIG_MEDIA_TUNER_FC0011=mCONFIG_MEDIA_TUNER_FC0012=mCONFIG_MEDIA_TUNER_FC0013=mCONFIG_MEDIA_TUNER_TDA18212=mCONFIG_MEDIA_TUNER_E4000=mCONFIG_MEDIA_TUNER_FC2580=mCONFIG_MEDIA_TUNER_M88RS6000T=mCONFIG_MEDIA_TUNER_TUA9001=mCONFIG_MEDIA_TUNER_SI2157=mCONFIG_MEDIA_TUNER_IT913X=mCONFIG_MEDIA_TUNER_R820T=mCONFIG_MEDIA_TUNER_QM1D1C0042=mCONFIG_MEDIA_TUNER_QM1D1B0004=m
#
# Multistandard (satellite) frontends
#CONFIG_DVB_STB0899=mCONFIG_DVB_STB6100=m
CONFIG_DVB_STV090x=mCONFIG_DVB_STV0910=m
CONFIG_DVB_STV6110x=mCONFIG_DVB_STV6111=mCONFIG_DVB_MXL5XX=mCONFIG_DVB_M88DS3103=m
#
# Multistandard (cable + terrestrial) frontends
#CONFIG_DVB_DRXK=mCONFIG_DVB_TDA18271C2DD=mCONFIG_DVB_SI2165=mCONFIG_DVB_MN88472=mCONFIG_DVB_MN88473=m
#
# DVB-S (satellite) frontends
#CONFIG_DVB_CX24110=mCONFIG_DVB_CX24123=mCONFIG_DVB_MT312=mCONFIG_DVB_ZL10036=mCONFIG_DVB_ZL10039=mCONFIG_DVB_S5H1420=mCONFIG_DVB_STV0288=mCONFIG_DVB_STB6000=mCONFIG_DVB_STV0299=mCONFIG_DVB_STV6110=mCONFIG_DVB_STV0900=mCONFIG_DVB_TDA8083=mCONFIG_DVB_TDA10086=mCONFIG_DVB_TDA8261=mCONFIG_DVB_VES1X93=mCONFIG_DVB_TUNER_ITD1000=mCONFIG_DVB_TUNER_CX24113=mCONFIG_DVB_TDA826X=mCONFIG_DVB_TUA6100=mCONFIG_DVB_CX24116=mCONFIG_DVB_CX24117=mCONFIG_DVB_CX24120=mCONFIG_DVB_SI21XX=mCONFIG_DVB_TS2020=mCONFIG_DVB_DS3000=mCONFIG_DVB_MB86A16=mCONFIG_DVB_TDA10071=m
#
# DVB-T (terrestrial) frontends
#CONFIG_DVB_SP8870=mCONFIG_DVB_SP887X=mCONFIG_DVB_CX22700=mCONFIG_DVB_CX22702=mCONFIG_DVB_DRXD=mCONFIG_DVB_L64781=mCONFIG_DVB_TDA1004X=mCONFIG_DVB_NXT6000=mCONFIG_DVB_MT352=mCONFIG_DVB_ZL10353=mCONFIG_DVB_DIB3000MB=mCONFIG_DVB_DIB3000MC=mCONFIG_DVB_DIB7000M=mCONFIG_DVB_DIB7000P=mCONFIG_DVB_TDA10048=mCONFIG_DVB_AF9013=mCONFIG_DVB_EC100=mCONFIG_DVB_STV0367=mCONFIG_DVB_CXD2820R=mCONFIG_DVB_CXD2841ER=mCONFIG_DVB_RTL2830=mCONFIG_DVB_RTL2832=mCONFIG_DVB_SI2168=mCONFIG_DVB_GP8PSK_FE=m
#
# DVB-C (cable) frontends
#CONFIG_DVB_VES1820=mCONFIG_DVB_TDA10021=mCONFIG_DVB_TDA10023=mCONFIG_DVB_STV0297=m
#
# ATSC (North American/Korean Terrestrial/Cable DTV) frontends #CONFIG_DVB_NXT200X=mCONFIG_DVB_OR51211=mCONFIG_DVB_OR51132=mCONFIG_DVB_BCM3510=mCONFIG_DVB_LGDT330X=mCONFIG_DVB_LGDT3305=mCONFIG_DVB_LGDT3306A=mCONFIG_DVB_LG2160=mCONFIG_DVB_S5H1409=mCONFIG_DVB_AU8522=mCONFIG_DVB_AU8522_DTV=mCONFIG_DVB_AU8522_V4L=mCONFIG_DVB_S5H1411=m
#
# ISDB-T (terrestrial) frontends
#CONFIG_DVB_S921=mCONFIG_DVB_DIB8000=mCONFIG_DVB_MB86A20S=m
#
# ISDB-S (satellite) & ISDB-T (terrestrial) frontends #CONFIG_DVB_TC90522=m
#
# Digital terrestrial only tuners/PLL
#CONFIG_DVB_PLL=mCONFIG_DVB_TUNER_DIB0070=mCONFIG_DVB_TUNER_DIB0090=m
#
# SEC control devices for DVB-S
#CONFIG_DVB_DRX39XYJ=mCONFIG_DVB_LNBH25=mCONFIG_DVB_LNBP21=mCONFIG_DVB_LNBP22=mCONFIG_DVB_ISL6405=mCONFIG_DVB_ISL6421=mCONFIG_DVB_ISL6423=mCONFIG_DVB_A8293=mCONFIG_DVB_LGS8GXX=mCONFIG_DVB_ATBM8830=m
CONFIG_DVB_TDA665x=mCONFIG_DVB_IX2505V=mCONFIG_DVB_M88RS2000=mCONFIG_DVB_AF9033=m
#
# Common Interface (EN50221) controller drivers
#CONFIG_DVB_CXD2099=m
#
# Tools to develop new frontends
#CONFIG_DVB_DUMMY_FE=m
#
# Graphics support
#CONFIG_AGP=yCONFIG_AGP_AMD64=yCONFIG_AGP_INTEL=yCONFIG_AGP_SIS=yCONFIG_AGP_VIA=yCONFIG_INTEL_GTT=yCONFIG_VGA_ARB=yCONFIG_VGA_ARB_MAX_GPUS=64CONFIG_VGA_SWITCHEROO=yCONFIG_DRM=mCONFIG_DRM_MIPI_DSI=yCONFIG_DRM_DP_AUX_CHARDEV=yCONFIG_DRM_DEBUG_SELFTEST=mCONFIG_DRM_KMS_HELPER=mCONFIG_DRM_KMS_FB_HELPER=yCONFIG_DRM_FBDEV_EMULATION=yCONFIG_DRM_FBDEV_OVERALLOC=100
# CONFIG_DRM_FBDEV_LEAK_PHYS_SMEM is not setCONFIG_DRM_LOAD_EDID_FIRMWARE=y
# CONFIG_DRM_DP_CEC is not setCONFIG_DRM_TTM=mCONFIG_DRM_VM=yCONFIG_DRM_SCHED=m
#
# I2C encoder or helper chips
#CONFIG_DRM_I2C_CH7006=mCONFIG_DRM_I2C_SIL164=m
# CONFIG_DRM_I2C_NXP_TDA998X is not set
# CONFIG_DRM_I2C_NXP_TDA9950 is not set
#
# ARM devices
#CONFIG_DRM_RADEON=m
# CONFIG_DRM_RADEON_USERPTR is not setCONFIG_DRM_AMDGPU=m
# CONFIG_DRM_AMDGPU_SI is not set
# CONFIG_DRM_AMDGPU_CIK is not set
# CONFIG_DRM_AMDGPU_USERPTR is not set
# CONFIG_DRM_AMDGPU_GART_DEBUGFS is not set
#
# ACP (Audio CoProcessor) Configuration
#
# CONFIG_DRM_AMD_ACP is not set
#
# Display Engine Configuration
#CONFIG_DRM_AMD_DC=yCONFIG_DRM_AMD_DC_DCN1_0=yCONFIG_DRM_AMD_DC_DCN1_01=y
# CONFIG_DEBUG_KERNEL_DC is not set
# CONFIG_HSA_AMD is not set
#
# AMD Library routines
#CONFIG_CHASH=m
# CONFIG_CHASH_STATS is not set
# CONFIG_CHASH_SELFTEST is not setCONFIG_DRM_NOUVEAU=mCONFIG_NOUVEAU_DEBUG=5CONFIG_NOUVEAU_DEBUG_DEFAULT=3
# CONFIG_NOUVEAU_DEBUG_MMU is not setCONFIG_DRM_NOUVEAU_BACKLIGHT=y
# CONFIG_DRM_NOUVEAU_SVM is not setCONFIG_DRM_I915=m
# CONFIG_DRM_I915_ALPHA_SUPPORT is not setCONFIG_DRM_I915_CAPTURE_ERROR=yCONFIG_DRM_I915_COMPRESS_ERROR=yCONFIG_DRM_I915_USERPTR=yCONFIG_DRM_I915_GVT=yCONFIG_DRM_I915_GVT_KVMGT=m
#
# drm/i915 Debugging # # CONFIG_DRM_I915_WERROR is not set # CONFIG_DRM_I915_DEBUG is not set # CONFIG_DRM_I915_SW_FENCE_DEBUG_OBJECTS is not set # CONFIG_DRM_I915_SW_FENCE_CHECK_DAG is not set # CONFIG_DRM_I915_DEBUG_GUC is not set # CONFIG_DRM_I915_SELFTEST is not set # CONFIG_DRM_I915_LOW_LEVEL_TRACEPOINTS is not set # CONFIG_DRM_I915_DEBUG_VBLANK_EVADE is not set # CONFIG_DRM_I915_DEBUG_RUNTIME_PM is not set CONFIG_DRM_VGEM=m # CONFIG_DRM_VKMS is not set CONFIG_DRM_VMWGFX=m CONFIG_DRM_VMWGFX_FBCON=y CONFIG_DRM_GMA500=m CONFIG_DRM_GMA600=y CONFIG_DRM_GMA3600=y CONFIG_DRM_UDL=m CONFIG_DRM_AST=m CONFIG_DRM_MGAG200=m CONFIG_DRM_CIRRUS_QEMU=m CONFIG_DRM_QXL=m CONFIG_DRM_BOCHS=m CONFIG_DRM_VIRTIO_GPU=m CONFIG_DRM_PANEL=y
3,280 lines not shown · raw message
Part of this message is not shown here. View the raw message.