| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Use the updated memcpy_mcsafe() implementation to define
copy_user_mcsafe() and copy_to_iter_mcsafe(). The most significant
difference from typical copy_to_iter() is that the ITER_KVEC and
ITER_BVEC iterator types can fail to complete a full transfer.
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: hch@lst.de
Cc: linux-fsdevel@vger.kernel.org
Cc: linux-nvdimm@lists.01.org
Link: http://lkml.kernel.org/r/152539239150.31796.9189779163576449784.stgit@dwillia2-desk3.amr.corp.intel.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
|
|
If we have multiple callers of sbq_wake_up(), we can end up in a
situation where the wait_cnt will continually go more and more
negative. Consider the case where our wake batch is 1, hence
wait_cnt will start out as 1.
wait_cnt == 1
CPU0 CPU1
atomic_dec_return(), cnt == 0
atomic_dec_return(), cnt == -1
cmpxchg(-1, 0) (succeeds)
[wait_cnt now 0]
cmpxchg(0, 1) (fails)
This ends up with wait_cnt being 0, we'll wakeup immediately
next time. Going through the same loop as above again, and
we'll have wait_cnt -1.
For the case where we have a larger wake batch, the only
difference is that the starting point will be higher. We'll
still end up with continually smaller batch wakeups, which
defeats the purpose of the rolling wakeups.
Always reset the wait_cnt to the batch value. Then it doesn't
matter who wins the race. But ensure that whomever does win
the race is the one that increments the ws index and wakes up
our batch count, loser gets to call __sbq_wake_up() again to
account his wakeups towards the next active wait state index.
Fixes: 6c0ca7ae292a ("sbitmap: fix wakeup hang after sbq resize")
Reviewed-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Pull dma-mapping fix from Christoph Hellwig:
"Just one little fix from Jean to avoid a harmless but very annoying
warning, especially for the drm code"
* tag 'dma-mapping-4.17-5' of git://git.infradead.org/users/hch/dma-mapping:
swiotlb: silent unwanted warning "buffer is full"
|
|
If DMA_ATTR_NO_WARN is passed to swiotlb_alloc_buffer(), it should be
passed further down to swiotlb_tbl_map_single(). Otherwise we escape
half of the warnings but still log the other half.
This is one of the multiple causes of spurious warnings reported at:
https://bugs.freedesktop.org/show_bug.cgi?id=104082
Signed-off-by: Jean Delvare <jdelvare@suse.de>
Fixes: 0176adb00406 ("swiotlb: refactor coherent buffer allocation")
Cc: Christoph Hellwig <hch@lst.de>
Cc: Christian König <christian.koenig@amd.com>
Cc: Michel Dänzer <michel@daenzer.net>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: stable@vger.kernel.org # v4.16
|
|
test_find_first_bit() is intentionally sub-optimal, and may cause soft
lockup due to long time of run on some systems. So decrease length of
bitmap to traverse to avoid lockup.
With the change below, time of test execution doesn't exceed 0.2 seconds
on my testing system.
Link: http://lkml.kernel.org/r/20180420171949.15710-1-ynorov@caviumnetworks.com
Fixes: 4441fca0a27f5 ("lib: test module for find_*_bit() functions")
Signed-off-by: Yury Norov <ynorov@caviumnetworks.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
Reorder Kconfig entries, so that menuconfig displays proper indentation.
Link: http://lkml.kernel.org/r/alpine.LRH.2.02.1804251601160.30569@file01.intranet.prod.int.rdu2.redhat.com
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Tested-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
|
|
percpu_ida() decouples disabling interrupts from the locking operations.
This breaks some assumptions if the locking operations are replaced like
they are under -RT.
The same locking can be achieved by avoiding local_irq_save() and using
spin_lock_irqsave() instead. percpu_ida_alloc() gains one more preemption
point because after unlocking the fastpath and before the pool lock is
acquired, the interrupts are briefly enabled.
Link: http://lkml.kernel.org/r/20180504153218.7301-1-bigeasy@linutronix.de
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Nicholas Bellinger <nab@linux-iscsi.org>
Cc: Shaohua Li <shli@fb.com>
Cc: Kent Overstreet <kent.overstreet@gmail.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
|
|
idr.c needs xarray.h
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
|
|
Improve the scalability of the IDA by using the per-IDA xa_lock rather
than the global simple_ida_lock. IDAs are not typically used in
performance-sensitive locations, but since we have this lock anyway, we
can use it. It is also a step towards converting the IDA from the radix
tree to the XArray.
Link: http://lkml.kernel.org/r/20180331125332.GF13332@bombadil.infradead.org
Signed-off-by: Matthew Wilcox <mawilcox@microsoft.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
|
|
Use BITS_TO_LONGS() macro to avoid calculation of reminder (bits %
BITS_PER_LONG) On ARM64 it saves 5 instruction for function - 16 before
and 11 after.
Link: http://lkml.kernel.org/r/20180411145914.6011-1-ynorov@caviumnetworks.com
Signed-off-by: Yury Norov <ynorov@caviumnetworks.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
|
|
Fix a race in the multi-order iteration code which causes the kernel to
hit a GP fault. This was first seen with a production v4.15 based kernel
(4.15.6-300.fc27.x86_64) utilizing a DAX workload which used order 9 PMD
DAX entries.
The race has to do with how we tear down multi-order sibling entries when
we are removing an item from the tree. Remember for example that an order
2 entry looks like this:
struct radix_tree_node.slots[] = [entry][sibling][sibling][sibling]
where 'entry' is in some slot in the struct radix_tree_node, and the three
slots following 'entry' contain sibling pointers which point back to
'entry.'
When we delete 'entry' from the tree, we call :
radix_tree_delete()
radix_tree_delete_item()
__radix_tree_delete()
replace_slot()
replace_slot() first removes the siblings in order from the first to the
last, then at then replaces 'entry' with NULL. This means that for a
brief period of time we end up with one or more of the siblings removed,
so:
struct radix_tree_node.slots[] = [entry][NULL][sibling][sibling]
This causes an issue if you have a reader iterating over the slots in the
tree via radix_tree_for_each_slot() while only under
rcu_read_lock()/rcu_read_unlock() protection. This is a common case in
mm/filemap.c.
The issue is that when __radix_tree_next_slot() => skip_siblings() tries
to skip over the sibling entries in the slots, it currently does so with
an exact match on the slot directly preceding our current slot. Normally
this works:
V preceding slot
struct radix_tree_node.slots[] = [entry][sibling][sibling][sibling]
^ current slot
This lets you find the first sibling, and you skip them all in order.
But in the case where one of the siblings is NULL, that slot is skipped
and then our sibling detection is interrupted:
V preceding slot
struct radix_tree_node.slots[] = [entry][NULL][sibling][sibling]
^ current slot
This means that the sibling pointers aren't recognized since they point
all the way back to 'entry', so we think that they are normal internal
radix tree pointers. This causes us to think we need to walk down to a
struct radix_tree_node starting at the address of 'entry'.
In a real running kernel this will crash the thread with a GP fault when
you try and dereference the slots in your broken node starting at 'entry'.
We fix this race by fixing the way that skip_siblings() detects sibling
nodes. Instead of testing against the preceding slot we instead look for
siblings via is_sibling_entry() which compares against the position of the
struct radix_tree_node.slots[] array. This ensures that sibling entries
are properly identified, even if they are no longer contiguous with the
'entry' they point to.
Link: http://lkml.kernel.org/r/20180503192430.7582-6-ross.zwisler@linux.intel.com
Fixes: 148deab223b2 ("radix-tree: improve multiorder iterators")
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Reported-by: CR, Sapthagirish <sapthagirish.cr@intel.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
|
|
I had neglected to increment the error counter when the tests failed,
which made the tests noisy when they fail, but not actually return an
error code.
Link: http://lkml.kernel.org/r/20180509114328.9887-1-mpe@ellerman.id.au
Fixes: 3cc78125a081 ("lib/test_bitmap.c: add optimisation tests")
Signed-off-by: Matthew Wilcox <mawilcox@microsoft.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Reported-by: Michael Ellerman <mpe@ellerman.id.au>
Tested-by: Michael Ellerman <mpe@ellerman.id.au>
Reviewed-by: Kees Cook <keescook@chromium.org>
Cc: Yury Norov <ynorov@caviumnetworks.com>
Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: <stable@vger.kernel.org> [4.13+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
|
|
test_find_first_bit() is intentionally sub-optimal, and may cause soft
lockup due to long time of run on some systems. So decrease length of
bitmap to traverse to avoid lockup.
With the change below, time of test execution doesn't exceed 0.2 seconds
on my testing system.
Link: http://lkml.kernel.org/r/20180420171949.15710-1-ynorov@caviumnetworks.com
Fixes: 4441fca0a27f5 ("lib: test module for find_*_bit() functions")
Signed-off-by: Yury Norov <ynorov@caviumnetworks.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
|
|
Make sure the user passed the right value to
sbitmap_queue_min_shallow_depth().
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
The sbitmap queue wake batch is calculated such that once allocations
start blocking, all of the bits which are already allocated must be
enough to fulfill the batch counters of all of the waitqueues. However,
the shallow allocation depth can break this invariant, since we block
before our full depth is being utilized. Add
sbitmap_queue_min_shallow_depth(), which saves the minimum shallow depth
the sbq will use, and update sbq_calc_wake_batch() to take it into
account.
Acked-by: Paolo Valente <paolo.valente@linaro.org>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
swiotlb use physical address of bounce buffer when do map and unmap,
therefore, related comment should be updated.
Signed-off-by: Yisheng Xie <xieyisheng1@huawei.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
swiotlb now selects the DMA_DIRECT_OPS config symbol, so this will
always be true.
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
This way we have one central definition of it, and user can select it as
needed. The new option is not user visible, which is the behavior
it had in most architectures, with a few notable exceptions:
- On x86_64 and mips/loongson3 it used to be user selectable, but
defaulted to y. It now is unconditional, which seems like the right
thing for 64-bit architectures without guaranteed availablity of
IOMMUs.
- on powerpc the symbol is user selectable and defaults to n, but
many boards select it. This change assumes no working setup
required a manual selection, but if that turned out to be wrong
we'll have to add another select statement or two for the respective
boards.
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Define this symbol if the architecture either uses 64-bit pointers or the
PHYS_ADDR_T_64BIT is set. This covers 95% of the old arch magic. We only
need an additional select for Xen on ARM (why anyway?), and we now always
set ARCH_DMA_ADDR_T_64BIT on mips boards with 64-bit physical addressing
instead of only doing it when highmem is set.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: James Hogan <jhogan@kernel.org>
|
|
This way we have one central definition of it, and user can select it as
needed. Note that we now also always select it when CONFIG_DMA_API_DEBUG
is select, which fixes some incorrect checks in a few network drivers.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
|
|
This way we have one central definition of it, and user can select it as
needed.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
|
|
This way we have one central definition of it, and user can select it as
needed.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
|
|
This avoids selecting IOMMU_HELPER just for this function. And we only
use it once or twice in normal builds so this often even is a size
reduction.
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
This function is only used by built-in code.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
|
|
This code is only used by sparc, and all new iommu drivers should use the
drivers/iommu/ framework. Also remove the unused exports.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: David S. Miller <davem@davemloft.net>
Reviewed-by: Anshuman Khandual <khandual@linux.vnet.ibm.com>
|
|
There is no arch specific code required for dma-debug, so there is no
need to opt into the support either.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
|
|
Only used by the AMD GART and Intel VT-D drivers, which must be built in.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
|
|
Just keep a single variable with a descriptive name instead of two
with confusing names.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
|
|
Most mainstream architectures are using 65536 entries, so lets stick to
that. If someone is really desperate to override it that can still be
done through <asm/dma-mapping.h>, but I'd rather see a really good
rationale for that.
dma_debug_init is now called as a core_initcall, which for many
architectures means much earlier, and provides dma-debug functionality
earlier in the boot process. This should be safe as it only relies
on the memory allocator already being available.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Marek Szyprowski <m.szyprowski@samsung.com>
Reviewed-by: Robin Murphy <robin.murphy@arm.com>
|
|
Minor conflict, a CHECK was placed into an if() statement
in net-next, whilst a newline was added to that CHECK
call in 'net'. Thanks to Daniel for the merge resolution.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This was used by the ide, scsi and networking code in the past to
determine if they should bounce payloads. Now that the dma mapping
always have to support dma to all physical memory (thanks to swiotlb
for non-iommu systems) there is no need to this crude hack any more.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Palmer Dabbelt <palmer@sifive.com> (for riscv)
Reviewed-by: Jens Axboe <axboe@kernel.dk>
|
|
As the recent swiotlb bug revealed, we seem to have given up the direct
DMA allocation too early and felt back to swiotlb allocation. The reason
is that swiotlb allocator expected that dma_direct_alloc() would try
harder to get pages even below 64bit DMA mask with GFP_DMA32, but the
function doesn't do that but only deals with GFP_DMA case.
This patch adds a similar fallback reallocation with GFP_DMA32 as we've
done with GFP_DMA. The condition is that the coherent mask is smaller
than 64bit (i.e. some address limitation), and neither GFP_DMA nor
GFP_DMA32 is set beforehand.
Signed-off-by: Takashi Iwai <tiwai@suse.de>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Smatch complains here:
lib/swiotlb.c:730 swiotlb_alloc_buffer()
warn: variable dereferenced before check 'dev' (see line 716)
"dev" isn't ever NULL in this function so we can just remove the check.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
Overlapping changes in selftests Makefile.
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
Remove all eBPF tests involving LD_ABS/LD_IND from test_bpf.ko. Reason
is that the eBPF tests from test_bpf module do not go via BPF verifier
and therefore any instruction rewrites from verifier cannot take place.
Therefore, move them into test_verifier which runs out of user space,
so that verfier can rewrite LD_ABS/LD_IND internally in upcoming patches.
It will have the same effect since runtime tests are also performed from
there. This also allows to finally unexport bpf_skb_vlan_{push,pop}_proto
and keep it internal to core kernel.
Additionally, also add further cBPF LD_ABS/LD_IND test coverage into
test_bpf.ko suite.
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
|
|
Make n signed to avoid leaking the pages array if __pipe_get_pages()
fails to allocate any pages.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
It returns -EFAULT and happens to be a helper for pipe_get_pages()
whose return type is ssize_t.
Signed-off-by: Ilya Dryomov <idryomov@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
The result was printing the warning only when we were explicitly asked
not to.
Cc: stable@vger.kernel.org
Fixes: 0176adb004065d6815a8e67946752df4cd947c5b "swiotlb: refactor
coherent buffer allocation"
Signed-off-by: Michel Dänzer <michel.daenzer@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>.
Signed-off-by: Christoph Hellwig <hch@lst.de>
|
|
commit 07e98962fa77 ("kobject: Send hotplug events in all network namespaces")
enabled sending hotplug events into all network namespaces back in 2010.
Over time the set of uevents that get sent into all network namespaces has
shrunk. We have now reached the point where hotplug events for all devices
that carry a namespace tag are filtered according to that namespace.
Specifically, they are filtered whenever the namespace tag of the kobject
does not match the namespace tag of the netlink socket.
Currently, only network devices carry namespace tags (i.e. network
namespace tags). Hence, uevents for network devices only show up in the
network namespace such devices are created in or moved to.
However, any uevent for a kobject that does not have a namespace tag
associated with it will not be filtered and we will broadcast it into all
network namespaces. This behavior stopped making sense when user namespaces
were introduced.
This patch simplifies and fixes couple of things:
- Split codepath for sending uevents by kobject namespace tags:
1. Untagged kobjects - uevent_net_broadcast_untagged():
Untagged kobjects will be broadcast into all uevent sockets recorded
in uevent_sock_list, i.e. into all network namespacs owned by the
intial user namespace.
2. Tagged kobjects - uevent_net_broadcast_tagged():
Tagged kobjects will only be broadcast into the network namespace they
were tagged with.
Handling of tagged kobjects in 2. does not cause any semantic changes.
This is just splitting out the filtering logic that was handled by
kobj_bcast_filter() before.
Handling of untagged kobjects in 1. will cause a semantic change. The
reasons why this is needed and ok have been discussed in [1]. Here is a
short summary:
- Userspace ignores uevents from network namespaces that are not owned by
the intial user namespace:
Uevents are filtered by userspace in a user namespace because the
received uid != 0. Instead the uid associated with the event will be
65534 == "nobody" because the global root uid is not mapped.
This means we can safely and without introducing regressions modify the
kernel to not send uevents into all network namespaces whose owning
user namespace is not the initial user namespace because we know that
userspace will ignore the message because of the uid anyway.
I have a) verified that is is true for every udev implementation out
there b) that this behavior has been present in all udev
implementations from the very beginning.
- Thundering herd:
Broadcasting uevents into all network namespaces introduces significant
overhead.
All processes that listen to uevents running in non-initial user
namespaces will end up responding to uevents that will be meaningless
to them. Mainly, because non-initial user namespaces cannot easily
manage devices unless they have a privileged host-process helping them
out. This means that there will be a thundering herd of activity when
there shouldn't be any.
- Removing needless overhead/Increasing performance:
Currently, the uevent socket for each network namespace is added to the
global variable uevent_sock_list. The list itself needs to be protected
by a mutex. So everytime a uevent is generated the mutex is taken on
the list. The mutex is held *from the creation of the uevent (memory
allocation, string creation etc. until all uevent sockets have been
handled*. This is aggravated by the fact that for each uevent socket
that has listeners the mc_list must be walked as well which means we're
talking O(n^2) here. Given that a standard Linux workload usually has
quite a lot of network namespaces and - in the face of containers - a
lot of user namespaces this quickly becomes a performance problem (see
"Thundering herd" above). By just recording uevent sockets of network
namespaces that are owned by the initial user namespace we
significantly increase performance in this codepath.
- Injecting uevents:
There's a valid argument that containers might be interested in
receiving device events especially if they are delegated to them by a
privileged userspace process. One prime example are SR-IOV enabled
devices that are explicitly designed to be handed of to other users
such as VMs or containers.
This use-case can now be correctly handled since
commit 692ec06d7c92 ("netns: send uevent messages"). This commit
introduced the ability to send uevents from userspace. As such we can
let a sufficiently privileged (CAP_SYS_ADMIN in the owning user
namespace of the network namespace of the netlink socket) userspace
process make a decision what uevents should be sent. This removes the
need to blindly broadcast uevents into all user namespaces and provides
a performant and safe solution to this problem.
- Filtering logic:
This patch filters by *owning user namespace of the network namespace a
given task resides in* and not by user namespace of the task per se.
This means if the user namespace of a given task is unshared but the
network namespace is kept and is owned by the initial user namespace a
listener that is opening the uevent socket in that network namespace
can still listen to uevents.
- Fix permission for tagged kobjects:
Network devices that are created or moved into a network namespace that
is owned by a non-initial user namespace currently are send with
INVALID_{G,U}ID in their credentials. This means that all current udev
implementations in userspace will ignore the uevent they receive for
them. This has lead to weird bugs whereby new devices showing up in such
network namespaces were not recognized and did not get IPs assigned etc.
This patch adjusts the permission to the appropriate {g,u}id in the
respective user namespace. This way udevd is able to correctly handle
such devices.
- Simplify filtering logic:
do_one_broadcast() already ensures that only listeners in mc_list receive
uevents that have the same network namespace as the uevent socket itself.
So the filtering logic in kobj_bcast_filter is not needed (see [3]). This
patch therefore removes kobj_bcast_filter() and replaces
netlink_broadcast_filtered() with the simpler netlink_broadcast()
everywhere.
[1]: https://lkml.org/lkml/2018/4/4/739
[2]: https://lkml.org/lkml/2018/4/26/767
[3]: https://lkml.org/lkml/2018/4/26/738
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
This patch adds alloc_uevent_skb() in preparation for follow up patches.
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux
Pull errseq infrastructure fix from Jeff Layton:
"The PostgreSQL developers recently had a spirited discussion about the
writeback error handling in Linux, and reached out to us about a
behavoir change to the code that bit them when the errseq_t changes
were merged.
When we changed to using errseq_t for tracking writeback errors, we
lost the ability for an application to see a writeback error that
occurred before the open on which the fsync was issued. This was
problematic for PostgreSQL which offloads fsync calls to a completely
separate process from the DB writers.
This patch restores that ability. If the errseq_t value in the inode
does not have the SEEN flag set, then we just return 0 for the sample.
That ensures that any recorded error is always delivered at least
once.
Note that we might still lose the error if the inode gets evicted from
the cache before anything can reopen it, but that was the case before
errseq_t was merged. At LSF/MM we had some discussion about keeping
inodes with unreported writeback errors around in the cache for longer
(possibly indefinitely), but that's really a separate problem"
* tag 'errseq-v4.17' of git://git.kernel.org/pub/scm/linux/kernel/git/jlayton/linux:
errseq: Always report a writeback error once
|
