When there are many DM multipath devices it really helps to have
additional context for which DM device a failed or reinstated path is
part of.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add more DMDEBUG that shows arguments passed and caller, and another
that shows state of related flags at end of queue_if_no_path().
Also add queue_if_no_path DMDEBUG to multipath_resume().
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Do not allow saving disabled queue_if_no_path if already saved as
enabled; implies multiple suspends (which shouldn't ever happen). Log
if this unlikely scenario is ever triggered.
Also, only write MPATHF_SAVED_QUEUE_IF_NO_PATH during presuspend or if
"fail_if_no_path" message. MPATHF_SAVED_QUEUE_IF_NO_PATH is no longer
always modified, e.g.: even if queue_if_no_path()'s save_old_value
argument wasn't set. This just implies a bit tighter control over
the management of MPATHF_SAVED_QUEUE_IF_NO_PATH. Side-effect is
multipath_resume() doesn't reset MPATHF_QUEUE_IF_NO_PATH unless
MPATHF_SAVED_QUEUE_IF_NO_PATH was set (during presuspend); and at that
time the MPATHF_SAVED_QUEUE_IF_NO_PATH bit gets cleared. So
MPATHF_SAVED_QUEUE_IF_NO_PATH's use is much more narrow in scope.
Last, but not least, do _not_ disable queue_if_no_path during noflush
suspend. There is no need/benefit to saving off queue_if_no_path via
MPATHF_SAVED_QUEUE_IF_NO_PATH and clearing MPATHF_QUEUE_IF_NO_PATH for
noflush suspend -- by avoiding this needless queue_if_no_path flag
churn there is less potential for MPATHF_QUEUE_IF_NO_PATH to get lost.
Which avoids potential for IOs to be errored back up to userspace
during DM multipath's handling of path failures.
That said, this last change papers over a reported issue concerning
request-based dm-multipath's interaction with blk-mq, relative to
suspend and resume: multipath_endio is being called _before_
multipath_resume. This should never happen if DM suspend's
blk_mq_quiesce_queue() + dm_wait_for_completion() is genuinely waiting
for all inflight blk-mq requests to complete. Similarly:
drivers/md/dm.c:__dm_resume() clearly calls dm_table_resume_targets()
_before_ dm_start_queue()'s blk_mq_unquiesce_queue() is called. If
the queue isn't even restarted until after multipath_resume(); the BIG
question that still needs answering is: how can multipath_end_io beat
multipath_resume in a race!?
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Remove micro-optimization that infers device is between presuspend and
resume (was done purely to avoid call to dm_noflush_suspending, which
isn't expensive anyway).
Remove flags argument since they are no longer checked.
And remove must_push_back_bio() since it was simply a call to
__must_push_back().
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
When specifying several devices the superblock location must be
checked to ensure the devices are specified in the correct order.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Prefer full zones when selecting the next zone for reclaim.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
per-device reclaim should select zones on that device only.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
When allocating a zone, pass in an indicator on which device the zone
should be allocated; this increases performance for a multi-device
setup because reclaim will now allocate zones on the device for which
reclaim is running.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Remove the hard-coded limit of two devices and support an unlimited
number of additional zoned devices.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Random and sequential zones should be part of the respective
device structure to make arbitration between devices possible.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Instead of having one reclaim workqueue for the entire set we should
be allocating a reclaim workqueue per device; doing so will reduce
contention and should boost performance for a multi-device setup.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add a metadata pointer within struct dmz_dev and use it as argument
for blkdev_report_zones() instead of the metadata itself.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Add a pointer, to the containing device, within struct dm_zone and
kill dmz_zone_to_dev().
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Checking the tertiary superblock just consists of validating UUIDs,
crcs, and the generation number; it doesn't have contents which would
be required during the actual operation.
So allocate a temporary superblock when checking tertiary devices to
avoid having to store it together with the 'real' superblocks.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The zones array is getting really large, and large arrays tend to
wreak havoc with the CPU caches. So convert it to xarray to become
more cache friendly.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Colin Ian King <colin.king@canonical.com> # fix leak in dmz_insert
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Instead of counting the number of reserved zones in dmz_free_zone(),
mark the zone as 'reserved' during allocation and simplify
dmz_free_zone().
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Instead of just reporting the errno, add some more verbose debugging
message in the reclaim path.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
The secondary superblock must reside on the same device as the primary
superblock, so there is no need to re-calculate the device.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Reviewed-by: Damien Le Moal <damien.lemoal@wdc.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Use dm_bufio_forget_buffers instead of a block-by-block loop that
calls dm_bufio_forget. dm_bufio_forget_buffers is faster than the loop
because it searches for used buffers using rb-tree.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Introduce a function forget_buffer_locked that forgets a range of
buffers. It is more efficient than calling forget_buffer in a loop.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
dm-bufio uses unnatural ordering in the rb-tree - blocks with smaller
numbers were put to the right node and blocks with bigger numbers were
put to the left node.
Reverse that logic so that it's natural.
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
When trying to upgrade the iopen glock from a shared to an exclusive lock in
gfs2_evict_inode, abort the wait if there is contention on the corresponding
inode glock: in that case, the inode must still be in active use on another
node, and we're not guaranteed to get the iopen glock anytime soon.
To make this work even better, when we notice contention on the iopen glock and
we can't evict the corresponsing inode and release the iopen glock immediately,
poke the inode glock. The other node(s) trying to acquire the lock can then
abort instead of timing out.
Thanks to Heinz Mauelshagen for pointing out a locking bug in a previous
version of this patch.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
In delete_work_func, if the iopen glock still has an inode attached,
limit the inode lookup to that specific generation number: in the likely
case that the inode was deleted on the node on which the inode's link
count dropped to zero, we can skip verifying the on-disk block type and
reading in the inode. The same applies if another node that had the
inode open managed to delete the inode before us.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Move the inode generation number check from gfs2_lookup_by_inum into
gfs2_inode_lookup: gfs2_inode_lookup may be able to decide that an inode with
the given inode generation number cannot exist without having to verify the
block type or reading the inode from disk.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Use a zero no_formal_ino instead of a NULL pointer to indicate that any inode
generation number will qualify: a valid inode never has a zero no_formal_ino.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
When an inode's link count drops to zero and the inode is cached on
other nodes, the current behavior of gfs2 is to immediately give up and
to rely on the other node(s) to delete the inode if there is iopen glock
contention. This leads to resource group glock bouncing and the loss of
caching. With the previous patches in place, we can fix that by not
giving up immediately.
When the inode is still open on other nodes, those nodes won't be able
to evict the inode and give up the iopen glock. In that case, our lock
conversion request will time out. The unlink system call will block for
the duration of the iopen lock conversion request. We're also holding
the inode glock in EX mode for an extended duration, so other nodes
won't be able to make progress on the inode, either.
This is worse than what we had before, but we can prevent other nodes
from getting stuck by aborting our iopen locking request if there is
contention on the inode glock. This will the the subject of a future
patch.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
When there's contention on the iopen glock, it means that the link count
of the corresponding inode has dropped to zero on a remote node which is
now trying to delete the inode. In that case, try to evict the inode so
that the iopen glock will be released, which will allow the remote node
to do its job.
When the inode is still open locally, the inode's reference count won't
drop to zero and so we'll keep holding the inode and its iopen glock.
The remote node will time out its request to grab the iopen glock, and
when the inode is finally closed locally, we'll try to delete it
ourself.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
This requires flushing delayed work items in gfs2_make_fs_ro (which is called
before unmounting a filesystem).
When inodes are deleted and then recreated, pending gl_delete work items would
have no effect because the inode generations will have changed, so we can
cancel any pending gl_delete works before reusing iopen glocks.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
When deleting an inode, keep track of the generation of the deleted inode in
the inode glock Lock Value Block (LVB). When trying to delete an inode
remotely, check the last-known inode generation against the deleted inode
generation to skip duplicate remote deletes. This avoids taking the resource
group glock in order to verify the block type.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Pull x86 mm updates from Ingo Molnar:
"Misc changes:
- Unexport various PAT primitives
- Unexport per-CPU tlbstate and uninline TLB helpers"
* tag 'x86-mm-2020-06-05' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (23 commits)
x86/tlb/uv: Add a forward declaration for struct flush_tlb_info
x86/cpu: Export native_write_cr4() only when CONFIG_LKTDM=m
x86/tlb: Restrict access to tlbstate
xen/privcmd: Remove unneeded asm/tlb.h include
x86/tlb: Move PCID helpers where they are used
x86/tlb: Uninline nmi_uaccess_okay()
x86/tlb: Move cr4_set_bits_and_update_boot() to the usage site
x86/tlb: Move paravirt_tlb_remove_table() to the usage site
x86/tlb: Move __flush_tlb_all() out of line
x86/tlb: Move flush_tlb_others() out of line
x86/tlb: Move __flush_tlb_one_kernel() out of line
x86/tlb: Move __flush_tlb_one_user() out of line
x86/tlb: Move __flush_tlb_global() out of line
x86/tlb: Move __flush_tlb() out of line
x86/alternatives: Move temporary_mm helpers into C
x86/cr4: Sanitize CR4.PCE update
x86/cpu: Uninline CR4 accessors
x86/tlb: Uninline __get_current_cr3_fast()
x86/mm: Use pgprotval_t in protval_4k_2_large() and protval_large_2_4k()
x86/mm: Unexport __cachemode2pte_tbl
...
This adds checks for gfs2_log_flush being stuck, similarly to the check
in gfs2_ail1_flush. To faciliate this and make the strings easy to grep
we move the ail1 emptying to its own function, empty_ail1_list.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Instructions starting with 0f18 up to 0f1f are reserved nops, except those
that were assigned to MPX. These include the endbr markers used by CET.
List them correctly in the opcode table.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Marcelo reports that kvm selftests fail to build with
"make ARCH=x86_64":
gcc -Wall -Wstrict-prototypes -Wuninitialized -O2 -g -std=gnu99
-fno-stack-protector -fno-PIE -I../../../../tools/include
-I../../../../tools/arch/x86_64/include -I../../../../usr/include/
-Iinclude -Ilib -Iinclude/x86_64 -I.. -c lib/kvm_util.c
-o /var/tmp/20200604202744-bin/lib/kvm_util.o
In file included from lib/kvm_util.c:11:
include/x86_64/processor.h:14:10: fatal error: asm/msr-index.h: No such
file or directory
#include <asm/msr-index.h>
^~~~~~~~~~~~~~~~~
compilation terminated.
"make ARCH=x86", however, works. The problem is that arch specific headers
for x86_64 live in 'tools/arch/x86/include', not in
'tools/arch/x86_64/include'.
Fixes: 66d69e081b ("selftests: fix kvm relocatable native/cross builds and installs")
Reported-by: Marcelo Bandeira Condotta <mcondotta@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20200605142028.550068-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Before this patch, asserts based on glocks did not print the glock with
the error. This patch introduces a new macro, gfs2_glock_assert_withdraw
which first prints the glock, then takes the assert.
This also changes a few glock asserts to the new macro.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
This patch makes the glock dumps in debugfs print the number of pages
(nrpages) for address space glocks. This will aid in debugging.
Signed-off-by: Bob Peterson <rpeterso@redhat.com>
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Under some circumstances, rxrpc will fail a transmit a packet through the
underlying UDP socket (ie. UDP sendmsg returns an error). This may result
in a call getting stuck.
In the instance being seen, where AFS tries to send a probe to the Volume
Location server, tracepoints show the UDP Tx failure (in this case returing
error 99 EADDRNOTAVAIL) and then nothing more:
afs_make_vl_call: c=0000015d VL.GetCapabilities
rxrpc_call: c=0000015d NWc u=1 sp=rxrpc_kernel_begin_call+0x106/0x170 [rxrpc] a=00000000dd89ee8a
rxrpc_call: c=0000015d Gus u=2 sp=rxrpc_new_client_call+0x14f/0x580 [rxrpc] a=00000000e20e4b08
rxrpc_call: c=0000015d SEE u=2 sp=rxrpc_activate_one_channel+0x7b/0x1c0 [rxrpc] a=00000000e20e4b08
rxrpc_call: c=0000015d CON u=2 sp=rxrpc_kernel_begin_call+0x106/0x170 [rxrpc] a=00000000e20e4b08
rxrpc_tx_fail: c=0000015d r=1 ret=-99 CallDataNofrag
The problem is that if the initial packet fails and the retransmission
timer hasn't been started, the call is set to completed and an error is
returned from rxrpc_send_data_packet() to rxrpc_queue_packet(). Though
rxrpc_instant_resend() is called, this does nothing because the call is
marked completed.
So rxrpc_notify_socket() isn't called and the error is passed back up to
rxrpc_send_data(), rxrpc_kernel_send_data() and thence to afs_make_call()
and afs_vl_get_capabilities() where it is simply ignored because it is
assumed that the result of a probe will be collected asynchronously.
Fileserver probing is similarly affected via afs_fs_get_capabilities().
Fix this by always issuing a notification in __rxrpc_set_call_completion()
if it shifts a call to the completed state, even if an error is also
returned to the caller through the function return value.
Also put in a little bit of optimisation to avoid taking the call
state_lock and disabling softirqs if the call is already in the completed
state and remove some now redundant rxrpc_notify_socket() calls.
Fixes: f5c17aaeb2 ("rxrpc: Calls should only have one terminal state")
Reported-by: Gerry Seidman <gerry@auristor.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Marc Dionne <marc.dionne@auristor.com>
Move the handling of call completion out of line so that the next patch can
add more code in that area.
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Marc Dionne <marc.dionne@auristor.com>
To add SW BOOST support for CPPC, we need to get the max frequency of
boost mode and non-boost mode. ACPI spec 6.2 section 8.4.7.1 describes
the following two CPC registers.
"Highest performance is the absolute maximum performance an individual
processor may reach, assuming ideal conditions. This performance level
may not be sustainable for long durations, and may only be achievable if
other platform components are in a specific state; for example, it may
require other processors be in an idle state.
Nominal Performance is the maximum sustained performance level of the
processor, assuming ideal operating conditions. In absence of an
external constraint (power, thermal, etc.) this is the performance level
the platform is expected to be able to maintain continuously. All
processors are expected to be able to sustain their nominal performance
state simultaneously."
To add SW BOOST support for CPPC, we can use Highest Performance as the
max performance in boost mode and Nominal Performance as the max
performance in non-boost mode. If the Highest Performance is greater
than the Nominal Performance, we assume SW BOOST is supported.
The current CPPC driver does not support SW BOOST and use 'Highest
Performance' as the max performance the CPU can achieve. 'Nominal
Performance' is used to convert 'performance' to 'frequency'. That
means, if firmware enable boost and provide a value for Highest
Performance which is greater than Nominal Performance, boost feature is
enabled by default.
Because SW BOOST is disabled by default, so, after this patch, boost
feature is disabled by default even if boost is enabled by firmware.
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Suggested-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
[ rjw: Subject ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Macro 'for_each_active_policy()' is defined internally. To avoid some
cpufreq driver needing this macro to iterate over all the policies in
'.set_boost' callback, we redefine '.set_boost' to act on only one
policy and pass the policy as an argument.
'cpufreq_boost_trigger_state()' iterates over all the policies to set
boost for the system.
This is preparation for adding SW BOOST support for CPPC.
To protect Boost enable/disable by sysfs from CPU online/offline,
add 'cpu_hotplug_lock' before calling '.set_boost' for each CPU.
Also move the lock from 'set_boost()' to 'store_cpb()' in
acpi_cpufreq.
Signed-off-by: Xiongfeng Wang <wangxiongfeng2@huawei.com>
Suggested-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
[ rjw: Subject & changelog ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
'swsusp_header_init()' is only called via 'core_initcall'.
It can be marked as __init to save a few bytes of memory.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
[ rjw: Subject ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
