| Age | Commit message (Collapse) | Author |
|---|
|
There's a small window where a deadlock can happen between fallocate and
mmap. This is described in detail by Filipe:
"""
When doing a fallocate operation we lock the inode, flush delalloc within
the target range, wait for any ordered extents to complete and then lock
the file range. Before we lock the range and after we flush delalloc,
there is a time window where another task can come in and do a memory
mapped write for a page within the fallocate range.
This means that after fallocate locks the range, there can be a dirty page
in the range. More often than not, this does not cause any problem.
The exception is when we are low on available metadata space, because an
fallocate operation needs to start a transaction while holding the file
range locked, either through btrfs_prealloc_file_range() or through the
call to btrfs_fallocate_update_isize(). If that's the case, we can end up
in a deadlock. The following list of steps explains how that happens:
1) A fallocate operation starts, locks the inode, flushes delalloc in the
range and waits for ordered extents in the range to complete;
2) Before the fallocate task locks the file range, another task does a
memory mapped write for a page in the fallocate target range. This is
possible since memory mapped writes do not (and can not) lock the
inode;
3) The fallocate task locks the file range. At this point there is one
dirty page in the range (due to the memory mapped write);
4) When the fallocate task attempts to start a transaction, it blocks when
attempting to reserve metadata space, since we are low on available
metadata space. Before blocking (wait on its reservation ticket), it
starts the async reclaim task (if not running already);
5) The async reclaim task is not able to release space through any other
means, so it decides to flush delalloc for inodes with dirty pages.
It finds that the inode used in the fallocate operation has a dirty
page and therefore queues a job (fs_info->flush_workers workqueue) to
flush delalloc for that inode and waits on that job to complete;
6) The flush job blocks when attempting to lock the file range because
it is currently locked by the fallocate task;
7) The fallocate task keeps waiting for its metadata reservation, waiting
for a wakeup on its reservation ticket. The async reclaim task is
waiting on the flush job, which in turn is waiting for locking the file
range that is currently locked by the fallocate task. So unless some
other task is able to release enough metadata space, for example an
ordered extent for some other inode completes, we end up in a deadlock
between all these tasks.
When this happens stack traces like the following show up in dmesg/syslog:
INFO: task kworker/u16:11:1810830 blocked for more than 120 seconds.
Tainted: G B W 5.10.0-rc4-btrfs-next-73 #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:kworker/u16:11 state:D stack: 0 pid:1810830 ppid: 2 flags:0x00004000
Workqueue: btrfs-flush_delalloc btrfs_work_helper [btrfs]
Call Trace:
__schedule+0x5d1/0xcf0
schedule+0x45/0xe0
lock_extent_bits+0x1e6/0x2d0 [btrfs]
? finish_wait+0x90/0x90
btrfs_invalidatepage+0x32c/0x390 [btrfs]
? __mod_memcg_state+0x8e/0x160
__extent_writepage+0x2d4/0x400 [btrfs]
extent_write_cache_pages+0x2b2/0x500 [btrfs]
? lock_release+0x20e/0x4c0
? trace_hardirqs_on+0x1b/0xf0
extent_writepages+0x43/0x90 [btrfs]
? lock_acquire+0x1a3/0x490
do_writepages+0x43/0xe0
? __filemap_fdatawrite_range+0xa4/0x100
__filemap_fdatawrite_range+0xc5/0x100
btrfs_run_delalloc_work+0x17/0x40 [btrfs]
btrfs_work_helper+0xf1/0x600 [btrfs]
process_one_work+0x24e/0x5e0
worker_thread+0x50/0x3b0
? process_one_work+0x5e0/0x5e0
kthread+0x153/0x170
? kthread_mod_delayed_work+0xc0/0xc0
ret_from_fork+0x22/0x30
INFO: task kworker/u16:1:2426217 blocked for more than 120 seconds.
Tainted: G B W 5.10.0-rc4-btrfs-next-73 #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:kworker/u16:1 state:D stack: 0 pid:2426217 ppid: 2 flags:0x00004000
Workqueue: events_unbound btrfs_async_reclaim_metadata_space [btrfs]
Call Trace:
__schedule+0x5d1/0xcf0
? kvm_clock_read+0x14/0x30
? wait_for_completion+0x81/0x110
schedule+0x45/0xe0
schedule_timeout+0x30c/0x580
? _raw_spin_unlock_irqrestore+0x3c/0x60
? lock_acquire+0x1a3/0x490
? try_to_wake_up+0x7a/0xa20
? lock_release+0x20e/0x4c0
? lock_acquired+0x199/0x490
? wait_for_completion+0x81/0x110
wait_for_completion+0xab/0x110
start_delalloc_inodes+0x2af/0x390 [btrfs]
btrfs_start_delalloc_roots+0x12d/0x250 [btrfs]
flush_space+0x24f/0x660 [btrfs]
btrfs_async_reclaim_metadata_space+0x1bb/0x480 [btrfs]
process_one_work+0x24e/0x5e0
worker_thread+0x20f/0x3b0
? process_one_work+0x5e0/0x5e0
kthread+0x153/0x170
? kthread_mod_delayed_work+0xc0/0xc0
ret_from_fork+0x22/0x30
(...)
several tasks waiting for the inode lock held by the fallocate task below
(...)
RIP: 0033:0x7f61efe73fff
Code: Unable to access opcode bytes at RIP 0x7f61efe73fd5.
RSP: 002b:00007ffc3371bbe8 EFLAGS: 00000202 ORIG_RAX: 000000000000013c
RAX: ffffffffffffffda RBX: 00007ffc3371bea0 RCX: 00007f61efe73fff
RDX: 00000000ffffff9c RSI: 0000560fbd5d90a0 RDI: 00000000ffffff9c
RBP: 00007ffc3371beb0 R08: 0000000000000001 R09: 0000000000000003
R10: 0000560fbd5d7ad0 R11: 0000000000000202 R12: 0000000000000001
R13: 000000000000005e R14: 00007ffc3371bea0 R15: 00007ffc3371beb0
task:fdm-stress state:D stack: 0 pid:2508243 ppid:2508153 flags:0x00000000
Call Trace:
__schedule+0x5d1/0xcf0
? _raw_spin_unlock_irqrestore+0x3c/0x60
schedule+0x45/0xe0
__reserve_bytes+0x4a4/0xb10 [btrfs]
? finish_wait+0x90/0x90
btrfs_reserve_metadata_bytes+0x29/0x190 [btrfs]
btrfs_block_rsv_add+0x1f/0x50 [btrfs]
start_transaction+0x2d1/0x760 [btrfs]
btrfs_replace_file_extents+0x120/0x930 [btrfs]
? btrfs_fallocate+0xdcf/0x1260 [btrfs]
btrfs_fallocate+0xdfb/0x1260 [btrfs]
? filename_lookup+0xf1/0x180
vfs_fallocate+0x14f/0x440
ioctl_preallocate+0x92/0xc0
do_vfs_ioctl+0x66b/0x750
? __do_sys_newfstat+0x53/0x60
__x64_sys_ioctl+0x62/0xb0
do_syscall_64+0x33/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xa9
"""
Fix this by disallowing mmaps from happening while we're doing any of
the fallocate operations on this inode.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Darrick reported a potential issue to me where we could allow mmap
writes after validating a page range matched in the case of dedupe.
Generally we rely on lock page -> lock extent with the ordered flush to
protect us, but this is done after we check the pages because we use the
generic helpers, so we could modify the page in between doing the check
and locking the range.
There also exists a deadlock, as described by Filipe
"""
When cloning a file range, we lock the inodes, flush any delalloc within
the respective file ranges, wait for any ordered extents and then lock the
file ranges in both inodes. This means that right after we flush delalloc
and before we lock the file ranges, memory mapped writes can come in and
dirty pages in the file ranges of the clone operation.
Most of the time this is harmless and causes no problems. However, if we
are low on available metadata space, we can later end up in a deadlock
when starting a transaction to replace file extent items. This happens if
when allocating metadata space for the transaction, we need to wait for
the async reclaim thread to release space and the reclaim thread needs to
flush delalloc for the inode that got the memory mapped write and has its
range locked by the clone task.
Basically what happens is the following:
1) A clone operation locks inodes A and B, flushes delalloc for both
inodes in the respective file ranges and waits for any ordered extents
in those ranges to complete;
2) Before the clone task locks the file ranges, another task does a
memory mapped write (which does not lock the inode) for one of the
inodes of the clone operation. So now we have a dirty page in one of
the ranges used by the clone operation;
3) The clone operation locks the file ranges for inodes A and B;
4) Later, when iterating over the file extents of inode A, the clone
task attempts to start a transaction. There's not enough available
free metadata space, so the async reclaim task is started (if not
running already) and we wait for someone to wake us up on our
reservation ticket;
5) The async reclaim task is not able to release space by any other
means and decides to flush delalloc for the inode of the clone
operation;
6) The workqueue job used to flush the inode blocks when starting
delalloc for the inode, since the file range is currently locked by
the clone task;
7) But the clone task is waiting on its reservation ticket and the async
reclaim task is waiting on the flush job to complete, which can't
progress since the clone task has the file range locked. So unless
some other task is able to release space, for example an ordered
extent for some other inode completes, we have a deadlock between all
these tasks;
When this happens stack traces like the following show up in dmesg/syslog:
INFO: task kworker/u16:11:1810830 blocked for more than 120 seconds.
Tainted: G B W 5.10.0-rc4-btrfs-next-73 #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:kworker/u16:11 state:D stack: 0 pid:1810830 ppid: 2 flags:0x00004000
Workqueue: btrfs-flush_delalloc btrfs_work_helper [btrfs]
Call Trace:
__schedule+0x5d1/0xcf0
schedule+0x45/0xe0
lock_extent_bits+0x1e6/0x2d0 [btrfs]
? finish_wait+0x90/0x90
btrfs_invalidatepage+0x32c/0x390 [btrfs]
? __mod_memcg_state+0x8e/0x160
__extent_writepage+0x2d4/0x400 [btrfs]
extent_write_cache_pages+0x2b2/0x500 [btrfs]
? lock_release+0x20e/0x4c0
? trace_hardirqs_on+0x1b/0xf0
extent_writepages+0x43/0x90 [btrfs]
? lock_acquire+0x1a3/0x490
do_writepages+0x43/0xe0
? __filemap_fdatawrite_range+0xa4/0x100
__filemap_fdatawrite_range+0xc5/0x100
btrfs_run_delalloc_work+0x17/0x40 [btrfs]
btrfs_work_helper+0xf1/0x600 [btrfs]
process_one_work+0x24e/0x5e0
worker_thread+0x50/0x3b0
? process_one_work+0x5e0/0x5e0
kthread+0x153/0x170
? kthread_mod_delayed_work+0xc0/0xc0
ret_from_fork+0x22/0x30
INFO: task kworker/u16:1:2426217 blocked for more than 120 seconds.
Tainted: G B W 5.10.0-rc4-btrfs-next-73 #1
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:kworker/u16:1 state:D stack: 0 pid:2426217 ppid: 2 flags:0x00004000
Workqueue: events_unbound btrfs_async_reclaim_metadata_space [btrfs]
Call Trace:
__schedule+0x5d1/0xcf0
? kvm_clock_read+0x14/0x30
? wait_for_completion+0x81/0x110
schedule+0x45/0xe0
schedule_timeout+0x30c/0x580
? _raw_spin_unlock_irqrestore+0x3c/0x60
? lock_acquire+0x1a3/0x490
? try_to_wake_up+0x7a/0xa20
? lock_release+0x20e/0x4c0
? lock_acquired+0x199/0x490
? wait_for_completion+0x81/0x110
wait_for_completion+0xab/0x110
start_delalloc_inodes+0x2af/0x390 [btrfs]
btrfs_start_delalloc_roots+0x12d/0x250 [btrfs]
flush_space+0x24f/0x660 [btrfs]
btrfs_async_reclaim_metadata_space+0x1bb/0x480 [btrfs]
process_one_work+0x24e/0x5e0
worker_thread+0x20f/0x3b0
? process_one_work+0x5e0/0x5e0
kthread+0x153/0x170
? kthread_mod_delayed_work+0xc0/0xc0
ret_from_fork+0x22/0x30
(...)
several other tasks blocked on inode locks held by the clone task below
(...)
RIP: 0033:0x7f61efe73fff
Code: Unable to access opcode bytes at RIP 0x7f61efe73fd5.
RSP: 002b:00007ffc3371bbe8 EFLAGS: 00000202 ORIG_RAX: 000000000000013c
RAX: ffffffffffffffda RBX: 00007ffc3371bea0 RCX: 00007f61efe73fff
RDX: 00000000ffffff9c RSI: 0000560fbd604690 RDI: 00000000ffffff9c
RBP: 00007ffc3371beb0 R08: 0000000000000002 R09: 0000560fbd5d75f0
R10: 0000560fbd5d81f0 R11: 0000000000000202 R12: 0000000000000002
R13: 000000000000000b R14: 00007ffc3371bea0 R15: 00007ffc3371beb0
task: fdm-stress state:D stack: 0 pid:2508234 ppid:2508153 flags:0x00004000
Call Trace:
__schedule+0x5d1/0xcf0
? _raw_spin_unlock_irqrestore+0x3c/0x60
schedule+0x45/0xe0
__reserve_bytes+0x4a4/0xb10 [btrfs]
? finish_wait+0x90/0x90
btrfs_reserve_metadata_bytes+0x29/0x190 [btrfs]
btrfs_block_rsv_add+0x1f/0x50 [btrfs]
start_transaction+0x2d1/0x760 [btrfs]
btrfs_replace_file_extents+0x120/0x930 [btrfs]
? lock_release+0x20e/0x4c0
btrfs_clone+0x3e4/0x7e0 [btrfs]
? btrfs_lookup_first_ordered_extent+0x8e/0x100 [btrfs]
btrfs_clone_files+0xf6/0x150 [btrfs]
btrfs_remap_file_range+0x324/0x3d0 [btrfs]
do_clone_file_range+0xd4/0x1f0
vfs_clone_file_range+0x4d/0x230
? lock_release+0x20e/0x4c0
ioctl_file_clone+0x8f/0xc0
do_vfs_ioctl+0x342/0x750
__x64_sys_ioctl+0x62/0xb0
do_syscall_64+0x33/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xa9
"""
Fix both of these issues by excluding mmaps from happening we are doing
any sort of remap, which prevents this race completely.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
A few places we intermix btrfs_inode_lock with a inode_unlock, and some
places we just use inode_lock/inode_unlock instead of btrfs_inode_lock.
None of these places are using this incorrectly, but as we adjust some
of these callers it would be nice to keep everything consistent, so
convert everybody to use btrfs_inode_lock/btrfs_inode_unlock.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
We need to be able to exclude page_mkwrite from happening concurrently
with certain operations. To facilitate this, add a i_mmap_lock to our
inode, down_read() it in our mkwrite, and add a new ILOCK flag to
indicate that we want to take the i_mmap_lock as well. I used pahole to
check the size of the btrfs_inode, the sizes are as follows
no lockdep:
before: 1120 (3 per 4k page)
after: 1160 (3 per 4k page)
lockdep:
before: 2072 (1 per 4k page)
after: 2224 (1 per 4k page)
We're slightly larger but it doesn't change how many objects we can fit
per page.
Reviewed-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
The parameter mirror is not used and does not make sense for checksum
verification of the given bio.
Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
force_cow can be calculated from inode and does not need to be passed as
an argument.
This simplifies run_delalloc_nocow() call from btrfs_run_delalloc_range()
A new function, should_nocow() checks if the range should be NOCOWed or
not. The function returns true iff either BTRFS_INODE_NODATA or
BTRFS_INODE_PREALLOC, but is not a defrag extent.
Tested-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Fix the following coccicheck warnings:
./fs/btrfs/volumes.c:1462:10-11: WARNING: return of 0/1 in function
'dev_extent_hole_check_zoned' with return type bool.
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Currently we do not do btree read ahead when doing an incremental send,
however we know that we will read and process any node or leaf in the
send root that has a generation greater than the generation of the parent
root. So triggering read ahead for such nodes and leafs is beneficial
for an incremental send.
This change does that, triggers read ahead of any node or leaf in the
send root that has a generation greater then the generation of the
parent root. As for the parent root, no readahead is triggered because
knowing in advance which nodes/leaves are going to be read is not so
linear and there's often a large time window between visiting nodes or
leaves of the parent root. So I opted to leave out the parent root,
and triggering read ahead for its nodes/leaves seemed to have not made
significant difference.
The following test script was used to measure the improvement on a box
using an average, consumer grade, spinning disk and with 16GiB of ram:
$ cat test.sh
#!/bin/bash
DEV=/dev/sdj
MNT=/mnt/sdj
MKFS_OPTIONS="--nodesize 16384" # default, just to be explicit
MOUNT_OPTIONS="-o max_inline=2048" # default, just to be explicit
mkfs.btrfs -f $MKFS_OPTIONS $DEV > /dev/null
mount $MOUNT_OPTIONS $DEV $MNT
# Create files with inline data to make it easier and faster to create
# large btrees.
add_files()
{
local total=$1
local start_offset=$2
local number_jobs=$3
local total_per_job=$(($total / $number_jobs))
echo "Creating $total new files using $number_jobs jobs"
for ((n = 0; n < $number_jobs; n++)); do
(
local start_num=$(($start_offset + $n * $total_per_job))
for ((i = 1; i <= $total_per_job; i++)); do
local file_num=$((start_num + $i))
local file_path="$MNT/file_${file_num}"
xfs_io -f -c "pwrite -S 0xab 0 2000" $file_path > /dev/null
if [ $? -ne 0 ]; then
echo "Failed creating file $file_path"
break
fi
done
) &
worker_pids[$n]=$!
done
wait ${worker_pids[@]}
sync
echo
echo "btree node/leaf count: $(btrfs inspect-internal dump-tree -t 5 $DEV | egrep '^(node|leaf) ' | wc -l)"
}
initial_file_count=500000
add_files $initial_file_count 0 4
echo
echo "Creating first snapshot..."
btrfs subvolume snapshot -r $MNT $MNT/snap1
echo
echo "Adding more files..."
add_files $((initial_file_count / 4)) $initial_file_count 4
echo
echo "Updating 1/50th of the initial files..."
for ((i = 1; i < $initial_file_count; i += 50)); do
xfs_io -c "pwrite -S 0xcd 0 20" $MNT/file_$i > /dev/null
done
echo
echo "Creating second snapshot..."
btrfs subvolume snapshot -r $MNT $MNT/snap2
umount $MNT
echo 3 > /proc/sys/vm/drop_caches
blockdev --flushbufs $DEV &> /dev/null
hdparm -F $DEV &> /dev/null
mount $MOUNT_OPTIONS $DEV $MNT
echo
echo "Testing full send..."
start=$(date +%s)
btrfs send $MNT/snap1 > /dev/null
end=$(date +%s)
echo
echo "Full send took $((end - start)) seconds"
umount $MNT
echo 3 > /proc/sys/vm/drop_caches
blockdev --flushbufs $DEV &> /dev/null
hdparm -F $DEV &> /dev/null
mount $MOUNT_OPTIONS $DEV $MNT
echo
echo "Testing incremental send..."
start=$(date +%s)
btrfs send -p $MNT/snap1 $MNT/snap2 > /dev/null
end=$(date +%s)
echo
echo "Incremental send took $((end - start)) seconds"
umount $MNT
Before this change, incremental send duration:
with $initial_file_count == 200000: 51 seconds
with $initial_file_count == 500000: 168 seconds
After this change, incremental send duration:
with $initial_file_count == 200000: 39 seconds (-26.7%)
with $initial_file_count == 500000: 125 seconds (-29.4%)
For $initial_file_count == 200000 there are 62600 nodes and leaves in the
btree of the first snapshot, and 77759 nodes and leaves in the btree of
the second snapshot. The root nodes were at level 2.
While for $initial_file_count == 500000 there are 152476 nodes and leaves
in the btree of the first snapshot, and 190511 nodes and leaves in the
btree of the second snapshot. The root nodes were at level 2 as well.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
When doing a full send we know that we are going to be reading every node
and leaf of the send root, so we benefit from enabling read ahead for the
btree.
This change enables read ahead for full send operations only, incremental
sends will have read ahead enabled in a different way by a separate patch.
The following test script was used to measure the improvement on a box
using an average, consumer grade, spinning disk and with 16GiB of RAM:
$ cat test.sh
#!/bin/bash
DEV=/dev/sdj
MNT=/mnt/sdj
MKFS_OPTIONS="--nodesize 16384" # default, just to be explicit
MOUNT_OPTIONS="-o max_inline=2048" # default, just to be explicit
mkfs.btrfs -f $MKFS_OPTIONS $DEV > /dev/null
mount $MOUNT_OPTIONS $DEV $MNT
# Create files with inline data to make it easier and faster to create
# large btrees.
add_files()
{
local total=$1
local start_offset=$2
local number_jobs=$3
local total_per_job=$(($total / $number_jobs))
echo "Creating $total new files using $number_jobs jobs"
for ((n = 0; n < $number_jobs; n++)); do
(
local start_num=$(($start_offset + $n * $total_per_job))
for ((i = 1; i <= $total_per_job; i++)); do
local file_num=$((start_num + $i))
local file_path="$MNT/file_${file_num}"
xfs_io -f -c "pwrite -S 0xab 0 2000" $file_path > /dev/null
if [ $? -ne 0 ]; then
echo "Failed creating file $file_path"
break
fi
done
) &
worker_pids[$n]=$!
done
wait ${worker_pids[@]}
sync
echo
echo "btree node/leaf count: $(btrfs inspect-internal dump-tree -t 5 $DEV | egrep '^(node|leaf) ' | wc -l)"
}
initial_file_count=500000
add_files $initial_file_count 0 4
echo
echo "Creating first snapshot..."
btrfs subvolume snapshot -r $MNT $MNT/snap1
echo
echo "Adding more files..."
add_files $((initial_file_count / 4)) $initial_file_count 4
echo
echo "Updating 1/50th of the initial files..."
for ((i = 1; i < $initial_file_count; i += 50)); do
xfs_io -c "pwrite -S 0xcd 0 20" $MNT/file_$i > /dev/null
done
echo
echo "Creating second snapshot..."
btrfs subvolume snapshot -r $MNT $MNT/snap2
umount $MNT
echo 3 > /proc/sys/vm/drop_caches
blockdev --flushbufs $DEV &> /dev/null
hdparm -F $DEV &> /dev/null
mount $MOUNT_OPTIONS $DEV $MNT
echo
echo "Testing full send..."
start=$(date +%s)
btrfs send $MNT/snap1 > /dev/null
end=$(date +%s)
echo
echo "Full send took $((end - start)) seconds"
umount $MNT
echo 3 > /proc/sys/vm/drop_caches
blockdev --flushbufs $DEV &> /dev/null
hdparm -F $DEV &> /dev/null
mount $MOUNT_OPTIONS $DEV $MNT
echo
echo "Testing incremental send..."
start=$(date +%s)
btrfs send -p $MNT/snap1 $MNT/snap2 > /dev/null
end=$(date +%s)
echo
echo "Incremental send took $((end - start)) seconds"
umount $MNT
Before this change, full send duration:
with $initial_file_count == 200000: 165 seconds
with $initial_file_count == 500000: 407 seconds
After this change, full send duration:
with $initial_file_count == 200000: 149 seconds (-10.2%)
with $initial_file_count == 500000: 353 seconds (-14.2%)
For $initial_file_count == 200000 there are 62600 nodes and leaves in the
btree of the first snapshot, while for $initial_file_count == 500000 there
are 152476 nodes and leaves. The roots were at level 2.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
btrfs_block_rsv_add can return only ENOSPC since it's called with
NO_FLUSH modifier. This so simplify the logic in
btrfs_delayed_inode_reserve_metadata to exploit this invariant.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
[ add assert and comment ]
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
It's only used for tracepoint to obtain the inode number, but we already
have the ino from btrfs_delayed_node::inode_id.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
It's no longer expected to call this function with an open transaction
so all the workarounds concerning this can be removed. In fact it'll
constitute a bug to call this function with a transaction already held
so WARN in this case.
Reviewed-by: Qu Wenruo <wqu@suse.com>
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Drop function declarations at the beginning of the file scrub.c. These
functions are defined before they are used in the same file and don't
need forward declaration.
No functional changes.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
btrfs_extent_readonly() checks if the block group is readonly, the bool
return type should be used.
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
btrfs_extent_readonly() is used by can_nocow_extent() in inode.c. So
move it from extent-tree.c to inode.c and declare it as static.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
btrfs_inc_block_group_ro wants to ensure that the current transaction is
not running dirty block groups, if it is it waits and loops again.
That logic is currently implemented using a goto label. Actually using
a proper do {} while() construct doesn't hurt readability nor does it
introduce excessive nesting and makes the relevant code stand out by
being encompassed in the loop construct. No functional changes.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
No point in duplicating the functionality just use the generic helper
that has the same semantics.
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
There is small error in comment about BTRFS_ORDERED_* flags, added in
commit 3c198fe06449 ("btrfs: rework the order of
btrfs_ordered_extent::flags") but the fixup did not get merged in time.
The 4 types are for ordered extent itself, not for direct io.
Only 3 types support direct io, REGULAR/NOCOW/PREALLOC.
Fix the comment to reflect that.
Signed-off-by: Qu Wenruo <wqu@suse.com>
Reviewed-by: David Sterba <dsterba@suse.com>
Signed-off-by: David Sterba <dsterba@suse.com>
|
|
<patrice.chotard@foss.st.com>
Patrice Chotard <patrice.chotard@foss.st.com>:
From: Patrice Chotard <patrice.chotard@foss.st.com>
Christophe Kerello (1):
spi: stm32-qspi: fix pm_runtime usage_count counter
Patrice Chotard (2):
spi: stm32-qspi: Trigger DMA only if more than 4 bytes to transfer
spi: stm32-qspi: Add dirmap support
drivers/spi/spi-stm32-qspi.c | 106 +++++++++++++++++++++++++++--------
1 file changed, 84 insertions(+), 22 deletions(-)
--
2.17.1
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux into arm/dt
More Qualcomm DTS updates for 5.13
This adds CPUfreq, interconnect providers, IPC, remoteproc and IPA to
the SDX55 platform and then adds board files for the Telit FN980 TLB and
Thundercomm TurboX T55.
* tag 'qcom-dts-for-5.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux:
ARM: dts: qcom: sdx55: add IPA information
ARM: dts: qcom: sdx55: Add basic devicetree support for Thundercomm T55
dt-bindings: arm: qcom: Add binding for Thundercomm T55 kit
ARM: dts: qcom: sdx55: Add basic devicetree support for Telit FN980 TLB
dt-bindings: arm: qcom: Add binding for Telit FN980 TLB board
ARM: dts: qcom: sdx55: Add Modem remoteproc node
ARM: dts: qcom: Fix node name for NAND controller node
ARM: dts: qcom: sdx55: Add interconnect nodes
ARM: dts: qcom: sdx55: Add SCM node
dt-bindings: firmware: scm: Add compatible for SDX55
ARM: dts: qcom: sdx55: Add IMEM and PIL info region
ARM: dts: qcom: sdx55: Add modem SMP2P node
ARM: dts: qcom: sdx55: Add CPUFreq support
ARM: dts: qcom: sdx55: Add support for APCS block
ARM: dts: qcom: sdx55: Add support for A7 PLL clock
Link: https://lore.kernel.org/r/20210419150956.860423-1-bjorn.andersson@linaro.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Update iommu property in lpass cpu node for supporting
simultaneous playback on headset and speaker.
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Signed-off-by: V Sujith Kumar Reddy <vsujithk@codeaurora.org>
Signed-off-by: Srinivasa Rao Mandadapu <srivasam@codeaurora.org>
Link: https://lore.kernel.org/r/20210406163330.11996-1-srivasam@codeaurora.org
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
|
|
Match what's downstream for this board.
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Cc: Srinivasa Rao Mandadapu <srivasam@codeaurora.org>
Cc: Ajit Pandey <ajitp@codeaurora.org>
Cc: Judy Hsiao <judyhsiao@chromium.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Cc: Stephen Boyd <swboyd@chromium.org>
Cc: Matthias Kaehlcke <mka@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20210315133924.v2.2.If218189eff613a6c48ba12d75fad992377d8f181@changeid
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
|
|
This was present downstream. Add upstream too. NOTE: upstream I
managed to get some sort of halfway state and got one pinctrl entry in
the coachz-r1 device tree. Remove that as part of this since it's now
in the dtsi.
Reviewed-by: Matthias Kaehlcke <mka@chromium.org>
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Cc: Srinivasa Rao Mandadapu <srivasam@codeaurora.org>
Cc: Ajit Pandey <ajitp@codeaurora.org>
Cc: Judy Hsiao <judyhsiao@chromium.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Cc: Stephen Boyd <swboyd@chromium.org>
Cc: Matthias Kaehlcke <mka@chromium.org>
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Link: https://lore.kernel.org/r/20210315133924.v2.1.I601a051cad7cfd0923e55b69ef7e5748910a6096@changeid
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
|
|
M5Stack are releasing a new widget based on the
SigmaStar SSD202D. We have some support for the
SSD202D so lets add a dts for it.
Signed-off-by: Daniel Palmer <daniel@0x0f.com>
Link: https://m5stack-store.myshopify.com/products/unitv2-ai-camera-gc2145
Link: https://lore.kernel.org/r/20210417011015.2105280-4-daniel@0x0f.com'
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Add a compatible for the M5Stack UnitV2 that is based on the
SigmaStar SSD202D (inifinity2m).
Signed-off-by: Daniel Palmer <daniel@0x0f.com>
Link: https://lore.kernel.org/r/20210417011015.2105280-3-daniel@0x0f.com'
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
M5Stack make various modules for STEM, Makers, IoT.
Their UnitV2 is based on a SigmaStar SSD202D SoC which
we already have some minimal support for so add a
prefix in preparation for UnitV2 board support.
Signed-off-by: Daniel Palmer <daniel@0x0f.com>
Link: https://m5stack.com/
Link: https://lore.kernel.org/r/20210417011015.2105280-2-daniel@0x0f.com'
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Fix unit names to make dtbs_check happy.
Signed-off-by: Matthias Brugger <matthias.bgg@gmail.com>
Reviewed-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
Link: https://lore.kernel.org/r/20210414144643.17435-2-matthias.bgg@kernel.org
Link: https://lore.kernel.org/r/20210416143923.23406-3-matthias.bgg@kernel.org'
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Fix unit names to make dtbs_check happy.
Signed-off-by: Matthias Brugger <matthias.bgg@gmail.com>
Reviewed-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
Link: https://lore.kernel.org/r/20210414144643.17435-1-matthias.bgg@kernel.org
Link: https://lore.kernel.org/r/20210416143923.23406-2-matthias.bgg@kernel.org'
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux-mem-ctrl into arm/drivers
Memory controller drivers for v5.13, part two
1. Renesas RPC: fix possible NULL pointer.
2. Exynos5422 DMC: add proper error checking for clk_prepare.
3. Mediatek SMI: use device-links instead of explicit PM runtime calls.
* tag 'memory-controller-drv-5.13-2' of git://git.kernel.org/pub/scm/linux/kernel/git/krzk/linux-mem-ctrl:
memory: mtk-smi: Add device-link between smi-larb and smi-common
memory: samsung: exynos5422-dmc: handle clk_set_parent() failure
memory: renesas-rpc-if: fix possible NULL pointer dereference of resource
Link: https://lore.kernel.org/r/20210415065514.7385-1-krzysztof.kozlowski@canonical.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
If the setup callback failed, but the controller has auto_runtime_pm
and set_cs, the setup failure could be missed.
Signed-off-by: Joe Burmeister <joe.burmeister@devtank.co.uk>
Link: https://lore.kernel.org/r/20210419130631.4586-1-joe.burmeister@devtank.co.uk
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
When initially probing the SPI slave device, the call for disabling an
SPI device without the SPI_CS_HIGH flag is not applied, as the
condition for checking whether or not the state to be applied equals the
one currently set evaluates to true.
This however might not necessarily be the case, as the chipselect might
be active.
Add a force flag to spi_set_cs which allows to override this
early exit condition. Set it to false everywhere except when called
from spi_setup to sync up the initial CS state.
Fixes commit d40f0b6f2e21 ("spi: Avoid setting the chip select if we don't
need to")
Signed-off-by: David Bauer <mail@david-bauer.net>
Link: https://lore.kernel.org/r/20210416195956.121811-1-mail@david-bauer.net
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Add stm32_qspi_dirmap_read() and stm32_qspi_dirmap_create()
to get dirmap support.
Update the exec_op callback which doens't allow anymore memory map
access. Memory map access are only available through the dirmap_read
callback.
Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>
Link: https://lore.kernel.org/r/20210419121541.11617-4-patrice.chotard@foss.st.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
In order to optimize accesses to spi flashes, trigger a DMA only
if more than 4 bytes has to be transferred.
DMA transfer preparation's cost becomes negligible above 4 bytes to
transfer. Below this threshold, indirect transfer give more throughput.
mtd_speedtest shows that page write throughtput increases :
- from 779 to 853 KiB/s (~9.5%) with s25fl512s SPI-NOR.
- from 5283 to 5666 KiB/s (~7.25%) with Micron SPI-NAND.
Signed-off-by: Christophe Kerello <christophe.kerello@foss.st.com>
Signed-off-by: Patrice Chotard <patrice.chotard@foss.st.com>
Link: https://lore.kernel.org/r/20210419121541.11617-3-patrice.chotard@foss.st.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
pm_runtime usage_count counter is not well managed.
pm_runtime_put_autosuspend callback drops the usage_counter but this
one has never been increased. Add pm_runtime_get_sync callback to bump up
the usage counter. It is also needed to use pm_runtime_force_suspend and
pm_runtime_force_resume APIs to handle properly the clock.
Fixes: 9d282c17b023 ("spi: stm32-qspi: Add pm_runtime support")
Signed-off-by: Christophe Kerello <christophe.kerello@foss.st.com>
Cc: stable@vger.kernel.org
Link: https://lore.kernel.org/r/20210419121541.11617-2-patrice.chotard@foss.st.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Fix a regression caused by making the 486SX separately selectable in
Kconfig, for which the HIGHMEM64G setting has not been updated and
therefore has become exposed as a user-selectable option for the M486SX
configuration setting unlike with original M486 and all the other
settings that choose non-PAE-enabled processors:
High Memory Support
> 1. off (NOHIGHMEM)
2. 4GB (HIGHMEM4G)
3. 64GB (HIGHMEM64G)
choice[1-3?]:
With the fix in place the setting is now correctly removed:
High Memory Support
> 1. off (NOHIGHMEM)
2. 4GB (HIGHMEM4G)
choice[1-2?]:
[ bp: Massage commit message. ]
Fixes: 87d6021b8143 ("x86/math-emu: Limit MATH_EMULATION to 486SX compatibles")
Signed-off-by: Maciej W. Rozycki <macro@orcam.me.uk>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: stable@vger.kernel.org # v5.5+
Link: https://lkml.kernel.org/r/alpine.DEB.2.21.2104141221340.44318@angie.orcam.me.uk
|
|
Fix the following coccicheck warning:
./arch/m68k/include/asm/sun3xflop.h:109:2-3: Unneeded semicolon
Signed-off-by: Wan Jiabing <wanjiabing@vivo.com>
Link: https://lore.kernel.org/r/20210415031450.23379-1-wanjiabing@vivo.com
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
|
|
Add touchscreen info for the Teclast Tbook 11 tablet. This includes info
for getting the firmware directly from the UEFI, so that the user does
not need to manually install the firmware in /lib/firmware/silead.
This change will make the touchscreen on these devices work OOTB,
without requiring any manual setup.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210417173105.4134-1-hdegoede@redhat.com
|
|
Alder PCH-P is based on Tiger Lake PCH.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Rajneesh Bhardwaj <irenic.rajneesh@gmail.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-10-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
Just like Ice Lake, Tiger Lake uses Cannon Lake's LTR information
and supports a few additional registers. Hence add the LTR registers
specific to Tiger Lake to the cnp_ltr_show_map[].
Also adjust the number of LTR IPs for Tiger Lake to the correct amount.
Signed-off-by: Gayatri Kammela <gayatri.kammela@intel.com>
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Rajneesh Bhardwaj <irenic.rajneesh@gmail.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-9-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
By default the Low Power Mode (LPM or sub-state) status registers will
latch condition status on every entry into Package C10. This is
configurable in the PMC to allow latching on any achievable sub-state. Add
a debugfs file to support this.
Also add the option to clear the status registers to 0. Clearing the status
registers before testing removes ambiguity around when the current values
were set.
The new file, latch_lpm_mode, looks like this:
[c10] S0i2.0 S0i3.0 S0i2.1 S0i3.1 S0i3.2 clear
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-8-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
Add the debugfs file, substate_requirements, to view the low power mode
(LPM) requirements for each enabled mode alongside the last latched status
of the condition.
After this patch, the new file will look like this:
Element | S0i2.0 | S0i3.0 | S0i2.1 | S0i3.1 | S0i3.2 | Status |
USB2PLL_OFF_STS | Required | Required | Required | Required | Required | |
PCIe/USB3.1_Gen2PLL_OFF_STS | Required | Required | Required | Required | Required | |
PCIe_Gen3PLL_OFF_STS | Required | Required | Required | Required | Required | Yes |
OPIOPLL_OFF_STS | Required | Required | Required | Required | Required | Yes |
OCPLL_OFF_STS | Required | Required | Required | Required | Required | Yes |
MainPLL_OFF_STS | | Required | | Required | Required | |
Signed-off-by: Gayatri Kammela <gayatri.kammela@intel.com>
Co-developed-by: David E. Box <david.e.box@linux.intel.com>
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-7-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
Platforms that support low power modes (LPM) such as Tiger Lake maintain
requirements for each sub-state that a readable in the PMC. However, unlike
LPM status registers, requirement registers are not memory mapped but are
available from an ACPI _DSM. Collect the requirements for Tiger Lake using
the _DSM method and store in a buffer.
Signed-off-by: Gayatri Kammela <gayatri.kammela@intel.com>
Co-developed-by: David E. Box <david.e.box@linux.intel.com>
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-6-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
Modify the low power mode (LPM or sub-state) residency counters to display
in microseconds just like the slp_s0_residency counter. The granularity of
the counter is approximately 30.5us per tick. Double this value then divide
by two to maintain accuracy.
Signed-off-by: Gayatri Kammela <gayatri.kammela@intel.com>
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Rajneesh Bhardwaj <irenic.rajneesh@gmail.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-5-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
The current implementation of pmc_core_substate_res_show() is written
specifically for Tiger Lake. However, new platform will also have
sub-states and may support different modes. Therefore rewrite the code to
handle sub-states generically.
Obtain the number and type of enabled states form the PMC. Use the Low
Power Mode (LPM) priority register to store the states in order from
shallowest to deepest for displays. Add a for_each macro to simplify
this. While changing the sub-state display it makes sense to show only the
"enabled" sub-states instead of showing all possible ones. After this
patch, the debugfs file looks like this:
Substate Residency
S0i2.0 0
S0i3.0 0
S0i2.1 9329279
S0i3.1 0
S0i3.2 0
Suggested-by: David E. Box <david.e.box@linux.intel.com>
Signed-off-by: Gayatri Kammela <gayatri.kammela@intel.com>
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Rajneesh Bhardwaj <irenic.rajneesh@gmail.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-4-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
The intel_pmc_core driver did not always bind to a device which meant it
lacked a struct device that could be used to maintain driver data. So a
global instance of struct pmc_dev was used for this purpose and functions
accessed this directly. Since the driver now binds to an ACPI device,
remove the global pmc_dev in favor of one that is allocated during probe.
Modify users of the global to obtain the object by argument instead.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Rajneesh Bhardwaj <irenic.rajneesh@gmail.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-3-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
The DMI callbacks, used for quirks, currently access the PMC by getting
the address a global pmc_dev struct. Instead, have the callbacks set a
global quirk specific variable. In probe, after calling dmi_check_system(),
pass pmc_dev to a function that will handle each quirk if its variable
condition is met. This allows removing the global pmc_dev later.
Signed-off-by: David E. Box <david.e.box@linux.intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Rajneesh Bhardwaj <irenic.rajneesh@gmail.com>
Link: https://lore.kernel.org/r/20210417031252.3020837-2-david.e.box@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
|
|
alignment in block mode
Since commit e085b51c74cc ("mmc: meson-gx: check for scatterlist size alignment in block mode"),
support for SDIO SD_IO_RW_EXTENDED transferts are properly filtered but some driver
like brcmfmac still gives a block sg buffer size not aligned with SDIO block,
triggerring a WARN_ONCE() with scary stacktrace even if the transfer works fine
but with possible degraded performances.
Simply replace with dev_warn_once() to inform user this should be fixed to avoid
degraded performance.
This should be ultimately fixed in brcmfmac, but since it's only a performance issue
the warning should be removed.
Fixes: e085b51c74cc ("mmc: meson-gx: check for scatterlist size alignment in block mode")
Reported-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Link: https://lore.kernel.org/r/20210416094347.2015896-1-narmstrong@baylibre.com
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
|
