Propagate the rate changes to MFG_BG3D's parent on MT8186 to allow
for proper GPU DVFS.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-10-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
The main/univpll clocks are used as clock sources for multiple
peripherals of different kind, some of which are critical (like AXIs);
a rate change on any of these two will produce a rate change on many
devices and that's likely to produce system instability if not done
correctly: this is the reason why we have (a lot of) "fixed factor"
main/univpll divider clocks, used by MUX clocks to provide different
rates based on PLL output dividers.
Following what was done on clk-mt8186-topckgen and also preventing the
same GPU DVFS issue, drop CLK_SET_RATE_PARENT from the aforementioned
clocks.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-9-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
The main/univpll clocks are used as clock sources for multiple
peripherals of different kind, some of which are critical (like AXIs);
a rate change on any of these two will produce a rate change on many
devices and that's likely to produce system instability if not done
correctly: this is the reason why we have (a lot of) "fixed factor"
main/univpll divider clocks, used by MUX clocks to provide different
rates based on PLL output dividers.
Following what was done on clk-mt8186-topckgen and also preventing the
same GPU DVFS issue, drop CLK_SET_RATE_PARENT from the aforementioned
clocks.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-8-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
The main/sys/univpll clocks are used as clock sources for multiple
peripherals of different kind, some of which are critical (like AXIs);
a rate change on any of these two will produce a rate change on many
devices and that's likely to produce system instability if not done
correctly: this is the reason why we have (a lot of) "fixed factor"
main/sys/univpll divider clocks, used by MUX clocks to provide
different rates based on PLL output dividers.
Following what was done on clk-mt8186-topckgen and also preventing the
same GPU DVFS issue, drop CLK_SET_RATE_PARENT from the aforementioned
clocks.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-7-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
The main/sys/univpll clocks are used as clock sources for multiple
peripherals of different kind, some of which are critical (like AXIs);
a rate change on any of these two will produce a rate change on many
devices and that's likely to produce system instability if not done
correctly: this is the reason why we have (a lot of) "fixed factor"
main/sys/univpll divider clocks, used by MUX clocks to provide
different rates based on PLL output dividers.
Following what was done on clk-mt8186-topckgen and also preventing the
same GPU DVFS issue, drop CLK_SET_RATE_PARENT from the aforementioned
clocks.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-6-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
The syspll and univpll clocks are used as clock sources for multiple
peripherals of different kind, some of which are critical (like AXIs);
a rate change on any of these two will produce a rate change on many
devices and that's likely to produce system instability if not done
correctly: this is the reason why we have (a lot of) "fixed factor"
sys/univpll divider clocks, used by MUX clocks to provide different
rates based on PLL output dividers.
Following what was done on clk-mt8186-topckgen and also solving the
same GPU DVFS issue, drop CLK_SET_RATE_PARENT from the aforementioned
clocks.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-5-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
There's no need to split each FACTOR entry in two lines, as each of
them does fit in one line just fine.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-4-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
The mainpll and univpll clocks are used as clock sources for multiple
peripherals of different kind, some of which are critical (like AXIs);
a rate change on any of these two will produce a rate change on many
devices and that's likely to produce system instability if not done
correctly: this is the reason why we have "fixed factor" clocks, used
by MUX clocks to provide different rates based on PLL output dividers.
Though, there's one fundamental issue that must be resolved somehow:
When performing GPU DVFS, we get a rate request that will try to change
the frequency of MAINPLL due to the CLK_TOP_MFG mux having clk26m,
mfgpll (the GPU dedicated PLL), mainpll_d3, mainpll_d5 (fixed factor
dividers) as possible parents.
In order to solve that, there are two ways:
1. Add new "fake" mainpll_d3_fixed, mainpll_d5_fixed clocks, clones
of mainpll_d3, mainpll_d5 clocks, for the only purpose of not
declaring CLK_SET_RATE_PARENT; or
2. Simply drop said flag from the original dividers.
After some careful validation, I cannot see anything calling a rate
change request during runtime for MAINPLL, nor for UNIVPLL (which would,
again, mean that we're reclocking lots of peripherals at once!), so it
is safe *and sane* to simply remove the CLK_SET_RATE_PARENT flag to all
of the main/univpll fixed factor divider clocks.
Besides, if for any (doubtful) reason main/univpll rate change will be
required in the future, it's still possible to call that on the PLL main
clocks, so we're still covered anyway.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-3-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
Before this change, every mtk_fixed_factor clock forced clock flags to
CLK_SET_RATE_PARENT: while this is harmless in some cases, it may not
be desired in some others, especially when performing clock muxing on
a clock having multiple parents of which one is a dedicated PLL and the
others are not.
This is especially seen on the GPU clocks on some SoCs, where we are
muxing between multiple parents: a fixed clock (crystal), a programmable
GPU PLL and one or more dividers for the MAINPLL, used for a number of
devices; it happens that when a rate change is called for the GPU, the
clock framework will try to satisfy the rate request by using one of the
MAINPLL dividers, which have CLK_SET_RATE_PARENT and will set the rate
on MAINPLL itself - overclocking or underclocking many devices in the
system - and making it to lock up.
Logically, it should be harmless (and would only reduce possible bugs)
to change all of the univpll and mainpll related fixed factor clocks
to not declare the CLK_SET_RATE_PARENT by default but, on some SoCs,
this is also used for dividers of other PLLs for which a rate change
based on the divider may be desired, hence introduce a new FACTOR_FLAGS()
macro to use custom flags (or none) on selected fixed factor clocks.
Signed-off-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Chen-Yu Tsai <wenst@chromium.org>
Link: https://lore.kernel.org/r/20221024102307.33722-2-angelogioacchino.delregno@collabora.com
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
On XE_LPM+ platforms the media engines are carved out into a separate
GT but have a common GGTMMADR address range which essentially makes
the GGTT address space to be shared between media and render GT. As a
result any updates in GGTT shall invalidate TLB of GTs sharing it and
similarly any operation on GGTT requiring an action on a GT will have to
involve all GTs sharing it. setup_private_pat was being done on a per
GGTT based as that doesn't touch any GGTT structures moved it to per GT
based.
BSPEC: 63834
v2:
1. Add details to commit msg
2. includes fix for failure to add item to ggtt->gt_list, as suggested
by Lucas
3. as ggtt_flush() is used only for ggtt drop i915_is_ggtt check within
it.
4. setup_private_pat moved out of intel_gt_tiles_init
v3:
1. Move out for_each_gt from i915_driver.c (Jani Nikula)
v4: drop using RCU primitives on ggtt->gt_list as it is not an RCU list
(Matt Roper)
Cc: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Aravind Iddamsetty <aravind.iddamsetty@intel.com>
Reviewed-by: Matt Roper <matthew.d.roper@intel.com>
Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20221122070126.4813-1-aravind.iddamsetty@intel.com
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQTFp0I1jqZrAX+hPRXbK58LschIgwUCY4AC5QAKCRDbK58LschI
g1e0AQCfAqduTy7mYd02jDNCV0wLphNp9FbPiP9OrQT37ABpKAEA1ulj1X59bX3d
HnZdDKuatcPZT9MV5hDLM7MFJ9GjOA4=
=fNmM
-----END PGP SIGNATURE-----
Daniel Borkmann says:
====================
bpf-next 2022-11-25
We've added 101 non-merge commits during the last 11 day(s) which contain
a total of 109 files changed, 8827 insertions(+), 1129 deletions(-).
The main changes are:
1) Support for user defined BPF objects: the use case is to allocate own
objects, build own object hierarchies and use the building blocks to
build own data structures flexibly, for example, linked lists in BPF,
from Kumar Kartikeya Dwivedi.
2) Add bpf_rcu_read_{,un}lock() support for sleepable programs,
from Yonghong Song.
3) Add support storing struct task_struct objects as kptrs in maps,
from David Vernet.
4) Batch of BPF map documentation improvements, from Maryam Tahhan
and Donald Hunter.
5) Improve BPF verifier to propagate nullness information for branches
of register to register comparisons, from Eduard Zingerman.
6) Fix cgroup BPF iter infra to hold reference on the start cgroup,
from Hou Tao.
7) Fix BPF verifier to not mark fentry/fexit program arguments as trusted
given it is not the case for them, from Alexei Starovoitov.
8) Improve BPF verifier's realloc handling to better play along with dynamic
runtime analysis tools like KASAN and friends, from Kees Cook.
9) Remove legacy libbpf mode support from bpftool,
from Sahid Orentino Ferdjaoui.
10) Rework zero-len skb redirection checks to avoid potentially breaking
existing BPF test infra users, from Stanislav Fomichev.
11) Two small refactorings which are independent and have been split out
of the XDP queueing RFC series, from Toke Høiland-Jørgensen.
12) Fix a memory leak in LSM cgroup BPF selftest, from Wang Yufen.
13) Documentation on how to run BPF CI without patch submission,
from Daniel Müller.
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
====================
Link: https://lore.kernel.org/r/20221125012450.441-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The fileserver probing code attempts to work out the best fileserver to
use for a volume by retrieving the RTT calculated by AF_RXRPC for the
probe call sent to each server and comparing them. Sometimes, however,
no RTT estimate is available and rxrpc_kernel_get_srtt() returns false,
leading good fileservers to be given an RTT of UINT_MAX and thus causing
the rotation algorithm to ignore them.
Fix afs_select_fileserver() to ignore rxrpc_kernel_get_srtt()'s return
value and just take the estimated RTT it provides - which will be capped
at 1 second.
Fixes: 1d4adfaf65 ("rxrpc: Make rxrpc_kernel_get_srtt() indicate validity")
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Marc Dionne <marc.dionne@auristor.com>
Tested-by: Marc Dionne <marc.dionne@auristor.com>
cc: linux-afs@lists.infradead.org
Link: https://lore.kernel.org/r/166965503999.3392585.13954054113218099395.stgit@warthog.procyon.org.uk/
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Rename deadline_is_seq_writes() to deadline_is_seq_write() (remove the
"s" plural) to more correctly reflect the fact that this function tests
a single request, not multiple requests.
Fixes: 015d02f485 ("block: mq-deadline: Do not break sequential write streams to zoned HDDs")
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
Link: https://lore.kernel.org/r/20221126025550.967914-2-damien.lemoal@opensource.wdc.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
I got the following report while doing device(mscc-miim) load test
with CONFIG_OF_UNITTEST and CONFIG_OF_DYNAMIC enabled:
OF: ERROR: memory leak, expected refcount 1 instead of 2,
of_node_get()/of_node_put() unbalanced - destroy cset entry:
attach overlay node /spi/soc@0/mdio@7107009c/ethernet-phy@0
If the 'fwnode' is not an acpi node, the refcount is get in
fwnode_mdiobus_phy_device_register(), but it has never been
put when the device is freed in the normal path. So call
fwnode_handle_put() in phy_device_release() to avoid leak.
If it's an acpi node, it has never been get, but it's put
in the error path, so call fwnode_handle_get() before
phy_device_register() to keep get/put operation balanced.
Fixes: bc1bee3b87 ("net: mdiobus: Introduce fwnode_mdiobus_register_phy()")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://lore.kernel.org/r/20221124150130.609420-1-yangyingliang@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The skb is delivered to netif_rx() which may free it, after calling this,
dereferencing skb may trigger use-after-free.
Fixes: f421436a59 ("net/hsr: Add support for the High-availability Seamless Redundancy protocol (HSRv0)")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20221125075724.27912-1-yuehaibing@huawei.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As the call trace shows, the original skb was freed in tipc_msg_validate(),
and dereferencing the old skb cb would cause an use-after-free crash.
BUG: KASAN: use-after-free in tipc_crypto_rcv_complete+0x1835/0x2240 [tipc]
Call Trace:
<IRQ>
tipc_crypto_rcv_complete+0x1835/0x2240 [tipc]
tipc_crypto_rcv+0xd32/0x1ec0 [tipc]
tipc_rcv+0x744/0x1150 [tipc]
...
Allocated by task 47078:
kmem_cache_alloc_node+0x158/0x4d0
__alloc_skb+0x1c1/0x270
tipc_buf_acquire+0x1e/0xe0 [tipc]
tipc_msg_create+0x33/0x1c0 [tipc]
tipc_link_build_proto_msg+0x38a/0x2100 [tipc]
tipc_link_timeout+0x8b8/0xef0 [tipc]
tipc_node_timeout+0x2a1/0x960 [tipc]
call_timer_fn+0x2d/0x1c0
...
Freed by task 47078:
tipc_msg_validate+0x7b/0x440 [tipc]
tipc_crypto_rcv_complete+0x4b5/0x2240 [tipc]
tipc_crypto_rcv+0xd32/0x1ec0 [tipc]
tipc_rcv+0x744/0x1150 [tipc]
This patch fixes it by re-fetching the skb cb from the new allocated skb
after calling tipc_msg_validate().
Fixes: fc1b6d6de2 ("tipc: introduce TIPC encryption & authentication")
Reported-by: Shuang Li <shuali@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Link: https://lore.kernel.org/r/1b1cdba762915325bd8ef9a98d0276eb673df2a5.1669398403.git.lucien.xin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Matthieu Baerts says:
====================
mptcp: More fixes for 6.1
Patch 1 makes sure data received after a close will still be processed
and acked as exepected. This is a regression for a commit introduced
in v5.11.
Patch 2 fixes a kernel deadlock found when working on validating TFO
with a listener MPTCP socket. This is not directly linked to TFO but
it is easier to reproduce the issue with it. This fixes a bug introduced
by a commit from v6.0.
====================
Link: https://lore.kernel.org/r/20221128154239.1999234-1-matthieu.baerts@tessares.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
All of the subflows of a msk will be orphaned in mptcp_close(), which
means the subflows are in DEAD state. After then, DATA_FIN will be sent,
and the other side will response with a DATA_ACK for this DATA_FIN.
However, if the other side still has pending data, the data that received
on these subflows will not be passed to the msk, as they are DEAD and
subflow_data_ready() will not be called in tcp_data_ready(). Therefore,
these data can't be acked, and they will be retransmitted again and again,
until timeout.
Fix this by setting ssk->sk_socket and ssk->sk_wq to 'NULL', instead of
orphaning the subflows in __mptcp_close(), as Paolo suggested.
Fixes: e16163b6e2 ("mptcp: refactor shutdown and close")
Reviewed-by: Biao Jiang <benbjiang@tencent.com>
Reviewed-by: Mengen Sun <mengensun@tencent.com>
Signed-off-by: Menglong Dong <imagedong@tencent.com>
Reviewed-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Matthieu Baerts <matthieu.baerts@tessares.net>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This patch changes the reported ethtool statistics for the lan9303
family of parts covered by this driver.
The TxUnderRun statistic label is renamed to RxShort to accurately
reflect what stat the device is reporting. I did not reorder the
statistics as that might cause problems with existing user code that
are expecting the stats at a certain offset.
Fixes: a1292595e0 ("net: dsa: add new DSA switch driver for the SMSC-LAN9303")
Signed-off-by: Jerry Ray <jerry.ray@microchip.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Link: https://lore.kernel.org/r/20221128193559.6572-1-jerry.ray@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Third, and hopefully final, set of fixes for v6.1. We are marking the
rsi driver as orphan, have some Information Element parsing fixes to
wilc1000 driver and three small fixes to the stack.
-----BEGIN PGP SIGNATURE-----
iQFFBAABCgAvFiEEiBjanGPFTz4PRfLobhckVSbrbZsFAmOEnFsRHGt2YWxvQGtl
cm5lbC5vcmcACgkQbhckVSbrbZuhVQf/c3MdS++rufVGBcplf7zG5MG8j6pJS9bx
ymuyqsqKoLuPDjLemFIbpcEkkLjt63KYNa8A9wkQ8vSfeg0wPVEym0b234q3XMb0
aAlRBurFAYdIql+aiZI2FxOlKfdLC6d/Sjr/FOe25QfX4H4FxBfZM94nobLawu/d
YBGiI29pauEFqZqBU4I3MAtVbk8/gDIVU318KQavpLuPJZNX5F83M025aokDCVj+
3l69MvphYMVvfOQKS8gLFBBJ41WpSrjIkuH0oQW1fmVc8b6Kiu9wGahOdxDm4JE6
iDiy8LdkvqpXehPMsHPqL/JLMByuJ+q/esZUNoX4vlITordNm/Nj0w==
=iNqa
-----END PGP SIGNATURE-----
Merge tag 'wireless-2022-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless
Kalle Valo says:
====================
wireless fixes for v6.1
Third, and hopefully final, set of fixes for v6.1. We are marking the
rsi driver as orphan, have some Information Element parsing fixes to
wilc1000 driver and three small fixes to the stack.
* tag 'wireless-2022-11-28' of git://git.kernel.org/pub/scm/linux/kernel/git/wireless/wireless:
wifi: mac8021: fix possible oob access in ieee80211_get_rate_duration
wifi: cfg80211: don't allow multi-BSSID in S1G
wifi: cfg80211: fix buffer overflow in elem comparison
wifi: wilc1000: validate number of channels
wifi: wilc1000: validate length of IEEE80211_P2P_ATTR_CHANNEL_LIST attribute
wifi: wilc1000: validate length of IEEE80211_P2P_ATTR_OPER_CHANNEL attribute
wifi: wilc1000: validate pairwise and authentication suite offsets
MAINTAINERS: mark rsi wifi driver as orphan
====================
Link: https://lore.kernel.org/r/20221128113513.6F459C433C1@smtp.kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a new error injection knob so that we can arbitrarily slow down
pagecache writes to test for race conditions and aberrant reclaim
behavior if the writeback mechanisms are slow to issue writeback. This
will enable functional testing for the ifork sequence counters
introduced in commit 304a68b9c6 ("xfs: use iomap_valid method to
detect stale cached iomaps") that fixes write racing with reclaim
writeback.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Add a new error injection knob so that we can arbitrarily slow down
writeback to test for race conditions and aberrant reclaim behavior if
the writeback mechanisms are slow to issue writeback. This will enable
functional testing for the ifork sequence counters introduced in commit
745b3f76d1 ("xfs: maintain a sequence count for inode fork
manipulations").
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
This patch series fixes a data corruption that occurs in a specific
multi-threaded write workload. The workload combined
racing unaligned adjacent buffered writes with low memory conditions
that caused both writeback and memory reclaim to race with the
writes.
The result of this was random partial blocks containing zeroes
instead of the correct data. The underlying problem is that iomap
caches the write iomap for the duration of the write() operation,
but it fails to take into account that the extent underlying the
iomap can change whilst the write is in progress.
The short story is that an iomap can span mutliple folios, and so
under low memory writeback can be cleaning folios the write()
overlaps. Whilst the overlapping data is cached in memory, this
isn't a problem, but because the folios are now clean they can be
reclaimed. Once reclaimed, the write() does the wrong thing when
re-instantiating partial folios because the iomap no longer reflects
the underlying state of the extent. e.g. it thinks the extent is
unwritten, so it zeroes the partial range, when in fact the
underlying extent is now written and so it should have read the data
from disk. This is how we get random zero ranges in the file
instead of the correct data.
The gory details of the race condition can be found here:
https://lore.kernel.org/linux-xfs/20220817093627.GZ3600936@dread.disaster.area/
Fixing the problem has two aspects. The first aspect of the problem
is ensuring that iomap can detect a stale cached iomap during a
write in a race-free manner. We already do this stale iomap
detection in the writeback path, so we have a mechanism for
detecting that the iomap backing the data range may have changed
and needs to be remapped.
In the case of the write() path, we have to ensure that the iomap is
validated at a point in time when the page cache is stable and
cannot be reclaimed from under us. We also need to validate the
extent before we start performing any modifications to the folio
state or contents. Combine these two requirements together, and the
only "safe" place to validate the iomap is after we have looked up
and locked the folio we are going to copy the data into, but before
we've performed any initialisation operations on that folio.
If the iomap fails validation, we then mark it stale, unlock the
folio and end the write. This effectively means a stale iomap
results in a short write. Filesystems should already be able to
handle this, as write operations can end short for many reasons and
need to iterate through another mapping cycle to be completed. Hence
the iomap changes needed to detect and handle stale iomaps during
write() operations is relatively simple...
However, the assumption is that filesystems should already be able
to handle write failures safely, and that's where the second
(first?) part of the problem exists. That is, handling a partial
write is harder than just "punching out the unused delayed
allocation extent". This is because mmap() based faults can race
with writes, and if they land in the delalloc region that the write
allocated, then punching out the delalloc region can cause data
corruption.
This data corruption problem is exposed by generic/346 when iomap is
converted to detect stale iomaps during write() operations. Hence
write failure in the filesytems needs to handle the fact that the
write() in progress doesn't necessarily own the data in the page
cache over the range of the delalloc extent it just allocated.
As a result, we can't just truncate the page cache over the range
the write() didn't reach and punch all the delalloc extent. We have
to walk the page cache over the untouched range and skip over any
dirty data region in the cache in that range. Which is ....
non-trivial.
That is, iterating the page cache has to handle partially populated
folios (i.e. block size < page size) that contain data. The data
might be discontiguous within a folio. Indeed, there might be
*multiple* discontiguous data regions within a single folio. And to
make matters more complex, multi-page folios mean we just don't know
how many sub-folio regions we might have to iterate to find all
these regions. All the corner cases between the conversions and
rounding between filesystem block size, folio size and multi-page
folio size combined with unaligned write offsets kept breaking my
brain.
However, if we convert the code to track the processed
write regions by byte ranges instead of fileystem block or page
cache index, we could simply use mapping_seek_hole_data() to find
the start and end of each discrete data region within the range we
needed to scan. SEEK_DATA finds the start of the cached data region,
SEEK_HOLE finds the end of the region. These are byte based
interfaces that understand partially uptodate folio regions, and so
can iterate discrete sub-folio data regions directly. This largely
solved the problem of discovering the dirty regions we need to keep
the delalloc extent over.
However, to use mapping_seek_hole_data() without needing to export
it, we have to move all the delalloc extent cleanup to the iomap
core and so now the iomap core can clean up delayed allocation
extents in a safe, sane and filesystem neutral manner.
With all this done, the original data corruption never occurs
anymore, and we now have a generic mechanism for ensuring that page
cache writes do not do the wrong thing when writeback and reclaim
change the state of the physical extent and/or page cache contents
whilst the write is in progress.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
-----BEGIN PGP SIGNATURE-----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=cv9i
-----END PGP SIGNATURE-----
Merge tag 'xfs-iomap-stale-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs into xfs-6.2-mergeB
xfs, iomap: fix data corruption due to stale cached iomaps
This patch series fixes a data corruption that occurs in a specific
multi-threaded write workload. The workload combined
racing unaligned adjacent buffered writes with low memory conditions
that caused both writeback and memory reclaim to race with the
writes.
The result of this was random partial blocks containing zeroes
instead of the correct data. The underlying problem is that iomap
caches the write iomap for the duration of the write() operation,
but it fails to take into account that the extent underlying the
iomap can change whilst the write is in progress.
The short story is that an iomap can span mutliple folios, and so
under low memory writeback can be cleaning folios the write()
overlaps. Whilst the overlapping data is cached in memory, this
isn't a problem, but because the folios are now clean they can be
reclaimed. Once reclaimed, the write() does the wrong thing when
re-instantiating partial folios because the iomap no longer reflects
the underlying state of the extent. e.g. it thinks the extent is
unwritten, so it zeroes the partial range, when in fact the
underlying extent is now written and so it should have read the data
from disk. This is how we get random zero ranges in the file
instead of the correct data.
The gory details of the race condition can be found here:
https://lore.kernel.org/linux-xfs/20220817093627.GZ3600936@dread.disaster.area/
Fixing the problem has two aspects. The first aspect of the problem
is ensuring that iomap can detect a stale cached iomap during a
write in a race-free manner. We already do this stale iomap
detection in the writeback path, so we have a mechanism for
detecting that the iomap backing the data range may have changed
and needs to be remapped.
In the case of the write() path, we have to ensure that the iomap is
validated at a point in time when the page cache is stable and
cannot be reclaimed from under us. We also need to validate the
extent before we start performing any modifications to the folio
state or contents. Combine these two requirements together, and the
only "safe" place to validate the iomap is after we have looked up
and locked the folio we are going to copy the data into, but before
we've performed any initialisation operations on that folio.
If the iomap fails validation, we then mark it stale, unlock the
folio and end the write. This effectively means a stale iomap
results in a short write. Filesystems should already be able to
handle this, as write operations can end short for many reasons and
need to iterate through another mapping cycle to be completed. Hence
the iomap changes needed to detect and handle stale iomaps during
write() operations is relatively simple...
However, the assumption is that filesystems should already be able
to handle write failures safely, and that's where the second
(first?) part of the problem exists. That is, handling a partial
write is harder than just "punching out the unused delayed
allocation extent". This is because mmap() based faults can race
with writes, and if they land in the delalloc region that the write
allocated, then punching out the delalloc region can cause data
corruption.
This data corruption problem is exposed by generic/346 when iomap is
converted to detect stale iomaps during write() operations. Hence
write failure in the filesytems needs to handle the fact that the
write() in progress doesn't necessarily own the data in the page
cache over the range of the delalloc extent it just allocated.
As a result, we can't just truncate the page cache over the range
the write() didn't reach and punch all the delalloc extent. We have
to walk the page cache over the untouched range and skip over any
dirty data region in the cache in that range. Which is ....
non-trivial.
That is, iterating the page cache has to handle partially populated
folios (i.e. block size < page size) that contain data. The data
might be discontiguous within a folio. Indeed, there might be
*multiple* discontiguous data regions within a single folio. And to
make matters more complex, multi-page folios mean we just don't know
how many sub-folio regions we might have to iterate to find all
these regions. All the corner cases between the conversions and
rounding between filesystem block size, folio size and multi-page
folio size combined with unaligned write offsets kept breaking my
brain.
However, if we convert the code to track the processed
write regions by byte ranges instead of fileystem block or page
cache index, we could simply use mapping_seek_hole_data() to find
the start and end of each discrete data region within the range we
needed to scan. SEEK_DATA finds the start of the cached data region,
SEEK_HOLE finds the end of the region. These are byte based
interfaces that understand partially uptodate folio regions, and so
can iterate discrete sub-folio data regions directly. This largely
solved the problem of discovering the dirty regions we need to keep
the delalloc extent over.
However, to use mapping_seek_hole_data() without needing to export
it, we have to move all the delalloc extent cleanup to the iomap
core and so now the iomap core can clean up delayed allocation
extents in a safe, sane and filesystem neutral manner.
With all this done, the original data corruption never occurs
anymore, and we now have a generic mechanism for ensuring that page
cache writes do not do the wrong thing when writeback and reclaim
change the state of the physical extent and/or page cache contents
whilst the write is in progress.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
* tag 'xfs-iomap-stale-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/dgc/linux-xfs:
xfs: drop write error injection is unfixable, remove it
xfs: use iomap_valid method to detect stale cached iomaps
iomap: write iomap validity checks
xfs: xfs_bmap_punch_delalloc_range() should take a byte range
iomap: buffered write failure should not truncate the page cache
xfs,iomap: move delalloc punching to iomap
xfs: use byte ranges for write cleanup ranges
xfs: punching delalloc extents on write failure is racy
xfs: write page faults in iomap are not buffered writes
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQTFp0I1jqZrAX+hPRXbK58LschIgwUCY4ACtQAKCRDbK58LschI
gyIcAP4nE/chC+gYDOleloK2tQlQawM5Sa4kwHjFzBdPD3tRrAEAs6y2fJjb5vZo
OXIFKhXv5Xo3knmynkTSVSVOvB48IAE=
=Cme8
-----END PGP SIGNATURE-----
Merge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Daniel Borkmann says:
====================
bpf 2022-11-25
We've added 10 non-merge commits during the last 8 day(s) which contain
a total of 7 files changed, 48 insertions(+), 30 deletions(-).
The main changes are:
1) Several libbpf ringbuf fixes related to probing for its availability,
size overflows when mmaping a 2G ringbuf and rejection of invalid
reservationsizes, from Hou Tao.
2) Fix a buggy return pointer in libbpf for attach_raw_tp function,
from Jiri Olsa.
3) Fix a local storage BPF map bug where the value's spin lock field
can get initialized incorrectly, from Xu Kuohai.
4) Two follow-up fixes in kprobe_multi BPF selftests for BPF CI,
from Jiri Olsa.
* tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf:
selftests/bpf: Make test_bench_attach serial
selftests/bpf: Filter out default_idle from kprobe_multi bench
bpf: Set and check spin lock value in sk_storage_map_test
bpf: Do not copy spin lock field from user in bpf_selem_alloc
libbpf: Check the validity of size in user_ring_buffer__reserve()
libbpf: Handle size overflow for user ringbuf mmap
libbpf: Handle size overflow for ringbuf mmap
libbpf: Use page size as max_entries when probing ring buffer map
bpf, perf: Use subprog name when reporting subprog ksymbol
libbpf: Use correct return pointer in attach_raw_tp
====================
Link: https://lore.kernel.org/r/20221125001034.29473-1-daniel@iogearbox.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
This reverts commit f72cd76b05.
This patch is so broken, it hurts. Apparently no one reviewed it and it
passed the build testing (because the code was compiled out), but it was
obviously never compile-tested, since it produces the following build
error, due to an incomplete conversion where an extra argument was left,
although the function being called was left:
stmmac_main.c: In function ‘stmmac_cmdline_opt’:
stmmac_main.c:7586:28: error: too many arguments to function ‘sysfs_streq’
7586 | } else if (sysfs_streq(opt, "pause:", 6)) {
| ^~~~~~~~~~~
In file included from ../include/linux/bitmap.h:11,
from ../include/linux/cpumask.h:12,
from ../include/linux/smp.h:13,
from ../include/linux/lockdep.h:14,
from ../include/linux/mutex.h:17,
from ../include/linux/notifier.h:14,
from ../include/linux/clk.h:14,
from ../drivers/net/ethernet/stmicro/stmmac/stmmac_main.c:17:
../include/linux/string.h:185:13: note: declared here
185 | extern bool sysfs_streq(const char *s1, const char *s2);
| ^~~~~~~~~~~
What's even worse is that the patch is flat out wrong. The stmmac_cmdline_opt()
function does not parse sysfs input, but cmdline input such as
"stmmaceth=tc:1,pause:1". The pattern of using strsep() followed by
strncmp() for such strings is not unique to stmmac, it can also be found
mainly in drivers under drivers/video/fbdev/.
With strncmp("tc:", 3), the code matches on the "tc:1" token properly.
With sysfs_streq("tc:"), it doesn't.
Fixes: f72cd76b05 ("net: stmmac: use sysfs_streq() instead of strncmp()")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Acked-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Link: https://lore.kernel.org/r/20221125105304.3012153-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add CDC-ECM support for LARA-L6.
LARA-L6 module can be configured (by AT interface) in three different
USB modes:
* Default mode (Vendor ID: 0x1546 Product ID: 0x1341) with 4 serial
interfaces
* RmNet mode (Vendor ID: 0x1546 Product ID: 0x1342) with 4 serial
interfaces and 1 RmNet virtual network interface
* CDC-ECM mode (Vendor ID: 0x1546 Product ID: 0x1343) with 4 serial
interface and 1 CDC-ECM virtual network interface
In CDC-ECM mode LARA-L6 exposes the following interfaces:
If 0: Diagnostic
If 1: AT parser
If 2: AT parser
If 3: AT parset/alternative functions
If 4: CDC-ECM interface
Signed-off-by: Davide Tronchin <davide.tronchin.94@gmail.com>
Link: https://lore.kernel.org/r/20221124112811.3548-1-davide.tronchin.94@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
When the kernel receives a route deletion request from user space it
tries to delete a route that matches the route attributes specified in
the request.
If only prefix information is specified in the request, the kernel
should delete the first matching FIB alias regardless of its associated
FIB info. However, an error is currently returned when the FIB info is
backed by a nexthop object:
# ip nexthop add id 1 via 192.0.2.2 dev dummy10
# ip route add 198.51.100.0/24 nhid 1
# ip route del 198.51.100.0/24
RTNETLINK answers: No such process
Fix by matching on such a FIB info when legacy nexthop attributes are
not specified in the request. An earlier check already covers the case
where a nexthop ID is specified in the request.
Add tests that cover these flows. Before the fix:
# ./fib_nexthops.sh -t ipv4_fcnal
...
TEST: Delete route when not specifying nexthop attributes [FAIL]
Tests passed: 11
Tests failed: 1
After the fix:
# ./fib_nexthops.sh -t ipv4_fcnal
...
TEST: Delete route when not specifying nexthop attributes [ OK ]
Tests passed: 12
Tests failed: 0
No regressions in other tests:
# ./fib_nexthops.sh
...
Tests passed: 228
Tests failed: 0
# ./fib_tests.sh
...
Tests passed: 186
Tests failed: 0
Cc: stable@vger.kernel.org
Reported-by: Jonas Gorski <jonas.gorski@gmail.com>
Tested-by: Jonas Gorski <jonas.gorski@gmail.com>
Fixes: 493ced1ac4 ("ipv4: Allow routes to use nexthop objects")
Fixes: 6bf92d70e6 ("net: ipv4: fix route with nexthop object delete warning")
Fixes: 61b91eb33a ("ipv4: Handle attempt to delete multipath route when fib_info contains an nh reference")
Signed-off-by: Ido Schimmel <idosch@nvidia.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Reviewed-by: David Ahern <dsahern@kernel.org>
Link: https://lore.kernel.org/r/20221124210932.2470010-1-idosch@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The EFI page table is initially created as a copy of the kernel page table.
With VMAP_STACK enabled, kernel stacks are allocated in the vmalloc area:
if the stack is allocated in a new PGD (one that was not present at the
moment of the efi page table creation or not synced in a previous vmalloc
fault), the kernel will take a trap when switching to the efi page table
when the vmalloc kernel stack is accessed, resulting in a kernel panic.
Fix that by updating the efi kernel mappings before switching to the efi
page table.
Signed-off-by: Alexandre Ghiti <alexghiti@rivosinc.com>
Fixes: b91540d52a ("RISC-V: Add EFI runtime services")
Tested-by: Emil Renner Berthing <emil.renner.berthing@canonical.com>
Reviewed-by: Atish Patra <atishp@rivosinc.com>
Link: https://lore.kernel.org/r/20221121133303.1782246-1-alexghiti@rivosinc.com
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
The tracing subsystem now has its own mailing list (although patches
should also be sent to LKML) as well as a new patchwork entry for kernel
related tracing patches.
Update the MAINTAINERS file to reflect the changes.
Link: https://lore.kernel.org/linux-trace-kernel/20221017140513.14b9ce2e@gandalf.local.home
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
The DeviceTree Specification v0.3 specifies that the cache node
'compatible' and 'cache-level' properties are 'required'. Cf.
s3.8 Multi-level and Shared Cache Nodes
The 'cache-unified' property should be present if one of the
properties for unified cache is present ('cache-size', ...).
Update the Device Trees accordingly.
Acked-by: William Zhang <william.zhang@broadcom.com>
Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
Link: https://lore.kernel.org/r/20221122163208.3810985-3-pierre.gondois@arm.com
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
The DeviceTree Specification v0.3 specifies that the cache node
'compatible' and 'cache-level' properties are 'required'. Cf.
s3.8 Multi-level and Shared Cache Nodes
The 'cache-unified' property should be present if one of the
properties for unified cache is present ('cache-size', ...).
Update the Device Trees accordingly.
Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
Link: https://lore.kernel.org/r/20221122163208.3810985-2-pierre.gondois@arm.com
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
The node names should be generic and DT schema expects certain pattern:
bcm4708-asus-rt-ac68u.dtb: leds: 'logo', 'power', 'usb2', 'usb3' do not match any of the regexes: '(^led-[0-9a-f]$|led)', 'pinctrl-[0-9]+'
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20221125144128.477059-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
The conditions reference the symbol SBI_V01, which does not exist. The
correct symbol is RISCV_SBI_V01.
Fixes: e623715f3d ("RISC-V: Increase range and default value of NR_CPUS")
Signed-off-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
Reviewed-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20221126061557.3541-1-samuel@sholland.org
Signed-off-by: Palmer Dabbelt <palmer@rivosinc.com>
- Fix osnoise duration type to 64bit not 32bit.
- Have histogram triggers be able to handle an unexpected NULL pointer
for the record event, that can happen when the histogram first starts up.
- Clear out ring buffers when dynamic events are removed, as the type
that is saved in the ring buffer is used to read the event, and a
stale type that is reused by another event could cause use after free
issues.
- Trivial comment fix.
- Fix memory leak in user_event_create().
-----BEGIN PGP SIGNATURE-----
iIoEABYIADIWIQRRSw7ePDh/lE+zeZMp5XQQmuv6qgUCY4UPShQccm9zdGVkdEBn
b29kbWlzLm9yZwAKCRAp5XQQmuv6qo3qAQCz2R814gCo54WEDcvRUmAmF9BSE2b5
wtaDfeyXQYbaMwEA4+cuNoMMlQvbVbmcss/Uh7RpRCB8+24iypeiEZNtIgg=
=GPrG
-----END PGP SIGNATURE-----
Merge tag 'trace-v6.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace
Pull tracing fixes from Steven Rostedt:
- Fix osnoise duration type to 64bit not 32bit
- Have histogram triggers be able to handle an unexpected NULL pointer
for the record event, which can happen when the histogram first
starts up
- Clear out ring buffers when dynamic events are removed, as the type
that is saved in the ring buffer is used to read the event, and a
stale type that is reused by another event could cause use after free
issues
- Trivial comment fix
- Fix memory leak in user_event_create()
* tag 'trace-v6.1-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing: Free buffers when a used dynamic event is removed
tracing: Add tracing_reset_all_online_cpus_unlocked() function
tracing: Fix race where histograms can be called before the event
tracing/osnoise: Fix duration type
tracing/user_events: Fix memory leak in user_event_create()
tracing/hist: add in missing * in comment blocks
With the changes to scan the page cache for dirty data to avoid data
corruptions from partial write cleanup racing with other page cache
operations, the drop writes error injection no longer works the same
way it used to and causes xfs/196 to fail. This is because xfs/196
writes to the file and populates the page cache before it turns on
the error injection and starts failing -overwrites-.
The result is that the original drop-writes code failed writes only
-after- overwriting the data in the cache, followed by invalidates
the cached data, then punching out the delalloc extent from under
that data.
On the surface, this looks fine. The problem is that page cache
invalidation *doesn't guarantee that it removes anything from the
page cache* and it doesn't change the dirty state of the folio. When
block size == page size and we do page aligned IO (as xfs/196 does)
everything happens to align perfectly and page cache invalidation
removes the single page folios that span the written data. Hence the
followup delalloc punch pass does not find cached data over that
range and it can punch the extent out.
IOWs, xfs/196 "works" for block size == page size with the new
code. I say "works", because it actually only works for the case
where IO is page aligned, and no data was read from disk before
writes occur. Because the moment we actually read data first, the
readahead code allocates multipage folios and suddenly the
invalidate code goes back to zeroing subfolio ranges without
changing dirty state.
Hence, with multipage folios in play, block size == page size is
functionally identical to block size < page size behaviour, and
drop-writes is manifestly broken w.r.t to this case. Invalidation of
a subfolio range doesn't result in the folio being removed from the
cache, just the range gets zeroed. Hence after we've sequentially
walked over a folio that we've dirtied (via write data) and then
invalidated, we end up with a dirty folio full of zeroed data.
And because the new code skips punching ranges that have dirty
folios covering them, we end up leaving the delalloc range intact
after failing all the writes. Hence failed writes now end up
writing zeroes to disk in the cases where invalidation zeroes folios
rather than removing them from cache.
This is a fundamental change of behaviour that is needed to avoid
the data corruption vectors that exist in the old write fail path,
and it renders the drop-writes injection non-functional and
unworkable as it stands.
As it is, I think the error injection is also now unnecessary, as
partial writes that need delalloc extent are going to be a lot more
common with stale iomap detection in place. Hence this patch removes
the drop-writes error injection completely. xfs/196 can remain for
testing kernels that don't have this data corruption fix, but those
that do will report:
xfs/196 3s ... [not run] XFS error injection drop_writes unknown on this kernel.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Now that iomap supports a mechanism to validate cached iomaps for
buffered write operations, hook it up to the XFS buffered write ops
so that we can avoid data corruptions that result from stale cached
iomaps. See:
https://lore.kernel.org/linux-xfs/20220817093627.GZ3600936@dread.disaster.area/
or the ->iomap_valid() introduction commit for exact details of the
corruption vector.
The validity cookie we store in the iomap is based on the type of
iomap we return. It is expected that the iomap->flags we set in
xfs_bmbt_to_iomap() is not perturbed by the iomap core and are
returned to us in the iomap passed via the .iomap_valid() callback.
This ensures that the validity cookie is always checking the correct
inode fork sequence numbers to detect potential changes that affect
the extent cached by the iomap.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
A recent multithreaded write data corruption has been uncovered in
the iomap write code. The core of the problem is partial folio
writes can be flushed to disk while a new racing write can map it
and fill the rest of the page:
writeback new write
allocate blocks
blocks are unwritten
submit IO
.....
map blocks
iomap indicates UNWRITTEN range
loop {
lock folio
copyin data
.....
IO completes
runs unwritten extent conv
blocks are marked written
<iomap now stale>
get next folio
}
Now add memory pressure such that memory reclaim evicts the
partially written folio that has already been written to disk.
When the new write finally gets to the last partial page of the new
write, it does not find it in cache, so it instantiates a new page,
sees the iomap is unwritten, and zeros the part of the page that
it does not have data from. This overwrites the data on disk that
was originally written.
The full description of the corruption mechanism can be found here:
https://lore.kernel.org/linux-xfs/20220817093627.GZ3600936@dread.disaster.area/
To solve this problem, we need to check whether the iomap is still
valid after we lock each folio during the write. We have to do it
after we lock the page so that we don't end up with state changes
occurring while we wait for the folio to be locked.
Hence we need a mechanism to be able to check that the cached iomap
is still valid (similar to what we already do in buffered
writeback), and we need a way for ->begin_write to back out and
tell the high level iomap iterator that we need to remap the
remaining write range.
The iomap needs to grow some storage for the validity cookie that
the filesystem provides to travel with the iomap. XFS, in
particular, also needs to know some more information about what the
iomap maps (attribute extents rather than file data extents) to for
the validity cookie to cover all the types of iomaps we might need
to validate.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
All the callers of xfs_bmap_punch_delalloc_range() jump through
hoops to convert a byte range to filesystem blocks before calling
xfs_bmap_punch_delalloc_range(). Instead, pass the byte range to
xfs_bmap_punch_delalloc_range() and have it do the conversion to
filesystem blocks internally.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
iomap_file_buffered_write_punch_delalloc() currently invalidates the
page cache over the unused range of the delalloc extent that was
allocated. While the write allocated the delalloc extent, it does
not own it exclusively as the write does not hold any locks that
prevent either writeback or mmap page faults from changing the state
of either the page cache or the extent state backing this range.
Whilst xfs_bmap_punch_delalloc_range() already handles races in
extent conversion - it will only punch out delalloc extents and it
ignores any other type of extent - the page cache truncate does not
discriminate between data written by this write or some other task.
As a result, truncating the page cache can result in data corruption
if the write races with mmap modifications to the file over the same
range.
generic/346 exercises this workload, and if we randomly fail writes
(as will happen when iomap gets stale iomap detection later in the
patchset), it will randomly corrupt the file data because it removes
data written by mmap() in the same page as the write() that failed.
Hence we do not want to punch out the page cache over the range of
the extent we failed to write to - what we actually need to do is
detect the ranges that have dirty data in cache over them and *not
punch them out*.
To do this, we have to walk the page cache over the range of the
delalloc extent we want to remove. This is made complex by the fact
we have to handle partially up-to-date folios correctly and this can
happen even when the FSB size == PAGE_SIZE because we now support
multi-page folios in the page cache.
Because we are only interested in discovering the edges of data
ranges in the page cache (i.e. hole-data boundaries) we can make use
of mapping_seek_hole_data() to find those transitions in the page
cache. As we hold the invalidate_lock, we know that the boundaries
are not going to change while we walk the range. This interface is
also byte-based and is sub-page block aware, so we can find the data
ranges in the cache based on byte offsets rather than page, folio or
fs block sized chunks. This greatly simplifies the logic of finding
dirty cached ranges in the page cache.
Once we've identified a range that contains cached data, we can then
iterate the range folio by folio. This allows us to determine if the
data is dirty and hence perform the correct delalloc extent punching
operations. The seek interface we use to iterate data ranges will
give us sub-folio start/end granularity, so we may end up looking up
the same folio multiple times as the seek interface iterates across
each discontiguous data region in the folio.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Commit d2825fa936 ("crypto: sm3,sm4 - move into crypto directory") moves
the SM3 and SM4 stand-alone library and the algorithm implementation for
the Crypto API into the same directory, and the corresponding relationship
of Kconfig is modified, CONFIG_CRYPTO_SM3/4 corresponds to the stand-alone
library of SM3/4, and CONFIG_CRYPTO_SM3/4_GENERIC corresponds to the
algorithm implementation for the Crypto API. Therefore, it is necessary
for this module to depend on the correct algorithm.
Fixes: d2825fa936 ("crypto: sm3,sm4 - move into crypto directory")
Cc: Jason A. Donenfeld <Jason@zx2c4.com>
Cc: stable@vger.kernel.org # v5.19+
Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Signed-off-by: Mimi Zohar <zohar@linux.ibm.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQSQHSd0lITzzeNWNm3h3BK/laaZPAUCY4SQxQAKCRDh3BK/laaZ
PAWSAP9WHL4ejQtVu2NMaEhZyIxs3weXLrFMPcQqOJ5JZhrgGAD/d6JufR/4jKWK
Sf/VLPsDlXsvPyCMJOSZAsQ5Bt1reA4=
=57DJ
-----END PGP SIGNATURE-----
Merge tag 'fuse-fixes-6.1-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse
Pull fuse fix from Miklos Szeredi:
"Fix a regression introduced in -rc4"
* tag 'fuse-fixes-6.1-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
fuse: lock inode unconditionally in fuse_fallocate()
If user_event_trace_register() fails within user_event_parse() the
call's print_fmt member is not freed. Add kfree call to fix this.
Link: https://lkml.kernel.org/r/20221123183248.554-1-beaub@linux.microsoft.com
Fixes: aa3b2b4c66 ("user_events: Add print_fmt generation support for basic types")
Signed-off-by: Beau Belgrave <beaub@linux.microsoft.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Through this node, you can control the background discard
to run more aggressively or not aggressively when reach the
utilization rate of the space.
Signed-off-by: Yangtao Li <frank.li@vivo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Do cleanup in f2fs_tuning_parameters() and __init_discard_policy(),
let's use macro instead of number.
Suggested-by: Chao Yu <chao@kernel.org>
Signed-off-by: Yangtao Li <frank.li@vivo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Under the current logic, after the discard thread wakes up, it will not
run according to the expected policy, but will use the expected policy
before sleep. Move the strategy selection to after the thread wakes up,
so that the running state of the thread meets expectations.
Signed-off-by: Yangtao Li <frank.li@vivo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
When f2fs chooses GC victim in large section & LFS mode,
next_victim_seg[gc_type] is referenced first. After segment is freed,
next_victim_seg[gc_type] has the next segment number.
However, next_victim_seg[gc_type] still has the last segment number
even after the last segment of section is freed. In this case, when f2fs
chooses a victim for the next GC round, the last segment of previous victim
section is chosen as a victim.
Initialize next_victim_seg[gc_type] to NULL_SEGNO for the last segment in
large section.
Fixes: e3080b0120 ("f2fs: support subsectional garbage collection")
Signed-off-by: Yonggil Song <yonggil.song@samsung.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Use f2fs_do_truncate_blocks() to truncate all blocks in-batch in
__complete_revoke_list().
Signed-off-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Since __queue_discard_cmd() never returns an error,
let's make it return void.
Signed-off-by: Yangtao Li <frank.li@vivo.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
