Start processing two new entry types in addition to current ones:
* Learned unicast tunnel entry
* Aged-out unicast tunnel entry
In both cases the device reports on a new {MAC, FID, IP address} tuple
that was learned / aged-out. Based on this notification, the driver
instructs the device to add / delete the entry to / from its database.
The driver also makes sure to notify the bridge and VxLAN drivers about
the new entry.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
FDB notifications for entries learned from an NVE tunnel contain the IP
address of the remote VTEP. In the case of IPv4 underlay, the IP address
is specified as-is. IPv6 addresses on the other hand, are specified as
handles which then need to be used to query the actual address from the
device.
Only IPv4 underlay is currently supported, so we cannot receive
notifications for IPv6 addresses and therefore an error is returned when
one tries to resolve such an address.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When processing a notification about a new FDB entry learned from a
VxLAN tunnel, the driver is provided with the FID index among other
parameters.
The driver potentially needs to update the bridge and VxLAN drivers
about the new entry using a pointer to the VxLAN device and the
corresponding VNI.
These two parameters are stored in the FID, so add a new function that
allows looking up a FID based on its index.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The driver periodically polls for new FDB entries learned by the device.
In the case of an FDB entry learned from a VxLAN tunnel, the
notification includes the IP of the remote VTEP, the filtering
identifier (FID) and the source MAC address of the overlay packet.
Assuming learning is enabled in the VxLAN and bridge drivers, the driver
needs to generate a notification and update them about the new FDB
entry.
Store the ifindex of the NVE device in the FID so that the driver will
be able to update the VxLAN and bridge drivers using it.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Will be used to process learned FDB records from an NVE tunnel.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow querying bridge port flags so that drivers capable of performing
VxLAN learning will update the bridge driver only if learning is enabled
on its bridge port corresponding to the VxLAN device.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In a similar fashion to the bridge device, allow changing the ageing
time of the VxLAN device by scheduling its timer to fire if the ageing
time changed.
One use case is selftests where learning / ageing of VxLAN FDB entries
is tested. The default ageing time is 5 minutes, which is too long for a
simple selftest.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to allow devices to signal learning events to VXLAN, introduce
two new switchdev messages: SWITCHDEV_VXLAN_FDB_ADD_TO_BRIDGE and
SWITCHDEV_VXLAN_FDB_DEL_TO_BRIDGE.
Listen to these notifications in the vxlan driver. The FDB entries
learned this way have an NTF_EXT_LEARNED flag, and only entries marked
as such can be unlearned by the _DEL_ event. They are also immediately
marked as offloaded. This is the same behavior that the bridge driver
observes.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When an external learning event collides with an user-added entry, the
user-added entry shouldn't be taken over. Otherwise on an unlearn event
the entry would be completely lost, even though the user added it by
hand.
Therefore skip update of FDB flags and state for these cases. This is in
accordance with the bridge behavior.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The VXLAN driver needs to differentiate between FDB entries learned by
the VXLAN driver, and those added by the user. The latter ones shouldn't
be taken over by external learning events. This is in accordance with
bridge behavior.
Therefore, extend the flags bitfield to 16 bits and add a new private
NTF flag to mark the user-added entries.
This seems preferable to adding a dedicated boolean, because passing the
flag, unlike passing e.g. a true, makes it clear what the meaning of the
bit is.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In a following patch, vxlan is extended to allow hardware FDB learning.
For FDB entries learned this way, switchdev notifications should not be
sent again, because the driver already knows about these entries.
To that end, add an argument vxlan_fdb_notify() to determine whether
the switchdev notifications should be sent. Propagate the argument to
all call sites transitively, eventually passing true in all root calls.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This argument is necessary for vxlan_fdb_delete(), the API of which is
prescribed by ndo_fdb_del, but __vxlan_fdb_delete() doesn't need it.
Therefore drop it.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In the original ftmac100_interrupt(), the interrupts are only disabled when
the condition "netif_running(netdev)" is true. However, this condition
causes kerenl hang in the following case. When the user requests to
disable the network device, kernel will clear the bit __LINK_STATE_START
from the dev->state and then call the driver's ndo_stop function. Network
device interrupts are not blocked during this process. If an interrupt
occurs between clearing __LINK_STATE_START and stopping network device,
kernel cannot disable the interrupts due to the condition
"netif_running(netdev)" in the ISR. Hence, kernel will hang due to the
continuous interruption of the network device.
In order to solve the above problem, the interrupts of the network device
should always be disabled in the ISR without being restricted by the
condition "netif_running(netdev)".
[V2]
Remove unnecessary curly braces.
Signed-off-by: Vincent Chen <vincentc@andestech.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The value of pitches is not correct while calling mode_set.
The issue we found so far on following system:
- Debian8 with XFCE Desktop
- Ubuntu with KDE Desktop
- SUSE15 with KDE Desktop
Signed-off-by: Y.C. Chen <yc_chen@aspeedtech.com>
Cc: <stable@vger.kernel.org>
Tested-by: Jean Delvare <jdelvare@suse.de>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Ursula Braun says:
====================
net/smc: fixes 2018-11-12
here is V4 of some net/smc fixes in different areas for the net tree.
v1->v2:
do not define 8-byte alignment for union smcd_cdc_cursor in
patch 4/5 "net/smc: atomic SMCD cursor handling"
v2->v3:
stay with 8-byte alignment for union smcd_cdc_cursor in
patch 4/5 "net/smc: atomic SMCD cursor handling", but get rid of
__packed for struct smcd_cdc_msg
v3->v4:
get rid of another __packed for struct smc_cdc_msg in
patch 4/5 "net/smc: atomic SMCD cursor handling"
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
In smc_wr_tx_put_slot() field pend->idx is used after being
cleared. That means always idx 0 is cleared in the wr_tx_mask.
This results in a broken administration of available WR send
payload buffers.
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Running uperf tests with SMCD on LPARs results in corrupted cursors.
SMCD cursors should be treated atomically to fix cursor corruption.
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a SMC-D link group is freed, a shutdown signal should be sent to
the peer to indicate that the link group is invalid. This patch adds the
shutdown signal to the SMC code.
Signed-off-by: Hans Wippel <hwippel@linux.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When searching for an existing link group the queue pair number is also
to be taken into consideration. When the SMC server sends a new number
in a CLC packet (keeping all other values equal) then a new link group
is to be created on the SMC client side.
Signed-off-by: Karsten Graul <kgraul@linux.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In case of a non-blocking SMC socket, the initial CLC handshake is
performed over a blocking TCP connection in a worker. If the SMC socket
is released, smc_release has to wait for the blocking CLC socket
operations (e.g., kernel_connect) inside the worker.
This patch aborts a CLC connection when the respective non-blocking SMC
socket is released to avoid waiting on socket operations or timeouts.
Signed-off-by: Hans Wippel <hwippel@linux.ibm.com>
Signed-off-by: Ursula Braun <ubraun@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
First set of fixes for 4.20, this time we have quite a few them but
all very small.
ath9k
* fix a locking regression found by a static checker
wlcore
* fix a crash which was a regression with wakeirq handling
brcm80211
* yet another fix for 160 MHz channel handling
mt76
* fix a longstaning build problem when CONFIG_LEDS_CLASS is disabled
* don't use uninitialised mutex
iwlwifi
* do note that the iwlwifi merge tag (commit 4ec321c146) seems to
contain wrong list of changes so ignore that
* fix ACPI data handling, a memory leak and other smaller fixes
ath10k
* fix a crash during suspend which was a recent regression
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
iQEcBAABAgAGBQJb9CR9AAoJEG4XJFUm622bMdAH/2z0jY38RTPqLYH7BKfIzQkB
M8GC4bi0c3NexkwS+S0xLr6A5FyTcbi2pKOqdZ2eXEoiopso5OocWXYnIw0IDA9C
ll5aI4DfIoqTIVBvYWp/Rnar67oRAfND5z3iOnu0C34KNiMbUY2WfJ6RUTo+gjZ4
Wokaf6EWOCODDtL2atBtvT96JSlM7t398aUksKzcOGX+0XPGYfyho/wQGZIYE9iQ
Exgmp9KyBBlQugu+EyGH+Dxt8J+1sSveIJXnfaCb9avtpo+C/SG7+pnJhX8ij7fH
+pYuWOq/YyliUVHVxfpK5+lVJB3ICGezbZ6xp3G3MURBqdX/E2oouFJvyKrESPc=
=je2r
-----END PGP SIGNATURE-----
Merge tag 'wireless-drivers-for-davem-2018-11-20' of git://git.kernel.org/pub/scm/linux/kernel/git/kvalo/wireless-drivers
Kalle Valo says:
====================
wireless-drivers fixes for 4.20
First set of fixes for 4.20, this time we have quite a few them but
all very small.
ath9k
* fix a locking regression found by a static checker
wlcore
* fix a crash which was a regression with wakeirq handling
brcm80211
* yet another fix for 160 MHz channel handling
mt76
* fix a longstaning build problem when CONFIG_LEDS_CLASS is disabled
* don't use uninitialised mutex
iwlwifi
* do note that the iwlwifi merge tag (commit 4ec321c146) seems to
contain wrong list of changes so ignore that
* fix ACPI data handling, a memory leak and other smaller fixes
ath10k
* fix a crash during suspend which was a recent regression
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jean-Louis reported a TCP regression and bisected to recent SACK
compression.
After a loss episode (receiver not able to keep up and dropping
packets because its backlog is full), linux TCP stack is sending
a single SACK (DUPACK).
Sender waits a full RTO timer before recovering losses.
While RFC 6675 says in section 5, "Algorithm Details",
(2) If DupAcks < DupThresh but IsLost (HighACK + 1) returns true --
indicating at least three segments have arrived above the current
cumulative acknowledgment point, which is taken to indicate loss
-- go to step (4).
...
(4) Invoke fast retransmit and enter loss recovery as follows:
there are old TCP stacks not implementing this strategy, and
still counting the dupacks before starting fast retransmit.
While these stacks probably perform poorly when receivers implement
LRO/GRO, we should be a little more gentle to them.
This patch makes sure we do not enable SACK compression unless
3 dupacks have been sent since last rcv_nxt update.
Ideally we should even rearm the timer to send one or two
more DUPACK if no more packets are coming, but that will
be work aiming for linux-4.21.
Many thanks to Jean-Louis for bisecting the issue, providing
packet captures and testing this patch.
Fixes: 5d9f4262b7 ("tcp: add SACK compression")
Reported-by: Jean-Louis Dupond <jean-louis@dupond.be>
Tested-by: Jean-Louis Dupond <jean-louis@dupond.be>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Michał Mirosław says:
====================
VLAN tag handling cleanup
This is a cleanup set after VLAN_TAG_PRESENT removal. The CFI bit
handling is made similar to how other tag fields are used.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace open-coded bitfield manipulation with skb_vlan_tag_*() helpers.
This also enables correctly passing of VLAN.CFI bit.
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Abstract CFI/DEI bit access consistently with other VLAN tag fields.
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a packet is trapped and the corresponding SKB marked as
already-forwarded, it retains this marking even after it is forwarded
across veth links into another bridge. There, since it ingresses the
bridge over veth, which doesn't have offload_fwd_mark, it triggers a
warning in nbp_switchdev_frame_mark().
Then nbp_switchdev_allowed_egress() decides not to allow egress from
this bridge through another veth, because the SKB is already marked, and
the mark (of 0) of course matches. Thus the packet is incorrectly
blocked.
Solve by resetting offload_fwd_mark() in skb_scrub_packet(). That
function is called from tunnels and also from veth, and thus catches the
cases where traffic is forwarded between bridges and transformed in a
way that invalidates the marking.
Fixes: 6bc506b4fb ("bridge: switchdev: Add forward mark support for stacked devices")
Fixes: abf4bb6b63 ("skbuff: Add the offload_mr_fwd_mark field")
Signed-off-by: Petr Machata <petrm@mellanox.com>
Suggested-by: Ido Schimmel <idosch@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Nikita V. Shirokov says:
====================
In this patch series I'm adding a helper for libbpf which would allow
it to load map-in-map(BPF_MAP_TYPE_ARRAY_OF_MAPS and BPF_MAP_TYPE_HASH_OF_MAPS).
First patch contains new helper + explains proposed workflow second patch
contains tests which also could be used as example usage.
v4->v5:
- naming: renamed everything to map_in_map instead of mapinmap
- start to return nonzero val if set_inner_map_fd failed
v3->v4:
- renamed helper to set_inner_map_fd
- now we set this value only if it haven't
been set before and only for (array|hash) of maps
v2->v3:
- fixing typo in patch description
- initializing inner_map_fd to -1 by default
v1->v2:
- addressing nits
- removing const identifier from fd in new helper
- starting to check return val for bpf_map_update_elem
====================
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
adding test/example of bpf_map__set_inner_map_fd usage
Signed-off-by: Nikita V. Shirokov <tehnerd@tehnerd.com>
Acked-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
idea is pretty simple. for specified map (pointed by struct bpf_map)
we would provide descriptor of already loaded map, which is going to be
used as a prototype for inner map. proposed workflow:
1) open bpf's object (bpf_object__open)
2) create bpf's map which is going to be used as a prototype
3) find (by name) map-in-map which you want to load and update w/
descriptor of inner map w/ a new helper from this patch
4) load bpf program w/ bpf_object__load
Signed-off-by: Nikita V. Shirokov <tehnerd@tehnerd.com>
Acked-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Use recently added capability check.
See commit 23499442c3 ("bpf: libbpf: retry map creation without
the name") for rationale.
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Instead, check for a newly created caps.name bpf_object capability.
If kernel doesn't support names, don't specify the attribute.
See commit 23499442c3 ("bpf: libbpf: retry map creation without
the name") for rationale.
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
It currently only checks whether kernel supports map/prog names.
This capability check will be used in the next two commits to
skip setting prog/map names.
Suggested-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Wrap headers in extern "C", to turn off C++ mangling.
This simplifies including libbpf in c++ and linking against it.
v2 changes:
* do the same for btf.h
v3 changes:
* test_libbpf.cpp to test for possible future c++ breakages
Signed-off-by: Stanislav Fomichev <sdf@google.com>
Acked-by: Alexei Starovoitov <ast@kernel.org>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Commit 2993e0515b ("tools/bpf: add support to read .BTF.ext sections")
added support to read .BTF.ext sections from an object file, create
and pass prog_btf_fd and func_info to the kernel.
The program btf_fd (prog->btf_fd) is initialized to be -1 to please
zclose so we do not need special handling dur prog close.
Passing -1 to the kernel, however, will cause loading error.
Passing btf_fd 0 to the kernel if prog->btf_fd is invalid
fixed the problem.
Fixes: 2993e0515b ("tools/bpf: add support to read .BTF.ext sections")
Reported-by: Andrey Ignatov <rdna@fb.com>
Reported-by: Emre Cantimur <haydum@fb.com>
Tested-by: Andrey Ignatov <rdna@fb.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
This week is a bit bigger than I expected. That's my fault, as I missed
a few patches while I was at Plumbers last week. We have:
* A fix to a quite embarassing issue where raw_copy_to_user() was
implemented with asm_copy_from_user() (and vice versa).
* Improvements to our makefile to allow flat binaries to be generated.
* A build fix that predeclares "struct module" at the top of
<asm/module.h>, which triggers warnings later in that header.
* The addition of our own <uapi/asm/unistd> header, which is necessary
to align our stat ABI on 32-bit systems.
* A fix to avoid printing a warning when the S or U bits are set in
print_isa().
I already have one patch in the queue for next week.
-----BEGIN PGP SIGNATURE-----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=xGZd
-----END PGP SIGNATURE-----
Merge tag 'riscv-for-linus-4.20-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux
Pull RISC-V fixes from Palmer Dabbelt:
"This week is a bit bigger than I expected. That's my fault, as I
missed a few patches while I was at Plumbers last week. We have:
- A fix to a quite embarassing issue where raw_copy_to_user() was
implemented with asm_copy_from_user() (and vice versa).
- Improvements to our makefile to allow flat binaries to be
generated.
- A build fix that predeclares "struct module" at the top of
<asm/module.h>, which triggers warnings later in that header.
- The addition of our own <uapi/asm/unistd> header, which is
necessary to align our stat ABI on 32-bit systems.
- A fix to avoid printing a warning when the S or U bits are set in
print_isa().
I already have one patch in the queue for next week"
* tag 'riscv-for-linus-4.20-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux:
RISC-V: recognize S/U mode bits in print_isa
riscv: add asm/unistd.h UAPI header
riscv: fix warning in arch/riscv/include/asm/module.h
RISC-V: Build flat and compressed kernel images
RISC-V: Fix raw_copy_{to,from}_user()
Address community comment.
Remove obsolete IGC_ERR define and use dev_err method.
Suggested by Joe Perches.
Signed-off-by: Sasha Neftin <sasha.neftin@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Now that synchronize_rcu() waits for preempt-disable regions of code
as well as RCU read-side critical sections, synchronize_sched() can be
replaced by synchronize_rcu(). This commit therefore makes this change.
Signed-off-by: "Paul E. McKenney" <paulmck@linux.ibm.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
While reviewing code, I noticed that Eric Dumazet recommends that
drivers check the return code of napi_complete_done, and use that
to decide to enable interrupts or not when exiting poll. One of
the Intel drivers was already fixed (ixgbe).
Upon looking at the Intel drivers as a whole, we are handling our
polling and NAPI exit in a few different ways based on whether we
have multiqueue and whether we have Tx cleanup included. Several
drivers had the bug of exiting NAPI with return 0, which appears
to mess up the accounting in the stack.
Consolidate all the NAPI routines to do best known way of exiting
and to just mostly look like each other.
1) check return code of napi_complete_done to control interrupt enable
2) return the actual amount of work done.
3) return budget immediately if need NAPI poll again
Tested the changes on e1000e with a high interrupt rate set, and
it shows about an 8% reduction in the CPU utilization when busy
polling because we aren't re-enabling interrupts when we're about
to be polled.
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
The #define for NETIF_F_GSO_UDP_L4 was incorrect in the
documentation, fix it by making it match the actual code.
Signed-off-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
The line continuation unintentionally adds whitespace so
instead use a coalesced format to remove the whitespace.
Miscellanea:
o Use a more typical style for ternaries and arguments
for this logging message
Signed-off-by: Joe Perches <joe@perches.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Acked-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
When we read the EOF page of the file via readpages, we need
to zero the region beyond EOF that we either do not read or
should not contain data so that mmap does not expose stale data to
user applications.
However, iomap_adjust_read_range() fails to detect EOF correctly,
and so fsx on 1k block size filesystems fails very quickly with
mapreads exposing data beyond EOF. There are two problems here.
Firstly, when calculating the end block of the EOF byte, we have
to round the size by one to avoid a block aligned EOF from reporting
a block too large. i.e. a size of 1024 bytes is 1 block, which in
index terms is block 0. Therefore we have to calculate the end block
from (isize - 1), not isize.
The second bug is determining if the current page spans EOF, and so
whether we need split it into two half, one for the IO, and the
other for zeroing. Unfortunately, the code that checks whether
we should split the block doesn't actually check if we span EOF, it
just checks if the read spans the /offset in the page/ that EOF
sits on. So it splits every read into two if EOF is not page
aligned, regardless of whether we are reading the EOF block or not.
Hence we need to restrict the "does the read span EOF" check to
just the page that spans EOF, not every page we read.
This patch results in correct EOF detection through readpages:
xfs_vm_readpages: dev 259:0 ino 0x43 nr_pages 24
xfs_iomap_found: dev 259:0 ino 0x43 size 0x66c00 offset 0x4f000 count 98304 type hole startoff 0x13c startblock 1368 blockcount 0x4
iomap_readpage_actor: orig pos 323584 pos 323584, length 4096, poff 0 plen 4096, isize 420864
xfs_iomap_found: dev 259:0 ino 0x43 size 0x66c00 offset 0x50000 count 94208 type hole startoff 0x140 startblock 1497 blockcount 0x5c
iomap_readpage_actor: orig pos 327680 pos 327680, length 94208, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 331776 pos 331776, length 90112, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 335872 pos 335872, length 86016, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 339968 pos 339968, length 81920, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 344064 pos 344064, length 77824, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 348160 pos 348160, length 73728, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 352256 pos 352256, length 69632, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 356352 pos 356352, length 65536, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 360448 pos 360448, length 61440, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 364544 pos 364544, length 57344, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 368640 pos 368640, length 53248, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 372736 pos 372736, length 49152, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 376832 pos 376832, length 45056, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 380928 pos 380928, length 40960, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 385024 pos 385024, length 36864, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 389120 pos 389120, length 32768, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 393216 pos 393216, length 28672, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 397312 pos 397312, length 24576, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 401408 pos 401408, length 20480, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 405504 pos 405504, length 16384, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 409600 pos 409600, length 12288, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 413696 pos 413696, length 8192, poff 0 plen 4096, isize 420864
iomap_readpage_actor: orig pos 417792 pos 417792, length 4096, poff 0 plen 3072, isize 420864
iomap_readpage_actor: orig pos 420864 pos 420864, length 1024, poff 3072 plen 1024, isize 420864
As you can see, it now does full page reads until the last one which
is split correctly at the block aligned EOF, reading 3072 bytes and
zeroing the last 1024 bytes. The original version of the patch got
this right, but it got another case wrong.
The EOF detection crossing really needs to the the original length
as plen, while it starts at the end of the block, will be shortened
as up-to-date blocks are found on the page. This means "orig_pos +
plen" no longer points to the end of the page, and so will not
correctly detect EOF crossing. Hence we have to use the length
passed in to detect this partial page case:
xfs_filemap_fault: dev 259:1 ino 0x43 write_fault 0
xfs_vm_readpage: dev 259:1 ino 0x43 nr_pages 1
xfs_iomap_found: dev 259:1 ino 0x43 size 0x2cc00 offset 0x2c000 count 4096 type hole startoff 0xb0 startblock 282 blockcount 0x4
iomap_readpage_actor: orig pos 180224 pos 181248, length 4096, poff 1024 plen 2048, isize 183296
xfs_iomap_found: dev 259:1 ino 0x43 size 0x2cc00 offset 0x2cc00 count 1024 type hole startoff 0xb3 startblock 285 blockcount 0x1
iomap_readpage_actor: orig pos 183296 pos 183296, length 1024, poff 3072 plen 1024, isize 183296
Heere we see a trace where the first block on the EOF page is up to
date, hence poff = 1024 bytes. The offset into the page of EOF is
3072, so the range we want to read is 1024 - 3071, and the range we
want to zero is 3072 - 4095. You can see this is split correctly
now.
This fixes the stale data beyond EOF problem that fsx quickly
uncovers on 1k block size filesystems.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
It returns EINVAL when the operation is not supported by the
filesystem. Fix it to return EOPNOTSUPP to be consistent with
the man page and clone_file_range().
Clean up the inconsistent error return handling while I'm there.
(I know, lipstick on a pig, but every little bit helps...)
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
When doing direct IO to a pipe for do_splice_direct(), then pipe is
trivial to fill up and overflow as it can only hold 16 pages. At
this point bio_iov_iter_get_pages() then returns -EFAULT, and we
abort the IO submission process. Unfortunately, iomap_dio_rw()
propagates the error back up the stack.
The error is converted from the EFAULT to EAGAIN in
generic_file_splice_read() to tell the splice layers that the pipe
is full. do_splice_direct() completely fails to handle EAGAIN errors
(it aborts on error) and returns EAGAIN to the caller.
copy_file_write() then completely fails to handle EAGAIN as well,
and so returns EAGAIN to userspace, having failed to copy the data
it was asked to.
Avoid this whole steaming pile of fail by having iomap_dio_rw()
silently swallow EFAULT errors and so do short reads.
To make matters worse, iomap_dio_actor() has a stale data exposure
bug bio_iov_iter_get_pages() fails - it does not zero the tail block
that it may have been left uncovered by partial IO. Fix the error
handling case to drop to the sub-block zeroing rather than
immmediately returning the -EFAULT error.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
If we are doing sub-block dio that extends EOF, we need to zero
the unused tail of the block to initialise the data in it it. If we
do not zero the tail of the block, then an immediate mmap read of
the EOF block will expose stale data beyond EOF to userspace. Found
with fsx running sub-block DIO sizes vs MAPREAD/MAPWRITE operations.
Fix this by detecting if the end of the DIO write is beyond EOF
and zeroing the tail if necessary.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
When we write into an unwritten extent via direct IO, we dirty
metadata on IO completion to convert the unwritten extent to
written. However, when we do the FUA optimisation checks, the inode
may be clean and so we issue a FUA write into the unwritten extent.
This means we then bypass the generic_write_sync() call after
unwritten extent conversion has ben done and we don't force the
modified metadata to stable storage.
This violates O_DSYNC semantics. The window of exposure is a single
IO, as the next DIO write will see the inode has dirty metadata and
hence will not use the FUA optimisation. Calling
generic_write_sync() after completion of the second IO will also
sync the first write and it's metadata.
Fix this by avoiding the FUA optimisation when writing to unwritten
extents.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Long saga. There have been days spent following this through dead end
after dead end in multi-GB event traces. This morning, after writing
a trace-cmd wrapper that enabled me to be more selective about XFS
trace points, I discovered that I could get just enough essential
tracepoints enabled that there was a 50:50 chance the fsx config
would fail at ~115k ops. If it didn't fail at op 115547, I stopped
fsx at op 115548 anyway.
That gave me two traces - one where the problem manifested, and one
where it didn't. After refining the traces to have the necessary
information, I found that in the failing case there was a real
extent in the COW fork compared to an unwritten extent in the
working case.
Walking back through the two traces to the point where the CWO fork
extents actually diverged, I found that the bad case had an extra
unwritten extent in it. This is likely because the bug it led me to
had triggered multiple times in those 115k ops, leaving stray
COW extents around. What I saw was a COW delalloc conversion to an
unwritten extent (as they should always be through
xfs_iomap_write_allocate()) resulted in a /written extent/:
xfs_writepage: dev 259:0 ino 0x83 pgoff 0x17000 size 0x79a00 offset 0 length 0
xfs_iext_remove: dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/2 offset 32 block 152 count 20 flag 1 caller xfs_bmap_add_extent_delay_real
xfs_bmap_pre_update: dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/1 offset 1 block 4503599627239429 count 31 flag 0 caller xfs_bmap_add_extent_delay_real
xfs_bmap_post_update: dev 259:0 ino 0x83 state RC|LF|RF|COW cur 0xffff888247b899c0/1 offset 1 block 121 count 51 flag 0 caller xfs_bmap_add_ex
Basically, Cow fork before:
0 1 32 52
+H+DDDDDDDDDDDD+UUUUUUUUUUU+
PREV RIGHT
COW delalloc conversion allocates:
1 32
+uuuuuuuuuuuu+
NEW
And the result according to the xfs_bmap_post_update trace was:
0 1 32 52
+H+wwwwwwwwwwwwwwwwwwwwwwww+
PREV
Which is clearly wrong - it should be a merged unwritten extent,
not an unwritten extent.
That lead me to look at the LEFT_FILLING|RIGHT_FILLING|RIGHT_CONTIG
case in xfs_bmap_add_extent_delay_real(), and sure enough, there's
the bug.
It takes the old delalloc extent (PREV) and adds the length of the
RIGHT extent to it, takes the start block from NEW, removes the
RIGHT extent and then updates PREV with the new extent.
What it fails to do is update PREV.br_state. For delalloc, this is
always XFS_EXT_NORM, while in this case we are converting the
delayed allocation to unwritten, so it needs to be updated to
XFS_EXT_UNWRITTEN. This LF|RF|RC case does not do this, and so
the resultant extent is always written.
And that's the bug I've been chasing for a week - a bmap btree bug,
not a reflink/dedupe/copy_file_range bug, but a BMBT bug introduced
with the recent in core extent tree scalability enhancements.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
On a sub-page block size filesystem, fsx is failing with a data
corruption after a series of operations involving copying a file
with the destination offset beyond EOF of the destination of the file:
8093(157 mod 256): TRUNCATE DOWN from 0x7a120 to 0x50000 ******WWWW
8094(158 mod 256): INSERT 0x25000 thru 0x25fff (0x1000 bytes)
8095(159 mod 256): COPY 0x18000 thru 0x1afff (0x3000 bytes) to 0x2f400
8096(160 mod 256): WRITE 0x5da00 thru 0x651ff (0x7800 bytes) HOLE
8097(161 mod 256): COPY 0x2000 thru 0x5fff (0x4000 bytes) to 0x6fc00
The second copy here is beyond EOF, and it is to sub-page (4k) but
block aligned (1k) offset. The clone runs the EOF zeroing, landing
in a pre-existing post-eof delalloc extent. This zeroes the post-eof
extents in the page cache just fine, dirtying the pages correctly.
The problem is that xfs_reflink_remap_prep() now truncates the page
cache over the range that it is copying it to, and rounds that down
to cover the entire start page. This removes the dirty page over the
delalloc extent from the page cache without having written it back.
Hence later, when the page cache is flushed, the page at offset
0x6f000 has not been written back and hence exposes stale data,
which fsx trips over less than 10 operations later.
Fix this by changing xfs_reflink_remap_prep() to use
xfs_flush_unmap_range().
Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
