The only difference between iwl_pcie_napi_poll_msix_shared() and
iwl_pcie_napi_poll_msix() is when we have a shared queue and nothing
in the rx queue. This case doesn't affect CPU performance, so we can
merge the two functions.
Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210411124417.9d1b61ef53a5.I60b33d5379cf7c12f1de30fc3fd4cefc38220141@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
We cannot lock a mutex while we're in an RCU critical section. At
the same time, we're accessing data structures that are protected
by the mvm->mutex anyway, so just move the entire locking here to
use only that.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210411124417.4d27bd36e10e.I1fd8e8fe442c41a5deaa560452b598ed7a60ada5@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Power save (PS) should only be enabled when we reach the max phy rate.
Before we reach it (MCS_9) for VHT, we should keep trying to improve the
throughput.
Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210411124417.2a2fb9f9c25e.I7c7bbcfbdc1d35d2c3338778fb397dd5b08ea0e8@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
umac_error_table In TLV address was read using mask on MSB
but on the same table in alive message it was without which
caused mismatch in devices with different memory region MSB
Signed-off-by: Roee Goldfiner <roee.h.goldfiner@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210411124417.179fee442115.Ib6eabe86cfda0b6044f07c07448c366b6e07e53d@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
In case the device is stopped any usage of hw queues needs to be
reallocated in fw due to fw reset after device stop, so all driver
internal queue should also be freed, and if we don't free the next usage
would leak the old memory and get in recover flows
"iwlwifi 0000:00:03.0: dma_pool_destroy iwlwifi:bc" warning.
Also warn about trying to reuse an internal allocated queue.
Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210411124417.c72d2f0355c4.Ia3baff633b9b9109f88ab379ef0303aa152c16bf@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
If the firmware is dead, the Tx queues won't drain, but
leaving a print in the log is enough, no need to WARN.
If the firmware is dead, we must already have printed enough
information in the log anyway.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210411124417.9a438b2320a9.I4aa897178df82acefd80173d76dd6849ad1bcdc1@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Debug message "firmware didn't ACK the reset - continue anyway\n"
in fw_reset_handshake() is classified as error, however this is not
an error as it is ignored. So, change it to info message for proper
classification of debug messages.
Signed-off-by: Ravi Darsi <ravi.darsi@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210331121101.449b3092c330.I515edcc41913ca7fbe4a4de923671d120d5618c6@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Instead of implementing the DSM evaluation function in the MVM code,
refactor it so it can be generalized and part of the common ACPI
implementation.
Signed-off-by: Harish Mitty <harish.mitty@intel.com>
[reworded subject and commit message]
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210331121101.a24af3551aac.I8e6bd5eb05f853b6331fa4823750f7ba8ffe46e6@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Read the new BIOS DSM and Pass to FW if to disable\enable
11ax for Ukraine according to the BIOS key. this is
needed to enable OEMs to control enable/disable 11ax in Ukraine
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210331121101.055654e7707e.If7eaf9839cd5d59729a235df07e31668b408e740@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Add support for ppag in China by reading revision 2 of the ppag table
from ACPI, and passing the data to the FW.
This is needed to enable OEMs to control ppag enablement
in China.
Signed-off-by: Miri Korenblit <miriam.rachel.korenblit@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210331121101.69af388d0dce.I8cfddf9e6837bf394b00390181b4b774ded19acd@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
When doing scan while 6GHz channels are not enabled, the 6GHz band
is not scanned. Thus, if there are no APs on the 2GHz and 5GHz bands
(that will allow discovery of geographic location etc. that would
allow enabling the 6GHz channels) but there are non collocated APs
on 6GHz PSC channels these would never be discovered.
To overcome this, FW added support for performing passive UHB scan
in case no APs were discovered during scan on the 2GHz and 5GHz
channels.
Add support for enabling such scan when the following conditions are
met:
- 6GHz channels are supported but not enabled by regulatory.
- Station interface is not associated or less than a defined time
interval passed from the last resume or HW reset flows.
- At least 4 channels are included in the scan request
- The scan request includes the widlcard SSID.
- At least 50 minutes passed from the last 6GHz passive scan.
Signed-off-by: Ilan Peer <ilan.peer@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210331121101.7c7bd00e0aeb.Ib226ad57e416b43a710c36a78a617d4243458b99@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
When we do queue sync, it's confusing that we have the structures
declared in the FW API header files that aren't really firmware,
and the union is also confusing - especially now in the code that
checks the size on the return.
So rework this: change the type of sync and whether to do it in a
synchronous fashion to arguments, and build the data structure in
the function, so we don't need the union.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210331121101.f62833fd9893.I612d7ac1c655ec4880329360e15d207698c750bc@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
We use mvm->queue_sync_state to wait for synchronous queue sync
messages, but if an async one happens inbetween we shouldn't
clear mvm->queue_sync_state after sending the async one, that
can run concurrently (at least from the CPU POV) with another
synchronous queue sync.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210331121101.d11c9bcdb4aa.I0772171dbaec87433a11513e9586d98b5d920b5f@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
If we (for example) have a trans_cfg entry in the PCI IDs table,
but then don't find a full cfg entry for it in the info table,
we fall through to the code that treats the PCI ID table entry
as a full cfg entry. This obviously causes crashes later, e.g.
when trying to build the firmware name string.
Avoid such crashes by using the low bit of the pointer as a tag
for trans_cfg entries (automatically using a macro that checks
the type when assigning) and then checking that before trying to
use the data as a full entry - if it's just a partial entry at
that point, fail.
Since we're adding some macro magic, also check that the type is
in fact either struct iwl_cfg_trans_params or struct iwl_cfg,
failing compilation ("initializer element is not constant") if
it isn't.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210330162204.6f69fe6e4128.I921d4ae20ef5276716baeeeda0b001cf25b9b968@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
For simplicity we assume that msix has 2 IRQ lines one used for rx data
called msix_non_share, and another used for one bit flags messages
(alive, hw error, sw error, rx data flag) called msix_share.
Every time the FW has data to send it puts it on the RX queue and HW
turns on the flags in msix_share (inta_fw) indicating about rx data,
and HW sends an interrupt a bit later to the msix_non_share _unless_
the msix_shared RX data bit was cleared.
Currently in the code every time we get an msix_shared we clear all bits
including rx data queue bits.
So we can have a race
----------------------------------------------------
DRIVER | HW | FW
----------------------------------------------------
- send host cmd to FW | |
| | - handle message
| | and put a response
| | on the RX queue
| - RX flag on |
| | - send alive msix
| - alive flag on |
| - interrupt |
| msix_share driver |
- handle msix_shared | |
and clear all flags | |
bits | |
| - don't send an |
| interrupt on |
| msix_non_shared |
| (driver cleared) |
- driver timeout on | |
waiting for host cmd | |
respond | |
| |
----------------------------------------------------
The change is to clear only the msi_shared flags that are handled in
the msix_shared flow, which will cause the hardware to send an interrupt
on the msix_non_share line as well, when it has data.
Signed-off-by: Mordechay Goodstein <mordechay.goodstein@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210330162204.a1cdda2fa270.I02a82312679f4541f30bb8db8747a797dbb70ee7@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Add support for different combinations of Bz
and CRFs.
Note: As of now we do not know the exact values
for ltr_delay and xtal_latency, so for now use the
worst case scenario values until the actual values
are clarified.
Signed-off-by: Matti Gottlieb <matti.gottlieb@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210330162204.caac8d996532.I6a22d6decb106cd50d7954b19236b69d685dcc39@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
"Fully associated" means that we heard a beacon with the DTIM
information and the firmware is configured to track the beacons.
Since the firmware needs to track the beacons for the CSA, we
can't configure the firmware for CSA before it knows when the
beacons are expected otherwise we'd get ASSERT 301D.
If we are required to start CSA before we told the firmware
when the beacons are expected to arrive, just report a
failure and let mac80211 disconnect.
Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210330162204.9adaedeb59e4.Idaad6aaf3f9759d023b8e00b10064915c0db9aa3@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
We currently have a special, separate, code path to acquire NIC
access for the in-flight host-command workaround on 7000 series
hardware. However, the normal code path here has grown a number
of additional workarounds/semantics over time, such as reprobing
the device if things fail.
Rather than try to replicate any of this logic, call the normal
grab_nic_access logic for the workaround.
This changes the spinlock to _bh, but that's OK since it's just
redundant, we already have soft-IRQs disabled when we get here,
and so didn't (have to) do it again. Since it's only for commands
there's however no point in making the code more complex just to
not use _bh here.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210330162204.d196fc6ffb23.Idc1ce3ce9fed9178beee7e5409bc669f79b06a0d@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Most devices don't set the apmg_wake_up_wa flag, so we don't do
anything for them. Avoid taking the spinlock for every command
unless the device needs this workaround.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210330162204.1ab60af3f318.I51cc202f68a2a953223e70c3e8610343412961b6@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
When moving to the new channel, we block TX until we hear the
first beacon. if it is not heard, we proceed to disconnect.
Since TX is blocked (without mac80211 being aware of it) the frame
is stuck, resulting with queue hang.
Instead, reenable TX before reporting on the connection loss.
As we are on the new channel, there is no problem with that,
even if the original CSA had quiet mode.
Signed-off-by: Sara Sharon <sara.sharon@intel.com>
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
Link: https://lore.kernel.org/r/iwlwifi.20210330162204.eb4f2ff1b863.Ib16238106b33d58b2b7688dc6297018b915ecef4@changeid
Signed-off-by: Luca Coelho <luciano.coelho@intel.com>
fix indentation of last line in if condition.
Fixes: af6afdb63f (staging: rtl8723bs: split long lines)
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Fabio Aiuto <fabioaiuto83@gmail.com>
Link: https://lore.kernel.org/r/20210414081739.2990-1-fabioaiuto83@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
For cloned connections cuse_channel_release() will be called more than
once, resulting in use after free.
Prevent device cloning for CUSE, which does not make sense at this point,
and highly unlikely to be used in real life.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
get_user_ns() is done twice (once in virtio_fs_get_tree() and once in
fuse_conn_init()), resulting in a reference leak.
Also looks better to use fsc->user_ns (which *should* be the
current_user_ns() at this point).
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
If an incoming FUSE request can't fit on the virtqueue, the request is
placed onto a workqueue so a worker can try to resubmit it later where
there will (hopefully) be space for it next time.
This is fine for requests that aren't larger than a virtqueue's maximum
capacity. However, if a request's size exceeds the maximum capacity of the
virtqueue (even if the virtqueue is empty), it will be doomed to a life of
being placed on the workqueue, removed, discovered it won't fit, and placed
on the workqueue yet again.
Furthermore, from section 2.6.5.3.1 (Driver Requirements: Indirect
Descriptors) of the virtio spec:
"A driver MUST NOT create a descriptor chain longer than the Queue
Size of the device."
To fix this, limit the number of pages FUSE will use for an overall
request. This way, each request can realistically fit on the virtqueue
when it is decomposed into a scattergather list and avoid violating section
2.6.5.3.1 of the virtio spec.
Signed-off-by: Connor Kuehl <ckuehl@redhat.com>
Reviewed-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
In fuse when a direct/write-through write happens we invalidate attrs
because that might have updated mtime/ctime on server and cached
mtime/ctime will be stale.
What about page writeback path. Looks like we don't invalidate attrs
there. To be consistent, invalidate attrs in writeback path as well. Only
exception is when writeback_cache is enabled. In that case we strust local
mtime/ctime and there is no need to invalidate attrs.
Recently users started experiencing failure of xfstests generic/080,
geneirc/215 and generic/614 on virtiofs. This happened only newer "stat"
utility and not older one. This patch fixes the issue.
So what's the root cause of the issue. Here is detailed explanation.
generic/080 test does mmap write to a file, closes the file and then checks
if mtime has been updated or not. When file is closed, it leads to
flushing of dirty pages (and that should update mtime/ctime on server).
But we did not explicitly invalidate attrs after writeback finished. Still
generic/080 passed so far and reason being that we invalidated atime in
fuse_readpages_end(). This is called in fuse_readahead() path and always
seems to trigger before mmaped write.
So after mmaped write when lstat() is called, it sees that atleast one of
the fields being asked for is invalid (atime) and that results in
generating GETATTR to server and mtime/ctime also get updated and test
passes.
But newer /usr/bin/stat seems to have moved to using statx() syscall now
(instead of using lstat()). And statx() allows it to query only ctime or
mtime (and not rest of the basic stat fields). That means when querying
for mtime, fuse_update_get_attr() sees that mtime is not invalid (only
atime is invalid). So it does not generate a new GETATTR and fill stat
with cached mtime/ctime. And that means updated mtime is not seen by
xfstest and tests start failing.
Invalidating attrs after writeback completion should solve this problem in
a generic manner.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
When posix access ACL is set, it can have an effect on file mode and it can
also need to clear SGID if.
- None of caller's group/supplementary groups match file owner group.
AND
- Caller is not priviliged (No CAP_FSETID).
As of now fuser server is responsible for changing the file mode as
well. But it does not know whether to clear SGID or not.
So add a flag FUSE_SETXATTR_ACL_KILL_SGID and send this info with SETXATTR
to let file server know that sgid needs to be cleared as well.
Reported-by: Luis Henriques <lhenriques@suse.de>
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fuse client needs to send additional information to file server when it
calls SETXATTR(system.posix_acl_access), so add extra flags field to the
structure.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
With commit f8425c9396 ("fuse: 32-bit user space ioctl compat for fuse
device") the matching constraints for the FUSE_DEV_IOC_CLONE ioctl command
are relaxed, limited to the testing of command type and number. As Arnd
noticed, this is wrong as it wouldn't ensure the correctness of the data
size or direction for the received FUSE device ioctl.
Fix by bringing back the comparison of the ioctl received by the FUSE
device to the originally generated FUSE_DEV_IOC_CLONE.
Fixes: f8425c9396 ("fuse: 32-bit user space ioctl compat for fuse device")
Reported-by: Arnd Bergmann <arnd@kernel.org>
Signed-off-by: Alessio Balsini <balsini@android.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
All callers of fuse_short_read already set the .page_zeroing flag, so no
need to do the tail zeroing again.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
There are two modes for write(2) and friends in fuse:
a) write through (update page cache, send sync WRITE request to userspace)
b) buffered write (update page cache, async writeout later)
The write through method kept all the page cache pages locked that were
used for the request. Keeping more than one page locked is deadlock prone
and Qian Cai demonstrated this with trinity fuzzing.
The reason for keeping the pages locked is that concurrent mapped reads
shouldn't try to pull possibly stale data into the page cache.
For full page writes, the easy way to fix this is to make the cached page
be the authoritative source by marking the page PG_uptodate immediately.
After this the page can be safely unlocked, since mapped/cached reads will
take the written data from the cache.
Concurrent mapped writes will now cause data in the original WRITE request
to be updated; this however doesn't cause any data inconsistency and this
scenario should be exceedingly rare anyway.
If the WRITE request returns with an error in the above case, currently the
page is not marked uptodate; this means that a concurrent read will always
read consistent data. After this patch the page is uptodate between
writing to the cache and receiving the error: there's window where a cached
read will read the wrong data. While theoretically this could be a
regression, it is unlikely to be one in practice, since this is normal for
buffered writes.
In case of a partial page write to an already uptodate page the locking is
also unnecessary, with the above caveats.
Partial write of a not uptodate page still needs to be handled. One way
would be to read the complete page before doing the write. This is not
possible, since it might break filesystems that don't expect any READ
requests when the file was opened O_WRONLY.
The other solution is to serialize the synchronous write with reads from
the partial pages. The easiest way to do this is to keep the partial pages
locked. The problem is that a write() may involve two such pages (one head
and one tail). This patch fixes it by only locking the partial tail page.
If there's a partial head page as well, then split that off as a separate
WRITE request.
Reported-by: Qian Cai <cai@lca.pw>
Link: https://lore.kernel.org/linux-fsdevel/4794a3fa3742a5e84fb0f934944204b55730829b.camel@lca.pw/
Fixes: ea9b9907b8 ("fuse: implement perform_write")
Cc: <stable@vger.kernel.org> # v2.6.26
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>