Use mgmt_txrx_frame_hex_dump to dump mgmt frame and optimize
frame dump and logs
Change-Id: I56f244fecfa2602c6b763f5734d36199b8c3f165
CRs-Fixed: 3838935
IO config can be modified due to access to shared memory.
This change scopes the data locally so as to avoid
vulnerability of count being modified by external
means while executing due to being in shared memory.
CRs-Fixed: 3777635
Change-Id: Ia5dd9138dcf8449e2d800aca9ffed73d9c4ba3ea
Signed-off-by: Akash Puliyadi Jegannathan <quic_apuliyad@quicinc.com>
I2C cmd can be modified due to access to shared memory.
This change scopes the data locally so as to avoid
vulnerability of count being modified by external
means while executing due to being in shared memory.
CRs-Fixed: 3777534
Change-Id: I4637f49db67d1bd1d5ca418435e3627b5652f604
Signed-off-by: Akash Puliyadi Jegannathan <quic_apuliyad@quicinc.com>
Enable the compilation of smmu proxy driver on Matrix.
Change-Id: I8abfeb5e34fb2a583f77773d425bee289ffcd601
Signed-off-by: Songrui Wu <quic_songrui@quicinc.com>
Currently, when monitor interface is going down, buffer ring filters
are not being reset.
To fix this, set mv_dev to NULL after filter reset.
Change-Id: I7555acd6b4a54a362e36a43a970ab1c75e3c24c8
CRs-Fixed: 3841235
When roamed from MLO to SLO, clean up vdev1 link, and disable RSO,
but RSO isn't re-enabled when disconnect completed for vdev0 link in
same MLD existed, can't roam until next reconnect.
To fix it, When roamed from MLO to SLO, clean up vdev1 link, don't
disable RSO since it's internal disconnect, no wmi like vdev stop is
sent to F/W too, vdev1 is stopped by F/W already.
Change-Id: Ib83b15352e91cb8ef73fd42bc9a5e1c6181f4ea9
CRs-Fixed: 3844460
(cherry picked from commit ecbd818bb6)
RSO stop isn't sent to F/W before link vdev stop when MLO
disconnect, F/W will assert later
To fix it, Send RSO stop to assoc vdev before link vdev stop when MLO
disconnect.
RSO stop for internal link cleanup has no side effect, only RSO disable
clears the RCL in firmware.
Change-Id: Id11da42ebebf0d9966974cc913cf6618cea0cfbb
CRs-Fixed: 3835214
(cherry picked from commit b03b971b86)
Curr_micbias2 val should be same with mb->micba2_mv.
for some customers, their micbias is not 1800mv,
so we should set the curr_micbias as mb->micb2_mv which
was the actual micbias val the customer use.
remove the dev_up check from micbias, since with this check,
Headset may be detected as a incorrect type.
Change-Id: Ia5a1fd434a09617a57096531c69956479c69f079
Signed-off-by: Yuhui Zhao <quic_yuhuzhao@quicinc.com>
Decrement the smcinvoke server object refcount in mutex context so that
we never get an object which is being freed.
Change-Id: I16c9c7f2518e8f2d7749d4556f5c7bf2ef5878af
Signed-off-by: Nishant Pandey <quic_nishpand@quicinc.com>
Color features are disabled during encoder disable, but reg dma
flush is not triggered. This change does reg dma flush for the
features to be disabled during encoder disable.
Change-Id: Ia74d4c43ad7b699f0097b49d86ad59529c0b3230
Signed-off-by: Anjaneya Prasad Musunuri <quic_aprasad@quicinc.com>
If there is only 1 AP candidate, and it has been added to denylist by F/W,
it is not allowed to connect, it's bad user experience to end users.
To fix it, if there is only 1 AP candidate, and it has been added to
denylist by F/W, but deny reason is neither critical nor certificate
required, such as REASON_STA_KICKOUT, need allow user to connect it if no
other candidates.
Change-Id: I2ae66ceba943dcfbc927568684218b7120c1e490
CRs-Fixed: 3785252
In the early stages of fastrpc_internal_invoke, we validate the user CID
and handle failure cases. However, in the error scenario, an invalid CID
can lead to issues when accessing the channel mutex. To prevent this, we
should validate the CID before accessing the channel mutex via fastrpc
user structure.
Change-Id: Ic1f7ae01a749b57c9b9e69210314d694ebcf300b
Signed-off-by: Santosh <quic_ssakore@quicinc.com>
In the early stages of fastrpc_internal_invoke, we validate the user CID
and handle failure cases. However, in the error scenario, an invalid CID
can lead to issues when accessing the channel mutex. To prevent this, we
should validate the CID before accessing the channel mutex via fastrpc
user structure.
Change-Id: Ic1f7ae01a749b57c9b9e69210314d694ebcf300b
Signed-off-by: Santosh <quic_ssakore@quicinc.com>
When roam happened in F/W and send roam stats to host driver,
host driver send these roam stats info to user space by event
without cache them.
Change-Id: I772c0a5035896715204f6eee277090ed1f33e97c
CRs-Fixed: 3790270
When partial fb is sent from userspace to dump only partial
frame roi, cwb crop feature needs to be enabled.
'commit fded005881
("disp: msm: sde: add support for CWB + single LM partial update")'
is overriding this crop ctrl enable bit. Due to this cwb xin is idle
always and wb timeouts are seen. Add changes to move the cwb setup
with false to happen early to fix overriding the crop ctrl enablement.
Change-Id: If187319d59026f43db71e6655a74bfd62acf2dc3
Signed-off-by: Akhil Jaiswal <quic_akhijais@quicinc.com>
Currently, in print_debug_data, kref_put is being called inside the
global lock, and the same lock is taken in the release callback of
kref_put, leading to spinlock recursion. There is no need to get and
put the reference for the fastrpce file inside this function because
we have already taken the reference inside the update_ramdump_status
while adding the init memory entry to the chan->initmems list.
Moreover, the same list will be used in print_debug_data.
Signed-off-by: Abhishek Singh <quic_abhishes@quicinc.com>
Change-Id: Ifdc8b3e0c2bbc5cc4237eedaa24c8cd766262dfe
(cherry picked from commit 3463a894b8)
- When the Client doesn't configure priority for a session
set Default session priority to high for SKU1 device
Change-Id: I0d02b5da4294bbb92dad299db3f5487b5126d6b1
Signed-off-by: Rajathi S <quic_rajathis@quicinc.com>
This change checks for hdr capability of connected receiver based on
vsc support of the same. If vsc support is not present, the parsing of
hdr extented block is avoided.
Change-Id: I5db5d1fe3e287123241ade10b698bb32f430ec43
Signed-off-by: Soutrik Mukhopadhyay <quic_mukhopad@quicinc.com>
When roamed from MLO to SLO, clean up vdev1 link, and disable RSO,
but RSO isn't re-enabled when disconnect completed for vdev0 link in
same MLD existed, can't roam until next reconnect.
To fix it, When roamed from MLO to SLO, clean up vdev1 link, don't
disable RSO since it's internal disconnect, no wmi like vdev stop is
sent to F/W too, vdev1 is stopped by F/W already.
Change-Id: Ib83b15352e91cb8ef73fd42bc9a5e1c6181f4ea9
CRs-Fixed: 3844460
RSO stop isn't sent to F/W before link vdev stop when MLO
disconnect, F/W will assert later
To fix it, Send RSO stop to assoc vdev before link vdev stop when MLO
disconnect.
RSO stop for internal link cleanup has no side effect, only RSO disable
clears the RCL in firmware.
Change-Id: Id11da42ebebf0d9966974cc913cf6618cea0cfbb
CRs-Fixed: 3835214
Currently the bitmap of action frames that can wake up the host
is sent via the wmi command WMI_WOW_SET_ACTION_WAKE_UP_CMDID.
This bitmap doesn't include EHT action frames so currently the
host doesn't wake up for EHT action frames.
Enable wake up for EHT action frames during runtime/system suspend.
CRs-Fixed: 3811796
Change-Id: I2c6ab7b7fba5572a02ff0db0f5198f2e9138a97e
Currently EPCS is enabled only via the vendor command hook,
add the ini configuration to enable/disable the EPCS
CRs-Fixed: 3838932
Change-Id: I1a887fa2a91dd68877d96169d39ea868ee675241
