Commit graph

982,081 commits

Author SHA1 Message Date
Mao Wenan
9060cb719e net: crypto set sk to NULL when af_alg_release.
KASAN has found use-after-free in sockfs_setattr.
The existed commit 6d8c50dcb0 ("socket: close race condition between sock_close()
and sockfs_setattr()") is to fix this simillar issue, but it seems to ignore
that crypto module forgets to set the sk to NULL after af_alg_release.

KASAN report details as below:
BUG: KASAN: use-after-free in sockfs_setattr+0x120/0x150
Write of size 4 at addr ffff88837b956128 by task syz-executor0/4186

CPU: 2 PID: 4186 Comm: syz-executor0 Not tainted xxx + #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
1.10.2-1ubuntu1 04/01/2014
Call Trace:
 dump_stack+0xca/0x13e
 print_address_description+0x79/0x330
 ? vprintk_func+0x5e/0xf0
 kasan_report+0x18a/0x2e0
 ? sockfs_setattr+0x120/0x150
 sockfs_setattr+0x120/0x150
 ? sock_register+0x2d0/0x2d0
 notify_change+0x90c/0xd40
 ? chown_common+0x2ef/0x510
 chown_common+0x2ef/0x510
 ? chmod_common+0x3b0/0x3b0
 ? __lock_is_held+0xbc/0x160
 ? __sb_start_write+0x13d/0x2b0
 ? __mnt_want_write+0x19a/0x250
 do_fchownat+0x15c/0x190
 ? __ia32_sys_chmod+0x80/0x80
 ? trace_hardirqs_on_thunk+0x1a/0x1c
 __x64_sys_fchownat+0xbf/0x160
 ? lockdep_hardirqs_on+0x39a/0x5e0
 do_syscall_64+0xc8/0x580
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x462589
Code: f7 d8 64 89 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 48 89 f8 48 89
f7 48 89 d6 48 89
ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3
48 c7 c1 bc ff ff
ff f7 d8 64 89 01 48
RSP: 002b:00007fb4b2c83c58 EFLAGS: 00000246 ORIG_RAX: 0000000000000104
RAX: ffffffffffffffda RBX: 000000000072bfa0 RCX: 0000000000462589
RDX: 0000000000000000 RSI: 00000000200000c0 RDI: 0000000000000007
RBP: 0000000000000005 R08: 0000000000001000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00007fb4b2c846bc
R13: 00000000004bc733 R14: 00000000006f5138 R15: 00000000ffffffff

Allocated by task 4185:
 kasan_kmalloc+0xa0/0xd0
 __kmalloc+0x14a/0x350
 sk_prot_alloc+0xf6/0x290
 sk_alloc+0x3d/0xc00
 af_alg_accept+0x9e/0x670
 hash_accept+0x4a3/0x650
 __sys_accept4+0x306/0x5c0
 __x64_sys_accept4+0x98/0x100
 do_syscall_64+0xc8/0x580
 entry_SYSCALL_64_after_hwframe+0x49/0xbe

Freed by task 4184:
 __kasan_slab_free+0x12e/0x180
 kfree+0xeb/0x2f0
 __sk_destruct+0x4e6/0x6a0
 sk_destruct+0x48/0x70
 __sk_free+0xa9/0x270
 sk_free+0x2a/0x30
 af_alg_release+0x5c/0x70
 __sock_release+0xd3/0x280
 sock_close+0x1a/0x20
 __fput+0x27f/0x7f0
 task_work_run+0x136/0x1b0
 exit_to_usermode_loop+0x1a7/0x1d0
 do_syscall_64+0x461/0x580
 entry_SYSCALL_64_after_hwframe+0x49/0xbe

Syzkaller reproducer:
r0 = perf_event_open(&(0x7f0000000000)={0x0, 0x70, 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0,
0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, @perf_config_ext}, 0x0, 0x0,
0xffffffffffffffff, 0x0)
r1 = socket$alg(0x26, 0x5, 0x0)
getrusage(0x0, 0x0)
bind(r1, &(0x7f00000001c0)=@alg={0x26, 'hash\x00', 0x0, 0x0,
'sha256-ssse3\x00'}, 0x80)
r2 = accept(r1, 0x0, 0x0)
r3 = accept4$unix(r2, 0x0, 0x0, 0x0)
r4 = dup3(r3, r0, 0x0)
fchownat(r4, &(0x7f00000000c0)='\x00', 0x0, 0x0, 0x1000)

Fixes: 6d8c50dcb0 ("socket: close race condition between sock_close() and sockfs_setattr()")
Signed-off-by: Mao Wenan <maowenan@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-02-18 12:01:24 -08:00
Hans Verkuil
4f14e3272f extended-controls.rst: split up per control class
The extended-controls.rst file had become too big. Split it up: each
control class reference gets its own rst file, and this file just
describes the Extended Control API.

Each control class reference is also moved up one level into the
table of contents to make it easier to find e.g. the codec control
reference.

Finally I rearranged the order so that all camera-related control
classes are grouped together, ditto for codec/jpeg and fm-rx/tx.

The ext-ctrls-codec.rst is still pretty big and it is a candidate
to split up further in the future, possibly per codec.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:55:41 -05:00
Sylwester Nawrocki
625c731d1b ARM: dts: exynos: Add support for secondary DAI to Odroid XU4
This patch extends DAPM routing and adds secondary CPU DAI entry
to support the secondary audio PCM interface on Odroid XU4.

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
2019-02-18 20:52:25 +01:00
Sylwester Nawrocki
885b005d23 ARM: dts: exynos: Add support for secondary DAI to Odroid XU3
This patch extends DAPM routing and adds secondary CPU DAI entry
to support the secondary audio PCM interface on Odroid XU3.

Signed-off-by: Sylwester Nawrocki <s.nawrocki@samsung.com>
Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
2019-02-18 20:51:58 +01:00
Hans Verkuil
c6f5c7c237 media: dvb-core: fix epoll() by calling poll_wait first
The epoll function expects that whenever the poll file op is
called, the poll_wait function is also called. That didn't
always happen in dvb_demux_poll(), dvb_dvr_poll() and
dvb_ca_en50221_io_poll(). Fix this, otherwise epoll()
can timeout when it shouldn't.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Reviewed-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:48:37 -05:00
Hans Verkuil
bb436cbeb9 media: videobuf: fix epoll() by calling poll_wait first
The epoll function expects that whenever the poll file op is
called, the poll_wait function is also called. That didn't
always happen in videobuf_poll_stream(). Fix this, otherwise
epoll() would timeout when it shouldn't.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:48:13 -05:00
Hans Verkuil
96ebc0ca6f media: v4l2-mem2mem: add q->error check to v4l2_m2m_poll()
The v4l2_m2m_poll function didn't check whether q->error
was set for either of the two queues. Add support for this.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:47:55 -05:00
Hans Verkuil
dd8695e4e1 media: v4l2-mem2mem: fix epoll() by calling poll_wait first
The epoll function expects that whenever the poll file op is
called, the poll_wait function is also called. That didn't
always happen in v4l2_m2m_poll(). Fix this, otherwise
epoll() would timeout when it shouldn't.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:47:04 -05:00
Hans Verkuil
398d768025 media: v4l2-ctrls.c: fix epoll() by calling poll_wait first
The epoll function expects that whenever the poll file op is
called, the poll_wait function is also called. That didn't
always happen in v4l2_ctrl_poll(). Fix this, otherwise epoll()
would timeout when it shouldn't.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:46:42 -05:00
Hans Verkuil
914c686864 media: vb2: fix epoll() by calling poll_wait first
The epoll function expects that whenever the poll file op is
called, the poll_wait function is also called. That didn't
always happen in vb2_core_poll() and vb2_poll(). Fix this,
otherwise epoll() would timeout when it shouldn't.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Reported-by: Yi Qingliang <niqingliang2003@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:46:20 -05:00
Hans Verkuil
f4dd471b5c media: media-request: fix epoll() by calling poll_wait first
The epoll function expects that whenever the poll file op is
called, the poll_wait function is also called. That didn't
always happen in media_request_poll(). Fix this, otherwise
epoll() would timeout when it shouldn't.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:45:33 -05:00
Hans Verkuil
b7990bcfb6 media: cec: fix epoll() by calling poll_wait first
The epoll function expects that whenever the poll file op is
called, the poll_wait function is also called. That didn't
always happen in cec_poll(). Fix this, otherwise epoll()
would timeout when it shouldn't.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Reviewed-by: Sean Young <sean@mess.org>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:44:45 -05:00
Hans Verkuil
7e4e716244 media: vb2: keep track of timestamp status
If a stream is stopped, or if a USERPTR/DMABUF buffer is queued
backed by a different user address or dmabuf fd, then the timestamp
should be skipped by vb2_find_timestamp since the memory it refers
to is no longer valid.

So keep track of a 'copied_timestamp' state: it is set when the
timestamp is copied from an output to a capture buffer, and is
cleared when it is no longer valid.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Reviewed-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:44:20 -05:00
Hans Verkuil
cfc7740835 media: vb2: replace bool by bitfield in vb2_buffer
The bool type is not recommended for use in structs, so replace these
by bitfields.

Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Reviewed-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:43:28 -05:00
Felix Fietkau
9f68847340 mt76: mt76x2: simplify per-chain signal strength handling
There is no need to use a for loop here, supported chips can only support
up to 2 chains.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 20:41:50 +01:00
David S. Miller
8bbed40f10 Merge git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nf-next
Pablo Neira Ayuso says:

====================
Netfilter/IPVS updates for net-next

The following patchset contains Netfilter/IPVS updates for you net-next
tree:

1) Missing NFTA_RULE_POSITION_ID netlink attribute validation,
   from Phil Sutter.

2) Restrict matching on tunnel metadata to rx/tx path, from wenxu.

3) Avoid indirect calls for IPV6=y, from Florian Westphal.

4) Add two indirections to prepare merger of IPV4 and IPV6 nat
   modules, from Florian Westphal.

5) Broken indentation in ctnetlink, from Colin Ian King.

6) Patches to use struct_size() from netfilter and IPVS,
   from Gustavo A. R. Silva.

7) Display kernel splat only once in case of racing to confirm
   conntrack from bridge plus nfqueue setups, from Chieh-Min Wang.

8) Skip checksum validation for layer 4 protocols that don't need it,
   patch from Alin Nastac.

9) Sparse warning due to symbol that should be static in CLUSTERIP,
   from Wei Yongjun.

10) Add new toggle to disable SDP payload translation when media
    endpoint is reachable though the same interface as the signalling
    peer, from Alin Nastac.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2019-02-18 11:38:30 -08:00
Vivek Kasireddy
98617b4567 media: imx-pxp: Start using the format VUYA32 instead of YUV32 (v2)
Buffers generated with YUV32 format seems to be incorrect, hence use
VUYA32 instead.

Changes from v1:
Add both formats VUYA32 and VUYX32 but associate only VUYX32 to the
output queue as the alpha channel of buffers is ignored on this queue.

Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
Reviewed-by: Philipp Zabel <p.zabel@pengutronix.de>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:34:47 -05:00
Vivek Kasireddy
6377bb7d18 media: vivid: Add definitions for the 32-bit packed YUV formats
Enable vivid to make use of the following formats:
 V4L2_PIX_FMT_AYUV32
 V4L2_PIX_FMT_XYUV32
 V4L2_PIX_FMT_VUYA32
 V4L2_PIX_FMT_VUYX32

Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:32:42 -05:00
Vivek Kasireddy
10a2bc7e63 media: v4l2-tpg-core: Add support for 32-bit packed YUV formats (v2)
Add support for the following formats to tpg:
 V4L2_PIX_FMT_AYUV32
 V4L2_PIX_FMT_XYUV32
 V4L2_PIX_FMT_VUYA32
 V4L2_PIX_FMT_VUYX32

Changes from v1:
Remove the duplicate case value

Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:31:23 -05:00
Vivek Kasireddy
a7fe4ca72b media: v4l: Add 32-bit packed YUV formats
The formats added in this patch include:
 V4L2_PIX_FMT_AYUV32
 V4L2_PIX_FMT_XYUV32
 V4L2_PIX_FMT_VUYA32
 V4L2_PIX_FMT_VUYX32

These formats enable the trasmission of alpha channel data to other
drivers and userspace applications in addition to YUV data. For
example, buffers generated by drivers in one of these formats
can be used by the Weston compositor to display as a texture or
flipped directly onto the overlay planes with the help of a DRM
driver.

Signed-off-by: Vivek Kasireddy <vivek.kasireddy@intel.com>
Signed-off-by: Hans Verkuil <hverkuil-cisco@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
2019-02-18 14:31:01 -05:00
Felix Fietkau
f2f6a47b50 mt76: fix resetting software IV flag on key delete
It needs to be unset instead of set

Fixes: 2340523646 ("mt76: fix transmission of encrypted management frames")
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 20:27:52 +01:00
Felix Fietkau
906d2d3f87 mt76: fix corrupted software generated tx CCMP PN
Since ccmp_pn is u8 *, the second half needs to start at array index 4
instead of 0. Fixes a connection stall after a certain amount of traffic

Fixes: 2340523646 ("mt76: fix transmission of encrypted management frames")
Cc: stable@vger.kernel.org
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 20:27:52 +01:00
Felix Fietkau
b231cd7f55 mt76: fix tx status timeout processing
Remove bogus check for MT_PACKET_ID_NO_ACK in mt76_tx_status_skb_get, which
is already handled in callers and turns timeout calls into no-ops
Do not clean up pending status items on reordering of tx status information
if the timeout is not reached yet

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 20:23:46 +01:00
Ramalingam C
35c0272502 drm/audio: declaration of struct device
Header has used the references to struct device without it definition
or declaration. Hence resulting in compilation warning such as

  "'struct device' declared inside parameter list..."

This changes adds a declaration to struct device in the header to avoid
any such warnings.

Signed-off-by: Ramalingam C <ramalingam.c@intel.com>
cc: Takashi Iwai <tiwai@suse.de>
cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Acked-by: Dave Airlie <airlied@redhat.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Link: https://patchwork.freedesktop.org/patch/msgid/1550293499-5560-1-git-send-email-ramalingam.c@intel.com
2019-02-18 20:19:28 +01:00
Ben Hutchings
d0ff23c110 mt76: Use the correct hweight8() function
mt76_init_stream_cap() and mt76_get_txpower() call __sw_hweight8()
directly, but that's only defined if CONFIG_GENERIC_HWEIGHT is
enabled.  The function that works on all architectures is hweight8().

Fixes: 551e1ef4d2 ("mt76: add mt76_init_stream_cap routine")
Fixes: 9313faacbb ("mt76: move mt76x02_get_txpower to mt76 core")
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 20:10:28 +01:00
Stanislaw Gruszka
3eeb7c062d mt76x02u: remove bogus check and comment padding
In mt76x02u_skb_dma_info() pad is always non-zero. Patch removes
bogus check and add comments to the function.

Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 20:03:22 +01:00
Lorenzo Bianconi
4bfff1ec25 mt76: usb: use dev_err_ratelimited instead of dev_err in mt76u_complete_rx
During device removal the driver can report multiple error messages.
Use dev_err_ratelimited instead of dev_err to display urb errors

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 20:01:59 +01:00
Lorenzo Bianconi
c2908a0dfd mt76: usb: introduce disable_usb_sg parameter
Add disable_usb_sg module parameter to disable scatter-gather on demand

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:55:55 +01:00
Lorenzo Bianconi
f752294946 mt76: usb: use a linear buffer for tx/rx datapath if sg is not supported
Use linear fragment and not a single usb scatter-gather buffer in mt76u
{tx,rx} datapath if the usb controller has sg data length constraints

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:55:54 +01:00
Lorenzo Bianconi
d704d16f20 mt76: usb: do not use sg buffers for mcu messages
Do not use scatter-gather buffers for mcu commands.
Introduce mt76u_buf_alloc and mt76u_buf_alloc_sg routines.

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:55:53 +01:00
Lorenzo Bianconi
63a7de5dfb mt76: usb: move mt76u_check_sg in usb.c
Move mt76u_check_sg routine in usb.c and introduce sg_en variable
in mt76_usb in order to check if scatter-gather is supported by
mt76u layer

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:55:52 +01:00
Lorenzo Bianconi
b3098121c4 mt76: usb: do not run mt76u_queues_deinit twice
Do not call mt76u_queues_deinit routine in mt76u_alloc_queues error path
since it will be run in mt76x0u_register_device or
mt76x2u_register_device error path. Current implementation triggers the
following kernel warning:

[   67.005516] WARNING: CPU: 2 PID: 761 at lib/refcount.c:187 refcount_sub_and_test_checked+0xa4/0xb8
[   67.019513] refcount_t: underflow; use-after-free.
[   67.099872] Hardware name: BCM2835
[   67.106268] Backtrace:
[   67.111584] [<8010c91c>] (dump_backtrace) from [<8010cc00>] (show_stack+0x20/0x24)
[   67.124974]  r6:60000013 r5:ffffffff r4:00000000 r3:a50bade6
[   67.132226] [<8010cbe0>] (show_stack) from [<807ca5f4>] (dump_stack+0xc8/0x114)
[   67.141225] [<807ca52c>] (dump_stack) from [<8011e65c>] (__warn+0xf4/0x120)
[   67.149849]  r9:000000bb r8:804d0138 r7:00000009 r6:8099dc84 r5:00000000 r4:b66c7b58
[   67.160767] [<8011e568>] (__warn) from [<8011e6d0>] (warn_slowpath_fmt+0x48/0x50)
[   67.171436]  r9:7f65e128 r8:80d1419c r7:80c0bac4 r6:b97b3044 r5:b7368e00 r4:00000000
[   67.182433] [<8011e68c>] (warn_slowpath_fmt) from [<804d0138>] (refcount_sub_and_test_checked+0xa4/0xb8)
[   67.195221]  r3:80c91c25 r2:8099dc94
[   67.200370]  r4:00000000
[   67.204397] [<804d0094>] (refcount_sub_and_test_checked) from [<804d0164>] (refcount_dec_and_test_checked+0x18/0x1c)
[   67.218046]  r4:b7368e00 r3:00000001
[   67.223125] [<804d014c>] (refcount_dec_and_test_checked) from [<805db49c>] (usb_free_urb+0x20/0x4c)
[   67.235358] [<805db47c>] (usb_free_urb) from [<7f639804>] (mt76u_buf_free+0x98/0xac [mt76_usb])
[   67.247302]  r4:00000001 r3:00000001
[   67.252468] [<7f63976c>] (mt76u_buf_free [mt76_usb]) from [<7f639ef8>] (mt76u_queues_deinit+0x44/0x100 [mt76_usb])
[   67.266102]  r8:b8fe8600 r7:b5dac480 r6:b5dace20 r5:00000001 r4:00000000 r3:00000080
[   67.277132] [<7f639eb4>] (mt76u_queues_deinit [mt76_usb]) from [<7f65c040>] (mt76x0u_cleanup+0x40/0x4c [mt76x0u])
[   67.290737]  r7:b5dac480 r6:b8fe8600 r5:ffffffea r4:b5dace20
[   67.298069] [<7f65c000>] (mt76x0u_cleanup [mt76x0u]) from [<7f65c564>] (mt76x0u_probe+0x1f0/0x354 [mt76x0u])
[   67.311174]  r4:b5dace20 r3:00000000
[   67.316312] [<7f65c374>] (mt76x0u_probe [mt76x0u]) from [<805e0b6c>] (usb_probe_interface+0x104/0x240)
[   67.328915]  r7:00000000 r6:7f65e034 r5:b6634800 r4:b8fe8620
[   67.336276] [<805e0a68>] (usb_probe_interface) from [<8056a8bc>] (really_probe+0x224/0x2f8)
[   67.347965]  r10:b65f0a00 r9:00000019 r8:7f65e034 r7:80d3e124 r6:00000000 r5:80d3e120
[   67.359175]  r4:b8fe8620 r3:805e0a68
[   67.364384] [<8056a698>] (really_probe) from [<8056ab60>] (driver_probe_device+0x6c/0x180)
[   67.375974]  r10:b65f0a00 r9:7f65e2c0 r8:b8fe8620 r7:00000000 r6:7f65e034 r5:7f65e034
[   67.387170]  r4:b8fe8620 r3:00000000
[   67.392378] [<8056aaf4>] (driver_probe_device) from [<8056ad54>] (__driver_attach+0xe0/0xe4)
[   67.404097]  r9:7f65e2c0 r8:7f65d22c r7:00000000 r6:b8fe8654 r5:7f65e034 r4:b8fe8620
[   67.415122] [<8056ac74>] (__driver_attach) from [<8056880c>] (bus_for_each_dev+0x68/0xa0)
[   67.426628]  r6:8056ac74 r5:7f65e034 r4:00000000 r3:00000027
[   67.434017] [<805687a4>] (bus_for_each_dev) from [<8056a1cc>] (driver_attach+0x28/0x30)
[   67.445394]  r6:80c6ddc8 r5:b7368f80 r4:7f65e034
[   67.451703] [<8056a1a4>] (driver_attach) from [<80569c24>] (bus_add_driver+0x194/0x21c)
[   67.463081] [<80569a90>] (bus_add_driver) from [<8056b504>] (driver_register+0x8c/0x124)
[   67.474560]  r7:80c6ddc8 r6:7f65e034 r5:00000000 r4:7f65e034
[   67.481964] [<8056b478>] (driver_register) from [<805df510>] (usb_register_driver+0x74/0x140)
[   67.493901]  r5:00000000 r4:7f65e000
[   67.499131] [<805df49c>] (usb_register_driver) from [<7f661024>] (mt76x0_driver_init+0x24/0x1000 [mt76x0u])
[   67.512258]  r9:00000001 r8:7f65e308 r7:00000000 r6:80c08d48 r5:7f661000 r4:7f65e2c0
[   67.523404] [<7f661000>] (mt76x0_driver_init [mt76x0u]) from [<80102f6c>] (do_one_initcall+0x4c/0x210)
[   67.536142] [<80102f20>] (do_one_initcall) from [<801ae63c>] (do_init_module+0x6c/0x21c)
[   67.547639]  r8:7f65e308 r7:80c08d48 r6:b65f0ac0 r5:7f65e2c0 r4:7f65e2c0
[   67.556129] [<801ae5d0>] (do_init_module) from [<801ad68c>] (load_module+0x1d10/0x2304)

Fixes: b40b15e152 ("mt76: add usb support to mt76 layer")
Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:55:51 +01:00
Lorenzo Bianconi
cb83585e11 mt76: usb: fix possible memory leak in mt76u_buf_free
Move q->ndesc initialization before the for loop in mt76u_alloc_rx
since otherwise allocated urbs will not be freed in mt76u_buf_free
Double-check scatterlist pointer in mt76u_buf_free

Fixes: b40b15e152 ("mt76: add usb support to mt76 layer")
Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:55:50 +01:00
Lorenzo Bianconi
bdba48d57a mt76: usb: fix possible NULL pointer dereference in mt76u_mcu_deinit
Fix possible NULL pointer dereference in mt76u_mcu_deinit routine that
can occur if initialization path fails before calling mt76u_mcu_init_rx

Fixes: ee676cd501 ("mt76: add driver code for MT76x2u based devices")
Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:55:49 +01:00
Stanislaw Gruszka
5de4db8fcb mt76x02u: use usb_bulk_msg to upload firmware
We don't need to send firmware data asynchronously, much simpler is just
use synchronous usb_bulk_msg().

Tested-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Lorenzo Bianconi
1ffe410ee3 mt76x0: init: introduce mt76x0_init_txpower routine
Add mt76x0_init_txpower in order to initialize max_power per channel
at device bootstrap. Modify mt76x0_get_tx_power_per_rate and
mt76x0_get_power_info signature in order to compute tx power for
non-operating channels

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Lorenzo Bianconi
0bee1ff692 mt76x0: phy: report target_power in debugfs
Initialize target_power variable in mt76x0_phy_set_txpower in order to
report target_power in debugfs

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Ryder Lee
bceac167b0 mt76: change the return type of mt76_dma_attach()
There is no need to retun 0 in mt76_dma_attach(), so switch it to void.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Lorenzo Bianconi
ecd25b547d mt76x2: remove mt76x2_alloc_device routine
Remove mt76x2_alloc_device since it just runs mt76_alloc_device.
Move mt76_alloc_device call in mt76x2_probe

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Lorenzo Bianconi
fa2355265c mt76x0: remove mt76x0u_alloc_device routine
Remove mt76x0u_alloc_device since it just runs mt76_alloc_device.
Move mt76_alloc_device call in mt76x0u_probe and in mt76x0e_probe

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Lorenzo Bianconi
3bfaa974de mt76x2u: remove mt76x2u_alloc_device routine
Remove mt76x2u_alloc_device since it just runs mt76_alloc_device.
Move mt76_alloc_device call in mt76x2u_probe

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Lorenzo Bianconi
c0f7b25ab8 mt76: move alloc_device common code in mt76_alloc_device
Move mt76x{0,2} alloc_device common code in mt76_alloc_device and
remove duplicated code

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Felix Fietkau
f1906fb249 mt76: increase ED/CCA tx block threshold
Block only when the busy time reaches 92%, as lower values can be reached with
heavy 802.11 traffic as well.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:34 +01:00
Felix Fietkau
ccdaf7b4f2 mt76: measure the time between mt76x02_edcca_check runs
Based on system load and time needed by other calibration runs, the time
between dev->mac_work runs can vary quite a bit.
Calculate busy time based on the actual time difference in order to avoid
potentially over-estimating busy time, which could lead to unnecessary tx
blocking.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:33 +01:00
Felix Fietkau
c15b7cef2a mt76: clear CCA timer stats in mt76x02_edcca_init
Avoid triggering too early

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:33 +01:00
Felix Fietkau
2e4050240a mt76: run MAC work every 100ms
ED/CCA Tx blocking checks need to be run every 100 ms in order to avoid
triggering too late and keeping tx blocking on for too long

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:33 +01:00
Felix Fietkau
20c06572c1 mt76: explicitly disable energy detect cca during scan
Avoid reusing the previous channel's tx blocking state

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:33 +01:00
Felix Fietkau
1564fa92fb mt76: mt76x2: avoid running DPD calibration if tx is blocked
Doing so could lead to hangs

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:33 +01:00
Felix Fietkau
128b75bf5c mt76: fix software encryption issues
Software encrypted packets can be passed not just through the drv_tx callback,
but also through the intermediate tx queue.
In order to deal with that, move the override to mt76x02_mac_write_txwi and
also take care of filling in the per-packet rate information

Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:33 +01:00
Lorenzo Bianconi
e6287c337c mt76: remove add_buf pointer in mt76_queue_ops
Remove add_buf function pointer in mt76_queue_ops data structure since
it is no longer used

Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
2019-02-18 19:54:33 +01:00