We move all the probe stuff to kthread so that it won't block
the system to go on probing other drivers. But that introduced
a bug that PM calls would not be removed by driver core. As each
platform driver uses the same PM callbacks for all device instances.
So add device_release_driver if it fails to probe devices.
Fixes: 79ac46bdea ("PCI: rockchip: dw: Add kthread to probe PCIe devices")
Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Change-Id: I9d6ba448a87defa8d924927f1bfcff51c889e1a0
When phy status is none, it mean that the power off
process is already be run. if the phy power off
function is called again before power on, ignore it.
Signed-off-by: Zhang Yubing <yubing.zhang@rock-chips.com>
Change-Id: I34c3f2b92076f7c27a1537d3c78000781de61179
when we set mode change flag and the phy is idle, here
will not clear the mode change flag when power on, we
need clear it.
Signed-off-by: Zhang Yubing <yubing.zhang@rock-chips.com>
Change-Id: I900d5c37a2e83212eb6461d714a653c93da923c7
Enable lt6911uxc HDMI to MIPI CSI-2 bridge driver for hdmi-in
application. Which found on rk3568-evb2-lp4x-v10.
Signed-off-by: Dingxian Wen <shawn.wen@rock-chips.com>
Change-Id: I4ce22f3040dbe55c3236ac3d0ac62f7821011c8d
v0.01.01: fix if plugin_gpio was not used
v0.01.02: modify driver init level to late_initcall
Signed-off-by: Jianwei Fan <jianwei.fan@rock-chips.com>
Change-Id: I3b15bed365552aa0dbaf8a8126400597caa96ed6
Add add heap - "system-dma32" to allocate pages with GFP_DMA32, support cache,
and add heap - "system-uncached-dma32" to allocate pages with GFP_DMA32, but
unsupport cache
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>
Change-Id: I93bb7c17fc038f52940ddc4facf836637c917bea
rkvenc2 encoder need rcb buffer for H.264 above 4K.
Signed-off-by: Herman Chen <herman.chen@rock-chips.com>
Change-Id: Ic1b9b5fafa3e4488d8e993e21335ca8bd7cd5c6f
To classify errors:
RX/TX error count < 128 is CAN_STATE_ERROR_WARNING,
An error frame is sent, enters the active error state.
RX/TX error count < 256 is CAN_STATE_ERROR_PASSIVE,
An error frame is sent, enter passive error state.
RX/TX error count >= 256 is CAN_STATE_BUS_OFF,
can bus off need to restart.
Signed-off-by: Elaine Zhang <zhangqing@rock-chips.com>
Change-Id: I4a85af3c86c5dfd14e4274832b0fb55ac09c98f7
Remove executable bit on abi_gki_aarch64_mtk file.
This file was added in 55d7c4eca6, the mode of
abi_gki_aarch64.xml was fixed in eb02ea0e, this
patch fixes the other file that from the
55d7c4eca6 patch.
Bug: 149040612
Fixes: 55d7c4eca6 ("ANDROID: Update symbol list for mtk")
Signed-off-by: Pat Tjin <pattjin@google.com>
Change-Id: I3800136eb7d7f54a89508660ae40e662ee5ac571
We observed the following deadlock in the stress test under low
memory scenario:
Thread A Thread B
- erofs_shrink_scan
- erofs_try_to_release_workgroup
- erofs_workgroup_try_to_freeze -- A
- z_erofs_do_read_page
- z_erofs_collection_begin
- z_erofs_register_collection
- erofs_insert_workgroup
- xa_lock(&sbi->managed_pslots) -- B
- erofs_workgroup_get
- erofs_wait_on_workgroup_freezed -- A
- xa_erase
- xa_lock(&sbi->managed_pslots) -- B
To fix this, it needs to hold xa_lock before freezing the workgroup
since xarray will be touched then. So let's hold the lock before
accessing each workgroup, just like what we did with the radix tree
before.
[ Gao Xiang: Jianhua Hao also reports this issue at
https://lore.kernel.org/r/b10b85df30694bac8aadfe43537c897a@xiaomi.com ]
Link: https://lore.kernel.org/r/20211118135844.3559-1-huangjianan@oppo.com
Fixes: 64094a0441 ("erofs: convert workstn to XArray")
Reviewed-by: Chao Yu <chao@kernel.org>
Reviewed-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Signed-off-by: Huang Jianan <huangjianan@oppo.com>
Reported-by: Jianhua Hao <haojianhua1@xiaomi.com>
Signed-off-by: Gao Xiang <xiang@kernel.org>
Signed-off-by: haojianhua1 <haojianhua1@xiaomi.com>
Change-Id: Icdfb520c073de8773d44cff64fd5f7a8499cdc85
Signed-off-by: haojianhua1 <haojianhua1@xiaomi.com>
Bug:208318427
(cherry picked from commit 57bbeacdbe)
Signed-off-by: Gao Xiang <xiang@kernel.org>
Change-Id: Ie282b4362898fa8e0c829b8c2454887e953a2610
Signed-off-by: haojianhua1 <haojianhua1@xiaomi.com>
Without initialization, it will be random data and hard for
vendor hook to decide.
Bug: 207739506
Change-Id: I278772d87eea38c03a40d4f0bef20ac8644e2ecd
Signed-off-by: Maria Yu <quic_aiquny@quicinc.com>
Nothing protects the access to the per_cpu variable sd_llc_id. When testing
the same CPU (i.e. this_cpu == that_cpu), a race condition exists with
update_top_cache_domain(). One scenario being:
CPU1 CPU2
==================================================================
per_cpu(sd_llc_id, CPUX) => 0
partition_sched_domains_locked()
detach_destroy_domains()
cpus_share_cache(CPUX, CPUX) update_top_cache_domain(CPUX)
per_cpu(sd_llc_id, CPUX) => 0
per_cpu(sd_llc_id, CPUX) = CPUX
per_cpu(sd_llc_id, CPUX) => CPUX
return false
ttwu_queue_cond() wouldn't catch smp_processor_id() == cpu and the result
is a warning triggered from ttwu_queue_wakelist().
Avoid a such race in cpus_share_cache() by always returning true when
this_cpu == that_cpu.
Fixes: 518cd62341 ("sched: Only queue remote wakeups when crossing cache boundaries")
Reported-by: Jing-Ting Wu <jing-ting.wu@mediatek.com>
Signed-off-by: Vincent Donnefort <vincent.donnefort@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Link: https://lore.kernel.org/r/20211104175120.857087-1-vincent.donnefort@arm.com
Bug: 204726704
Change-Id: Ib6e59187b6d7d7dcabae84e3541d5fbe0dfc400a
(cherry picked from commit 42dc938a59)
Signed-off-by: Jing-Ting Wu <Jing-Ting.Wu@mediatek.com>
1. Fixup timeout when running multi thread,
because fd can not pass on other thread.
2. Tide some code
Signed-off-by: Li Huang <putin.li@rock-chips.com>
Change-Id: I68b0557138f583ab8bb341bf5bbe2ec145d2884c
This reverts commit 6b7fb9a6c9.
The audio regbank of hdmi-qp is drived by audio interface clk, and had
fixed by commits as follows:
[1] 3a7f369b5c ("arm64: dts: rockchip: rk3588: Add aud clk for hdmi nodes")
[2] e108ff9f6f ("drm/rockchip: dw_hdmi: Handle aud clk for hdmi qp")
OTOH, to make android hal happy, we still allow to access hdmi audio even
hdmi is plugged out.
"hdmi-audio-codec: ASoC: error at snd_soc_dai_startup on i2s-hifi: -19"
Android/Linux should reopen hdmi sound card depends on HPD plug event
to bring back to normal state.
Change-Id: I654fe18a04b750f57c8bd52ef4948a476bc1fa50
Signed-off-by: Sugar Zhang <sugar.zhang@rock-chips.com>
The separation of pinctrl and gpio drivers created a tiny window where
a pinconfig setting might produce a null-pointer dereference.
The affected device were rk3288-veyron devices in this case.
Pinctrl-hogs are claimed when the pinctrl driver is registered,
at which point their pinconfig settings will be applied.
At this time the now separate gpio devices will not have been created
yet and the matching driver won't have probed yet, making the gpio->foo()
call run into a null-ptr.
As probing is not really guaranteed to have been completed at a specific
time, introduce a queue that can hold the output settings until the gpio
driver has probed and will (in a separate patch) fetch the elements
of the list.
We expect the gpio driver to empty the list, but will nevertheless empty
it ourself on remove if that didn't happen.
Fixes: 9ce9a02039 ("pinctrl/rockchip: drop the gpio related codes")
Signed-off-by: Heiko Stuebner <heiko@sntech.de>
Link: https://lore.kernel.org/r/20210913224926.1260726-4-heiko@sntech.de
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
(cherry picked from commit e7165b1dff)
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>
Change-Id: Id179c7eaacfb41565e2427ad82b55d11c2be8c4d
Some difference made during driver upstream review, this patch sync the
driver with upstream version.
Signed-off-by: Jianqun Xu <jay.xu@rock-chips.com>
Change-Id: I0da05d04f54bfc8e83afa28692c658b0b4d2604b
