dw_mci write cmd index to CMD register to trigger sending cmd by BIU.
However, if device fall into panic holding cmd/data line to low level cause BIU
cannnot send out cmd forever. So no cmd_done_int will come. AND, cmd response timeout
only valid after cmd been sent. Nothing to break this loop, we need a s/w recovery from
STATE_SENDING_CMD to STATE_IDLE, and the pending one reported as -ETIMEOUT, let caller
decide howto again.
Reported-by: roger.hu <hwg@rock-chips.com>
Signed-off-by: lintao <lintao@rock-chips.com>
Reviewed-and-tested-by: roger.hu <hwg@rock-chips.com>
1.Audi vpu_combo contain hevc and vpu,it need to switch
when hevc on or vpu on,but there was a issue could cause
hevc or vpu failed,so current vpu driver disable/enable
iommu each frame to avoid failed,for these,a lot of log
produced,so change dev_info to dev_dbg when iommu attach/deattach
2.AudiB has fixed the vop read problem,we use soc_is_rk3126 or
soc_is_rk3128 instead of cpu_is_rk312x to identify Audi
This patch adds the support of security labels for f2fs, which will be used
by Linus Security Models (LSMs).
Quote from http://en.wikipedia.org/wiki/Linux_Security_Modules:
"Linux Security Modules (LSM) is a framework that allows the Linux kernel to
support a variety of computer security models while avoiding favoritism toward
any single security implementation. The framework is licensed under the terms of
the GNU General Public License and is standard part of the Linux kernel since
Linux 2.6. AppArmor, SELinux, Smack and TOMOYO Linux are the currently accepted
modules in the official kernel.".
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
(cherry picked from commit 8ae8f1627f)
I found a bug when testing power-off-recovery as follows.
[Bug Scenario]
1. create a file
2. fsync the file
3. reboot w/o any sync
4. try to recover the file
- found its fsync mark
- found its dentry mark
: try to recover its dentry
- get its file name
- get its parent inode number
: here we got zero value
The reason why we get the wrong parent inode number is that we didn't
synchronize the inode page with its newly created inode information perfectly.
Especially, previous f2fs stores fi->i_pino and writes it to the cached
node page in a wrong order, which incurs the zero-valued i_pino during the
recovery.
So, this patch modifies the creation flow to fix the synchronization order of
inode page with its inode.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
(cherry picked from commit 44a83ff6a8)
This patch is for passing a locked node page to get_dnode_of_data.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
(cherry picked from commit 64aa7ed98d)
1.
schedule_delayed_work(&pcd->check_id_work,msecs_to_jiffies(HZ))
should be schedule_delayed_work(&pcd->check_id_work,HZ)
2.fix coding-style to pass checkpatch.pl
3.set otg controler working on foece host/device mode to avoid
mode_missmatch_interrupts
If same layer is closed twice, atv_layer_cnt may be
set to zero, so vop go to standy. Now win0、win1、hwc
is mapped to atv_layer_cnt bit[0-2]. When open/close
layer, set coresponding atv_layer_cnt bit to one/zero.
The driver supports 32 parameters of each mipi cmds
when initialization the mipi screen.
But the users may send a cmds longer the 32 parameters sometimes,
and it will result in that the array out of range.
DTYPE_GEN_SWRITE_2P:
Generic Short WRITE Packet with 2 parameters(one command
and one parameter). But the users may send the packet more
than 2 parameters.
DTYPE_GEN_SWRITE_1P:
Generic Short WRITE Packet with 1 parameters(one command
without parameter).
DTYPE_GEN_SWRITE_0P:
Generic Short WRITE Packet with 0 parameters(nop packet
without command and parameter).
Take following sequence to avoid picture flash when swiching resolution:
1. Send AV_MUTE GCP packet.
2. Power off hdmi phy.
3. Sleep 2 second to wait TV process no TMDS signal.
4. Configure video and audio registers.
5. Power on HDMI phy.
6. Send CLEAR_MUTE GCP packet.
before release fence, we should make sure extend lcdc config take
effect. if not, maybe GPU would access the buffer which hdmi using,
and cause tearing.
Signed-off-by: Mark Yao <mark.yao@rock-chips.com>
1. Remove ehci_port_power() in ehci-rockchip, since commit
c73cee7 the ehci_port_power() function no longer needed,
ehci hub driver taking care of power supply.
2. Fix 64-bit address pointer warnning.
3. Register rk-rhci driver by ehci_init_driver() for reducing
useless code.
4. Fix kernrl crash caused by null pointer dereference ,
When 'echo 0 > ehci_power' usb_remove_hcd() run twice. Set rk
ehci->host_enabled 0 to stop connect_detect thread.
