Current implementation executes key notify function in TimerHandler
at TPL_NOTIFY. The code change is to make key notify function
executed at TPL_CALLBACK to reduce the time occupied at TPL_NOTIFY.
The code will signal KeyNotify process event if the key pressed
matches any key registered and insert the KeyData to the EFI Key
queue for notify, then the KeyNotify process handler will invoke
key notify functions at TPL_CALLBACK.
Cc: Ruiyu Ni <Ruiyu.ni@intel.com>
Cc: Michael Kinney <michael.d.kinney@intel.com>
Cc: Feng Tian <feng.tian@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Star Zeng <star.zeng@intel.com>
Reviewed-by: Ruiyu Ni <Ruiyu.ni@intel.com>
This reverts commit 31ae446b1a.
Changing the receive FIFO depth in Terminal driver Start() is not
recommended.
A new PCD PcdUartDefaultReceiveFifoDepth was added and
MdeModulePkg/SerialDxe driver uses the PCD as the default receive
FIFO depth.
Signed-off-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Star Zeng <star.zeng@intel.com>
Acked-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Ryan Harkin <ryan.harkin@linaro.org>
The Serial IO protocol instances provided by SerialDxe and consumed by
TerminalDxe come with a Mode.ReceiveFifoDepth=1 default setting, as
required by UEFI 2.5.
Although TerminalDxe calls EFI_SERIAL_IO_PROTOCOL.SetAttributes() in the
TerminalDriverBindingStart() and TerminalConInTimerHandler() functions, it
only does so to change the Mode.Timeout member. Other members of Mode,
including Mode.ReceiveFifoDepth, are preserved.
On some platforms this causes the UART that underlies TerminalDxe not to
have enough room for bursts of scan codes, which translates to broken
parsing of escape sequences, e.g. cursor movement keys.
According to the UEFI spec, passing ReceiveFifoDepth=0 to
EFI_SERIAL_IO_PROTOCOL.SetAttributes() "will use the device's default FIFO
depth". While TerminalDxe could try to configure a receive FIFO depth that
matches the longest escape sequence it wishes to parse, in practice the
device-specific default FIFO depth -- which may well differ from the
spec-mandated SerialIo->Mode.ReceiveFifoDepth=1 default -- seems to work.
Hence let's just set that.
This issue was exposed by SVN r18971 / git commit 921e987b2b
("ArmPlatformPkg: Use SerialDxe in MdeModulePkg instead of EmbeddedPkg").
In that conversion, MdeModulePkg's SerialDxe started to initialize
Mode.ReceiveFifoDepth to 1 (in conformance with the spec), unlike the
prior, non-conformant initialization to 0 in EmbeddedPkg's SerialDxe.
Since TerminalDxe would never change ReceiveFifoDepth from the new default
value 1, and the ArmPlatformPkg/Drivers/PL011Uart library instance,
underlying SerialDxe through SerialPortLib, would obey it too, they would
collectively effect a receive queue depth of 1, rather than the default 16
or 32. This broke cursor keys on the ARM FVP and Juno platforms.
It is the client of EFI_SERIAL_IO_PROTOCOL that is responsible for
modifying the attributes, if the defaults are not appropriate, hence this
patch modifies TerminalDxe.
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Ryan Harkin <ryan.harkin@linaro.org>
Cc: Leif Lindholm <leif.lindholm@linaro.org>
Cc: Star Zeng <star.zeng@intel.com>
Reported-by: Ryan Harkin <ryan.harkin@linaro.org>
Reference: http://thread.gmane.org/gmane.comp.bios.edk2.devel/4779/focus=6553
Reference: http://thread.gmane.org/gmane.comp.bios.edk2.devel/6594
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Tested-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Ryan Harkin <ryan.harkin@linaro.org>
Reviewed-by: Star Zeng <star.zeng@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@19701 6f19259b-4bc3-4df7-8a09-765794883524
1. Get default terminal type from PCD rather than using PCANSI
directly in BuildTeminalDevpath;
2. Only terminal type is needed to create an TerminalDev instance, so
remove the useless code of creating and freeing DefaultNode.
3. Some white space refining.
Cc: Feng Tian <feng.tian@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Heyi Guo <heyi.guo@linaro.org>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Feng Tian <feng.tian@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18027 6f19259b-4bc3-4df7-8a09-765794883524
This is bug fix for TerminalDxe: NullRemaining should be set to FALSE
by fault and then be set to TRUE conditionally.
Cc: Feng Tian <feng.tian@intel.com>
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Heyi Guo <heyi.guo@linaro.org>
Reviewed-by: Ruiyu Ni <ruiyu.ni@intel.com>
Reviewed-by: Feng Tian <feng.tian@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@18026 6f19259b-4bc3-4df7-8a09-765794883524
Accept the VT220 escape code [3~ as backspace for TtyTerm terminals. This is
sent by many Linux terminals by default. Also accept VT220 function keys
F1-F12, and VT100 F1-F4 keys as these are commonly sent by Linux terminals.
The VT220 escape codes are longer, and variable length so a new state is added
to the state machine along with a variable to construct the multibyte escape
sequence.
There are currently no ambiguous escape sequence prefixes accepted, so the TTY
terminal accepts escape sequences for a variety of terminals. The goal is to
'just work' with as many terminals as possible, rather than properly emulating
any specific terminal. Backspace, Del, and F10 have been tested on xterm,
rxvt, tmux, and screen.
Note: The existing vt100 function key handling does not match the vt100
documentation that I found, so I added the TTY terminal handling
of VT100 F1-F4 (really PF1-PF4 on vt100) separately. The vt100
has no F5-F10 keys, so I don't know what the current vt100 code
is based on.
Contributed-under: TianoCore Contribution Agreement 1.0
Signed-off-by: Roy Franz <roy.franz@linaro.org>
Reviewed-by: Feng Tian <feng.tian@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@17897 6f19259b-4bc3-4df7-8a09-765794883524
This patch a adds new terminal type, TtyTerm, to TerminalDxe. This terminal
type provides a place to add support for various *nix terminals that don't
behave like standard VT terminals. The goal is to 'just work' with as many
terminals as possible, rather than properly emulating any one specific
terminal.
Signed-off-by: Roy Franz <roy.franz@linaro.org>
Contributed-under: TianoCore Contribution Agreement 1.0
Reviewed-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Reviewed-by: Feng Tian <feng.tian@intel.com>
git-svn-id: https://svn.code.sf.net/p/edk2/code/trunk/edk2@17895 6f19259b-4bc3-4df7-8a09-765794883524
The others changes include:
1. Check RemainingDevicePath at beginning of Supported(), make sure it has been verified before Start() is called.
2. Check IO protocol firstly rather than EfiDevicePathProtocolGuid, reduce the times entering into Start() function because EfiDevicePathProtocolGuid existed on most of handle.
3. If no any child device is created on last time, and RemainingDevicePath is valid Uart Devcie path, go on creating child device handle based on this RemainingDevicePath.
git-svn-id: https://edk2.svn.sourceforge.net/svnroot/edk2/trunk/edk2@9262 6f19259b-4bc3-4df7-8a09-765794883524
Star Zeng
Gary Lin
Ruiyu Ni
Laszlo Ersek
Heyi Guo
Roy Franz
Elvin Li
ydong10
li-elvin
niruiyu
jljusten
mdkinney
hhtian
vanjeff
xli24
gikidy
qhuang8
jji4
qwang12
AJFISH
xgu3