This patch fix an issue that the driver may cause "nobody cared" IRQ
when this driver detects the overrun flag only.
Signed-off-by: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com>
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Commit 8b6ff84c2d ("serial: sh-sci: Fix R-Car SCIF and HSCIF
overrun handling") added overrun handling for (H)SCIF using the SCLSR
register, but also accidentally added a bogus call to
sci_handle_fifo_overrun() in the receive interrupt path.
Remove it again.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reorder sampling_rate assignment for consistency in all cases of the
switch statement.
Avoid using the ternary conditional operator to make it more clear that
the value is overridden by platform data.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
serial_core.c was moved from drivers/serial/ to drivers/tty/serial/ a
while ago. Remove the path to make it move-proof.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
SCIF_ERROR_CLEAR includes SCIFA_ORER, which exists only on SCIFA/SCIFB
and SCIF on sh7705/sh7720/sh7721.
To fix this:
1. Remove SCIFA_ORER from the definition of SCIF_ERROR_CLEAR,
2. During initialization, store the error clear mask to use,
incorporating the overrun bit only if it applies to the SCxSR
register.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Expressions involving "BIT(...)" create values of type "long", which is
64-bit on 64-bit. Hence "~BIT(...)" no longer fits in 32-bit, which
will cause future compiler warnings when assigning to 32-bit variables:
drivers/tty/serial/sh-sci.c: In function 'sci_init_single':
drivers/tty/serial/sh-sci.h:58:25: warning: large integer implicitly truncated to unsigned type [-Woverflow]
#define SCI_ERROR_CLEAR ~(SCI_RESERVED | SCI_PER | SCI_FER | SCI_ORER)
^
drivers/tty/serial/sh-sci.c:2325:27: note: in expansion of macro 'SCI_ERROR_CLEAR'
sci_port->error_clear = SCI_ERROR_CLEAR;
As these values are (at most) 32-bit register values anyway, cast them
to "u32" at the definition level to prevent such compiler warnings.
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The #ifdef logic to clear SCxSR bits using RMW on SCIFA/SCIFB and SCIF
variants with some SCIFA features (sh7705/SH7720/sh7721) has several
drawbacks:
- It wasn't updated for newer R-Mobile variants (APE6),
- It doesn't correctly handle SoCs with both SCIF and SCIFA/B (e.g.
R-Car Gen2, but also legacy sh7723/sh7724),
- It doesn't play well with ARM multi-platform kernels: on R-Car Gen2,
SCIF/SCIFA/SCIFB/HSCIF were handled differently, depending on
whether r8a7740 or sh73a0 support was enabled or not,
Replace the #ifdef logic by runtime logic to fix this.
SCIFA/SCIFB and SCIF on sh7705/sh7720/sh7721 use RMW to clear error
bits, other variants use plain stores, as before.
Note that this changes behavior for SCIFA on sh7723/sh7724 (these SoCs
have both SCIF and SCIFA), which didn't use RMW before.
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The following race conditions can happen when a serial port is used
as console.
Case1: CPU_B is used to detect an interrupt from a serial port,
but it can have interrupts disabled during the waiting time.
Case2: CPU_B clears UART_IER just after CPU_A sets UART_IER and then
a serial port may not make an interrupt.
Case3: CPU_A sets UART_IER just after CPU_B clears UART_IER.
This is an unexpected behavior for serial8250_console_write().
CPU_A [autoconfig_irq] | CPU_B [serial8250_console_write]
----------------------------|---------------------------------------
|
probe_irq_on() | spin_lock_irqsave(&port->lock,)
serial_outp(,UART_IER,0x0f) | serial_out(,UART_IER,0)
udelay(20); | uart_console_write()
probe_irq_off() |
| spin_unlock_irqrestore(&port->lock,)
Case1 and 2 can make autoconfig_irq() failed.
In these cases, the console doesn't work in interrupt mode and
"input overrun" (which can make operation mistakes) can happen
on some systems. Especially in the Case1, It is known that the
problem happens with high rate every boot once it occurs
because the boot sequence is always almost same.
port mutex makes sure that the autoconfig operation is exclusive of
any other concurrent HW access except by the console operation.
console lock is required in autoconfig_irq().
Signed-off-by: Taichi Kageyama <t-kageyama@cp.jp.nec.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Reviewed-by: Peter Hurley <peter@hurleysoftware.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This fixes receiving broken characters on the console from an MPC5125
system when systemd comes up which repeatedly opens and shuts down the
console device.
Trial and error with the needed interval showed that 500 us are good
enough most of the time when using 38400 Bd, so I think 1 ms is a good
compromise between fixing the issue and not penalize faster setups too
much.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Kconfig currently controlling compilation of this code is:
8250/Kconfig:config SERIAL_8250_LPC18XX
8250/Kconfig: bool "NXP LPC18xx/43xx serial port support"
...meaning that it currently is not being built as a module by anyone.
When targetting orphaned modular code in non-modular drivers, this
came up. Joachim indicated that the driver was actually meant to
be tristate but ended up bool by accident. So here we make it
tristate instead of removing the modular code that was essentially
orphaned.
Suggested-by: Joachim Eastwood <manabian@gmail.com>
Cc: Jiri Slaby <jslaby@suse.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Acked-by: Joachim Eastwood <manabian@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Kconfig for this driver is currently:
config SERIAL_MPSC
bool "Marvell MPSC serial port support"
...meaning that it currently is not being built as a module by anyone.
Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only.
Since module_init translates to device_initcall in the non-modular
case, the init ordering remains unchanged with this commit.
We leave some tags like MODULE_AUTHOR for documentation purposes.
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Cc: linux-serial@vger.kernel.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Kconfig currently controlling compilation of this code is:
drivers/tty/hvc/Kconfig:config HVC_DRIVER
drivers/tty/hvc/Kconfig: bool
...meaning that it currently is not being built as a module by anyone.
Lets remove the modular code that is essentially orphaned, so that
when reading the driver there is no doubt it is builtin-only, even
though someone bothered to comment that the code was not used.
Unlike other changes, this driver binds in w/o using module_init,
so we dont have init ordering concerns with this commit.
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Acked-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: linuxppc-dev@lists.ozlabs.org
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Kconfig currently controlling compilation of this code is:
config.debug:config MAGIC_SYSRQ
bool "Magic SysRq key"
...meaning that it currently is not being built as a module by anyone.
Lets remove the traces of modularity we can so that when reading the
driver there is less doubt it is builtin-only.
Since module_init translates to device_initcall in the non-modular
case, the init ordering remains unchanged with this commit.
We don't delete the module.h include since other parts of the file are
using content from there.
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The Kconfig currently controlling compilation of this code is:
drivers/tty/Kconfig:config LEGACY_PTYS
drivers/tty/Kconfig: bool "Legacy (BSD) PTY support"
...and:
drivers/tty/Kconfig:config UNIX98_PTYS
drivers/tty/Kconfig: bool "Unix98 PTY support" if EXPERT
combined with this:
obj-$(CONFIG_LEGACY_PTYS) += pty.o
obj-$(CONFIG_UNIX98_PTYS) += pty.o
...meaning that it currently is not being built as a module by anyone.
Lets remove the traces of modularity we can so that when reading the
driver there is less doubt it is builtin-only.
Since module_init translates to device_initcall in the non-modular
case, the init ordering remains unchanged with this commit.
We don't delete the module.h include since other parts of the file are
using content from there.
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jiri Slaby <jslaby@suse.com>
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CIR type serial ports aren't real serial ports.
This is just a way to prevent legacy 8250 serial
driver from probing and eventually binding some
resources.
Since in current state such ports aren't providing
any real functionality and it is not possible
to change their type via setserial/ioctl(TIOCSSERIAL)
(due to UPF_FIXED_PORT flag set on them)
it is simpler and cleaner to not register them at all
with serial core.
Print a short message in this case so it is known
to user what has happened.
This way checks for PORT_8250_CIR in serial port
callbacks can be removed too, since they won't
ever be called.
Signed-off-by: Maciej Szmigiero <mail@maciej.szmigiero.name>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The function tty_insert_flip_string() returns an int and as such it
might fail. So the result is that I kindly asked to insert 48 bytes and
the function only insterted 32.
I have no idea what to do with the remaining 16 so I think dropping them
is the only option. I also increase the buf_overrun counter so userpace
has a clue that we lost bytes.
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Tested-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull strscpy string copy function implementation from Chris Metcalf.
Chris sent this during the merge window, but I waffled back and forth on
the pull request, which is why it's going in only now.
The new "strscpy()" function is definitely easier to use and more secure
than either strncpy() or strlcpy(), both of which are horrible nasty
interfaces that have serious and irredeemable problems.
strncpy() has a useless return value, and doesn't NUL-terminate an
overlong result. To make matters worse, it pads a short result with
zeroes, which is a performance disaster if you have big buffers.
strlcpy(), by contrast, is a mis-designed "fix" for strlcpy(), lacking
the insane NUL padding, but having a differently broken return value
which returns the original length of the source string. Which means
that it will read characters past the count from the source buffer, and
you have to trust the source to be properly terminated. It also makes
error handling fragile, since the test for overflow is unnecessarily
subtle.
strscpy() avoids both these problems, guaranteeing the NUL termination
(but not excessive padding) if the destination size wasn't zero, and
making the overflow condition very obvious by returning -E2BIG. It also
doesn't read past the size of the source, and can thus be used for
untrusted source data too.
So why did I waffle about this for so long?
Every time we introduce a new-and-improved interface, people start doing
these interminable series of trivial conversion patches.
And every time that happens, somebody does some silly mistake, and the
conversion patch to the improved interface actually makes things worse.
Because the patch is mindnumbing and trivial, nobody has the attention
span to look at it carefully, and it's usually done over large swatches
of source code which means that not every conversion gets tested.
So I'm pulling the strscpy() support because it *is* a better interface.
But I will refuse to pull mindless conversion patches. Use this in
places where it makes sense, but don't do trivial patches to fix things
that aren't actually known to be broken.
* 'strscpy' of git://git.kernel.org/pub/scm/linux/kernel/git/cmetcalf/linux-tile:
tile: use global strscpy() rather than private copy
string: provide strscpy()
Make asm/word-at-a-time.h available on all architectures
Support for TSYS01 temperature sensor
Signed-off-by: Ludovic Tancerel <ludovic.tancerel@maplehightech.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
Measurement specialties drivers common part.
These functions are used by further drivers
in the patchset: TSYS01, TSYS02D, HTU21, MS5637, MS8607
Signed-off-by: Ludovic Tancerel <ludovic.tancerel@maplehightech.com>
Signed-off-by: Jonathan Cameron <jic23@kernel.org>
It was observed that setting TIF_MEMDIE before sending SIGKILL at
oom_kill_process() allows memory reserves to be depleted by allocations
which are not needed for terminating the OOM victim.
This patch reverts commit 6bc2b856bb ("staging: android: lowmemorykiller:
set TIF_MEMDIE before send kill sig"), for oom_kill_process() was updated
to send SIGKILL before setting TIF_MEMDIE.
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We don't want to swap bytes that we're reading and writing to the
FIFOs when we're running on a big-endian CPU. Doing so causes
problems like where the qcom-spmi-iadc driver can't detect the
type of device because the bytes are all mixed up. Use the raw IO
accessors for these API instead, and collapse pmic_arb_base_read()
into the byte reading API so that we aren't tempted to read non-FIFO
data like commands with that function.
Cc: Andy Gross <agross@codeaurora.org>
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Acked-by: Bjorn Andersson <bjorn.andersson@sonymobile.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Get notified immediately when a balloon target is set, instead of waiting for
up to one second.
The up-to 1 second gap could be long enough to cause swapping inside of the
VM that receives the VM.
Acked-by: Andy King <acking@vmware.com>
Signed-off-by: Xavier Deguillard <xdeguillard@vmware.com>
Tested-by: Siva Sankar Reddy B <sankars@vmware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Unify the behavior of the first start of the balloon and a reset. Also on
unload, declare that the balloon driver does not have any capabilities
anymore.
Acked-by: Andy King <acking@vmware.com>
Signed-off-by: Xavier Deguillard <xdeguillard@vmware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2m ballooning significantly reduces the hypervisor side (and guest side)
overhead of ballooning and unballooning.
hypervisor only:
balloon unballoon
4 KB 2 GB/s 2.6 GB/s
2 MB 54 GB/s 767 GB/s
Use 2 MB pages as the hypervisor is alwys 64bit and 2 MB is the smallest
supported super-page size.
The code has to run on older versions of ESX and old balloon drivers run on
newer version of ESX. Hence match the capabilities with the host before 2m
page ballooning could be enabled.
Signed-off-by: Xavier Deguillard <xdeguillard@vmware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When VMware's hypervisor requests a VM to reclaim memory this is preferrably done
via ballooning. If the balloon driver does not return memory fast enough, more
drastic methods, such as hypervisor-level swapping are needed. These other methods
cause performance issues, e.g. hypervisor-level swapping requires the hypervisor to
swap in a page syncronously while the virtual CPU is blocked.
Hence it is in the interest of the VM to balloon memory as fast as possible. The
problem with doing this is that the VM might end up doing nothing else than
ballooning and the user might notice that the VM is stalled, esp. when the VM has
only a single virtual CPU.
This is less of a problem if the VM and the hypervisor perform balloon operations
faster. Also the balloon driver yields regularly, hence on a single virtual CPU
the Linux scheduler should be able to properly time-slice between ballooning and
other tasks.
Testing Done: quickly ballooned a lot of pages while wathing if there are any
perceived hickups (periods of non-responsiveness) in the execution of the
linux VM. No such hickups were seen.
Signed-off-by: Xavier Deguillard <xdeguillard@vmware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This helps with debugging vmw_balloon behavior, as it is clear what
functionality is enabled.
Acked-by: Andy King <acking@vmware.com>
Signed-off-by: Xavier Deguillard <xdeguillard@vmware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead of waiting for the next GET_TARGET command, we can react faster
by exploiting the fact that each hypervisor call also returns the
balloon target.
Signed-off-by: Xavier Deguillard <xdeguillard@vmware.com>
Acked-by: Dmitry Torokhov <dtor@vmware.com>
Signed-off-by: Philip P. Moltmann <moltmann@vmware.com>
Acked-by: Andy King <acking@vmware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Introduce a new capability to the driver that allow sending 512 pages in
one hypervisor call. This reduce the cost of the driver when reclaiming
memory.
Signed-off-by: Xavier Deguillard <xdeguillard@vmware.com>
Acked-by: Dmitry Torokhov <dtor@vmware.com>
Signed-off-by: Philip P. Moltmann <moltmann@vmware.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If kzalloc() fails then gms is NULL and we are returning NULL, but the
functions which called this function gru_register_mmu_notifier() are not
expecting NULL as the return. They are expecting either a valid pointer
or the error code in ERR_PTR.
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Acked-by: Dimitri Sivanich <sivanich@sgi.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Remove the function tfh_restart() that is not used anywhere.
This was partially found by using a static code analysis program called cppcheck.
Signed-off-by: Rickard Strandqvist <rickard_strandqvist@spectrumdigital.se>
Acked-by: Dimitri Sivanich <sivanich@sgi.com>
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To be future-proof and for better readability the time comparisons are
modified to use time_before() instead of plain, error-prone math.
Signed-off-by: Manuel Schölling <manuel.schoelling@gmx.de>
Acked-by: Dimitri Sivanich <sivanich@sgi.com>
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch adds the SCIF kernel node QP control messages required to
enable SCIF RMAs. Examples of such node QP control messages include
registration, unregistration, remote memory allocation requests,
remote memory unmap and SCIF remote fence requests.
The patch also updates the SCIF driver with minor changes required to
enable SCIF RMAs by adding the new files to the build, initializing
RMA specific information during SCIF endpoint creation, reserving SCIF
DMA channels, initializing SCIF RMA specific global data structures,
adding the IOCTL hooks required for SCIF RMAs and updating RMA
specific debugfs hooks.
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Nikhil Rao <nikhil.rao@intel.com>
Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch implements the fence APIs required to synchronize
DMAs. SCIF provides an interface to return a "mark" for all DMAs
programmed at the instant the API was called. Users can then "wait" on
the mark provided previously by blocking inside the kernel. Upon
receipt of a DMA completion interrupt the waiting thread is woken
up. There is also an interface to signal DMA completion by polling for
a location to be updated via a "signal" cookie to avoid the interrupt
overhead in the mark/wait interface. SCIF allows programming fences on
both the local and the remote node for both the mark/wait or the fence
signal APIs.
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Nikhil Rao <nikhil.rao@intel.com>
Signed-off-by: Jacek Lawrynowicz <jacek.lawrynowicz@intel.com>
Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
SCIF allows users to read from or write to registered remote memory
via CPU copies or DMA. The API verifies that both local and remote
windows are valid before initiating the CPU or DMA transfers. SCIF has
optimized algorithms for handling byte aligned as well as cache line
aligned DMA engines. A registration cache is maintained to avoid the
overhead of pinning pages repeatedly if buffers are reused. The
registration cache is invalidated upon receipt of MMU notifier
callbacks. SCIF windows are destroyed and the pages are unpinned only
once all prior DMAs initiated using that window are drained. Users can
request synchronous DMA operations as well as tail byte ordering if
required. CPU copies are always performed synchronously.
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Nikhil Rao <nikhil.rao@intel.com>
Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch implements the SCIF mmap/munmap interface. A similar
capability is provided to kernel clients via the
scif_get_pages()/scif_put_pages() APIs. The SCIF mmap interface
queries to check if a window is valid and then remaps the local
virtual address to the remote physical pages. These mappings are
subsequently destroyed upon receipt of the VMA close operation or
scif_get_pages(). This functionality allows SCIF users to directly
access remote memory without any driver interaction once the mappings
are created thereby providing bare-metal PCIe latency. These mappings
are zapped to avoid RMA accesses from user space, if a Coprocessor is
reset.
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Nikhil Rao <nikhil.rao@intel.com>
Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch adds the implementation for operations performed on the
list of SCIF windows. Examples of such operations includes adding the
windows to the list of registered (or cached) windows, querying the
list of self or remote windows and unregistering windows. The query
operation is used by SCIF APIs which initiate DMAs, CPU copies or
fences to ensure that a window remains valid during a transfer.
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Nikhil Rao <nikhil.rao@intel.com>
Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch implements the SCIF APIs required to pin and unpin
pages. SCIF registration locks down the pages. It then sends a remote
window allocation request to the peer. Once the peer has allocated
memory, the local SCIF endpoint copies the pinned page information to
the peer and notifies the peer once the copy has complete. The peer
upon receipt of the registration notification adds the new remote
window to its list. At this point the window page information is
available on both self and remote nodes so that they can start
performing SCIF DMAs, CPU copies and fences. The unregistration API
tears down the registration at both self and remote nodes.
Reviewed-by: Nikhil Rao <nikhil.rao@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This patch adds the internal data structures required to perform SCIF
RMAs. The data structures required to maintain per SCIF endpoint, RMA
information are contained in scif_endpt_rma_info. scif_pinned_pages
describes a set of SCIF pinned pages maintained locally. The
scif_window is a data structure which contains all the fields required
to describe a SCIF registered window on self and remote nodes. It
contains an offset which is used as a key to perform SCIF DMAs and CPU
copies between self and remote registered windows.
Reviewed-by: Nikhil Rao <nikhil.rao@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Sudeep Dutt <sudeep.dutt@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
