This commit accomplishes a few things:
1) Properly prevent multiple sets of threads from running at once using
a mutex. Lots of race issues existed with the thread_cleanup.
2) The mutex allows us to ensure that threads are finished before
tearing down the device or module.
3) Don't use kthread_stop when the threads can exit by themselves, as
this is counter-indicated by the kthread_create documentation. Threads
now wait for kthread_stop to occur.
4) Writing to the run file now blocks until the threads are complete.
The test can then be safely interrupted by a SIGINT.
Also, while I was at it:
5) debugfs_run_write shouldn't return 0 in the early check cases as this
could cause debugfs_run_write to loop undesirably.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
When debugging performance problems, if some issue causes the ntb
hardware to be significantly slower than expected, ntb_perf will
hang requiring a reboot because it only schedules once every 4GB.
Instead, schedule based on jiffies so it will not hang the CPU if
the transfer is slow.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
I'm working on hardware that currently has a limited number of
scratchpad registers and ntb_ndev fails with no clue as to why. I
feel it is better to fail early and provide a reasonable error message
then to fail later on.
The same is done to ntb_perf, but it doesn't currently require enough
spads to actually fail. I've also removed the unused SPAD_MSG and
SPAD_ACK enums so that MAX_SPAD accurately reflects the number of
spads used.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
We allocate some memory window buffers when the link comes up, then we
provide debugfs files to read/write each side of the link.
This is useful for debugging the mapping when writing new drivers.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
On my system, dma_alloc_coherent won't produce memory anywhere
near the size of the BAR. So I needed a way to limit this.
It's pretty much copied straight from ntb_transport.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Currently we only allocate a fixed default number of descriptors for the tx
and rx side. We should dynamically resize it to the number of descriptors
resides in the transport rings. We should know the number of transmit
descriptors at initializaiton. We will allocate the default number of
descriptors for receive side and allocate additional ones when we know the
actual max entries for receive.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Allen Hubbe <allen.hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
On hardware with 32 scratchpad registers the spad field in ntb tool
could chop off the end. The maximum buffer size is increased from
256 to 15 times the number or scratchpads.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
If you tried to write two spads in one line, as per the example:
root@peer# echo '0 0x01010101 1 0x7f7f7f7f' > $DBG_DIR/peer_spad
then the CPU would freeze in an infinite loop.
This wasn't immediately obvious but 'pos' was not incrementing the
buffer, so after reading the second pair of values, 'pos' would once
again be 3 and it would re-read the second pair of values ad infinitum.
Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Acked-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Kernel zero day testing warned about address space confusion. A virtual
iomem address was used where a physical address is expected. The
offending functions implement an optional part of the api, so they are
removed. They can be added later, after testing.
Fixes: a1b3695820
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Acked-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The clean up routine when we failed to allocate kthread is not cleaning
up all the threads, only the same one over and over again.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
kthread_create_no_node() returns error pointers, never NULL. Fix check so
it handles error correctly.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
kmalloc can fail and we should check for NULL before using the pointer
returned by kmalloc.
Signed-off-by: Sudip Mukherjee <sudip.mukherjee@codethink.co.uk>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The perf tool is missing the setup of translation window. Adding call to
setup the translation window for backed memory.
Signed-off-by: John Kading <john.kading@gd-ms.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Instead of keep trying to go through the init routine when we aren't able
to allocate memory, we should just stop and go down.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
We can leave tasklet spinning forever if we disable the tasklet during
qp shutdown and the tasklets are still being kicked off. This hopefully
should avoid that race condition.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reported-by: Alex Depoutovitch <alex@pernixdata.com>
Tested-by: Alex Depoutovitch <alex@pernixdata.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The ntb driver assigns between pointers an __iomem tokens, and
also casts them to 64-bit integers, which results in compiler
warnings on 32-bit systems:
drivers/ntb/test/ntb_perf.c: In function 'perf_copy':
drivers/ntb/test/ntb_perf.c:213:10: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
vbase = (u64)(u64 *)mw->vbase;
^
drivers/ntb/test/ntb_perf.c:214:14: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
dst_vaddr = (u64)(u64 *)dst;
^
This adds __iomem annotations where needed and changes the temporary
variables to iomem pointers to avoid casting them to u64. I did not
see the problem in linux-next earlier, but it show showed up in
4.5-rc1.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Fixes: 8a7b6a778a ("ntb: ntb perf tool")
Signed-off-by: Jon Mason <jdmason@kudzu.us>
If the parameter given to the macro is replaced throughout the macro as
it is evaluated. The intent is that the macro parameter should replace
the only the first parameter to container_of(). However, the way the
macro was written, it would also inadvertantly replace a structure field
name. If a parameter of any other name is given to the macro, it will
fail to compile, if the structure does not contain a field of the same
name. At worst, it will compile, and hide improper access of an
unintended field in the structure.
Change the macro parameter name, so it does not conflict with the
structure field name.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
This adds support for AMD's PCI-Express Non-Transparent Bridge
(NTB) device on the Zeppelin platform. The driver connnects to the
standard NTB sub-system interface, with modification to add hooks
for power management in a separate patch. The AMD NTB device has 3
memory windows, 16 doorbell, 16 scratch-pad registers, and supports
up to 16 PCIe lanes running a Gen3 speeds.
Signed-off-by: Xiangliang Yu <Xiangliang.Yu@amd.com>
Reviewed-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Providing raw performance data via a tool that directly access data from
NTB w/o any software overhead. This allows measurement of the hardware
performance limit. In revision one we are only doing single direction
CPU and DMA writes. Eventually we will provide bi-directional writes.
The measurement using DMA engine for NTB performance measure does
not measure the raw performance of DMA engine over NTB due to software
overhead. But it should provide the peak performance through the Linux DMA
driver.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Tested-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The transport right now does not handle the case where we run out of DMA
descriptors. We just fail when we do not succeed. Adding code to retry for
a bit attempting to use the DMA engine instead of instantly fail to CPU
copy.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The lower bits read from a BAR register will contain property bits
that we do not care about. Clear those so that we can use the BAR
values for limit and xlat registers.
Reported-by: Conrad Meyer <cem@freebsd.org>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The transmit overrun avoidance error path in ntb_process_tx accidentally
swapped the first two values being passed to the tx_handler client.
This could result in crashes in the ntb_netdev (or other out-of-tree NTB
clients).
Reported-by: Alex Depoutovitch <alex@pernixdata.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
resource_size_t may be 32-bit wide on some architectures, which causes
this warning when building the NTB code:
drivers/ntb/ntb_transport.c: In function 'ntb_transport_link_work':
drivers/ntb/ntb_transport.c:828:46: warning: right shift count >= width of type [-Wshift-count-overflow]
The warning is harmless but can be avoided by using the upper_32_bits()
macro.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Fixes: e26a5843f7 ("NTB: Split ntb_hw_intel and ntb_transport drivers")
Signed-off-by: Jon Mason <jdmason@kudzu.us>
There is no need for the upstream and downstream addresses to be different
for the NTB configs. Go to using a single set of address. It is still
possible to configure them differently using module parameter override
however.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Acked and Tested-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Order of operations issue with the QP Num and MW count, which would
result in the receive buffer pointer being invalid if there are more
than 1 MW. Corrected with parenthesis to enforce the proper order of
operations.
Reported-by: John I. Kading <John.Kading@gd-ms.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
These variables were not used anywhere. So remove them.
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
We were accessing nt->mw_vec after freeing it. Fix the error path so
that we free nt->mw_vec after we have finished using it.
Signed-off-by: Sudip Mukherjee <sudip@vectorindia.org>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
smatch detected an issue in the function ntb_transport_max_size() where
we could be dereferencing a dma channel pointer when it is NULL.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Check that b2b_mw_idx is in range of the number of memory windows when
initializing the device. The workaround is considered to be in effect
only if the device b2b_idx is exactly UINT_MAX, instead of any index
past the last memory window.
Only print B2B MW workaround information in debugfs if the workaround is
in effect.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Allocate two DMA channels, one for TX operation and one for RX
operation, instead of having one DMA channel for everything. This
provides slightly better performance, and also will make error handling
cleaner later on.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The dma_sync_wait can hurt the performance of workloads mixed with both
large and small frames. Large frames will be copied using the dma
engine. Small frames will be copied by the cpu. The dma_sync_wait
prevents the cpu and dma engine copying in parallel.
In the period where the cpu is copying, the dma engine is stopped. The
dma engine is not doing any useful work to copy large frames during that
time, and the additional time to restart the dma engine for the next
large frame. This will decrease the throughput for the portion of a
workload with large frames.
In the period where the dma engine is copying, the cpu is held up
waiting for dma to complete. The small frames processing will be
delayed until the dma is complete. The RX frames are completed
in-order, and the processing of small frames takes very little time, so
dma_sync_wait may have an insignificant impact on the respose time of
frames. The more significant impact is to the system, because the delay
in dma_sync_wait is implemented as busy non-blocking wait. This can
prevent the delayed core from doing any useful work, even if it could be
processing work for other drivers, unrelated to transport RX processing.
After applying the earlier patch to fix out-of-order RX acknoledgement,
the dma_sync_wait is no longer necessary. Remove it, so that cpu memcpy
will proceed immediately for small frames, in parallel with ongoing dma
for large frames. Do not hold up the cpu from doing work while dma is
in progress. The prior fix will continue to ensure in-order completion
of the RX frames to the upper layer, and in-order delivery of the RX
acknoledgement.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Make QP stats info more readable for debugging purposes. Also add an
entry to indicate whether DMA is being used.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The list should be added from the bottom and not the top in order to
ensure the transport is provided in the same order to clients as ntb
devices are discovered.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Adding PCI Device IDs for B2B (back to back), RP (root port, primary),
and TB (transparent bridge, secondary) devices.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Right now if we push the NTB really hard, we start dropping packets due
to not able to process the packets fast enough. We need to st:qop the
upper layer from flooding us when that happens.
A timer is necessary in order to restart the queue once the resource has
been processed on the receive side. Due to the way NTB is setup, the
resources on the tx side are tied to the processing of the rx side and
there's no async way to know when the rx side has released those
resources.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Avoid any chance of format string expansion when calling dev_set_name.
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Remove early dereference of a pointer that is checked later in the code.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
A plain 32 bit integer will overflow for values over 4GiB.
Change the plain integer size to the appropriate size type in
ntb_set_mw. Change the type of the size parameter and two local
variables used for size.
Even if there is no overflow, a size of zero is invalid here.
Reported-by: Juyoung Jung <jjung@micron.com>
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Schedule to receive on QP link up, to make sure that the doorbell is
properly cleared for interrupts.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
When the remote side is not up, we do not have all the context for the
transport, and that causes NULL ptr access. Have the debugfs reads check
to see if transport is up before we make access.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Currently the debugfs does not have files for all NTB transport queue
pairs. When there are multiple NTBs present in a system, the QP names
of the last transport clobber the names of previously added transport
QPs. Only the last added QPs can be observed via debugfs.
Create a directory per NTB transport to associate the QPs with that
transport. Name the directory the same as the PCI device.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
It was possible for a synchronous update of the RX index in the error
case to get ahead of the asynchronous RX index update in the normal
case. Change the RX processing to preserve an RX completion order.
There were two error cases. First, if a buffer is not present to
receive data, there would be no queue entry to preserve the RX
completion order. Instead of dropping the RX frame, leave the RX frame
in the ring. Schedule RX processing when RX entries are enqueued, in
case there are RX frames waiting in the ring to be received.
Second, if a buffer is too small to receive data, drop the frame in the
ring, mark the RX entry as done, and indicate the error in the RX entry
length. Check for a negative length in the receive callback in
ntb_netdev, and count occurrences as rx_length_errors.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
When split BAR is enabled, the driver needs to dump out the split BAR
registers rather than the original 64bit BAR registers.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The unsafe doorbell and scratchpad access should display reason when
WARN is called. Otherwise we get a stack dump without any explanation.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Printouts driver name and version to indicate what is being loaded.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Benchmarking showed a significant performance increase with the MTU size
to 64k instead of 16k. Change the driver default to 64k.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Instead of using the platform code names, use the correct platform names
to identify the respective Intel NTB hardware.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Disable DMA usage by default, since the CPU provides much better
performance with write combining. Provide a module parameter to enable
DMA usage when offloading the memcpy is preferred.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Changing the memory window BAR mappings to write combining significantly
boosts the performance. We will also use memcpy that uses non-temporal
store, which showed performance improvement when doing non-cached
memcpys.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Allocate memory and request the DMA channel for the same NUMA node as
the NTB device.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
When the ntb transport is connecting and waiting for the peer, the debug
console receives lots of debug level messages about the remote qp link
status being down. Rate limit those messages.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
This is a simple debugging driver that enables the doorbell and
scratch pad registers to be read and written from the debugfs. This
tool enables more complicated debugging to be scripted from user space.
This driver may be used to test that your ntb hardware and drivers are
functioning at a basic level.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
This is a simple ping pong driver that exercises the scratch pads and
doorbells of the ntb hardware. This driver may be used to test that
your ntb hardware and drivers are functioning at a basic level.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Add module parameters for the addresses to be used in B2B topology.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Reset the link stats when the link goes down. In particular, the TX and
RX index and count must be reset, or else the TX side will be sending
packets to the RX side where the RX side is not expecting them. Reset
all the stats, to be consistent.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
On link down, don't advance RX index to the next entry. The next entry
should never be valid after receiving the link down flag.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The same message "qp %d: Link Down\n" was printed at two locations in
ntb_transport. Change the messages so they are distinct.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Set errata flags for the specific device IDs to which they apply,
instead of the whole Xeon hardware class.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Link training should be enabled in the driver probe for root port mode.
We should not have to wait for transport to be loaded for this to
happen. Otherwise the ntb device will not show up on the transparent
bridge side of the link.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The transport was writing and then reading the peer scratch pad,
essentially reading what it just wrote instead of exchanging any
information with the peer. The transport expects the peer values to be
the same as the local values, so this issue was not obvious.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Change ntb_hw_intel to use the new NTB hardware abstraction layer.
Split ntb_transport into its own driver. Change it to use the new NTB
hardware abstraction layer.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Abstract the NTB device behind a programming interface, so that it can
support different hardware and client drivers.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
This patch only moves files to their new locations, before applying the
next two patches adding the NTB Abstraction layer. Splitting this patch
from the next is intended make distinct which code is changed only due
to moving the files, versus which are substantial code changes in adding
the NTB Abstraction layer.
Signed-off-by: Allen Hubbe <Allen.Hubbe@emc.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Commit ab760a0 (ntb: Adding split BAR support for Haswell platforms)
changed ntb_device's mw from a fixed-size array into a pointer that is
allocated based on limits.max_mw; however, on Atom platforms, max_mw
is not initialized until ntb_device_setup(), which happens after the
allocation.
Fill out max_mw in ntb_atom_detect() to match ntb_xeon_detect(); this
happens before the use of max_mw in the ndev->mw allocation.
Fixes a null pointer dereference on Atom platforms with ntb hardware.
v2: fix typo (mw_max should be max_mw)
Signed-off-by: Daniel Verkamp <daniel.verkamp@intel.com>
Acked-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The MW regbase and vbase(s) were not being freed if an error occurred
in the vbase allocation loop. This is corrected by updating the error
path for the allocation loop to err4.
Reported-by: Julia Lawall <julia.lawall@lip6.fr>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The PCI core now disables MSI and MSI-X for all devices during enumeration
regardless of CONFIG_PCI_MSI. Remove device-specific code to disable
MSI/MSI-X.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
On the Haswell platform, a split BAR option to allow creation of 2
32bit BARs (4 and 5) from the 64bit BAR 4. Adding support for this
new option.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Instead of using a module parameter, we should detect the errata via
PCI DID and then set an appropriate flag. This will be used for additional
errata later on.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
To simplify some of the platform detection code. Move the platform detection
to a function to be called earlier.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Move the platform detection function to separate functions to allow
easier maintenence.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
Create a debugfs entry for the NTB device to log the basic device info,
as well as display the error count on a number of registers.
Signed-off-by: Jon Mason <jon.mason@intel.com>
The NTB translate register must have the value to be BAR size aligned.
This alignment check make sure that the DMA memory allocated has the
proper alignment. Another requirement for NTB to function properly with
memory window BAR size greater or equal to 4M is to use the CMA feature
in 3.16 kernel with the appropriate CONFIG_CMA_ALIGNMENT and
CONFIG_CMA_SIZE_MBYTES set.
Signed-off-by: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Jon Mason <jdmason@kudzu.us>
The detection of an uneven number of queues on the given memory windows
was not correct. The mw_num is zero based and the mod should be
division to spread them evenly over the mw's.
Signed-off-by: Jon Mason <jon.mason@intel.com>
As result of deprecation of MSI-X/MSI enablement functions
pci_enable_msix() and pci_enable_msi_block() all drivers
using these two interfaces need to be updated to use the
new pci_enable_msi_range() or pci_enable_msi_exact()
and pci_enable_msix_range() or pci_enable_msix_exact()
interfaces.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
This is an cleanup effort to make ntb_setup_msix() more
readable - use ntb_setup_bwd_msix() to init MSI-Xs on
BWD hardware and ntb_setup_snb_msix() - on SNB hardware.
Function ntb_setup_snb_msix() also initializes MSI-Xs the
way it should has been done - looping pci_enable_msix()
until success or failure.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
In the code for Xeon devices in back-to-back mode with xeon_errata_workaround
disabled, the downstream device puts the wrong value in SNB_B2B_XLAT_OFFSETL
(SNB_MBAR01_DSD_ADDR vs. SNB_MBAR01_USD_ADDR).
This was spotted while reading code, since the typo has no practical effect,
at least for now: the low 32 bits of both constants are actually identical
anyway. However, it's clearer and safer to use the right name.
Signed-off-by: Roland Dreier <roland@purestorage.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
NTB-RP Link Up issue, Xeon Doorbell errata workaround, ntb_transport
link down race, and correct dmaengine_get/put usage. Also, clean-ups
to remove duplicate defines and document a hardware errata. Finally,
some changes to improve performance.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)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=nOG2
-----END PGP SIGNATURE-----
Merge tag 'ntb-3.13' of git://github.com/jonmason/ntb
Pull non-transparent bridge updates from Jon Mason:
"NTB driver bug fixes to address a missed call to pci_enable_msix,
NTB-RP Link Up issue, Xeon Doorbell errata workaround, ntb_transport
link down race, and correct dmaengine_get/put usage.
Also, clean-ups to remove duplicate defines and document a hardware
errata. Finally, some changes to improve performance"
* tag 'ntb-3.13' of git://github.com/jonmason/ntb:
NTB: Disable interrupts and poll under high load
NTB: Enable Snoop on Primary Side
NTB: Document HW errata
NTB: remove duplicate defines
NTB: correct dmaengine_get/put usage
NTB: Fix ntb_transport link down race
ntb: Fix missed call to pci_enable_msix()
NTB: Fix NTB-RP Link Up
NTB: Xeon Doorbell errata workaround
Enable Snoop from Primary to Secondary side on BAR23 and BAR45 on all
TLPs. Previously, Snoop was only enabled from Secondary to Primary
side. This can have a performance improvement on some workloads.
Also, make the code more obvious about how the link is being enabled.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Remove duplicate defines in drivers/ntb/ntb_regs.h
Signed-off-by: Michael Opdenacker <michael.opdenacker@free-electrons.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
dmaengine_get() causes the initialization of the per-cpu channel tables.
It needs to be called prior to dma_find_channel().
Initial version by Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
A WARN_ON is being hit in ntb_qp_link_work due to the NTB transport link
being down while the ntb qp link is still active. This is caused by the
transport link being brought down prior to the qp link worker thread
being terminated. To correct this, shutdown the qp's prior to bringing
the transport link down. Also, only call the qp worker thread if it is
in interrupt context, otherwise call the function directly.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Current MSI-X enablement code assumes MSI-Xs were successfully
allocated in case less than requested vectors were available.
That assumption is wrong, since MSI-Xs should be enabled with
a repeated call to pci_enable_msix(). This update fixes this.
Signed-off-by: Alexander Gordeev <agordeev@redhat.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
The Xeon NTB-RP setup, the transparent side does not get a link up/down
interrupt. Since the presence of a NTB device on the transparent side
means that we have a NTB link up, we can work around the lack of an
interrupt by simply calling the link up function to notify the upper
layers.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Modifications to the 14th bit of the B2BDOORBELL register will not be
mirrored to the remote system due to a hardware issue. To get around
the issue, shrink the number of available doorbell bits by 1. The max
number of doorbells was being used as a way to referencing the Link
Doorbell bit. Since this would no longer work, the driver must now
explicitly reference that bit.
This does not affect the xeon_errata_workaround case, as it is not using
the b2bdoorbell register.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Use the generic unmap object to unmap dma buffers.
Cc: Vinod Koul <vinod.koul@intel.com>
Cc: Tomasz Figa <t.figa@samsung.com>
Cc: Dave Jiang <dave.jiang@intel.com>
Signed-off-by: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Acked-by: Jon Mason <jon.mason@intel.com>
[djbw: fix up unmap len, and GFP flags]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
PCI core will initialize device MSI/MSI-X capability in
pci_msi_init_pci_dev(). So device driver should use
pci_dev->msi_cap/msix_cap to determine whether the device
support MSI/MSI-X instead of using
pci_find_capability(pci_dev, PCI_CAP_ID_MSI/MSIX).
Access to PCIe device config space again will consume more time.
Signed-off-by: Yijing Wang <wangyijing@huawei.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
Add support for Non-Transparent Bridge connected to a PCI-E Root Port on
the remote system (also known as NTB-RP mode). This allows for a NTB
enabled system to be connected to a non-NTB enabled system/slot.
Modifications to the registers and BARs/MWs on the Secondary side by the
remote system are reflected into registers on the Primary side for the
local system. Similarly, modifications of registers and BARs/MWs on
Primary side by the local system are reflected into registers on the
Secondary side for the Remote System. This allows communication between
the 2 sides via these registers and BARs/MWs.
Note: there is not a fix for the Xeon Errata (that was already worked
around in NTB-B2B mode) for NTB-RP mode. Due to this limitation, NTB-RP
will not work on the Secondary side with the Xeon Errata workaround
enabled. To get around this, disable the workaround via the
xeon_errata_workaround=0 modparm. However, this can cause the hang
described in the errata.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Many variable names in the NTB driver refer to the primary or secondary
side. However, these variables will be used to access the reverse case
when in NTB-RP mode. Make these names more generic in anticipation of
NTB-RP support.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Allocate and use a DMA engine channel to transmit and receive data over
NTB. If none is allocated, fall back to using the CPU to transfer data.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Correct the issues on NTB that prevented it from working on x86_32 and
modify the Kconfig to allow it to be permitted to be used in that
environment as well.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Add support for new Intel NTB devices on upcoming Xeon hardware. Since
the Xeon hardware design is already in place in the driver, all that is
needed are the new device ids.
Remove the device IDs for NTB devs running in Transparent Bridge mode,
as this driver is not being used for those devices.
Rename the device IDs for NTB devs running in NTB-RP mode to better
identify their usage model. "PS" to denote the Primary Side of NTB, and
"SS" to denote the secondary side. The primary side is the interface
exposed to the local system, and the secondary side is the interface
exposed to the remote system.
Signed-off-by: Jon Mason <jon.mason@intel.com>
The BWD NTB device will drop the link if an error is encountered on the
point-to-point PCI bridge. The link will stay down until all errors are
cleared and the link is re-established. On link down, check to see if
the error is detected, if so do the necessary housekeeping to try and
recover from the error and reestablish the link.
There is a potential race between the 2 NTB devices recovering at the
same time. If the times are synchronized, the link will not recover and the
driver will be stuck in this loop forever. Add a random interval to the
recovery time to prevent this race.
Signed-off-by: Jon Mason <jon.mason@intel.com>
There is a Xeon hardware errata related to writes to SDOORBELL or
B2BDOORBELL in conjunction with inbound access to NTB MMIO Space, which
may hang the system. To workaround this issue, use one of the memory
windows to access the interrupt and scratch pad registers on the remote
system. This bypasses the issue, but removes one of the memory windows
from use by the transport. This reduction of MWs necessitates adding
some logic to determine the number of available MWs.
Since some NTB usage methodologies may have unidirectional traffic, the
ability to disable the workaround via modparm has been added.
See BF113 in
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-c5500-c3500-spec-update.pdf
See BT119 in
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-e5-family-spec-update.pdf
Signed-off-by: Jon Mason <jon.mason@intel.com>
Debugfs was setup in NTB to only have a single debugfs directory. This
resulted in the leaking of debugfs directories and files when multiple
NTB devices were present, due to each device stomping on the variables
containing the previous device's values (thus preventing them from being
freed on cleanup). Correct this by creating a secondary directory of
the PCI BDF for each device present, and nesting the previously existing
information in those directories.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Due to ambiguous documentation, the USD/DSD identification is backward
when compared to the setting in BIOS. Correct the bits to match the
BIOS setting.
Signed-off-by: Jon Mason <jon.mason@intel.com>
The NTB Xeon hardware has 16 scratch pad registers and 16 back-to-back
scratch pad registers. Correct the #define to represent this and update
the variable names to reflect their usage.
Signed-off-by: Jon Mason <jon.mason@intel.com>
If an error is encountered in ntb_device_setup, it is possible that the
spci_cmd isn't populated. Writes to the offset can result in a NULL
pointer dereference. This issue is easily encountered by running in
NTB-RP mode, as it currently is not supported and will generate an
error. To get around this issue, return if an error is encountered
prior to attempting to write to the spci_cmd offset.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Fix issue with adding multiple ntb client devices to the ntb virtual
bus. Previously, multiple devices would be added with the same name,
resulting in crashes. To get around this issue, add a unique number to
the device when it is added.
Signed-off-by: Jon Mason <jon.mason@intel.com>
The system will appear to lockup for long periods of time due to the NTB
driver spending too much time in memcpy. Avoid this by reducing the
number of packets that can be serviced on a given interrupt.
Signed-off-by: Jon Mason <jon.mason@intel.com>
The ring logic of the NTB receive buffer/transmit memory window requires
there to be at least 2 payload sized allotments. For the minimal size
case, split the buffer into two and set the transport_mtu to the
appropriate size.
Signed-off-by: Jon Mason <jon.mason@intel.com>
If the NTB link toggles, the driver could stop receiving due to the
tx_index not being set to 0 on the transmitting size on a link-up event.
This is due to the driver expecting the incoming data to start at the
beginning of the receive buffer and not at a random place.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Each link-up will allocate a new NTB receive buffer when the NTB
properties are negotiated with the remote system. These allocations did
not check for existing buffers and thus did not free them. Now, the
driver will check for an existing buffer and free it if not of the
correct size, before trying to alloc a new one.
Signed-off-by: Jon Mason <jon.mason@intel.com>
64bit BAR sizes are permissible with an NTB device. To support them
various modifications and clean-ups were required, most significantly
using 2 32bit scratch pad registers for each BAR.
Also, modify the driver to allow more than 2 Memory Windows.
Signed-off-by: Jon Mason <jon.mason@intel.com>
->remote_rx_info and ->rx_info are struct ntb_rx_info pointers. If we
add sizeof(struct ntb_rx_info) then it goes too far.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
These tests are off by one. If "mw" is equal to NTB_NUM_MW then we
would go beyond the end of the ndev->mw[] array.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
Correct instances of variable dereferencing before checking its value on
the functions exported to the client drivers. Also, add sanity checks
for all exported functions.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Jon Mason <jon.mason@intel.com>
Address the sparse warnings and resulting fallout
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Atomic readq and writeq do not exist by default on some 32bit
architectures, thus causing compile errors due to non-existent symbols.
Since NTB has not been tested 32bit, disable x86_32 support until such
time as this and any other issues can be properly discovered.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If the NTB client driver enqueues the maximum number of rx buffers, it
will not be able to re-enqueue another in its callback handler due to a
lack of free entries. This can be avoided by adding the current entry
to the free queue prior to calling the client callback handler. With
this change, ntb_netdev will no longer encounter a rx error on its first
packet.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
CPU reads across NTB are slow(er) and can hang the local system if an
ECC error is encountered on the remote. To work around the need for a
read, have the remote side write its current position in the rx buffer
to the local system's buffer and use that to see if there is room when
transmitting.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Correct gcc warning of using too much stack debugfs_read. This is done
by kmallocing the buffer instead of using the char array on stack.
Also, shrinking the buffer to something closer to what is currently
being used.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Declare ntb_bus_type static to remove it from name space, and remove
unused ntb_get_max_spads function. Found via `make namespacecheck`.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use simple_open for debugfs instead of recreating it in the NTB driver.
Caught by coccicheck.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Move all cancel_delayed_work_sync to a work thread to prevent sleeping
in interrupt context (when the NTB link goes down). Caught via
'Sleep inside atomic section checking'
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since it is possible for the memory windows on the two NTB connected
systems to be different sizes, the divergent sizes must be accounted for
in the segmentation of the MW's on each side. Create separate size
variables and initialization as necessary.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
mmiowb is not sufficient to flush the data and is causing data
corruption. Change to wmb and the data corruption is no more.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Attempts to probe client ntb drivers without ntb hardware present will
result in null pointer dereference due to the lack of the ntb bus device
being registers. Check to see if this is the case, and fail all calls
by the clients registering their drivers.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
These are now gone from the kernel, so remove them from the newly-added
drivers before they start to cause build errors for people.
Cc: Jon Mason <jon.mason@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A PCI-Express non-transparent bridge (NTB) is a point-to-point PCIe bus
connecting 2 systems, providing electrical isolation between the two subsystems.
A non-transparent bridge is functionally similar to a transparent bridge except
that both sides of the bridge have their own independent address domains. The
host on one side of the bridge will not have the visibility of the complete
memory or I/O space on the other side of the bridge. To communicate across the
non-transparent bridge, each NTB endpoint has one (or more) apertures exposed to
the local system. Writes to these apertures are mirrored to memory on the
remote system. Communications can also occur through the use of doorbell
registers that initiate interrupts to the alternate domain, and scratch-pad
registers accessible from both sides.
The NTB device driver is needed to configure these memory windows, doorbell, and
scratch-pad registers as well as use them in such a way as they can be turned
into a viable communication channel to the remote system. ntb_hw.[ch]
determines the usage model (NTB to NTB or NTB to Root Port) and abstracts away
the underlying hardware to provide access and a common interface to the doorbell
registers, scratch pads, and memory windows. These hardware interfaces are
exported so that other, non-mainlined kernel drivers can access these.
ntb_transport.[ch] also uses the exported interfaces in ntb_hw.[ch] to setup a
communication channel(s) and provide a reliable way of transferring data from
one side to the other, which it then exports so that "client" drivers can access
them. These client drivers are used to provide a standard kernel interface
(i.e., Ethernet device) to NTB, such that Linux can transfer data from one
system to the other in a standard way.
Signed-off-by: Jon Mason <jon.mason@intel.com>
Reviewed-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Logan Gunthorpe