Reserved encapsulated signals which were not released hold the context
refcount, leading to a failure when killing the user process on device
reset or device fini.
Add the release of these left signals in the CS roll-back process.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
hl_encaps_handle_do_release_sob() can be called only when the last
reference to the context object is released and hl_ctx_do_release() is
initiated, and therefore it shouldn't call hl_ctx_put().
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Add the ASIC-specific code for Gaudi2. Supply (almost) all of the
function callbacks that the driver's common code need to initialize,
finalize and submit workloads to the Gaudi2 ASIC.
It also contains the code to initialize the F/W of the Gaudi2 ASIC
and to receive events from the F/W.
It contains new debugfs entry to dump razwi events. razwi is a case
where the device's engines create a transaction that reaches an
invalid destination.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In future ASICs, it would be possible to have a non-idle
device when context is released. We thus need to postpone the
scrubbing. Postpone it to hpriv release if reset is not executed
or to device late init if reset is executed.
Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
We use scrub_device_mem only to scrub the entire SRAM and entire
DRAM. Therefore there is no need to send addr and size
args to the callback.
Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
hl_get_compute_ctx() is used to get the pointer to the compute context
from the hpriv object.
The function is called in code paths that are not necessarily initiated
by user, so it is possible that a context release process will happen in
parallel.
This can lead to a race condition in which hl_get_compute_ctx()
retrieves the context pointer, and just before it increments the context
refcount, the context object is released and a freed memory is accessed.
To avoid this race, add a mutex to protect the context pointer in hpriv.
With this lock, hl_get_compute_ctx() will be able to detect if the
context has been released or is about to be released.
struct hl_ctx_mgr has a mutex for contexts IDR with a similar "ctx_lock"
name, so rename it to just "lock" to avoid a confusion with the new
lock.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Often, the user is not interested in the completion timestamp of all
command submissions.
A common situation is, for example, when the user submits a burst of,
possibly, several thousands of commands, then request the completion
timestamp of only couple of specific key commands from all the burst.
The problem is that currently, the outcome of the early commands may be
lost, due to a large amount of later commands, that the user does not
really care about.
This patch creates a separate store with the outcomes of commands the
user has mark explicitly as interested in. This store does not mix the
marked commands with the unmarked ones, hence the data there will
survive for much longer.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Need to allow user retrieve data during reset and afterwards without
the need to reopen the device.
Did it by seperating the user peocesses list into two lists:
1. fpriv_list which contains list of user processes that opened
the device (currently only one).
2. fpriv_ctrl_list which contains list of user processes that opened
the control device. This processes in this list shall not be
killed during reset, only when the device is suddenly removed from
PCI chain.
Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
It was an error to save the compute context's pointer in the device
structure, as it allowed its use without proper ref-cnt.
Change the variable to a flag that only indicates whether there is
an active compute context. Code that needs the pointer will now
be forced to use proper internal APIs to get the pointer.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
There are multiple places where the code needs to get the context's
pointer and increment its ref cnt. This is the proper way instead
of using the compute context pointer in the device structure.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Pass the user's context pointer into the etr configuration function
to extract its ASID.
Using the compute_ctx pointer is an error as it is just an indication
of whether a user has opened the compute device.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Compute context pointer in hdev shouldn't be used for fetching the
context's pointer.
If an object needs the context's pointer, it should get it while
incrementing its kref, and when the object is released, put it.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The driver supports only a single context. Therefore, no need to check
if the user context that is closed is the compute context. The user
context, if exists, is always the compute context.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
No need to check the return value if the following action is the same for
both cases. In addition, now that hl_ctx_free() doesn't print if the
context is not released, its name can be misleading as the context might
stay alive after it is executed with no indication for that.
Hence we can discard it and simply put the refcount.
Signed-off-by: Omer Shpigelman <oshpigelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
This warning is redundant as we will print a notice in case the device
is still in use after the FD was closed. No need to print the same
message per context.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
This commit is the second part of the encapsulated signals feature.
It contains the driver support for submission of cs with encapsulated
signals and the wait for them.
Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The signaling from within encapsulated OP capability is merged into the
existing stream architecture, such that one can trigger multiple
signaling from an encapsulated op, according to the time the event
was done in the graph execution and avoid the need to wait for the
whole encapsulated OP execution to be complete before the stream can
signal.
This commit implements only the reserve/unreserve part.
Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
When user sends multiple CSs, waiting for each CS is not efficient
as it involves many user-kernel context switches.
In order to address this issue we add support to "wait on multiple CSs"
using a new uAPI which can wait on maximum of 32 CSs. The new uAPI is
defined using a new flag - WAIT_FOR_MULTI_CS - in the wait_for_cs IOCTL.
The input parameters for this uAPI will be:
@seq: user pointer to an array of up to 32 CS's sequence numbers.
@seq_array_len: length of sequence array.
@timeout_us: timeout for waiting for any CS.
The output paramateres for this API will be:
@status: multi CS ioctl completion status (dedicated status was added as
well).
@flags: bitmap of output flags of the CS.
@cs_completion_map: bitmap for multi CS, if CS sequence that was placed
in index N in input seq array has completed- the N-th
bit in cs_completion_map will be 1, otherwise it will
be 0.
@timestamp_nsec: timestamp of the first completed CS
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
To add proper support for wait-for-multi-CS, locking the CS lock
for each CS fence in the list is not efficient.
Instead, this patch add support to lock the CS lock once to get all
required fences.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
If device is not idle after user closes the FD we must reset device
as next user that will try to open FD will encounter a non-functional
device.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Improve "vm" debugfs node to print also the virtual addresses which are
currently mapped to HW blocks in the device.
Signed-off-by: Sagiv Ozeri <sozeri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Currently this API uses single 64 bits mask for engines idle indication.
Recently, it was observed that more bits are needed for some ASICs.
This patch modifies the use of the idle mask and the idle_extensions
mask.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In order to support scnenarios in which driver needs access to
HW components but it cannot access them directly, we add support for
scheduling command buffers internally.
These command buffers will be transmitted upon next user command
submission context.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In order for reserving VA ranges for kernel memory, we need
to allow the VM module to be initiated with kernel context.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In cases of multi-tenants, administrators may want to prevent data
leakage between users running on the same device one after another.
To do that the driver can scrub the internal memory (both SRAM and
DRAM) after a user finish to use the memory.
Because in GAUDI the driver allows only one application to use the
device at a time, it can scrub the memory when user app close FD.
In future devices where we have MMU on the DRAM, we can scrub the DRAM
memory with a finer granularity (page granularity) when the user
allocates the memory.
This feature is not supported in Goya.
To allow users that want to debug their applications, we add a kernel
module parameter to load the driver with this feature disabled.
Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Implement sync stream collective for GAUDI. Need to allocate additional
resources for that and add ctx_fini() to clean up those resources.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The device should be idle after a context is closed. If not, print a
notice.
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
During debugging of error we sometimes need to know whether the error
happened when a user context was open. Add debug prints when opening and
closing user contexts.
Reviewed-by: Tomer Tayar <ttayar@habana.ai>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
There are cases in which the device should access the host memory of a
CB through the device MMU, and thus this memory should be mapped.
The patch adds a flag to the CB IOCTL, in which a user can ask the
driver to perform the mapping when creating a CB.
The mapping is allowed only if a dedicated VA range was allocated for
the specific ASIC.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Add a missing free of the cs_pending array in the error flow of context
initialization.
Fixes: c16d45f42b ("habanalabs: Use pending CS amount per ASIC")
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Create a device MMU-mapped internal command buffer pool, in order to allow
the driver to allocate CBs for the signal/wait operations
that are fetched by the queues when they are configured with the user's
address space ID.
We must pre-map this internal pool due to performance issues.
This pool is needed for future ASIC support and it is currently unused in
GOYA and GAUDI.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
For internal needs of our CI we need to move all the common code into a
common folder instead of putting them in the root folder of the driver.
Same applies to the common header files under include/
Signed-off-by: Oded Gabbay <oded.gabbay@gmail.com>
Reviewed-by: Omer Shpigelman <oshpigelman@habana.ai>
