The driver will be able to send notification events towards
a user process, using user's registered event file descriptor.
The driver uses the notification mechanism to inform the
user about an occurred event.
A user thread can wait until a notification is received from
the driver.
The driver stores the occurred event until the user reads it,
using HL_INFO_GET_EVENTS - new ioctl opcode in the INFO ioctl.
Gaudi specific implementation includes sending a notification
on a TPC assertion event that is received from f/w.
Signed-off-by: Tal Cohen <talcohen@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Scenario:
1. During hard reset, driver executes device_kill_open_processes.
2. Drivers file descriptor is not closed yet (user process is alive),
hence we are starting loop on all open file descriptors.
3. Just before getting task struct of user process, according to
pid, SIGKILL is sent to the user process, hence get_pid_task
fails, driver prints a warning and device_kill_open_processes
returns an error.
4. Returned error causing driver fini do disable the device object
of the process which causes a kernel crash.
The fix is to handle this case not as an error and continue fini flow
as normal, since the killed process (by the SIGKILL) will release its
resources just like it will do when the driver sends him the sigkill.
Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With the new code required for the flow added, we can now switch
to using the new memory manager infrastructure, removing the old code.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The new unified memory manager uses page offset to pass buffer handle
during the mmap operation. One problem with this approach is that it
requires the handle to always be divisible by the page size, else, the
user would not be able to pass it correctly as an argument to the mmap
system call.
Previously, this was achieved by shifting the handle left after alloc
operation, and shifting it right before get operation. This was done in
the user code. This creates code duplication, and, what's worse,
requires some knowledge from the user regarding the handle internal
structure, hurting the encapsulation.
This patch encloses all the page shifts inside memory manager functions.
This way, the user can take the handle as a black box, and simply use
it, without any concert about how it actually works.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The habanalabs HW requires memory resources to be used by its
internal hardware structures. These structures are allocated and
initialized by the driver. We would like to use the device HBM for
that purpose. This memory is io-remapped and accessed using the
writel()/writeb()/writew() commands.
Since some of the HW structures are one byte in size we need to
add support for the writeb() and readb() functions in the driver.
Signed-off-by: Moti Haimovski <mhaimovski@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Instead of using for_each_sg when iterating sgt that contains dma
entries, use the more proper for_each_sgtable_dma_sg macro.
In addition, both Goya and Gaudi have the exact same implementation
of the asic function that encapsulate the usage of this macro, so
it is better to move that implementation to the common code.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Only a hard-reset is an unexpected event which should be notify in
the kernel log. Other resets are normal operations and therefore
we should not pollute the log with them.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When accessing the configuration registers through debugfs,
it is only allowed to access aligned address.
Fail if address is not aligned.
Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is a preparation for unifying the code of accessing device memory
through debugfs. Add struct fields and callbacks that will later
be used in debugfs code and will reduce code duplication
among the different read{32,64}/write{32,64} callbacks of
every asic.
Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With the introduction of the unified memory manager infrastructure, the
timestamp buffers can be converted to use it.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In order to be more informative during device open, we are adding a
new return code -EAGAIN that indicates device is still going through
resource reclaiming and hence it cannot be used yet.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Several H/W events can be sent adjacently, even due to a single error.
If a hard-reset is triggered as part of handling one of these events,
the following events won't be handled.
The debug info from these missed events is important, sometimes even
more important than the one that was handled.
To allow handling these close events, add an option to delay a device
reset and use it when resetting due to H/W events.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In case user application was interrupted while some cs still in-flight
or in the middle of completion handling in driver, the
last refcount of the kernel private data for the user process
will not be put in the fd close flow, but in the cs completion
workqueue context.
This means that the device reset-upon-device-release will be called
from that context. During the reset flow, the driver flushes all the cs
workqueue to ensure that any scheduled work has run to completion,
and since we are running from the completion context we will
have deadlock.
Therefore, we need to skip flushing the workqueue in those cases.
It is safe to do it because the user won't be able to release the device
unless the workqueues are already empty.
Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
For current devices there is a need to send the max power value to F/W
during device init, for example because there might be several card
types.
In future devices, this info will be programmed in the device's EEPROM
and will be read by F/W, and hence the driver should not send it.
Modify the sending of the relevant message to be done only for ASIC
types that need it.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The max_power variable which is used for calculating the device
utilization is the ASIC specific property which is set during init.
However, the max value can be modified via sysfs, and thus the updated
value in the device structure should be used instead.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Timestamp registration API allows the user to register
a timestamp record event which will make the driver set
timestamp when CQ counter reaches the target value
and write it to a specific location specified
by the user.
This is a non blocking API, unlike the wait_for_interrupt
which is a blocking one.
Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In case of soft reset failure, hard reset should be initiated, but
reset flags were not set to enable it, which caused another soft reset
followed by another failure.
Updated reset flags to enable hard reset flow in case of soft reset
failure.
Signed-off-by: Dani Liberman <dliberman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
For better maintainability, try to concentrate all the common functions
that communicate with the f/w in firmware_if.c
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Now that clock gating is permanently disabled in GAUDI, no need for
the ASIC functions of setting and disabling clock gating, as this
was a unique scenario in GAUDI.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
As hard-reset can be requested during soft-reset, driver must allow
it or else critical events received during soft-reset will be
ignored.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Atomic operations during reset are replaced by a spinlock in order
to have the ability to protect more than a single variable.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Unify variables related to device reset, which will help us to
add some new reset functionality in future patches.
Signed-off-by: Ofir Bitton <obitton@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>
The reset flags used by the reset thread are currently a mix of
hard-coded values and a specific flag which is passed from the context
that initiates the reset.
To make it easier to pass more flags in future from this context to the
reset thread, modify it to pass all the original reset flags to the
thread.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Driver should handle events during soft-reset as F/W is not
going through reset and it keeps sending events towards host.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The ASIC-specific soft_reset_late_init() is now called after either
soft-reset or reset-upon-device-release. Therefore, it needs a more
appropriate name.
No need to split it to two functions, as an ASIC either supports
soft-reset or reset-upon-device-release.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reset upon device release is not a soft-reset from user/system point
of view. As such, we shouldn't count that reset in the statistics we
gather and expose to the monitoring applications.
We also shouldn't print soft-reset when doing the reset upon device
release.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Changing the frequency automatically is only done in Goya. In future
ASICs this is done inside the firmware. Therefore, move the common code
into the Goya specific files.
Main changes as part of the commit are:
1. The thread for setting frequency is moved from device_late_init
to goya_late_init
2. hl_device_set_frequency is removed from hl_device_open as it is
not relevant for other ASICs and for Goya it is taken care by
the thread
3. hl_device_set_frequency is renamed as goya_set_frequency
Signed-off-by: Rajaravi Krishna Katta <rkatta@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Hard-reset is mutually exclusive with reset-on-device-release.
Therefore, if such a request arrives to the reset function, abort
the reset and return an error to the callee.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
If a device reset has started, there is a chance that the heartbeat
function will fail because the device is disabled at the beginning
of the reset function.
In that case, we don't want the error message to appear in the log.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In addition to the clock throttling reason, user should be able
to obtain also the start time and the duration of the throttling
event.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Up until now the driver stored indication if Linux was loaded on the
device CPU. This was needed in order to coordinate some tasks that are
performed by the Linux.
In future ASICs, many of those tasks will be performed by the boot
fit, so now we need the same indication of boot fit load status.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
There may be a situation where drivers receives continuous fatal H/W
error events from FW immediately post reset cycle.
This may be due to some fault on the silicon itself.
In such case its better to bypass reset cycle so we won't be stuck in
endless loop of resets.
This commit bypasses reset request in case driver received two back to
back FW fatal error before first occurrence of heartbeat event.
Signed-off-by: Bharat Jauhari <bjauhari@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The driver allows only a single process to open a device's FD at any
single time. This is done by checking "hdev->compute_ctx" under mutex.
Therefore, to prevent a race between the moment a user closes it's FD
and when another user tries to open the device, we need to make sure
that clearing this variable is the very last thing that is done in the
code of the FD's release.
I'm moving the idle check before clearing this variable and the
"reset on device release". btw, if the reset happens it will prevent
any other user from opening the device until the reset is finished.
An important thing to note is that we need to remove the user process
that is closing the device from the process list BEFORE calling the
reset function. That is to prevent a case where the reset code will
try to kill that user process and it is unnecessary as the process
doesn't hold any device/driver resources anymore.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reset to the device is not necessarily due to an error, so print it
as info instead of error.
In addition, print the type of reset we are doing:
- reset of the entire device (aka hard reset)
- reset of the device after user have released it (less than hard reset)
- lighter reset of an inference device (aka soft reset)
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Soft-reset is the procedure where we reset only the compute/DMA engines
of the device, without requiring the current user-space process to
release the device.
This type of reset can happen if TDR event occurred (a workload got
stuck) or by a root request through sysfs.
This is only relevant for inference ASICs, as there is no real-world
use-case to do that in training, because training runs on multiple
devices.
In addition, we also do (in certain ASICs) a reset upon device release.
That reset uses the same code as the soft-reset.
Therefore, to better differentiate between the two resets, it is better
to rename the soft-reset support as "inference soft-reset", to make
the code more self-explanatory.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
When the f/w runs in secured mode, it can reset the ASIC when certain
events occur. In unsecured mode, the driver asks the f/w to reset the
ASIC for those events.
We need to perform the entire reset procedure but without accessing the
ASIC. i.e. without halting the engines and without sending messages
to the f/w.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
On init, the disabled state is cleared right before hw_init and that
causes the device to report on "Operational" state before the device
initialization is finished. Although the char device is not yet exposed
to the user at this stage, the sysfs entries are exposed.
This can cause errors in monitoring applications that use the sysfs
entries.
In order to avoid this, a new state "in device creation" is introduced
to ne reported when the device is not disabled but is still in init
flow.
Signed-off-by: Omer Shpigelman <oshpigelman@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>
The driver quietly handles memory mappings that were not freed so no
need to print a warning about that when user closes the FD.
Accordingly, revise the text that is printed in case the device is
still in use after the user process closed the FD.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
There is a scenario where an ongoing soft reset would race with an
ongoing heartbeat routine, eventually causing heartbeat to fail and
thus to escalate into a hard reset.
With this fix, soft-reset procedure will disable heartbeat CPU messages
and flush the (ongoing) current one before continuing with reset code.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
If hard reset fails after the call to hw_fini and before loading the
linux image to the device, a subsequent call to hw_fini should
communicate via COMMS (or MSG_TO_CPU regs for old FW versions).
However, the driver still tries in this case to communicate via the GIC,
and thus no hard reset is actually done.
To avoid that, the patch clears the linux_loaded flag after every call
to hw_fini.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
To improve the user's ability to debug the case where a workload that
is part of executing training/inference of a topology is getting stuck,
we need to add a 'core dump' each time a CS times-out. The 'core dump'
shall contain all relevant Sync Manager information and corresponding
fence values.
The most recent dumps shall be accessible via debugfs, under
'state_dump' node. Reading from the node will provide the oldest dump
available. Writing an integer value X will discard X dumps, starting
with the oldest one, i.e. subsequent read will now return newer
dumps.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The previous function we used, find_get_pid(), wasn't good in case
the user process was run inside docker.
As a result, we didn't had the PID and we couldn't kill the user
process in case the device got stuck and we needed to reset the
device.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In device fini there was missing a call to release all pending user
interrupts. That can cause a process to be stuck inside the driver's
IOCTL of wait for interrupts, in case the device is removed or
simulator is killed at the same time.
In addition, also call to remove inactive codec job was missing.
Moreover, to prevent such errors in the future (where code is added
to reset path but not to device fini), we moved some common parts
to two dedicated functions:
cleanup_resources
take_release_locks
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
