IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
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>
Currently the SOB reset was in fence release function which happens
only at the CS wraparound during the CS allocation time.
In order to support the new encapsulated signals reservation feature,
we need to move the SOB reset to an earlier phase because this SOB
could reach it's max value very fast using the signal reservation.
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>
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>
Print the SM name instead of index because it is more informational for
the user to know the SM name instead of id when a SM interrupt occurs.
In addition, the index that is printed is of the SOB group, not
a specific SOB.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
State dump is relevant to the user in case of Sync Manager error, so
we need to trigger it in that case as well.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Each ASIC can have a different offset to add to a host dma address,
to enable the ASIC to access that host memory.
The usage for this can be common code so add this to the asic
property structure.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Recently, the size parameter in userptr structure was change to u64.
As a result, we need to change the type of the local range_size
in device_va_to_pa() to u64 to avoid overflow.
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>
In case of host-resident MMU, when the page tables pool is destroyed,
its pointer is not nullified correctly.
As a result, on a device fini which happens after a failing reset, the
already destroyed pool is accessed, which leads to a kernel panic.
The patch fixes the setting of the pool pointer to NULL.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
This function will be used for more mmap operations than just
mmaping CBs.
Signed-off-by: Zvika Yehudai <zyehudai@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
At the first stage, only gaudi core dump shall be implemented, not
including the status registers.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
With the infrastructure in place, monitors and fences dump shall be
implemented.
Signed-off-by: Yuri Nudelman <ynudelman@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>
Sometimes we may need to disable optimization of using huge pages
in our memory management code. Add such a flag to the function that
creates the list of physical pages that would be programmed into the
device MMU.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Increase the size variable in the userptr structure to 64-bit. That
variable describes the size of the memory allocation of the user that
is now being mapped into the device. The mapping can be larger than
4GB, so we need to support it.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Same as we handle it in the regular wait for CS, we need to handle the
case where the waiting for user interrupt was interrupted. In that case,
we need to return correct error code to the user.
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>
In case user interrupt arrived but the completion value is less than
the target value, we want to retry the wait.
However, before the retry we must reinitialize the completion object,
under spin-lock, so the wait function won't exit immediately because
the completion object is already completed (from the previous
interrupt).
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Update recent changes made in firmware header files, which contain
a minor COMMS protocol change and new error status definitions.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
A new user flag is required to make memory map hint mandatory, in
contrast to the current situation where it is best effort.
This is due to the requirement to map certain data to specific
pre-determined device virtual address ranges.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Add support for pre-determined driver-reserved device VA address ranges.
This is needed for future ASIC support where some contents must be
mapped into these pre-determined ranges because the H/W will be
configured using these ranges.
In case the user asks to map a VA without a hint address, avoid
allocating the device VA from the reserved ranges.
Make sure the validation checks of the hint address take into account
situation where the DRAM page size is not pow of 2.
Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
I got a NULL pointer dereference report when doing fuzz test:
Call Trace:
qp_release_pages+0xae/0x130
qp_host_unregister_user_memory.isra.25+0x2d/0x80
vmci_qp_broker_unmap+0x191/0x320
? vmci_host_do_alloc_queuepair.isra.9+0x1c0/0x1c0
vmci_host_unlocked_ioctl+0x59f/0xd50
? do_vfs_ioctl+0x14b/0xa10
? tomoyo_file_ioctl+0x28/0x30
? vmci_host_do_alloc_queuepair.isra.9+0x1c0/0x1c0
__x64_sys_ioctl+0xea/0x120
do_syscall_64+0x34/0xb0
entry_SYSCALL_64_after_hwframe+0x44/0xae
When a queue pair is created by the following call, it will not
register the user memory if the page_store is NULL, and the
entry->state will be set to VMCIQPB_CREATED_NO_MEM.
vmci_host_unlocked_ioctl
vmci_host_do_alloc_queuepair
vmci_qp_broker_alloc
qp_broker_alloc
qp_broker_create // set entry->state = VMCIQPB_CREATED_NO_MEM;
When unmapping this queue pair, qp_host_unregister_user_memory() will
be called to unregister the non-existent user memory, which will
result in a null pointer reference. It will also change
VMCIQPB_CREATED_NO_MEM to VMCIQPB_CREATED_MEM, which should not be
present in this operation.
Only when the qp broker has mem, it can unregister the user
memory when unmapping the qp broker.
Only when the qp broker has no mem, it can register the user
memory when mapping the qp broker.
Fixes: 06164d2b72 ("VMCI: queue pairs implementation.")
Cc: stable <stable@vger.kernel.org>
Reported-by: Hulk Robot <hulkci@huawei.com>
Reviewed-by: Jorgen Hansen <jhansen@vmware.com>
Signed-off-by: Wang Hai <wanghai38@huawei.com>
Link: https://lore.kernel.org/r/20210818124845.488312-1-wanghai38@huawei.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When scsi_dispatch_cmd was moved to scsi_lib.c and made static, some
compilers (i.e., at least gcc 8.4.0) decided to compile this
inline. This is a problem for lkdtm.ko, which inserted a kprobe
on this function for the SCSI_DISPATCH_CMD crashpoint.
Move this crashpoint one function up the call chain to
scsi_queue_rq. Though this is also a static function, it should never be
inlined because it is assigned as a structure entry. Therefore,
kprobe_register should always be able to find it.
Fixes: 82042a2cdb ("scsi: move scsi_dispatch_cmd to scsi_lib.c")
Acked-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Kevin Mitchell <kevmitch@arista.com>
Link: https://lore.kernel.org/r/20210819022940.561875-2-kevmitch@arista.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
General Electric Healthcare's PPD has a secondary processor from
NXP's Kinetis K20 series. That device has two SPI chip selects:
The main interface's behaviour depends on the loaded firmware
and is currently unused.
The secondary interface can be used to update the firmware using
EzPort protocol. This is implemented by this driver using the
kernel's firmware API. The firmware is being flashed into
non-volatile flash memory, so it is enough to flash it once
and not on every boot. Flashing will wear the flash memory
(it has a life time of at least 10k programming cycles). At
the same time only occasional FW updates are expected (like e.g.
a BIOS update). Thus the firmware update is triggered via sysfs
instead of doing it in the driver's probe routine like many
other drivers.
Signed-off-by: Sebastian Reichel <sebastian.reichel@collabora.com>
Link: https://lore.kernel.org/r/20210802172309.164365-4-sebastian.reichel@collabora.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
On Tegra186 and later, a portion of the SYSRAM may be reserved for use
by TZ. Non-TZ memory accesses to this portion, including speculative
accesses, trigger SErrors that bring down the system. This does also
happen in practice occasionally (due to speculative accesses).
To fix the issue, add a flag to the SRAM driver to only map the
device tree-specified reserved areas depending on a flag set
based on the compatibility string. This would not affect non-Tegra
systems that rely on the entire thing being memory mapped.
If 64K pages are being used, we cannot exactly map the 4K regions
that are placed in SYSRAM - ioremap code instead aligns to closest
64K pages. However, since in practice the non-accessible memory area
is 64K aligned, these mappings do not overlap with the non-accessible
memory area and things work out.
Reviewed-by: Mian Yousaf Kaukab <ykaukab@suse.de>
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Link: https://lore.kernel.org/r/20210715103423.1811101-1-mperttunen@nvidia.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
