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>
There is code related to hard-reset, which is done in gaudi specific
code. However, this code can be used by future ASICs and therefore it
is better to move it to the common code section.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
We add support for NIC DERR ECC error events, in case this error
is received a device reset will be performed.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In preparation for a new feature that allows the user to reserve
signals ahead of submissions, we need to change a current assumption
in the code.
Currently, the driver uses 2 SOBs to support signal CS. When the first
SOB reaches max value, the driver switches to the other one and assumes
that when it will need to switch back to the first one, all of the
signals have already been handled.
This assumption won't hold when the new feature will be added, because
using signal reservation, the driver can reach the max SOB value very
fast.
The change is to add a validity check when submitting a signal CS, to
make sure the previous SOB is available (all the signals attached to
it indeed finished).
Signed-off-by: farah kassabri <fkassabri@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
We introduce a new type of reset which is reset upon device release.
This reset is very similar to soft reset except the fact it is
performed only upon device release and not upon user sysfs request
nor TDR.
The purpose of this reset is to make sure the device is returned to
IDLE state after the current user has finished working with the device.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
To be able to debug long-running CS better, without changing the
userspace code, we are adding a new option through debugfs interface
to skip the reset of the device in case of CS timeout.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Currently driver sends fc interrupt id to FW instead of using
cpu interrupt id. We intend to fix that and keep backward
compatibility by using the same interrupt values.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
There was a rogue #ifdef that crept into the upstream code for
backwards compatibility which isn't needed of course.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In case QMAN has an error and stop_on_err is true, print specific
information of the "offending" command buffer batch.
If the error occurred on one of the higher CPs, the CQ pointer and size
will be printed along with (up to) last 8 PQEs of the stream.
If the error occurred in the lower CP, the CQ pointer and size will be
printed along with (up to) last 8 PQEs of ALL upper CPs as we have no
way to know which upper CP sent the job there.
This is done so higher SW levels will be able to debug their CS by
extracting the raw data of the offending command buffer batch and
examine those offline to detect the issue.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In a system with multiple ASICs, there is a need to provide monitoring
tools with information on how long a device was opened and how many
times a device was opened.
Therefore, we add a new opcode to the INFO ioctl to provide that
information.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Update STMTCSR and STMSYNCR values in order to reduce amount of sync
packets
Signed-off-by: Tal Albo <talbo@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
fix the following smatch warnings:
goya_pin_memory_before_cs()
warn: '&userptr->job_node' not removed from list
gaudi_pin_memory_before_cs()
warn: '&userptr->job_node' not removed from list
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
pin_user_pages_fast() might fail and return a negative number, or pin
less pages than requested and return the number of the pages that were
pinned.
For the latter, it is informative to print also the memory size and the
number of requested pages.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
If an error occurs after a 'pci_enable_pcie_error_reporting()' call, it
must be undone by a corresponding 'pci_disable_pcie_error_reporting()'
call, as already done in the remove function.
Fixes: 2e5eda4681 ("habanalabs: PCIe Advanced Error Reporting support")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Firmware in habanalabs devices is composed of several components.
During device initialization, we read these versions from the device.
Print them during device initialization to allow better visibility in
automated systems.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Hard reset flow on PLDM might take more than 2 minutes.
Hence add a dedicated hard reset timeout of 6 minutes for PLDM.
Signed-off-by: Omer Shpigelman <oshpigelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In current code, for dynamic f/w loading flow, DRAM scrambling is
enabled post Linux fit image is loaded to the card. This can cause the
device CPU to go into reset state.
The correct sequence should be:
1. Load boot fit image
2. Enable scrambling
3. Load Linux fit image
This commit aligns the DRAM scrambling enabling with the static f/w load
flow.
Signed-off-by: Bharat Jauhari <bjauhari@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
If there is an error in the QMAN/engine, there is no point of trying
to continue running the workload. It is better to stop to allow the
user to debug the program.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In case we have EQ fault we would like to know about it.
For this, a status bitmask was added in which EQ_FAULT bit is
set by FW in case of EQ fault.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
When there is an ECC error in the HBM, return a standard error code,
-EIO in this case, and not a positive value.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
When converting virtual address to physical we need to add correct
offset to the physical page.
For this we need to use mask that include ALL bits of page offset.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Get rid of the need to check if boot_dev_sts is valid on every access
to value read from these registers.
This is done by storing the register value in hdev props ONLY if
register is enabled.
This way if register is NOT enabled all capability bits will not be set.
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>
Sometimes it is useful to allow the command to continue running despite
the timeout occurred, to differentiate between really stuck or just very
time consuming commands. This can be achieved by passing a new debug
flag alongside the cs, HL_CS_FLAGS_SKIP_RESET_ON_TIMEOUT.
Anyway, if the timeout occurred, a warning print shall be issued,
however this shall not fail the submission.
Signed-off-by: Yuri Nudelman <ynudelman@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In order for driver to be aware of process or thread crashes inside
GAUDI's CPU, we introduce a new event which contains all relevant
information. Upon event reception, driver will dump information and
will reset the device.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In the collective wait, we put jobs on the QMANs of all the NICs. The
code takes into account if a port is disabled only in case of PCI card.
When this info arrives from the f/w, the code doesn't take it into
account, and it tries to schedule jobs on NICs that aren't enabled and
thats a bug.
To fix this, after the f/w sends us the list of disabled ports, we
update the state of the QMANs according to that list. In addition,
we need to update the HW_CAP bits so the collective wait operation
will not try to use those QMANs. We also need to update the collective
master monitor mask.
Moreover, we need to add a protection for such future cases and in case
the user will try to submit work to those QMANs.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Current implementation uses a single interrupt interface towards
FW, this interface is causing races between interrupt types.
We split this interface to interface per interrupt type.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
There is no dependency when probing multiple devices so indicate to the
kernel that it can probe our devices in ASYNC fashion.
This shortens insmod of the driver from ~2 minutes to 20 seconds on
a system with 8 devices.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Update the QM stop on error masks to also stop on ARB errors.
Signed-off-by: Tomer Tayar <ttayar@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
nic_ports_mask is used by the networking part of the driver.
In the compute part, we use the HW_CAP bits to select what is active
and what is not.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
This fix was applied since there was an incorrect reported CPU ID to GIC
such that an error in MME2 QMAN aliased to be an arriving from DMA0_QM.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
After all the latest changes to the reset code, there were some
redundancy and errors in the flows.
If the Linux FIT is loaded to the ASIC CPU, we need to communicate
with it only via GIC. If it is not loaded, we need to either use
COMMS protocol (for newer f/w) or MSG_TO_CPU register (for older f/w).
In addition, if we halted the device CPU then we need to mark that
the driver will do the reset, regardless of the capabilities.
Also, to prevent false errors, we need to keep track whether the
device CPU was already halted. If so, we shouldn't try to halt it
again.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Using negative logic (i.e. fw_security_disabled) is confusing.
Modify the flag to use positive logic (fw_security_enabled).
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
This is needed because legacy FW 'communication' protocol will soon
become obsolete.
Because COMMS is a boot protocol, communicating through it is supported
only until Linux is loaded to the device CPU, where in that case we
will fallback to the former implementation.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Security is set based on PCI ID, and after reading preboot status bits.
GIC usage is set in both scenarios since GIC can't be used when security
is enabled.
Moreover, writing to GIC/SP is enabled only after Linux is fully loaded.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
On newer releases, host won't be able to trigger an interrupt directly
to the ASIC GIC controller.
To be able to decide whether GIC can/not be used, we must read device's
preboot status bits in a stage that precedes the possible first use of
GIC (when device is in dirty state).
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
To harden the event queue mechanism, we add a running index to the
control header of the entry.
The firmware writes the index in each entry and the driver verifies
that the index of the current entry is larger by 1 of the index of
the previous entry.
In case it isn't, the driver will treat the entry as if it wasn't
valid (it won't process it but won't skip it).
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Scrubbing memory after every unmap is very costly in terms of
performance. If a user wants it he can enable it but the default
should prioritize performance.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
As iATU configuration is done by FW, driver should not try and
move HBM bar.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Reading of GIC privileged status will be done after F/W is loaded,
because privileged GIC capability is only available with the correct
ARMCP version, and after it's loaded.
Such versions necessarily support COMMS, so GIC alternatives (SP regs)
will be read directly from dynamic regs.
As well, initiation of DMA QMANs will occur after F/W is loaded
since it depends on GIC configuration.
In case F/W isn't loaded there's no problem since either way
there won't be any GIC IRQ handling.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Fix issue in which the input to the function is_asic_secured was device
PCI_IDS number instead of the asic_type enumeration.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
LKD should provide hard reset cause to preboot prior to
loading any FW components (in case needed).
Current implementation is based on the new FW 'COMMS' protocol
In cased 'COMMS' is disabled - reset cause won't be sent.
Currently, only 2 reset causes are shared: HEARTBEAT & TDR.
Sending the reset cause will provide the missing watchdog
info that the firmware needs to provide to the BMC.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
An information print notifying on starting to load the f/w was removed
by mistake when moving to the new dynamic f/w loading mechanism.
Restore that print as the F/W loading usually takes between 10 to 20
seconds and this print helps the user know the status of the driver
load.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Due to new security restrictions, GIC controller can no
longer be accessed from user/kernel.
To monitor that, a new status bit will be read from preboot
caps, indicating whether direct access to GIC is blocked.
In case it is blocked, driver will use scratchpad registers
instead of using GIC interface on two main scenarios:
The first of which LKD triggers interrupts to F/W through GIC,
and the second of when LKD configures all engines/QMANs
to write to GIC when they want to report an error.
From F/W perspective, it will poll on all SPs, and once IRQ
number is retrieved, SP register is cleared, and it will perform the
write to the GIC to trigger the IRQ handler.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Maintain both STS1 and ERR1 registers used for status communication
with F/W.
Those are not maintained as we currently have less than 31
statuses/error defined and so LKD did not refer to those register.
The reason to read them now is to try to support future f/w versions
with current driver.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
When attempting to read FW component's version we should break if input
FW component is invalid in order to avoid using uninitialized
destination pointer.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
When setting "DMA mask from FW" we are reading PSOC_GLOBAL_CONF register
which is allowed only once FW has done it's iATU configuration.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Some users might want to implement their own policy of when the device
is unusable so we need to ignore this status in the driver and continue
loading as normal.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Implementing dynamic linux image load to the device.
This patch also implements the FW communication steps during the
boot-fit.
This patch also enables the dynamic protocol based on the compatibility
flag.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Implementing dynamic boot fit image load to the device.
Note that some necessary adjustment were added to the static loader as
well so that both loaders can co-exist.
as this is not the final FW load stage the dynamic FW load is still
forced to be non functional.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Hint address failure that results in a valid mapping with an address
that was allocated by the driver is not a real failure.
Therefore, the driver shouldn't notify about this in kernel log. The
user is responsible to check the returned address.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Indicate "progress" instead of "error" when reporting progress status.
Change "u-boot stopped by user" to "Cannot boot" message as
CPU_BOOT_STATUS_UBOOT_NOT_READY may indicate a fatal error that prevent
u-boot from loading firmware.
Signed-off-by: Guy Nisan <gnisan@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
iATU (internal Address Translation Unit of the PCI controller)
configuration is being done by FW right after driver enables
the PCI device. Hence, driver must add a minor sleep afterwards
in order to make sure FW finishes configuring iATU regions.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Update the common and GAUDI firmware header files to the latest version.
The latest version use the correct endianness types so this commit also
contains minor changes to the code to use the correct conversions when
reading/writing to the firmware structures.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
LKD has interfaces in which it receives device address.
For instance the debugfs_read/write variants receives device address for
CFG/SRAM/DRAM for read/write and need to translate to the mapped PCI BAR
address.
In addition, the dynamic FW load protocol dictates that the address to
which the LKD will copy the image for the next FW component will be
received as a device address and can be placed either in SRAM or DRAM.
We need to distinguish those regions as the access methods to those
regions are different (in DRAM we possibly need to set the BAR base).
Looking forward this code will be used to remove duplicated code in the
debugfs_read/write that search the memory region for the input device
address.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
First stage of the dynamic FW load protocol is to reset the protocol to
avoid residues from former load cycles.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Instead of using multiple ASIC specific copies of functions to read the
FW version use single common one that gets ASIC specific arguments.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Use mmu cache range invalidation instead of entire cache invalidation
because it yields better performance.
In GOYA and GAUDI, always use entire cache invalidation because these
ASICs don't support range invalidation.
Signed-off-by: Alon Mizrahi <amizrahi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Replace multiple arguments to init device CPU function by passing
firmware loader managing structure that is initialized per ASIC with
the loader parameters.
In addition, the FW loader management structure is now part of the
habanalabs device, this way the loader parameters will be able to be
communicated across various boot stages.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
This refactor is needed due to the dynamic FW load in which requesting
the FW file (and getting its attributes) is not immediately followed by
copying FW file content.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Start the skeleton for the dynamic F/W load by marking current preboot
code path as legacy.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
On PLDM, in case of NIC hangs, the ELBI reset to take much longer than
expected. As a result an increase in the ELBI reset timeout is required.
Signed-off-by: Moti Haimovski <mhaimovski@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
The variable err is being assigned a value that is never read, the
assignment is redundant and can be removed. Also remove some empty
lines.
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Addresses-Coverity: ("Unused value")
Link: https://lore.kernel.org/r/20210603131210.84763-1-colin.king@canonical.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Our code analyzer reported a uaf.
In gaudi_memset_device_memory, cb is get via hl_cb_kernel_create()
with 2 refcount.
If hl_cs_allocate_job() failed, the execution runs into release_cb
branch. One ref of cb is dropped by hl_cb_put(cb) and could be freed
if other thread also drops one ref. Then cb is used by cb->id later,
which is a potential uaf.
My patch add a variable 'id' to accept the value of cb->id before the
hl_cb_put(cb) is called, to avoid the potential uaf.
Fixes: 423815bf02 ("habanalabs/gaudi: remove PCI access to SM block")
Signed-off-by: Lv Yunlong <lyl2019@mail.ustc.edu.cn>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Wait for interrupt timeout calculation is wrong, hence timeout occurs
when user waits on an interrupt with certain timeout values.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In case firmware has a bug and erroneously reports a status error
(e.g. device unusable) during boot, allow the user to tell the driver
to continue the boot regardless of the error status.
This will be done via kernel parameter which exposes a mask. The
user that loads the driver can decide exactly which status error to
ignore and which to take into account. The bitmask is according to
defines in hl_boot_if.h
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
This error indicates a problem in the security initialization inside
the f/w so we need to stop the device loading because it won't be
usable.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
If we read all FF from the boot status register, then something is
totally wrong and there is no point of reading specific errors.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Currently the user cannot interpret the PLL information based on index
as its exposed as an integer.
This commit exposes ASIC specific PLL indexes and maps it to a generic
FW compatible index.
Signed-off-by: Bharat Jauhari <bjauhari@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In the case where size is zero the while loop never assigns rc and the
return value is uninitialized. Fix this by initializing rc to zero.
Fixes: 639781dcab ("habanalabs/gaudi: add debugfs to DMA from the device")
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Addresses-Coverity: ("Uninitialized scalar variable")
Link: https://lore.kernel.org/r/20210412161012.1628202-1-colin.king@canonical.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
We need to print a message to the kernel log in case we encounter
an unknown error in the f/w boot to help the user understand what
happened.
In addition, we shouldn't print unknown error in case of known errors.
Moreover, in case of warnings/info, we shouldn't return -EIO that will
fail the initialization and mark the device as disabled
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
update files to latest version from F/W team.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
As part of the securing GAUDI, the F/W will configure the PCI iATU
regions. If the driver identifies a secured PCI ID, it will know to
skip iATU configuration in a very early stage.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
As F/ security indication must be available before driver approaches
PCI bus, F/W security should be derived from PCI id rather than be
fetched during boot handshake with F/W.
Signed-off-by: Ofir Bitton <obitton@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
DRAM scrubbing can take time hence it adds to latency during allocation.
To minimize latency during initialization, scrubbing is moved to release
call.
In case scrubbing fails it means the device is in a bad state,
hence HARD reset is initiated.
Signed-off-by: Bharat Jauhari <bjauhari@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
In order to minimize hard coded values between F/W and the driver, we
send msi-x indexes dynamically to the F/W.
Signed-off-by: Ohad Sharabi <osharabi@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Clearing QM errors by the driver will prevent these H/W blocks from
stopping in case they are configured to stop on errors, so perform this
clearing only if this mode is not in use.
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 multiple ECC errors, FW will set the DEVICE_UNUSABLE bit.
On boot-up, the driver will therefore fail inserting the device.
Signed-off-by: Koby Elbaz <kelbaz@habana.ai>
Reviewed-by: Oded Gabbay <ogabbay@kernel.org>
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Prefer the use of strscpy when copying the ASIC name into a char array,
to prevent accidentally exceeding the array's length.
In addition, strlcpy is frowned upon so replace it.
Signed-off-by: Oded Gabbay <ogabbay@kernel.org>
Yuri Nudelman