On Thu, Sep 10, 2020 at 11:03 PM Jakub Kicinski wrote:
> On Thu, 10 Sep 2020 19:11:16 +0300 Oded Gabbay wrote:
> > +module_param(nic_rx_poll, int, 0444);
> MODULE_PARM_DESC(nic_rx_poll,
> > + "Enable NIC Rx polling mode (0 = no, 1 = yes, default no)");
>
> If your chip does not support IRQ
On Thu, Sep 10, 2020 at 11:31 PM Jakub Kicinski wrote:
> On Thu, 10 Sep 2020 23:17:59 +0300 Oded Gabbay wrote:
> > > Doesn't seem like this one shows any more information than can be
> > > queried with ethtool, right?
> > correct, it just displays it in a format that is much more readable
>
>
From: Oded Gabbay
Sent: Tuesday, 16 July 2019 12:02
> This patch adds some in-code documentation on the different opcodes of
> the INFO IOCTL.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
me the CS)
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
unnecessary.
>
> This patch caps the maximum timeout of the simulator to 10 seconds, which
> is our largest timeout in the code. That is more then enough for anything the
> simulator is doing.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
MMU cache invalidation timeout indicates that the device is unstable and
therefore unusable.
Hence in such case do hard reset and return an error to the user if was
called from ioctl.
In addition, change the print to error level and rephrase its text.
Signed-off-by: Omer Shpigelman
---
drivers
On Sun, May 31, 2020 at 12:16 PM, Oded Gabbay wrote:
> Use the u64_to_user_ptr(x) kernel macro to correctly cast u64 to void*
>
> Reported-by: kbuild test robot
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
omatically when
> reaching 125 degrees.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
> to make the properties appear in sysfs only for GOYA cards.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
This patch improves the security in the Debug IOCTL.
It adds checks that:
- The register index value is in the allowed range for all opcodes.
- The event types number is in the allowed range in SPMU enable.
- The events number is in the allowed range in SPMU disable.
Signed-off-by: Omer
structure.
Note that in contrary to the input structure, the user can provide an
output structure with changing size or no size at all. Therefore the user
output structure validation is already done in the Debug logic later on.
Signed-off-by: Omer Shpigelman
---
drivers/misc/habanalabs
; over the last period of 100-1000ms. The return value is 0-100, representing as
> percentage the total utilization rate.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
ndling function.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
g on the device. If the F/W is old
> and doesn't supply a card name, a default card name is displayed as the
> sensors group name.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
r", etc.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
From: Oded Gabbay
Sent: Monday, 2 September 2019 10:50
> If the initialization of a device failed, the driver prints an error message
> with
> the id of the device. The device index on the file system is that id divided
> by
> 2.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
This patch changes the allocation of the host memory pages array to use
vmalloc if needed. This in order to support mapping of large memory
chunks.
Signed-off-by: Omer Shpigelman
---
drivers/misc/habanalabs/memory.c | 12 +++-
1 file changed, 3 insertions(+), 9 deletions(-)
diff --git
This patch adds support in mapping of vmalloc memory.
In contrary to user memory, vmalloc memory is already pinned and has no
vm_area structure. Therefore a new capability was needed in order to map
this memory.
Mapping vmalloc memory is needed for Gaudi ASIC.
Signed-off-by: Omer Shpigelman
On Thu, Oct 10, 2019 at 5:10 PM Christoph Hellwig wrote:
> On Thu, Oct 10, 2019 at 02:06:22PM +0000, Omer Shpigelman wrote:
> > This patch adds support in mapping of vmalloc memory.
> > In contrary to user memory, vmalloc memory is already pinned and has
> > no vm_area struc
This patch changes the allocation of the host memory pages array to use
vmalloc if needed. This in order to support mapping of large memory
chunks.
Signed-off-by: Omer Shpigelman
---
drivers/misc/habanalabs/memory.c | 12 +++-
1 file changed, 3 insertions(+), 9 deletions(-)
diff --git
In contrary to user memory, kernel memory is already pinned and has no
vm_area structure. Therefore we need a new code path to map this memory.
This is a pre-requisite patch for upstreaming future ASIC support
Signed-off-by: Omer Shpigelman
---
Changes in v2:
- use a boolean parameter
On Thu, Oct 23, 2019 at 12:21 PM Christoph Hellwig wrote:
> On Tue, Oct 22, 2019 at 06:30:35AM +0000, Omer Shpigelman wrote:
> > Changes in v2:
> > - use a boolean parameter to indicate for kernel address rather than
> >calling is_vmalloc_addr, as kernel and user
ff-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
On Mon, Jul 13, 2020 at 06:54 PM, Oded Gabbay wrote:
> Update the boot interface file from the latest version from firmware.
> Defines for secure boot were added.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
Coresight is not supported on simulator, therefore add a boolean for
checking that (currently used by un-upstreamed code).
Signed-off-by: Omer Shpigelman
---
drivers/misc/habanalabs/goya/goya.c | 1 +
drivers/misc/habanalabs/habanalabs.h | 2 ++
2 files changed, 3 insertions(+)
diff --git
Remove an old hash that is not in use anymore.
Signed-off-by: Omer Shpigelman
---
drivers/misc/habanalabs/habanalabs.h | 2 --
drivers/misc/habanalabs/mmu.c| 1 -
2 files changed, 3 deletions(-)
diff --git a/drivers/misc/habanalabs/habanalabs.h
b/drivers/misc/habanalabs/habanalabs.h
Rephrase F/W error message to make it more understandable to ordinary
users.
Signed-off-by: Omer Shpigelman
---
drivers/misc/habanalabs/firmware_if.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/misc/habanalabs/firmware_if.c
b/drivers/misc/habanalabs
ed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
gt; Each bit represents a different engine, according to gaudi_engine_id enum.
>
> See debugfs documentation for more details.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
On Tue, Jun 16, 2020 at 9:13 AM, Oded Gabbay wrote:
> We no longer need to initialize the rate limiters in GAUDI A1.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
On Tue, Jun 16, 2020 at 09:13 AM, Oded Gabbay wrote:
> MAP/UNMAP are done also for device memory.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
tly align
> the structure to 8 bytes.
>
> Signed-off-by: Oded Gabbay
Reviewed-by: Omer Shpigelman
32 matches
Mail list logo