From: Ira Weiny <ira.we...@intel.com>
Attributed ID was declared as an int while the value should really be big
endian 16.
Fixes: 35c4cbb17811 ("IB/core: Create get_perf_mad function in sysfs.c")
Reported-by: Bart Van Assche <bart.vanass...@sandisk.com>
Signed-off
From: Ira Weiny <ira.we...@intel.com>
This was a single patch before. The change to dev_dbg required a precursor
patch to add the dd_dev_dbg which is consistent with the other dev_* macros
which automatically use struct hfi1_devdata.
Changes from V4:
Fix build error which aros
From: Ira Weiny <ira.we...@intel.com>
To be used in future patches add dd_dev_dbg. dd_* functions properly decode
the hfi1_devdata structure used throughout the driver
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/hfi.h | 4
1 file changed,
From: Dean Luick <dean.lu...@intel.com>
Correctly set half-swing for integrated devices. A0 needs all fields set for
CcePcieCtrl. B0 and later only need a few fields set.
Reviewed-by: Stuart Summers <john.s.summ...@intel.com>
Signed-off-by: Dean Luick <dean.lu...@intel.com>
tation of the interaction
with user space and makes use of these functions.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/Kconfig| 1 +
drivers/staging/rdma/hfi1/Makefile | 2 +-
drivers/stagi
From: Mitko Haralanov <mitko.harala...@intel.com>
The HFI1_CAP_TID_UNMAP comment was incorrectly implying the
opposite of what capability actually did. Correct this error.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
gt;
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/hfi.h | 20
1 file changed, 20 insertions(+)
diff --git a/drivers/staging/rdma/hfi1/hfi.h b/drivers/staging/rdma/hfi1/hfi.h
index 12760bb9e056..905bb40febb3 100644
--- a/drivers/staging/rdma/hfi1/h
irtual buffer, program the RcvArray group entries with
these physical chunks, and unprogram the RcvArray groups.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
Changes from V1:
Add comment to program_rcvarray
the PSM library.
This function is currently unused and is being added
in preparation for the TID caching feature.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 67
time.
This would have caused other complications and ugliness.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/file_ops.c | 448 +++
drivers/staging/rdma/hfi1/hfi
ed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 263 ++-
1 file changed, 259 insertions(+), 4 deletions(-)
diff --git a/drivers/staging/rdma/hfi1/user_exp_rcv.c
From: Ira Weiny <ira.we...@intel.com>
Expected receives work by user-space libraries (PSM) calling into the driver
with information about the user's receive buffer and have the driver DMA-map
that buffer and program the HFI to receive data directly into it.
This is an expensive ope
quired resources and structures.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 154 +--
1 file changed, 144 insertions(+), 10 deletions(-)
diff --git a/dr
flush of the write-combining buffer.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 64
1 file changed, 64 insertions(+)
diff --git a/drivers
From: Mitko Haralanov <mitko.harala...@intel.com>
The exp_lock lock does not need to be a spinlock as
all its uses are in process context and allowing the
process to sleep when the mutex is contended might
be benefitial.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Revi
From: Mitko Haralanov <mitko.harala...@intel.com>
There is no need to use a separate variable for a
return value and a label when returning right away
would do just as well.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.co
From: Mitko Haralanov <mitko.harala...@intel.com>
Convert bit definitions to use BIT() macro as per checkpatch.pl
requirements.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
include/uapi/rdma/hf
ntains the
free code path. This is done on purpose to help with patch
reviews as the programming code path is much longer.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/u
r file.
The command is also added to the switch() statement in file_ops.c
for completeness and in preparation for its usage later.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/file_ops.c | 1 +
inc
time.
This would have caused other complications and ugliness.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/file_ops.c | 448 +++
drivers/staging/rdma/hfi1/hfi
viewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 263 ++-
1 file changed, 259 insertions(+), 4 deletions(-)
diff --git a/drivers/staging/rdma/hfi1/user_exp_rcv.c
the PSM library.
This function is currently unused and is being added
in preparation for the TID caching feature.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 67
quired resources and structures.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 154 +--
1 file changed, 144 insertions(+), 10 deletions(-)
diff --git a/dr
From: Mitko Haralanov <mitko.harala...@intel.com>
Convert bit definitions to use BIT() macro as per checkpatch.pl
requirements.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
include/uapi/rdma/hf
flush of the write-combining buffer.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 64
1 file changed, 64 insertions(+)
diff --git a/drivers
From: Mitko Haralanov <mitko.harala...@intel.com>
The exp_lock lock does not need to be a spinlock as
all its uses are in process context and allowing the
process to sleep when the mutex is contended might
be benefitial.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mit
irtual buffer, program the RcvArray group entries with
these physical chunks, and unprogram the RcvArray groups.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/user_exp_rcv.c | 310 +
From: Mitko Haralanov <mitko.harala...@intel.com>
The HFI1_CAP_TID_UNMAP comment was incorrectly implying the
opposite of what capability actually did. Correct this error.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
From: Mitko Haralanov <mitko.harala...@intel.com>
There is no need to use a separate variable for a
return value and a label when returning right away
would do just as well.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.co
From: Mitko Haralanov <mitko.harala...@intel.com>
In preparation for adding the TID caching support, there is a set
of headers, structures, and variables which will be needed. This
commit adds them to the hfi.h header file.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-o
ntains the
free code path. This is done on purpose to help with patch
reviews as the programming code path is much longer.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/u
From: Ira Weiny <ira.we...@intel.com>
Expected receives work by user-space libraries (PSM) calling into the driver
with information about the user's receive buffer and have the driver DMA-map
that buffer and program the HFI to receive data directly into it.
This is an expensive ope
tation of the interaction
with user space and makes use of these functions.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/Kconfig| 1 +
drivers/staging/rdma/hfi1/Makefile | 2 +-
drivers/stagi
r file.
The command is also added to the switch() statement in file_ops.c
for completeness and in preparation for its usage later.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers/staging/rdma/hfi1/file_ops.c | 1 +
inc
based on the defaults for the recv queue size. However, it was kept
fixed such that increasing those queue sizes would not adversely affect
fairness in the future.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Dean Luick <dean.lu...@intel.com>
---
drivers/infiniba
From: Mitko Haralanov <mitko.harala...@intel.com>
Clean-up unnecessary goto statements based on feedback from the
mailing list on previous patch submissions.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers
From: Mitko Haralanov <mitko.harala...@intel.com>
There is no need to cleck if the packet queue is allocated
when cleaning up a user context. The hfi1_user_sdma_free_queues()
function already does all the required checks.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-o
From: Mitko Haralanov <mitko.harala...@intel.com>
It is possible for an SDMA transmission error to happen
during the processing of an user SDMA transfer. In that
case it is better to detect it early and abort any further
attempts to send more packets.
Reviewed-by: Ira Weiny <ira.we...@
From: Mitko Haralanov <mitko.harala...@intel.com>
Convert hfi1_get_user_pages() to use get_user_pages_fast(),
which is much fatster. The mm semaphore is still taken to
update the pinned page count but is for a much shorter
amount of time.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
From: Ira Weiny <ira.we...@intel.com>
Various improvements to the SDMA engine code.
---
Changes from V1:
Fix off by one error in the last patch
Mitko Haralanov (5):
staging/rdma/hfi1: Convert to use get_user_pages_fast
staging/rdma/hfi1: Unconditionally clean-up SDMA
errupt context. Therefore, it is necessary to put
all the completed SDMA tx requests onto a separate list (txcmp) and
offload the actual clean-up and unpinning work to a workqueue.
Reviewed-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.
intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/file_ops.c | 8 +--
drivers/staging/rdma/hfi1/hfi.h| 4 +-
drivers/staging/rdma/hfi1/user_pages.c | 97 +-
3 files changed, 32 insertions(+), 77 deleti
From: Mitko Haralanov <mitko.harala...@intel.com>
There is no need to cleck if the packet queue is allocated
when cleaning up a user context. The hfi1_user_sdma_free_queues()
function already does all the required checks.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-o
From: Mitko Haralanov <mitko.harala...@intel.com>
It is possible for an SDMA transmission error to happen
during the processing of an user SDMA transfer. In that
case it is better to detect it early and abort any further
attempts to send more packets.
Reviewed-by: Ira Weiny <ira.we...@
errupt context. Therefore, it is necessary to put
all the completed SDMA tx requests onto a separate list (txcmp) and
offload the actual clean-up and unpinning work to a workqueue.
Reviewed-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Reviewed-by: Ira Weiny <ira.we...@intel.
From: Ira Weiny <ira.we...@intel.com>
Various improvements to the SDMA engine code.
Mitko Haralanov (5):
staging/rdma/hfi1: Convert to use get_user_pages_fast
staging/rdma/hfi1: Unconditionally clean-up SDMA queues
staging/rdma/hfi1: Clean-up unnecessary goto statements
staging/rdm
From: Mitko Haralanov <mitko.harala...@intel.com>
Clean-up unnecessary goto statements based on feedback from the
mailing list on previous patch submissions.
Reviewed-by: Ira Weiny <ira.we...@intel.com>
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
---
drivers
From: Ira Weiny <ira.we...@intel.com>
Use BIT macros rather than shifts.
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/staging/rdma/hfi1/diag.c b/drivers/stagi
From: Ira Weiny <ira.we...@intel.com>
Place logical operators at the end of the previous line when using a multi-line
statement. Found by checkpatch --strict
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 6 +++---
1 file changed, 3 inse
From: Ira Weiny <ira.we...@intel.com>
It was identified that hfi1_ioctl may sleep with a spin lock held.
This was identified publicly here:
http://www.spinics.net/lists/linux-rdma/msg29926.html
As well as by our internal development.
This series cleans up the code and parameter
From: Ira Weiny <ira.we...@intel.com>
sizeof should use the variable rather than the struct definition to ensure that
type changes are properly accounted for.
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 2 +-
1 file changed, 1 insertion(+)
From: Ira Weiny <ira.we...@intel.com>
physState, linkState, and devState should be phys_state, link_state, and
dev_state
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdm
From: Ira Weiny <ira.we...@intel.com>
Else statements should continue using braces even if there is only 1 line in
the block. Found by checkpatch --strict
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 6 --
1 file changed, 4 inse
From: Ira Weiny <ira.we...@intel.com>
Set link state is not supported remove from the switch statement and allow the
default to return -ENOTTY
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from v1:
From: Ira Weiny <ira.we...@intel.com>
Rather than have a switch in a large else clause make the parameter checks
return immediately.
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma
Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from V1:
Squashed removal of goto which is safe after this patch
staging/rdma/hfi1: remove unneeded goto done
drivers/stagi
From: Ira Weiny <ira.we...@intel.com>
Fix line alignment in various places as caught by checkpatch --strict.
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 16
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/dri
From: Ira Weiny <ira.we...@intel.com>
Final clean up of the if/then/else clause for the parameter checks of
hfi1_ioctl
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from v1:
SETLINKSTATE is
From: Ira Weiny <ira.we...@intel.com>
Use !foo rather than (foo == NULL) as recommended by checkpatch --strict
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 20 ++--
1 file changed, 10 insertions(+), 10 deletions(-)
diff --g
Queue
Reviewed-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from V1:
Adjust to add the done label which can't be removed until a later patch
From: Ira Weiny <ira.we...@intel.com>
To be used in future patches add dd_dev_dbg. dd_* functions properly decode
the hfi1_devdata structure used throughout the driver
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/hfi.h | 4
1 file changed,
From: Dean Luick <dean.lu...@intel.com>
Correctly set half-swing for integrated devices. A0 needs all fields set for
CcePcieCtrl. B0 and later only need a few fields set.
Reviewed-by: Stuart Summers <john.s.summ...@intel.com>
Signed-off-by: Dean Luick <dean.lu...@intel.com>
From: Ira Weiny <ira.we...@intel.com>
This was a single patch before. The change to dev_dbg required a precursor
patch to add the dd_dev_dbg which is consistent with the other dev_* macros
which automatically use struct hfi1_devdata.
Dean Luick (1):
staging/rdma/hfi1: set Gen3 half
From: Ira Weiny <ira.we...@intel.com>
Rather than have a switch in a large else clause make the parameter checks
return immediately.
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma
From: Ira Weiny <ira.we...@intel.com>
Final clean up of the if/then/else clause for the parameter checks of
hfi1_ioctl
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from v1:
SETLINKSTATE is
Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from V1:
Squashed removal of goto which is safe after this patch
staging/rdma/hfi1: remove unneeded goto done
drivers/stagi
From: Dean Luick <dean.lu...@intel.com>
Correctly set half-swing for integrated devices. A0 needs all fields set for
CcePcieCtrl. B0 and later only need a few fields set.
Reviewed-by: Stuart Summers <john.s.summ...@intel.com>
Signed-off-by: Dean Luick <dean.lu...@intel.com>
From: Ira Weiny <ira.we...@intel.com>
It was identified that hfi1_ioctl may sleep with a spin lock held.
This was identified publicly here:
http://www.spinics.net/lists/linux-rdma/msg29926.html
As well as by our internal development.
This series cleans up the code and parameter
From: Jubin John <jubin.j...@intel.com>
Correct the checks on diag.c with the latest checkpatch
Reviewed-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Reviewed-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Jubin John <jubin.j...@intel.com>
Queue
Reviewed-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from V1:
Adjust to add the done label which can't be removed until a later patch
From: Ira Weiny <ira.we...@intel.com>
physState, linkState, and devState should be phys_state, link_state, and
dev_state
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdm
el.com>
Signed-off-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/chip.c | 74 -
drivers/staging/rdma/hfi1/driver.c | 108 -
drivers/staging/rdma/hfi
niszyn <mike.marcinis...@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 27 ++-
drivers/staging/rdma/hfi1/hfi.h | 20 ++
drivers/stagin
From: Jubin John <jubin.j...@intel.com>
Using the latest checkpatch correct the checks on diag.c
Reviewed-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Reviewed-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Jubin John <jubin.j...@intel.com>
From: Ira Weiny <ira.we...@intel.com>
physState, linkState, and devState should be phys_state, link_state, and
dev_state
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdm
From: Ira Weiny <ira.we...@intel.com>
It was identified that hfi1_ioctl may sleep with a spin lock held.
This was identified publicly here:
http://www.spinics.net/lists/linux-rdma/msg29926.html
As well as by our internal development.
This series cleans up the code and parameter
From: Ira Weiny <ira.we...@intel.com>
Rather than have a switch in a large else clause make the parameter checks
return immediately.
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma
From: Ira Weiny <ira.we...@intel.com>
Set link state was not supported and so we can return early in the parameter
checks rather than falling through the switch clause.
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.co
From: Ira Weiny <ira.we...@intel.com>
Final clean up of the if/then/else clause for the parameter checks of
hfi1_ioctl
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdm
Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 59
1 file changed, 23 insertions(+), 36 deletions(-)
diff --git a/drivers/staging/rdma/hfi1/diag.c b/drivers/stagin
From: Ira Weiny <ira.we...@intel.com>
This goto done is followed by an if (ret) break in the outer switch clause. It
is unnecessary.
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1
Queue
Reported-by: Alexey Khoroshilov <khoroshi...@ispras.ru>
Reviewed-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/diag.c | 10
com>
Signed-off-by: Vennila Megavannan<vennila.megavan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from V1:
Redo commit message as well as Kconfig help to make it clear what the
compile and module options do.
Changes from V2:
Re
From: Dean Luick <dean.lu...@intel.com>
Correctly set half-swing for integrated devices. A0 needs all fields set for
CcePcieCtrl. B0 and later only need a few fields set.
Reviewed-by: Stuart Summers <john.s.summ...@intel.com>
Signed-off-by: Dean Luick <dean.lu...@intel.com>
number of sdma engines and
with WQ_CPU_INTENSIVE and WQ_HIGHPRI specified.
Then change send to use the new scheduler which no longer needs to get the
s_lock
Reviewed-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marcinis...@intel.com>
is...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
This patch was previously sent as #9/9 but 5/9 failed.
I'm resending 7,8, and 9 as they did not explicitly depend on 5 and 6.
drivers/staging/rdma/hfi1/qp.c| 27 +--
drivers/staging/rdma/hfi1
From: Mike Marciniszyn <mike.marcinis...@intel.com>
Move hfi1_migrate_qp from ruc.c to qp.[hc] in prep for modifying the QP
workqueue.
Signed-off-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
This patch was previously se
ed-off-by: Vennila Megavannan<vennila.megavan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from V1:
Redo commit message as well as Kconfig help to make it clear what the
compile and module options do.
---
drivers/s
From: Ira Weiny <ira.we...@intel.com>
The following 3 patches fix the ECN detection and add a module parameter to
turn the prescan of the receive queue on and off.
Arthur Kepner (2):
staging/rdma/hfi1: don't cache "prescan head"
staging/rdma/hfi1: optionally prescan rx que
a CNP, or adjusting the injection rate). Afterward, the packet will
be processed normally.
Reviewed-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Arthur Kepner <arthur.kep...@intel.com>
Signed-off-by: Ira We
From: Dean Luick <dean.lu...@intel.com>
Correctly set half-swing for integrated devices. A0 needs all fields set for
CcePcieCtrl. B0 and later only need a few fields set.
Reviewed-by: Stuart Summers <john.s.summ...@intel.com>
Signed-off-by: Dean Luick <dean.lu...@intel.com>
Arthur Kepner <arthur.kep...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/driver.c | 13 +++--
drivers/staging/rdma/hfi1/hfi.h| 13 -
2 files changed, 3 insertions(+), 23 deletions(-)
diff --git a/drivers/staging
inis...@intel.com>
Signed-off-by: Vennila Megavannan<vennila.megavan...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/Kconfig | 14 +++---
drivers/staging/rdma/hfi1/driver.c | 24 +---
2 files changed, 24 in
tation of the interaction with user
space and makes use of these functions.
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/Kconfig| 1 +
drivers/staging/rdma/hfi1/Makefile | 2 +-
drivers
is...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/qp.c| 27 +--
drivers/staging/rdma/hfi1/qp.h| 1 -
drivers/staging/rdma/hfi1/ruc.c | 10 ++
drivers/staging/rdma/hfi1/ud.c| 1 +
drivers/stagin
From: Ira Weiny <ira.we...@intel.com>
Remove the following macros in favor of explicit use of struct hfi1_filedata and
various sub structures.
ctxt_fp
subctxt_fp
tidcursor_fp
user_sdma_pkt_fp
user_sdma_comp_fp
Reviewed-by: Mitko Haralanov <mitko.harala...@intel.com>
Signed-off-b
gned-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
Changes from V3:
Reworked based on the removal of the file pointer macros
Split out some prep patches and code clean up
Changes from V2:
Fix random Kc
number of sdma engines and
with WQ_CPU_INTENSIVE and WQ_HIGHPRI specified.
Then change send to use the new scheduler which no longer needs to get the
s_lock
Reviewed-by: Dennis Dalessandro <dennis.dalessan...@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marcinis...@intel.com>
From: Mike Marciniszyn <mike.marcinis...@intel.com>
Move hfi1_migrate_qp from ruc.c to qp.[hc] in prep for modifying the QP
workqueue.
Signed-off-by: Mike Marciniszyn <mike.marcinis...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma
From: Mitko Haralanov <mitko.harala...@intel.com>
These includes were not used in file_ops.c
Signed-off-by: Mitko Haralanov <mitko.harala...@intel.com>
Signed-off-by: Ira Weiny <ira.we...@intel.com>
---
drivers/staging/rdma/hfi1/file_ops.c | 10 --
1 file changed, 10
1 - 100 of 889 matches
Mail list logo
