On 10/26/2017 12:12 PM, SF Markus Elfring wrote:
From: Markus Elfring
Date: Thu, 26 Oct 2017 17:54:15 +0200
Add a jump target so that a bit of exception handling can be better reused
at the end of this function.
I'm not sure this is that big of a win. I mean you aren't really making
the code
On 10/27/2017 4:53 PM, SF Markus Elfring wrote:
This change adds nothing except code churn.
I guess that the shown change possibility can reduce the object code size
for the affected function.
Do you care for such an detail?
Regards,
Markus
You guess? Well perhaps you should find out for ce
idea to depend on internal behaviors like this.
To address the above issues, this patch removes the the fast exit and
Typo above too many "the".
adds an explicit synchronize_rcu().
Signed-off-by: Tejun Heo
Cc: Dennis Dalessandro
Cc: Mike Marciniszyn
Cc: linux-r...@vger.kernel.org
C
On 4/11/2018 3:32 AM, Jia-Ju Bai wrote:
i40iw_add_mqh_4() is never called in atomic context, because it
calls rtnl_lock() that can sleep.
Despite never getting called from atomic context,
i40iw_add_mqh_4() calls kzalloc() with GFP_ATOMIC,
which does not sleep for allocation.
GFP_ATOMIC is not ne
completion and data coalescing, on Intel 5x00 and 7300
Thanks!
Reviewed-by: Dennis Dalessandro
fers, thanks to Jerome's feedback.
Also:
-- Dennis, thanks for your earlier review, and I have not yet added your
Reviewed-by tag, because this revision changes the things that you had
previously reviewed, thus potentially requiring another look.
This spin looks fine to me.
it any worse I don't think.
Signed-off-by: David Howells
cc: Dennis Dalessandro
cc: Mike Marciniszyn
cc: linux-r...@vger.kernel.org
Reviewed-by: Dennis Dalessandro
On 4/8/2021 6:00 PM, Stephen Rothwell wrote:
Hi all,
Commit
042a00f93aad ("IB/{ipoib,hfi1}: Add a timeout handler for rdma_netdev")
is missing a Signed-off-by from its author.
Doh! That's my fault. I must have fat fingered the delete button instead
of editing the line when I was converti
Add a trace message to HFI1s user IOCTL handling. This allows debugging
of which IOCTLs are being handled by the driver.
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/file_ops.c |2 ++
drivers/staging/rdma/hfi1/trace.c|1 +
drivers/staging
Consolidate all the IOCTL defines which are supported by the driver into
a common location. This header file is made available to user space and
will now allow user applications to determine IOCTL numbers directly.
Reviewed-by: Mitko Haralanov
Signed-off-by: Dennis Dalessandro
---
drivers
Reviewed-by: Mike Marciniszyn
Reviewed-by: Dean Luick
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/diag.c | 107 ++
drivers/staging/rdma/hfi1/eprom.c| 121 ++
drivers/staging/rdma/hfi1/eprom.h| 16
Remove the write() handler for user space commands now that ioctl
handling is available. User apps will need to change to use ioctl from
this point forward.
Reviewed-by: Mitko Haralanov
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/file_ops.c | 245
The HFI1_CMD_SDMA_STATUS_UPD command was never implemented it has no
reason to live in the driver. Remove it.
Reviewed-by: Mitko Haralanov
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/file_ops.c |3 ---
include/uapi/rdma/hfi/hfi1_user.h|1
s software version to sysfs for this exact purpose.
Reviewed-by: Mitko Haralanov
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/sysfs.c |8
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/drivers/staging/rdma/hfi1/sysfs.c
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/common.h |6 +
drivers/staging/rdma/hfi1/diag.c | 13 --
drivers/staging/rdma/hfi1/file_ops.c | 224 ++
drivers/staging/rdma/hfi1/hfi.h | 14 ++
include/uapi/rdma
needed so that user space applications (psm) can determine if it needs to do
ioctl() or write().
This patch applies on the latest hfi1 patch set "Fix link and other issues".
Patches can also be viewed on GitHub at:
https://github.com/ddalessa/kernel/tree/for-4.7
---
Dennis Dale
On Thu, Apr 14, 2016 at 10:45:50AM -0600, Jason Gunthorpe wrote:
On Thu, Apr 14, 2016 at 08:41:35AM -0700, Dennis Dalessandro wrote:
This patch series removes the write() interface for user access in favor of an
ioctl() based approach. This is in response to the complaint that we had
different
On Thu, Apr 14, 2016 at 12:05:40PM -0600, Jason Gunthorpe wrote:
There are some pretty obvious paths to make this saner that could only
be a few weeks away, we haven't even had the first conversations
yet. I think you are completely wrong there is no 'line of sight'
It certainly can't be years.
On Mon, May 23, 2016 at 12:06:01PM +0530, Shyam Saini wrote:
Fixed following checkpatch.pl warnings:
Prefer 'unsigned int' to bare use of 'unsigned'
Prefer READ_ONCE() over ACCESS_ONCE()
In general please don't fix two types of issues in the same patch. Also, as
to the unsinged -> unsigned i
On Mon, May 23, 2016 at 12:06:02PM +0530, Shyam Saini wrote:
Fixed following checkpatch.pl warnings:
Prefer 'unsigned int' to bare use of 'unsigned'
Prefer READ_ONCE() over ACCESS_ONCE()
Same response [1] as to your previous patch. If you do spin the READ_ONCE
bit please squash the two.
a/msg35319.html
---
Dennis Dalessandro (5):
IB/hfi1: Export drivers user sw version via sysfs
IB/hfi1: Remove unused user command
IB/hfi1: Add ioctl() interface for user commands
IB/hfi1: Remove write(), use ioctl() for user cmds
IB/hfi1: Add trace message in user IOCTL hand
s software version to sysfs for this exact purpose.
Reviewed-by: Christoph Hellwig
Reviewed-by: Mitko Haralanov
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/sysfs.c |8
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/drive
The HFI1_CMD_SDMA_STATUS_UPD command was never implemented it has no
reason to live in the driver. Remove it.
Reviewed-by: Christoph Hellwig
Reviewed-by: Mitko Haralanov
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/file_ops.c |3 ---
include/uapi
Add a trace message to HFI1s user IOCTL handling. This allows debugging
of which IOCTLs are being handled by the driver.
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/file_ops.c |2 ++
drivers/staging/rdma/hfi1/trace.c|1 +
drivers/staging
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/common.h |6 +
drivers/staging/rdma/hfi1/diag.c | 13 --
drivers/staging/rdma/hfi1/file_ops.c | 224 ++
drivers/staging/rdma/hfi1/hfi.h | 14 ++
include/uapi/rdma
Remove the write() handler for user space commands now that ioctl
handling is available. User apps will need to change to use ioctl from
this point forward.
Reviewed-by: Mitko Haralanov
Signed-off-by: Dennis Dalessandro
---
drivers/staging/rdma/hfi1/file_ops.c | 249
On Thu, May 12, 2016 at 11:34:45AM -0600, Jason Gunthorpe wrote:
On Thu, May 12, 2016 at 10:18:27AM -0700, Dennis Dalessandro wrote:
There is also a driver software version being exported via a sysfs
file. This is needed so that user space applications (psm) can
determine if it needs to do
On Thu, May 12, 2016 at 11:43:32AM -0600, Jason Gunthorpe wrote:
On Thu, May 12, 2016 at 10:18:47AM -0700, Dennis Dalessandro wrote:
+ case HFI1_IOCTL_EP_INFO:
+ case HFI1_IOCTL_EP_ERASE_CHIP:
+ case HFI1_IOCTL_EP_ERASE_RANGE:
+ case HFI1_IOCTL_EP_READ_RANGE
On Thu, May 12, 2016 at 01:25:08PM -0600, Jason Gunthorpe wrote:
On Thu, May 12, 2016 at 03:07:38PM -0400, Dennis Dalessandro wrote:
>>There is also a driver software version being exported via a sysfs
>>file. This is needed so that user space applications (psm) can
>>determ
On Thu, May 12, 2016 at 03:28:21PM -0600, Jason Gunthorpe wrote:
On Thu, May 12, 2016 at 03:48:16PM -0400, Doug Ledford wrote:
were going to rip out the EEPROM code. In any case, the best fix would
be to rebase the two series that are remaining and move any "rip out
things like eeprom support"
On Sun, Feb 21, 2016 at 04:30:39PM +0100, Jannik Becher wrote:
Fixed a coding style issue.
Signed-off-by: Jannik Becher
---
drivers/staging/rdma/hfi1/verbs_mcast.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/staging/rdma/hfi1/verbs_mcast.c
b/drivers/staging
On Wed, Apr 20, 2016 at 02:36:16PM -0600, Jason Gunthorpe wrote:
On Thu, Apr 14, 2016 at 01:52:44PM -0400, Dennis Dalessandro wrote:
The eprom device is for low level programming of the eprom on the
chip. We do not use i2c for this because the eprom is directly
attached to the chip and not
On Sat, Apr 16, 2016 at 09:09:40AM +0300, Leon Romanovsky wrote:
On Fri, Apr 15, 2016 at 07:28:01PM -0400, Ira Weiny wrote:
On Sat, Apr 16, 2016 at 12:23:28AM +0300, Leon Romanovsky wrote:
Do you have a technical reason that this patch series does not fix the
write/writev issue brought up by Al?
On Sat, Apr 16, 2016 at 08:19:17PM +0100, Al Viro wrote:
While we are at it, could the person who'd come up with ui_lseek() be located
and made to stand up and explain the rationale behind the SEEK_END semantics
therein? To quote the manpage (and paraphrase just about any introductory
textbook)
m to me, thanks.
Acked-by: Dennis Dalessandro
On 8/18/2019 3:29 PM, Wenwen Wang wrote:
In fault_opcodes_read(), 'data' is not deallocated if debugfs_file_get()
fails, leading to a memory leak. To fix this bug, introduce the 'free_data'
label to free 'data' before returning the error.
Signed-off-by: W
ata' before returning the error.
Signed-off-by: Wenwen Wang
Acked-by: Dennis Dalessandro
d->element.port_num);
- if (record->event != IB_EVENT_PORT_ERR ||
+ if (record->event != IB_EVENT_PORT_ERR &&
record->event != IB_EVENT_PORT_ACTIVE)
return;
Missed this in code review. Thanks!
Acked-by: Dennis Dalessandro
On 5/9/2019 5:12 PM, Andrea Parri wrote:
On Tue, Apr 30, 2019 at 01:16:57AM +0200, Andrea Parri wrote:
Hi Mike,
This barrier only applies to the read-modify-write operations; in
particular, it does not apply to the atomic_read() primitive.
Replace the barrier with an smp_mb().
This is one o
On 7/8/2019 12:08 PM, Jason Gunthorpe wrote:
On Mon, Jul 08, 2019 at 12:57:25PM +1000, Stephen Rothwell wrote:
Hi all,
After merging the rdma tree, today's linux-next build (x86_64
allmodconfig) failed like this:
In file included from :32:
./usr/include/rdma/rvt-abi.h:13:10: fatal error: rdma/
ter-gather will be stuck with that.
Here's a differential patch with that, that shows a nice little cleanup in
a couple of IB places, and as you point out, it also provides the hooks for
performance upgrades (via batching) in the future.
Does this API look about right?
I'm on board with tha
: Dennis Dalessandro
On 3/3/2018 8:03 AM, Arushi Singhal wrote:
Continue at the bottom of a loop are removed.
Issue found using drop_continue.cocci Coccinelle script.
Signed-off-by: Arushi Singhal
---
drivers/infiniband/hw/qib/qib_init.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/infiniband/hw/qi
On 3/3/2018 9:07 AM, Julia Lawall wrote:
diff --git a/drivers/infiniband/hw/qib/qib_init.c
b/drivers/infiniband/hw/qib/qib_init.c
index 3990f38..1bd3b32 100644
--- a/drivers/infiniband/hw/qib/qib_init.c
+++ b/drivers/infiniband/hw/qib/qib_init.c
@@ -681,7 +681,6 @@ int qib_init(struct qib_devdat
-arm-
m...@vger.kernel.org; linux-arm-ker...@lists.infradead.org; Mike
Marciniszyn ; Dennis Dalessandro
; Doug Ledford ;
open list:HFI1 DRIVER ; open list ; Deucher, Alexander
Subject: Re: [PATCH 1/2] IB/hfi1: Try slot reset before secondary bus reset
[+cc Alex, who might know why DRM drivers have
On 4/13/2018 1:27 PM, Greg Thelen wrote:
Allow INFINIBAND without INFINIBAND_ADDR_TRANS.
Signed-off-by: Greg Thelen
Cc: Tarick Bedeir
Change-Id: I6fbbf8a432e467710fa65e4904b7d61880b914e5
Forgot to remove the Gerrit thing.
-Denny
On 4/17/2018 10:32 AM, Bjorn Helgaas wrote:
[+cc Bartlomiej, who recently changed do_pcie_gen3_transition(), LKML]
On Mon, Apr 16, 2018 at 11:53:43PM -0700, Christoph Hellwig wrote:
On Mon, Apr 16, 2018 at 07:28:23PM -0500, Frederick Lawler wrote:
The IB/hfi1 driver uses custom macros to confi
dd->rcvhdrtail_dummy_physaddr);
- dd->rcvhdrtail_dummy_kvaddr = NULL;
+ dd->rcvhdrtail_dummy_kvaddr = NULL;
}
for (ctxt = 0; tmp && ctxt < dd->num_rcv_contexts; ctxt++) {
Acked-by: Dennis Dalessandro
On 12/20/2016 09:02 AM, Geliang Tang wrote:
To make the code clearer, use rb_entry() instead of container_of() to
deal with rbtree.
Signed-off-by: Geliang Tang
Reviewed-by: Dennis Dalessandro
Thanks
-Denny
On Thu, Jun 16, 2016 at 03:08:17PM -0500, Bjorn Helgaas wrote:
As far as I can tell, none of these PCI questions were raised on
linux-pci, so we never even had a chance to have a conversation about
them.
I'll let Ashutosh handle the technical details here since he is most
familiar with the cod
On 02/10/2017 04:04 PM, SF Markus Elfring wrote:
From: Markus Elfring
Date: Fri, 10 Feb 2017 21:01:55 +0100
Reuse existing functionality from memdup_user() instead of keeping
duplicate source code.
Signed-off-by: Markus Elfring
Thanks for the patch, but this one is already taken care of alo
if ((linkcap & PCI_EXP_LNKCAP_SLS) != PCI_EXP_LNKCAP_SLS_8_0GB) {
Reviewed-by: Dennis Dalessandro
: Bolarinwa Olayemi Saheed
Looks like we may have had a problem when calling
pci_read_config_dword() from the init dd path and doing a check for < 0
to bail. So this looks like goodness to me.
Reviewed-by: Dennis Dalessandro
the later so do the same thing here.
Fixes: 6d72344cf6c4 ("IB/ipoib: Increase ipoib Datagram mode MTU's upper limit")
Link: https://github.com/ClangBuiltLinux/linux/issues/1062
Link:
https://lore.kernel.org/linux-rdma/20200527040350.GA3118979@ubuntu-s3-xlarge-x86/
Signed-off-
On 6/12/2020 3:54 PM, Aditya Pakki wrote:
In case of failure of alloc_ud_wq_attr, the memory allocated by
rvt_alloc_rq() is not freed. The patch fixes this issue by
calling rvt_free_rq().
Signed-off-by: Aditya Pakki
---
drivers/infiniband/sw/rdmavt/qp.c | 1 +
1 file changed, 1 insertion(+)
s), GFP_KERNEL);
if (!rdi->ports)
ib_dealloc_device(&rdi->ibdev);
Acked-by: Dennis Dalessandro
On 10/10/2020 4:57 AM, Xianting Tian wrote:
In architecture like powerpc, we can have cpus without any local memory
attached to it. In such cases the node does not have real memory.
Use local_memory_node(), which is guaranteed to have memory.
local_memory_node is a noop in other architectures th
, NUM_NETDEV_MAP_ENTRIES)) {
- dd_dev_err(dd, "Not enought RMT entries used = %d\n",
+ dd_dev_err(dd, "Not enough RMT entries used = %d\n",
rmt_start);
return false;
}
Thanks!
Acked-by: Dennis Dalessandro
d\n");
- kfree(dd->dummy_netdev);
+ free_netdev(dd->dummy_netdev);
dd->dummy_netdev = NULL;
}
}
For the kfree->free_netdev, you probably want to add:
Reported-by: kbuild test robot
Reported-by: Dan Carpenter
Also can add:
Reviewed-by: Dennis Dalessandro
Thanks
On 10/16/2020 10:11 AM, Jason Gunthorpe wrote:
On Mon, Oct 12, 2020 at 08:36:57AM -0400, Dennis Dalessandro wrote:
On 10/10/2020 4:57 AM, Xianting Tian wrote:
In architecture like powerpc, we can have cpus without any local memory
attached to it. In such cases the node does not have real
ugh.
The macro name "struct_size" is misleading to me. Maybe that's just me,
and in any case...
Reviewed-by: Dennis Dalessandro
eof(*pkt) + sizeof(pkt->addr[0])*n;
+ pktsize = struct_size(pkt, addr, n);
/*
* Determine if this is tid-sdma or just sdma.
Again, same minor objection but correct patch none the less.
Reviewed-by: Dennis Dalessandro
(struct _vls_ectrs);
+ response_data_size = struct_size(req, port[0].vls, num_vls);
if (response_data_size > sizeof(pmp->data)) {
pmp->mad_hdr.status |= IB_SMP_INVALID_FIELD;
Reviewed-by: Dennis Dalessandro
ned-off-by: Aditya Pakki
---
v1: Fix incorrect order of rvt_free_rq and free_ud_wq_attr.
Suggested by Dennis Dalessandro.
---
drivers/infiniband/sw/rdmavt/qp.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/sw/rdmavt/qp.c
b/drivers/infiniband/sw/r
e sdma descriptors are there */
tx->txreq.num_desc = 0;
/* Set the header type */
tx->phdr.hdr.hdr_type = priv->hdr_type;
Thanks
Acked-by: Dennis Dalessandro
tx->txq = txp->txq;
Thanks
Acked-by: Dennis Dalessandro
On 04/05/2017 09:50 AM, SF Markus Elfring wrote:
From: Markus Elfring
Date: Wed, 5 Apr 2017 15:34:32 +0200
A few update suggestions were taken into account
from static source code analysis.
Markus Elfring (5):
Use kcalloc() in qib_init_iba7322_funcs()
Use kmalloc_array() in qib_init_7322_v
On 04/23/2017 03:33 AM, Leon Romanovsky wrote:
On Sat, Apr 22, 2017 at 04:36:19PM +0200, SF Markus Elfring wrote:
From: Markus Elfring
Date: Sat, 22 Apr 2017 13:26:49 +0200
Replace the specification of data types by pointer dereferences
as the parameter for the operator "sizeof" to make the co
On 04/24/2017 10:35 AM, Christoph Hellwig wrote:
On Mon, Apr 24, 2017 at 02:16:31PM +, Byczkowski, Jakub wrote:
Tested-by: Jakub Byczkowski
Are you (and Doug) ok with queueing this up in the PCI tree?
We are fine however Doug wants to handle it.
-Denny
On 7/13/2017 6:13 PM, Colin King wrote:
From: Colin Ian King
Trivial fix to spelling mistake, rename variable 'continious'
to the correct spelling 'continuous'
Signed-off-by: Colin Ian King
---
Thanks!
Acked-by: Dennis Dalessandro
locking is done for performance reasons, not for
correctness, so use the _nort() variants to avoid the above problem.
Suggested-by: Julia Cartwright
Cc: Clark Williams
Cc: Dean Luick
Cc: Dennis Dalessandro
Cc: Doug Ledford
Cc: Kaike Wan
Cc: Leon Romanovsky
Cc: linux-r...@vger.kernel.org
Cc:
On 10/3/2017 11:49 AM, Arnaldo Carvalho de Melo wrote:
Cc: Clark Williams
Cc: Dean Luick
Cc: Dennis Dalessandro
Cc: Doug Ledford
Cc: Julia Cartwright
Cc: Kaike Wan
Cc: Leon Romanovsky
Cc: linux-r...@vger.kernel.org
Cc: Peter Zijlstra
Cc: Sebastian Andrzej Siewior
Cc: Sebastian Sanchez
if it's useful for you I'm sure
it is for others as well, so I have no problem with it.
Reviewed-by: Dennis Dalessandro
On 10/11/2017 11:31 AM, Dennis Dalessandro wrote:
On 10/11/2017 8:41 AM, Thomas Bogendoerfer wrote:
Provide information about used firmware files via modinfo.
Signed-off-by: Thomas Bogendoerfer
---
drivers/infiniband/hw/hfi1/firmware.c | 5 +
1 file changed, 5 insertions(+)
diff --git
On 10/11/2017 11:49 AM, Leon Romanovsky wrote:
On Wed, Oct 11, 2017 at 11:31:18AM -0400, Dennis Dalessandro wrote:
On 10/11/2017 8:41 AM, Thomas Bogendoerfer wrote:
Provide information about used firmware files via modinfo.
Signed-off-by: Thomas Bogendoerfer
---
drivers/infiniband/hw/hfi1
On 10/11/2017 5:35 PM, Don Hiatt wrote:
On 10/11/2017 1:56 PM, Christos Gkekas wrote:
Clean up unused bypass variables in verbs.
Signed-off-by: Christos Gkekas
---
drivers/infiniband/hw/hfi1/verbs.c | 4
1 file changed, 4 deletions(-)
diff --git a/drivers/infiniband/hw/hfi1/verbs.c
eviewed-by: Dennis Dalessandro
limit reached for
0x%0x traps\n",
trap->data.generic_type);
kfree(trap);
}
Thanks!
Acked-by: Dennis Dalessandro
On 11/28/2017 1:15 AM, Leon Romanovsky wrote:
On Mon, Nov 27, 2017 at 11:36:37AM -0700, Jason Gunthorpe wrote:
On Mon, Nov 27, 2017 at 10:41:21AM +0800, Wei Hu (Xavier) wrote:
This patch-set introduces three patches to fix about the memory
related issues.
If these patches are for-rc can you r
On 11/28/2017 2:33 PM, Jason Gunthorpe wrote:
On Tue, Nov 28, 2017 at 06:50:44PM +, Bart Van Assche wrote:
On Tue, 2017-11-28 at 11:39 -0700, Jason Gunthorpe wrote:
On Tue, Nov 28, 2017 at 08:20:09AM -0500, Dennis Dalessandro wrote:
I could resubmit just the series, or you could just pick
data")
Signed-off-by: Nicolai Stange
Looks OK to me.
Reviewed-by: Dennis Dalessandro
81 matches
Mail list logo
