Hi,
that is ok with us.
We will probably send the patches on Monday.
regards Stefan
On Thursday 19 July 2007 10:11, Tziporet Koren wrote:
Hi,
In the OFED meeting this week there was a discussion on OFED 1.2.1 and 1.2.c
We may want to have only one release (1.2.c) that will include 1.2.1.
This patch adds a check to ensure the asm-ppc directory is present in
/lib/modules/K_VER/build/include/ in order to build ib_ehca on
rhel-4.5 systems.
Regards Stefan
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
build_env.sh | 13 -
1 files changed, 12 insertions(+), 1
Hi Michael,
On Saturday 11 August 2007 20:10, Michael S. Tsirkin wrote:
Quoting Stefan Roscher [EMAIL PROTECTED]:
Subject: [PATCH] IB/ehca: fix bugs to support rhel 4.5 in OFED 1.2.c-11
After OFED-1.2.c uses another kernelcode base
Another base as compared to what?
Compared
We do? What is the problem you are solving? Is the build broken?
Yes ehca in OFED-1.2.c does not compile on rhel-4.5 without these patches.
Can you post the error message? Esp the cache patch looks like
it can't be fixing compile errors sine you replace a struct
by an identical
On Tuesday 14 August 2007 15:54, Michael S. Tsirkin wrote:
diff -Nurp
ofa_kernel-1.2.c/kernel_patches/backport/2.6.9_U5/backport_ehca_4_rhel45_mutex.patch
ofa_kernel-1.2.c_new/kernel_patches/backport/2.6.9_U5/backport_ehca_4_rhel45_mutex.patch
---
Please pull for OFED 1.3 the following branch for libehca.
git://git.openfabrics.org/~/scm/libehca.git
branch: ofed_1_3
Thanks Stefan,
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Please pull for OFED 1.3 the following branch for libehca.
git://git.openfabrics.org/~/scm/libehca.git
branch: ofed_1_3
Thanks Stefan,
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
This patch backport the xxx_hotcpu_notifier() interface to older
kernelverisons.
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
2.6.16/include/linux/cpu.h|7 +++
2.6.16_sles10/include/linux/cpu.h |7 +++
2.6.16_sles10_sp1/include/linux/cpu.h |7
This patch backports the sg_page() interface to older kernelversions.
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
2.6.16/include/linux/scatterlist.h|7 +++
2.6.16_sles10/include/linux/scatterlist.h |7 +++
2.6.16_sles10_sp1/include/linux/scatterlist.h
This patch deletes the backport file hvcall.h for RHEL-5.1.
The backport is no longer needed because RHEL-5.1 already contains the needed
hvcalls.
regards Stefan
Signed-off-by: Stefan Roscher stefan.roscher at de.ibm.com
---
hvcall.h | 32
1 file changed, 32
This patch fixes a migth_sleep kernel warning with using sdp + ehca device
driver.
We have to change the kmem_cache_alloc() attribute from GFP_KERNEL to
GFP_ATOMIC because sdp is running within a spinlock
during ah allocation.
regards Stefan
Signed-off-by: Stefan Roscher stefan.roscher
Hi Vladimir,
On Wednesday 21 November 2007 09:23:37 Vladimir Sokolovsky wrote:
Hi Nam,
Please check the following compilation error:
make[4]: ***
[/tmp/ofa_1_3_dev_kernel-20071121-0957_check/drivers/infiniband/hw/ehca/ehca_eq.o]
Error 1
I checked this error and found out that the patch:
Please pull for OFED 1.3 the following branch for libehca.
git://git.openfabrics.org/~/scm/libehca.git
branch: ofed_1_3
Thanks Stefan,
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
Please pull for OFED 1.3 the following branch for libehca.
git://git.openfabrics.org/~/scm/libehca.git
branch: ofed_1_3
Thanks Stefan,
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
This issue occurs only on the two kernels mentioned above.
My Question is , is this the bug you described here:
https://bugs.openfabrics.org/show_bug.cgi?id=807
or should I open a new one?
Kind Regards
Stefan Roscher ___
ewg mailing list
ewg
He Vladimir,
Please pull for OFED 1.3 the following branch for libehca.
git://git.openfabrics.org/~/scm/libehca.git
branch: ofed_1_3
Thanks Stefan,
___
ewg mailing list
ewg@lists.openfabrics.org
, we will first save
all properties of this node, then delete the node
and add the node again with an updated location code. This patch adds this
workaround in openibd script.
On all other kernels and distributions the original functionallity is unchanged.
Signed-off-by: Stefan Roscher
On Wednesday 19 December 2007 10:50:36 Hoang-Nam Nguyen wrote:
Hi Vlad!
Please add attached libehca.spec.in to the libehca git tree.
Applied. Thanks for providing this.
Nam
___
ewg mailing list
ewg@lists.openfabrics.org
Hi Stefan,
Actually, the problem is in your git tree. See,
http://www.openfabrics.org/git/?p=~hnguyen/libehca.git;a=blob;f=libehca.spec.in;h=0edbf3ea34ad115f181c8df4e165b9ab13bce936;hb=ofed_1_3
Run 'dos2unix libehca.spec.in', commit the changes and I will pull into
my git tree.
Done!
Hi Vlad,
the problem still exists in OFED-1.3-20080107-0600.
Is the spec.in file updated in this build?
regards Stefan
On Monday 07 January 2008 16:39:15 Vladimir Sokolovsky wrote:
Stefan Roscher wrote:
Hi Stefan,
Actually, the problem is in your git tree. See,
http://www.openfabrics.org
Hi Vlad!
Please pull from
git://git.openfabrics.org/~hnguyen/libehca ofed_1_3
to get the following small patches for ofed-1.3:
0001remove white spaces and add a missing function header
0002add proper IPD calculation, remove unneccessary query_port()
0003add unsolicited ACK
On Wednesday 13 February 2008 09:04:53 Or Gerlitz wrote:
Pradeep Satyanarayana wrote:
This patch fixes -fail to destroy ipoib rx QP
(https://bugs.openfabrics.org/show_bug.cgi?id=906)
Hence the usecnt issue reported previously on ehca is solved and allows the
qp to be destroyed.
As
Please pull the following branch for libehca.
git://git.openfabrics.org/~/scm/libehca.git
branch: ofed_1_3
changes:
* move libehca version to 1.2
* add update ChangeLog
Thanks Stefan,
___
ewg mailing list
ewg@lists.openfabrics.org
On Monday 18 February 2008 23:49:30 Roland Dreier wrote:
Here is the link to the patch sent previously:
http://lists.openfabrics.org/pipermail/general/2008-February/046463.html
OK, applied, although that link points to an HTML-mangled version of
the patch, and I also had to figure out
Hi,
this patch changes the version number for ehca, to seperate it from the one in
kernel upstream.
It will apply against vlads ofed-1.3 git tree.
Please pick it for OFED-1.3 release.
regards Stefan
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
ehca_main.c |2 +-
1 file changed, 1
On Monday 24 March 2008 21:45:01 Tziporet Koren wrote:
* New vendor HW support - non was reported so far (IBM and Chelsio
- do you have something?)
* OpenSM:
We are not planning to include new features for ehca in OFED-1.4.
regards Stefan
Hi,
is there any schedule for the OFED-1.3.1 release? When should we start to send
some minor bugfixes for ehca? Would the kernel-base be the same 2.6.24
or will it change to 2.6.25?
regards Stefan
___
ewg mailing list
ewg@lists.openfabrics.org
Also, introduce a few inline helper functions to make the code more readable.
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_hca.c | 128
1 files changed, 80 insertions(+), 48 deletions(-)
diff --git a/drivers/infiniband/hw
Signed-off-by: Stefan Roscher stefan.roscher at de.ibm.com
---
drivers/infiniband/hw/ehca/ehca_classes.h |5
drivers/infiniband/hw/ehca/ehca_cq.c | 10
drivers/infiniband/hw/ehca/ehca_main.c| 36 +++-
drivers/infiniband/hw/ehca/ehca_qp.c
Also remove duplicate assignment of local_ca_ack_delay
and change min_t check for local_ca_ack_delay to u8 instead of int.
Signed-off-by: Stefan Roscher stefan.roscher at de.ibm.com
---
drivers/infiniband/hw/ehca/ehca_hca.c |7 +++
1 files changed, 3 insertions(+), 4 deletions(-)
diff
This is necessary because, in a multicore environment, a race between
uverbs async handler and destroy QP could occur.
Signed-off-by: Stefan Roscher stefan.roscher at de.ibm.com
---
We are not sure if this should be fixed in the driver or in uverbs itself.
Roland, what's your opinion about
On Wednesday 07 May 2008 17:32:03 Roland Dreier wrote:
We are not sure if this should be fixed in the driver or in uverbs itself.
Roland, what's your opinion about this?
Would be nice to be able to fix it in uverbs but I don't see how. In
particular a kernel consumer has to have the
Signed-off-by: Stefan Roscher stefan.roscher at de.ibm.com
---
ehca_0041_Fix_wrong_return_types.patch | 38 +
1 file changed, 38 insertions(+)
diff -Nurp
ofa_kernel-1.3_old/kernel_patches/fixes/ehca_0041_Fix_wrong_return_types.patch
ofa_kernel-1.3_new
This patchset contains two changes for IB/ehca and ibmebus.
The first patch enables ibmebus_request_irq() to optionally return the
IRQ number, which is used by the second patch to trigger EOI in case of
lost interrupts.
They should apply cleanly against 2.6.26 git tree.
Thanks
Stefan
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_classes.h |1 +
drivers/infiniband/hw/ehca/ehca_eq.c |6 --
drivers/infiniband/hw/ehca/ehca_main.c| 12 ++--
3 files changed, 15 insertions(+), 4 deletions(-)
diff --git a/drivers
On Tuesday 10 June 2008 00:28:16 Paul Mackerras wrote:
Stefan Roscher writes:
This patchset contains two changes for IB/ehca and ibmebus.
The first patch enables ibmebus_request_irq() to optionally return the
IRQ number, which is used by the second patch to trigger EOI in case
Hi Roland,
On Tuesday 10 June 2008 18:18:50 Roland Dreier wrote:
So just to be clear: this is a workaround for a hardware/firmware bug?
Yes it is.
OK, so paulus et al... does it seem like a good approach to call H_EOI
from driver code (given that this driver makes tons of other
During corner case testing, we noticed that some versions of ehca
do not properly transition to interrupt done in special load situations.
This can be resolved by periodically triggering EOI through H_EOI,
if eqes are pending.
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
As firmware team
Hi Vlad,
during the installation of the OFED-1.4 package I noticed that the installation
script shows the user the missing packages/dependencies in a inconvenient and
circular way.
If packages like gcc++, tcl, bison, tk and many more were missing, the script
only shows one missing package at
On Sunday 03 August 2008 15:24:54 Vladimir Sokolovsky wrote:
Hi Stefan,
I did the required change,
Please check in the OFED-1.4-20080804-0600 build.
Regards,
Vladimir
Great,
works fine for me. Thank you!
regards Stefan
___
ewg mailing
patches at the end of the week.
regards Stefan Roscher
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
This patch adds the missing backport patches for the new kernelbase 2.6.27
It should cleanly apply against Vlad's git tree.
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
.../ehca_02_revert_interface_change.patch | 28
Hi Vlad,
During the creation of the backport patches I noticed we have no rh-4.7
backport directory yet.
Will OFED-1.4 support RH-4.7?
regards Stefan
___
ewg mailing list
ewg@lists.openfabrics.org
parameters are now used as initial values for those variables.
If a user selects an invalid number of CQs or QPs we don't print an error any
longer, instead we will inform the user with a warning and set the values to
the respective maximum supported by the HW.
Signed-off-by: Stefan Roscher [EMAIL
will apply cleanly on top of 2.6.28 git tree.
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_irq.c |7 +--
drivers/infiniband/hw/ehca/ehca_qp.c |5 +
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/ehca_irq.c
Will send a patch for 1323 tomorrow.
Mit freundlichen Grüßen / Kind regards
Stefan Roscher
eHCA/eHEA Linux Driver Development
IBM Systems Technology Group, Systems Software Development / FW I/O
Firmware Entwicklung 2
Closed the Bug, fixed in daily build
Mit freundlichen Grüßen / Kind regards
Stefan Roscher
eHCA/eHEA Linux Driver Development
IBM Systems Technology Group, Systems Software Development / FW I/O
Firmware Entwicklung 2
This fix enables ehca device driver to generate flush workcompletions even if
the application don't request work completions for all work requests. The
current implementation of ehca will generate flush work completions for
the wrong work requests if an application uses non signaled work
.
Signed-off-by: Stefan Roscher [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_classes.h |7 +++
drivers/infiniband/hw/ehca/ehca_qp.c | 12 ++--
drivers/infiniband/hw/ehca/ehca_reqs.c| 13 ++---
3 files changed, 19 insertions(+), 13 deletions(-)
diff --git
This patchset contains performance improvments for ehca driver.
It will skip code which is not necessary for userspace queue pairs
and will replace vmalloc() calls with kmalloc().
Because of this fundamental code change we will also increment the version
number.
They should apply cleanly against
From: Anton Blanchard antonb at au1.ibm.com
To improve performance of driver ressource allocation,
replace the vmalloc() call with kmalloc().
Signed-off-by: Stefan Roscher stefan.roscher at de.ibm.com
---
drivers/infiniband/hw/ehca/ipz_pt_fn.c |6 +++---
1 files changed, 3 insertions(+), 3
. These changes will improve the performance of creation
of userspace queue pairs.
Signed-off-by: Stefan Roscher stefan.roscher at de.ibm.com
---
drivers/infiniband/hw/ehca/ehca_qp.c | 94 ++--
drivers/infiniband/hw/ehca/hcp_if.c |6 +-
drivers/infiniband/hw/ehca
Signed-off-by: Stefan Roscher stefan.rosc...@de.ibm.com
---
drivers/infiniband/hw/ehca/ehca_main.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/ehca_main.c
b/drivers/infiniband/hw/ehca/ehca_main.c
index 368311c..85905ab 100644
/ Kind regards
Stefan Roscher
eHCA/eHEA Linux Driver Development
IBM Systems Technology Group, Systems Software Development / FW I/O
Firmware Entwicklung 2
In case of large queue pairs there is the possibillity of allocation failures
due to memory fragmentationo with kmalloc().To ensure the memory is allocated
even
if kmalloc() can not find chunks which are big enough, we try to allocate the
memory
with vmalloc().
Signed-off-by: Stefan Roscher
On Wednesday 22 April 2009 04:10:18 pm michael wrote:
Hi,
I don't take the point, if it is not import use the vmalloc. Why you try
with a kmalloc
alloc first? and why do not use kzalloc?
Because kmalloc() is faster than vmalloc() causing a huge performance win
when someone allocates a
Hi Roland,
On Wednesday 12 May 2010 06:40:16 pm Roland Dreier wrote:
* added get_huge_page_size() to read the huge page size from
/proc/meminfo. This is done at ibv_fork_init() time.
That doesn't work on systems that have multiple huge page sizes (eg
powerpc). You can compare the
57 matches
Mail list logo
