, against Roland's git. Please
review the changes and apply the patches for 2.6.24 if they are okay.
Regards,
Joachim
--
Joachim Fenkes -- eHCA Linux Driver Developer and Hardware Tamer
IBM Deutschland Entwicklung GmbH -- Dept. 3627 (I/O Firmware Dev. 2)
Schoenaicher Strasse 220 -- 71032
From: Stefan Roscher [EMAIL PROTECTED]
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/hcp_if.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/hcp_if.c
b/drivers/infiniband/hw/ehca/hcp_if.c
index 24f4541
From: Stefan Roscher [EMAIL PROTECTED]
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_cq.c |7 ++-
drivers/infiniband/hw/ehca/ehca_main.c |2 +-
drivers/infiniband/hw/ehca/ehca_qp.c |9 +++--
drivers/infiniband/hw/ehca
Change hvcall trace output towards better readability: reg numbers instead
of argument numbers, return code as signed decimal instead of unsigned hex.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/hcp_if.c | 57 ++
1 files
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_tools.h |9 +++--
1 files changed, 3 insertions(+), 6 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/ehca_tools.h
b/drivers/infiniband/hw/ehca/ehca_tools.h
index 57c77a7..f9b264b 100644
Some firmware levels exhibit a race condition between H_ALLOC_RESOURCE(MR)
and H_FREE_RESOURCE(MR). Work around this problem by locking these hvCalls
against each other.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/hcp_if.c | 28 +---
1
Nobody needed the SVNEHCA_ prefix anyway.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_main.c |6 --
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/ehca_main.c
b/drivers/infiniband/hw/ehca/ehca_main.c
Roland Dreier [EMAIL PROTECTED] wrote on 13.09.2007 06:33:45:
-#define HCA_CAP_MR_PGSIZE_4K 1
-#define HCA_CAP_MR_PGSIZE_64K 2
-#define HCA_CAP_MR_PGSIZE_1M 4
-#define HCA_CAP_MR_PGSIZE_16M 8
+#define HCA_CAP_MR_PGSIZE_4K 0x8000
+#define HCA_CAP_MR_PGSIZE_64K 0x4000
the hugetlb flag in ehca_reg_user_mr()
The patches should apply cleanly, in order, on top of my previous 12-patch
set. Please review the changes and apply the patches for 2.6.24 if they are
okay.
Regards,
Joachim
--
Joachim Fenkes -- eHCA Linux Driver Developer and Hardware Tamer
IBM Deutschland
During ib_umem_get(), determine whether all pages from the memory region are
hugetlb pages and report this in the hugetlb field. Low-level driver can
use this information if they need it.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/core/umem.c | 20
...because, on virtualized hardware like System p, we can't be sure that the
physical pages behind them are contiguous.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_mrmw.c | 25 +++--
1 files changed, 15 insertions(+), 10 deletions
Roland Dreier [EMAIL PROTECTED] wrote on 12.09.2007 22:21:54:
What happens if someone runs the new driver with older firmware? Or
what if someone upgrades the firmware without updating the driver?
Thanks for pointing our noses to this. Your comment triggered some further
internal discussions
hca_cap_mr_pgsize and fixes a problem with ib_srp
[5/5] enables large page MRs by default
I built the patches on top of Roland's for-2.6.24 git branch. Please review
and queue them for 2.6.24-rc1 if you're okay with them. Thanks!
Cheers,
Joachim
--
Joachim Fenkes -- eHCA Linux Driver Developer
Because hardware reports the SRQ token in RWQEs of SRQ base QPs, supply the
base QP token as SRQ token, so we can properly find the SRQ base QP.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_qp.c |4 +++-
1 files changed, 3 insertions(+), 1 deletions
Simplify ehca_encode_hwpage_size(), fixing an infinite loop for pgsize == 0
in the process. Fix the bug in alloc_fmr() that triggered the loop.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_mrmw.c | 15 ---
1 files changed, 4 insertions(+), 11
ehca_shca.hca_cap_mr_pgsize now contains all supported page sizes ORed
together. This makes some checks easier to code and understand, plus we can
return this value verbatim in query_hca(), fixing a problem with SRP
(reported by Anton Blanchard -- thanks!).
Signed-off-by: Joachim Fenkes [EMAIL
is changed so that it applies that patch if, and only if, it
is run on a ppc64 architecture and the kernel version implies that the
kernel has the ibmebus bug.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
ofed_scripts/ofed_patch.sh | 49
be circumvented by, prior to loading the
eHCA driver, changing the location codes of the offending devices so that
they're not the same anymore. This patch adds that circumvention to openibd,
with an additional check to make sure we're on the right architecture and
kernel version.
Signed-off-by: Joachim
The formula would yield -1 for this, which is wrong in a bad way (max
throttling). Clamp to 0, which is the correct value.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
This fixes another regression introduced in rc3.
Please review and apply for 2.6.24-rc4. Thanks!
drivers/infiniband/hw
Roland Dreier [EMAIL PROTECTED] wrote on 06.12.2007 19:27:09:
+ ehca_lock_hcalls =
!(cur_cpu_spec-cpu_user_features
+ PPC_FEATURE_ARCH_2_05);
We already talked about this yesterday, but I still feel that
checking the
instruction
Firmware would round up the number of SGEs to four, because the WQE
structure holds four SGEs. For SRQ, only three are supported, so return a
fixed value instead.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
The patch will apply cleanly on top of Roland's git. Please review and apply
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
This addresses a comment of Roland and bumps the version number. If it's not
too late, please apply for 2.6.24. Thanks!
drivers/infiniband/hw/ehca/ehca_classes.h |1 +
drivers/infiniband/hw/ehca/ehca_main.c|2 +-
drivers/infiniband
an empty (size 0) RDMA Read into
the message stream every now and then if the consumer doesn't post them
frequently enough.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_classes.h |5 ++
drivers/infiniband/hw/ehca/ehca_qp.c | 14 +++-
drivers/infiniband
From: Hoang-Nam Nguyen hnguyen at de.ibm.com
Signed-off-by: Hoang-Nam Nguyen [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_classes.h |2 +-
drivers/infiniband/hw/ehca/ehca_main.c|6 +++---
2 files changed, 4 insertions(+), 4 deletions(-)
diff --git
From: Hoang-Nam Nguyen hnguyen at de.ibm.com
Signed-off-by: Hoang-Nam Nguyen [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_qp.c |5 -
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/ehca_qp.c
b/drivers/infiniband/hw/ehca/ehca_qp.c
index
-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_reqs.c |4
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/ehca_reqs.c
b/drivers/infiniband/hw/ehca/ehca_reqs.c
index 3aacc8c..2ce8cff 100644
--- a/drivers/infiniband/hw/ehca
From: Hoang-Nam Nguyen [EMAIL PROTECTED]
This patch enables ehca to redirect any PMA queries to the
actual PMA QP.
Signed-off-by: Hoang-Nam Nguyen [EMAIL PROTECTED]
Reviewed-by: Joachim Fenkes [EMAIL PROTECTED]
Reviewed-by: Christoph Raisch [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca
and apply for 2.6.25 if you think they're okay.
Thanks and regards,
Joachim
--
Joachim Fenkes -- eHCA Linux Driver Developer and Hardware Tamer
IBM Deutschland Entwicklung GmbH -- Dept. 3627 (I/O Firmware Dev. 2)
Schoenaicher Strasse 220 -- 71032 Boeblingen -- Germany
eMail: [EMAIL
Please review these patches and queue them for inclusion into 2.6.26 if you
think they're okay.
Thanks!
Joachim
--
Joachim Fenkes -- eHCA Linux Driver Developer and Hardware Tamer
IBM Deutschland Entwicklung GmbH -- Dept. 3627 (I/O Firmware Dev. 2)
Schoenaicher Strasse 220 -- 71032 Boeblingen
...as required by IB Spec, C10-29.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_classes.h |1 +
drivers/infiniband/hw/ehca/ehca_qp.c |3 +++
drivers/infiniband/hw/ehca/ehca_reqs.c|5 +
3 files changed, 9 insertions(+), 0 deletions
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_irq.c|2 +-
drivers/infiniband/hw/ehca/ehca_main.c | 14 ++--
drivers/infiniband/hw/ehca/ehca_mrmw.c | 16 ++
drivers/infiniband/hw/ehca/ehca_qp.c | 12
drivers
Always enable large page support; didn't seem to cause problems for anyone.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_main.c | 22 +++---
1 files changed, 3 insertions(+), 19 deletions(-)
diff --git a/drivers/infiniband/hw/ehca
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_main.c | 37 +++
1 files changed, 18 insertions(+), 19 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/ehca_main.c
b/drivers/infiniband/hw/ehca/ehca_main.c
index ab02ac8
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
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 45fe35a..6504897 100644
--- a/drivers
On Monday 21 April 2008 10:04, Joachim Fenkes wrote:
+ if (unlikely(my_qp-state != IB_QPS_RTS)) {
+ ehca_err(qp-device, QP not in RTS state qpn=%x, qp-qp_num);
+ return -EINVAL;
+ }
Myself, I'm not very happy with using EINVAL, but I can't think of a more
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
On Friday 06 June 2008 22:20, Dotan Barak wrote:
I checked the code in the ehca driver and noticed that
post RR to a QP is being accepted in any state (including the RESET state).
You're right, this is only consistent -- thanks for pointing
This gives ehca an autogenerated modalias and therefore enables automatic
loading.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_main.c |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/infiniband/hw/ehca/ehca_main.c
b
Hi,
this patchset contains eHCA backports for several kernels as well as some
fixes to other backports that broke on powerpc.
[1/5] fixes core for 2.6.24
[2/5] fixes core for RHEL5.1 and RHEL 5.2
[3/5] adds ehca backports for 2.6.2[345]
[4/5] adds ehca backports for RHEL5.1 and RHEL5.2
[5/5]
From: Alexander Schmidt [EMAIL PROTECTED]
The backport kobject.h was causing an include recursion (included slab.h
which in turn needed kobject.h), so move the function bodies to their own .c
file in core, which is the only place where the new functions are needed.
Signed-off-by: Joachim Fenkes
From: Alexander Schmidt [EMAIL PROTECTED]
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
.../2.6.18-EL5.1/include/linux/dma-mapping.h |2 +-
.../2.6.18-EL5.2/include/linux/dma-mapping.h |2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/kernel_addons
From: Alexander Schmidt [EMAIL PROTECTED]
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
.../2.6.18-EL5.1/ehca-01-ibmebus_loc_code.patch| 186
.../2.6.18-EL5.2/ehca-01-ibmebus_loc_code.patch| 186
2 files changed, 372 insertions(+), 0
From: Alexander Schmidt [EMAIL PROTECTED]
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
.../ehca_01_ibmebus_loc_code.patch | 154
.../ehca_01_ibmebus_loc_code.patch | 154
2 files changed, 308 insertions(+), 0
Also, fix atomic64_t backport on powerpc64.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
.../2.6.9_U6/include/asm-powerpc/atomic64.h| 61 +++
.../backport/2.6.9_U6/include/asm/atomic.h |4 +
.../2.6.9_U6/ehca_01_ibmebus_loc_code.patch| 186
Vladimir Sokolovsky [EMAIL PROTECTED] wrote on 15.07.2008 17:54:19:
Applied,
Thanks!
Should the same be done for 2.6.9_U4 and 2.6.9_U5?
I'd rather not; we haven't tested those.
Regards,
Joachim
___
ewg mailing list
ewg@lists.openfabrics.org
[1/2] fixes spurious PATH_MIG events with certain FW versions
[2/2] inserts a default value for Local CA ACK Delay
Please review these patches and queue them for inclusion into the kernel if
you think they're okay.
Thanks!
Joachim
___
ewg mailing
Some firmware versions report a Local CA ACK Delay of 0. In that case,
return a more sensible default value of 12 (- 16 msec) instead.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_hca.c |4 +++-
1 files changed, 3 insertions(+), 1 deletions(-)
diff
Hi, Vlad,
please pull from ~hnguyen/libehca.git -- there's a new fix for OFED 1.4 in
there.
Thanks and regards,
Joachim
___
ewg mailing list
ewg@lists.openfabrics.org
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/ewg
A previous fix introduced a regression where port activation events were
dropped unconditionally if port autodetection was not enabled. Fixed.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
Roland -- this patch is made against your for-linus branch. Please review
and apply if you think it's
Roland Dreier [EMAIL PROTECTED] wrote on 10.11.2008 21:36:23:
A previous fix introduced a regression where port activation events
were
dropped unconditionally if port autodetection was not enabled. Fixed.
Is this a fix to IB/ehca: Remove reference to special QP in case of
port
Our suppress DMEM code works in these kernels only if
CONFIG_HOTPLUG_SPARSE_MEMORY is disabled, so patch it out again.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
Vlad, please apply this one for RC5 -- thanks!
Joachim
diff -Nurp a/kernel_patches/backport/2.6.24
From: Michael Ellerman [EMAIL PROTECTED]
shca_list_lock is taken from softirq context in ehca_poll_eqs, so we need to
lock IRQ safe elsewhere.
Signed-off-by: Michael Ellerman [EMAIL PROTECTED]
Acked-by: Joachim Fenkes [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_main.c | 17
The error message printed when the eHCA driver prevents memory hotplug is
misleading -- the user might think that hot-removing the lhca, hotplugging
memory, then hot-adding the lhca again will work, but it doesn't.
Signed-off-by: Joachim Fenkes [EMAIL PROTECTED]
---
Vlad, please apply
Roland Dreier [EMAIL PROTECTED] wrote on 26.11.2008 00:13:51:
That's too bad... I applied this patch but out of curiousity, why
doesn't the hot-remove/hot-add work? I would have thought that
re-registering all of memory after the hot-add would do the right thing.
That's right, but right now,
All the fields in the control block are nicely right-aligned, so no masking
is necessary.
Signed-off-by: Joachim Fenkes fen...@de.ibm.com
---
drivers/infiniband/hw/ehca/ehca_classes_pSeries.h | 28 -
drivers/infiniband/hw/ehca/ehca_qp.c | 18 +++-
2
between ehca, mlx4 and mthca in all possible
combinations.
Signed-off-by: Joachim Fenkes fen...@de.ibm.com
---
Vlad, please queue this patch for OFED 1.5 if Hal doesn't object -- thanks!
Regards,
Joachim
libibmad/include/infiniband/mad.h |9 +++
libibmad/src/gs.c |6
On Tuesday 30 June 2009 00:01, Hal Rosenstock wrote:
On Mon, Jun 29, 2009 at 8:10 AM, Joachim Fenkesfen...@de.ibm.com wrote:
Previously, libibmad reacted to GSI MAD responses with a redirect status
by throwing an error. IBM eHCA adapters use redirection, so most
infiniband_diags tools
between ehca, mlx4 and mthca in all possible
combinations.
Signed-off-by: Joachim Fenkes fen...@de.ibm.com
---
On Tuesday 30 June 2009 16:14, Hal Rosenstock wrote:
On Tue, Jun 30, 2009 at 8:04 AM, Joachim Fenkesfen...@de.ibm.com wrote:
On Tuesday 30 June 2009 00:01, Hal Rosenstock wrote:
On Mon
between ehca, mlx4 and mthca in all possible
combinations.
Signed-off-by: Joachim Fenkes fen...@de.ibm.com
---
Hi, Hal and Jason,
here's an updated patch that will bail on GID-routed redirection. Also, I
moved the redirection itself into its own function so it can easily be
included into RMPP as well
Hal Rosenstock hal.rosenst...@gmail.com wrote on 01.07.2009 15:59:41:
+static int redirect_port(ib_portid_t *port, uint8_t *mad)
+{
+ port-lid = mad_get_field(mad, 64, IB_CPI_REDIRECT_LID_F);
+ if (!port-lid) {
+ IBWARN(GID-based redirection is not supported);
between ehca, mlx4 and mthca in all possible
combinations.
Signed-off-by: Joachim Fenkes fen...@de.ibm.com
---
After all has been said and done, here's the hopefully last iteration of the
patch, with the hex display of the redirect LID replaced by decimal.
Any objections against this patch?
Regards
Hal Rosenstock hal.rosenst...@gmail.com wrote on 07.07.2009 17:23:18:
+static int redirect_port(ib_portid_t *port, uint8_t *mad)
+{
+ port-lid = mad_get_field(mad, 64, IB_CPI_REDIRECT_LID_F);
+ if (!port-lid) {
+ IBWARN(GID-based redirection is not supported);
Hal Rosenstock hal.rosenst...@gmail.com wrote on 08.07.2009 15:24:53:
I suppose this can be fixed down the road.
Is that an ACK? ;)
Indeed, a qualified ACK :-)
Cool, thanks!
This patch should make its way into OFED 1.5... so who should pull it?
You? Vlad? Someone not on CC? Whoever,
Hal Rosenstock hal.rosenst...@gmail.com wrote on 08.07.2009 18:48:29:
This patch should make its way into OFED 1.5... so who should pull it?
You? Vlad? Someone not on CC? Whoever, please apply for OFED 1.5 --
thanks!
Sasha is the management maintainer. Userspace trees for OFED 1.5
The old code used a lot of hardcoded values, which might not be valid in all
environments (especially routed fabrics or partitioned subnets). Copy as
much information as possible from the incoming request to prevent that.
Signed-off-by: Joachim Fenkes fen...@de.ibm.com
---
Hal, Jason -- here's
Hal Rosenstock hal.rosenst...@gmail.com wrote on 26.08.2009 17:15:03:
Thanks for doing this. It looks sane to me. The only issue I recall that
appears to be remaining is a better setting of
ClassPortInfo:RespTimeValue
rather than hardcoding. Perhaps using the value from PortInfo is the way
Hal Rosenstock hal.rosenst...@gmail.com wrote on 27.08.2009 15:31:40:
I don't think it should be hard coded. IMO it would be better to default
to 18
and somehow able to be adjusted (via a (dynamic) module parameter ?).
I don't see how making this a parameter would benefit any end user, while
Was reporting CQE flags in the wrong bit positions, causing consumers to
miss incoming immediate data.
Signed-off-by: Joachim Fenkes fen...@de.ibm.com
---
Please review and queue for 2.6.32 if you think it's okay. Thanks!
Joachim
drivers/infiniband/hw/ehca/ehca_reqs.c |6 +-
1 files
Vladimir Sokolovsky v...@dev.mellanox.co.il wrote on 09.05.2010
11:37:08:
Can you please create the latest.txt file in the same directory
where libehca tarballs placed, such that it will include the name of
the tarball that should go into OFED-1.5.2?
This way the new libehca library will be
Steve Wise sw...@opengridcomputing.com wrote on 21.07.2010 15:12:49:
Already been done a long time ago. From cxgb3_release_notes.txt:
Oops, sorry, didn't check the docs first. I duped my bug to yours now.
Regards,
Joachim
___
ewg mailing list
69 matches
Mail list logo