I agree with Segher. We already know we have opportunities to do a better
job with shrink-wrapping (pushing this kind of useless activity down past
early exits), so having examples of code to look at to improve this would
be useful.
-- Bill
Bill Schmidt, Ph.D.
Linux on Power Toolchain
IBM
applying this on top.
Untested, but seems obvious.
Fix a typo in setting COMMON_USER2_POWER7 bits to .cpu_user_features2
cpu specs table.
Signed-off-by: Will Schmidt will_schm...@vnet.ibm.com
diff --git a/arch/powerpc/kernel/cputable.c
b/arch/powerpc/kernel/cputable.c
index c60bbec..51eecb5
Mike:
It is not a permanent damage thing.
A read disturb does no permanent damage to the chip
but if the read disturb event involves more bits than
can be corrected by your ECC code, it can do permanent
damage to the *DATA* you've stored in that block.
For this reason, a good flash
and work occurring on Power.
Signed-off-by: Will Schmidt will_schm...@vnet.ibm.com
cc: Olof Johansson o...@lixom.net
cc: Anton Blanchard an...@samba.org
cc: Benjamin Herrenschmidt b...@kernel.crashing.org
---
v2 - Per feedback from Olof, the code is reworked to utilize kernel
parameter runtime checks
and work occurring on Power.
Signed-off-by: Will Schmidt will_schm...@vnet.ibm.com
cc: Olof Johansson o...@lixom.net
cc: Anton Blanchard an...@samba.org
cc: Benjamin Herrenschmidt b...@kernel.crashing.org
---
v2 - Per feedback from Olof, the code is reworked to utilize kernel
parameter runtime
On Sat, 2010-09-25 at 22:49 -0500, Olof Johansson wrote:
On Fri, Sep 24, 2010 at 04:44:15PM -0500, Will Schmidt wrote:
Add a config option for the (batched) MULTITCE and BULK_REMOVE h-calls.
By default, these options are on and are beneficial for performance and
throughput reasons
is useful for some of the PREEMPT_RT related investigation and
work occurring on Power.
Signed-off-by: Will Schmidt will_schm...@vnet.ibm.com
cc: Anton Blanchard an...@samba.org
cc: Benjamin Herrenschmidt b...@kernel.crashing.org
---
diff --git a/arch/powerpc/platforms/pseries/Kconfig
b/arch
. Note that the 32bit code already exports these symbols.
Signed-off-By: Sonny Rao sonny...@us.ibm.com
Acked-by: Alexander Schmidt al...@linux.vnet.ibm.com
Index: common/arch/powerpc/mm/init_64.c
===
--- common.orig/arch/powerpc/mm
dvh...@linux.vnet.ibm.com wrote on 09/02/2010 01:04:28 AM:
Subject
Re: [PATCH][RFC] preempt_count corruption across H_CEDE call with
CONFIG_PREEMPT on pseries
With this in place, we no longer see the preempt_count dropping below
zero. However, if I offline/online a CPU about 246 times I
Ankita Garg ank...@in.ibm.com wrote on 08/19/2010 10:58:24 AM:
Subject
Re: [PATCH][RFC] preempt_count corruption across H_CEDE call with
CONFIG_PREEMPT on pseries
Hi Darren,
On Thu, Jul 22, 2010 at 11:24:13AM -0700, Darren Hart wrote:
With some instrumentation we were able to determine
On Sun, 2010-07-11 at 02:49 -0500, Milton Miller wrote:
On Fri, 09 Jul 2010 about 08:55:01 -, Will Schmidt wrote:
We've been seeing some issues with userspace randomly SIGSEGV'ing while
running the -RT kernels on POWER7 based systems. After lots of
debugging, head scratching
dvh...@linux.vnet.ibm.com wrote on 07/22/2010 06:57:18 PM:
Subject
Re: [PATCH][RFC] preempt_count corruption across H_CEDE call with
CONFIG_PREEMPT on pseries
On 07/22/2010 03:25 PM, Benjamin Herrenschmidt wrote:
On Thu, 2010-07-22 at 11:24 -0700, Darren Hart wrote:
1) How can the
call to
arch_leave_lazy_mmu_mode().
The patch here is what I am currently running with, on both POWER6 and
POWER7 systems, successfully.
Signed-off-by: Will Schmidt will_schm...@vnet.ibm.com
CC:Ben Herrenschmidt b...@kernel.crashing.org
CC:Thomas Gleixner t...@linutronix.de
---
diff
On Thu, 2010-05-20 at 11:05 +0200, Jan-Bernd Themann wrote:
Hi Thomas
Re: [PATCH RT] ehea: make receive irq handler non-threaded (IRQF_NODELAY)
On Thu, 20 May 2010, Jan-Bernd Themann wrote:
Thought more about that. The case at hand (ehea) is nasty:
The driver does _NOT_
On Thu, 2010-05-20 at 16:45 +0200, Thomas Gleixner wrote:
On Thu, 20 May 2010, Darren Hart wrote:
On 05/20/2010 01:14 AM, Thomas Gleixner wrote:
On Thu, 20 May 2010, Jan-Bernd Themann wrote:
Thought more about that. The case at hand (ehea) is nasty:
The driver does _NOT_
On Thu, 11 Mar 2010 11:57:53 -0800
Roland Dreier rdre...@cisco.com wrote:
Seems fine... adding EHCA guys just to make sure.
.node is being removed
Signed-off-by: Grant Likely grant.lik...@secretlab.ca
Acked-by: Alexander Schmidt al...@linux.vnet.ibm.com
On Wed, 2009-11-25 at 16:19 -0500, Steven Rostedt wrote:
On Wed, 2009-11-25 at 10:12 -0600, Will Schmidt wrote:
Tested-by: Will Schmidt will_schm...@vnet.ibm.com
Signed-off-by: Will Schmidt will_schm...@vnet.ibm.com
Isn't it assumed that the one that made the patch also tested it? Well I
The tb_total and purr_total values reported via the hcall_stats code
should be cumulative, rather than being replaced by the latest delta tb
or purr value.
Tested-by: Will Schmidt will_schm...@vnet.ibm.com
Signed-off-by: Will Schmidt will_schm...@vnet.ibm.com
---
[ This is a touch-up to the [3
This patch sets the port autodetect mode as default for the ehca driver. The
autodetect code has been in the kernel for several releases now and has
proved to be stable.
---
Roland, please queue this change for 2.6.32 if you are okay with it.
drivers/infiniband/hw/ehca/ehca_main.c |8
the version number for this
patch, so please also apply the following:
ehca: Increment version number for DMEM toleration
Signed-off-by: Alexander Schmidt al...@linux.vnet.ibm.com
---
drivers/infiniband/hw/ehca/ehca_main.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- infiniband.git.orig
On Tue, 16 Jun 2009 09:10:39 -0700
Roland Dreier rdre...@cisco.com wrote:
Yeah, the notifier code remains untouched as we still do not allow dynamic
memory operations _while_ our module is loaded. The patch allows the
driver to
cope with DMEM operations that happened before the
Hi Roland,
thank you for taking a look at the code!
On Fri, 12 Jun 2009 21:50:58 -0700
Roland Dreier rdre...@cisco.com wrote:
OK, one major issue with this patch and a few minor nits.
First, the major issue is that I don't see anything in the patch that
changes the code in
From: Hannes Hering heri...@de.ibm.com
This patch implements toleration of dynamic memory operations and 16 GB
gigantic pages. Toleration means that the driver can cope with dynamic
memory operations that happened before the driver was loaded. While using the
ehca driver, dynamic memory
Hi Roland,
did you have a chance to take a look at the patchset and will you apply it, or
are there any outstanding issues we need to address?
Regards,
Alex
On Wed, 22 Apr 2009 16:02:28 +0200
Stefan Roscher ossro...@linux.vnet.ibm.com wrote:
In case of large queue pairs there is the
On Tue, 28 Apr 2009 07:01:32 -0700
Roland Dreier rdre...@cisco.com wrote:
did you have a chance to take a look at the patchset and will you apply
it, or
are there any outstanding issues we need to address?
I guess it's OK, but definitely 2.6.31 material. I guess I'll stick it
error list of its
respective CQ. If the error QP list of a CQ is not empty, poll_cq()
generates flush CQEs before polling the actual CQ.
Signed-off-by: Alexander Schmidt [EMAIL PROTECTED]
---
Applies on top of 2.6.27-rc3, please consider this for 2.6.28.
drivers/infiniband/hw/ehca/ehca_classes.h
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Dear,
I some questions about the PowerPC cache model.
How many caches are there?
How many caches are in multi-core systems? do you have a L1 cache per
cpu and one shared L2? or is there a victim L3 cache?
can you tell me something about the path
success
[4/5] check idr_find() return value
[5/5] discard double CQE for one WR
They all apply on top of 2.6.27-rc1. If possible, we would like to get
them into 2.6.27.
Regards,
Alexander Schmidt
___
Linuxppc-dev mailing list
Linuxppc-dev@ozlabs.org
https
Rename the poll_cq_one_read_cqe goto label to what it actually does,
repoll.
Signed-off-by: Alexander Schmidt [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_reqs.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
--- infiniband.git.orig/drivers/infiniband/hw/ehca
to the application.
Signed-off-by: Alexander Schmidt [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_classes.h |9 +
drivers/infiniband/hw/ehca/ehca_qes.h |1
drivers/infiniband/hw/ehca/ehca_qp.c | 34 +-
drivers/infiniband/hw/ehca/ehca_reqs.c| 54
When the ehca driver detects an invalid opcode in a CQE, it currently
passes the CQE to the application and returns with success. This patch
changes the CQE handling to discard CQEs with invalid opcodes and to
continue reading the next CQE from the CQ.
Signed-off-by: Alexander Schmidt [EMAIL
be found.
Signed-off-by: Alexander Schmidt [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_reqs.c |4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--- infiniband.git.orig/drivers/infiniband/hw/ehca/ehca_reqs.c
+++ infiniband.git/drivers/infiniband/hw/ehca/ehca_reqs.c
@@ -680,8
. In the reset state, however, it is not allowed to post SQ WQEs,
which confuses applications like ib_mad.
The solution for this problem is to immediately set the qp state as requested
by modify_qp(), even when the call is cached.
Signed-off-by: Alexander Schmidt [EMAIL PROTECTED]
---
drivers
/5] rename goto label in ehca_poll_cq_one()
[3/5] repoll on invalid opcode instead of returning success
[4/5] check idr_find() return value
[5/5] discard double CQE for one WR
They all apply on top of 2.6.27-rc1. If possible, we would like to get
them into 2.6.27.
Regards,
Alexander Schmidt
be found.
Signed-off-by: Alexander Schmidt [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_reqs.c |4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
--- infiniband.git.orig/drivers/infiniband/hw/ehca/ehca_reqs.c
+++ infiniband.git/drivers/infiniband/hw/ehca/ehca_reqs.c
@@ -680,8
When the ehca driver detects an invalid opcode in a CQE, it currently passes
the CQE to the application and returns with success. This patch changes the
CQE handling to discard CQEs with invalid opcodes and to continue reading the
next CQE from the CQ.
Signed-off-by: Alexander Schmidt [EMAIL
to the application.
Signed-off-by: Alexander Schmidt [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_classes.h |9 +
drivers/infiniband/hw/ehca/ehca_qes.h |1
drivers/infiniband/hw/ehca/ehca_qp.c | 34 +-
drivers/infiniband/hw/ehca/ehca_reqs.c| 54
Rename the poll_cq_one_read_cqe goto label to what it actually does,
repoll.
Signed-off-by: Alexander Schmidt [EMAIL PROTECTED]
---
drivers/infiniband/hw/ehca/ehca_reqs.c |6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
--- infiniband.git.orig/drivers/infiniband/hw/ehca
-By: Will Schmidt [EMAIL PROTECTED]
---
This is a resend.. this latest respin is updated to apply on top of Mikeys
slb_mmu_size change.
(earlier Updates made per comments from Olof and Ben and Paul).
This version adds padding around the ESID and VSID fields, and the LLP bits
are displayed too
On Fri, 2007-11-30 at 16:56 +1100, Stephen Rothwell wrote:
On Tue, 27 Nov 2007 18:15:59 -0600 Will Schmidt [EMAIL PROTECTED] wrote:
(resending with the proper from addr this time).
I'm seeing some funky behavior on power5/power6 partitions with this
patch.A /sbin/reboot is now
On Wed, 2007-11-28 at 14:18 -0600, Linas Vepstas wrote:
On Tue, Nov 27, 2007 at 06:15:59PM -0600, Will Schmidt wrote:
(resending with the proper from addr this time).
I'm seeing some funky behavior on power5/power6 partitions with this
patch.A /sbin/reboot is now behaving much
Hi Folks,
I've been seeing a boot hang/crash on power3 systems for a few weeks.
(hangs on a 270, drops to SP on a p610). This afternoon I got around
to tracking it down to the changes in
commit d9c2340052278d8eb2ffb16b0484f8f794def4de
Do not depend on MAX_ORDER when grouping pages by
-By: Will Schmidt [EMAIL PROTECTED]
---
(Updates made per comments from Olof and Ben and Paul).
This version adds padding around the ESID and VSID fields, and the LLP bits
are displayed too.
Counting bits, the VSID output looks to be as large as 51 bits, which requires
up to 13 spaces. This doesnt
1800 59e7ef80dc80 256M ESID=1 VSID= 59e7ef80d LLP
bits: 0
15 1000 59e7ef80dc80
Tested on power5 and power6.
Signed-Off-By: Will Schmidt [EMAIL PROTECTED]
---
This version adds padding around the ESID and VSID fields, and the LLP bits
are displayed too
for deciding when to call GET_ESID
and GET_ESID_1T.
This also happens to fix a duplicate-slb-entry inspired machine-check
exception I was seeing when trying to run java on a power6 partition.
Tested on power6 and power5.
Signed-Off-By: Will Schmidt [EMAIL PROTECTED]
---
arch/powerpc/mm/slb.c | 33
or #define DEBUG_LOW nearby.
Signed-Off-By: Will Schmidt [EMAIL PROTECTED]
---
arch/powerpc/mm/hash_utils_64.c |1 +
arch/powerpc/mm/slb.c |1 +
arch/powerpc/platforms/cell/smp.c |1 +
arch/powerpc/platforms/pseries/firmware.c |1 +
4 files
-By: Will Schmidt [EMAIL PROTECTED]
---
arch/powerpc/xmon/xmon.c | 20 ++--
1 files changed, 14 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c
index 121b04d..97984f3 100644
--- a/arch/powerpc/xmon/xmon.c
+++ b/arch/powerpc/xmon/xmon.c
On Sat, 2007-10-27 at 14:19 +1000, Benjamin Herrenschmidt wrote:
On Fri, 2007-10-26 at 15:46 -0500, Will Schmidt wrote:
[Powerpc] fix switch_slb handling of 1T ESID values
Now that we have 1TB segment size support, we need to be using the
GET_ESID_1T macro when comparing ESID values
-entry inspired machine-check
exception I was seeing when trying to run java on a power6 partition.
Tested on power6 and power5.
Signed-Off-By: Will Schmidt [EMAIL PROTECTED]
---
There is a similar bit of code in stab.c switch_stab(). Should this change
also be made there?
---
arch/powerpc
On Sun, 2007-10-21 at 14:44 +0200, Jens Axboe wrote:
On Fri, Oct 19 2007, Will Schmidt wrote:
Hi Jens, Stephen, and Everyone else.
...
You need this, will remember to fix that up for the new branch as well.
diff --git a/drivers/block/viodasd.c b/drivers/block/viodasd.c
index e824b67
On Wed, 2007-10-03 at 13:13 +1000, Paul Mackerras wrote:
Will Schmidt writes:
I still need to test this code for performance issues, and this version
could still use some cosmetic touchups, so I dont think we want this to
go into a tree yet. I am reposting this primarily to indicate
Hi Paul,
just a few questions.
On Wed, 2007-08-01 at 12:04 +1000, Paul Mackerras wrote:
This makes the kernel use 1TB segments for all kernel mappings and for
user addresses of 1TB and above, on machines which support them
(currently POWER5+ and POWER6). We don't currently use 1TB
On Fri, 2007-08-03 at 08:37 +1000, Benjamin Herrenschmidt wrote:
Is there technical reason why the 'local' variable remains at the end of
the parm list for these? In other cases 'ssize' simply gets added to
the end of the parm list.
Looks nicer to have psize and ssize together :-)
On Wed, 2007-08-01 at 16:02 +1000, Michael Neuling wrote:
--- linux-2.6-ozlabs.orig/arch/powerpc/kernel/entry_64.S
+++ linux-2.6-ozlabs/arch/powerpc/kernel/entry_64.S
@@ -389,7 +389,9 @@ BEGIN_FTR_SECTION
ld r9,PACA_SLBSHADOWPTR(r13)
li r12,0
std
On Thu, 2007-07-26 at 11:56 +1000, Stephen Rothwell wrote:
Hi Will,
On Wed, 25 Jul 2007 11:55:31 -0500 Will Schmidt [EMAIL PROTECTED] wrote:
cmpwi 0,r24,0 /* Are we processor 0? */
- beq .__start_initialization_iSeries /* Start up the first processor
55 matches
Mail list logo