From: Dave Olson
This is version 5 of the patch, incorporating Michael Ellerman's
suggested changes from March 26, 2015. v5 removes the __cpuinit
that had crept in from my patch based on an older kernel, but is
otherwise the same as v4.
This problem appears to have been introd
From: Dave Olson
This is version 4 of the patch, incorporating Michael Ellerman's
suggested changes from March 26, 2015.
This problem appears to have been introduced in 2.6.29 by
the commit in the Fixes line below.
This caused lscpu to error out on e500v2 devices, and probably
Some workloads take a lot of TLB misses despite using traditional
hugepages. Handle these TLB misses in the asm fastpath rather than
going through a bunch of C code. With this patch I measured around a
5x speedup in handling hugepage TLB misses.
Signed-off-by: Scott Wood
---
arch/powerpc/mm/tl
The patch to add it_page_shift incorrectly changed the increment of
uaddr to use it_page_shift, rather then (1 << it_page_shift).
This broke booting on at least some Cell blades, as the iommu was
basically non-functional.
Fixes: 3a553170d35d ("powerpc/iommu: Add it_page_shift field to determine i
The conversion from __get_cpu_var() to this_cpu_ptr() in iic_setup_cpu()
is wrong. It causes an oops at boot.
We need the per-cpu address of struct cpu_iic, not cpu_iic.regs->prio.
Sparse noticed this, because we pass a non-iomem pointer to out_be64(),
but we obviously don't check the sparse resu
From: Wood Scott-B07421
Sent: Friday, April 3, 2015 0:03
To: Zhao Chenhui-B35336
Cc: linuxppc-dev@lists.ozlabs.org; devicet...@vger.kernel.org;
linux-ker...@vger.kernel.org; Jin Zhengxiong-R64188
Subject: Re: [3/4] powerpc: support CPU hotplug for e500mc,
> -Original Message-
> From: Wood Scott-B07421
> Sent: Friday, April 03, 2015 7:56 AM
> To: Tang Yuantian-B29983
> Cc: Wood Scott-B07421; linuxppc-dev@lists.ozlabs.org; Xie Shaohui-B21989
> Subject: Re: [PATCH] powerpc/config: add SCSI configs for
> corenet64_smp_defconfig
>
> On Tue, 2015
On Thu, 2015-04-02 at 16:37 -0500, Lijun Pan wrote:
> Merge Freescale's Internal mpc85xx_defconfig with upstream's.
> Run savedefconfig to get the minimal change.
What makes the internal config preferred without any explanation of why
we want or don't want specific config items? Especially where
On Thu, 2015-04-02 at 16:17 +0530, sand...@freescale.com wrote:
> From: Sandeep Singh
>
> This controller is available on many Freescale SOCs like MPC8315, P1020,
> P1010,
> P1022 and P1024
>
> Signed-off-by: Sandeep Singh
> Signed-off-by: Poonam Aggrwal
> ---
> Documentation/devicetree/bind
From: Sowmini Varadhan
Date: Thu, 2 Apr 2015 17:54:53 -0400
> the other question that comes to my mind is: the whole lazy_flush
> optimization probably works best when there is exactly one pool,
> and no large pools. In most other cases, we'd end up doing a lazy_flush
> when we wrap within our po
On Tue, 2015-03-31 at 18:00 +0800, yuantian.t...@freescale.com wrote:
> From: Tang Yuantian
>
> Otherwise there wil be no SCSI device nodes.
>
> Signed-off-by: Shaohui Xie
> Signed-off-by: Tang Yuantian
> ---
> arch/powerpc/configs/corenet64_smp_defconfig | 7 +++
> 1 file changed, 7 inse
On Tue, 2015-03-31 at 21:58 +0200, Filip Brozović wrote:
> On 3/31/2015 7:54 PM, Scott Wood wrote:
> > This breaks multiplatform support. You need to determine this at
> > runtime.
>
> Understood, but I'm unsure of how to do this exactly. Would it be
> appropriate to define another array, snum_i
On Fri, 2015-03-27 at 19:07 +0200, Purcareata Bogdan wrote:
> On 27.02.2015 03:05, Scott Wood wrote:
> > On Thu, 2015-02-26 at 14:31 +0100, Sebastian Andrzej Siewior wrote:
> >> On 02/26/2015 02:02 PM, Paolo Bonzini wrote:
> >>>
> >>>
> >>> On 24/02/2015 00:27, Scott Wood wrote:
> This isn't a
Highlights include BMan device tree nodes, an MSI erratum workaround, a
couple minor performance improvements, config updates, and misc
fixes/cleanup.
The following changes since commit 6eca8933d3ff17bff39d5f10a2a22366d8622fa6:
powerpc/kernel: Rename copy_thread() 'arg' argument to 'kthread_arg
On Thu, 2015-04-02 at 18:15 -0400, Sowmini Varadhan wrote:
> On (04/03/15 08:57), Benjamin Herrenschmidt wrote:
> >
> > > > I only just noticed too, you completely dropped the code to honor
> > > > the dma mask. Why that ? Some devices rely on this.
>
> /* Sowmini's comment about this coming
On (04/03/15 08:57), Benjamin Herrenschmidt wrote:
>
> > > I only just noticed too, you completely dropped the code to honor
> > > the dma mask. Why that ? Some devices rely on this.
/* Sowmini's comment about this coming from sparc origins.. */
> Probably, not that many devices have limits
Merge Freescale's Internal corenet64_smp_defconfig with upstream's.
Run savedefconfig to get the minimal change.
Signed-off-by: Lijun Pan
---
arch/powerpc/configs/corenet64_smp_defconfig | 42
1 file changed, 24 insertions(+), 18 deletions(-)
diff --git a/arch/power
Merge Freescale's Internal corenet32_smp_defconfig with upstream's.
Run savedefconfig to get the minimal change.
Signed-off-by: Lijun Pan
---
arch/powerpc/configs/corenet32_smp_defconfig | 14 +-
1 file changed, 5 insertions(+), 9 deletions(-)
diff --git a/arch/powerpc/configs/coren
On Thu, 2015-04-02 at 17:54 -0400, Sowmini Varadhan wrote:
> the other question that comes to my mind is: the whole lazy_flush
> optimization probably works best when there is exactly one pool,
> and no large pools. In most other cases, we'd end up doing a lazy_flush
> when we wrap within our pool
On Fri, 2015-04-03 at 09:01 +1100, Benjamin Herrenschmidt wrote:
> On Fri, 2015-04-03 at 08:57 +1100, Benjamin Herrenschmidt wrote:
>
> > No, the last argument to iommu_area_alloc() which is passed from the
> > callers when doing consistent allocs. Basically, the DMA api mandates
> > that consiste
On Fri, 2015-04-03 at 08:57 +1100, Benjamin Herrenschmidt wrote:
> No, the last argument to iommu_area_alloc() which is passed from the
> callers when doing consistent allocs. Basically, the DMA api mandates
> that consistent allocs are naturally aligned (to their own size), we
> implement that on
On Thu, 2015-04-02 at 17:43 -0400, Sowmini Varadhan wrote:
> On (04/03/15 07:54), Benjamin Herrenschmidt wrote:
> > > + limit = pool->end;
> > > +
> > > + /* The case below can happen if we have a small segment appended
> > > + * to a large, or when the previous alloc was at the very end of
> > >
Merge Freescale's Internal mpc85xx_defconfig with upstream's.
Run savedefconfig to get the minimal change.
Signed-off-by: Lijun Pan
---
arch/powerpc/configs/mpc85xx_defconfig | 28 +++-
1 file changed, 7 insertions(+), 21 deletions(-)
diff --git a/arch/powerpc/configs/mp
Merge Freescale's Internal mpc85xx_smp_defconfig with upstream's.
Run savedefconfig to get the minimal change.
Signed-off-by: Lijun Pan
---
arch/powerpc/configs/mpc85xx_smp_defconfig | 10 --
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/arch/powerpc/configs/mpc85xx_smp_d
On Tue, 2015-03-31 at 10:40 -0400, Sowmini Varadhan wrote:
> + if (largealloc) {
> + pool = &(iommu->large_pool);
> + spin_lock_irqsave(&pool->lock, flags);
> + pool_nr = 0; /* to keep compiler happy */
> + } else {
> + /* pick out pool_nr */
the other question that comes to my mind is: the whole lazy_flush
optimization probably works best when there is exactly one pool,
and no large pools. In most other cases, we'd end up doing a lazy_flush
when we wrap within our pool itself, losing the benefit of that
optimization.
Given that the
On (04/03/15 07:54), Benjamin Herrenschmidt wrote:
> > + limit = pool->end;
> > +
> > + /* The case below can happen if we have a small segment appended
> > +* to a large, or when the previous alloc was at the very end of
> > +* the available space. If so, go back to the beginning and f
On Fri, 2015-04-03 at 07:22 +1100, Benjamin Herrenschmidt wrote:
> On Thu, 2015-04-02 at 12:21 -0400, David Miller wrote:
> > From: Sowmini Varadhan
> > Date: Thu, 2 Apr 2015 08:51:52 -0400
> >
> > > do I need to resubmit this without the RFC tag? Perhaps I should
> > > have dropped that some tim
Should have sent this with the other comments, but found it hiding on my
desktop...
On Sat, 2015-03-28 at 01:55 +1100, Alexey Kardashevskiy wrote:
> In order to support memory pre-registration, we need a way to track
> the use of every registered memory region and only allow unregistration
> if a
On Thu, 2015-04-02 at 12:21 -0400, David Miller wrote:
> From: Sowmini Varadhan
> Date: Thu, 2 Apr 2015 08:51:52 -0400
>
> > do I need to resubmit this without the RFC tag? Perhaps I should
> > have dropped that some time ago.
>
> I want to hear from the powerpc folks whether they can positively
On Tue, Mar 24, 2015 at 11:55:09AM -0500, Emil Medve wrote:
> From: Kumar Gala
>
> Signed-off-by: Kumar Gala
> Signed-off-by: Geoff Thorpe
> Signed-off-by: Hai-Ying Wang
> Signed-off-by: Chunhe Lan
> Signed-off-by: Poonam Aggrwal
> [Emil Medve: Sync with the upstream binding]
> Signed-off-by
From: Sowmini Varadhan
Date: Thu, 2 Apr 2015 08:51:52 -0400
> do I need to resubmit this without the RFC tag? Perhaps I should
> have dropped that some time ago.
I want to hear from the powerpc folks whether they can positively
adopt the new generic code or not.
_
On Thu, 2015-04-02 at 06:16 -0500, Zhao Chenhui-B35336 wrote:
>
>
> From: Wood Scott-B07421
> Sent: Tuesday, March 31, 2015 10:07
> To: Zhao Chenhui-B35336
> Cc: linuxppc-dev@lists.ozlabs.org; devicet...@vger.kernel.org;
> linux-ker...@vger.kernel.org; Jin
On 04/01/2015 07:36 AM, Igal.Liberman wrote:
From: Igal Liberman
The FMan FLib provides the basic API used by the FMan drivers to
configure and control the FMan hardware.
Signed-off-by: Igal Liberman
---
drivers/soc/fsl/Kconfig |2 +
drivers/soc/fsl/Makefile |2 +
drive
On Thu, 2015-04-02 at 05:33 -0500, Zhao Chenhui-B35336 wrote:
> > +static int rcpm_v2_plat_enter_state(int state)
> > +{
> > + u32 *pmcsr_reg = &rcpm_v2_regs->powmgtcsr;
> > + int ret = 0;
> > + int result;
> > +
> > + switch (state) {
> > + case PLAT_PM_LPM20:
> > +
On Tuesday 31 March 2015 16:00:42 Daniel Axtens wrote:
> Introduces the pci_controller_ops structure.
> Add pci_controller_ops.dma_dev_setup, shadowing ppc_md.pci_dma_dev_setup.
> Add a shim, and change the callsites to use the shim.
>
> Signed-off-by: Daniel Axtens
>
> ---
>
> v1 --> v2:
> -
On Thu, Apr 02, 2015 at 02:12:47PM +0200, Ingo Molnar wrote:
> Okay. I'll resurrect the fix with a hotplug_cpu__tick_pull() name -
> agreed?
Sure.
> > > Also, I improved the changelog (attached below), but decided
> > > against applying it until these questions are cleared - please use
> > > t
On (03/31/15 23:12), David Miller wrote:
>
> It's much more amortized with smart buffering strategies, which are
> common on current generation networking cards.
>
> There you only eat one map/unmap per "PAGE_SIZE / rx_pkt_size".
>
> Maybe the infiniband stuff is doing things very suboptimally,
On 04/02/2015 05:42 PM, Ingo Molnar wrote:
>
> * Peter Zijlstra wrote:
>
>> On Thu, Apr 02, 2015 at 12:42:27PM +0200, Ingo Molnar wrote:
>>> So why not use a suitable CPU_DOWN* notifier for this, instead of open
>>> coding it all into a random place in the hotplug machinery?
>>
>> Because notif
Hi Sandeep,
Small nit inside.
On 04/02/2015 01:47 PM, sand...@freescale.com wrote:
> From: Sandeep Singh
>
> This controller is available on many Freescale SOCs like MPC8315, P1020,
> P1010,
> P1022 and P1024
>
> Signed-off-by: Sandeep Singh
> Signed-off-by: Poonam Aggrwal
> ---
> Document
* Peter Zijlstra wrote:
> On Thu, Apr 02, 2015 at 12:42:27PM +0200, Ingo Molnar wrote:
> > So why not use a suitable CPU_DOWN* notifier for this, instead of open
> > coding it all into a random place in the hotplug machinery?
>
> Because notifiers are crap? ;-) [...]
No doubt - but I didn't f
On Thu, Apr 02, 2015 at 12:42:27PM +0200, Ingo Molnar wrote:
> So why not use a suitable CPU_DOWN* notifier for this, instead of open
> coding it all into a random place in the hotplug machinery?
Because notifiers are crap? ;-) Its entirely impossible to figure out
what's happening to core code i
On 04/02/2015 05:01 PM, Ingo Molnar wrote:
>
> * Preeti U Murthy wrote:
>
>> On 04/02/2015 04:12 PM, Ingo Molnar wrote:
>>>
>>> * Preeti U Murthy wrote:
>>>
It was found when doing a hotplug stress test on POWER, that the machine
either hit softlockups or rcu_sched stall warnings. Th
From: Wood Scott-B07421
Sent: Tuesday, March 31, 2015 10:35
To: Zhao Chenhui-B35336
Cc: linuxppc-dev@lists.ozlabs.org; devicet...@vger.kernel.org;
linux-ker...@vger.kernel.org; Jin Zhengxiong-R64188
Subject: Re: [4/4] powerpc/85xx: support sleep feature o
* Preeti U Murthy wrote:
> On 04/02/2015 04:12 PM, Ingo Molnar wrote:
> >
> > * Preeti U Murthy wrote:
> >
> >> It was found when doing a hotplug stress test on POWER, that the machine
> >> either hit softlockups or rcu_sched stall warnings. The issue was
> >> traced to commit 7cba160ad789a
On 04/02/2015 04:12 PM, Ingo Molnar wrote:
>
> * Preeti U Murthy wrote:
>
>> It was found when doing a hotplug stress test on POWER, that the machine
>> either hit softlockups or rcu_sched stall warnings. The issue was
>> traced to commit 7cba160ad789a powernv/cpuidle: Redesign idle states
>> m
From: Wood Scott-B07421
Sent: Tuesday, March 31, 2015 10:07
To: Zhao Chenhui-B35336
Cc: linuxppc-dev@lists.ozlabs.org; devicet...@vger.kernel.org;
linux-ker...@vger.kernel.org; Jin Zhengxiong-R64188
Subject: Re: [3/4] powerpc: support CPU hotplug for e500
On 04/01/2015 06:19 PM, Madalin Bucur wrote:
> This is the second version of a patch series that adds
> the Ethernet driver for the Freescale QorIQ Data Path
> Acceleration Architecture (DPAA).
>
> This second version includes the changes requested by
> Kumar Gala. The PM ops patch was removed as
From: Sandeep Singh
Add device tree for TDM on QorIQ chips.
Signed-off-by: Sandeep Singh
Signed-off-by: Poonam Aggrwal
---
arch/powerpc/boot/dts/fsl/qoriq-tdm1.0.dtsi | 41 +
1 file changed, 41 insertions(+)
create mode 100644 arch/powerpc/boot/dts/fsl/qoriq-tdm1.
From: Sandeep Singh
Added dtsi file for Freescale TDM controller.
This controller is available on many Freescale SOCs like MPC8315, P1020, P1010,
P1022 and P1024
Signed-off-by: Sandeep Singh
Signed-off-by: Poonam Aggrwal
---
arch/powerpc/boot/dts/fsl/pq3-tdm1.0-0.dtsi | 41 +++
From: Sandeep Singh
This controller is available on many Freescale SOCs like MPC8315, P1020, P1010,
P1022 and P1024
Signed-off-by: Sandeep Singh
Signed-off-by: Poonam Aggrwal
---
Documentation/devicetree/bindings/tdm/fsl-tdm.txt | 65 +
1 files changed, 65 insertions(+),
Just a few nits.
This series is posted as an RFC, so this might not be what you're
expecting right now. But as these messages got tangled up in my mail
filter anyhow, I thought I might as well bother you with these nits now.
On Wed, 2015-04-01 at 19:19 +0300, Madalin Bucur wrote:
> --- /dev/null
On Tue, Mar 03, 2015 at 02:26:22PM +0800, xuelin@freescale.com wrote:
> From: Xuelin Shi
>
> The RaidEngine is a new FSL hardware used for Raid5/6 acceration.
> This patch enables the RaidEngine functionality and provides
> hardware offloading capability for memcpy, xor and pq computation.
>
* Preeti U Murthy wrote:
> It was found when doing a hotplug stress test on POWER, that the machine
> either hit softlockups or rcu_sched stall warnings. The issue was
> traced to commit 7cba160ad789a powernv/cpuidle: Redesign idle states
> management, which exposed the cpu down race with hrtim
From: Wood Scott-B07421
Sent: Tuesday, March 31, 2015 9:30
To: Zhao Chenhui-B35336
Cc: linuxppc-dev@lists.ozlabs.org; devicet...@vger.kernel.org;
linux-ker...@vger.kernel.org; Jin Zhengxiong-R64188
Subject: Re: [2/4] powerpc/rcpm: add RCPM driver
On Thu,
From: Wood Scott-B07421
Sent: Tuesday, March 31, 2015 9:10
To: Zhao Chenhui-B35336
Cc: linuxppc-dev@lists.ozlabs.org; devicet...@vger.kernel.org;
linux-ker...@vger.kernel.org; Jin Zhengxiong-R64188
Subject: Re: [1/4] powerpc/cache: add cache flush operatio
Hi Linus,
Please pull one powerpc fix for 4.0:
The following changes since commit e42391cd048809d903291d07f86ed3934ce138e9:
Linux 4.0-rc6 (2015-03-29 15:26:31 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux.git tags/powerpc-4.0-4
for
"Aneesh Kumar K.V" writes:
> Michael Ellerman writes:
>
> With patch:
> sys: 0m11.3258
>
> ie, a -0.7% impact
>
> If that impact is high we could possibly put that tracepoint within #ifdef
> CONFIG_DEBUG_VM ?
Since the ebizzy runs results were not stable, I did a micro benchmark
to m
58 matches
Mail list logo
