On 24/03/16 02:58, Torsten Duwe wrote:
> Since nobody liked the extra stack frame nor its workarounds, here is
> the next attempt. Assumptions:
>
> 1. Heuristics are bad. The better they are, the more subtly the
>way they might fail.
>
> 2. The TOC pointer is usually dividable by 4, if not by
IBM POWER8 NVlink systems come with Tesla K40-ish GPUs each of which
also has a couple of fast speed links (NVLink). The interface to links
is exposed as an emulated PCI bridge which is included into the same
IOMMU group as the corresponding GPU.
In the kernel, NPUs get a separate PHB of the PNV_P
On Wed, 2016-03-23 at 21:39 +1100, Michael Ellerman wrote:
> On Wed, 2016-03-23 at 13:52 +1100, Russell Currey wrote:
> >
> > On Wed, 2016-03-23 at 11:38 +1100, Michael Ellerman wrote:
> > >
> > > On Tue, 2016-22-03 at 00:34:55 UTC, Russell Currey wrote:
> > > >
> > > >
> > > > +
On Tuesday, March 22, 2016 07:55:07 PM Viresh Kumar wrote:
> On 22-03-16, 18:57, Shilpasri G Bhat wrote:
> > Create sysfs attributes to export throttle information in
> > /sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory. The
> > newly added sysfs files are as follows:
> >
> > 1)/sys/d
> On Mar 23, 2016, at 2:50 PM, Uma Krishnan wrote:
>
> From: "Manoj N. Kumar"
>
> While profiling the cxlflash_queuecommand() path under a heavy load it
> was found that number of retries to find cmd_room was fairly high.
>
> There are two problems with the current back-off:
> a) It starts wit
> On Mar 23, 2016, at 2:50 PM, Uma Krishnan wrote:
>
> From: "Manoj N. Kumar"
>
> While running 'sg_reset -H' back to back the following exception was seen:
>
> [ 735.115695] Faulting instruction address: 0xd98c0864
> cpu 0x0: Vector: 300 (Data Access) at [c00afa80]
>pc: d
From: "Manoj N. Kumar"
While profiling the cxlflash_queuecommand() path under a heavy load it
was found that number of retries to find cmd_room was fairly high.
There are two problems with the current back-off:
a) It starts with a udelay of 0
b) It backs-off linearly
Tried several approaches (a
From: "Manoj N. Kumar"
While running 'sg_reset -H' back to back the following exception was seen:
[ 735.115695] Faulting instruction address: 0xd98c0864
cpu 0x0: Vector: 300 (Data Access) at [c00afa80]
pc: d98c0864: cxlflash_async_err_irq+0x84/0x5c0 [cxlflash]
lr
This series has a couple of patches in cxlflash.
The first patch is critical as it fixes a regression that was introduced
by Commit 6ded8b3cbd9a ("cxlflash: Unmap problem state area before
detaching master context") as part of the first SCSI push for v4.6.
The second patch addresses a performance
On Wednesday 23 March 2016 07:00 PM, Michal Hocko wrote:
> On Wed 23-03-16 17:37:18, Vaishali Thakkar wrote:
>> Current code fails to ignore the 'hugepages=' parameters when unsupported
>> hugepagesize is specified. With this patchset, introduce new architecture
>> independent routine hugetlb_bad
> -Original Message-
> From: Linuxppc-dev [mailto:linuxppc-dev-
> bounces+varun.sethi=freescale@lists.ozlabs.org] On Behalf Of Madalin-
> Cristian Bucur
> Sent: Wednesday, March 23, 2016 3:29 PM
> To: j...@8bytes.org; Andy Fleming
> Cc: io...@lists.linux-foundation.org; linuxppc-dev@
Since nobody liked the extra stack frame nor its workarounds, here is
the next attempt. Assumptions:
1. Heuristics are bad. The better they are, the more subtly the
way they might fail.
2. The TOC pointer is usually dividable by 4, if not by 8. An odd
value never occurs.
Conclusively, thi
On Wed 23-03-16 17:37:18, Vaishali Thakkar wrote:
> Current code fails to ignore the 'hugepages=' parameters when unsupported
> hugepagesize is specified. With this patchset, introduce new architecture
> independent routine hugetlb_bad_size to handle such command line options.
> And then call it in
Update the setup_hugepagesz function to call the routine
hugetlb_bad_size when unsupported hugepage size is found.
Misc:
- Silent checkpatch.pl's 80 characters and printk warning
Signed-off-by: Vaishali Thakkar
Reviewed-by: Mike Kravetz
Reviewed-by: Naoya Horiguchi
Cc: Hillf Danton
Cc: Mic
Current code fails to ignore the 'hugepages=' parameters when unsupported
hugepagesize is specified. With this patchset, introduce new architecture
independent routine hugetlb_bad_size to handle such command line options.
And then call it in architecture specific code.
Changes since v1:
-
> -Original Message-
> From: Andy Fleming
> To: j...@8bytes.org
> Cc: io...@lists.linux-foundation.org, linuxppc-dev@lists.ozlabs.org
> Subject: [v2] powerpc: Fix incorrect PPC32 PAMU dependency
>
> The Freescale PAMU can be enabled on both 32 and 64-bit Power
> chips. Commit 477ab7a19cec8
On Wed, 2016-03-23 at 13:52 +1100, Russell Currey wrote:
> On Wed, 2016-03-23 at 11:38 +1100, Michael Ellerman wrote:
> > On Tue, 2016-22-03 at 00:34:55 UTC, Russell Currey wrote:
> > >
> > > + case RTAS_EXTENDED_DELAY_MIN+2:
> > > + mwait = 100;
> > > + brea
17 matches
Mail list logo