From: Julia Lawall
Initialize ret before returning on failure, as done elsewhere in the
function.
A simplified version of the semantic match that finds this problem is as
follows: (http://coccinelle.lip6.fr/)
//
(
if@p1 (\(ret < 0\|ret != 0\))
{ ... return ret; }
|
ret@p1 = 0
)
... when != re
In iperf experiments running linux as the Tx side (TCP client) with
10 threads results in a severe performance drop when TSO is disabled,
indicating a weakness in the software that can be avoided by using
the scalable IOMMU arena DMA allocation.
Baseline numbers before this patch:
with default
Investigation of multithreaded iperf experiments on an ethernet
interface show the iommu->lock as the hottest lock identified by
lockstat, with something of the order of 21M contentions out of
27M acquisitions, and an average wait time of 26 us for the lock.
This is not efficient. A more scalable
Note that this conversion is only being done to consolidate the
code and ensure that the common code provides the sufficient
abstraction. It is not expected to result in any noticeable
performance improvement, as there is typically one ldc_iommu
per vnet_port, and each one has 8k entries, with a ty
Addresses latest BenH comments: need_flush checks, add support
for dma mask and align_order.
Sowmini Varadhan (3):
Break up monolithic iommu table/lock into finer graularity pools and
lock
Make sparc64 use scalable lib/iommu-common.c functions
Make LDC use common iommu poll management f
On Sun, Apr 5, 2015 at 1:38 PM, Julia Lawall wrote:
> From: Julia Lawall
>
> Initialize ret before returning on failure, as done elsewhere in the
> function.
> diff --git a/sound/ppc/snd_ps3.c b/sound/ppc/snd_ps3.c
> index 1aa52ef..9b18b52 100644
> --- a/sound/ppc/snd_ps3.c
> +++ b/sound/ppc/snd
Please ignore. Wrong patch set.
On Sun, 5 Apr 2015, Julia Lawall wrote:
> From: Julia Lawall
>
> Initialize ret before returning on failure, as done elsewhere in the
> function.
>
> A simplified version of the semantic match that finds this problem is as
> follows: (http://coccinelle.lip6.fr/)
On Sun, 2015-04-05 at 07:49 -0400, Sowmini Varadhan wrote:
> + if (likely(pass == 0)) {
> + /* First failure, rescan from the beginning.
> */
> + pool->hint = pool->start;
> + set_flush(iommu);
> +
These structs can be const, so make them const.
Miscellanea:
o Add static to struct uart_ops declaration in bfin_sport_uart
Signed-off-by: Joe Perches
---
On Sun, 2015-04-05 at 00:04 +0200, Stefan Agner wrote:
> On 2015-04-04 19:47, Joe Perches wrote:
> > Done with:
> > sed -r -i -e 's/\bstruct
On (04/05/15 22:26), Benjamin Herrenschmidt wrote:
>
> So you decided to keep the logic here that updates the hint instead of
> just getting rid of need_flush alltogether ?
>
> Out of curiosity, what's the rationale ? Did you find a reason why
> resetting the hint in those two cases (rather than
On Sun, 5 Apr 2015, Joe Perches wrote:
> drivers/tty/serial/dz.c | 2 +-
> drivers/tty/serial/zs.c | 2 +-
For these verified that the change works correctly, thanks for your work.
Acked-by: Maciej W. Rozycki
Maciej
___
On Mon, 2015-04-06 at 01:09 +0200, Stefan Agner wrote:
> On 2015-04-05 20:24, Joe Perches wrote:
> > Still compiled only x86 and not cross-compiled or tested.
>
> Cross-compiled with multi_v7_defconfig
[]
> > drivers/tty/serial/fsl_lpuart.c | 4 ++--
>
> Tested that driver on actual h
12 matches
Mail list logo
