Re: [PATCH net] rds: Reintroduce statistics counting

2017-08-08 Thread Shamir Rabinovitch
rds_ib_recv_refill(conn, 0, GFP_KERNEL); > + rds_ib_stats_inc(s_ib_rx_refill_from_thread); > } > > return ret; > -- > 2.9.3 > Reviewed-by: Shamir Rabinovitch <shamir.rabinovi...@oracle.com>

[PATCH RDS v1] rds: debug messages are enabled by default

2016-10-27 Thread shamir . rabinovitch
From: shamir rabinovitch <shamir.rabinovi...@oracle.com> rds use Kconfig option called "RDS_DEBUG" to enable rds debug messages. This option cause the rds Makefile to add -DDEBUG to the rds gcc command line. When CONFIG_DYNAMIC_DEBUG is enabled, the "DEBUG" macr

[PATCH v4 1/2] RDS: memory allocated must be align to 8

2016-04-07 Thread Shamir Rabinovitch
Fix issue in 'rds_ib_cong_recv' when accessing unaligned memory allocated by 'rds_page_remainder_alloc' using uint64_t pointer. Signed-off-by: Shamir Rabinovitch <shamir.rabinovi...@oracle.com> --- net/rds/page.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/n

[PATCH v4 2/2] RDS: fix congestion map corruption for PAGE_SIZE > 4k

2016-04-07 Thread Shamir Rabinovitch
When PAGE_SIZE > 4k single page can contain 2 RDS fragments. If 'rds_ib_cong_recv' ignore the RDS fragment offset in to the page it then read the data fragment as far congestion map update and lead to corruption of the RDS connection far congestion map. Signed-off-by: Shamir Rabinovi

Re: [PATCH v2] rds: rds-stress show all zeros after few minutes

2016-04-03 Thread Shamir Rabinovitch
On Thu, Mar 31, 2016 at 04:02:46PM -0400, David Miller wrote: > From: shamir rabinovitch <shamir.rabinovi...@oracle.com> > Date: Thu, 31 Mar 2016 02:29:22 -0400 > > > Issue can be seen on platforms that use 8K and above page size > > while rds fragment size is 4K. On

[PATCH v3 1/2] RDS: fix "Kernel unaligned access" on sparc64

2016-04-03 Thread Shamir Rabinovitch
Sparc64 expect 64 bit types to be aligned to 8. This patch ensure that memory allocated by 'rds_page_remainder_alloc' will be aligned to 8. This patch fix issue found in 'rds_ib_cong_recv' when accessing unaligned memory using uint64_t pointer. Signed-off-by: Shamir Rabinovitch <shamir.rabin

[PATCH v3 2/2] RDS: fix congestion map corruption for PAGE_SIZE > 8k

2016-04-03 Thread Shamir Rabinovitch
for Sparc64 platforms: 001f8f8 RDS: fix "Kernel unaligned access" on sparc64 Signed-off-by: Shamir Rabinovitch <shamir.rabinovi...@oracle.com> Reviewed-by: Wengang Wang <wen.gang.w...@oracle.com> Reviewed-by: Ajaykumar Hotchandani <ajaykumar.hotchand...@oracle.com>

[PATCH v2] rds: rds-stress show all zeros after few minutes

2016-03-31 Thread shamir rabinovitch
to account lead to reading the data fragment as congestion map fragment and hang of the rds transmit due to far congestion map corruption. Signed-off-by: shamir rabinovitch <shamir.rabinovi...@oracle.com> Reviewed-by: Wengang Wang <wen.gang.w...@oracle.com> Reviewed-by: Ajaykumar

Re: [PATCH] rds: rds-stress show all zeros after few minutes

2016-03-31 Thread Shamir Rabinovitch
On Thu, Mar 31, 2016 at 05:23:30PM +0300, Sergei Shtylyov wrote: > On 3/31/2016 5:20 PM, Shamir Rabinovitch wrote: > > >>>Issue can be seen on platforms that use 8K and above page size > >>>while rds fragment size is 4K. On those platforms single page is >

Re: [PATCH] rds: rds-stress show all zeros after few minutes

2016-03-31 Thread Shamir Rabinovitch
On Thu, Mar 31, 2016 at 04:13:35PM +0300, Sergei Shtylyov wrote: > Hello. > > On 3/31/2016 3:50 AM, shamir rabinovitch wrote: > > >Issue can be seen on platforms that use 8K and above page size > >while rds fragment size is 4K. On those platforms single page is > &g

[PATCH] rds: rds-stress show all zeros after few minutes

2016-03-31 Thread shamir rabinovitch
;santosh.shilim...@oracle.com> Tested-by: Anand Bibhuti <anand.bibh...@oracle.com> Signed-off-by: shamir rabinovitch <shamir.rabinovi...@oracle.com> --- net/rds/ib_recv.c |2 +- net/rds/iw_recv.c |2 +- net/rds/page.c|5 +++-- 3 files changed, 5 insertions(+), 4 deletions(-) di