* [PATCH] x86: Use online node real index in calulate_tbl_offset()
@ 2010-10-29 2:18 Yinghai Lu
2010-10-29 3:02 ` H. Peter Anvin
2010-10-31 3:38 ` Shaohua Li
0 siblings, 2 replies; 4+ messages in thread
From: Yinghai Lu @ 2010-10-29 2:18 UTC (permalink / raw)
To: H. Peter Anvin, Ingo Molnar, Thomas Gleixner
Cc: Shaohua Li, Eric Dumazet, Andrew Morton, linux-kernel
Found one numa system that doesn't have ram installed in first socket
hang during executing init scripts.
bisect to:
|commit 932967202182743c01a2eee4bdfa2c42697bc586
|Author: Shaohua Li <shaohua.li@intel.com>
|Date: Wed Oct 20 11:07:03 2010 +0800
|
| x86: Spread tlb flush vector between nodes
It turns out when first socket is not online could have cpus on node1
tlb_offset set to bigger than NUM_INVALIDATE_TLB_VECTORS.
that could affect systems like 4 sockets, but socket 2 doesn't
have installed, sockets 3 will get too big tlb_offset.
Need to use real online node idx.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
---
arch/x86/mm/tlb.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
Index: linux-2.6/arch/x86/mm/tlb.c
===================================================================
--- linux-2.6.orig/arch/x86/mm/tlb.c
+++ linux-2.6/arch/x86/mm/tlb.c
@@ -223,7 +223,7 @@ void native_flush_tlb_others(const struc
static void __cpuinit calculate_tlb_offset(void)
{
- int cpu, node, nr_node_vecs;
+ int cpu, node, nr_node_vecs, idx = 0;
/*
* we are changing tlb_vector_offset for each CPU in runtime, but this
* will not cause inconsistency, as the write is atomic under X86. we
@@ -239,7 +239,7 @@ static void __cpuinit calculate_tlb_offs
nr_node_vecs = NUM_INVALIDATE_TLB_VECTORS/nr_online_nodes;
for_each_online_node(node) {
- int node_offset = (node % NUM_INVALIDATE_TLB_VECTORS) *
+ int node_offset = (idx % NUM_INVALIDATE_TLB_VECTORS) *
nr_node_vecs;
int cpu_offset = 0;
for_each_cpu(cpu, cpumask_of_node(node)) {
@@ -248,6 +248,7 @@ static void __cpuinit calculate_tlb_offs
cpu_offset++;
cpu_offset = cpu_offset % nr_node_vecs;
}
+ idx++;
}
}
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] x86: Use online node real index in calulate_tbl_offset()
2010-10-29 2:18 [PATCH] x86: Use online node real index in calulate_tbl_offset() Yinghai Lu
@ 2010-10-29 3:02 ` H. Peter Anvin
2010-10-31 3:38 ` Shaohua Li
1 sibling, 0 replies; 4+ messages in thread
From: H. Peter Anvin @ 2010-10-29 3:02 UTC (permalink / raw)
To: Yinghai Lu
Cc: Ingo Molnar, Thomas Gleixner, Shaohua Li, Eric Dumazet,
Andrew Morton, linux-kernel
Shaohua, does this look right to you?
-hpa
On 10/28/2010 07:18 PM, Yinghai Lu wrote:
>
> Found one numa system that doesn't have ram installed in first socket
> hang during executing init scripts.
>
> bisect to:
>
> |commit 932967202182743c01a2eee4bdfa2c42697bc586
> |Author: Shaohua Li <shaohua.li@intel.com>
> |Date: Wed Oct 20 11:07:03 2010 +0800
> |
> | x86: Spread tlb flush vector between nodes
>
> It turns out when first socket is not online could have cpus on node1
> tlb_offset set to bigger than NUM_INVALIDATE_TLB_VECTORS.
>
> that could affect systems like 4 sockets, but socket 2 doesn't
> have installed, sockets 3 will get too big tlb_offset.
>
> Need to use real online node idx.
>
> Signed-off-by: Yinghai Lu <yinghai@kernel.org>
>
> ---
> arch/x86/mm/tlb.c | 5 +++--
> 1 file changed, 3 insertions(+), 2 deletions(-)
>
> Index: linux-2.6/arch/x86/mm/tlb.c
> ===================================================================
> --- linux-2.6.orig/arch/x86/mm/tlb.c
> +++ linux-2.6/arch/x86/mm/tlb.c
> @@ -223,7 +223,7 @@ void native_flush_tlb_others(const struc
>
> static void __cpuinit calculate_tlb_offset(void)
> {
> - int cpu, node, nr_node_vecs;
> + int cpu, node, nr_node_vecs, idx = 0;
> /*
> * we are changing tlb_vector_offset for each CPU in runtime, but this
> * will not cause inconsistency, as the write is atomic under X86. we
> @@ -239,7 +239,7 @@ static void __cpuinit calculate_tlb_offs
> nr_node_vecs = NUM_INVALIDATE_TLB_VECTORS/nr_online_nodes;
>
> for_each_online_node(node) {
> - int node_offset = (node % NUM_INVALIDATE_TLB_VECTORS) *
> + int node_offset = (idx % NUM_INVALIDATE_TLB_VECTORS) *
> nr_node_vecs;
> int cpu_offset = 0;
> for_each_cpu(cpu, cpumask_of_node(node)) {
> @@ -248,6 +248,7 @@ static void __cpuinit calculate_tlb_offs
> cpu_offset++;
> cpu_offset = cpu_offset % nr_node_vecs;
> }
> + idx++;
> }
> }
>
--
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel. I don't speak on their behalf.
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH] x86: Use online node real index in calulate_tbl_offset()
2010-10-29 2:18 [PATCH] x86: Use online node real index in calulate_tbl_offset() Yinghai Lu
2010-10-29 3:02 ` H. Peter Anvin
@ 2010-10-31 3:38 ` Shaohua Li
1 sibling, 0 replies; 4+ messages in thread
From: Shaohua Li @ 2010-10-31 3:38 UTC (permalink / raw)
To: Yinghai Lu
Cc: H. Peter Anvin, Ingo Molnar, Thomas Gleixner, Eric Dumazet,
Andrew Morton, linux-kernel
On Fri, 2010-10-29 at 10:18 +0800, Yinghai Lu wrote:
> Found one numa system that doesn't have ram installed in first socket
> hang during executing init scripts.
>
> bisect to:
>
> |commit 932967202182743c01a2eee4bdfa2c42697bc586
> |Author: Shaohua Li <shaohua.li@intel.com>
> |Date: Wed Oct 20 11:07:03 2010 +0800
> |
> | x86: Spread tlb flush vector between nodes
>
> It turns out when first socket is not online could have cpus on node1
> tlb_offset set to bigger than NUM_INVALIDATE_TLB_VECTORS.
>
> that could affect systems like 4 sockets, but socket 2 doesn't
> have installed, sockets 3 will get too big tlb_offset.
>
> Need to use real online node idx.
>
Thanks for catching it up.
Acked-by: Shaohua Li <shaohua.li@intel.com>
^ permalink raw reply [flat|nested] 4+ messages in thread
* [PATCH] x86: Use online node real index in calulate_tbl_offset()
@ 2010-11-13 18:52 Yinghai Lu
0 siblings, 0 replies; 4+ messages in thread
From: Yinghai Lu @ 2010-11-13 18:52 UTC (permalink / raw)
To: H. Peter Anvin, Ingo Molnar, Thomas Gleixner, Andrew Morton
Cc: linux-kernel, Linus Torvalds, Shaohua Li
--- resending, hope in can make into -rc2.
Found one numa system that doesn't have ram installed with first socket
hang during executing init scripts.
bisect to:
|commit 932967202182743c01a2eee4bdfa2c42697bc586
|Author: Shaohua Li <shaohua.li@intel.com>
|Date: Wed Oct 20 11:07:03 2010 +0800
|
| x86: Spread tlb flush vector between nodes
It turns out when first socket is not online could have cpus on node1
tlb_offset set to bigger than NUM_INVALIDATE_TLB_VECTORS.
that could affect systems like 4 sockets, but socket 2 doesn't
have installed, sockets 3 will get too big tlb_offset.
Need to use real online node idx.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Acked-by: Shaohua Li <shaohua.li@intel.com>
---
arch/x86/mm/tlb.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
Index: linux-2.6/arch/x86/mm/tlb.c
===================================================================
--- linux-2.6.orig/arch/x86/mm/tlb.c
+++ linux-2.6/arch/x86/mm/tlb.c
@@ -223,7 +223,7 @@ void native_flush_tlb_others(const struc
static void __cpuinit calculate_tlb_offset(void)
{
- int cpu, node, nr_node_vecs;
+ int cpu, node, nr_node_vecs, idx = 0;
/*
* we are changing tlb_vector_offset for each CPU in runtime, but this
* will not cause inconsistency, as the write is atomic under X86. we
@@ -239,7 +239,7 @@ static void __cpuinit calculate_tlb_offs
nr_node_vecs = NUM_INVALIDATE_TLB_VECTORS/nr_online_nodes;
for_each_online_node(node) {
- int node_offset = (node % NUM_INVALIDATE_TLB_VECTORS) *
+ int node_offset = (idx % NUM_INVALIDATE_TLB_VECTORS) *
nr_node_vecs;
int cpu_offset = 0;
for_each_cpu(cpu, cpumask_of_node(node)) {
@@ -248,6 +248,7 @@ static void __cpuinit calculate_tlb_offs
cpu_offset++;
cpu_offset = cpu_offset % nr_node_vecs;
}
+ idx++;
}
}
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2010-11-13 18:53 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-10-29 2:18 [PATCH] x86: Use online node real index in calulate_tbl_offset() Yinghai Lu
2010-10-29 3:02 ` H. Peter Anvin
2010-10-31 3:38 ` Shaohua Li
2010-11-13 18:52 Yinghai Lu
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.