* Re: ppc: RECLAIM_DISTANCE 10?
From: Michal Hocko @ 2014-02-19 9:19 UTC (permalink / raw)
To: David Rientjes; +Cc: linuxppc-dev, Anton Blanchard, LKML, linux-mm
In-Reply-To: <alpine.DEB.2.02.1402190017480.7280@chino.kir.corp.google.com>
On Wed 19-02-14 00:20:21, David Rientjes wrote:
> On Wed, 19 Feb 2014, Michal Hocko wrote:
>
> > > I strongly suspect that the patch is correct since powerpc node distances
> > > are different than the architectures you're talking about and get doubled
> > > for every NUMA domain that the hardware supports.
> >
> > Even if the units of the distance is different on PPC should every NUMA
> > machine have zone_reclaim enabled? That doesn't right to me.
> >
>
> In my experience on powerpc it's very correct, there's typically a
> significant latency in remote access and we don't have the benefit of a
> SLIT that actually defines the locality between proximity domains like we
> do on other architectures.
Interesting. So is the PPC NUMA basically about local vs. very distant?
Should REMOTE_DISTANCE reflect that as well? Or can we have
distance < REMOTE_DISTANCE and it would still make sense to have
zone_reclaim enabled?
[...]
Thanks!
--
Michal Hocko
SUSE Labs
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: Michal Hocko @ 2014-02-19 8:33 UTC (permalink / raw)
To: David Rientjes
Cc: linux-mm, Nishanth Aravamudan, linuxppc-dev, Anton Blanchard,
LKML
In-Reply-To: <alpine.DEB.2.02.1402181737530.17521@chino.kir.corp.google.com>
On Tue 18-02-14 17:43:38, David Rientjes wrote:
> On Tue, 18 Feb 2014, Nishanth Aravamudan wrote:
>
> > How about the following?
> >
> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > index 5de4337..1a0eced 100644
> > --- a/mm/page_alloc.c
> > +++ b/mm/page_alloc.c
> > @@ -1854,7 +1854,8 @@ static void __paginginit init_zone_allows_reclaim(int nid)
> > int i;
> >
> > for_each_online_node(i)
> > - if (node_distance(nid, i) <= RECLAIM_DISTANCE)
> > + if (node_distance(nid, i) <= RECLAIM_DISTANCE ||
> > + !NODE_DATA(i)->node_present_pages)
> > node_set(i, NODE_DATA(nid)->reclaim_nodes);
> > else
> > zone_reclaim_mode = 1;
>
> [ I changed the above from NODE_DATA(nid) -> NODE_DATA(i) as you caught
> so we're looking at the right code. ]
>
> That can't be right, it would allow reclaiming from a memoryless node. I
> think what you want is
>
> for_each_online_node(i) {
> if (!node_present_pages(i))
> continue;
> if (node_distance(nid, i) <= RECLAIM_DISTANCE) {
> node_set(i, NODE_DATA(nid)->reclaim_nodes);
> continue;
> }
> /* Always try to reclaim locally */
> zone_reclaim_mode = 1;
> }
>
> but we really should be able to do for_each_node_state(i, N_MEMORY) here
> and memoryless nodes should already be excluded from that mask.
Agreed! Actually the code I am currently interested in is based on 3.0
kernel where zone_reclaim_mode is set in build_zonelists which relies on
find_next_best_node which iterates only N_HIGH_MEMORY nodes which should
have non 0 present pages.
[...]
--
Michal Hocko
SUSE Labs
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: Michal Hocko @ 2014-02-19 8:23 UTC (permalink / raw)
To: Nishanth Aravamudan; +Cc: linux-mm, linuxppc-dev, Anton Blanchard, LKML
In-Reply-To: <20140218233404.GB10844@linux.vnet.ibm.com>
On Tue 18-02-14 15:34:05, Nishanth Aravamudan wrote:
> Hi Michal,
>
> On 18.02.2014 [10:06:58 +0100], Michal Hocko wrote:
> > Hi,
> > I have just noticed that ppc has RECLAIM_DISTANCE reduced to 10 set by
> > 56608209d34b (powerpc/numa: Set a smaller value for RECLAIM_DISTANCE to
> > enable zone reclaim). The commit message suggests that the zone reclaim
> > is desirable for all NUMA configurations.
> >
> > History has shown that the zone reclaim is more often harmful than
> > helpful and leads to performance problems. The default RECLAIM_DISTANCE
> > for generic case has been increased from 20 to 30 around 3.0
> > (32e45ff43eaf mm: increase RECLAIM_DISTANCE to 30).
>
> Interesting.
>
> > I strongly suspect that the patch is incorrect and it should be
> > reverted. Before I will send a revert I would like to understand what
> > led to the patch in the first place. I do not see why would PPC use only
> > LOCAL_DISTANCE and REMOTE_DISTANCE distances and in fact machines I have
> > seen use different values.
> >
> > Anton, could you comment please?
>
> I'll let Anton comment here, but in looking into this issue in working
> on CONFIG_HAVE_MEMORYLESS_NODE support, I realized that any LPAR with
> memoryless nodes will set zone_reclaim_mode to 1. I think we want to
> ignore memoryless nodes when we set up the reclaim mode like the
> following? I'll send it as a proper patch if you agree?
Funny enough, ppc memoryless node setup is what led me to this code.
We had a setup like this:
node 0 cpus: 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
node 0 size: 0 MB
node 0 free: 0 MB
node 2 cpus:
node 2 size: 7168 MB
node 2 free: 6019 MB
node distances:
node 0 2
0: 10 40
2: 40 10
Which ends up enabling zone_reclaim although there is only a single node
with memory. Not that RECLAIM_DISTANCE would make any difference here as
the distance is even above default RECLAIM_DISTANCE.
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 5de4337..4f6ff6f 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -1853,8 +1853,9 @@ static void __paginginit init_zone_allows_reclaim(int nid)
> {
> int i;
>
> - for_each_online_node(i)
> - if (node_distance(nid, i) <= RECLAIM_DISTANCE)
> + for_each_online_node(i) {
> + if (node_distance(nid, i) <= RECLAIM_DISTANCE ||
> + local_memory_node(nid) != nid)
> node_set(i, NODE_DATA(nid)->reclaim_nodes);
> else
> zone_reclaim_mode = 1;
>
> Note, this won't actually do anything if CONFIG_HAVE_MEMORYLESS_NODES is
> not set, but if it is, I think semantically it will indicate that
> memoryless nodes *have* to reclaim remotely.
>
> And actually the above won't work, because the callpath is
>
> start_kernel -> setup_arch -> paging_init [-> free_area_init_nodes ->
> free_area_init_node -> init_zone_allows_reclaim] which is called before
> build_all_zonelists. This is a similar ordering problem as I'm having
> with the MEMORYLESS_NODE support, will work on it.
I think you just want for_each_node_state(nid, N_MEMORY) and skip all
the memory less nodes, no?
--
Michal Hocko
SUSE Labs
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: David Rientjes @ 2014-02-19 8:20 UTC (permalink / raw)
To: Michal Hocko; +Cc: linuxppc-dev, Anton Blanchard, LKML, linux-mm
In-Reply-To: <20140219081644.GA14783@dhcp22.suse.cz>
On Wed, 19 Feb 2014, Michal Hocko wrote:
> > I strongly suspect that the patch is correct since powerpc node distances
> > are different than the architectures you're talking about and get doubled
> > for every NUMA domain that the hardware supports.
>
> Even if the units of the distance is different on PPC should every NUMA
> machine have zone_reclaim enabled? That doesn't right to me.
>
In my experience on powerpc it's very correct, there's typically a
significant latency in remote access and we don't have the benefit of a
SLIT that actually defines the locality between proximity domains like we
do on other architectures. We have had significant issues with thp, for
example, being allocated remotely instead of pages locally, much more
drastic than on our x86 machines, particularly AMD machines.
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: Michal Hocko @ 2014-02-19 8:16 UTC (permalink / raw)
To: David Rientjes; +Cc: linuxppc-dev, Anton Blanchard, LKML, linux-mm
In-Reply-To: <alpine.DEB.2.02.1402181424490.20772@chino.kir.corp.google.com>
On Tue 18-02-14 14:27:11, David Rientjes wrote:
> On Tue, 18 Feb 2014, Michal Hocko wrote:
>
> > Hi,
> > I have just noticed that ppc has RECLAIM_DISTANCE reduced to 10 set by
> > 56608209d34b (powerpc/numa: Set a smaller value for RECLAIM_DISTANCE to
> > enable zone reclaim). The commit message suggests that the zone reclaim
> > is desirable for all NUMA configurations.
> >
> > History has shown that the zone reclaim is more often harmful than
> > helpful and leads to performance problems. The default RECLAIM_DISTANCE
> > for generic case has been increased from 20 to 30 around 3.0
> > (32e45ff43eaf mm: increase RECLAIM_DISTANCE to 30).
> >
> > I strongly suspect that the patch is incorrect and it should be
> > reverted. Before I will send a revert I would like to understand what
> > led to the patch in the first place. I do not see why would PPC use only
> > LOCAL_DISTANCE and REMOTE_DISTANCE distances and in fact machines I have
> > seen use different values.
> >
>
> I strongly suspect that the patch is correct since powerpc node distances
> are different than the architectures you're talking about and get doubled
> for every NUMA domain that the hardware supports.
Even if the units of the distance is different on PPC should every NUMA
machine have zone_reclaim enabled? That doesn't right to me.
--
Michal Hocko
SUSE Labs
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: David Rientjes @ 2014-02-19 1:43 UTC (permalink / raw)
To: Nishanth Aravamudan
Cc: Michal Hocko, linux-mm, linuxppc-dev, Anton Blanchard, LKML
In-Reply-To: <20140218235800.GC10844@linux.vnet.ibm.com>
On Tue, 18 Feb 2014, Nishanth Aravamudan wrote:
> How about the following?
>
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 5de4337..1a0eced 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -1854,7 +1854,8 @@ static void __paginginit init_zone_allows_reclaim(int nid)
> int i;
>
> for_each_online_node(i)
> - if (node_distance(nid, i) <= RECLAIM_DISTANCE)
> + if (node_distance(nid, i) <= RECLAIM_DISTANCE ||
> + !NODE_DATA(i)->node_present_pages)
> node_set(i, NODE_DATA(nid)->reclaim_nodes);
> else
> zone_reclaim_mode = 1;
[ I changed the above from NODE_DATA(nid) -> NODE_DATA(i) as you caught
so we're looking at the right code. ]
That can't be right, it would allow reclaiming from a memoryless node. I
think what you want is
for_each_online_node(i) {
if (!node_present_pages(i))
continue;
if (node_distance(nid, i) <= RECLAIM_DISTANCE) {
node_set(i, NODE_DATA(nid)->reclaim_nodes);
continue;
}
/* Always try to reclaim locally */
zone_reclaim_mode = 1;
}
but we really should be able to do for_each_node_state(i, N_MEMORY) here
and memoryless nodes should already be excluded from that mask.
> @@ -4901,13 +4902,13 @@ void __paginginit free_area_init_node(int nid, unsigned long *zones_size,
>
> pgdat->node_id = nid;
> pgdat->node_start_pfn = node_start_pfn;
> - init_zone_allows_reclaim(nid);
> #ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP
> get_pfn_range_for_nid(nid, &start_pfn, &end_pfn);
> #endif
> calculate_node_totalpages(pgdat, start_pfn, end_pfn,
> zones_size, zholes_size);
>
> + init_zone_allows_reclaim(nid);
> alloc_node_mem_map(pgdat);
> #ifdef CONFIG_FLAT_NODE_MEM_MAP
> printk(KERN_DEBUG "free_area_init_node: node %d, pgdat %08lx, node_mem_map %08lx\n",
>
> I think it's safe to move init_zone_allows_reclaim, because I don't
> think any allocates are occurring here that could cause us to reclaim
> anyways, right? Moving it allows us to safely reference
> node_present_pages.
>
Yeah, this is fine.
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: Nishanth Aravamudan @ 2014-02-19 0:40 UTC (permalink / raw)
To: Michal Hocko; +Cc: linux-mm, linuxppc-dev, Anton Blanchard, LKML
In-Reply-To: <20140218235800.GC10844@linux.vnet.ibm.com>
On 18.02.2014 [15:58:00 -0800], Nishanth Aravamudan wrote:
> On 18.02.2014 [15:34:05 -0800], Nishanth Aravamudan wrote:
> > Hi Michal,
> >
> > On 18.02.2014 [10:06:58 +0100], Michal Hocko wrote:
> > > Hi,
> > > I have just noticed that ppc has RECLAIM_DISTANCE reduced to 10 set by
> > > 56608209d34b (powerpc/numa: Set a smaller value for RECLAIM_DISTANCE to
> > > enable zone reclaim). The commit message suggests that the zone reclaim
> > > is desirable for all NUMA configurations.
> > >
> > > History has shown that the zone reclaim is more often harmful than
> > > helpful and leads to performance problems. The default RECLAIM_DISTANCE
> > > for generic case has been increased from 20 to 30 around 3.0
> > > (32e45ff43eaf mm: increase RECLAIM_DISTANCE to 30).
> >
> > Interesting.
> >
> > > I strongly suspect that the patch is incorrect and it should be
> > > reverted. Before I will send a revert I would like to understand what
> > > led to the patch in the first place. I do not see why would PPC use only
> > > LOCAL_DISTANCE and REMOTE_DISTANCE distances and in fact machines I have
> > > seen use different values.
> > >
> > > Anton, could you comment please?
> >
> > I'll let Anton comment here, but in looking into this issue in working
> > on CONFIG_HAVE_MEMORYLESS_NODE support, I realized that any LPAR with
> > memoryless nodes will set zone_reclaim_mode to 1. I think we want to
> > ignore memoryless nodes when we set up the reclaim mode like the
> > following? I'll send it as a proper patch if you agree?
> >
> > diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> > index 5de4337..4f6ff6f 100644
> > --- a/mm/page_alloc.c
> > +++ b/mm/page_alloc.c
> > @@ -1853,8 +1853,9 @@ static void __paginginit init_zone_allows_reclaim(int nid)
> > {
> > int i;
> >
> > - for_each_online_node(i)
> > - if (node_distance(nid, i) <= RECLAIM_DISTANCE)
> > + for_each_online_node(i) {
> > + if (node_distance(nid, i) <= RECLAIM_DISTANCE ||
> > + local_memory_node(nid) != nid)
> > node_set(i, NODE_DATA(nid)->reclaim_nodes);
> > else
> > zone_reclaim_mode = 1;
> >
> > Note, this won't actually do anything if CONFIG_HAVE_MEMORYLESS_NODES is
> > not set, but if it is, I think semantically it will indicate that
> > memoryless nodes *have* to reclaim remotely.
> >
> > And actually the above won't work, because the callpath is
> >
> > start_kernel -> setup_arch -> paging_init [-> free_area_init_nodes ->
> > free_area_init_node -> init_zone_allows_reclaim] which is called before
> > build_all_zonelists. This is a similar ordering problem as I'm having
> > with the MEMORYLESS_NODE support, will work on it.
>
> How about the following?
>
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 5de4337..1a0eced 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -1854,7 +1854,8 @@ static void __paginginit init_zone_allows_reclaim(int nid)
> int i;
>
> for_each_online_node(i)
> - if (node_distance(nid, i) <= RECLAIM_DISTANCE)
> + if (node_distance(nid, i) <= RECLAIM_DISTANCE ||
> + !NODE_DATA(nid)->node_present_pages)
err s/nid/i/ above.
-Nish
^ permalink raw reply
* Re: [PATCH][v2] powerpc/fsl: Add/update miscellaneous missing bindings
From: Scott Wood @ 2014-02-19 0:20 UTC (permalink / raw)
To: Harninder Rai; +Cc: devicetree, linuxppc-dev
In-Reply-To: <1392276554-10368-1-git-send-email-harninder.rai@freescale.com>
On Thu, 2014-02-13 at 12:59 +0530, Harninder Rai wrote:
> Missing bindings were found on running checkpatch.pl on bsc9132
> device tree. This patch add/update the following
>
> - Add bindings for L2 cache controller
> - Add bindings for memory controller
> - Update bindings for USB controller
>
> Signed-off-by: Harninder Rai <harninder.rai@freescale.com>
> ---
> Changes since base version:
> Incorporated Scott's comments
> - Rename l2cc.txt to l2cache.txt
> - Add information about ePAPR compliance
> - Add missing "cache" in compatible
> - Miscellaneous minors
>
> .../devicetree/bindings/powerpc/fsl/l2cache.txt | 26 ++++++++++++++++++++
> .../devicetree/bindings/powerpc/fsl/mem-ctrlr.txt | 16 ++++++++++++
> Documentation/devicetree/bindings/usb/fsl-usb.txt | 2 +
> 3 files changed, 44 insertions(+), 0 deletions(-)
> create mode 100644 Documentation/devicetree/bindings/powerpc/fsl/l2cache.txt
> create mode 100644 Documentation/devicetree/bindings/powerpc/fsl/mem-ctrlr.txt
>
> diff --git a/Documentation/devicetree/bindings/powerpc/fsl/l2cache.txt b/Documentation/devicetree/bindings/powerpc/fsl/l2cache.txt
> new file mode 100644
> index 0000000..79ef4a1
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/powerpc/fsl/l2cache.txt
> @@ -0,0 +1,26 @@
> +Freescale L2 Cache Controller
> +
> +L2 cache is present in Freescale's QorIQ and QorIQ Qonverge platforms.
> +The cache bindings explained below are ePAPR compliant
> +
> +Required Properties:
> +
> +- compatible : Should include "fsl,chip-l2-cache-controller" and "cache"
> + where chip is the processor (bsc9132, npc8572 etc.)
> +- reg : Address and size of L2 cache controller registers
> +- cache-size : Size of the entire L2 cache
> +- interrupts : Error interrupt of L2 controller
> +
> +Optional Properties:
> +
> +- cache-line-size : Size of L2 cache lines
cache-line-size is required as per ePAPR.
> diff --git a/Documentation/devicetree/bindings/powerpc/fsl/mem-ctrlr.txt b/Documentation/devicetree/bindings/powerpc/fsl/mem-ctrlr.txt
> new file mode 100644
> index 0000000..70b42bb
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/powerpc/fsl/mem-ctrlr.txt
> @@ -0,0 +1,16 @@
> +Freescale DDR memory controller
> +
> +Properties:
> +
> +- compatible : Should include "fsl,chip-memory-controller" where
> + chip is the processor (bsc9132, mpc8572 etc.)
Please also cover newer device trees that use
"fsl,qoriq-memory-controller" and don't have
"fsl,CHIP-memory-controller" (they also use a variant with the block
version, but since the block version is readable in a register I don't
think it's necessary to specify that here).
> diff --git a/Documentation/devicetree/bindings/usb/fsl-usb.txt b/Documentation/devicetree/bindings/usb/fsl-usb.txt
> index bd5723f..afa5809 100644
> --- a/Documentation/devicetree/bindings/usb/fsl-usb.txt
> +++ b/Documentation/devicetree/bindings/usb/fsl-usb.txt
> @@ -9,6 +9,8 @@ Required properties :
> - compatible : Should be "fsl-usb2-mph" for multi port host USB
> controllers, or "fsl-usb2-dr" for dual role USB controllers
> or "fsl,mpc5121-usb2-dr" for dual role USB controllers of MPC5121
> + Wherever applicable, the IP version of the USB controller should
> + also be mentioned (for eg. fsl-usb2-dr-v2.2 for bsc9132).
Please terminate the previous sentence with a period.
-Scott
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: Nishanth Aravamudan @ 2014-02-18 23:58 UTC (permalink / raw)
To: Michal Hocko; +Cc: linux-mm, linuxppc-dev, Anton Blanchard, LKML
In-Reply-To: <20140218233404.GB10844@linux.vnet.ibm.com>
On 18.02.2014 [15:34:05 -0800], Nishanth Aravamudan wrote:
> Hi Michal,
>
> On 18.02.2014 [10:06:58 +0100], Michal Hocko wrote:
> > Hi,
> > I have just noticed that ppc has RECLAIM_DISTANCE reduced to 10 set by
> > 56608209d34b (powerpc/numa: Set a smaller value for RECLAIM_DISTANCE to
> > enable zone reclaim). The commit message suggests that the zone reclaim
> > is desirable for all NUMA configurations.
> >
> > History has shown that the zone reclaim is more often harmful than
> > helpful and leads to performance problems. The default RECLAIM_DISTANCE
> > for generic case has been increased from 20 to 30 around 3.0
> > (32e45ff43eaf mm: increase RECLAIM_DISTANCE to 30).
>
> Interesting.
>
> > I strongly suspect that the patch is incorrect and it should be
> > reverted. Before I will send a revert I would like to understand what
> > led to the patch in the first place. I do not see why would PPC use only
> > LOCAL_DISTANCE and REMOTE_DISTANCE distances and in fact machines I have
> > seen use different values.
> >
> > Anton, could you comment please?
>
> I'll let Anton comment here, but in looking into this issue in working
> on CONFIG_HAVE_MEMORYLESS_NODE support, I realized that any LPAR with
> memoryless nodes will set zone_reclaim_mode to 1. I think we want to
> ignore memoryless nodes when we set up the reclaim mode like the
> following? I'll send it as a proper patch if you agree?
>
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 5de4337..4f6ff6f 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -1853,8 +1853,9 @@ static void __paginginit init_zone_allows_reclaim(int nid)
> {
> int i;
>
> - for_each_online_node(i)
> - if (node_distance(nid, i) <= RECLAIM_DISTANCE)
> + for_each_online_node(i) {
> + if (node_distance(nid, i) <= RECLAIM_DISTANCE ||
> + local_memory_node(nid) != nid)
> node_set(i, NODE_DATA(nid)->reclaim_nodes);
> else
> zone_reclaim_mode = 1;
>
> Note, this won't actually do anything if CONFIG_HAVE_MEMORYLESS_NODES is
> not set, but if it is, I think semantically it will indicate that
> memoryless nodes *have* to reclaim remotely.
>
> And actually the above won't work, because the callpath is
>
> start_kernel -> setup_arch -> paging_init [-> free_area_init_nodes ->
> free_area_init_node -> init_zone_allows_reclaim] which is called before
> build_all_zonelists. This is a similar ordering problem as I'm having
> with the MEMORYLESS_NODE support, will work on it.
How about the following?
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5de4337..1a0eced 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1854,7 +1854,8 @@ static void __paginginit init_zone_allows_reclaim(int nid)
int i;
for_each_online_node(i)
- if (node_distance(nid, i) <= RECLAIM_DISTANCE)
+ if (node_distance(nid, i) <= RECLAIM_DISTANCE ||
+ !NODE_DATA(nid)->node_present_pages)
node_set(i, NODE_DATA(nid)->reclaim_nodes);
else
zone_reclaim_mode = 1;
@@ -4901,13 +4902,13 @@ void __paginginit free_area_init_node(int nid, unsigned long *zones_size,
pgdat->node_id = nid;
pgdat->node_start_pfn = node_start_pfn;
- init_zone_allows_reclaim(nid);
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP
get_pfn_range_for_nid(nid, &start_pfn, &end_pfn);
#endif
calculate_node_totalpages(pgdat, start_pfn, end_pfn,
zones_size, zholes_size);
+ init_zone_allows_reclaim(nid);
alloc_node_mem_map(pgdat);
#ifdef CONFIG_FLAT_NODE_MEM_MAP
printk(KERN_DEBUG "free_area_init_node: node %d, pgdat %08lx, node_mem_map %08lx\n",
I think it's safe to move init_zone_allows_reclaim, because I don't
think any allocates are occurring here that could cause us to reclaim
anyways, right? Moving it allows us to safely reference
node_present_pages.
Thanks,
Nish
^ permalink raw reply related
* Re: [PATCH v3 3/3] powerpc/pseries: Report in kernel device tree update to drmgr
From: Benjamin Herrenschmidt @ 2014-02-18 23:47 UTC (permalink / raw)
To: Tyrel Datwyler; +Cc: nfont, linuxppc-dev, Tyrel Datwyler
In-Reply-To: <5303EC9E.7070507@gmail.com>
On Tue, 2014-02-18 at 15:28 -0800, Tyrel Datwyler wrote:
> In this case we get a double update which I clearly neglected to
> mention
> in the patch. The first patch in this series actually removes an
> unnecessary double update from the existing kernel implementation. The
> same information is returned by the update-nodes/properties call the
> second time around and aside from being a waste of a few cycles is
> harmless.
Thanks. Can you resend the patches with updated descriptions ? I will
try to send them to Linus by end of week.
Cheers,
Ben.
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: Nishanth Aravamudan @ 2014-02-18 23:34 UTC (permalink / raw)
To: Michal Hocko; +Cc: linux-mm, linuxppc-dev, Anton Blanchard, LKML
In-Reply-To: <20140218090658.GA28130@dhcp22.suse.cz>
Hi Michal,
On 18.02.2014 [10:06:58 +0100], Michal Hocko wrote:
> Hi,
> I have just noticed that ppc has RECLAIM_DISTANCE reduced to 10 set by
> 56608209d34b (powerpc/numa: Set a smaller value for RECLAIM_DISTANCE to
> enable zone reclaim). The commit message suggests that the zone reclaim
> is desirable for all NUMA configurations.
>
> History has shown that the zone reclaim is more often harmful than
> helpful and leads to performance problems. The default RECLAIM_DISTANCE
> for generic case has been increased from 20 to 30 around 3.0
> (32e45ff43eaf mm: increase RECLAIM_DISTANCE to 30).
Interesting.
> I strongly suspect that the patch is incorrect and it should be
> reverted. Before I will send a revert I would like to understand what
> led to the patch in the first place. I do not see why would PPC use only
> LOCAL_DISTANCE and REMOTE_DISTANCE distances and in fact machines I have
> seen use different values.
>
> Anton, could you comment please?
I'll let Anton comment here, but in looking into this issue in working
on CONFIG_HAVE_MEMORYLESS_NODE support, I realized that any LPAR with
memoryless nodes will set zone_reclaim_mode to 1. I think we want to
ignore memoryless nodes when we set up the reclaim mode like the
following? I'll send it as a proper patch if you agree?
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 5de4337..4f6ff6f 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1853,8 +1853,9 @@ static void __paginginit init_zone_allows_reclaim(int nid)
{
int i;
- for_each_online_node(i)
- if (node_distance(nid, i) <= RECLAIM_DISTANCE)
+ for_each_online_node(i) {
+ if (node_distance(nid, i) <= RECLAIM_DISTANCE ||
+ local_memory_node(nid) != nid)
node_set(i, NODE_DATA(nid)->reclaim_nodes);
else
zone_reclaim_mode = 1;
Note, this won't actually do anything if CONFIG_HAVE_MEMORYLESS_NODES is
not set, but if it is, I think semantically it will indicate that
memoryless nodes *have* to reclaim remotely.
And actually the above won't work, because the callpath is
start_kernel -> setup_arch -> paging_init [-> free_area_init_nodes ->
free_area_init_node -> init_zone_allows_reclaim] which is called before
build_all_zonelists. This is a similar ordering problem as I'm having
with the MEMORYLESS_NODE support, will work on it.
Thanks,
Nish
^ permalink raw reply related
* Re: [PATCH v3 3/3] powerpc/pseries: Report in kernel device tree update to drmgr
From: Tyrel Datwyler @ 2014-02-18 23:28 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Tyrel Datwyler; +Cc: nfont, linuxppc-dev
In-Reply-To: <1392596560.24061.12.camel@pasglop>
On 02/16/2014 04:22 PM, Benjamin Herrenschmidt wrote:
> On Fri, 2014-01-31 at 15:58 -0800, Tyrel Datwyler wrote:
>> Traditionally it has been drmgr's responsibilty to update the device tree
>> through the /proc/ppc64/ofdt interface after a suspend/resume operation.
>> This patchset however has modified suspend/resume ops to preform that update
>> entirely in the kernel during the resume. Therefore, a mechanism is required
>> for drmgr to determine who is responsible for the update. This patch adds a
>> show function to the "hibernate" attribute that returns 1 if the kernel
>> updates the device tree after the resume and 0 if drmgr is responsible.
>
> This is not right.
>
> We cannot change the kernel behaviour in a way that is incompatible with
> existing userspace, and unless you can explain me why that is not the
> case, it feels like this patch set does just that ....
>
> What happens if you have an older drmgr which still does the update
> itself ?
>
In this case we get a double update which I clearly neglected to mention
in the patch. The first patch in this series actually removes an
unnecessary double update from the existing kernel implementation. The
same information is returned by the update-nodes/properties call the
second time around and aside from being a waste of a few cycles is harmless.
Tyrel
> You can't just change user visible behaviours and interface without at
> least explaining why it's ok to do so at the very least (and ensuring
> that of course) without breaking existing userspace.
>
> Now it's possible that the double update caused by this series is
> harmless, in which case please explain it in the changeset, I certainly
> can't assume it and if it's not, you'll need some other way for drmgr to
> signal the kernel to indicate it supports the new behaviour before you
> enable it.
>
> Ben.
>
>> Signed-off-by: Tyrel Datwyler <tyreld@linux.vnet.ibm.com>
>> ---
>> arch/powerpc/platforms/pseries/suspend.c | 25 ++++++++++++++++++++++++-
>> 1 file changed, 24 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/powerpc/platforms/pseries/suspend.c b/arch/powerpc/platforms/pseries/suspend.c
>> index 1d9c580..b87b978 100644
>> --- a/arch/powerpc/platforms/pseries/suspend.c
>> +++ b/arch/powerpc/platforms/pseries/suspend.c
>> @@ -192,7 +192,30 @@ out:
>> return rc;
>> }
>>
>> -static DEVICE_ATTR(hibernate, S_IWUSR, NULL, store_hibernate);
>> +#define USER_DT_UPDATE 0
>> +#define KERN_DT_UPDATE 1
>> +
>> +/**
>> + * show_hibernate - Report device tree update responsibilty
>> + * @dev: subsys root device
>> + * @attr: device attribute struct
>> + * @buf: buffer
>> + *
>> + * Report whether a device tree update is performed by the kernel after a
>> + * resume, or if drmgr must coordinate the update from user space.
>> + *
>> + * Return value:
>> + * 0 if drmgr is to initiate update, and 1 otherwise
>> + **/
>> +static ssize_t show_hibernate(struct device *dev,
>> + struct device_attribute *attr,
>> + char *buf)
>> +{
>> + return sprintf(buf, "%d\n", KERN_DT_UPDATE);
>> +}
>> +
>> +static DEVICE_ATTR(hibernate, S_IWUSR | S_IRUGO,
>> + show_hibernate, store_hibernate);
>>
>> static struct bus_type suspend_subsys = {
>> .name = "power",
>
>
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev@lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/linuxppc-dev
>
^ permalink raw reply
* Re: ppc: RECLAIM_DISTANCE 10?
From: David Rientjes @ 2014-02-18 22:27 UTC (permalink / raw)
To: Michal Hocko; +Cc: linuxppc-dev, Anton Blanchard, LKML, linux-mm
In-Reply-To: <20140218090658.GA28130@dhcp22.suse.cz>
On Tue, 18 Feb 2014, Michal Hocko wrote:
> Hi,
> I have just noticed that ppc has RECLAIM_DISTANCE reduced to 10 set by
> 56608209d34b (powerpc/numa: Set a smaller value for RECLAIM_DISTANCE to
> enable zone reclaim). The commit message suggests that the zone reclaim
> is desirable for all NUMA configurations.
>
> History has shown that the zone reclaim is more often harmful than
> helpful and leads to performance problems. The default RECLAIM_DISTANCE
> for generic case has been increased from 20 to 30 around 3.0
> (32e45ff43eaf mm: increase RECLAIM_DISTANCE to 30).
>
> I strongly suspect that the patch is incorrect and it should be
> reverted. Before I will send a revert I would like to understand what
> led to the patch in the first place. I do not see why would PPC use only
> LOCAL_DISTANCE and REMOTE_DISTANCE distances and in fact machines I have
> seen use different values.
>
I strongly suspect that the patch is correct since powerpc node distances
are different than the architectures you're talking about and get doubled
for every NUMA domain that the hardware supports.
^ permalink raw reply
* Re: [RFC PATCH 2/3] topology: support node_numa_mem() for determining the fallback node
From: Nishanth Aravamudan @ 2014-02-18 22:22 UTC (permalink / raw)
To: Christoph Lameter
Cc: Han Pingtian, Matt Mackall, Pekka Enberg,
Linux Memory Management List, Paul Mackerras, Anton Blanchard,
David Rientjes, Joonsoo Kim, linuxppc-dev, Wanpeng Li
In-Reply-To: <alpine.DEB.2.10.1402181547210.3973@nuc>
On 18.02.2014 [15:49:22 -0600], Christoph Lameter wrote:
> On Tue, 18 Feb 2014, Nishanth Aravamudan wrote:
>
> > We use the topology provided by the hypervisor, it does actually reflect
> > where CPUs and memory are, and their corresponding performance/NUMA
> > characteristics.
>
> And so there are actually nodes without memory that have processors?
Virtually (topologically as indicated to Linux), yes. Physically, I
don't think they are, but they might be exhausted, which is we get sort
of odd-appearing NUMA configurations.
> Can the hypervisor or the linux arch code be convinced to ignore nodes
> without memory or assign a sane default node to processors?
I think this happens quite often, so I don't know that we want to ignore
the performance impact of the underlying NUMA configuration. I guess we
could special-case memoryless/cpuless configurations somewhat, but I
don't think there's any reason to do that if we can make memoryless-node
support work in-kernel?
> > > Ok then also move the memory of the local node somewhere?
> >
> > This happens below the OS, we don't control the hypervisor's decisions.
> > I'm not sure if that's what you are suggesting.
>
> You could also do this from the powerpc arch code by sanitizing the
> processor / node information that is then used by Linux.
I see what you're saying now, thanks!
-Nish
^ permalink raw reply
* Re: [RFC PATCH 2/3] topology: support node_numa_mem() for determining the fallback node
From: Christoph Lameter @ 2014-02-18 21:49 UTC (permalink / raw)
To: Nishanth Aravamudan
Cc: Han Pingtian, Matt Mackall, Pekka Enberg,
Linux Memory Management List, Paul Mackerras, Anton Blanchard,
David Rientjes, Joonsoo Kim, linuxppc-dev, Wanpeng Li
In-Reply-To: <20140218210923.GA28170@linux.vnet.ibm.com>
On Tue, 18 Feb 2014, Nishanth Aravamudan wrote:
> We use the topology provided by the hypervisor, it does actually reflect
> where CPUs and memory are, and their corresponding performance/NUMA
> characteristics.
And so there are actually nodes without memory that have processors?
Can the hypervisor or the linux arch code be convinced to ignore nodes
without memory or assign a sane default node to processors?
> > Ok then also move the memory of the local node somewhere?
>
> This happens below the OS, we don't control the hypervisor's decisions.
> I'm not sure if that's what you are suggesting.
You could also do this from the powerpc arch code by sanitizing the
processor / node information that is then used by Linux.
^ permalink raw reply
* Re: [RFC PATCH 2/3] topology: support node_numa_mem() for determining the fallback node
From: Nishanth Aravamudan @ 2014-02-18 21:09 UTC (permalink / raw)
To: Christoph Lameter
Cc: Han Pingtian, Matt Mackall, Pekka Enberg,
Linux Memory Management List, Paul Mackerras, Anton Blanchard,
David Rientjes, Joonsoo Kim, linuxppc-dev, Wanpeng Li
In-Reply-To: <alpine.DEB.2.10.1402181356120.2910@nuc>
On 18.02.2014 [13:58:20 -0600], Christoph Lameter wrote:
> On Tue, 18 Feb 2014, Nishanth Aravamudan wrote:
>
> >
> > Well, on powerpc, with the hypervisor providing the resources and the
> > topology, you can have cpuless and memoryless nodes. I'm not sure how
> > "fake" the NUMA is -- as I think since the resources are virtualized to
> > be one system, it's logically possible that the actual topology of the
> > resources can be CPUs from physical node 0 and memory from physical node
> > 2. I would think with KVM on a sufficiently large (physically NUMA
> > x86_64) and loaded system, one could cause the same sort of
> > configuration to occur for a guest?
>
> Ok but since you have a virtualized environment: Why not provide a fake
> home node with fake memory that could be anywhere? This would avoid the
> whole problem of supporting such a config at the kernel level.
We use the topology provided by the hypervisor, it does actually reflect
where CPUs and memory are, and their corresponding performance/NUMA
characteristics.
> Do not have a fake node that has no memory.
>
> > In any case, these configurations happen fairly often on long-running
> > (not rebooted) systems as LPARs are created/destroyed, resources are
> > DLPAR'd in and out of LPARs, etc.
>
> Ok then also move the memory of the local node somewhere?
This happens below the OS, we don't control the hypervisor's decisions.
I'm not sure if that's what you are suggesting.
Thanks,
Nish
^ permalink raw reply
* Re: [RFC PATCH 2/3] topology: support node_numa_mem() for determining the fallback node
From: Christoph Lameter @ 2014-02-18 19:58 UTC (permalink / raw)
To: Nishanth Aravamudan
Cc: Han Pingtian, Matt Mackall, Pekka Enberg,
Linux Memory Management List, Paul Mackerras, Anton Blanchard,
David Rientjes, Joonsoo Kim, linuxppc-dev, Wanpeng Li
In-Reply-To: <20140218172832.GD31998@linux.vnet.ibm.com>
On Tue, 18 Feb 2014, Nishanth Aravamudan wrote:
>
> Well, on powerpc, with the hypervisor providing the resources and the
> topology, you can have cpuless and memoryless nodes. I'm not sure how
> "fake" the NUMA is -- as I think since the resources are virtualized to
> be one system, it's logically possible that the actual topology of the
> resources can be CPUs from physical node 0 and memory from physical node
> 2. I would think with KVM on a sufficiently large (physically NUMA
> x86_64) and loaded system, one could cause the same sort of
> configuration to occur for a guest?
Ok but since you have a virtualized environment: Why not provide a fake
home node with fake memory that could be anywhere? This would avoid the
whole problem of supporting such a config at the kernel level.
Do not have a fake node that has no memory.
> In any case, these configurations happen fairly often on long-running
> (not rebooted) systems as LPARs are created/destroyed, resources are
> DLPAR'd in and out of LPARs, etc.
Ok then also move the memory of the local node somewhere?
> I might look into it, as it might have sped up testing these changes.
I guess that will be necessary in order to support the memoryless nodes
long term.
^ permalink raw reply
* Re: Anyone using SysRQ key sequences on console serial port ?
From: John Donnelly @ 2014-02-18 19:47 UTC (permalink / raw)
To: Scott Wood; +Cc: Paul Gortmaker, linuxppc-dev
In-Reply-To: <1392665821.6733.630.camel@snotra.buserror.net>
[-- Attachment #1: Type: text/plain, Size: 1900 bytes --]
I am enable to get one keyboard sequence responded to with the noted change
in the dts .
for instance:
SysRQ ( Break) c
Panics .. Which is a good response, and since it doesn't require a return
to user mode ( I suspect ) it appears to work.
Any other requests fail to report any information :
SysRQ (break ) l - list active processes
m - list memory
Any additional SysRQ are ignored., and the system appears hung.
On an reference Intel platform, multiple SyqRQ can be issued and the
system remains healthy .
On Mon, Feb 17, 2014 at 1:37 PM, Scott Wood <scottwood@freescale.com> wrote:
> On Sun, 2014-02-16 at 10:56 -0500, Paul Gortmaker wrote:
> > On Fri, Feb 14, 2014 at 3:42 PM, John Donnelly <john.d@servergy.com>
> wrote:
> > > Hi,
> > >
> > > I tried using the SysRq hotkey sequence on a serial console -
> > > 3.11.0-5-powerpc-e500mc system, by issuing a " break " and the system
> > > immediately wedges after displaying "SysRQ : HELP : " using both
> "Putty" and
> > > "Teraterm" terminal emulators. I know the system is dead because my
> ssh
> > > sessions stopped too.
> >
> > Yes it does work -- or at least it _did_ work. Make sure your dts has
> an entry
> >
> > compatible = "fsl,ns16550", "ns16550";
> >
> > since that enables a workaround I'd added for a hardware errata relating
> > to sending breaks over the serial console. What you describe above
> > makes me think you aren't getting the workaround enabled.
>
> Also make sure CONFIG_SERIAL_8250_FSL is enabled.
>
> -Scott
>
>
>
--
*Regards,*
* John.*
*--*
*o* Energy-efficiency is #1 reason data centers look to expand. -- Digital
Realty Trust
*o* Green Data Centers spending to increase 300% worldwide by 2016. --
Pike Research
*o *Data Centers have become as vital to the functioni
ng of society as power stations. -- The Economist
[-- Attachment #2: Type: text/html, Size: 4606 bytes --]
^ permalink raw reply
* Re: [RFC PATCH 2/3] topology: support node_numa_mem() for determining the fallback node
From: Nishanth Aravamudan @ 2014-02-18 17:28 UTC (permalink / raw)
To: Christoph Lameter
Cc: Han Pingtian, Matt Mackall, Pekka Enberg,
Linux Memory Management List, Paul Mackerras, Anton Blanchard,
David Rientjes, Joonsoo Kim, linuxppc-dev, Wanpeng Li
In-Reply-To: <alpine.DEB.2.10.1402181051560.1291@nuc>
On 18.02.2014 [10:57:09 -0600], Christoph Lameter wrote:
> On Mon, 17 Feb 2014, Joonsoo Kim wrote:
>
> > On Wed, Feb 12, 2014 at 10:51:37PM -0800, Nishanth Aravamudan wrote:
> > > Hi Joonsoo,
> > > Also, given that only ia64 and (hopefuly soon) ppc64 can set
> > > CONFIG_HAVE_MEMORYLESS_NODES, does that mean x86_64 can't have
> > > memoryless nodes present? Even with fakenuma? Just curious.
>
> x86_64 currently does not support memoryless nodes otherwise it would
> have set CONFIG_HAVE_MEMORYLESS_NODES in the kconfig. Memoryless nodes are
> a bit strange given that the NUMA paradigm is to have NUMA nodes (meaning
> memory) with processors. MEMORYLESS nodes means that we have a fake NUMA
> node without memory but just processors. Not very efficient. Not sure why
> people use these configurations.
Well, on powerpc, with the hypervisor providing the resources and the
topology, you can have cpuless and memoryless nodes. I'm not sure how
"fake" the NUMA is -- as I think since the resources are virtualized to
be one system, it's logically possible that the actual topology of the
resources can be CPUs from physical node 0 and memory from physical node
2. I would think with KVM on a sufficiently large (physically NUMA
x86_64) and loaded system, one could cause the same sort of
configuration to occur for a guest?
In any case, these configurations happen fairly often on long-running
(not rebooted) systems as LPARs are created/destroyed, resources are
DLPAR'd in and out of LPARs, etc.
> > I don't know, because I'm not expert on NUMA system :)
> > At first glance, fakenuma can't be used for testing
> > CONFIG_HAVE_MEMORYLESS_NODES. Maybe some modification is needed.
>
> Well yeah. You'd have to do some mods to enable that testing.
I might look into it, as it might have sped up testing these changes.
Thanks,
Nish
^ permalink raw reply
* Re: [RFC PATCH 2/3] topology: support node_numa_mem() for determining the fallback node
From: Nishanth Aravamudan @ 2014-02-18 17:22 UTC (permalink / raw)
To: Christoph Lameter
Cc: Han Pingtian, Matt Mackall, Pekka Enberg,
Linux Memory Management List, Paul Mackerras, Anton Blanchard,
David Rientjes, Joonsoo Kim, linuxppc-dev, Wanpeng Li
In-Reply-To: <alpine.DEB.2.10.1402121612270.8183@nuc>
On 12.02.2014 [16:16:11 -0600], Christoph Lameter wrote:
> Here is another patch with some fixes. The additional logic is only
> compiled in if CONFIG_HAVE_MEMORYLESS_NODES is set.
>
> Subject: slub: Memoryless node support
>
> Support memoryless nodes by tracking which allocations are failing.
> Allocations targeted to the nodes without memory fall back to the
> current available per cpu objects and if that is not available will
> create a new slab using the page allocator to fallback from the
> memoryless node to some other node.
>
> Signed-off-by: Christoph Lameter <cl@linux.com>
Tested-by: Nishanth Aravamudan <nacc@linux.vnet.ibm.com>
Acked-by: Nishanth Aravamudan <nacc@linux.vnet.ibm.com>
> Index: linux/mm/slub.c
> ===================================================================
> --- linux.orig/mm/slub.c 2014-02-12 16:07:48.957869570 -0600
> +++ linux/mm/slub.c 2014-02-12 16:09:22.198928260 -0600
> @@ -134,6 +134,10 @@ static inline bool kmem_cache_has_cpu_pa
> #endif
> }
>
> +#ifdef CONFIG_HAVE_MEMORYLESS_NODES
> +static nodemask_t empty_nodes;
> +#endif
> +
> /*
> * Issues still to be resolved:
> *
> @@ -1405,16 +1409,28 @@ static struct page *new_slab(struct kmem
> void *last;
> void *p;
> int order;
> + int alloc_node;
>
> BUG_ON(flags & GFP_SLAB_BUG_MASK);
>
> page = allocate_slab(s,
> flags & (GFP_RECLAIM_MASK | GFP_CONSTRAINT_MASK), node);
> - if (!page)
> + if (!page) {
> +#ifdef CONFIG_HAVE_MEMORYLESS_NODES
> + if (node != NUMA_NO_NODE)
> + node_set(node, empty_nodes);
> +#endif
> goto out;
> + }
>
> order = compound_order(page);
> - inc_slabs_node(s, page_to_nid(page), page->objects);
> + alloc_node = page_to_nid(page);
> +#ifdef CONFIG_HAVE_MEMORYLESS_NODES
> + node_clear(alloc_node, empty_nodes);
> + if (node != NUMA_NO_NODE && alloc_node != node)
> + node_set(node, empty_nodes);
> +#endif
> + inc_slabs_node(s, alloc_node, page->objects);
> memcg_bind_pages(s, order);
> page->slab_cache = s;
> __SetPageSlab(page);
> @@ -1722,7 +1738,7 @@ static void *get_partial(struct kmem_cac
> struct kmem_cache_cpu *c)
> {
> void *object;
> - int searchnode = (node == NUMA_NO_NODE) ? numa_node_id() : node;
> + int searchnode = (node == NUMA_NO_NODE) ? numa_mem_id() : node;
>
> object = get_partial_node(s, get_node(s, searchnode), c, flags);
> if (object || node != NUMA_NO_NODE)
> @@ -2117,8 +2133,19 @@ static void flush_all(struct kmem_cache
> static inline int node_match(struct page *page, int node)
> {
> #ifdef CONFIG_NUMA
> - if (!page || (node != NUMA_NO_NODE && page_to_nid(page) != node))
> + int page_node = page_to_nid(page);
> +
> + if (!page)
> return 0;
> +
> + if (node != NUMA_NO_NODE) {
> +#ifdef CONFIG_HAVE_MEMORYLESS_NODES
> + if (node_isset(node, empty_nodes))
> + return 1;
> +#endif
> + if (page_node != node)
> + return 0;
> + }
> #endif
> return 1;
> }
>
^ permalink raw reply
* Re: [RFC PATCH 2/3] topology: support node_numa_mem() for determining the fallback node
From: Christoph Lameter @ 2014-02-18 16:57 UTC (permalink / raw)
To: Joonsoo Kim
Cc: Han Pingtian, Nishanth Aravamudan, Matt Mackall, Pekka Enberg,
Linux Memory Management List, Paul Mackerras, Anton Blanchard,
David Rientjes, linuxppc-dev, Wanpeng Li
In-Reply-To: <20140217070051.GE3468@lge.com>
On Mon, 17 Feb 2014, Joonsoo Kim wrote:
> On Wed, Feb 12, 2014 at 10:51:37PM -0800, Nishanth Aravamudan wrote:
> > Hi Joonsoo,
> > Also, given that only ia64 and (hopefuly soon) ppc64 can set
> > CONFIG_HAVE_MEMORYLESS_NODES, does that mean x86_64 can't have
> > memoryless nodes present? Even with fakenuma? Just curious.
x86_64 currently does not support memoryless nodes otherwise it would
have set CONFIG_HAVE_MEMORYLESS_NODES in the kconfig. Memoryless nodes are
a bit strange given that the NUMA paradigm is to have NUMA nodes (meaning
memory) with processors. MEMORYLESS nodes means that we have a fake NUMA
node without memory but just processors. Not very efficient. Not sure why
people use these configurations.
> I don't know, because I'm not expert on NUMA system :)
> At first glance, fakenuma can't be used for testing
> CONFIG_HAVE_MEMORYLESS_NODES. Maybe some modification is needed.
Well yeah. You'd have to do some mods to enable that testing.
^ permalink raw reply
* Re: [RFC PATCH 2/3] topology: support node_numa_mem() for determining the fallback node
From: Christoph Lameter @ 2014-02-18 16:38 UTC (permalink / raw)
To: Joonsoo Kim
Cc: Han Pingtian, Nishanth Aravamudan, Matt Mackall, Pekka Enberg,
Linux Memory Management List, Paul Mackerras, Anton Blanchard,
David Rientjes, linuxppc-dev, Wanpeng Li
In-Reply-To: <20140217065257.GD3468@lge.com>
On Mon, 17 Feb 2014, Joonsoo Kim wrote:
> On Wed, Feb 12, 2014 at 04:16:11PM -0600, Christoph Lameter wrote:
> > Here is another patch with some fixes. The additional logic is only
> > compiled in if CONFIG_HAVE_MEMORYLESS_NODES is set.
> >
> > Subject: slub: Memoryless node support
> >
> > Support memoryless nodes by tracking which allocations are failing.
>
> I still don't understand why this tracking is needed.
Its an optimization to avoid calling the page allocator to figure out if
there is memory available on a particular node.
> All we need for allcation targeted to memoryless node is to fallback proper
> node, that it, numa_mem_id() node of targeted node. My previous patch
> implements it and use proper fallback node on every allocation code path.
> Why this tracking is needed? Please elaborate more on this.
Its too slow to do that on every alloc. One needs to be able to satisfy
most allocations without switching percpu slabs for optimal performance.
> > Allocations targeted to the nodes without memory fall back to the
> > current available per cpu objects and if that is not available will
> > create a new slab using the page allocator to fallback from the
> > memoryless node to some other node.
And what about the next alloc? Assuem there are N allocs from a memoryless
node this means we push back the partial slab on each alloc and then fall
back?
> > {
> > void *object;
> > - int searchnode = (node == NUMA_NO_NODE) ? numa_node_id() : node;
> > + int searchnode = (node == NUMA_NO_NODE) ? numa_mem_id() : node;
> >
> > object = get_partial_node(s, get_node(s, searchnode), c, flags);
> > if (object || node != NUMA_NO_NODE)
>
> This isn't enough.
> Consider that allcation targeted to memoryless node.
It will not common get there because of the tracking. Instead a per cpu
object will be used.
> get_partial_node() always fails even if there are some partial slab on
> memoryless node's neareast node.
Correct and that leads to a page allocator action whereupon the node will
be marked as empty.
> We should fallback to some proper node in this case, since there is no slab
> on memoryless node.
NUMA is about optimization of memory allocations. It is often *not* about
correctness but heuristics are used in many cases. F.e. see the zone
reclaim logic, zone reclaim mode, fallback scenarios in the page allocator
etc etc.
^ permalink raw reply
* [PATCH RFC 25/31] mmc: sdhci: convert sdhci_set_uhs_signaling() into a library function
From: Russell King @ 2014-02-18 15:11 UTC (permalink / raw)
To: linux-arm-kernel, linux-mmc
Cc: Barry Song, spear-devel, Stephen Warren, Anton Vorontsov,
Chris Ball, Michal Simek, Thierry Reding, Viresh Kumar, Ben Dooks,
linux-tegra, linuxppc-dev
In-Reply-To: <20140218150801.GD21483@n2100.arm.linux.org.uk>
Add sdhci_set_uhs_signaling() and always call the set_uhs_signaling
method. This avoids quirks being added into sdhci_set_uhs_signaling().
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
=-DO NOT APPLY-=
drivers/mmc/host/sdhci-acpi.c | 2 ++
drivers/mmc/host/sdhci-bcm-kona.c | 1 +
drivers/mmc/host/sdhci-bcm2835.c | 1 +
drivers/mmc/host/sdhci-cns3xxx.c | 1 +
drivers/mmc/host/sdhci-dove.c | 1 +
drivers/mmc/host/sdhci-of-arasan.c | 1 +
drivers/mmc/host/sdhci-of-esdhc.c | 1 +
drivers/mmc/host/sdhci-of-hlwd.c | 1 +
drivers/mmc/host/sdhci-pci.c | 1 +
drivers/mmc/host/sdhci-pltfm.c | 1 +
drivers/mmc/host/sdhci-pxav2.c | 1 +
drivers/mmc/host/sdhci-pxav3.c | 1 +
drivers/mmc/host/sdhci-s3c.c | 1 +
drivers/mmc/host/sdhci-sirf.c | 1 +
drivers/mmc/host/sdhci-spear.c | 1 +
drivers/mmc/host/sdhci-tegra.c | 1 +
drivers/mmc/host/sdhci.c | 42 +++++++++++++++++++++-----------------
drivers/mmc/host/sdhci.h | 1 +
18 files changed, 41 insertions(+), 19 deletions(-)
diff --git a/drivers/mmc/host/sdhci-acpi.c b/drivers/mmc/host/sdhci-acpi.c
index c20964e577e9..2ece31d88196 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-acpi.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-acpi.c
@@ -105,6 +105,7 @@ static const struct sdhci_ops sdhci_acpi_ops_dflt = {
.enable_dma = sdhci_acpi_enable_dma,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static const struct sdhci_ops sdhci_acpi_ops_int = {
@@ -112,6 +113,7 @@ static const struct sdhci_ops sdhci_acpi_ops_int = {
.enable_dma = sdhci_acpi_enable_dma,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
.hw_reset = sdhci_acpi_int_hw_reset,
};
diff --git a/drivers/mmc/host/sdhci-bcm-kona.c b/drivers/mmc/host/sdhci-bcm-kona.c
index 77b14b1a24bc..20096aaf14b4 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-bcm-kona.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-bcm-kona.c
@@ -211,6 +211,7 @@ static struct sdhci_ops sdhci_bcm_kona_ops = {
.platform_send_init_74_clocks = sdhci_bcm_kona_init_74_clocks,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
.card_event = sdhci_bcm_kona_card_event,
};
diff --git a/drivers/mmc/host/sdhci-bcm2835.c b/drivers/mmc/host/sdhci-bcm2835.c
index 74906d6008e1..46af9a439d7b 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-bcm2835.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-bcm2835.c
@@ -136,6 +136,7 @@ static const struct sdhci_ops bcm2835_sdhci_ops = {
.get_min_clock = bcm2835_sdhci_get_min_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static const struct sdhci_pltfm_data bcm2835_sdhci_pdata = {
diff --git a/drivers/mmc/host/sdhci-cns3xxx.c b/drivers/mmc/host/sdhci-cns3xxx.c
index 587d73ef33ff..14b74075589a 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-cns3xxx.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-cns3xxx.c
@@ -81,6 +81,7 @@ static const struct sdhci_ops sdhci_cns3xxx_ops = {
.set_clock = sdhci_cns3xxx_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static const struct sdhci_pltfm_data sdhci_cns3xxx_pdata = {
diff --git a/drivers/mmc/host/sdhci-dove.c b/drivers/mmc/host/sdhci-dove.c
index 4f4216115aed..8d6d0a293f17 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-dove.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-dove.c
@@ -89,6 +89,7 @@ static const struct sdhci_ops sdhci_dove_ops = {
.set_clock = sdhci_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static const struct sdhci_pltfm_data sdhci_dove_pdata = {
diff --git a/drivers/mmc/host/sdhci-of-arasan.c b/drivers/mmc/host/sdhci-of-arasan.c
index f0ee594f25d1..5bd1092310f2 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-of-arasan.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-of-arasan.c
@@ -57,6 +57,7 @@ static struct sdhci_ops sdhci_arasan_ops = {
.get_timeout_clock = sdhci_arasan_get_timeout_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static struct sdhci_pltfm_data sdhci_arasan_pdata = {
diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c
index c4f8cd3f83c8..fcaeae5f55b8 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-of-esdhc.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-of-esdhc.c
@@ -309,6 +309,7 @@ static const struct sdhci_ops sdhci_esdhc_ops = {
.adma_workaround = esdhci_of_adma_workaround,
.set_bus_width = esdhc_pltfm_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static const struct sdhci_pltfm_data sdhci_esdhc_pdata = {
diff --git a/drivers/mmc/host/sdhci-of-hlwd.c b/drivers/mmc/host/sdhci-of-hlwd.c
index a4a1f0f2c0a0..b341661369a2 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-of-hlwd.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-of-hlwd.c
@@ -61,6 +61,7 @@ static const struct sdhci_ops sdhci_hlwd_ops = {
.set_clock = sdhci_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static const struct sdhci_pltfm_data sdhci_hlwd_pdata = {
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index 01ee60d17b04..ba4eaf03200d 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pci.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pci.c
@@ -1062,6 +1062,7 @@ static const struct sdhci_ops sdhci_pci_ops = {
.enable_dma = sdhci_pci_enable_dma,
.set_bus_width = sdhci_pci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
.hw_reset = sdhci_pci_hw_reset,
};
diff --git a/drivers/mmc/host/sdhci-pltfm.c b/drivers/mmc/host/sdhci-pltfm.c
index 1fb89f44bd58..7e834fb78f42 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pltfm.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pltfm.c
@@ -48,6 +48,7 @@ static const struct sdhci_ops sdhci_pltfm_ops = {
.set_clock = sdhci_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
#ifdef CONFIG_OF
diff --git a/drivers/mmc/host/sdhci-pxav2.c b/drivers/mmc/host/sdhci-pxav2.c
index db5257bf032e..3c0f3c0a1cc8 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pxav2.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pxav2.c
@@ -116,6 +116,7 @@ static const struct sdhci_ops pxav2_sdhci_ops = {
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
.set_bus_width = pxav2_mmc_set_bus_width,
.reset = pxav2_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
#ifdef CONFIG_OF
diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c
index 13f8552ed0b3..4c5e52fbc6ee 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pxav3.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pxav3.c
@@ -174,6 +174,7 @@ static const struct sdhci_ops pxav3_sdhci_ops = {
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = pxav3_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static struct sdhci_pltfm_data sdhci_pxav3_pdata = {
diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
index 63496bd62196..a633418ddcfa 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-s3c.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-s3c.c
@@ -375,6 +375,7 @@ static struct sdhci_ops sdhci_s3c_ops = {
.get_min_clock = sdhci_s3c_get_min_clock,
.set_bus_width = sdhci_s3c_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static void sdhci_s3c_notify_change(struct platform_device *dev, int state)
diff --git a/drivers/mmc/host/sdhci-sirf.c b/drivers/mmc/host/sdhci-sirf.c
index 3b775348b470..17004531d089 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-sirf.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-sirf.c
@@ -32,6 +32,7 @@ static struct sdhci_ops sdhci_sirf_ops = {
.get_max_clock = sdhci_sirf_get_max_clk,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static struct sdhci_pltfm_data sdhci_sirf_pdata = {
diff --git a/drivers/mmc/host/sdhci-spear.c b/drivers/mmc/host/sdhci-spear.c
index f365a3fc7612..94ae64d23ef3 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-spear.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-spear.c
@@ -40,6 +40,7 @@ static const struct sdhci_ops sdhci_pltfm_ops = {
.set_clock = sdhci_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
/* gpio card detection interrupt handler */
diff --git a/drivers/mmc/host/sdhci-tegra.c b/drivers/mmc/host/sdhci-tegra.c
index a0a8b5cc3b0c..d06b6ff60432 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-tegra.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-tegra.c
@@ -156,6 +156,7 @@ static const struct sdhci_ops tegra_sdhci_ops = {
.set_clock = sdhci_set_clock,
.set_bus_width = tegra_sdhci_set_bus_width,
.reset = tegra_sdhci_reset,
=+DO NOT APPLY+= .set_uhs_signaling = sdhci_set_uhs_signaling,
};
static const struct sdhci_pltfm_data sdhci_tegra20_pdata = {
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 4b6ee2029604..d7e142ea55ef 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci.c
@@ -1404,6 +1404,28 @@ void sdhci_set_bus_width(struct sdhci_host *host, int width)
}
EXPORT_SYMBOL_GPL(sdhci_set_bus_width);
=+DO NOT APPLY+=void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing)
=+DO NOT APPLY+={
=+DO NOT APPLY+= u16 ctrl_2;
=+DO NOT APPLY+=
=+DO NOT APPLY+= ctrl_2 = sdhci_readw(host, SDHCI_HOST_CONTROL2);
=+DO NOT APPLY+= /* Select Bus Speed Mode for host */
=+DO NOT APPLY+= ctrl_2 &= ~SDHCI_CTRL_UHS_MASK;
=+DO NOT APPLY+= if ((timing == MMC_TIMING_MMC_HS200) ||
=+DO NOT APPLY+= (timing == MMC_TIMING_UHS_SDR104))
=+DO NOT APPLY+= ctrl_2 |= SDHCI_CTRL_UHS_SDR104;
=+DO NOT APPLY+= else if (timing == MMC_TIMING_UHS_SDR12)
=+DO NOT APPLY+= ctrl_2 |= SDHCI_CTRL_UHS_SDR12;
=+DO NOT APPLY+= else if (timing == MMC_TIMING_UHS_SDR25)
=+DO NOT APPLY+= ctrl_2 |= SDHCI_CTRL_UHS_SDR25;
=+DO NOT APPLY+= else if (timing == MMC_TIMING_UHS_SDR50)
=+DO NOT APPLY+= ctrl_2 |= SDHCI_CTRL_UHS_SDR50;
=+DO NOT APPLY+= else if (timing == MMC_TIMING_UHS_DDR50)
=+DO NOT APPLY+= ctrl_2 |= SDHCI_CTRL_UHS_DDR50;
=+DO NOT APPLY+= sdhci_writew(host, ctrl_2, SDHCI_HOST_CONTROL2);
=+DO NOT APPLY+=}
=+DO NOT APPLY+=EXPORT_SYMBOL_GPL(sdhci_set_uhs_signaling);
=+DO NOT APPLY+=
static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
{
unsigned long flags;
@@ -1507,25 +1529,7 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
host->ops->set_clock(host, host->clock);
}
- if (host->ops->set_uhs_signaling)
- host->ops->set_uhs_signaling(host, ios->timing);
- else {
- ctrl_2 = sdhci_readw(host, SDHCI_HOST_CONTROL2);
- /* Select Bus Speed Mode for host */
- ctrl_2 &= ~SDHCI_CTRL_UHS_MASK;
- if ((ios->timing == MMC_TIMING_MMC_HS200) ||
- (ios->timing == MMC_TIMING_UHS_SDR104))
- ctrl_2 |= SDHCI_CTRL_UHS_SDR104;
- else if (ios->timing == MMC_TIMING_UHS_SDR12)
- ctrl_2 |= SDHCI_CTRL_UHS_SDR12;
- else if (ios->timing == MMC_TIMING_UHS_SDR25)
- ctrl_2 |= SDHCI_CTRL_UHS_SDR25;
- else if (ios->timing == MMC_TIMING_UHS_SDR50)
- ctrl_2 |= SDHCI_CTRL_UHS_SDR50;
- else if (ios->timing == MMC_TIMING_UHS_DDR50)
- ctrl_2 |= SDHCI_CTRL_UHS_DDR50;
- sdhci_writew(host, ctrl_2, SDHCI_HOST_CONTROL2);
- }
=+DO NOT APPLY+= host->ops->set_uhs_signaling(host, ios->timing);
if (!(host->quirks2 & SDHCI_QUIRK2_PRESET_VALUE_BROKEN) &&
((ios->timing == MMC_TIMING_UHS_SDR12) ||
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 3179a8053019..7a35395e5f56 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci.h
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci.h
@@ -403,6 +403,7 @@ static inline bool sdhci_sdio_irq_enabled(struct sdhci_host *host)
void sdhci_set_clock(struct sdhci_host *host, unsigned int clock);
void sdhci_set_bus_width(struct sdhci_host *host, int width);
void sdhci_reset(struct sdhci_host *host, u8 mask);
=+DO NOT APPLY+=void sdhci_set_uhs_signaling(struct sdhci_host *host, unsigned timing);
#ifdef CONFIG_PM
extern int sdhci_suspend_host(struct sdhci_host *host);
--
1.8.3.1
^ permalink raw reply
* [PATCH RFC 19/31] mmc: sdhci: convert sdhci_set_clock() into a library function
From: Russell King @ 2014-02-18 15:10 UTC (permalink / raw)
To: linux-arm-kernel, linux-mmc
Cc: Barry Song, spear-devel, Stephen Warren, Anton Vorontsov,
Chris Ball, Michal Simek, Thierry Reding, Viresh Kumar, Ben Dooks,
linux-tegra, linuxppc-dev
In-Reply-To: <20140218150801.GD21483@n2100.arm.linux.org.uk>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
=-DO NOT APPLY-=
drivers/mmc/host/sdhci-acpi.c | 2 ++
drivers/mmc/host/sdhci-bcm-kona.c | 1 +
drivers/mmc/host/sdhci-bcm2835.c | 1 +
drivers/mmc/host/sdhci-cns3xxx.c | 3 +--
drivers/mmc/host/sdhci-dove.c | 1 +
drivers/mmc/host/sdhci-esdhc.h | 1 -
drivers/mmc/host/sdhci-of-arasan.c | 1 +
drivers/mmc/host/sdhci-of-hlwd.c | 1 +
drivers/mmc/host/sdhci-pci.c | 1 +
drivers/mmc/host/sdhci-pltfm.c | 1 +
drivers/mmc/host/sdhci-pxav2.c | 1 +
drivers/mmc/host/sdhci-pxav3.c | 1 +
drivers/mmc/host/sdhci-s3c.c | 19 ++++++++++++++-----
drivers/mmc/host/sdhci-sirf.c | 1 +
drivers/mmc/host/sdhci-spear.c | 1 +
drivers/mmc/host/sdhci-tegra.c | 1 +
drivers/mmc/host/sdhci.c | 17 ++++++-----------
drivers/mmc/host/sdhci.h | 1 +
include/linux/mmc/sdhci.h | 2 --
19 files changed, 36 insertions(+), 21 deletions(-)
diff --git a/drivers/mmc/host/sdhci-acpi.c b/drivers/mmc/host/sdhci-acpi.c
index 1ded6690be43..c20964e577e9 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-acpi.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-acpi.c
@@ -101,12 +101,14 @@ static void sdhci_acpi_int_hw_reset(struct sdhci_host *host)
}
static const struct sdhci_ops sdhci_acpi_ops_dflt = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.enable_dma = sdhci_acpi_enable_dma,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
};
static const struct sdhci_ops sdhci_acpi_ops_int = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.enable_dma = sdhci_acpi_enable_dma,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
diff --git a/drivers/mmc/host/sdhci-bcm-kona.c b/drivers/mmc/host/sdhci-bcm-kona.c
index 4c82ecb7a71e..77b14b1a24bc 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-bcm-kona.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-bcm-kona.c
@@ -205,6 +205,7 @@ static void sdhci_bcm_kona_init_74_clocks(struct sdhci_host *host,
}
static struct sdhci_ops sdhci_bcm_kona_ops = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.get_max_clock = sdhci_bcm_kona_get_max_clk,
.get_timeout_clock = sdhci_bcm_kona_get_timeout_clock,
.platform_send_init_74_clocks = sdhci_bcm_kona_init_74_clocks,
diff --git a/drivers/mmc/host/sdhci-bcm2835.c b/drivers/mmc/host/sdhci-bcm2835.c
index 289b1c80d5fc..74906d6008e1 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-bcm2835.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-bcm2835.c
@@ -131,6 +131,7 @@ static const struct sdhci_ops bcm2835_sdhci_ops = {
.read_l = bcm2835_sdhci_readl,
.read_w = bcm2835_sdhci_readw,
.read_b = bcm2835_sdhci_readb,
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
.get_min_clock = bcm2835_sdhci_get_min_clock,
.set_bus_width = sdhci_set_bus_width,
diff --git a/drivers/mmc/host/sdhci-cns3xxx.c b/drivers/mmc/host/sdhci-cns3xxx.c
index 416f4a4c2e35..587d73ef33ff 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-cns3xxx.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-cns3xxx.c
@@ -89,8 +89,7 @@ static const struct sdhci_pltfm_data sdhci_cns3xxx_pdata = {
SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK |
SDHCI_QUIRK_INVERTED_WRITE_PROTECT |
SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN |
- SDHCI_QUIRK_BROKEN_TIMEOUT_VAL |
- SDHCI_QUIRK_NONSTANDARD_CLOCK,
=+DO NOT APPLY+= SDHCI_QUIRK_BROKEN_TIMEOUT_VAL,
};
static int sdhci_cns3xxx_probe(struct platform_device *pdev)
diff --git a/drivers/mmc/host/sdhci-dove.c b/drivers/mmc/host/sdhci-dove.c
index 4c0191bff85d..4f4216115aed 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-dove.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-dove.c
@@ -86,6 +86,7 @@ static u32 sdhci_dove_readl(struct sdhci_host *host, int reg)
static const struct sdhci_ops sdhci_dove_ops = {
.read_w = sdhci_dove_readw,
.read_l = sdhci_dove_readl,
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
};
diff --git a/drivers/mmc/host/sdhci-esdhc.h b/drivers/mmc/host/sdhci-esdhc.h
index de69bddc3afc..3497cfaf683c 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-esdhc.h
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-esdhc.h
@@ -20,7 +20,6 @@
#define ESDHC_DEFAULT_QUIRKS (SDHCI_QUIRK_FORCE_BLK_SZ_2048 | \
SDHCI_QUIRK_NO_BUSY_IRQ | \
- SDHCI_QUIRK_NONSTANDARD_CLOCK | \
SDHCI_QUIRK_DATA_TIMEOUT_USES_SDCLK | \
SDHCI_QUIRK_PIO_NEEDS_DELAY)
diff --git a/drivers/mmc/host/sdhci-of-arasan.c b/drivers/mmc/host/sdhci-of-arasan.c
index faef21740584..f0ee594f25d1 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-of-arasan.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-of-arasan.c
@@ -52,6 +52,7 @@ static unsigned int sdhci_arasan_get_timeout_clock(struct sdhci_host *host)
}
static struct sdhci_ops sdhci_arasan_ops = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
.get_timeout_clock = sdhci_arasan_get_timeout_clock,
.set_bus_width = sdhci_set_bus_width,
diff --git a/drivers/mmc/host/sdhci-of-hlwd.c b/drivers/mmc/host/sdhci-of-hlwd.c
index fb01958cb18e..a4a1f0f2c0a0 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-of-hlwd.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-of-hlwd.c
@@ -58,6 +58,7 @@ static const struct sdhci_ops sdhci_hlwd_ops = {
.write_l = sdhci_hlwd_writel,
.write_w = sdhci_hlwd_writew,
.write_b = sdhci_hlwd_writeb,
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
};
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index 88a6c6b87d70..01ee60d17b04 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pci.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pci.c
@@ -1058,6 +1058,7 @@ static void sdhci_pci_hw_reset(struct sdhci_host *host)
}
static const struct sdhci_ops sdhci_pci_ops = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.enable_dma = sdhci_pci_enable_dma,
.set_bus_width = sdhci_pci_set_bus_width,
.reset = sdhci_reset,
diff --git a/drivers/mmc/host/sdhci-pltfm.c b/drivers/mmc/host/sdhci-pltfm.c
index bfbf467b61c7..1fb89f44bd58 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pltfm.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pltfm.c
@@ -45,6 +45,7 @@ unsigned int sdhci_pltfm_clk_get_max_clock(struct sdhci_host *host)
EXPORT_SYMBOL_GPL(sdhci_pltfm_clk_get_max_clock);
static const struct sdhci_ops sdhci_pltfm_ops = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
};
diff --git a/drivers/mmc/host/sdhci-pxav2.c b/drivers/mmc/host/sdhci-pxav2.c
index 2eee0c8b88eb..db5257bf032e 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pxav2.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pxav2.c
@@ -112,6 +112,7 @@ static void pxav2_mmc_set_bus_width(struct sdhci_host *host, int width)
}
static const struct sdhci_ops pxav2_sdhci_ops = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
.set_bus_width = pxav2_mmc_set_bus_width,
.reset = pxav2_reset,
diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c
index e6ea46c4d233..3d6b574528ea 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pxav3.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pxav3.c
@@ -170,6 +170,7 @@ static int pxav3_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs)
}
static const struct sdhci_ops pxav3_sdhci_ops = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.set_uhs_signaling = pxav3_set_uhs_signaling,
.platform_send_init_74_clocks = pxav3_gen_init_74_clocks,
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
index 18579555a5f7..63496bd62196 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-s3c.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-s3c.c
@@ -57,6 +57,8 @@ struct sdhci_s3c {
struct clk *clk_io;
struct clk *clk_bus[MAX_BUS_CLK];
=+DO NOT APPLY+=
=+DO NOT APPLY+= bool no_divider;
};
/**
@@ -69,6 +71,7 @@ struct sdhci_s3c {
*/
struct sdhci_s3c_drv_data {
unsigned int sdhci_quirks;
=+DO NOT APPLY+= bool no_divider;
};
static inline struct sdhci_s3c *to_s3c(struct sdhci_host *host)
@@ -153,7 +156,7 @@ static unsigned int sdhci_s3c_consider_clock(struct sdhci_s3c *ourhost,
* If controller uses a non-standard clock division, find the best clock
* speed possible with selected clock source and skip the division.
*/
- if (ourhost->host->quirks & SDHCI_QUIRK_NONSTANDARD_CLOCK) {
=+DO NOT APPLY+= if (ourhost->no_divider) {
rate = clk_round_rate(clksrc, wanted);
return wanted - rate;
}
@@ -191,8 +194,10 @@ static void sdhci_s3c_set_clock(struct sdhci_host *host, unsigned int clock)
host->mmc->actual_clock = 0;
/* don't bother if the clock is going off. */
- if (clock == 0)
=+DO NOT APPLY+= if (clock == 0) {
=+DO NOT APPLY+= sdhci_set_clock(host, clock);
return;
=+DO NOT APPLY+= }
for (src = 0; src < MAX_BUS_CLK; src++) {
delta = sdhci_s3c_consider_clock(ourhost, src, clock);
@@ -242,6 +247,8 @@ static void sdhci_s3c_set_clock(struct sdhci_host *host, unsigned int clock)
if (clock < 25 * 1000000)
ctrl |= (S3C_SDHCI_CTRL3_FCSEL3 | S3C_SDHCI_CTRL3_FCSEL2);
writel(ctrl, host->ioaddr + S3C_SDHCI_CONTROL3);
=+DO NOT APPLY+=
=+DO NOT APPLY+= sdhci_set_clock(host, clock);
}
/**
@@ -617,8 +624,10 @@ static int sdhci_s3c_probe(struct platform_device *pdev)
/* Setup quirks for the controller */
host->quirks |= SDHCI_QUIRK_NO_ENDATTR_IN_NOPDESC;
host->quirks |= SDHCI_QUIRK_NO_HISPD_BIT;
- if (drv_data)
=+DO NOT APPLY+= if (drv_data) {
host->quirks |= drv_data->sdhci_quirks;
=+DO NOT APPLY+= sc->no_divider = drv_data->no_divider;
=+DO NOT APPLY+= }
#ifndef CONFIG_MMC_SDHCI_S3C_DMA
@@ -667,7 +676,7 @@ static int sdhci_s3c_probe(struct platform_device *pdev)
* If controller does not have internal clock divider,
* we can use overriding functions instead of default.
*/
- if (host->quirks & SDHCI_QUIRK_NONSTANDARD_CLOCK) {
=+DO NOT APPLY+= if (sc->no_divider) {
sdhci_s3c_ops.set_clock = sdhci_cmu_set_clock;
sdhci_s3c_ops.get_min_clock = sdhci_cmu_get_min_clock;
sdhci_s3c_ops.get_max_clock = sdhci_cmu_get_max_clock;
@@ -813,7 +822,7 @@ static const struct dev_pm_ops sdhci_s3c_pmops = {
#if defined(CONFIG_CPU_EXYNOS4210) || defined(CONFIG_SOC_EXYNOS4212)
static struct sdhci_s3c_drv_data exynos4_sdhci_drv_data = {
- .sdhci_quirks = SDHCI_QUIRK_NONSTANDARD_CLOCK,
=+DO NOT APPLY+= .no_divider = true,
};
#define EXYNOS4_SDHCI_DRV_DATA ((kernel_ulong_t)&exynos4_sdhci_drv_data)
#else
diff --git a/drivers/mmc/host/sdhci-sirf.c b/drivers/mmc/host/sdhci-sirf.c
index 5d79e10e1ba2..3b775348b470 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-sirf.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-sirf.c
@@ -28,6 +28,7 @@ static unsigned int sdhci_sirf_get_max_clk(struct sdhci_host *host)
}
static struct sdhci_ops sdhci_sirf_ops = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.get_max_clock = sdhci_sirf_get_max_clk,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
diff --git a/drivers/mmc/host/sdhci-spear.c b/drivers/mmc/host/sdhci-spear.c
index ef990974a522..f365a3fc7612 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-spear.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-spear.c
@@ -37,6 +37,7 @@ struct spear_sdhci {
/* sdhci ops */
static const struct sdhci_ops sdhci_pltfm_ops = {
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.set_bus_width = sdhci_set_bus_width,
.reset = sdhci_reset,
};
diff --git a/drivers/mmc/host/sdhci-tegra.c b/drivers/mmc/host/sdhci-tegra.c
index 7754c0319fda..a0a8b5cc3b0c 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-tegra.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-tegra.c
@@ -153,6 +153,7 @@ static const struct sdhci_ops tegra_sdhci_ops = {
.read_l = tegra_sdhci_readl,
.read_w = tegra_sdhci_readw,
.write_l = tegra_sdhci_writel,
=+DO NOT APPLY+= .set_clock = sdhci_set_clock,
.set_bus_width = tegra_sdhci_set_bus_width,
.reset = tegra_sdhci_reset,
};
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index c370e6bd7ecd..ddd57bc235ab 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci.c
@@ -1112,19 +1112,13 @@ static u16 sdhci_get_preset_value(struct sdhci_host *host)
return preset;
}
-static void sdhci_set_clock(struct sdhci_host *host, unsigned int clock)
=+DO NOT APPLY+=void sdhci_set_clock(struct sdhci_host *host, unsigned int clock)
{
int div = 0; /* Initialized for compiler warning */
int real_div = div, clk_mul = 1;
u16 clk = 0;
unsigned long timeout;
- if (host->ops->set_clock) {
- host->ops->set_clock(host, clock);
- if (host->quirks & SDHCI_QUIRK_NONSTANDARD_CLOCK)
- return;
- }
-
host->mmc->actual_clock = 0;
sdhci_writew(host, 0, SDHCI_CLOCK_CONTROL);
@@ -1221,6 +1215,7 @@ static void sdhci_set_clock(struct sdhci_host *host, unsigned int clock)
clk |= SDHCI_CLOCK_CARD_EN;
sdhci_writew(host, clk, SDHCI_CLOCK_CONTROL);
}
=+DO NOT APPLY+=EXPORT_SYMBOL_GPL(sdhci_set_clock);
static int sdhci_set_power(struct sdhci_host *host, unsigned short power)
{
@@ -1439,7 +1434,7 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
sdhci_enable_preset_value(host, false);
if (!ios->clock || ios->clock != host->clock) {
- sdhci_set_clock(host, ios->clock);
=+DO NOT APPLY+= host->ops->set_clock(host, ios->clock);
host->clock = ios->clock;
}
@@ -1509,7 +1504,7 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
/* Re-enable SD Clock */
- sdhci_set_clock(host, host->clock);
=+DO NOT APPLY+= host->ops->set_clock(host, host->clock);
}
@@ -1553,7 +1548,7 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
}
/* Re-enable SD Clock */
- sdhci_set_clock(host, host->clock);
=+DO NOT APPLY+= host->ops->set_clock(host, host->clock);
} else
sdhci_writeb(host, ctrl, SDHCI_HOST_CONTROL);
@@ -2125,7 +2120,7 @@ static void sdhci_tasklet_finish(unsigned long param)
/* Some controllers need this kick or reset won't work here */
if (host->quirks & SDHCI_QUIRK_CLOCK_BEFORE_RESET)
/* This is to force an update */
- sdhci_set_clock(host, host->clock);
=+DO NOT APPLY+= host->ops->set_clock(host, host->clock);
/* Spec says we should do both at the same time, but Ricoh
controllers do not like that. */
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 7d84cb3b0e00..ac20195f667b 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci.h
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci.h
@@ -400,6 +400,7 @@ static inline bool sdhci_sdio_irq_enabled(struct sdhci_host *host)
return !!(host->flags & SDHCI_SDIO_IRQ_ENABLED);
}
=+DO NOT APPLY+=void sdhci_set_clock(struct sdhci_host *host, unsigned int clock);
void sdhci_set_bus_width(struct sdhci_host *host, int width);
void sdhci_reset(struct sdhci_host *host, u8 mask);
diff --git a/include/linux/mmc/sdhci.h b/include/linux/mmc/sdhci.h
index 801579313789..20daa39e3a68 100644
=-DO NOT APPLY-= a/include/linux/mmc/sdhci.h
=+DO NOT APPLY+= b/include/linux/mmc/sdhci.h
@@ -57,8 +57,6 @@ struct sdhci_host {
#define SDHCI_QUIRK_BROKEN_CARD_DETECTION (1<<15)
/* Controller reports inverted write-protect state */
#define SDHCI_QUIRK_INVERTED_WRITE_PROTECT (1<<16)
-/* Controller has nonstandard clock management */
-#define SDHCI_QUIRK_NONSTANDARD_CLOCK (1<<17)
/* Controller does not like fast PIO transfers */
#define SDHCI_QUIRK_PIO_NEEDS_DELAY (1<<18)
/* Controller has to be forced to use block size of 2048 bytes */
--
1.8.3.1
^ permalink raw reply
* [PATCH RFC 12/31] mmc: sdhci: convert reset into a library function
From: Russell King @ 2014-02-18 15:10 UTC (permalink / raw)
To: linux-arm-kernel, linux-mmc
Cc: Barry Song, spear-devel, Stephen Warren, Anton Vorontsov,
Chris Ball, Michal Simek, Thierry Reding, Viresh Kumar, Ben Dooks,
linux-tegra, linuxppc-dev
In-Reply-To: <20140218150801.GD21483@n2100.arm.linux.org.uk>
Rather than having platform_reset_enter/platform_reset_exit methods,
turn the core of the reset handling into a library function which
platforms can call at the appropriate moment in their (new) reset
method.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
=-DO NOT APPLY-=
drivers/mmc/host/sdhci-acpi.c | 2 ++
drivers/mmc/host/sdhci-bcm-kona.c | 1 +
drivers/mmc/host/sdhci-bcm2835.c | 1 +
drivers/mmc/host/sdhci-cns3xxx.c | 1 +
drivers/mmc/host/sdhci-dove.c | 1 +
drivers/mmc/host/sdhci-esdhc-imx.c | 1 +
drivers/mmc/host/sdhci-of-arasan.c | 1 +
drivers/mmc/host/sdhci-of-esdhc.c | 1 +
drivers/mmc/host/sdhci-of-hlwd.c | 1 +
drivers/mmc/host/sdhci-pci.c | 1 +
drivers/mmc/host/sdhci-pltfm.c | 1 +
drivers/mmc/host/sdhci-pxav2.c | 6 +++--
drivers/mmc/host/sdhci-pxav3.c | 6 +++--
drivers/mmc/host/sdhci-s3c.c | 1 +
drivers/mmc/host/sdhci-sirf.c | 1 +
drivers/mmc/host/sdhci-spear.c | 1 +
drivers/mmc/host/sdhci-tegra.c | 6 +++--
drivers/mmc/host/sdhci.c | 47 +++++++++++++++++++-------------------
drivers/mmc/host/sdhci.h | 4 ++--
19 files changed, 53 insertions(+), 31 deletions(-)
diff --git a/drivers/mmc/host/sdhci-acpi.c b/drivers/mmc/host/sdhci-acpi.c
index b138bfdd3843..1ded6690be43 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-acpi.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-acpi.c
@@ -103,11 +103,13 @@ static void sdhci_acpi_int_hw_reset(struct sdhci_host *host)
static const struct sdhci_ops sdhci_acpi_ops_dflt = {
.enable_dma = sdhci_acpi_enable_dma,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static const struct sdhci_ops sdhci_acpi_ops_int = {
.enable_dma = sdhci_acpi_enable_dma,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
.hw_reset = sdhci_acpi_int_hw_reset,
};
diff --git a/drivers/mmc/host/sdhci-bcm-kona.c b/drivers/mmc/host/sdhci-bcm-kona.c
index bf851a6e4e8e..4c82ecb7a71e 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-bcm-kona.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-bcm-kona.c
@@ -209,6 +209,7 @@ static struct sdhci_ops sdhci_bcm_kona_ops = {
.get_timeout_clock = sdhci_bcm_kona_get_timeout_clock,
.platform_send_init_74_clocks = sdhci_bcm_kona_init_74_clocks,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
.card_event = sdhci_bcm_kona_card_event,
};
diff --git a/drivers/mmc/host/sdhci-bcm2835.c b/drivers/mmc/host/sdhci-bcm2835.c
index 7ab69af979b4..289b1c80d5fc 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-bcm2835.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-bcm2835.c
@@ -134,6 +134,7 @@ static const struct sdhci_ops bcm2835_sdhci_ops = {
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
.get_min_clock = bcm2835_sdhci_get_min_clock,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static const struct sdhci_pltfm_data bcm2835_sdhci_pdata = {
diff --git a/drivers/mmc/host/sdhci-cns3xxx.c b/drivers/mmc/host/sdhci-cns3xxx.c
index 5e0cc9c47887..87af66bb1ea8 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-cns3xxx.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-cns3xxx.c
@@ -83,6 +83,7 @@ static const struct sdhci_ops sdhci_cns3xxx_ops = {
.get_max_clock = sdhci_cns3xxx_get_max_clk,
.set_clock = sdhci_cns3xxx_set_clock,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static const struct sdhci_pltfm_data sdhci_cns3xxx_pdata = {
diff --git a/drivers/mmc/host/sdhci-dove.c b/drivers/mmc/host/sdhci-dove.c
index 7eef4890fd22..4c0191bff85d 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-dove.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-dove.c
@@ -87,6 +87,7 @@ static const struct sdhci_ops sdhci_dove_ops = {
.read_w = sdhci_dove_readw,
.read_l = sdhci_dove_readl,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static const struct sdhci_pltfm_data sdhci_dove_pdata = {
diff --git a/drivers/mmc/host/sdhci-esdhc-imx.c b/drivers/mmc/host/sdhci-esdhc-imx.c
index 568239d84cbe..b1d74fa33c5f 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-esdhc-imx.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-esdhc-imx.c
@@ -888,6 +888,7 @@ static struct sdhci_ops sdhci_esdhc_ops = {
.get_ro = esdhc_pltfm_get_ro,
.set_bus_width = esdhc_pltfm_set_bus_width,
.set_uhs_signaling = esdhc_set_uhs_signaling,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static const struct sdhci_pltfm_data sdhci_esdhc_imx_pdata = {
diff --git a/drivers/mmc/host/sdhci-of-arasan.c b/drivers/mmc/host/sdhci-of-arasan.c
index 9bb1dd263a45..faef21740584 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-of-arasan.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-of-arasan.c
@@ -55,6 +55,7 @@ static struct sdhci_ops sdhci_arasan_ops = {
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
.get_timeout_clock = sdhci_arasan_get_timeout_clock,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static struct sdhci_pltfm_data sdhci_arasan_pdata = {
diff --git a/drivers/mmc/host/sdhci-of-esdhc.c b/drivers/mmc/host/sdhci-of-esdhc.c
index 86b8326e77c3..4530f9957f20 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-of-esdhc.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-of-esdhc.c
@@ -309,6 +309,7 @@ static const struct sdhci_ops sdhci_esdhc_ops = {
#endif
.adma_workaround = esdhci_of_adma_workaround,
.set_bus_width = esdhc_pltfm_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static const struct sdhci_pltfm_data sdhci_esdhc_pdata = {
diff --git a/drivers/mmc/host/sdhci-of-hlwd.c b/drivers/mmc/host/sdhci-of-hlwd.c
index 4d5d0015e392..fb01958cb18e 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-of-hlwd.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-of-hlwd.c
@@ -59,6 +59,7 @@ static const struct sdhci_ops sdhci_hlwd_ops = {
.write_w = sdhci_hlwd_writew,
.write_b = sdhci_hlwd_writeb,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static const struct sdhci_pltfm_data sdhci_hlwd_pdata = {
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index 59cc938b450d..88a6c6b87d70 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pci.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pci.c
@@ -1060,6 +1060,7 @@ static void sdhci_pci_hw_reset(struct sdhci_host *host)
static const struct sdhci_ops sdhci_pci_ops = {
.enable_dma = sdhci_pci_enable_dma,
.set_bus_width = sdhci_pci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
.hw_reset = sdhci_pci_hw_reset,
};
diff --git a/drivers/mmc/host/sdhci-pltfm.c b/drivers/mmc/host/sdhci-pltfm.c
index 40b0fe224f9d..bfbf467b61c7 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pltfm.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pltfm.c
@@ -46,6 +46,7 @@ EXPORT_SYMBOL_GPL(sdhci_pltfm_clk_get_max_clock);
static const struct sdhci_ops sdhci_pltfm_ops = {
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
#ifdef CONFIG_OF
diff --git a/drivers/mmc/host/sdhci-pxav2.c b/drivers/mmc/host/sdhci-pxav2.c
index d24c282e5eb8..2eee0c8b88eb 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pxav2.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pxav2.c
@@ -51,11 +51,13 @@
#define MMC_CARD 0x1000
#define MMC_WIDTH 0x0100
-static void pxav2_set_private_registers(struct sdhci_host *host, u8 mask)
=+DO NOT APPLY+=static void pxav2_reset(struct sdhci_host *host, u8 mask)
{
struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
struct sdhci_pxa_platdata *pdata = pdev->dev.platform_data;
=+DO NOT APPLY+= sdhci_reset(host, mask);
=+DO NOT APPLY+=
if (mask == SDHCI_RESET_ALL) {
u16 tmp = 0;
@@ -111,8 +113,8 @@ static void pxav2_mmc_set_bus_width(struct sdhci_host *host, int width)
static const struct sdhci_ops pxav2_sdhci_ops = {
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
- .platform_reset_exit = pxav2_set_private_registers,
.set_bus_width = pxav2_mmc_set_bus_width,
=+DO NOT APPLY+= .reset = pxav2_reset,
};
#ifdef CONFIG_OF
diff --git a/drivers/mmc/host/sdhci-pxav3.c b/drivers/mmc/host/sdhci-pxav3.c
index 89c932d9685e..e6ea46c4d233 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-pxav3.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-pxav3.c
@@ -57,11 +57,13 @@
#define SDCE_MISC_INT (1<<2)
#define SDCE_MISC_INT_EN (1<<1)
-static void pxav3_set_private_registers(struct sdhci_host *host, u8 mask)
=+DO NOT APPLY+=static void pxav3_reset(struct sdhci_host *host, u8 mask)
{
struct platform_device *pdev = to_platform_device(mmc_dev(host->mmc));
struct sdhci_pxa_platdata *pdata = pdev->dev.platform_data;
=+DO NOT APPLY+= sdhci_reset(host, mask);
=+DO NOT APPLY+=
if (mask == SDHCI_RESET_ALL) {
/*
* tune timing of read data/command when crc error happen
@@ -168,11 +170,11 @@ static int pxav3_set_uhs_signaling(struct sdhci_host *host, unsigned int uhs)
}
static const struct sdhci_ops pxav3_sdhci_ops = {
- .platform_reset_exit = pxav3_set_private_registers,
.set_uhs_signaling = pxav3_set_uhs_signaling,
.platform_send_init_74_clocks = pxav3_gen_init_74_clocks,
.get_max_clock = sdhci_pltfm_clk_get_max_clock,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = pxav3_reset,
};
static struct sdhci_pltfm_data sdhci_pxav3_pdata = {
diff --git a/drivers/mmc/host/sdhci-s3c.c b/drivers/mmc/host/sdhci-s3c.c
index 1c4ebd69d4e5..619047be00c9 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-s3c.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-s3c.c
@@ -366,6 +366,7 @@ static struct sdhci_ops sdhci_s3c_ops = {
.set_clock = sdhci_s3c_set_clock,
.get_min_clock = sdhci_s3c_get_min_clock,
.set_bus_width = sdhci_s3c_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static void sdhci_s3c_notify_change(struct platform_device *dev, int state)
diff --git a/drivers/mmc/host/sdhci-sirf.c b/drivers/mmc/host/sdhci-sirf.c
index 16fcd48f9556..5d79e10e1ba2 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-sirf.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-sirf.c
@@ -30,6 +30,7 @@ static unsigned int sdhci_sirf_get_max_clk(struct sdhci_host *host)
static struct sdhci_ops sdhci_sirf_ops = {
.get_max_clock = sdhci_sirf_get_max_clk,
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
static struct sdhci_pltfm_data sdhci_sirf_pdata = {
diff --git a/drivers/mmc/host/sdhci-spear.c b/drivers/mmc/host/sdhci-spear.c
index 549e9e0edb27..ef990974a522 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-spear.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-spear.c
@@ -38,6 +38,7 @@ struct spear_sdhci {
/* sdhci ops */
static const struct sdhci_ops sdhci_pltfm_ops = {
.set_bus_width = sdhci_set_bus_width,
=+DO NOT APPLY+= .reset = sdhci_reset,
};
/* gpio card detection interrupt handler */
diff --git a/drivers/mmc/host/sdhci-tegra.c b/drivers/mmc/host/sdhci-tegra.c
index feed799b827a..7754c0319fda 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci-tegra.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci-tegra.c
@@ -108,12 +108,14 @@ static unsigned int tegra_sdhci_get_ro(struct sdhci_host *host)
return mmc_gpio_get_ro(host->mmc);
}
-static void tegra_sdhci_reset_exit(struct sdhci_host *host, u8 mask)
=+DO NOT APPLY+=static void tegra_sdhci_reset(struct sdhci_host *host, u8 mask)
{
struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
struct sdhci_tegra *tegra_host = pltfm_host->priv;
const struct sdhci_tegra_soc_data *soc_data = tegra_host->soc_data;
=+DO NOT APPLY+= sdhci_reset(host, mask);
=+DO NOT APPLY+=
if (!(mask & SDHCI_RESET_ALL))
return;
@@ -152,7 +154,7 @@ static const struct sdhci_ops tegra_sdhci_ops = {
.read_w = tegra_sdhci_readw,
.write_l = tegra_sdhci_writel,
.set_bus_width = tegra_sdhci_set_bus_width,
- .platform_reset_exit = tegra_sdhci_reset_exit,
=+DO NOT APPLY+= .reset = tegra_sdhci_reset,
};
static const struct sdhci_pltfm_data sdhci_tegra20_pdata = {
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index ad744941fdf8..10540192d544 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci.c
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci.c
@@ -163,17 +163,9 @@ static void sdhci_disable_card_detection(struct sdhci_host *host)
sdhci_set_card_detection(host, false);
}
-static void sdhci_reset(struct sdhci_host *host, u8 mask)
=+DO NOT APPLY+=void sdhci_reset(struct sdhci_host *host, u8 mask)
{
unsigned long timeout;
- if (host->quirks & SDHCI_QUIRK_NO_CARD_NO_RESET) {
- if (!(sdhci_readl(host, SDHCI_PRESENT_STATE) &
- SDHCI_CARD_PRESENT))
- return;
- }
-
- if (host->ops->platform_reset_enter)
- host->ops->platform_reset_enter(host, mask);
sdhci_writeb(host, mask, SDHCI_SOFTWARE_RESET);
@@ -198,9 +190,18 @@ static void sdhci_reset(struct sdhci_host *host, u8 mask)
timeout--;
mdelay(1);
}
=+DO NOT APPLY+=}
=+DO NOT APPLY+=EXPORT_SYMBOL_GPL(sdhci_reset);
=+DO NOT APPLY+=
=+DO NOT APPLY+=static void sdhci_do_reset(struct sdhci_host *host, u8 mask)
=+DO NOT APPLY+={
=+DO NOT APPLY+= if (host->quirks & SDHCI_QUIRK_NO_CARD_NO_RESET) {
=+DO NOT APPLY+= if (!(sdhci_readl(host, SDHCI_PRESENT_STATE) &
=+DO NOT APPLY+= SDHCI_CARD_PRESENT))
=+DO NOT APPLY+= return;
=+DO NOT APPLY+= }
- if (host->ops->platform_reset_exit)
- host->ops->platform_reset_exit(host, mask);
=+DO NOT APPLY+= host->ops->reset(host, mask);
if (host->quirks & SDHCI_QUIRK_RESTORE_IRQS_AFTER_RESET) {
sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
@@ -218,9 +219,9 @@ static void sdhci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios);
static void sdhci_init(struct sdhci_host *host, int soft)
{
if (soft)
- sdhci_reset(host, SDHCI_RESET_CMD|SDHCI_RESET_DATA);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_CMD|SDHCI_RESET_DATA);
else
- sdhci_reset(host, SDHCI_RESET_ALL);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_ALL);
host->ier = SDHCI_INT_BUS_POWER | SDHCI_INT_DATA_END_BIT |
SDHCI_INT_DATA_CRC | SDHCI_INT_DATA_TIMEOUT |
@@ -962,8 +963,8 @@ static void sdhci_finish_data(struct sdhci_host *host)
* upon error conditions.
*/
if (data->error) {
- sdhci_reset(host, SDHCI_RESET_CMD);
- sdhci_reset(host, SDHCI_RESET_DATA);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_CMD);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_DATA);
}
sdhci_send_command(host, data->stop);
@@ -1585,7 +1586,7 @@ static void sdhci_do_set_ios(struct sdhci_host *host, struct mmc_ios *ios)
* it on each ios seems to solve the problem.
*/
if(host->quirks & SDHCI_QUIRK_RESET_CMD_DATA_ON_IOS)
- sdhci_reset(host, SDHCI_RESET_CMD | SDHCI_RESET_DATA);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_CMD | SDHCI_RESET_DATA);
mmiowb();
spin_unlock_irqrestore(&host->lock, flags);
@@ -2082,8 +2083,8 @@ static void sdhci_card_event(struct mmc_host *mmc)
pr_err("%s: Resetting controller.\n",
mmc_hostname(host->mmc));
- sdhci_reset(host, SDHCI_RESET_CMD);
- sdhci_reset(host, SDHCI_RESET_DATA);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_CMD);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_DATA);
host->mrq->cmd->error = -ENOMEDIUM;
tasklet_schedule(&host->finish_tasklet);
@@ -2151,8 +2152,8 @@ static void sdhci_tasklet_finish(unsigned long param)
/* Spec says we should do both at the same time, but Ricoh
controllers do not like that. */
- sdhci_reset(host, SDHCI_RESET_CMD);
- sdhci_reset(host, SDHCI_RESET_DATA);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_CMD);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_DATA);
}
host->mrq = NULL;
@@ -2789,7 +2790,7 @@ int sdhci_add_host(struct sdhci_host *host)
if (debug_quirks2)
host->quirks2 = debug_quirks2;
- sdhci_reset(host, SDHCI_RESET_ALL);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_ALL);
host->version = sdhci_readw(host, SDHCI_HOST_VERSION);
host->version = (host->version & SDHCI_SPEC_VER_MASK)
@@ -3272,7 +3273,7 @@ int sdhci_add_host(struct sdhci_host *host)
#ifdef SDHCI_USE_LEDS_CLASS
reset:
- sdhci_reset(host, SDHCI_RESET_ALL);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_ALL);
sdhci_writel(host, 0, SDHCI_INT_ENABLE);
sdhci_writel(host, 0, SDHCI_SIGNAL_ENABLE);
free_irq(host->irq, host);
@@ -3314,7 +3315,7 @@ void sdhci_remove_host(struct sdhci_host *host, int dead)
#endif
if (!dead)
- sdhci_reset(host, SDHCI_RESET_ALL);
=+DO NOT APPLY+= sdhci_do_reset(host, SDHCI_RESET_ALL);
sdhci_writel(host, 0, SDHCI_INT_ENABLE);
sdhci_writel(host, 0, SDHCI_SIGNAL_ENABLE);
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 0301f928eb11..7d84cb3b0e00 100644
=-DO NOT APPLY-= a/drivers/mmc/host/sdhci.h
=+DO NOT APPLY+= b/drivers/mmc/host/sdhci.h
@@ -285,8 +285,7 @@ struct sdhci_ops {
void (*platform_send_init_74_clocks)(struct sdhci_host *host,
u8 power_mode);
unsigned int (*get_ro)(struct sdhci_host *host);
- void (*platform_reset_enter)(struct sdhci_host *host, u8 mask);
- void (*platform_reset_exit)(struct sdhci_host *host, u8 mask);
=+DO NOT APPLY+= void (*reset)(struct sdhci_host *host, u8 mask);
int (*platform_execute_tuning)(struct sdhci_host *host, u32 opcode);
int (*set_uhs_signaling)(struct sdhci_host *host, unsigned int uhs);
void (*hw_reset)(struct sdhci_host *host);
@@ -402,6 +401,7 @@ static inline bool sdhci_sdio_irq_enabled(struct sdhci_host *host)
}
void sdhci_set_bus_width(struct sdhci_host *host, int width);
=+DO NOT APPLY+=void sdhci_reset(struct sdhci_host *host, u8 mask);
#ifdef CONFIG_PM
extern int sdhci_suspend_host(struct sdhci_host *host);
--
1.8.3.1
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox