linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Laurent Dufour <ldufour@linux.ibm.com>
To: Tyrel Datwyler <tyreld@linux.ibm.com>,
	mpe@ellerman.id.au, benh@kernel.crashing.org, paulus@samba.org
Cc: nathanl@linux.ibm.com, linuxppc-dev@lists.ozlabs.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] pseries/drmem: update LMBs after LPM
Date: Tue, 27 Apr 2021 19:28:51 +0200	[thread overview]
Message-ID: <c80a76eb-1e10-95f7-def1-5eb26d66efb2@linux.ibm.com> (raw)
In-Reply-To: <8b36065d-e4b4-bc0c-cf69-f01c91e49061@linux.ibm.com>

Le 27/04/2021 à 19:01, Tyrel Datwyler a écrit :
> On 4/27/21 8:01 AM, Laurent Dufour wrote:
>> After a LPM, the device tree node ibm,dynamic-reconfiguration-memory may be
>> updated by the hypervisor in the case the NUMA topology of the LPAR's
>> memory is updated.
>>
>> This is caught by the kernel, but the memory's node is updated because
>> there is no way to move a memory block between nodes.
>>
>> If later a memory block is added or removed, drmem_update_dt() is called
>> and it is overwriting the DT node to match the added or removed LMB. But
>> the LMB's associativity node has not been updated after the DT node update
>> and thus the node is overwritten by the Linux's topology instead of the
>> hypervisor one.
>>
>> Introduce a hook called when the ibm,dynamic-reconfiguration-memory node is
>> updated to force an update of the LMB's associativity.
>>
>> Signed-off-by: Laurent Dufour <ldufour@linux.ibm.com>
>> ---
>>   arch/powerpc/include/asm/drmem.h          |  1 +
>>   arch/powerpc/mm/drmem.c                   | 48 +++++++++++++++++++++++
>>   arch/powerpc/platforms/pseries/mobility.c |  9 +++++
>>   3 files changed, 58 insertions(+)
>>
>> diff --git a/arch/powerpc/include/asm/drmem.h b/arch/powerpc/include/asm/drmem.h
>> index bf2402fed3e0..55c2c25085b0 100644
>> --- a/arch/powerpc/include/asm/drmem.h
>> +++ b/arch/powerpc/include/asm/drmem.h
>> @@ -111,6 +111,7 @@ int drmem_update_dt(void);
>>   int __init
>>   walk_drmem_lmbs_early(unsigned long node, void *data,
>>   		      int (*func)(struct drmem_lmb *, const __be32 **, void *));
>> +void drmem_update_lmbs(void);
>>   #endif
>>
>>   static inline void invalidate_lmb_associativity_index(struct drmem_lmb *lmb)
>> diff --git a/arch/powerpc/mm/drmem.c b/arch/powerpc/mm/drmem.c
>> index 9af3832c9d8d..46074bdfdb3c 100644
>> --- a/arch/powerpc/mm/drmem.c
>> +++ b/arch/powerpc/mm/drmem.c
>> @@ -307,6 +307,54 @@ int __init walk_drmem_lmbs_early(unsigned long node, void *data,
>>   	return ret;
>>   }
>>
>> +/*
>> + * Update the LMB associativity index.
>> + */
>> +static int update_lmb(struct drmem_lmb *updated_lmb,
>> +		      __maybe_unused const __be32 **usm,
>> +		      __maybe_unused void *data)
>> +{
>> +	struct drmem_lmb *lmb;
>> +
>> +	/*
>> +	 * Brut force there may be better way to fetch the LMB
>> +	 */
>> +	for_each_drmem_lmb(lmb) {
>> +		if (lmb->drc_index != updated_lmb->drc_index)
>> +			continue;
>> +
>> +		lmb->aa_index = updated_lmb->aa_index;
>> +		break;
>> +	}
>> +	return 0;
>> +}
>> +
>> +/*
>> + * Update the LMB associativity index.
>> + *
>> + * This needs to be called when the hypervisor is updating the
>> + * dynamic-reconfiguration-memory node property.
>> + */
>> +void drmem_update_lmbs(void)
>> +{
>> +	struct device_node *node;
>> +	const __be32 *prop;
>> +
>> +	node = of_find_node_by_path("/ibm,dynamic-reconfiguration-memory");
>> +	if (!node)
>> +		return;
>> +
>> +	prop = of_get_property(node, "ibm,dynamic-memory", NULL);
>> +	if (prop) {
>> +		__walk_drmem_v1_lmbs(prop, NULL, NULL, update_lmb);
>> +	} else {
>> +		prop = of_get_property(node, "ibm,dynamic-memory-v2", NULL);
>> +		if (prop)
>> +			__walk_drmem_v2_lmbs(prop, NULL, NULL, update_lmb);
>> +	}
>> +
>> +	of_node_put(node);
>> +}
>>   #endif
>>
>>   static int init_drmem_lmb_size(struct device_node *dn)
>> diff --git a/arch/powerpc/platforms/pseries/mobility.c b/arch/powerpc/platforms/pseries/mobility.c
>> index ea4d6a660e0d..c68eccc6e8df 100644
>> --- a/arch/powerpc/platforms/pseries/mobility.c
>> +++ b/arch/powerpc/platforms/pseries/mobility.c
>> @@ -25,6 +25,7 @@
>>
>>   #include <asm/machdep.h>
>>   #include <asm/rtas.h>
>> +#include <asm/drmem.h>
>>   #include "pseries.h"
>>   #include "../../kernel/cacheinfo.h"
>>
>> @@ -237,6 +238,7 @@ int pseries_devicetree_update(s32 scope)
>>   	__be32 *data;
>>   	int update_nodes_token;
>>   	int rc;
>> +	bool drmem_updated = false;
>>
>>   	update_nodes_token = rtas_token("ibm,update-nodes");
>>   	if (update_nodes_token == RTAS_UNKNOWN_SERVICE)
>> @@ -271,6 +273,10 @@ int pseries_devicetree_update(s32 scope)
>>   					continue;
>>   				}
>>
>> +				if (!strcmp(np->full_name,
>> +					    "ibm,dynamic-reconfiguration-memory"))
>> +					drmem_updated = true;
> 
> Is there a reason that we can't use the existing pseries_memory_notifier()
> callback in pseries/hotplug-memory.c to trigger the drmem_update_lmbs() when
> either the ibm,dynamic-memory or ibm,dynamic-memory-v2 properties are updated?

Thanks a lot Tyrel!

That's far more elegant, I'll send a v2 soon.

Laurent.

> Something like:
> 
> static int pseries_memory_notifier(struct notifier_block *nb,
>                                     unsigned long action, void *data)
> {
>          struct of_reconfig_data *rd = data;
>          int err = 0;
> 
>          switch (action) {
>          case OF_RECONFIG_ATTACH_NODE:
>                  err = pseries_add_mem_node(rd->dn);
>                  break;
>          case OF_RECONFIG_DETACH_NODE:
>                  err = pseries_remove_mem_node(rd->dn);
>                  break;
> 	case OF_RECONFIG_UPDATE_PROPERTY:
> 		if (!strcmp(rd->dn->full_name, "ibm,dynamic-reconfiguration-memory"));
> 			drmem_update_lmbs(rd->prop);
> 		break;
>          }
>          return notifier_from_errno(err);
> }
> 
> Your drmem_update_lmbs() would need to be updated to take a property and to
> check the property name against ibm,dyanmic-memory[-v2].
> 
> -Tyrel
> 
>> +
>>   				switch (action) {
>>   				case DELETE_DT_NODE:
>>   					delete_dt_node(np);
>> @@ -293,6 +299,9 @@ int pseries_devicetree_update(s32 scope)
>>   	} while (rc == 1);
>>
>>   	kfree(rtas_buf);
>> +
>> +	if (drmem_updated)
>> +		drmem_update_lmbs();
>>   	return rc;
>>   }
>>
> 


  reply	other threads:[~2021-04-27 17:29 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-27 15:01 [PATCH] pseries/drmem: update LMBs after LPM Laurent Dufour
2021-04-27 17:01 ` Tyrel Datwyler
2021-04-27 17:28   ` Laurent Dufour [this message]
  -- strict thread matches above, loose matches on Subject: below --
2021-04-27 18:13 Laurent Dufour
2021-04-27 18:17 ` Laurent Dufour
2021-04-27 19:37 ` Tyrel Datwyler

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c80a76eb-1e10-95f7-def1-5eb26d66efb2@linux.ibm.com \
    --to=ldufour@linux.ibm.com \
    --cc=benh@kernel.crashing.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mpe@ellerman.id.au \
    --cc=nathanl@linux.ibm.com \
    --cc=paulus@samba.org \
    --cc=tyreld@linux.ibm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).