1. Patchew
  2. linux
  3. [PATCH v4 00/13] x86/resctrl: Support for AMD QoS new features and bug fix
  4. View patch

[PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD

Babu Moger posted 13 patches 3 years, 7 months ago
[PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
Posted by Babu Moger 3 years, 7 months ago 
AMD systems support zero CBM (capacity bit mask) for L3 allocation.
That is reflected in rdt_init_res_defs_amd() by:

	r->cache.arch_has_empty_bitmaps = true;

However given the unified code in cbm_validate(), checking for:
	val == 0 && !arch_has_empty_bitmaps

is not enough because of another check in cbm_validate():

	if ((zero_bit - first_bit) < r->cache.min_cbm_bits)

The default value of r->cache.min_cbm_bits = 1.

Leading to:

	$ cd /sys/fs/resctrl
	$ mkdir foo
	$ cd foo
	$ echo L3:0=0 > schemata
          -bash: echo: write error: Invalid argument
	$ cat /sys/fs/resctrl/info/last_cmd_status
	  Need at least 1 bits in the mask

Fix the issue by initializing the min_cbm_bits to 0 for AMD. Also,
remove the default setting of min_cbm_bits and initialize it separately.

After the fix
	$ cd /sys/fs/resctrl
	$ mkdir foo
	$ cd foo
	$ echo L3:0=0 > schemata
	$ cat /sys/fs/resctrl/info/last_cmd_status
	  ok

Link: https://lore.kernel.org/lkml/20220517001234.3137157-1-eranian@google.com/
Fixes: 316e7f901f5a ("x86/resctrl: Add struct rdt_cache::arch_has_{sparse, empty}_bitmaps")
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Babu Moger <babu.moger@amd.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
---
 arch/x86/kernel/cpu/resctrl/core.c |    8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
index bb1c3f5f60c8..a5c51a14fbce 100644
--- a/arch/x86/kernel/cpu/resctrl/core.c
+++ b/arch/x86/kernel/cpu/resctrl/core.c
@@ -66,9 +66,6 @@ struct rdt_hw_resource rdt_resources_all[] = {
 			.rid			= RDT_RESOURCE_L3,
 			.name			= "L3",
 			.cache_level		= 3,
-			.cache = {
-				.min_cbm_bits	= 1,
-			},
 			.domains		= domain_init(RDT_RESOURCE_L3),
 			.parse_ctrlval		= parse_cbm,
 			.format_str		= "%d=%0*x",
@@ -83,9 +80,6 @@ struct rdt_hw_resource rdt_resources_all[] = {
 			.rid			= RDT_RESOURCE_L2,
 			.name			= "L2",
 			.cache_level		= 2,
-			.cache = {
-				.min_cbm_bits	= 1,
-			},
 			.domains		= domain_init(RDT_RESOURCE_L2),
 			.parse_ctrlval		= parse_cbm,
 			.format_str		= "%d=%0*x",
@@ -877,6 +871,7 @@ static __init void rdt_init_res_defs_intel(void)
 			r->cache.arch_has_sparse_bitmaps = false;
 			r->cache.arch_has_empty_bitmaps = false;
 			r->cache.arch_has_per_cpu_cfg = false;
+			r->cache.min_cbm_bits = 1;
 		} else if (r->rid == RDT_RESOURCE_MBA) {
 			hw_res->msr_base = MSR_IA32_MBA_THRTL_BASE;
 			hw_res->msr_update = mba_wrmsr_intel;
@@ -897,6 +892,7 @@ static __init void rdt_init_res_defs_amd(void)
 			r->cache.arch_has_sparse_bitmaps = true;
 			r->cache.arch_has_empty_bitmaps = true;
 			r->cache.arch_has_per_cpu_cfg = true;
+			r->cache.min_cbm_bits = 0;
 		} else if (r->rid == RDT_RESOURCE_MBA) {
 			hw_res->msr_base = MSR_IA32_MBA_BW_BASE;
 			hw_res->msr_update = mba_wrmsr_amd;
Re: [PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
Posted by James Morse 3 years, 7 months ago 
Hi Babu,

On 07/09/2022 18:59, Babu Moger wrote:
> AMD systems support zero CBM (capacity bit mask) for L3 allocation.
> That is reflected in rdt_init_res_defs_amd() by:
> 
> 	r->cache.arch_has_empty_bitmaps = true;
> 
> However given the unified code in cbm_validate(), checking for:
> 	val == 0 && !arch_has_empty_bitmaps
> 
> is not enough because of another check in cbm_validate():
> 
> 	if ((zero_bit - first_bit) < r->cache.min_cbm_bits)

Right, the Intel version had this, but the AMD didn't. I evidently only thought about this
the !arch_has_empty_bitmaps way round! Sorry about that.


> The default value of r->cache.min_cbm_bits = 1.
> 
> Leading to:
> 
> 	$ cd /sys/fs/resctrl
> 	$ mkdir foo
> 	$ cd foo
> 	$ echo L3:0=0 > schemata
>           -bash: echo: write error: Invalid argument
> 	$ cat /sys/fs/resctrl/info/last_cmd_status
> 	  Need at least 1 bits in the mask
> 
> Fix the issue by initializing the min_cbm_bits to 0 for AMD. Also,
> remove the default setting of min_cbm_bits and initialize it separately.
> 
> After the fix
> 	$ cd /sys/fs/resctrl
> 	$ mkdir foo
> 	$ cd foo
> 	$ echo L3:0=0 > schemata
> 	$ cat /sys/fs/resctrl/info/last_cmd_status
> 	  ok
> 
> Link: https://lore.kernel.org/lkml/20220517001234.3137157-1-eranian@google.com/
> Fixes: 316e7f901f5a ("x86/resctrl: Add struct rdt_cache::arch_has_{sparse, empty}_bitmaps")

> Signed-off-by: Stephane Eranian <eranian@google.com>
> Signed-off-by: Babu Moger <babu.moger@amd.com>

Er, who is the author if this patch? If you are reposting Stephane's patch then there
needs to be a 'From: ' at the top of the email so that git preserves the ownership. You
may need some incantation of "git commit --amend --author=" to fix this in your tree.

As its a fix, have you posted this separately? Mixing fixes and new-code makes it hard for
the maintainer to spot what needs to be taken for the next -rc.


> diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
> index bb1c3f5f60c8..a5c51a14fbce 100644
> --- a/arch/x86/kernel/cpu/resctrl/core.c
> +++ b/arch/x86/kernel/cpu/resctrl/core.c
> @@ -66,9 +66,6 @@ struct rdt_hw_resource rdt_resources_all[] = {
>  			.rid			= RDT_RESOURCE_L3,
>  			.name			= "L3",
>  			.cache_level		= 3,
> -			.cache = {
> -				.min_cbm_bits	= 1,
> -			},
>  			.domains		= domain_init(RDT_RESOURCE_L3),
>  			.parse_ctrlval		= parse_cbm,
>  			.format_str		= "%d=%0*x",
> @@ -83,9 +80,6 @@ struct rdt_hw_resource rdt_resources_all[] = {
>  			.rid			= RDT_RESOURCE_L2,
>  			.name			= "L2",
>  			.cache_level		= 2,
> -			.cache = {
> -				.min_cbm_bits	= 1,
> -			},
>  			.domains		= domain_init(RDT_RESOURCE_L2),
>  			.parse_ctrlval		= parse_cbm,
>  			.format_str		= "%d=%0*x",
> @@ -877,6 +871,7 @@ static __init void rdt_init_res_defs_intel(void)
>  			r->cache.arch_has_sparse_bitmaps = false;
>  			r->cache.arch_has_empty_bitmaps = false;
>  			r->cache.arch_has_per_cpu_cfg = false;
> +			r->cache.min_cbm_bits = 1;
>  		} else if (r->rid == RDT_RESOURCE_MBA) {
>  			hw_res->msr_base = MSR_IA32_MBA_THRTL_BASE;
>  			hw_res->msr_update = mba_wrmsr_intel;
> @@ -897,6 +892,7 @@ static __init void rdt_init_res_defs_amd(void)
>  			r->cache.arch_has_sparse_bitmaps = true;
>  			r->cache.arch_has_empty_bitmaps = true;
>  			r->cache.arch_has_per_cpu_cfg = true;
> +			r->cache.min_cbm_bits = 0;
>  		} else if (r->rid == RDT_RESOURCE_MBA) {
>  			hw_res->msr_base = MSR_IA32_MBA_BW_BASE;
>  			hw_res->msr_update = mba_wrmsr_amd;
> 
> 


Reviewed-by: James Morse <james.morse@arm.com>

Thanks,

James
Re: [PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
Posted by Moger, Babu 3 years, 7 months ago 
Hi James,

On 9/9/22 12:00, James Morse wrote:
> Hi Babu,
>
> On 07/09/2022 18:59, Babu Moger wrote:
>> AMD systems support zero CBM (capacity bit mask) for L3 allocation.
>> That is reflected in rdt_init_res_defs_amd() by:
>>
>> 	r->cache.arch_has_empty_bitmaps = true;
>>
>> However given the unified code in cbm_validate(), checking for:
>> 	val == 0 && !arch_has_empty_bitmaps
>>
>> is not enough because of another check in cbm_validate():
>>
>> 	if ((zero_bit - first_bit) < r->cache.min_cbm_bits)
> Right, the Intel version had this, but the AMD didn't. I evidently only thought about this
> the !arch_has_empty_bitmaps way round! Sorry about that.

>
>
>> The default value of r->cache.min_cbm_bits = 1.
>>
>> Leading to:
>>
>> 	$ cd /sys/fs/resctrl
>> 	$ mkdir foo
>> 	$ cd foo
>> 	$ echo L3:0=0 > schemata
>>           -bash: echo: write error: Invalid argument
>> 	$ cat /sys/fs/resctrl/info/last_cmd_status
>> 	  Need at least 1 bits in the mask
>>
>> Fix the issue by initializing the min_cbm_bits to 0 for AMD. Also,
>> remove the default setting of min_cbm_bits and initialize it separately.
>>
>> After the fix
>> 	$ cd /sys/fs/resctrl
>> 	$ mkdir foo
>> 	$ cd foo
>> 	$ echo L3:0=0 > schemata
>> 	$ cat /sys/fs/resctrl/info/last_cmd_status
>> 	  ok
>>
>> Link: https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kernel.org%2Flkml%2F20220517001234.3137157-1-eranian%40google.com%2F&amp;data=05%7C01%7Cbabu.moger%40amd.com%7Cc5b955e9726344c550f008da9284dedd%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C637983396695085653%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=wIVM5%2BrCwfWDpIrLkDtycgoCd4PWMC3D8y%2FAjshIW%2FQ%3D&amp;reserved=0
>> Fixes: 316e7f901f5a ("x86/resctrl: Add struct rdt_cache::arch_has_{sparse, empty}_bitmaps")
>> Signed-off-by: Stephane Eranian <eranian@google.com>
>> Signed-off-by: Babu Moger <babu.moger@amd.com>
> Er, who is the author if this patch? If you are reposting Stephane's patch then there
> needs to be a 'From: ' at the top of the email so that git preserves the ownership. You

I can add Stephane's name in "From" if it is right thing to do. But this
patch was modified from the original version Stephane posted.

https://lore.kernel.org/lkml/20220517001234.3137157-1-eranian@google.com/


> may need some incantation of "git commit --amend --author=" to fix this in your tree.
>
> As its a fix, have you posted this separately? Mixing fixes and new-code makes it hard for
> the maintainer to spot what needs to be taken for the next -rc.

ok. I can send this separate in next version.


>
>
>> diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
>> index bb1c3f5f60c8..a5c51a14fbce 100644
>> --- a/arch/x86/kernel/cpu/resctrl/core.c
>> +++ b/arch/x86/kernel/cpu/resctrl/core.c
>> @@ -66,9 +66,6 @@ struct rdt_hw_resource rdt_resources_all[] = {
>>  			.rid			= RDT_RESOURCE_L3,
>>  			.name			= "L3",
>>  			.cache_level		= 3,
>> -			.cache = {
>> -				.min_cbm_bits	= 1,
>> -			},
>>  			.domains		= domain_init(RDT_RESOURCE_L3),
>>  			.parse_ctrlval		= parse_cbm,
>>  			.format_str		= "%d=%0*x",
>> @@ -83,9 +80,6 @@ struct rdt_hw_resource rdt_resources_all[] = {
>>  			.rid			= RDT_RESOURCE_L2,
>>  			.name			= "L2",
>>  			.cache_level		= 2,
>> -			.cache = {
>> -				.min_cbm_bits	= 1,
>> -			},
>>  			.domains		= domain_init(RDT_RESOURCE_L2),
>>  			.parse_ctrlval		= parse_cbm,
>>  			.format_str		= "%d=%0*x",
>> @@ -877,6 +871,7 @@ static __init void rdt_init_res_defs_intel(void)
>>  			r->cache.arch_has_sparse_bitmaps = false;
>>  			r->cache.arch_has_empty_bitmaps = false;
>>  			r->cache.arch_has_per_cpu_cfg = false;
>> +			r->cache.min_cbm_bits = 1;
>>  		} else if (r->rid == RDT_RESOURCE_MBA) {
>>  			hw_res->msr_base = MSR_IA32_MBA_THRTL_BASE;
>>  			hw_res->msr_update = mba_wrmsr_intel;
>> @@ -897,6 +892,7 @@ static __init void rdt_init_res_defs_amd(void)
>>  			r->cache.arch_has_sparse_bitmaps = true;
>>  			r->cache.arch_has_empty_bitmaps = true;
>>  			r->cache.arch_has_per_cpu_cfg = true;
>> +			r->cache.min_cbm_bits = 0;
>>  		} else if (r->rid == RDT_RESOURCE_MBA) {
>>  			hw_res->msr_base = MSR_IA32_MBA_BW_BASE;
>>  			hw_res->msr_update = mba_wrmsr_amd;
>>
>>
>
> Reviewed-by: James Morse <james.morse@arm.com>


Thank You.

Babu
Re: [PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
Posted by Reinette Chatre 3 years, 6 months ago 
Hi Babu,

On 9/12/2022 7:54 AM, Moger, Babu wrote:
> On 9/9/22 12:00, James Morse wrote:
>> On 07/09/2022 18:59, Babu Moger wrote:


>>> Fixes: 316e7f901f5a ("x86/resctrl: Add struct rdt_cache::arch_has_{sparse, empty}_bitmaps")
>>> Signed-off-by: Stephane Eranian <eranian@google.com>
>>> Signed-off-by: Babu Moger <babu.moger@amd.com>
>> Er, who is the author if this patch? If you are reposting Stephane's patch then there
>> needs to be a 'From: ' at the top of the email so that git preserves the ownership. You
> 
> I can add Stephane's name in "From" if it is right thing to do. But this
> patch was modified from the original version Stephane posted.
> 
> https://lore.kernel.org/lkml/20220517001234.3137157-1-eranian@google.com/

True, but also please consider what Stephane posted originally:
https://lore.kernel.org/lkml/20220516055055.2734840-1-eranian@google.com/

A "Co-developed-by" may help with attribution: 

Co-developed-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Babu Moger <babu.moger@amd.com>

Reinette
RE: [PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
Posted by Moger, Babu 3 years, 6 months ago 
[AMD Official Use Only - General]



> -----Original Message-----
> From: Reinette Chatre <reinette.chatre@intel.com>
> Sent: Friday, September 16, 2022 10:53 AM
> To: Moger, Babu <Babu.Moger@amd.com>; James Morse
> <james.morse@arm.com>; corbet@lwn.net; tglx@linutronix.de;
> mingo@redhat.com; bp@alien8.de
> Cc: fenghua.yu@intel.com; dave.hansen@linux.intel.com; x86@kernel.org;
> hpa@zytor.com; paulmck@kernel.org; akpm@linux-foundation.org;
> quic_neeraju@quicinc.com; rdunlap@infradead.org;
> damien.lemoal@opensource.wdc.com; songmuchun@bytedance.com;
> peterz@infradead.org; jpoimboe@kernel.org; pbonzini@redhat.com;
> chang.seok.bae@intel.com; pawan.kumar.gupta@linux.intel.com;
> jmattson@google.com; daniel.sneddon@linux.intel.com; Das1, Sandipan
> <Sandipan.Das@amd.com>; tony.luck@intel.com; linux-doc@vger.kernel.org;
> linux-kernel@vger.kernel.org; bagasdotme@gmail.com; eranian@google.com
> Subject: Re: [PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
> 
> Hi Babu,
> 
> On 9/12/2022 7:54 AM, Moger, Babu wrote:
> > On 9/9/22 12:00, James Morse wrote:
> >> On 07/09/2022 18:59, Babu Moger wrote:
> 
> 
> >>> Fixes: 316e7f901f5a ("x86/resctrl: Add struct
> >>> rdt_cache::arch_has_{sparse, empty}_bitmaps")
> >>> Signed-off-by: Stephane Eranian <eranian@google.com>
> >>> Signed-off-by: Babu Moger <babu.moger@amd.com>
> >> Er, who is the author if this patch? If you are reposting Stephane's
> >> patch then there needs to be a 'From: ' at the top of the email so
> >> that git preserves the ownership. You
> >
> > I can add Stephane's name in "From" if it is right thing to do. But
> > this patch was modified from the original version Stephane posted.
> >
> > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore
> > .kernel.org%2Flkml%2F20220517001234.3137157-1-
> eranian%40google.com%2F&
> >
> amp;data=05%7C01%7Cbabu.moger%40amd.com%7C69a28ad3fcfe444404a50
> 8da97fb
> >
> 82ee%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C6379894037788
> 16201%7
> >
> CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI
> 6Ik1
> >
> haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=OnLKz6cBiypVvv1x
> 8PSv1JUz
> > ugilG1Gpwgkcz55ydqI%3D&amp;reserved=0
> 
> True, but also please consider what Stephane posted originally:
> https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore.kern
> el.org%2Flkml%2F20220516055055.2734840-1-
> eranian%40google.com%2F&amp;data=05%7C01%7Cbabu.moger%40amd.com
> %7C69a28ad3fcfe444404a508da97fb82ee%7C3dd8961fe4884e608e11a82d994
> e183d%7C0%7C0%7C637989403778816201%7CUnknown%7CTWFpbGZsb3d8ey
> JWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7
> C3000%7C%7C%7C&amp;sdata=2mTjDYC9B%2Fr6HR6SlwToWXewsWub2rfPQp
> c9JIkETus%3D&amp;reserved=0
> 
> A "Co-developed-by" may help with attribution:
> 
> Co-developed-by: Stephane Eranian <eranian@google.com>
> Signed-off-by: Stephane Eranian <eranian@google.com>
> Signed-off-by: Babu Moger <babu.moger@amd.com>
> 
Yes, Sounds good to me.
Thanks
Babu
Re: [PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
Posted by Reinette Chatre 3 years, 6 months ago 
Hi Babu,

On 9/7/2022 10:59 AM, Babu Moger wrote:
> AMD systems support zero CBM (capacity bit mask) for L3 allocation.

Above mentions "for L3 allocation", but the change impacts
L3 as well as L2 allocation. Perhaps just
"cache allocation"?

> That is reflected in rdt_init_res_defs_amd() by:
> 
> 	r->cache.arch_has_empty_bitmaps = true;
> 
> However given the unified code in cbm_validate(), checking for:
> 	val == 0 && !arch_has_empty_bitmaps
> 
> is not enough because of another check in cbm_validate():
> 
> 	if ((zero_bit - first_bit) < r->cache.min_cbm_bits)
> 
> The default value of r->cache.min_cbm_bits = 1.
> 
> Leading to:
> 
> 	$ cd /sys/fs/resctrl
> 	$ mkdir foo
> 	$ cd foo
> 	$ echo L3:0=0 > schemata
>           -bash: echo: write error: Invalid argument
> 	$ cat /sys/fs/resctrl/info/last_cmd_status
> 	  Need at least 1 bits in the mask
> 
> Fix the issue by initializing the min_cbm_bits to 0 for AMD. Also,
> remove the default setting of min_cbm_bits and initialize it separately.
> 
> After the fix
> 	$ cd /sys/fs/resctrl
> 	$ mkdir foo
> 	$ cd foo
> 	$ echo L3:0=0 > schemata
> 	$ cat /sys/fs/resctrl/info/last_cmd_status
> 	  ok
> 
> Link: https://lore.kernel.org/lkml/20220517001234.3137157-1-eranian@google.com/
> Fixes: 316e7f901f5a ("x86/resctrl: Add struct rdt_cache::arch_has_{sparse, empty}_bitmaps")
> Signed-off-by: Stephane Eranian <eranian@google.com>
> Signed-off-by: Babu Moger <babu.moger@amd.com>
> Reviewed-by: Ingo Molnar <mingo@kernel.org>

(apart from the changelog nitpick)

Thank you for clarifying the way forward for this fix.

Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>

Reinette
RE: [PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
Posted by Moger, Babu 3 years, 6 months ago 
[AMD Official Use Only - General]



> -----Original Message-----
> From: Reinette Chatre <reinette.chatre@intel.com>
> Sent: Friday, September 16, 2022 10:53 AM
> To: Moger, Babu <Babu.Moger@amd.com>; corbet@lwn.net;
> tglx@linutronix.de; mingo@redhat.com; bp@alien8.de
> Cc: fenghua.yu@intel.com; dave.hansen@linux.intel.com; x86@kernel.org;
> hpa@zytor.com; paulmck@kernel.org; akpm@linux-foundation.org;
> quic_neeraju@quicinc.com; rdunlap@infradead.org;
> damien.lemoal@opensource.wdc.com; songmuchun@bytedance.com;
> peterz@infradead.org; jpoimboe@kernel.org; pbonzini@redhat.com;
> chang.seok.bae@intel.com; pawan.kumar.gupta@linux.intel.com;
> jmattson@google.com; daniel.sneddon@linux.intel.com; Das1, Sandipan
> <Sandipan.Das@amd.com>; tony.luck@intel.com; james.morse@arm.com;
> linux-doc@vger.kernel.org; linux-kernel@vger.kernel.org;
> bagasdotme@gmail.com; eranian@google.com
> Subject: Re: [PATCH v4 01/13] x86/resctrl: Fix min_cbm_bits for AMD
> 
> Hi Babu,
> 
> On 9/7/2022 10:59 AM, Babu Moger wrote:
> > AMD systems support zero CBM (capacity bit mask) for L3 allocation.
> 
> Above mentions "for L3 allocation", but the change impacts
> L3 as well as L2 allocation. Perhaps just "cache allocation"?
> 
> > That is reflected in rdt_init_res_defs_amd() by:
> >
> > 	r->cache.arch_has_empty_bitmaps = true;
> >
> > However given the unified code in cbm_validate(), checking for:
> > 	val == 0 && !arch_has_empty_bitmaps
> >
> > is not enough because of another check in cbm_validate():
> >
> > 	if ((zero_bit - first_bit) < r->cache.min_cbm_bits)
> >
> > The default value of r->cache.min_cbm_bits = 1.
> >
> > Leading to:
> >
> > 	$ cd /sys/fs/resctrl
> > 	$ mkdir foo
> > 	$ cd foo
> > 	$ echo L3:0=0 > schemata
> >           -bash: echo: write error: Invalid argument
> > 	$ cat /sys/fs/resctrl/info/last_cmd_status
> > 	  Need at least 1 bits in the mask
> >
> > Fix the issue by initializing the min_cbm_bits to 0 for AMD. Also,
> > remove the default setting of min_cbm_bits and initialize it separately.
> >
> > After the fix
> > 	$ cd /sys/fs/resctrl
> > 	$ mkdir foo
> > 	$ cd foo
> > 	$ echo L3:0=0 > schemata
> > 	$ cat /sys/fs/resctrl/info/last_cmd_status
> > 	  ok
> >
> > Link:
> > https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Flore
> > .kernel.org%2Flkml%2F20220517001234.3137157-1-
> eranian%40google.com%2F&
> >
> amp;data=05%7C01%7Cbabu.moger%40amd.com%7Cfdccc20e6a234fb3872a0
> 8da97fb
> >
> 94fc%7C3dd8961fe4884e608e11a82d994e183d%7C0%7C0%7C6379894040498
> 44076%7
> >
> CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI
> 6Ik1
> >
> haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&amp;sdata=DIpkBAkI7lAjj5QQ4
> 5VahssT
> > dWGj%2FcUwGJDiHXNYzz8%3D&amp;reserved=0
> > Fixes: 316e7f901f5a ("x86/resctrl: Add struct
> > rdt_cache::arch_has_{sparse, empty}_bitmaps")
> > Signed-off-by: Stephane Eranian <eranian@google.com>
> > Signed-off-by: Babu Moger <babu.moger@amd.com>
> > Reviewed-by: Ingo Molnar <mingo@kernel.org>
> 
> (apart from the changelog nitpick)
> 
> Thank you for clarifying the way forward for this fix.
> 
> Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>

Thank you
Babu Moger

Patchew 2.1-devel-ea86937

© 2016 - 2026 Red Hat, Inc.