[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ec170d40-7390-4cc7-9128-d200df7405ee@intel.com>
Date: Thu, 19 Sep 2024 08:35:19 -0700
From: Reinette Chatre <reinette.chatre@...el.com>
To: <babu.moger@....com>, <corbet@....net>, <tglx@...utronix.de>,
<mingo@...hat.com>, <bp@...en8.de>, <dave.hansen@...ux.intel.com>,
<x86@...nel.org>
CC: <fenghua.yu@...el.com>, <hpa@...or.com>, <paulmck@...nel.org>,
<thuth@...hat.com>, <xiongwei.song@...driver.com>, <ardb@...nel.org>,
<pawan.kumar.gupta@...ux.intel.com>, <daniel.sneddon@...ux.intel.com>,
<sandipan.das@....com>, <kai.huang@...el.com>, <peterz@...radead.org>,
<kan.liang@...ux.intel.com>, <pbonzini@...hat.com>, <xin3.li@...el.com>,
<ebiggers@...gle.com>, <alexandre.chartre@...cle.com>, <perry.yuan@....com>,
<tan.shaopeng@...itsu.com>, <james.morse@....com>, <tony.luck@...el.com>,
<maciej.wieczor-retman@...el.com>, <linux-doc@...r.kernel.org>,
<linux-kernel@...r.kernel.org>, <peternewman@...gle.com>,
<eranian@...gle.com>
Subject: Re: [PATCH 5/7] x86/resctrl: Add interface to enable/disable SDCIAE
Hi Babu,
On 9/18/24 1:10 PM, Moger, Babu wrote:
> On 9/13/24 15:51, Reinette Chatre wrote:
>> On 8/16/24 9:16 AM, Babu Moger wrote:
...
>>> + if (enable) {
>>> + ret = closid_alloc_sdciae(r);
>>> + if (ret < 0) {
>>> + rdt_last_cmd_puts("SDCIAE CLOSID is not available\n");
>>> + goto out_sdciae;
>>> + }
>>> + } else {
>>> + sdciae_closid = get_sdciae_closid(r);
>>> + closid_free(sdciae_closid);
>>> + }
>>
>>
>>> +
>>> + ret = resctrl_arch_set_sdciae_enabled(RDT_RESOURCE_L3, enable);
>>
>> I assume that once SDCIAE is enabled the I/O traffic will start flowing to
>> whatever
>> was the last CBM of the max CLOSID? Is this intended or should there be
>> some default
>> CBM that this feature should start with?
>
> It will start with whatever the last CBM for max CLOSID.
This seems arbitrary based on whatever allocation the previous resource group
using the CLOSID has. When a new resource group is created resctrl ensures
that it is created with all usable allocations, see rdtgroup_init_cat().
Letting cache injection start with whatever allocation remnant programmed
in a register does not seem ideal. What if, for example, after that resource
group was removed, a new exclusive resource group was created that overlaps
with that allocation?
Reinette
Powered by blists - more mailing lists