From: Babu Moger <babu.moger@amd.com>
To: <corbet@lwn.net>, <reinette.chatre@intel.com>,
<tglx@linutronix.de>, <mingo@redhat.com>, <bp@alien8.de>,
<dave.hansen@linux.intel.com>, <tony.luck@intel.com>,
<peternewman@google.com>
Cc: <fenghua.yu@intel.com>, <x86@kernel.org>, <hpa@zytor.com>,
<paulmck@kernel.org>, <akpm@linux-foundation.org>,
<thuth@redhat.com>, <rostedt@goodmis.org>,
<xiongwei.song@windriver.com>,
<pawan.kumar.gupta@linux.intel.com>,
<daniel.sneddon@linux.intel.com>, <jpoimboe@kernel.org>,
<perry.yuan@amd.com>, <andipan.das@amd.com>,
<kai.huang@intel.com>, <xiaoyao.li@intel.com>,
<seanjc@google.com>, <babu.moger@amd.com>, <xin3.li@intel.com>,
<andrew.cooper3@citrix.com>, <ebiggers@google.com>,
<mario.limonciello@amd.com>, <james.morse@arm.com>,
<tan.shaopeng@fujitsu.com>, <linux-doc@vger.kernel.org>,
<linux-kernel@vger.kernel.org>, <maciej.wieczor-retman@intel.com>,
<eranian@google.com>
Subject: [PATCH v10 16/24] x86/resctrl: Add interface to the assign counter
Date: Thu, 12 Dec 2024 14:15:19 -0600 [thread overview]
Message-ID: <a72e23d8fe43038cd319403ed68b657fb36e23df.1734034524.git.babu.moger@amd.com> (raw)
In-Reply-To: <cover.1734034524.git.babu.moger@amd.com>
The mbm_cntr_assign mode offers several counters that can be assigned
to an RMID, event pair and monitor the bandwidth as long as it is
assigned.
Counters are managed at the domain level. Introduce the interface to
allocate/free/assign the counters.
If the user requests assignments across all domains, some domains may
fail if they run out of counters. Ensure assignments continue in other
domains wherever possible.
Signed-off-by: Babu Moger <babu.moger@amd.com>
---
v10: Patch changed completely.
Counters are managed at the domain based on the discussion.
https://lore.kernel.org/lkml/CALPaoCj+zWq1vkHVbXYP0znJbe6Ke3PXPWjtri5AFgD9cQDCUg@mail.gmail.com/
Reset non-architectural MBM state.
Commit message update.
v9: Introduced new function resctrl_config_cntr to assign the counter, update
the bitmap and reset the architectural state.
Taken care of error handling(freeing the counter) when assignment fails.
Moved mbm_cntr_assigned_to_domain here as it used in this patch.
Minor text changes.
v8: Renamed rdtgroup_assign_cntr() to rdtgroup_assign_cntr_event().
Added the code to return the error if rdtgroup_assign_cntr_event fails.
Moved definition of MBM_EVENT_ARRAY_INDEX to resctrl/internal.h.
Updated typo in the comments.
v7: New patch. Moved all the FS code here.
Merged rdtgroup_assign_cntr and rdtgroup_alloc_cntr.
Adde new #define MBM_EVENT_ARRAY_INDEX.
---
arch/x86/kernel/cpu/resctrl/internal.h | 5 +-
arch/x86/kernel/cpu/resctrl/monitor.c | 4 +-
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 110 +++++++++++++++++++++++++
3 files changed, 116 insertions(+), 3 deletions(-)
diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
index 849bcfe4ea5b..70d2577fc377 100644
--- a/arch/x86/kernel/cpu/resctrl/internal.h
+++ b/arch/x86/kernel/cpu/resctrl/internal.h
@@ -704,5 +704,8 @@ unsigned int mon_event_config_index_get(u32 evtid);
int resctrl_arch_config_cntr(struct rdt_resource *r, struct rdt_mon_domain *d,
enum resctrl_event_id evtid, u32 rmid, u32 closid,
u32 cntr_id, bool assign);
-
+int rdtgroup_assign_cntr_event(struct rdt_resource *r, struct rdtgroup *rdtgrp,
+ struct rdt_mon_domain *d, enum resctrl_event_id evtid);
+struct mbm_state *get_mbm_state(struct rdt_mon_domain *d, u32 closid,
+ u32 rmid, enum resctrl_event_id evtid);
#endif /* _ASM_X86_RESCTRL_INTERNAL_H */
diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c
index f857af361af1..8823cd97ff1f 100644
--- a/arch/x86/kernel/cpu/resctrl/monitor.c
+++ b/arch/x86/kernel/cpu/resctrl/monitor.c
@@ -575,8 +575,8 @@ void free_rmid(u32 closid, u32 rmid)
list_add_tail(&entry->list, &rmid_free_lru);
}
-static struct mbm_state *get_mbm_state(struct rdt_mon_domain *d, u32 closid,
- u32 rmid, enum resctrl_event_id evtid)
+struct mbm_state *get_mbm_state(struct rdt_mon_domain *d, u32 closid,
+ u32 rmid, enum resctrl_event_id evtid)
{
u32 idx = resctrl_arch_rmid_idx_encode(closid, rmid);
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index e895d2415f22..1c8694a68cf4 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -1927,6 +1927,116 @@ int resctrl_arch_config_cntr(struct rdt_resource *r, struct rdt_mon_domain *d,
return 0;
}
+/*
+ * Configure the counter for the event, RMID pair for the domain.
+ */
+static int resctrl_config_cntr(struct rdt_resource *r, struct rdt_mon_domain *d,
+ enum resctrl_event_id evtid, u32 rmid, u32 closid,
+ u32 cntr_id, bool assign)
+{
+ struct mbm_state *m;
+ int ret;
+
+ ret = resctrl_arch_config_cntr(r, d, evtid, rmid, closid, cntr_id, assign);
+ if (ret)
+ return ret;
+
+ m = get_mbm_state(d, closid, rmid, evtid);
+ if (m)
+ memset(m, 0, sizeof(struct mbm_state));
+
+ return ret;
+}
+
+static bool mbm_cntr_assigned(struct rdt_resource *r, struct rdt_mon_domain *d,
+ struct rdtgroup *rdtgrp, enum resctrl_event_id evtid)
+{
+ int cntr_id;
+
+ for (cntr_id = 0; cntr_id < r->mon.num_mbm_cntrs; cntr_id++) {
+ if (d->cntr_cfg[cntr_id].rdtgrp == rdtgrp &&
+ d->cntr_cfg[cntr_id].evtid == evtid)
+ return true;
+ }
+
+ return false;
+}
+
+static int mbm_cntr_alloc(struct rdt_resource *r, struct rdt_mon_domain *d,
+ struct rdtgroup *rdtgrp, enum resctrl_event_id evtid)
+{
+ int cntr_id;
+
+ for (cntr_id = 0; cntr_id < r->mon.num_mbm_cntrs; cntr_id++) {
+ if (!d->cntr_cfg[cntr_id].rdtgrp) {
+ d->cntr_cfg[cntr_id].rdtgrp = rdtgrp;
+ d->cntr_cfg[cntr_id].evtid = evtid;
+ return cntr_id;
+ }
+ }
+
+ return -EINVAL;
+}
+
+static void mbm_cntr_free(struct rdt_resource *r, struct rdt_mon_domain *d,
+ struct rdtgroup *rdtgrp, enum resctrl_event_id evtid)
+{
+ int cntr_id;
+
+ for (cntr_id = 0; cntr_id < r->mon.num_mbm_cntrs; cntr_id++) {
+ if (d->cntr_cfg[cntr_id].rdtgrp == rdtgrp &&
+ d->cntr_cfg[cntr_id].evtid == evtid)
+ memset(&d->cntr_cfg[cntr_id], 0, sizeof(struct mbm_cntr_cfg));
+ }
+}
+
+/*
+ * Assign a hardware counter to event @evtid of group @rdtgrp.
+ * Counter will be assigned to all the domains if rdt_mon_domain is NULL
+ * else the counter will be assigned to specific domain.
+ */
+int rdtgroup_assign_cntr_event(struct rdt_resource *r, struct rdtgroup *rdtgrp,
+ struct rdt_mon_domain *d, enum resctrl_event_id evtid)
+{
+ int cntr_id, ret = 0;
+
+ if (!d) {
+ list_for_each_entry(d, &r->mon_domains, hdr.list) {
+ if (mbm_cntr_assigned(r, d, rdtgrp, evtid))
+ continue;
+
+ cntr_id = mbm_cntr_alloc(r, d, rdtgrp, evtid);
+ if (cntr_id < 0) {
+ rdt_last_cmd_puts("Domain Out of MBM assignable counters\n");
+ continue;
+ }
+
+ ret = resctrl_config_cntr(r, d, evtid, rdtgrp->mon.rmid,
+ rdtgrp->closid, cntr_id, true);
+ if (ret)
+ goto out_done_assign;
+ }
+ } else {
+ if (mbm_cntr_assigned(r, d, rdtgrp, evtid))
+ goto out_done_assign;
+
+ cntr_id = mbm_cntr_alloc(r, d, rdtgrp, evtid);
+ if (cntr_id < 0) {
+ rdt_last_cmd_puts("Domain Out of MBM assignable counters\n");
+ goto out_done_assign;
+ }
+
+ ret = resctrl_config_cntr(r, d, evtid, rdtgrp->mon.rmid,
+ rdtgrp->closid, cntr_id, true);
+ }
+
+out_done_assign:
+ if (ret)
+ mbm_cntr_free(r, d, rdtgrp, evtid);
+
+ return ret;
+}
+
/* rdtgroup information files for one cache resource. */
static struct rftype res_common_files[] = {
{
--
2.34.1
next prev parent reply other threads:[~2024-12-12 20:18 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-12-12 20:15 [PATCH v10 00/24] x86/resctrl : Support AMD Assignable Bandwidth Monitoring Counters (ABMC) Babu Moger
2024-12-12 20:15 ` [PATCH v10 01/24] x86/resctrl: Add __init attribute to functions called from resctrl_late_init() Babu Moger
2024-12-12 20:15 ` [PATCH v10 02/24] x86/cpufeatures: Add support for Assignable Bandwidth Monitoring Counters (ABMC) Babu Moger
2024-12-12 20:15 ` [PATCH v10 03/24] x86/resctrl: Add ABMC feature in the command line options Babu Moger
2024-12-12 20:15 ` [PATCH v10 04/24] x86/resctrl: Consolidate monitoring related data from rdt_resource Babu Moger
2024-12-12 20:15 ` [PATCH v10 05/24] x86/resctrl: Detect Assignable Bandwidth Monitoring feature details Babu Moger
2024-12-12 20:15 ` [PATCH v10 06/24] x86/resctrl: Introduce resctrl_file_fflags_init() to initialize fflags Babu Moger
2024-12-12 20:15 ` [PATCH v10 07/24] x86/resctrl: Add support to enable/disable AMD ABMC feature Babu Moger
2024-12-19 21:48 ` Reinette Chatre
2024-12-20 15:14 ` Moger, Babu
2024-12-20 17:16 ` Reinette Chatre
2024-12-12 20:15 ` [PATCH v10 08/24] x86/resctrl: Introduce the interface to display monitor mode Babu Moger
2024-12-19 21:59 ` Reinette Chatre
2024-12-20 15:31 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 09/24] x86/resctrl: Introduce interface to display number of monitoring counters Babu Moger
2024-12-19 22:03 ` Reinette Chatre
2024-12-20 15:41 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 10/24] x86/resctrl: Introduce mbm_total_cfg and mbm_local_cfg in struct rdt_hw_mon_domain Babu Moger
2024-12-12 20:15 ` [PATCH v10 11/24] x86/resctrl: Remove MSR reading of event configuration value Babu Moger
2024-12-19 22:12 ` Reinette Chatre
2024-12-20 16:09 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 12/24] x86/resctrl: Introduce cntr_cfg to track assignable counters at domain Babu Moger
2024-12-19 22:33 ` Reinette Chatre
2024-12-20 17:33 ` Moger, Babu
2024-12-20 20:58 ` Reinette Chatre
2024-12-12 20:15 ` [PATCH v10 13/24] x86/resctrl: Introduce interface to display number of free counters Babu Moger
2024-12-19 22:50 ` Reinette Chatre
2024-12-20 18:05 ` Moger, Babu
2024-12-20 18:32 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 14/24] x86/resctrl: Add data structures and definitions for ABMC assignment Babu Moger
2024-12-12 20:15 ` [PATCH v10 15/24] x86/resctrl: Implement resctrl_arch_config_cntr() to assign a counter with ABMC Babu Moger
2024-12-19 23:04 ` Reinette Chatre
2024-12-20 19:22 ` Moger, Babu
2024-12-20 21:41 ` Reinette Chatre
2024-12-20 22:28 ` Moger, Babu
2024-12-20 23:47 ` Reinette Chatre
2024-12-21 13:40 ` Moger, Babu
2024-12-12 20:15 ` Babu Moger [this message]
2024-12-12 23:37 ` [PATCH v10 16/24] x86/resctrl: Add interface to the assign counter Luck, Tony
2024-12-13 15:57 ` Moger, Babu
2024-12-13 16:24 ` Luck, Tony
2024-12-13 16:54 ` Moger, Babu
2024-12-18 22:01 ` Reinette Chatre
2024-12-19 19:45 ` Moger, Babu
2024-12-19 21:12 ` Reinette Chatre
2024-12-19 21:38 ` Moger, Babu
2024-12-19 21:45 ` Luck, Tony
2024-12-19 22:33 ` Moger, Babu
2024-12-19 23:22 ` Reinette Chatre
2024-12-20 20:34 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 17/24] x86/resctrl: Add the interface to unassign a counter Babu Moger
2024-12-19 23:32 ` Reinette Chatre
2024-12-20 21:38 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 18/24] x86/resctrl: Auto assign/unassign counters when mbm_cntr_assign is enabled Babu Moger
2024-12-19 23:39 ` Reinette Chatre
2024-12-21 13:45 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 19/24] x86/resctrl: Report "Unassigned" for MBM events in mbm_cntr_assign mode Babu Moger
2024-12-19 23:59 ` Reinette Chatre
2024-12-21 14:04 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 20/24] x86/resctrl: Introduce the interface to switch between monitor modes Babu Moger
2024-12-20 2:56 ` Reinette Chatre
2024-12-21 14:20 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 21/24] x86/resctrl: Configure mbm_cntr_assign mode if supported Babu Moger
2024-12-20 3:03 ` Reinette Chatre
2024-12-21 14:33 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 22/24] x86/resctrl: Update assignments on event configuration changes Babu Moger
2024-12-20 3:12 ` Reinette Chatre
2024-12-21 14:59 ` Moger, Babu
2024-12-23 16:20 ` Reinette Chatre
2025-01-13 20:03 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 23/24] x86/resctrl: Introduce interface to list assignment states of all the groups Babu Moger
2024-12-12 22:57 ` Luck, Tony
2024-12-13 15:23 ` Moger, Babu
2024-12-12 20:15 ` [PATCH v10 24/24] x86/resctrl: Introduce interface to modify assignment states of " Babu Moger
2024-12-20 3:23 ` Reinette Chatre
2024-12-21 15:28 ` Moger, Babu
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=a72e23d8fe43038cd319403ed68b657fb36e23df.1734034524.git.babu.moger@amd.com \
--to=babu.moger@amd.com \
--cc=akpm@linux-foundation.org \
--cc=andipan.das@amd.com \
--cc=andrew.cooper3@citrix.com \
--cc=bp@alien8.de \
--cc=corbet@lwn.net \
--cc=daniel.sneddon@linux.intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=ebiggers@google.com \
--cc=eranian@google.com \
--cc=fenghua.yu@intel.com \
--cc=hpa@zytor.com \
--cc=james.morse@arm.com \
--cc=jpoimboe@kernel.org \
--cc=kai.huang@intel.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=maciej.wieczor-retman@intel.com \
--cc=mario.limonciello@amd.com \
--cc=mingo@redhat.com \
--cc=paulmck@kernel.org \
--cc=pawan.kumar.gupta@linux.intel.com \
--cc=perry.yuan@amd.com \
--cc=peternewman@google.com \
--cc=reinette.chatre@intel.com \
--cc=rostedt@goodmis.org \
--cc=seanjc@google.com \
--cc=tan.shaopeng@fujitsu.com \
--cc=tglx@linutronix.de \
--cc=thuth@redhat.com \
--cc=tony.luck@intel.com \
--cc=x86@kernel.org \
--cc=xiaoyao.li@intel.com \
--cc=xin3.li@intel.com \
--cc=xiongwei.song@windriver.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
all inboxes | Powered by JetHome®