x86/resctrl: Move the resctrl filesystem code to /fs/resctrl
Subsystems (3)
James Morse
reviewerv3 cover x86/resctrl: Move the resctrl filesystem code to /fs/resctrl
Changes since v2?
Patches 3+13 have been merged, then split into six patches that also bring the format string and a few other parameters into the struct resctrl_schema. This makes for a cleaner arch/fs split.
Dave's checkpatch fixes to existing code have been broken out into a separate series to be posted shortly. The asm->linux changes got merged into the patch that makes this possible.
~
This is the final series that allows other architectures to implement resctrl. The final patch to move the code has been ommited, but can be generated using the python script at the end of the series.
The final move is a bit of a monster. I don't expect that to get merged as part of this series - we should wait for it to make less impact on other series.
Otherwise this series renames functions and moves code around. With the exception of invalid configurations for the configurable-events, there should be no changes in behaviour caused by this series.
The driving pattern is to make things like struct rdtgroup private to resctrl. Features like pseudo-lock aren't going to work on arm64, the ability to disable it at compile time is added.
After this, I can start posting the MPAM driver to make use of resctrl on arm64. (What's MPAM? See the cover letter of the first series. [1])
This series is based on v6.10-rc1 and can be retrieved from:
https://git.kernel.org/pub/scm/linux/kernel/git/morse/linux.git mpam/move_to_fs/v3
As ever - bugs welcome,
Thanks,
James
[v2] https://lore.kernel.org/r/[email protected]
[v1] https://lore.kernel.org/r/[email protected]
[1] https://lore.kernel.org/lkml/[email protected]/
James Morse (38):
x86/resctrl: Fix allocation of cleanest CLOSID on platforms with no
monitors
x86/resctrl: Add a helper to avoid reaching into the arch code
resource list
x86/resctrl: Add a schema format enum and use this for fflags x86/resctrl: Use schema type to determine how to parse schema values x86/resctrl: Use schema type to determine the schema format string x86/resctrl: Move data_width to be a schema property x86/resctrl: Add max_bw to struct resctrl_membw x86/resctrl: Generate default_ctrl instead of sharing it x86/resctrl: Add helper for setting CPU default properties x86/resctrl: Remove rdtgroup from update_cpu_closid_rmid() x86/resctrl: Export resctrl fs's init function x86/resctrl: Wrap resctrl_arch_find_domain() around rdt_find_domain() x86/resctrl: Move resctrl types to a separate header x86/resctrl: Add a resctrl helper to reset all the resources x86/resctrl: Move monitor exit work to a restrl exit call x86/resctrl: Move monitor init work to a resctrl init call x86/resctrl: Stop using the for_each_*_rdt_resource() walkers x86/resctrl: Export the is_mbm_*_enabled() helpers to asm/resctrl.h x86/resctrl: Add resctrl_arch_is_evt_configurable() to abstract BMEC x86/resctrl: Change mon_event_config_{read,write}() to be arch helpers x86/resctrl: Move mbm_cfg_mask to struct rdt_resource x86/resctrl: Add resctrl_arch_ prefix to pseudo lock functions x86/resctrl: Allow an architecture to disable pseudo lock x86/resctrl: Make prefetch_disable_bits belong to the arch code x86/resctrl: Make resctrl_arch_pseudo_lock_fn() take a plr x86/resctrl: Move thread_throttle_mode_init() to be managed by resctrl x86/resctrl: Move get_config_index() to a header x86/resctrl: Claim get_domain_from_cpu() for resctrl x86/resctrl: Describe resctrl's bitmap size assumptions x86/resctrl: Rename resctrl_sched_in() to begin with "resctrl_arch_" x86/resctrl: resctrl_exit() teardown resctrl but leave the mount point x86/resctrl: Drop __init/__exit on assorted symbols x86/resctrl: Move is_mba_sc() out of core.c
x86/resctrl: Add end-marker to the resctrl_event_id enum x86/resctrl: Remove a newline to avoid confusing the code move script fs/resctrl: Add boiler plate for external resctrl code x86/resctrl: Move the filesystem bits to headers visible to fs/resctrl x86/resctrl: Add python script to move resctrl code to /fs/resctrl
MAINTAINERS | 2 + arch/Kconfig | 8 + arch/x86/Kconfig | 12 +- arch/x86/include/asm/resctrl.h | 45 +- arch/x86/kernel/cpu/resctrl/Makefile | 5 +- arch/x86/kernel/cpu/resctrl/core.c | 158 ++--- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 67 +- arch/x86/kernel/cpu/resctrl/internal.h | 185 ++---- arch/x86/kernel/cpu/resctrl/monitor.c | 88 +-- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 65 +- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 288 +++++--- arch/x86/kernel/process_32.c | 2 +- arch/x86/kernel/process_64.c | 2 +- fs/Kconfig | 1 + fs/Makefile | 1 + fs/resctrl/Kconfig | 36 + fs/resctrl/Makefile | 3 + fs/resctrl/ctrlmondata.c | 0 fs/resctrl/internal.h | 0 fs/resctrl/monitor.c | 0 fs/resctrl/pseudo_lock.c | 0 fs/resctrl/rdtgroup.c | 0 fs/resctrl/trace.h | 0 include/linux/resctrl.h | 193 +++++- include/linux/resctrl_types.h | 59 ++ resctrl_copy_pasta.py | 766 ++++++++++++++++++++++ 26 files changed, 1546 insertions(+), 440 deletions(-) create mode 100644 fs/resctrl/Kconfig create mode 100644 fs/resctrl/Makefile create mode 100644 fs/resctrl/ctrlmondata.c create mode 100644 fs/resctrl/internal.h create mode 100644 fs/resctrl/monitor.c create mode 100644 fs/resctrl/pseudo_lock.c create mode 100644 fs/resctrl/rdtgroup.c create mode 100644 fs/resctrl/trace.h create mode 100644 include/linux/resctrl_types.h create mode 100644 resctrl_copy_pasta.py
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 1/38 x86/resctrl: Fix allocation of cleanest CLOSID on platforms with no monitors
commit 6eac36bb9eb0 ("x86/resctrl: Allocate the cleanest CLOSID by searching closid_num_dirty_rmid") added logic that causes resctrl to search for the CLOSID with the fewest dirty cache lines when creating a new control group, if requested by the arch code. This depends on the values read from the llc_occupancy counters. The logic is applicable to architectures where the CLOSID effectively forms part of the monitoring identifier and so do not allow complete freedom to choose an unused monitoring identifier for a given CLOSID.
This support missed that some platforms may not have these counters. This causes a NULL pointer dereference when creating a new control group as the array was not allocated by dom_data_init().
As this feature isn't necessary on platforms that don't have cache occupancy monitors, add this to the check that occurs when a new control group is allocated.
The existing code is not selected by any upstream platform, it makes no sense to backport this patch to stable.
---
Changes since v1:
* [Commit message only] Reword the first paragraph to make it clear that the issue being fixed wasn't directly associated with addition of a Kconfig option. (Actually, the option is not in Kconfig yet, and gets added later in this series.)
--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +2 -1
@@ -149,7 +149,8 @@ static int closid_alloc(void)lockdep_assert_held (&rdtgroup_mutex); -if (IS_ENABLED (CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID )) { +if (IS_ENABLED (CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID ) && +is_llc_occupancy_enabled ()) { cleanest_closid =resctrl_find_cleanest_closid ();if (cleanest_closid <0 )return cleanest_closid;
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 01/38] x86/resctrl: Fix allocation of cleanest CLOSID on platforms with no monitors
Hi James,
Quoted 2 lines · James Morse
commit
6eac36bb9eb0("x86/resctrl: Allocate the cleanest CLOSID by searchingclosid_num_dirty_rmid") added logic that causes resctrl to
There is a custom in x86 on how messages containing references to commits are formatted. For reference you can see how the recent fix 739c9765793e ("x86/resctrl: Don't try to free nonexistent RMIDs") was reformatted before merge. Applied to this patch, it may look something like:
Commit
6eac36bb9eb0 ("x86/resctrl: Allocate the cleanest CLOSID by searching closid_num_dirty_rmid")
added logic ...
Looking at, for example, commit e3ca96e479c9 ("x86/resctrl: Pass domain to target CPU"), it does seem ok to split the name of the commit.
Quoted 7 lines
search for the CLOSID with the fewest dirty cache lines when creating a new control group, if requested by the arch code. This depends on the values read from the
llc_occupancycounters. The logic is applicable to architectures where the CLOSID effectively forms part of the monitoring identifier and so do not allow complete freedom to choose an unused monitoring identifier for a given CLOSID.
Quoted 8 lines
This support missed that some platforms may not have these counters. This causes a NULL pointer dereference when creating a new control group as the array was not allocated by
dom_data_init().As this feature isn't necessary on platforms that don't have cache occupancy monitors, add this to the check that occurs when a new control group is allocated.
The snippet below does not belong in changelog and can be moved to maintainer notes.
Quoted 24 lines
The existing code is not selected by any upstream platform, it makes no sense to backport this patch to stable.
Fixes 6eac36bb9eb0 ("x86/resctrl: Allocate the cleanest CLOSID by searching closid_num_dirty_rmid")Signed-off-by James Morse <[email protected]>Tested-by Peter Newman <[email protected]>Tested-by Shaopeng Tan <[email protected]>Reviewed-by David Hildenbrand <[email protected]>---
Changes since v1:
* [Commit message only] Reword the first paragraph to make it clear that the issue being fixed wasn't directly associated with addition of aKconfigoption. (Actually, the option is not inKconfigyet, and gets added later in this series.)--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)arch/x86/kernel/cpu/resctrl/rdtgroup.c +0
-0
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int closid_alloc(void) · line 156
Quoted 9 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +2
-1@@ -149,7 +149,8 @@ static int closid_alloc(void)lockdep_assert_held (&rdtgroup_mutex); -if (IS_ENABLED (CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID )) { +if (IS_ENABLED (CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID ) && +is_llc_occupancy_enabled ()) { cleanest_closid =resctrl_find_cleanest_closid ();if (cleanest_closid <0 )return cleanest_closid;
With changelog updated:
| Reviewed-by: Reinette Chatre <[email protected]>
Reinette
James Morse
reviewerPATCH v3 2/38 x86/resctrl: Add a helper to avoid reaching into the arch code resource list
In reply to James MorseResctrl occasionally wants to know something about a specific resource, in these cases it reaches into the arch code's rdt_resources_all[] array.
Once the filesystem parts of resctrl are moved to /fs/, this means it will need visibility of the architecture specific struct rdt_hw_resource definition, and the array of all resources. All architectures would also need a r_resctrl member in this struct.
Instead, abstract this via a helper to allow architectures to do different things here. Move the level enum to the resctrl header and add a helper to retrieve the struct rdt_resource by 'rid'.
resctrl_arch_get_resource() should not return NULL for any value in the enum, it may instead return a dummy resource that is !alloc_enabled && !mon_enabled.
---
Changes since v1:
* Backed out non-functional renaming of "r" to "l3" in rdt_get_tree(), and unhoisted the assignment of r (as now is) back into the if () where it started out. There seem to be no uses of this variable outside this if().
* [Commit message only] Typo fix:
s/resctrl_hw_resource/rdt_hw_resource/g
--- arch/x86/kernel/cpu/resctrl/core.c | 10 +++++++++- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 2 +- arch/x86/kernel/cpu/resctrl/internal.h | 10 ---------- arch/x86/kernel/cpu/resctrl/monitor.c | 8 ++++---- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 12 ++++++------ include/linux/resctrl.h | 17 +++++++++++++++++ 6 files changed, 37 insertions(+), 22 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +9 -1
@@ -117,6 +117,14 @@ struct rdt_hw_resource rdt_resources_all[] = { }, }; +struct rdt_resource *resctrl_arch_get_resource (enum resctrl_res_level l) +{ +if (l >=RDT_NUM_RESOURCES ) +return NULL ; + +return &rdt_resources_all[l].r_resctrl; +} +/* * cache_alloc_hsw_probe() - Have to probe for Intel haswell server CPUs * as they do not have CPUID enumeration support for Cache allocation.
@@ -164,7 +172,7 @@ static inline void cache_alloc_hsw_probe(void)bool is_mba_sc (struct rdt_resource *r) {if (!r) -return rdt_resources_all[RDT_RESOURCE_MBA ].r_resctrl.membw.mba_sc; + r =resctrl_arch_get_resource (RDT_RESOURCE_MBA );/* * The software controller support is only applicable to MBA resource.
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1 -1
@@ -575,7 +575,7 @@ int rdtgroup_mondata_show(struct seq_file *m, void *arg) domid = md.u.domid; evtid = md.u.evtid; - r = &rdt_resources_all[resid].r_resctrl; + r =resctrl_arch_get_resource (resid); d =rdt_find_domain (r, domid,NULL );if (IS_ERR_OR_NULL (d)) { ret = -ENOENT ;
arch/x86/kernel/cpu/resctrl/internal.h +0 -10
@@ -468,16 +468,6 @@ extern struct rdt_hw_resource rdt_resources_all[];extern struct rdtgroup rdtgroup_default;extern struct dentry *debugfs_resctrl; -enum resctrl_res_level { -RDT_RESOURCE_L3 , -RDT_RESOURCE_L2 , -RDT_RESOURCE_MBA , -RDT_RESOURCE_SMBA , - -/* Must be the last */ -RDT_NUM_RESOURCES , -}; -static inline struct rdt_resource *resctrl_inc (struct rdt_resource *res) {struct rdt_hw_resource *hw_res =resctrl_to_arch_res (res);
arch/x86/kernel/cpu/resctrl/monitor.c +4 -4
@@ -322,7 +322,7 @@ static void limbo_release_entry(struct rmid_entry *entry)*/ void __check_limbo (struct rdt_domain *d,bool force_free) { -struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_L3 ].r_resctrl; +struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_L3 );u32 idx_limit =resctrl_arch_system_num_rmid_idx ();struct rmid_entry *entry;u32 idx, cur_idx =1 ;
@@ -478,7 +478,7 @@ int alloc_rmid(u32 closid)static void add_rmid_to_limbo (struct rmid_entry *entry) { -struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_L3 ].r_resctrl; +struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_L3 );struct rdt_domain *d;u32 idx;
@@ -680,7 +680,7 @@ static void update_mba_bw(struct rdtgroup *rgrp, struct rdt_domain *dom_mbm)if (!is_mbm_local_enabled ())return ; - r_mba = &rdt_resources_all[RDT_RESOURCE_MBA ].r_resctrl; + r_mba =resctrl_arch_get_resource (RDT_RESOURCE_MBA ); closid = rgrp->closid; rmid = rgrp->mon.rmid;
@@ -850,7 +850,7 @@ void mbm_handle_overflow(struct work_struct *work)if (!resctrl_mounted || !resctrl_arch_mon_capable ()) goto out_unlock; - r = &rdt_resources_all[RDT_RESOURCE_L3 ].r_resctrl; + r =resctrl_arch_get_resource (RDT_RESOURCE_L3 ); d = container_of(work, struct rdt_domain, mbm_over.work);list_for_each_entry (prgrp, &rdt_all_groups, rdtgroup_list) {
arch/x86/kernel/cpu/resctrl/rdtgroup.c +6 -6
@@ -2253,7 +2253,7 @@ static void l2_qos_cfg_update(void *arg)static inline bool is_mba_linear (void ) { -return rdt_resources_all[RDT_RESOURCE_MBA ].r_resctrl.membw.delay_linear; +return resctrl_arch_get_resource (RDT_RESOURCE_MBA )->membw.delay_linear; }static int set_cache_qos_cfg (int level,bool enable)
@@ -2341,7 +2341,7 @@ static void mba_sc_domain_destroy(struct rdt_resource *r,*/ static bool supports_mba_mbps (void ) { -struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_MBA ].r_resctrl; +struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_MBA );return (is_mbm_local_enabled () && r->alloc_capable &&is_mba_linear ());
@@ -2353,7 +2353,7 @@ static bool supports_mba_mbps(void)*/ static int set_mba_sc (bool mba_sc) { -struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_MBA ].r_resctrl; +struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_MBA );u32 num_closid =resctrl_arch_get_num_closid (r);struct rdt_domain *d;int i;
@@ -2701,7 +2701,7 @@ static int rdt_get_tree(struct fs_context *fc) resctrl_mounted = true;if (is_mbm_enabled ()) { - r = &rdt_resources_all[RDT_RESOURCE_L3 ].r_resctrl; + r =resctrl_arch_get_resource (RDT_RESOURCE_L3 );list_for_each_entry (dom, &r->domains, list)mbm_setup_overflow_handler (dom,MBM_OVERFLOW_INTERVAL ,RESCTRL_PICK_ANY_CPU );
@@ -3870,7 +3870,7 @@ static int rdtgroup_show_options(struct seq_file *seq, struct kernfs_root *kf)if (resctrl_arch_get_cdp_enabled (RDT_RESOURCE_L2 ))seq_puts (seq,",cdpl2" ); -if (is_mba_sc (&rdt_resources_all[RDT_RESOURCE_MBA ].r_resctrl)) +if (is_mba_sc (resctrl_arch_get_resource (RDT_RESOURCE_MBA )))seq_puts (seq,",mba_MBps" );if (resctrl_debug)
@@ -4060,7 +4060,7 @@ static void clear_childcpus(struct rdtgroup *r, unsigned int cpu)void resctrl_offline_cpu (unsigned int cpu) { -struct rdt_resource *l3 = &rdt_resources_all[RDT_RESOURCE_L3 ].r_resctrl; +struct rdt_resource *l3 =resctrl_arch_get_resource (RDT_RESOURCE_L3 );struct rdtgroup *rdtgrp;struct rdt_domain *d;
include/linux/resctrl.h +17 -0
@@ -36,6 +36,16 @@ enum resctrl_conf_type { CDP_DATA, }; +enum resctrl_res_level { +RDT_RESOURCE_L3 , +RDT_RESOURCE_L2 , +RDT_RESOURCE_MBA , +RDT_RESOURCE_SMBA , + +/* Must be the last */ +RDT_NUM_RESOURCES , +}; +#define CDP_NUM_TYPES (CDP_DATA + 1)/*
@@ -190,6 +200,13 @@ struct rdt_resource {bool cdp_capable; }; +/* +* Get the resource that exists at this level. If the level is not supported +* a dummy/not-capable resource can be returned. Levels >= RDT_NUM_RESOURCES +* will return NULL. +*/ +struct rdt_resource *resctrl_arch_get_resource (enum resctrl_res_level l); +/** * struct resctrl_schema - configuration abilities of a resource presented to * user-space
Signature · 1 line
2.39.2
Re: [PATCH v3 02/38] x86/resctrl: Add a helper to avoid reaching into the arch code resource list
Hi James,
Quoted 17 lines · James Morse
Resctrl occasionally wants to know something about a specific resource, in these cases it reaches into the arch code's rdt_resources_all[] array.
Once the filesystem parts of resctrl are moved to /fs/, this means it will need visibility of the architecture specific struct
rdt_hw_resourcedefinition, and the array of all resources. All architectures would also need ar_resctrlmember in this struct.Instead, abstract this via a helper to allow architectures to do different things here. Move the level enum to the resctrl header and add a helper to retrieve the struct
rdt_resourceby 'rid'.
resctrl_arch_get_resource()should not return NULL for any value in the enum, it may instead return a dummy resource that is !alloc_enabled && !mon_enabled.
Quoted 6 lines
Co-developed-by Dave Martin <[email protected]>Signed-off-by Dave Martin <[email protected]>Signed-off-by James Morse <[email protected]>Tested-by Peter Newman <[email protected]>Tested-by Shaopeng Tan <[email protected]>---
Reinette
James Morse
reviewerPATCH v3 3/38 x86/resctrl: Add a schema format enum and use this for fflags
In reply to James Morseresctrl has three types of control, these emerge from the way the architecture initialises a number of properties in struct rdt_resource.
A group of these properties need to be set the same on all architectures, it would be better to specify the format the schema entry should use, and allow resctrl to generate all the other properties it needs. This avoids architectures having divergant behaviour here.
Add a schema format enum, and as a first use, replace the fflags member of struct rdt_resource.
The MBA schema has a different format between AMD and Intel systems. The schema_fmt property is changed by __rdt_get_mem_config_amd() to enable the MBPS format.
---
Changes since v2:
* This patch is new
--- arch/x86/kernel/cpu/resctrl/core.c | 13 +++++++++---- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 17 +++++++++++++++-- include/linux/resctrl.h | 16 ++++++++++++++-- 3 files changed, 38 insertions(+), 8 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +9 -4
@@ -68,11 +68,11 @@ struct rdt_hw_resource rdt_resources_all[] = { .r_resctrl = { .rid = RDT_RESOURCE_L3, .name = "L3", + .schema_fmt = RESCTRL_SCHEMA_BITMAP, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_L3), .parse_ctrlval = parse_cbm, .format_str = "%d=%0*x", - .fflags = RFTYPE_RES_CACHE, }, .msr_base = MSR_IA32_L3_CBM_BASE, .msr_update = cat_wrmsr,
@@ -82,11 +82,11 @@ struct rdt_hw_resource rdt_resources_all[] = { .r_resctrl = { .rid = RDT_RESOURCE_L2, .name = "L2", + .schema_fmt = RESCTRL_SCHEMA_BITMAP, .cache_level = 2, .domains = domain_init(RDT_RESOURCE_L2), .parse_ctrlval = parse_cbm, .format_str = "%d=%0*x", - .fflags = RFTYPE_RES_CACHE, }, .msr_base = MSR_IA32_L2_CBM_BASE, .msr_update = cat_wrmsr,
@@ -96,11 +96,15 @@ struct rdt_hw_resource rdt_resources_all[] = { .r_resctrl = { .rid = RDT_RESOURCE_MBA, .name = "MB", +/* +* MBA schema_fmt is modified by +* __rdt_get_mem_config_amd() +*/ + .schema_fmt = RESCTRL_SCHEMA_PERCENTAGE, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_MBA), .parse_ctrlval = parse_bw, .format_str = "%d=%*u", - .fflags = RFTYPE_RES_MB, }, }, [RDT_RESOURCE_SMBA] =
@@ -108,11 +112,11 @@ struct rdt_hw_resource rdt_resources_all[] = { .r_resctrl = { .rid = RDT_RESOURCE_SMBA, .name = "SMBA", + .schema_fmt = RESCTRL_SCHEMA_MBPS, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_SMBA), .parse_ctrlval = parse_bw, .format_str = "%d=%*u", - .fflags = RFTYPE_RES_MB, }, }, };
@@ -253,6 +257,7 @@ static bool __rdt_get_mem_config_amd(struct rdt_resource *r)cpuid_count (0x80000020 , subleaf, &eax, &ebx, &ecx, &edx); hw_res->num_closid = edx +1 ; r->default_ctrl =1 << eax; + r->schema_fmt =RESCTRL_SCHEMA_MBPS ;/* AMD does not use delay */ r->membw.delay_linear = false;
arch/x86/kernel/cpu/resctrl/rdtgroup.c +15 -2
@@ -2162,6 +2162,19 @@ static int rdtgroup_mkdir_info_resdir(void *priv, char *name,return ret; } +static u32 fflags_from_resource (struct rdt_resource *r) +{ +switch (r->schema_fmt) { +case RESCTRL_SCHEMA_BITMAP : +return RFTYPE_RES_CACHE ; +case RESCTRL_SCHEMA_PERCENTAGE : +case RESCTRL_SCHEMA_MBPS : +return RFTYPE_RES_MB ; + } + +return WARN_ON_ONCE (1 ); +} +static int rdtgroup_create_info_dir (struct kernfs_node *parent_kn) {struct resctrl_schema *s;
@@ -2182,14 +2195,14 @@ static int rdtgroup_create_info_dir(struct kernfs_node *parent_kn)/* loop over enabled controls, these are all alloc_capable */ list_for_each_entry (s, &resctrl_schema_all, list) { r = s->res; - fflags = r->fflags |RFTYPE_CTRL_INFO ; + fflags =fflags_from_resource (r) |RFTYPE_CTRL_INFO ; ret =rdtgroup_mkdir_info_resdir (s, s->name, fflags);if (ret) goto out_destroy; }for_each_mon_capable_rdt_resource (r) { - fflags = r->fflags |RFTYPE_MON_INFO ; + fflags =fflags_from_resource (r) |RFTYPE_MON_INFO ;sprintf (name,"%s_MON" , r->name); ret =rdtgroup_mkdir_info_resdir (r, name, fflags);if (ret)
include/linux/resctrl.h +14 -2
@@ -160,6 +160,18 @@ struct resctrl_membw {struct rdt_parse_data ;struct resctrl_schema ; +/** +* enum resctrl_schema_fmt - The format user-space provides for a schema. +* @RESCTRL_SCHEMA_BITMAP: The schema is a bitmap in hex. +* @RESCTRL_SCHEMA_PERCENTAGE: The schema is a decimal percentage value. +* @RESCTRL_SCHEMA_MBPS: The schema is a decimal MBps value. +*/ +enum resctrl_schema_fmt { +RESCTRL_SCHEMA_BITMAP , +RESCTRL_SCHEMA_PERCENTAGE , +RESCTRL_SCHEMA_MBPS , +}; +/** * struct rdt_resource - attributes of a resctrl resource * @rid: The index of the resource
@@ -175,8 +187,8 @@ struct resctrl_schema;* @default_ctrl: Specifies default cache cbm or memory B/W percent. * @format_str: Per resource format string to show domain value * @parse_ctrlval: Per resource function pointer to parse control values +* @schema_fmt: Which format string and parser is used for this schema. * @evt_list: List of monitoring events -* @fflags: flags to choose base and info files * @cdp_capable: Is the CDP feature available on this resource */ struct rdt_resource {
@@ -195,8 +207,8 @@ struct rdt_resource {int (*parse_ctrlval)(struct rdt_parse_data *data,struct resctrl_schema *s,struct rdt_domain *d); +enum resctrl_schema_fmt schema_fmt;struct list_head evt_list; -unsigned long fflags;bool cdp_capable; };
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 03/38] x86/resctrl: Add a schema format enum and use this for fflags
Hi James,
Quoted 7 lines · James Morse
resctrl has three types of control, these emerge from the way the architecture initialises a number of properties in struct
rdt_resource.A group of these properties need to be set the same on all architectures, it would be better to specify the format the schema entry should use, and allow resctrl to generate all the other properties it needs. This avoids architectures having divergant behaviour here.
divergant -> divergent ?
Quoted 10 lines
Add a schema format enum, and as a first use, replace the fflags member of struct
rdt_resource.The MBA schema has a different format between AMD and Intel systems. The
schema_fmtproperty is changed by__rdt_get_mem_config_amd()to enable the MBPS format.Signed-off-by James Morse <[email protected]>---
...
Quoted 11 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +3
-0@@ -2162,6 +2162,19 @@ static int rdtgroup_mkdir_info_resdir(void *priv, char *name,return ret; } +static u32 fflags_from_resource (struct rdt_resource *r) +{ +switch (r->schema_fmt) {
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int rdtgroup_mkdir_info_resdir(void *priv, char *name, · line 2177
Quoted 10 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +10
-0+case RESCTRL_SCHEMA_BITMAP : +return RFTYPE_RES_CACHE ; +case RESCTRL_SCHEMA_PERCENTAGE : +case RESCTRL_SCHEMA_MBPS : +return RFTYPE_RES_MB ; + } + +return WARN_ON_ONCE (1 ); +} +
The fflags returned specifies which files will be associated with the resource in the "info" directory. Basing this on a property of the schema does not look right to me. I understand that many of the info files relate to, for example, information related to the bitmap used by the cache, but that is not the same for info files related to the MBA resource (all info files related to MBA resource are not about the schema property format).
I do not think the type of values of a schema should dictate which files appear in the info directory. Doesn't MPAM support percentage for cache resources and bitmaps for memory resources?
Can the fflags rather depend on the resource type itself, by using the rid?
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int rdtgroup_create_info_dir(struct kernfs_node *parent_kn) · line 2198
Quoted 9 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1
-1static int rdtgroup_create_info_dir (struct kernfs_node *parent_kn) {struct resctrl_schema *s;@@ -2182,14 +2195,14 @@ static int rdtgroup_create_info_dir(struct kernfs_node *parent_kn)/* loop over enabled controls, these are all alloc_capable */ list_for_each_entry (s, &resctrl_schema_all, list) { r = s->res; - fflags = r->fflags |RFTYPE_CTRL_INFO ; + fflags =fflags_from_resource (r) |RFTYPE_CTRL_INFO ;
(please watch for extra spaces)
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int rdtgroup_create_info_dir(struct kernfs_node *parent_kn) · line 2205
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1
-1ret =rdtgroup_mkdir_info_resdir (s, s->name, fflags);if (ret) goto out_destroy; }for_each_mon_capable_rdt_resource (r) { - fflags = r->fflags |RFTYPE_MON_INFO ; + fflags =fflags_from_resource (r) |RFTYPE_MON_INFO ;
(please watch for extra spaces)
Reinette
James Morse
reviewerRe: [PATCH v3 03/38] x86/resctrl: Add a schema format enum and use this for fflags
Hi Reinette,
Quoted 37 lines · Reinette Chatre
Quoted 7 lines · James Morse
resctrl has three types of control, these emerge from the way the architecture initialises a number of properties in struct
rdt_resource.A group of these properties need to be set the same on all architectures, it would be better to specify the format the schema entry should use, and allow resctrl to generate all the other properties it needs. This avoids architectures having divergant behaviour here.
divergant -> divergent ?
Quoted 26 lines
Add a schema format enum, and as a first use, replace the fflags member of struct
rdt_resource.The MBA schema has a different format between AMD and Intel systems. The
schema_fmtproperty is changed by__rdt_get_mem_config_amd()to enable the MBPS format.arch/x86/kernel/cpu/resctrl/rdtgroup.c +9
-0@@ -2162,6 +2162,19 @@ static int rdtgroup_mkdir_info_resdir(void *priv, char *name, return ret; } +static u32 fflags_from_resource(struct rdt_resource *r) +{ + switch (r->schema_fmt) { + case RESCTRL_SCHEMA_BITMAP: + return RFTYPE_RES_CACHE; + case RESCTRL_SCHEMA_PERCENTAGE: + case RESCTRL_SCHEMA_MBPS: + return RFTYPE_RES_MB; + } +
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int rdtgroup_mkdir_info_resdir(void *priv, char *name, · line 2177
Quoted 3 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +3
-0+ return WARN_ON_ONCE(1); +} +The fflags returned specifies which files will be associated with the resource in the "info" directory. Basing this on a property of the schema does not look right to me. I understand that many of the info files relate to, for example, information related to the bitmap used by the cache,
Do we agree that some of them are?
One reason for doing this is it decouples the parsing and management of bitmaps from "this is the L3 cache", which will make it much easier to support bitmaps on some other kind of resource.
Ultimately I'd like to expose these to user-space, so that user-space can work out how to configure resources it doesn't recognise. Today '100' could be a percentage, a bitmap, or a value in MB/s. Today some knowledge of the control type is needed to work this out.
Quoted 3 lines
but that is not the same for
info files related to the MBA resource (all info files related to MBA resource are not about the schema property format).
Hmmm, because the files min_bandwidth and bandwidth_gran both have bandwidth in their name?
I agree 'delay_linear' and 'thread_throttle_mode' are a bit strange.
Quoted 2 lines
I do not think the type of values of a schema should dictate which files appear in the info directory.
Longer term I think this will be a problem. We probably only have 3 types of control:
percentage, bitmap and MB/s... but if each resource on each architecture adds files here the list will quickly grow. User-space won't be able to work out how to configure a resource type it hadn't seen before.
This may not be the time - but I think eventually resctrl shouldn't have to care about what resources the architecture is presenting.
For these files, we may need to duplicate 'min_bandwidth' as 'min_percentage'. MBA would have both, but any new controls using percentage wouldn't expose them.
Quoted 2 lines
Doesn't MPAM support percentage for cache resources and bitmaps for memory resources?
It can have fixed-point-fractions and bitmaps for both caches and memory. Unfortunately everything in MPAM is optional - the driver converts whatever it finds for memory bandwidth to a percentage as that is what resctrl and user-space expect.
I can't do the same for cache controls as bitmaps implicitly isolate portions, something that can't be done with the fractional control. So far everyone has built the bitmaps because its the easiest implementation - but I have had requests to support the cache fixed-point-fraction. Doing it as a percentage is least invasive to resctrl...
Quoted 1 line
Can the fflags rather depend on the resource type itself, by using the rid?
Sure.
Thanks,
James
Quoted 20 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int rdtgroup_create_info_dir(struct kernfs_node *parent_kn) · line 2198
Quoted 6 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1
-1@@ -2182,14 +2195,14 @@ static int rdtgroup_create_info_dir(struct kernfs_node *parent_kn) /* loop over enabled controls, these are all alloc_capable */ list_for_each_entry(s, &resctrl_schema_all, list) { r = s->res; - fflags = r->fflags | RFTYPE_CTRL_INFO; + fflags = fflags_from_resource(r) | RFTYPE_CTRL_INFO;(please watch for extra spaces)
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int rdtgroup_create_info_dir(struct kernfs_node *parent_kn) · line 2205
Quoted 7 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1
-1ret = rdtgroup_mkdir_info_resdir(s, s->name, fflags); if (ret) goto out_destroy; } for_each_mon_capable_rdt_resource(r) { - fflags = r->fflags | RFTYPE_MON_INFO; + fflags = fflags_from_resource(r) | RFTYPE_MON_INFO;(please watch for extra spaces)
Reinette
Reinette Chatre
maintainerRe: [PATCH v3 03/38] x86/resctrl: Add a schema format enum and use this for fflags
Hi James,
Quoted 49 lines · James Morse
Hi Reinette,
Quoted 37 lines · Reinette Chatre
Quoted 7 lines · James Morse
resctrl has three types of control, these emerge from the way the architecture initialises a number of properties in struct
rdt_resource.A group of these properties need to be set the same on all architectures, it would be better to specify the format the schema entry should use, and allow resctrl to generate all the other properties it needs. This avoids architectures having divergant behaviour here.
divergant -> divergent ?
Quoted 26 lines
Add a schema format enum, and as a first use, replace the fflags member of struct
rdt_resource.The MBA schema has a different format between AMD and Intel systems. The
schema_fmtproperty is changed by__rdt_get_mem_config_amd()to enable the MBPS format.arch/x86/kernel/cpu/resctrl/rdtgroup.c +9
-0@@ -2162,6 +2162,19 @@ static int rdtgroup_mkdir_info_resdir(void *priv, char *name, return ret; } +static u32 fflags_from_resource(struct rdt_resource *r) +{ + switch (r->schema_fmt) { + case RESCTRL_SCHEMA_BITMAP: + return RFTYPE_RES_CACHE; + case RESCTRL_SCHEMA_PERCENTAGE: + case RESCTRL_SCHEMA_MBPS: + return RFTYPE_RES_MB; + } +Quoted 8 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int rdtgroup_mkdir_info_resdir(void *priv, char *name, · line 2177
Quoted 3 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +3
-0+ return WARN_ON_ONCE(1); +} +The fflags returned specifies which files will be associated with the resource in the "info" directory. Basing this on a property of the schema does not look right to me. I understand that many of the info files relate to, for example, information related to the bitmap used by the cache,
Quoted 5 lines
Do we agree that some of them are?
One reason for doing this is it decouples the parsing and management of bitmaps from "this is the L3 cache", which will make it much easier to support bitmaps on some other kind of resource.
The way I see it is that it changes the meaning of the RFTYPE_RES_CACHE flag from "this is a file related to the cache resource" to "this is a file containing a bitmap property". It prevents us from easily adding a file related to the cache resource, which the info directory is intended to contain.
Quoted 6 lines
Ultimately I'd like to expose these to user-space, so that user-space can work out how to configure resources it doesn't recognise. Today '100' could be a percentage, a bitmap, or a value in MB/s. Today some knowledge of the control type is needed to work this out.
Quoted 7 lines
Quoted 3 lines
but that is not the same for
info files related to the MBA resource (all info files related to MBA resource are not about the schema property format).Hmmm, because the files
min_bandwidthandbandwidth_granboth have bandwidth in their name?I agree '
delay_linear' and 'thread_throttle_mode' are a bit strange.
Right. This is not a clean association.
Quoted 9 lines
Quoted 4 lines
I do not think the type of values of a schema should dictate which files appear in the info directory.
Longer term I think this will be a problem. We probably only have 3 types of control:
percentage, bitmap and MB/s... but if each resource on each architecture adds files here the list will quickly grow. User-space won't be able to work out how to configure a resource type it hadn't seen before.
That is fair. This makes the type of control a property of the resource as is done in this series. Perhaps this can be exposed to user space via the info directory?
Possibly the files related to control can have new flags that that reflect the control type instead of the resource. For example, "bit_usage" currently has "RFTYPE_CTRL_INFO | RFTYPE_RES_CACHE" and that could be (for lack of better term) "RFTYPE_CTRL_INFO | RFTYPE_CTRL_BITMAP" to disconnect the control type from the resource. Doing so may then map nicely to the fflags_from_resource() in this patch that connects the schema format to the _control_ type flag. As we have found there is not a clear mapping between the control type and the resource type so I expect RFTYPE_RES_CACHE and RFTYPE_RES_MB to remain and be associated with files that contain information specific to that resource. This enables future additions of files containing cache specific (non-bitmap) properties to still be added (with RFTYPE_RES_CACHE flag) without impacting everything that uses a bitmap.
What do you think?
Quoted 19 lines
This may not be the time - but I think eventually resctrl shouldn't have to care about what resources the architecture is presenting.
For these files, we may need to duplicate 'min_bandwidth' as 'min_percentage'. MBA would have both, but any new controls using percentage wouldn't expose them.Quoted 2 lines
Doesn't MPAM support percentage for cache resources and bitmaps for memory resources?
It can have fixed-point-fractions and bitmaps for both caches and memory. Unfortunately everything in MPAM is optional - the driver converts whatever it finds for memory bandwidth to a percentage as that is what resctrl and user-space expect.
I can't do the same for cache controls as bitmaps implicitly isolate portions, something that can't be done with the fractional control. So far everyone has built the bitmaps because its the easiest implementation - but I have had requests to support the cache fixed-point-fraction. Doing it as a percentage is least invasive to resctrl...
Quoted 4 lines
Quoted 1 line
Can the fflags rather depend on the resource type itself, by using the rid?
Sure.
Reinette
James Morse
reviewerRe: [PATCH v3 03/38] x86/resctrl: Add a schema format enum and use this for fflags
Hi Reinette,
Quoted 50 lines · Reinette Chatre
Quoted 49 lines · James Morse
Quoted 37 lines · Reinette Chatre
Quoted 7 lines · James Morse
resctrl has three types of control, these emerge from the way the architecture initialises a number of properties in struct
rdt_resource.A group of these properties need to be set the same on all architectures, it would be better to specify the format the schema entry should use, and allow resctrl to generate all the other properties it needs. This avoids architectures having divergant behaviour here.
divergant -> divergent ?
Quoted 26 lines
Add a schema format enum, and as a first use, replace the fflags member of struct
rdt_resource.The MBA schema has a different format between AMD and Intel systems. The
schema_fmtproperty is changed by__rdt_get_mem_config_amd()to enable the MBPS format.arch/x86/kernel/cpu/resctrl/rdtgroup.c +9
-0@@ -2162,6 +2162,19 @@ static int rdtgroup_mkdir_info_resdir(void *priv, char *name, return ret; } +static u32 fflags_from_resource(struct rdt_resource *r) +{ + switch (r->schema_fmt) { + case RESCTRL_SCHEMA_BITMAP: + return RFTYPE_RES_CACHE; + case RESCTRL_SCHEMA_PERCENTAGE: + case RESCTRL_SCHEMA_MBPS: + return RFTYPE_RES_MB; + } +Quoted 8 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c static int rdtgroup_mkdir_info_resdir(void *priv, char *name, · line 2177
Quoted 3 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +3
-0+ return WARN_ON_ONCE(1); +} +The fflags returned specifies which files will be associated with the resource in the "info" directory. Basing this on a property of the schema does not look right to me. I understand that many of the info files relate to, for example, information related to the bitmap used by the cache,
Do we agree that some of them are?
Quoted 8 lines
Quoted 3 lines
One reason for doing this is it decouples the parsing and management of bitmaps from "this is the L3 cache", which will make it much easier to support bitmaps on some other kind of resource.
The way I see it is that it changes the meaning of the
RFTYPE_RES_CACHEflag from "this is a file related to the cache resource" to "this is a file containing a bitmap property". It prevents us from easily adding a file related to the cache resource, which the info directory is intended to contain.
I struggled to find something that is a property of a "cache control", but is neither a property of the control (e.g. bitmap size) or the cache. I guess the 'bit_usage' stuff is the best example.
Maybe we end up with two sets of flags - this will be for the distant future. Currently I taking your 'base fflags on resource id'.
Quoted 20 lines
Quoted 5 lines
Ultimately I'd like to expose these to user-space, so that user-space can work out how to configure resources it doesn't recognise. Today '100' could be a percentage, a bitmap, or a value in MB/s. Today some knowledge of the control type is needed to work this out.
Quoted 7 lines
Quoted 3 lines
but that is not the same for
info files related to the MBA resource (all info files related to MBA resource are not about the schema property format).Hmmm, because the files
min_bandwidthandbandwidth_granboth have bandwidth in their name?I agree '
delay_linear' and 'thread_throttle_mode' are a bit strange.Right. This is not a clean association.
Quoted 5 lines
Quoted 4 lines
I do not think the type of values of a schema should dictate which files appear in the info directory.
Quoted 7 lines
Quoted 4 lines
Longer term I think this will be a problem. We probably only have 3 types of control:
percentage, bitmap and MB/s... but if each resource on each architecture adds files here the list will quickly grow. User-space won't be able to work out how to configure a resource type it hadn't seen before.That is fair. This makes the type of control a property of the resource as is done in this series. Perhaps this can be exposed to user space via the info directory?
Yes, that is something I intend to look at. I eventually need to get MPAM's "cache capacity" controls working as there are a number of hardware platforms that have it. This would probably be a percentage control for 'L2' or 'L3', exposing an "info/schema_format" file makes the most sense. I can't convert the existing bitmap as it implies isolation, which this control format can't do, so it does need to be separate.
But! - to prevent confusing existing software, I don't think the L2/L3 should be touched - those will forever have to be implicitly a bitmap, so anything in this area would have to be an additional schema.
Quoted 12 lines
Possibly the files related to control can have new flags that that reflect the control type instead of the resource. For example, "
bit_usage" currently has "RFTYPE_CTRL_INFO|RFTYPE_RES_CACHE" and that could be (for lack of better term) "RFTYPE_CTRL_INFO|RFTYPE_CTRL_BITMAP" to disconnect the control type from the resource. Doing so may then map nicely to thefflags_from_resource()in this patch that connects the schema format to the _control_ type flag. As we have found there is not a clear mapping between the control type and the resource type so I expectRFTYPE_RES_CACHEandRFTYPE_RES_MBto remain and be associated with files that contain information specific to that resource. This enables future additions of files containing cache specific (non-bitmap) properties to still be added (withRFTYPE_RES_CACHEflag) without impacting everything that uses a bitmap.
Quoted 1 line
What do you think?
Makes sense!
Thanks,
James
James Morse
reviewerPATCH v3 4/38 x86/resctrl: Use schema type to determine how to parse schema values
In reply to James MorseResctrl's architecture code gets to specify a function pointer that is used when parsing schema entries. This is expected to be one of two helpers from the filesystem code.
Setting this function pointer allows the architecture code to change the ABI resctrl presents to user-space, and forces resctrl to expose these helpers.
Instead, use the schema format enum to choose which schema parser to use. This allows the helpers to be made static and the structs used for passing arguments moved out of shared headers.
---
Changes since v2:
* This patch is new
--- arch/x86/kernel/cpu/resctrl/core.c | 4 --- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 33 +++++++++++++++++++---- arch/x86/kernel/cpu/resctrl/internal.h | 10 ------- include/linux/resctrl.h | 5 ---- 4 files changed, 28 insertions(+), 24 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +0 -4
@@ -71,7 +71,6 @@ struct rdt_hw_resource rdt_resources_all[] = { .schema_fmt = RESCTRL_SCHEMA_BITMAP, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_L3), - .parse_ctrlval = parse_cbm, .format_str = "%d=%0*x", }, .msr_base = MSR_IA32_L3_CBM_BASE,
@@ -85,7 +84,6 @@ struct rdt_hw_resource rdt_resources_all[] = { .schema_fmt = RESCTRL_SCHEMA_BITMAP, .cache_level = 2, .domains = domain_init(RDT_RESOURCE_L2), - .parse_ctrlval = parse_cbm, .format_str = "%d=%0*x", }, .msr_base = MSR_IA32_L2_CBM_BASE,
@@ -103,7 +101,6 @@ struct rdt_hw_resource rdt_resources_all[] = { .schema_fmt = RESCTRL_SCHEMA_PERCENTAGE, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_MBA), - .parse_ctrlval = parse_bw, .format_str = "%d=%*u", }, },
@@ -115,7 +112,6 @@ struct rdt_hw_resource rdt_resources_all[] = { .schema_fmt = RESCTRL_SCHEMA_MBPS, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_SMBA), - .parse_ctrlval = parse_bw, .format_str = "%d=%*u", }, },
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +28 -5
@@ -23,6 +23,15 @@#include "internal.h" +struct rdt_parse_data { +struct rdtgroup *rdtgrp; +char *buf; +}; + +typedef int (ctrlval_parser_t )(struct rdt_parse_data *data, +struct resctrl_schema *s, +struct rdt_domain *d); +/* * Check whether MBA bandwidth percentage value is correct. The value is * checked against the minimum and max bandwidth values specified by the
@@ -59,8 +68,8 @@ static bool bw_validate(char *buf, unsigned long *data, struct rdt_resource *r)return true; } -int parse_bw (struct rdt_parse_data *data,struct resctrl_schema *s, -struct rdt_domain *d) +static int parse_bw (struct rdt_parse_data *data,struct resctrl_schema *s, +struct rdt_domain *d) {struct resctrl_staged_config *cfg;u32 closid = data->rdtgrp->closid;
@@ -138,8 +147,8 @@ static bool cbm_validate(char *buf, u32 *data, struct rdt_resource *r)* Read one cache bit mask (hex). Check that it is valid for the current * resource type. */ -int parse_cbm (struct rdt_parse_data *data,struct resctrl_schema *s, -struct rdt_domain *d) +static int parse_cbm (struct rdt_parse_data *data,struct resctrl_schema *s, +struct rdt_domain *d) {struct rdtgroup *rdtgrp = data->rdtgrp;struct resctrl_staged_config *cfg;
@@ -195,6 +204,19 @@ int parse_cbm(struct rdt_parse_data *data, struct resctrl_schema *s,return 0 ; } +static ctrlval_parser_t *get_parser (struct rdt_resource *r) +{ +switch (r->schema_fmt) { +case RESCTRL_SCHEMA_BITMAP : +return &parse_cbm; +case RESCTRL_SCHEMA_PERCENTAGE : +case RESCTRL_SCHEMA_MBPS : +return &parse_bw; + }; + +return NULL ; +} +/* * For each domain in this resource we expect to find a series of: * id=mask
@@ -204,6 +226,7 @@ int parse_cbm(struct rdt_parse_data *data, struct resctrl_schema *s,static int parse_line (char *line,struct resctrl_schema *s,struct rdtgroup *rdtgrp) { +ctrlval_parser_t *parse_ctrlval =get_parser (s->res);enum resctrl_conf_type t = s->conf_type;struct resctrl_staged_config *cfg;struct rdt_resource *r = s->res;
@@ -235,7 +258,7 @@ static int parse_line(char *line, struct resctrl_schema *s,if (d->id == dom_id) { data.buf = dom; data.rdtgrp = rdtgrp; -if (r->parse_ctrlval (&data, s, d)) +if (parse_ctrlval (&data, s, d))return -EINVAL ;if (rdtgrp->mode ==RDT_MODE_PSEUDO_LOCKSETUP ) { cfg = &d->staged_config[t];
arch/x86/kernel/cpu/resctrl/internal.h +0 -10
@@ -416,11 +416,6 @@ static inline bool is_mbm_event(int e) e <= QOS_L3_MBM_LOCAL_EVENT_ID); } -struct rdt_parse_data { -struct rdtgroup *rdtgrp; -char *buf; -}; -/** * struct rdt_hw_resource - arch private attributes of a resctrl resource * @r_resctrl: Attributes of the resource used directly by resctrl.
@@ -457,11 +452,6 @@ static inline struct rdt_hw_resource *resctrl_to_arch_res(struct rdt_resource *r return container_of(r, struct rdt_hw_resource, r_resctrl); } -int parse_cbm (struct rdt_parse_data *data,struct resctrl_schema *s, -struct rdt_domain *d); -int parse_bw (struct rdt_parse_data *data,struct resctrl_schema *s, -struct rdt_domain *d); -extern struct mutex rdtgroup_mutex;extern struct rdt_hw_resource rdt_resources_all[];
include/linux/resctrl.h +0 -5
@@ -157,7 +157,6 @@ struct resctrl_membw {u32 *mb_map; }; -struct rdt_parse_data ;struct resctrl_schema ;/**
@@ -186,7 +185,6 @@ enum resctrl_schema_fmt {* @data_width: Character width of data when displaying * @default_ctrl: Specifies default cache cbm or memory B/W percent. * @format_str: Per resource format string to show domain value -* @parse_ctrlval: Per resource function pointer to parse control values * @schema_fmt: Which format string and parser is used for this schema. * @evt_list: List of monitoring events * @cdp_capable: Is the CDP feature available on this resource
@@ -204,9 +202,6 @@ struct rdt_resource {int data_width;u32 default_ctrl;const char *format_str; -int (*parse_ctrlval)(struct rdt_parse_data *data, -struct resctrl_schema *s, -struct rdt_domain *d);enum resctrl_schema_fmt schema_fmt;struct list_head evt_list;bool cdp_capable;
Signature · 1 line
2.39.2
Re: [PATCH v3 04/38] x86/resctrl: Use schema type to determine how to parse schema values
Hi James,
Quoted 8 lines · James Morse
Resctrl's architecture code gets to specify a function pointer that is used when parsing schema entries. This is expected to be one of two helpers from the filesystem code.
Setting this function pointer allows the architecture code to change the ABI resctrl presents to user-space, and forces resctrl to expose these helpers.
Quoted 6 lines
Instead, use the schema format enum to choose which schema parser to use. This allows the helpers to be made static and the structs used for passing arguments moved out of shared headers.
Signed-off-by James Morse <[email protected]>---
Reinette
James Morse
reviewerPATCH v3 5/38 x86/resctrl: Use schema type to determine the schema format string
In reply to James MorseResctrl's architecture code gets to specify a format string that is used when printing schema entries. This is expected to be one of two values that the filesystem code supports.
Setting this format string allows the architecture code to change the ABI resctrl presents to user-space.
Instead, use the schema format enum to choose which format string to use.
---
Changes since v2:
* This patch is new.
--- arch/x86/kernel/cpu/resctrl/core.c | 4 ---- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 2 +- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 12 ++++++++++++ include/linux/resctrl.h | 4 ++-- 4 files changed, 15 insertions(+), 7 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +0 -4
@@ -71,7 +71,6 @@ struct rdt_hw_resource rdt_resources_all[] = { .schema_fmt = RESCTRL_SCHEMA_BITMAP, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_L3), - .format_str = "%d=%0*x", }, .msr_base = MSR_IA32_L3_CBM_BASE, .msr_update = cat_wrmsr,
@@ -84,7 +83,6 @@ struct rdt_hw_resource rdt_resources_all[] = { .schema_fmt = RESCTRL_SCHEMA_BITMAP, .cache_level = 2, .domains = domain_init(RDT_RESOURCE_L2), - .format_str = "%d=%0*x", }, .msr_base = MSR_IA32_L2_CBM_BASE, .msr_update = cat_wrmsr,
@@ -101,7 +99,6 @@ struct rdt_hw_resource rdt_resources_all[] = { .schema_fmt = RESCTRL_SCHEMA_PERCENTAGE, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_MBA), - .format_str = "%d=%*u", }, }, [RDT_RESOURCE_SMBA] =
@@ -112,7 +109,6 @@ struct rdt_hw_resource rdt_resources_all[] = { .schema_fmt = RESCTRL_SCHEMA_MBPS, .cache_level = 3, .domains = domain_init(RDT_RESOURCE_SMBA), - .format_str = "%d=%*u", }, }, };
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1 -1
@@ -483,7 +483,7 @@ static void show_doms(struct seq_file *s, struct resctrl_schema *schema, int clo ctrl_val =resctrl_arch_get_config (r, dom, closid, schema->conf_type); -seq_printf (s, r->format_str, dom->id, max_data_width, +seq_printf (s, schema->fmt_str, dom->id, max_data_width, ctrl_val); sep = true; }
arch/x86/kernel/cpu/resctrl/rdtgroup.c +12 -0
@@ -2597,6 +2597,18 @@ static int schemata_list_add(struct rdt_resource *r, enum resctrl_conf_type typeif (cl > max_name_width) max_name_width = cl; +switch (r->schema_fmt) { +case RESCTRL_SCHEMA_BITMAP : + s->fmt_str ="%d=%0*x" ; +break ; +case RESCTRL_SCHEMA_PERCENTAGE : + s->fmt_str ="%d=%0*u" ; +break ; +case RESCTRL_SCHEMA_MBPS : + s->fmt_str ="%d=%0*u" ; +break ; + } +INIT_LIST_HEAD (&s->list);list_add (&s->list, &resctrl_schema_all);
include/linux/resctrl.h +2 -2
@@ -184,7 +184,6 @@ enum resctrl_schema_fmt {* @name: Name to use in "schemata" file. * @data_width: Character width of data when displaying * @default_ctrl: Specifies default cache cbm or memory B/W percent. -* @format_str: Per resource format string to show domain value * @schema_fmt: Which format string and parser is used for this schema. * @evt_list: List of monitoring events * @cdp_capable: Is the CDP feature available on this resource
@@ -201,7 +200,6 @@ struct rdt_resource {char *name;int data_width;u32 default_ctrl; -const char *format_str;enum resctrl_schema_fmt schema_fmt;struct list_head evt_list;bool cdp_capable;
@@ -219,6 +217,7 @@ struct rdt_resource *resctrl_arch_get_resource(enum resctrl_res_level l);* user-space * @list: Member of resctrl_schema_all. * @name: The name to use in the "schemata" file. +* @fmt_str: Format string to show domain value * @conf_type: Whether this schema is specific to code/data. * @res: The resource structure exported by the architecture to describe * the hardware that is configured by this schema.
@@ -229,6 +228,7 @@ struct rdt_resource *resctrl_arch_get_resource(enum resctrl_res_level l);struct resctrl_schema {struct list_head list;char name[8 ]; +const char *fmt_str;enum resctrl_conf_type conf_type;struct rdt_resource *res;u32 num_closid;
Signature · 1 line
2.39.2
Re: [PATCH v3 05/38] x86/resctrl: Use schema type to determine the schema format string
Hi James,
Quoted 4 lines · James Morse
Resctrl's architecture code gets to specify a format string that is used when printing schema entries. This is expected to be one of two values that the filesystem code supports.
Quoted 8 lines
Setting this format string allows the architecture code to change the ABI resctrl presents to user-space.
Instead, use the schema format enum to choose which format string to use.
Signed-off-by James Morse <[email protected]>---
Reinette
James Morse
reviewerPATCH v3 6/38 x86/resctrl: Move data_width to be a schema property
In reply to James MorseThe resctrl architecture code gets to specify the width of the schema entries that are used by resctrl. These are determined by the schema format, e.g. percentage or bitmap.
Move this property into struct resctrl_schema and get the filesystem parts of resctrl to set it based on the schema format.
This allows rdt_init_padding() to be removed, its work can be done by schemata_list_add(), allowing max_name_width and max_data_width to be moved out of core.c which has no counterpart after the move to fs.
The logic for calculating max_name_width was moved in earlier patches, but the definition was not moved.
---
Changes since v2:
* This patch is new.
--- arch/x86/kernel/cpu/resctrl/core.c | 26 -------------------------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 11 +++++++++++ include/linux/resctrl.h | 4 ++-- 3 files changed, 13 insertions(+), 28 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +0 -26
@@ -44,12 +44,6 @@ static DEFINE_MUTEX(domain_list_lock);*/ DEFINE_PER_CPU(struct resctrl_pqr_state, pqr_state); -/* -* Used to store the max resource name width and max resource data width -* to display the schemata in a tabular format -*/ -int max_name_width, max_data_width; -/* * Global boolean for rdt_alloc which is true if any * resource allocation is enabled.
@@ -222,7 +216,6 @@ static bool __get_mem_config_intel(struct rdt_resource *r)return false; r->membw.arch_needs_linear = false; } - r->data_width =3 ;if (boot_cpu_has (X86_FEATURE_PER_THREAD_MBA )) r->membw.throttle_mode =THREAD_THROTTLE_PER_THREAD ;
@@ -262,8 +255,6 @@ static bool __rdt_get_mem_config_amd(struct rdt_resource *r) r->membw.throttle_mode =THREAD_THROTTLE_UNDEFINED ; r->membw.min_bw =0 ; r->membw.bw_gran =1 ; -/* Max value is 2048, Data width should be 4 in decimal */ - r->data_width =4 ; r->alloc_capable = true;
@@ -283,7 +274,6 @@ static void rdt_get_cache_alloc_cfg(int idx, struct rdt_resource *r) r->cache.cbm_len = eax.split.cbm_len +1 ; r->default_ctrl =BIT_MASK (eax.split.cbm_len +1 ) -1 ; r->cache.shareable_bits = ebx & r->default_ctrl; - r->data_width = (r->cache.cbm_len +3 ) /4 ;if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) r->cache.arch_has_sparse_bitmasks = ecx.split.noncont; r->alloc_capable = true;
@@ -631,20 +621,6 @@ static int resctrl_arch_offline_cpu(unsigned int cpu)return 0 ; } -/* -* Choose a width for the resource name and resource data based on the -* resource that has widest name and cbm. -*/ -static __init void rdt_init_padding (void ) -{ -struct rdt_resource *r; - -for_each_alloc_capable_rdt_resource (r) { -if (r->data_width > max_data_width) - max_data_width = r->data_width; - } -} -enum {RDT_FLAG_CMT ,RDT_FLAG_MBM_TOTAL ,
@@ -942,8 +918,6 @@ static int __init resctrl_late_init(void)if (!get_rdt_resources ())return -ENODEV ; -rdt_init_padding (); - state = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "x86/resctrl/cat:online:", resctrl_arch_online_cpu,
arch/x86/kernel/cpu/resctrl/rdtgroup.c +11 -0
@@ -58,6 +58,12 @@ static struct kernfs_node *kn_mongrp;/* Kernel fs node for "mon_data" directory under root */ static struct kernfs_node *kn_mondata; +/* +* Used to store the max resource name width and max resource data width +* to display the schemata in a tabular format +*/ +int max_name_width, max_data_width; +static struct seq_buf last_cmd_status;static char last_cmd_status_buf[512 ];
@@ -2600,15 +2606,20 @@ static int schemata_list_add(struct rdt_resource *r, enum resctrl_conf_type typeswitch (r->schema_fmt) {case RESCTRL_SCHEMA_BITMAP : s->fmt_str ="%d=%0*x" ; + s->data_width = (r->cache.cbm_len +3 ) /4 ;break ;case RESCTRL_SCHEMA_PERCENTAGE : s->fmt_str ="%d=%0*u" ; + s->data_width =3 ;break ;case RESCTRL_SCHEMA_MBPS : s->fmt_str ="%d=%0*u" ; + s->data_width =4 ;break ; } + max_data_width =max (max_data_width, s->data_width); +INIT_LIST_HEAD (&s->list);list_add (&s->list, &resctrl_schema_all);
include/linux/resctrl.h +2 -2
@@ -182,7 +182,6 @@ enum resctrl_schema_fmt {* @membw: If the component has bandwidth controls, their properties. * @domains: RCU list of all domains for this resource * @name: Name to use in "schemata" file. -* @data_width: Character width of data when displaying * @default_ctrl: Specifies default cache cbm or memory B/W percent. * @schema_fmt: Which format string and parser is used for this schema. * @evt_list: List of monitoring events
@@ -198,7 +197,6 @@ struct rdt_resource {struct resctrl_membw membw;struct list_head domains;char *name; -int data_width;u32 default_ctrl;enum resctrl_schema_fmt schema_fmt;struct list_head evt_list;
@@ -218,6 +216,7 @@ struct rdt_resource *resctrl_arch_get_resource(enum resctrl_res_level l);* @list: Member of resctrl_schema_all. * @name: The name to use in the "schemata" file. * @fmt_str: Format string to show domain value +* @data_width: Character width of data when displaying * @conf_type: Whether this schema is specific to code/data. * @res: The resource structure exported by the architecture to describe * the hardware that is configured by this schema.
@@ -229,6 +228,7 @@ struct resctrl_schema {struct list_head list;char name[8 ];const char *fmt_str; +int data_width;enum resctrl_conf_type conf_type;struct rdt_resource *res;u32 num_closid;
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 06/38] x86/resctrl: Move data_width to be a schema property
Hi James,
Quoted 11 lines · James Morse
The resctrl architecture code gets to specify the width of the schema entries that are used by resctrl. These are determined by the schema format, e.g. percentage or bitmap.
Move this property into struct
resctrl_schemaand get the filesystem parts of resctrl to set it based on the schema format.This allows
rdt_init_padding()to be removed, its work can be done byschemata_list_add(), allowingmax_name_widthandmax_data_widthto be moved out ofcore.cwhich has no counterpart after the move to fs.
Please do write commit messages in imperative mood.
Quoted 91 lines
The logic for calculating
max_name_widthwas moved in earlier patches, but the definition was not moved.Signed-off-by James Morse <[email protected]>---
Changes since v2:
* This patch is new.--- arch/x86/kernel/cpu/resctrl/core.c | 26 -------------------------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 11 +++++++++++ include/linux/resctrl.h | 4 ++-- 3 files changed, 13 insertions(+), 28 deletions(-)arch/x86/kernel/cpu/resctrl/core.c +0
-26@@ -44,12 +44,6 @@ static DEFINE_MUTEX(domain_list_lock);*/ DEFINE_PER_CPU(struct resctrl_pqr_state, pqr_state); -/* -* Used to store the max resource name width and max resource data width -* to display the schemata in a tabular format -*/ -int max_name_width, max_data_width; -/* * Global boolean for rdt_alloc which is true if any * resource allocation is enabled. @@ -222,7 +216,6 @@ static bool __get_mem_config_intel(struct rdt_resource *r)return false; r->membw.arch_needs_linear = false; } - r->data_width =3 ;if (boot_cpu_has (X86_FEATURE_PER_THREAD_MBA )) r->membw.throttle_mode =THREAD_THROTTLE_PER_THREAD ;@@ -262,8 +255,6 @@ static bool __rdt_get_mem_config_amd(struct rdt_resource *r) r->membw.throttle_mode =THREAD_THROTTLE_UNDEFINED ; r->membw.min_bw =0 ; r->membw.bw_gran =1 ; -/* Max value is 2048, Data width should be 4 in decimal */ - r->data_width =4 ; r->alloc_capable = true;@@ -283,7 +274,6 @@ static void rdt_get_cache_alloc_cfg(int idx, struct rdt_resource *r) r->cache.cbm_len = eax.split.cbm_len +1 ; r->default_ctrl =BIT_MASK (eax.split.cbm_len +1 ) -1 ; r->cache.shareable_bits = ebx & r->default_ctrl; - r->data_width = (r->cache.cbm_len +3 ) /4 ;if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) r->cache.arch_has_sparse_bitmasks = ecx.split.noncont; r->alloc_capable = true;@@ -631,20 +621,6 @@ static int resctrl_arch_offline_cpu(unsigned int cpu)return 0 ; } -/* -* Choose a width for the resource name and resource data based on the -* resource that has widest name and cbm. -*/ -static __init void rdt_init_padding (void ) -{ -struct rdt_resource *r; - -for_each_alloc_capable_rdt_resource (r) { -if (r->data_width > max_data_width) - max_data_width = r->data_width; - } -} -enum {RDT_FLAG_CMT ,RDT_FLAG_MBM_TOTAL ,@@ -942,8 +918,6 @@ static int __init resctrl_late_init(void)if (!get_rdt_resources ())return -ENODEV ; -rdt_init_padding (); - state = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "x86/resctrl/cat:online:", resctrl_arch_online_cpu,arch/x86/kernel/cpu/resctrl/rdtgroup.c +0
-0
arch/x86/kernel/cpu/resctrl/rdtgroup.c static struct kernfs_node *kn_mongrp; · line 64
Quoted 7 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +4
-0@@ -58,6 +58,12 @@ static struct kernfs_node *kn_mongrp;/* Kernel fs node for "mon_data" directory under root */ static struct kernfs_node *kn_mondata; +/* +* Used to store the max resource name width and max resource data width +* to display the schemata in a tabular format +*/
I understand that you just copied existing text here, but since you touch this line could you please have this sentence end with a period?
Quoted 55 lines
arch/x86/kernel/cpu/resctrl/rdtgroup.c +7
-0+int max_name_width, max_data_width; +static struct seq_buf last_cmd_status;static char last_cmd_status_buf[512 ];@@ -2600,15 +2606,20 @@ static int schemata_list_add(struct rdt_resource *r, enum resctrl_conf_type typeswitch (r->schema_fmt) {case RESCTRL_SCHEMA_BITMAP : s->fmt_str ="%d=%0*x" ; + s->data_width = (r->cache.cbm_len +3 ) /4 ;break ;case RESCTRL_SCHEMA_PERCENTAGE : s->fmt_str ="%d=%0*u" ; + s->data_width =3 ;break ;case RESCTRL_SCHEMA_MBPS : s->fmt_str ="%d=%0*u" ; + s->data_width =4 ;break ; } + max_data_width =max (max_data_width, s->data_width); +INIT_LIST_HEAD (&s->list);list_add (&s->list, &resctrl_schema_all);include/linux/resctrl.h +1
-2@@ -182,7 +182,6 @@ enum resctrl_schema_fmt {* @membw: If the component has bandwidth controls, their properties. * @domains: RCU list of all domains for this resource * @name: Name to use in "schemata" file. -* @data_width: Character width of data when displaying * @default_ctrl: Specifies default cache cbm or memory B/W percent. * @schema_fmt: Which format string and parser is used for this schema. * @evt_list: List of monitoring events @@ -198,7 +197,6 @@ struct rdt_resource {struct resctrl_membw membw;struct list_head domains;char *name; -int data_width;u32 default_ctrl;enum resctrl_schema_fmt schema_fmt;struct list_head evt_list;@@ -218,6 +216,7 @@ struct rdt_resource *resctrl_arch_get_resource(enum resctrl_res_level l);* @list: Member of resctrl_schema_all. * @name: The name to use in the "schemata" file. * @fmt_str: Format string to show domain value +* @data_width: Character width of data when displaying * @conf_type: Whether this schema is specific to code/data. * @res: The resource structure exported by the architecture to describe * the hardware that is configured by this schema.
include/linux/resctrl.h struct resctrl_schema { · line 234
Quoted 7 lines
include/linux/resctrl.h +1
-0@@ -229,6 +228,7 @@ struct resctrl_schema {struct list_head list;char name[8 ];const char *fmt_str; +int data_width;enum resctrl_conf_type conf_type;struct rdt_resource *res;u32 num_closid;
Reinette
James Morse
reviewerPATCH v3 7/38 x86/resctrl: Add max_bw to struct resctrl_membw
In reply to James Morse__rdt_get_mem_config_amd() and __get_mem_config_intel() both use the default_ctrl property as a maximum value. This is because the MBA schema works differently between these platforms. Doing this complicates determining whether the default_ctrl property belongs to the arch code, or can be derived from the schema format.
Add a max_bw property for AMD platforms to specify their maximum MBA bandwidth. This isn't needed for other schema formats.
This will allow the default_ctrl to be generated from the schema properties when it is needed.
---
Changes since v2:
* This patch is new.
--- arch/x86/kernel/cpu/resctrl/core.c | 2 ++ arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 9 +++++---- include/linux/resctrl.h | 2 ++ 3 files changed, 9 insertions(+), 4 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +2 -0
@@ -206,6 +206,7 @@ static bool __get_mem_config_intel(struct rdt_resource *r) hw_res->num_closid = edx.split.cos_max +1 ; max_delay = eax.split.max_delay +1 ; r->default_ctrl =MAX_MBA_BW ; + r->membw.max_bw =MAX_MBA_BW ; r->membw.arch_needs_linear = true;if (ecx &MBA_IS_LINEAR ) { r->membw.delay_linear = true;
@@ -243,6 +244,7 @@ static bool __rdt_get_mem_config_amd(struct rdt_resource *r) hw_res->num_closid = edx +1 ; r->default_ctrl =1 << eax; r->schema_fmt =RESCTRL_SCHEMA_MBPS ; + r->membw.max_bw =1 << eax;/* AMD does not use delay */ r->membw.delay_linear = false;
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +5 -4
@@ -57,10 +57,10 @@ static bool bw_validate(char *buf, unsigned long *data, struct rdt_resource *r)return false; } -if ((bw < r->membw.min_bw || bw > r->default_ctrl) && +if ((bw < r->membw.min_bw || bw > r->membw.max_bw) && !is_mba_sc (r)) {rdt_last_cmd_printf ("MB value %ld out of range [%d,%d]\n" , bw, - r->membw.min_bw, r->default_ctrl); + r->membw.min_bw, r->membw.max_bw);return false; }
@@ -108,8 +108,9 @@ static int parse_bw(struct rdt_parse_data *data, struct resctrl_schema *s,*/ static bool cbm_validate (char *buf,u32 *data,struct rdt_resource *r) { -unsigned long first_bit, zero_bit, val; +u32 supported_bits =BIT_MASK (r->cache.cbm_len +1 ) -1 ;unsigned int cbm_len = r->cache.cbm_len; +unsigned long first_bit, zero_bit, val;int ret; ret =kstrtoul (buf,16 , &val);
@@ -118,7 +119,7 @@ static bool cbm_validate(char *buf, u32 *data, struct rdt_resource *r)return false; } -if ((r->cache.min_cbm_bits >0 && val ==0 ) || val > r->default_ctrl) { +if ((r->cache.min_cbm_bits >0 && val ==0 ) || val > supported_bits) {rdt_last_cmd_puts ("Mask out of range\n" );return false; }
include/linux/resctrl.h +2 -0
@@ -139,6 +139,7 @@ enum membw_throttle_mode {/** * struct resctrl_membw - Memory bandwidth allocation related data * @min_bw: Minimum memory bandwidth percentage user can request +* @max_bw: Maximum memory bandwidth value, used as the reset value * @bw_gran: Granularity at which the memory bandwidth is allocated * @delay_linear: True if memory B/W delay is in linear scale * @arch_needs_linear: True if we can't configure non-linear resources
@@ -149,6 +150,7 @@ enum membw_throttle_mode {*/ struct resctrl_membw {u32 min_bw; +u32 max_bw;u32 bw_gran;u32 delay_linear;bool arch_needs_linear;
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 8/38 x86/resctrl: Generate default_ctrl instead of sharing it
In reply to James MorseThe struct rdt_resource default_ctrl is used by both the architecture code for resetting the hardware controls, and by the filesystem parts of resctrl to report to user-space.
This means the value has to be shared, but might not match the properties of the control. e.g. a percentage greater than 100.
Instead, determine the default control value from a shared helper resctrl_get_default_ctrl() that uses the schema properties to determine the correct value.
---
Changes since v2:
* This patch is new.
--- arch/x86/kernel/cpu/resctrl/core.c | 16 +++++++--------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 6 +++--- include/linux/resctrl.h | 21 +++++++++++++++++++-- 3 files changed, 29 insertions(+), 14 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +7 -9
@@ -137,7 +137,10 @@ static inline void cache_alloc_hsw_probe(void) {struct rdt_hw_resource *hw_res = &rdt_resources_all[RDT_RESOURCE_L3 ];struct rdt_resource *r = &hw_res->r_resctrl; -u64 max_cbm =BIT_ULL_MASK (20 ) -1 , l3_cbm_0; +u64 max_cbm, l3_cbm_0; + + r->cache.cbm_len =20 ; + max_cbm =resctrl_get_default_ctrl (r);if (wrmsrl_safe (MSR_IA32_L3_CBM_BASE , max_cbm))return ;
@@ -149,8 +152,6 @@ static inline void cache_alloc_hsw_probe(void)return ; hw_res->num_closid =4 ; - r->default_ctrl = max_cbm; - r->cache.cbm_len =20 ; r->cache.shareable_bits =0xc0000 ; r->cache.min_cbm_bits =2 ; r->cache.arch_has_sparse_bitmasks = false;
@@ -205,7 +206,6 @@ static bool __get_mem_config_intel(struct rdt_resource *r)cpuid_count (0x00000010 ,3 , &eax.full, &ebx, &ecx, &edx.full); hw_res->num_closid = edx.split.cos_max +1 ; max_delay = eax.split.max_delay +1 ; - r->default_ctrl =MAX_MBA_BW ; r->membw.max_bw =MAX_MBA_BW ; r->membw.arch_needs_linear = true;if (ecx &MBA_IS_LINEAR ) {
@@ -242,7 +242,6 @@ static bool __rdt_get_mem_config_amd(struct rdt_resource *r)cpuid_count (0x80000020 , subleaf, &eax, &ebx, &ecx, &edx); hw_res->num_closid = edx +1 ; - r->default_ctrl =1 << eax; r->schema_fmt =RESCTRL_SCHEMA_MBPS ; r->membw.max_bw =1 << eax;
@@ -274,8 +273,7 @@ static void rdt_get_cache_alloc_cfg(int idx, struct rdt_resource *r)cpuid_count (0x00000010 , idx, &eax.full, &ebx, &ecx.full, &edx.full); hw_res->num_closid = edx.split.cos_max +1 ; r->cache.cbm_len = eax.split.cbm_len +1 ; - r->default_ctrl =BIT_MASK (eax.split.cbm_len +1 ) -1 ; - r->cache.shareable_bits = ebx & r->default_ctrl; + r->cache.shareable_bits = ebx &resctrl_get_default_ctrl (r);if (boot_cpu_data.x86_vendor ==X86_VENDOR_INTEL ) r->cache.arch_has_sparse_bitmasks = ecx.split.noncont; r->alloc_capable = true;
@@ -322,7 +320,7 @@ static u32 delay_bw_map(unsigned long bw, struct rdt_resource *r)return MAX_MBA_BW - bw;pr_warn_once ("Non Linear delay-bw map not supported but queried\n" ); -return r->default_ctrl; +return resctrl_get_default_ctrl (r); }static void mba_wrmsr_intel (struct msr_param *m)
@@ -419,7 +417,7 @@ static void setup_default_ctrlval(struct rdt_resource *r, u32 *dc)* For Memory Allocation: Set b/w requested to 100% */ for (i =0 ; i < hw_res->num_closid; i++, dc++) - *dc = r->default_ctrl; + *dc =resctrl_get_default_ctrl (r); }static void domain_free (struct rdt_hw_domain *hw_dom)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +3 -3
@@ -970,7 +970,7 @@ static int rdt_default_ctrl_show(struct kernfs_open_file *of,struct resctrl_schema *s = of->kn->parent->priv;struct rdt_resource *r = s->res; -seq_printf (seq,"%x\n" , r->default_ctrl); +seq_printf (seq,"%x\n" ,resctrl_get_default_ctrl (r));return 0 ; }
@@ -2869,7 +2869,7 @@ static int reset_all_ctrls(struct rdt_resource *r) hw_dom =resctrl_to_arch_dom (d);for (i =0 ; i < hw_res->num_closid; i++) - hw_dom->ctrl_val[i] = r->default_ctrl; + hw_dom->ctrl_val[i] =resctrl_get_default_ctrl (r); msr_param.dom = d;smp_call_function_any (&d->cpu_mask, rdt_ctrl_update, &msr_param,1 ); }
@@ -3340,7 +3340,7 @@ static void rdtgroup_init_mba(struct rdt_resource *r, u32 closid) } cfg = &d->staged_config[CDP_NONE ]; - cfg->new_ctrl = r->default_ctrl; + cfg->new_ctrl =resctrl_get_default_ctrl (r); cfg->have_new_ctrl = true; } }
include/linux/resctrl.h +19 -2
@@ -184,7 +184,6 @@ enum resctrl_schema_fmt {* @membw: If the component has bandwidth controls, their properties. * @domains: RCU list of all domains for this resource * @name: Name to use in "schemata" file. -* @default_ctrl: Specifies default cache cbm or memory B/W percent. * @schema_fmt: Which format string and parser is used for this schema. * @evt_list: List of monitoring events * @cdp_capable: Is the CDP feature available on this resource
@@ -199,7 +198,6 @@ struct rdt_resource {struct resctrl_membw membw;struct list_head domains;char *name; -u32 default_ctrl;enum resctrl_schema_fmt schema_fmt;struct list_head evt_list;bool cdp_capable;
@@ -236,6 +234,25 @@ struct resctrl_schema {u32 num_closid; }; +/** +* resctrl_get_default_ctrl() - Return the default control value for this +* resource. +* @r: The resource whose default control type is queried. +*/ +static inline u32 resctrl_get_default_ctrl (struct rdt_resource *r) +{ +switch (r->schema_fmt) { +case RESCTRL_SCHEMA_BITMAP : +return BIT_MASK (r->cache.cbm_len) -1 ; +case RESCTRL_SCHEMA_PERCENTAGE : +return 100u ; +case RESCTRL_SCHEMA_MBPS : +return r->membw.max_bw; + } + +return WARN_ON_ONCE (1 ); +} +/* The number of closid supported by this resource regardless of CDP */ u32 resctrl_arch_get_num_closid (struct rdt_resource *r);int resctrl_arch_update_domains (struct rdt_resource *r,u32 closid);
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 9/38 x86/resctrl: Add helper for setting CPU default properties
In reply to James Morserdtgroup_rmdir_ctrl() and rdtgroup_rmdir_mon() set the per-CPU pqr_state for CPUs that were part of the rmdir()'d group.
Another architecture might not have a 'pqr_state', its hardware may need the values in a different format. MPAM's equivalent of RMID values are not unique, and always need the CLOSID to be provided too.
There is only one caller that modifies a single value, (rdtgroup_rmdir_mon()). MPAM always needs both CLOSID and RMID for the hardware value as these are written to the same system register.
As rdtgroup_rmdir_mon() has the CLOSID on hand, only provide a helper to set both values. These values are read by __resctrl_sched_in(), but may be written by a different CPU without any locking, add READ/WRTE_ONCE() to avoid torn values.
---
Changes since v1:
* In rdtgroup_rmdir_mon(), (re)set CPU default closid based on the parent control group, to avoid the appearance of referencing something that we're in the process of destroying (even if it doesn't make a difference because the victim mon group necessarily has the same closid as the parent control group).
Update comment to match.
No (intentional) functional change.
--- arch/x86/include/asm/resctrl.h | 14 +++++++++++--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 20 ++++++++++++++------ 2 files changed, 25 insertions(+), 9 deletions(-)
arch/x86/include/asm/resctrl.h +11 -3
@@ -4,8 +4,9 @@#ifdef CONFIG_X86_CPU_RESCTRL -#include <linux/sched.h> #include <linux/jump_label.h> +#include <linux/percpu.h> +#include <linux/sched.h> /* * This value can never be a valid CLOSID, and is used when mapping a
@@ -96,8 +97,8 @@ static inline void resctrl_arch_disable_mon(void)static inline void __resctrl_sched_in (struct task_struct *tsk) {struct resctrl_pqr_state *state =this_cpu_ptr (&pqr_state); -u32 closid = state->default_closid; -u32 rmid = state->default_rmid; +u32 closid =READ_ONCE (state->default_closid); +u32 rmid =READ_ONCE (state->default_rmid);u32 tmp;/*
@@ -132,6 +133,13 @@ static inline unsigned int resctrl_arch_round_mon_val(unsigned int val)return val * scale; } +static inline void resctrl_arch_set_cpu_default_closid_rmid (int cpu,u32 closid, +u32 rmid) +{ +WRITE_ONCE (per_cpu (pqr_state.default_closid, cpu), closid); +WRITE_ONCE (per_cpu (pqr_state.default_rmid, cpu), rmid); +} +static inline void resctrl_arch_set_closid_rmid (struct task_struct *tsk,u32 closid,u32 rmid) {
arch/x86/kernel/cpu/resctrl/rdtgroup.c +14 -6
@@ -3652,14 +3652,21 @@ static int rdtgroup_mkdir(struct kernfs_node *parent_kn, const char *name,static int rdtgroup_rmdir_mon (struct rdtgroup *rdtgrp,cpumask_var_t tmpmask) {struct rdtgroup *prdtgrp = rdtgrp->mon.parent; +u32 closid, rmid;int cpu;/* Give any tasks back to the parent group */ rdt_move_group_tasks (rdtgrp, prdtgrp, tmpmask); -/* Update per cpu rmid of the moved CPUs first */ +/* +* Update per cpu closid/rmid of the moved CPUs first. +* Note: the closid will not change, but the arch code still needs it. +*/ + closid = prdtgrp->closid; + rmid = prdtgrp->mon.rmid;for_each_cpu (cpu, &rdtgrp->cpu_mask) -per_cpu (pqr_state.default_rmid, cpu) = prdtgrp->mon.rmid; +resctrl_arch_set_cpu_default_closid_rmid (cpu, closid, rmid); +/* * Update the MSR on moved CPUs and CPUs which have moved * task running on them.
@@ -3692,6 +3699,7 @@ static int rdtgroup_ctrl_remove(struct rdtgroup *rdtgrp)static int rdtgroup_rmdir_ctrl (struct rdtgroup *rdtgrp,cpumask_var_t tmpmask) { +u32 closid, rmid;int cpu;/* Give any tasks back to the default group */
@@ -3702,10 +3710,10 @@ static int rdtgroup_rmdir_ctrl(struct rdtgroup *rdtgrp, cpumask_var_t tmpmask) &rdtgroup_default.cpu_mask, &rdtgrp->cpu_mask);/* Update per cpu closid and rmid of the moved CPUs first */ -for_each_cpu (cpu, &rdtgrp->cpu_mask) { -per_cpu (pqr_state.default_closid, cpu) = rdtgroup_default.closid; -per_cpu (pqr_state.default_rmid, cpu) = rdtgroup_default.mon.rmid; - } + closid = rdtgroup_default.closid; + rmid = rdtgroup_default.mon.rmid; +for_each_cpu (cpu, &rdtgrp->cpu_mask) +resctrl_arch_set_cpu_default_closid_rmid (cpu, closid, rmid);/* * Update the MSR on moved CPUs and CPUs which have moved
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 10/38 x86/resctrl: Remove rdtgroup from update_cpu_closid_rmid()
In reply to James Morseupdate_cpu_closid_rmid() takes a struct rdtgroup as an argument, which it uses to update the local CPUs default pqr values. This is a problem once the resctrl parts move out to /fs/, as the arch code cannot poke around inside struct rdtgroup.
Rename update_cpu_closid_rmid() as resctrl_arch_sync_cpu_closid_rmid() to be used as the target of an IPI, and pass the effective CLOSID and RMID in a new struct.
---
Changes since v2:
* Change the function name in the commit message to match.
Changes since v1:
* To clarify the meanings of the new helper and struct:
Rename resctrl_arch_sync_cpu_default() to resctrl_arch_sync_cpu_closid_rmid();
Rename struct resctrl_cpu_sync to struct resctrl_cpu_defaults;
Flesh out the comment block in <linux/resctrl.h>.
No functional change.
--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 17 +++++++++++++---- include/linux/resctrl.h | 22 ++++++++++++++++++++++ 2 files changed, 35 insertions(+), 4 deletions(-)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +13 -4
@@ -347,13 +347,13 @@ static int rdtgroup_cpus_show(struct kernfs_open_file *of,* from update_closid_rmid() is protected against __switch_to() because * preemption is disabled. */ -static void update_cpu_closid_rmid (void *info) +void resctrl_arch_sync_cpu_closid_rmid (void *info) { -struct rdtgroup *r = info; +struct resctrl_cpu_defaults *r = info;if (r) {this_cpu_write (pqr_state.default_closid, r->closid); -this_cpu_write (pqr_state.default_rmid, r->mon.rmid); +this_cpu_write (pqr_state.default_rmid, r->rmid); }/*
@@ -368,11 +368,20 @@ static void update_cpu_closid_rmid(void *info)* Update the PGR_ASSOC MSR on all cpus in @cpu_mask, * * Per task closids/rmids must have been set up before calling this function. +* @r may be NULL. */ static void update_closid_rmid (const struct cpumask *cpu_mask,struct rdtgroup *r) { -on_each_cpu_mask (cpu_mask, update_cpu_closid_rmid, r,1 ); +struct resctrl_cpu_defaults defaults, *p =NULL ; + +if (r) { + defaults.closid = r->closid; + defaults.rmid = r->mon.rmid; + p = &defaults; + } + +on_each_cpu_mask (cpu_mask, resctrl_arch_sync_cpu_closid_rmid, p,1 ); } static int cpus_mon_write(struct rdtgroup *rdtgrp, cpumask_var_t newmask,
include/linux/resctrl.h +22 -0
@@ -234,6 +234,28 @@ struct resctrl_schema {u32 num_closid; }; +struct resctrl_cpu_defaults { +u32 closid; +u32 rmid; +}; + +/** +* resctrl_arch_sync_cpu_closid_rmid() - Refresh this CPU's CLOSID and RMID. +* Call via IPI. +* @info: If non-NULL, a pointer to a struct resctrl_cpu_defaults +* specifying the new CLOSID and RMID for tasks in the default +* resctrl ctrl and mon group when running on this CPU. If NULL, +* this CPU is not re-assigned to a different default group. +* +* Propagates reassignment of CPUs and/or tasks to different resctrl groups +* when requested by the resctrl core code. +* +* This function records the per-cpu defaults specified by @info (if any), +* and then reconfigures the CPU's hardware CLOSID and RMID for subsequent +* execution based on @current, in the same way as during a task switch. +*/ +void resctrl_arch_sync_cpu_closid_rmid (void *info); +/** * resctrl_get_default_ctrl() - Return the default control value for this * resource.
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 11/38 x86/resctrl: Export resctrl fs's init function
In reply to James Morserdtgroup_init() needs exporting so that arch code can call it once it lives in core code. As this is one of the few functions exported, rename it to have "resctrl" in the name. The same goes for the exit call.
x86's arch code init functions for RDT are renamed to have an arch prefix to make it clear these are part of the architecture code.
---
Changes since v1:
* Rename stale rdtgroup_init() to resctrl_init() in arch/x86/kernel/cpu/resctrl/monitor.c comments.
No functional change.
* [Commit message only] Minor rewording to avoid "impersonating code".
* [Commit message only] Typo fix:
s/to have the resctrl/to have resctrl/ in commit message.
--- arch/x86/kernel/cpu/resctrl/core.c | 12 ++++++------ arch/x86/kernel/cpu/resctrl/internal.h | 3 --- arch/x86/kernel/cpu/resctrl/monitor.c | 2 +- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 8 ++++---- include/linux/resctrl.h | 3 +++ 5 files changed, 14 insertions(+), 14 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +6 -6
@@ -902,7 +902,7 @@ void resctrl_cpu_detect(struct cpuinfo_x86 *c) } } -static int __init resctrl_late_init (void ) +static int __init resctrl_arch_late_init (void ) {struct rdt_resource *r;int state, ret;
@@ -925,7 +925,7 @@ static int __init resctrl_late_init(void)if (state <0 )return state; - ret =rdtgroup_init (); + ret =resctrl_init ();if (ret) {cpuhp_remove_state (state);return ret;
@@ -941,18 +941,18 @@ static int __init resctrl_late_init(void)return 0 ; } -late_initcall (resctrl_late_init); +late_initcall (resctrl_arch_late_init); -static void __exit resctrl_exit(void) +static void __exit resctrl_arch_exit(void) {struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_L3 ].r_resctrl;cpuhp_remove_state (rdt_online); -rdtgroup_exit (); +resctrl_exit ();if (r->mon_capable)rdt_put_mon_l3_config (); } -__exitcall (resctrl_exit); +__exitcall (resctrl_arch_exit);
arch/x86/kernel/cpu/resctrl/internal.h +0 -3
@@ -301,9 +301,6 @@ extern struct list_head rdt_all_groups;extern int max_name_width, max_data_width; -int __init rdtgroup_init (void ); -void __exit rdtgroup_exit(void); -/** * struct rftype - describe each file in the resctrl file system * @name: File name
arch/x86/kernel/cpu/resctrl/monitor.c +1 -1
@@ -952,7 +952,7 @@ static int dom_data_init(struct rdt_resource *r)/* * RESCTRL_RESERVED_CLOSID and RESCTRL_RESERVED_RMID are special and * are always allocated. These are used for the rdtgroup_default -* control group, which will be setup later in rdtgroup_init(). +* control group, which will be setup later in resctrl_init(). */ idx =resctrl_arch_rmid_idx_encode (RESCTRL_RESERVED_CLOSID ,RESCTRL_RESERVED_RMID );
arch/x86/kernel/cpu/resctrl/rdtgroup.c +4 -4
@@ -4146,14 +4146,14 @@ void resctrl_offline_cpu(unsigned int cpu) }/* -* rdtgroup_init - rdtgroup initialization +* resctrl_init - resctrl filesystem initialization * * Setup resctrl file system including set up root, create mount point, -* register rdtgroup filesystem, and initialize files under root directory. +* register resctrl filesystem, and initialize files under root directory. * * Return: 0 on success or -errno */ -int __init rdtgroup_init (void ) +int __init resctrl_init (void ) {int ret =0 ;
@@ -4201,7 +4201,7 @@ int __init rdtgroup_init(void)return ret; } -void __exit rdtgroup_exit(void) +void __exit resctrl_exit(void) {debugfs_remove_recursive (debugfs_resctrl);unregister_filesystem (&rdt_fs_type);
include/linux/resctrl.h +3 -0
@@ -369,4 +369,7 @@ void resctrl_arch_reset_rmid_all(struct rdt_resource *r, struct rdt_domain *d);extern unsigned int resctrl_rmid_realloc_threshold;extern unsigned int resctrl_rmid_realloc_limit; +int __init resctrl_init (void ); +void __exit resctrl_exit(void); +#endif /* _RESCTRL_H */
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 12/38 x86/resctrl: Wrap resctrl_arch_find_domain() around rdt_find_domain()
In reply to James Morserdt_find_domain() finds a domain given a resource and a cache-id. It's not quite right for the resctrl arch API as it also returns the position to insert a new domain, which is needed when bringing a domain online in the arch code.
Wrap rdt_find_domain() in another function resctrl_arch_find_domain() in order to avoid the unnecessary argument outside the arch code.
---
Changes since v1:
* [Commit message only] Minor rewording to avoid "impersonating code".
* [Commit message only] Typo fix:
s/in a another/in another/ in commit message.
--- arch/x86/kernel/cpu/resctrl/core.c | 9 +++++++-- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 2 +- arch/x86/kernel/cpu/resctrl/internal.h | 2 -- include/linux/resctrl.h | 2 ++ 4 files changed, 10 insertions(+), 5 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +7 -2
@@ -381,8 +381,8 @@ void rdt_ctrl_update(void *arg)* caller, return the first domain whose id is bigger than the input id. * The domain list is sorted by id in ascending order. */ -struct rdt_domain *rdt_find_domain (struct rdt_resource *r,int id, -struct list_head **pos) +static struct rdt_domain *rdt_find_domain (struct rdt_resource *r,int id, +struct list_head **pos) {struct rdt_domain *d;struct list_head *l;
@@ -406,6 +406,11 @@ struct rdt_domain *rdt_find_domain(struct rdt_resource *r, int id,return NULL ; } +struct rdt_domain *resctrl_arch_find_domain (struct rdt_resource *r,int id) +{ +return rdt_find_domain (r, id,NULL ); +} +static void setup_default_ctrlval (struct rdt_resource *r,u32 *dc) {struct rdt_hw_resource *hw_res =resctrl_to_arch_res (r);
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +1 -1
@@ -600,7 +600,7 @@ int rdtgroup_mondata_show(struct seq_file *m, void *arg) evtid = md.u.evtid; r =resctrl_arch_get_resource (resid); - d =rdt_find_domain (r, domid,NULL ); + d =resctrl_arch_find_domain (r, domid);if (IS_ERR_OR_NULL (d)) { ret = -ENOENT ; goto out;
arch/x86/kernel/cpu/resctrl/internal.h +0 -2
@@ -535,8 +535,6 @@ void rdtgroup_kn_unlock(struct kernfs_node *kn);int rdtgroup_kn_mode_restrict (struct rdtgroup *r,const char *name);int rdtgroup_kn_mode_restore (struct rdtgroup *r,const char *name,umode_t mask); -struct rdt_domain *rdt_find_domain (struct rdt_resource *r,int id, -struct list_head **pos);ssize_t rdtgroup_schemata_write (struct kernfs_open_file *of,char *buf,size_t nbytes,loff_t off); int rdtgroup_schemata_show(struct kernfs_open_file *of,
include/linux/resctrl.h +2 -0
@@ -277,6 +277,8 @@ static inline u32 resctrl_get_default_ctrl(struct rdt_resource *r)/* The number of closid supported by this resource regardless of CDP */ u32 resctrl_arch_get_num_closid (struct rdt_resource *r); + +struct rdt_domain *resctrl_arch_find_domain (struct rdt_resource *r,int id);int resctrl_arch_update_domains (struct rdt_resource *r,u32 closid);/*
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 13/38 x86/resctrl: Move resctrl types to a separate header
In reply to James MorseWhen resctrl is fully factored into core and per-arch code, each arch will need to use some resctrl common definitions in order to define its own specializations and helpers. Following conventional practice, it would be desirable to put the dependent arch definitions in an <asm/resctrl.h> header that is included by the common <linux/resctrl.h> header. However, this can make it awkward to avoid a circular dependency between <linux/resctrl.h> and the arch header.
To avoid such dependencies, move the affected common types and constants into a new header that does not need to depend on <linux/resctrl.h> or on the arch headers.
The same logic applies to the monitor-configuration defines, move these too.
Some kind of enumeration for events is needed between the filesystem and architecture code. Take the x86 definition as its convenient for x86.
The definition of enum resctrl_event_id is need to allow the architecture code to define resctrl_arch_event_is_free_running(), resctrl_arch_set_cdp_enabled(), resctrl_arch_mon_ctx_alloc() and resctrl_arch_mon_ctx_free().
The definition of enum resctrl_res_level is needed to allow the architecture code to define resctrl_arch_set_cdp_enabled() and resctrl_arch_get_cdp_enabled().
The bits for mbm_local_bytes_config et al are ABI, and must be the same on all architectures. These are documented in Documentation/arch/x86/resctrl.rst
The maintainers entry for these headers was missed when resctrl.h was created. Add a wildcard entry to match both resctrl.h and resctrl_types.h.
---
Changes since v2:
* Added to the commit message why each of these things is necessary.
* Moved the enum resctrl_conf_type back to resctrl.h - this week arm's CDP emulation code gets away without this...
Changes since v1:
* [Commit message only] Rewrite commit message to clarify the the rationale for refactoring the headers in this way.
--- MAINTAINERS | 1 + arch/x86/kernel/cpu/resctrl/internal.h | 24 ------------ include/linux/resctrl.h | 21 +--------- include/linux/resctrl_types.h | 54 ++++++++++++++++++++++++++ 4 files changed, 56 insertions(+), 44 deletions(-) create mode 100644 include/linux/resctrl_types.h
MAINTAINERS +1 -0
@@ -18859,6 +18859,7 @@ S: Supported F: Documentation/arch/x86/resctrl* F: arch/x86/include/asm/resctrl.h F: arch/x86/kernel/cpu/resctrl/ +F: include/linux/resctrl*.h F: tools/testing/selftests/resctrl/ READ-COPY UPDATE (RCU)
arch/x86/kernel/cpu/resctrl/internal.h +0 -24
@@ -32,30 +32,6 @@*/ #define MBM_CNTR_WIDTH_OFFSET_MAX (62 - MBM_CNTR_WIDTH_BASE) -/* Reads to Local DRAM Memory */ -#define READS_TO_LOCAL_MEM BIT(0) - -/* Reads to Remote DRAM Memory */ -#define READS_TO_REMOTE_MEM BIT(1) - -/* Non-Temporal Writes to Local Memory */ -#define NON_TEMP_WRITE_TO_LOCAL_MEM BIT(2) - -/* Non-Temporal Writes to Remote Memory */ -#define NON_TEMP_WRITE_TO_REMOTE_MEM BIT(3) - -/* Reads to Local Memory the system identifies as "Slow Memory" */ -#define READS_TO_LOCAL_S_MEM BIT(4) - -/* Reads to Remote Memory the system identifies as "Slow Memory" */ -#define READS_TO_REMOTE_S_MEM BIT(5) - -/* Dirty Victims to All Types of Memory */ -#define DIRTY_VICTIMS_TO_ALL_MEM BIT(6) - -/* Max event bits supported */ -#define MAX_EVT_CONFIG_BITS GENMASK(6, 0) -/** * cpumask_any_housekeeping() - Choose any CPU in @mask, preferring those that * aren't marked nohz_full
include/linux/resctrl.h +1 -20
@@ -5,6 +5,7 @@#include <linux/kernel.h> #include <linux/list.h> #include <linux/pid.h> +#include <linux/resctrl_types.h> /* CLOSID, RMID value used by the default control group */ #define RESCTRL_RESERVED_CLOSID 0
@@ -36,28 +37,8 @@ enum resctrl_conf_type { CDP_DATA, }; -enum resctrl_res_level { -RDT_RESOURCE_L3 , -RDT_RESOURCE_L2 , -RDT_RESOURCE_MBA , -RDT_RESOURCE_SMBA , - -/* Must be the last */ -RDT_NUM_RESOURCES , -}; -#define CDP_NUM_TYPES (CDP_DATA + 1) -/* -* Event IDs, the values match those used to program IA32_QM_EVTSEL before -* reading IA32_QM_CTR on RDT systems. -*/ -enum resctrl_event_id { -QOS_L3_OCCUP_EVENT_ID =0x01 , -QOS_L3_MBM_TOTAL_EVENT_ID =0x02 , -QOS_L3_MBM_LOCAL_EVENT_ID =0x03 , -}; -/** * struct resctrl_staged_config - parsed configuration to be applied * @new_ctrl: new ctrl value to be loaded
include/linux/resctrl_types.h new file +54 -0
54 lines not on this page · raw message
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 13/38] x86/resctrl: Move resctrl types to a separate header
Hi James,
Quoted 12 lines · James Morse
When resctrl is fully factored into core and per-arch code, each arch will need to use some resctrl common definitions in order to define its own specializations and helpers. Following conventional practice, it would be desirable to put the dependent arch definitions in an <
asm/resctrl.h> header that is included by the common <linux/resctrl.h> header. However, this can make it awkward to avoid a circular dependency between <linux/resctrl.h> and the arch header.To avoid such dependencies, move the affected common types and constants into a new header that does not need to depend on <
linux/resctrl.h> or on the arch headers.
Quoted 8 lines
The same logic applies to the monitor-configuration defines, move these too.
Some kind of enumeration for events is needed between the filesystem and architecture code. Take the x86 definition as its convenient for x86.
The definition of enum
resctrl_event_idis need to allow the architecture
"is need" -> "is needed" ?
Quoted 1 line
code to define
resctrl_arch_event_is_free_running(),
Cannot find resctrl_arch_event_is_free_running()
Quoted 1 line
resctrl_arch_set_cdp_enabled(),resctrl_arch_mon_ctx_alloc()and
resctrl_arch_set_cdp_enabled() should not need enum resctrl_event_id
Quoted 23 lines
resctrl_arch_mon_ctx_free().The definition of enum
resctrl_res_levelis needed to allow the architecture code to defineresctrl_arch_set_cdp_enabled()andresctrl_arch_get_cdp_enabled().The bits for
mbm_local_bytes_configet al are ABI, and must be the same on all architectures. These are documented inDocumentation/arch/x86/resctrl.rstThe maintainers entry for these headers was missed when
resctrl.hwas created. Add a wildcard entry to match bothresctrl.handresctrl_types.h.Signed-off-by James Morse <[email protected]>Tested-by Peter Newman <[email protected]>Tested-by Shaopeng Tan <[email protected]>---
Changes since v2:
* Added to the commit message why each of these things is necessary.
* Moved the enumresctrl_conf_typeback toresctrl.h- this week arm's CDP emulation code gets away without this...
Quoted 8 lines
Changes since v1:
* [Commit message only] Rewrite commit message to clarify the the rationale for refactoring the headers in this way.--- MAINTAINERS | 1 + arch/x86/kernel/cpu/resctrl/internal.h | 24 ------------ include/linux/resctrl.h | 21 +--------- include/linux/resctrl_types.h | 54 ++++++++++++++++++++++++++
Considering the motivation I also expected to see a change in arch/x86/include/asm/resctrl.h that adds the #include of the new file.
Reinette
James Morse
reviewerRe: [PATCH v3 13/38] x86/resctrl: Move resctrl types to a separate header
Hi Reinette,
Quoted 20 lines · Reinette Chatre
Quoted 19 lines · James Morse
When resctrl is fully factored into core and per-arch code, each arch will need to use some resctrl common definitions in order to define its own specializations and helpers. Following conventional practice, it would be desirable to put the dependent arch definitions in an <
asm/resctrl.h> header that is included by the common <linux/resctrl.h> header. However, this can make it awkward to avoid a circular dependency between <linux/resctrl.h> and the arch header.To avoid such dependencies, move the affected common types and constants into a new header that does not need to depend on <
linux/resctrl.h> or on the arch headers.The same logic applies to the monitor-configuration defines, move these too.
Some kind of enumeration for events is needed between the filesystem and architecture code. Take the x86 definition as its convenient for x86.
Quoted 8 lines
Quoted 1 line
The definition of enum
resctrl_event_idis need to allow the architecture"is need" -> "is needed" ?
Quoted 1 line
code to define
resctrl_arch_event_is_free_running(),Cannot find
resctrl_arch_event_is_free_running()
Sorry - this will show up after the MPAM driver:
{
MPAM has an additional piece of hardware that needs to be allocated to read the memory bandwidth counters. resctrl expects these things to be pre-allocated and free running from the start of time.
I have some patches to explicitly tell resctrl this, so that the resctrl interface to these things can be used by perf to query the 'mbm' counters, even if the files are not exposed.
}
Quoted 3 lines
Quoted 1 line
resctrl_arch_set_cdp_enabled(),resctrl_arch_mon_ctx_alloc()and
resctrl_arch_set_cdp_enabled()should not need enumresctrl_event_id
Sorry, wrong list.
Quoted 14 lines
Quoted 13 lines
resctrl_arch_mon_ctx_free().The definition of enum
resctrl_res_levelis needed to allow the architecture code to defineresctrl_arch_set_cdp_enabled()andresctrl_arch_get_cdp_enabled().The bits for
mbm_local_bytes_configet al are ABI, and must be the same on all architectures. These are documented inDocumentation/arch/x86/resctrl.rstThe maintainers entry for these headers was missed when
resctrl.hwas created. Add a wildcard entry to match bothresctrl.handresctrl_types.h.
Quoted 8 lines
Quoted 5 lines
---
MAINTAINERS | 1 +arch/x86/kernel/cpu/resctrl/internal.h| 24 ------------ include/linux/resctrl.h | 21 +--------- include/linux/resctrl_types.h | 54 ++++++++++++++++++++++++++Considering the motivation I also expected to see a change in
arch/x86/include/asm/resctrl.hthat adds the #include of the new file.
It gets added in a later patch - but I'll move it here.
Thanks,
James
James Morse
reviewerPATCH v3 14/38 x86/resctrl: Add a resctrl helper to reset all the resources
In reply to James MorseOn umount(), resctrl resets each resource back to its default configuration. It only ever does this for all resources in one go.
reset_all_ctrls() is architecture specific as it works with struct rdt_hw_resource.
Add an architecture helper to reset all resources.
---
Changes since v1:
* Rename the for_each_capable_rdt_resource() introduced in the new function resctrl_arch_reset_resources(), back to for_each_alloc_capable_rdt_resource() as it was in the original code.
The change looked unintentional; and presumably a resource that does not support resource allocation doesn't have any properties to reset...
--- arch/x86/include/asm/resctrl.h | 2 ++ arch/x86/kernel/cpu/resctrl/rdtgroup.c | 16 +++++++++++----- 2 files changed, 13 insertions(+), 5 deletions(-)
arch/x86/include/asm/resctrl.h +2 -0
@@ -15,6 +15,8 @@*/ #define X86_RESCTRL_EMPTY_CLOSID ((u32)~0) +void resctrl_arch_reset_resources (void ); +/** * struct resctrl_pqr_state - State cache for the PQR MSR * @cur_rmid: The cached Resource Monitoring ID
arch/x86/kernel/cpu/resctrl/rdtgroup.c +11 -5
@@ -2886,6 +2886,14 @@ static int reset_all_ctrls(struct rdt_resource *r)return 0 ; } +void resctrl_arch_reset_resources (void ) +{ +struct rdt_resource *r; + +for_each_alloc_capable_rdt_resource (r) +reset_all_ctrls (r); +} +/* * Move tasks from one to the other group. If @from is NULL, then all tasks * in the systems are moved unconditionally (used for teardown).
@@ -2995,16 +3003,14 @@ static void rmdir_all_sub(void)static void rdt_kill_sb (struct super_block *sb) { -struct rdt_resource *r; -cpus_read_lock ();mutex_lock (&rdtgroup_mutex);rdt_disable_ctx (); -/*Put everything back to default values. */ -for_each_alloc_capable_rdt_resource (r) -reset_all_ctrls (r); +/* Put everything back to default values. */ +resctrl_arch_reset_resources (); +rmdir_all_sub ();rdt_pseudo_lock_release (); rdtgroup_default.mode =RDT_MODE_SHAREABLE ;
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 15/38 x86/resctrl: Move monitor exit work to a restrl exit call
In reply to James Morserdt_put_mon_l3_config() is called via the architecture's resctrl_arch_exit() call, and appears to free the rmid_ptrs[] and closid_num_dirty_rmid[] arrays. In reality this code is marked __exit, and is removed by the linker as resctl can't be built as a module.
To separate the filesystem and architecture parts of resctrl, this free()ing work needs to be triggered by the filesystem, as these structures belong to the filesystem code.
Rename rdt_put_mon_l3_config() resctrl_mon_resource_exit() and call it from resctrl_exit(). The kfree() is currently dependent on r->mon_capable. resctrl_mon_resource_init() takes no arguments, so resctrl_mon_resource_exit() shouldn't take any either. Add the check to dom_data_exit(), making it take the resource as an argument. This makes it more symmetrical with dom_data_init().
---
Changes since v2:
* Dropped __exit as needed in the next patch.
Change since v1:
* [Commit message only] Typo fixes:
s/restrl/resctrl/g
s/resctl/resctrl/g
* [Commit message only] Reword second paragraph to remove reference to the MPAM error interrupt, which provides background rationale for a later patch rather than for this patch, and so it is not really relevant here.
--- arch/x86/kernel/cpu/resctrl/core.c | 5 ----- arch/x86/kernel/cpu/resctrl/internal.h | 2 +- arch/x86/kernel/cpu/resctrl/monitor.c | 12 ++++++++---- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 2 ++ 4 files changed, 11 insertions(+), 10 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +0 -5
@@ -950,14 +950,9 @@ late_initcall(resctrl_arch_late_init); static void __exit resctrl_arch_exit(void) { -struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_L3 ].r_resctrl; -cpuhp_remove_state (rdt_online);resctrl_exit (); - -if (r->mon_capable) -rdt_put_mon_l3_config (); }__exitcall (resctrl_arch_exit);
arch/x86/kernel/cpu/resctrl/internal.h +1 -1
@@ -535,7 +535,7 @@ void closid_free(int closid);int alloc_rmid (u32 closid);void free_rmid (u32 closid,u32 rmid);int rdt_get_mon_l3_config (struct rdt_resource *r); -void __exit rdt_put_mon_l3_config(void); +void resctrl_mon_resource_exit (void );bool __init rdt_cpu_has (int flag);void mon_event_count (void *info);int rdtgroup_mondata_show (struct seq_file *m,void *arg);
arch/x86/kernel/cpu/resctrl/monitor.c +8 -4
@@ -965,10 +965,12 @@ static int dom_data_init(struct rdt_resource *r)return err; } -static void __exit dom_data_exit(void) +static void dom_data_exit (struct rdt_resource *r) { -mutex_lock (&rdtgroup_mutex); +if (!r->mon_capable) +return ; +mutex_lock (&rdtgroup_mutex);if (IS_ENABLED (CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID )) {kfree (closid_num_dirty_rmid); closid_num_dirty_rmid =NULL ;
@@ -1075,9 +1077,11 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)return 0 ; } -void __exit rdt_put_mon_l3_config(void) +void resctrl_mon_resource_exit (void ) { -dom_data_exit (); +struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_L3 ); + +dom_data_exit (r); }void __init intel_rdt_mbm_apply_quirk (void )
arch/x86/kernel/cpu/resctrl/rdtgroup.c +2 -0
@@ -4212,4 +4212,6 @@ void __exit resctrl_exit(void)debugfs_remove_recursive (debugfs_resctrl);unregister_filesystem (&rdt_fs_type);sysfs_remove_mount_point (fs_kobj,"resctrl" ); + +resctrl_mon_resource_exit (); }
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 15/38] x86/resctrl: Move monitor exit work to a restrl exit call
Hi James,
Quoted 6 lines · James Morse
rdt_put_mon_l3_config()is called via the architecture'sresctrl_arch_exit()call, and appears to free the rmid_ptrs[] and closid_num_dirty_rmid[] arrays. In reality this code is marked__exit, and is removed by the linker as resctl can't be built as a module.
Quoted 7 lines
To separate the filesystem and architecture parts of resctrl, this free()ing work needs to be triggered by the filesystem, as these structures belong to the filesystem code.
Rename
rdt_put_mon_l3_config()resctrl_mon_resource_exit()and call it fromresctrl_exit(). Thekfree()is currently dependent onr->mon_capable.resctrl_mon_resource_init()
resctrl_mon_resource_init() does not exist at this point making this motivation difficult to follow.
Quoted 8 lines
takes no arguments, so
resctrl_mon_resource_exit()shouldn't take any either. Add the check todom_data_exit(), making it take the resource as an argument. This makes it more symmetrical withdom_data_init().Signed-off-by James Morse <[email protected]>Tested-by Peter Newman <[email protected]>Tested-by Shaopeng Tan <[email protected]>
Quoted 8 lines
---
Changes since v2:
* Dropped__exitas needed in the next patch.Change since v1:
* [Commit message only] Typo fixes:
s/restrl/resctrl/g
s/resctl/resctrl/g
Something went wrong here since the subject and changelog still contains the terms that were intended to be replaced.
Quoted 50 lines
* [Commit message only] Reword second paragraph to remove reference to the MPAM error interrupt, which provides background rationale for a later patch rather than for this patch, and so it is not really relevant here.
--- arch/x86/kernel/cpu/resctrl/core.c | 5 ----- arch/x86/kernel/cpu/resctrl/internal.h | 2 +- arch/x86/kernel/cpu/resctrl/monitor.c | 12 ++++++++---- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 2 ++ 4 files changed, 11 insertions(+), 10 deletions(-)arch/x86/kernel/cpu/resctrl/core.c +0
-5@@ -950,14 +950,9 @@ late_initcall(resctrl_arch_late_init); static void __exit resctrl_arch_exit(void) { -struct rdt_resource *r = &rdt_resources_all[RDT_RESOURCE_L3 ].r_resctrl; -cpuhp_remove_state (rdt_online);resctrl_exit (); - -if (r->mon_capable) -rdt_put_mon_l3_config (); }__exitcall (resctrl_arch_exit);arch/x86/kernel/cpu/resctrl/internal.h +1
-1@@ -535,7 +535,7 @@ void closid_free(int closid);int alloc_rmid (u32 closid);void free_rmid (u32 closid,u32 rmid);int rdt_get_mon_l3_config (struct rdt_resource *r); -void __exit rdt_put_mon_l3_config(void); +void resctrl_mon_resource_exit (void );bool __init rdt_cpu_has (int flag);void mon_event_count (void *info);int rdtgroup_mondata_show (struct seq_file *m,void *arg);arch/x86/kernel/cpu/resctrl/monitor.c +0
-0@@ -965,10 +965,12 @@ static int dom_data_init(struct rdt_resource *r)return err;
arch/x86/kernel/cpu/resctrl/monitor.c static int dom_data_init(struct rdt_resource *r) · line 973
Quoted 10 lines
arch/x86/kernel/cpu/resctrl/monitor.c +4
-2} -static void __exit dom_data_exit(void) +static void dom_data_exit (struct rdt_resource *r) { -mutex_lock (&rdtgroup_mutex); +if (!r->mon_capable) +return ; +mutex_lock (&rdtgroup_mutex);
I know there has been a bit of back&forth on whether the mutex is needed here. With this change moving dom_data_exit() out from __exit I think the locking should aim to be consistent with existing runtime and thus the check of r->mon_capable should be with mutex held. Having this little snippet outside mutex will just cause confusion. Do you have motivation for needing this be done outside of mutex? I think it ended up this way with this patch aiming to keep existing flow exactly, but that ended up as convenience in a flow where mutex was not really needed at all.
Reinette
James Morse
reviewerRe: [PATCH v3 15/38] x86/resctrl: Move monitor exit work to a restrl exit call
Hi Reinette,
Quoted 11 lines · Reinette Chatre
Quoted 10 lines · James Morse
rdt_put_mon_l3_config()is called via the architecture'sresctrl_arch_exit()call, and appears to free the rmid_ptrs[] and closid_num_dirty_rmid[] arrays. In reality this code is marked__exit, and is removed by the linker as resctl can't be built as a module.To separate the filesystem and architecture parts of resctrl, this free()ing work needs to be triggered by the filesystem, as these structures belong to the filesystem code.
Quoted 6 lines
Quoted 3 lines
Rename
rdt_put_mon_l3_config()resctrl_mon_resource_exit()and call it fromresctrl_exit(). Thekfree()is currently dependent onr->mon_capable.resctrl_mon_resource_init()
resctrl_mon_resource_init()does not exist at this point making this motivation difficult to follow.
Right - I re-ordered the patches to make the diffs simpler. I'll drop this paragraph.
Quoted 26 lines
Quoted 10 lines
takes no arguments, so
resctrl_mon_resource_exit()shouldn't take any either. Add the check todom_data_exit(), making it take the resource as an argument. This makes it more symmetrical withdom_data_init().arch/x86/kernel/cpu/resctrl/monitor.c +0
-0arch/x86/kernel/cpu/resctrl/monitor.c static int dom_data_init(struct rdt_resource *r) · line 972
Quoted 9 lines
arch/x86/kernel/cpu/resctrl/monitor.c +3
-1@@ -965,10 +965,12 @@ static int dom_data_init(struct rdt_resource *r) return err; } -static void __exit dom_data_exit(void) +static void dom_data_exit(struct rdt_resource *r) { - mutex_lock(&rdtgroup_mutex); + if (!r->mon_capable) + return; + mutex_lock(&rdtgroup_mutex);I know there has been a bit of back&forth on whether the mutex is needed here. With this change moving
dom_data_exit()out from__exitI think the locking should aim to be consistent with existing runtime and thus the check ofr->mon_capableshould be with mutex held. Having this little snippet outside mutex will just cause confusion.
Quoted 1 line
Do you have motivation for needing this be done outside of mutex?
Just to avoid sleeping - then returning having done nothing.
Quoted 4 lines
I think it
ended up this way with this patch aiming to keep existing flow exactly, but that ended up as convenience in a flow where mutex was not really needed at all.
Sure - I don't think it matters either way. Its not a path where performance matters. The property is read-only from when resctrl_init() is called. I'll make it look like resctrl_offline_domain().
Thanks,
James
Re: [PATCH v3 15/38] x86/resctrl: Move monitor exit work to a restrl exit call
In reply to James MorseQuoted 1 line · James Morse
__exit, and is removed by the linker as resctl can't be built
Typo fix: resctl -> resctrl
Also in the commit message subject line:
Typo fix: restrl -> resctrl
Quoted 4 lines
Change since v1:
* [Commit message only] Typo fixes:
s/restrl/resctrl/g
s/resctl/resctrl/g
Oh? I'm not sure how convinced I am. :-)
-Carl
James Morse
reviewerPATCH v3 16/38 x86/resctrl: Move monitor init work to a resctrl init call
In reply to James Morserdt_get_mon_l3_config() is called from the architecture's resctrl_arch_late_init(), and initialises both architecture specific fields, such as hw_res->mon_scale and resctrl filesystem fields by calling dom_data_init().
To separate the filesystem and architecture parts of resctrl, this function needs splitting up.
Add resctrl_mon_resource_init() to do the filesystem specific work, and call it from resctrl_init(). This runs later, but is still before the filesystem is mounted and the rmid_ptrs[] array can be used.
---
Changes since v2:
* Added error handling for the case sysfs files can't be created.
--- arch/x86/kernel/cpu/resctrl/internal.h | 1 + arch/x86/kernel/cpu/resctrl/monitor.c | 24 +++++++++++++++++------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 9 ++++++++- 3 files changed, 26 insertions(+), 8 deletions(-)
arch/x86/kernel/cpu/resctrl/internal.h +1 -0
@@ -542,6 +542,7 @@ int rdtgroup_mondata_show(struct seq_file *m, void *arg);void mon_event_read (struct rmid_read *rr,struct rdt_resource *r,struct rdt_domain *d,struct rdtgroup *rdtgrp,int evtid,int first); +int resctrl_mon_resource_init (void );void mbm_setup_overflow_handler (struct rdt_domain *dom,unsigned long delay_ms,int exclude_cpu);
arch/x86/kernel/cpu/resctrl/monitor.c +17 -7
@@ -1016,12 +1016,28 @@ static void l3_mon_evt_init(struct rdt_resource *r)list_add_tail (&mbm_local_event.list, &r->evt_list); } +int resctrl_mon_resource_init (void ) +{ +struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_L3 ); +int ret; + +if (!r->mon_capable) +return 0 ; + + ret =dom_data_init (r); +if (ret) +return ret; + +l3_mon_evt_init (r); + +return 0 ; +} +int __init rdt_get_mon_l3_config (struct rdt_resource *r) {unsigned int mbm_offset = boot_cpu_data.x86_cache_mbm_width_offset;struct rdt_hw_resource *hw_res =resctrl_to_arch_res (r);unsigned int threshold; -int ret; resctrl_rmid_realloc_limit = boot_cpu_data.x86_cache_size *1024 ; hw_res->mon_scale = boot_cpu_data.x86_cache_occ_scale;
@@ -1049,10 +1065,6 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)*/ resctrl_rmid_realloc_threshold =resctrl_arch_round_mon_val (threshold); - ret =dom_data_init (r); -if (ret) -return ret; -if (rdt_cpu_has (X86_FEATURE_BMEC )) {u32 eax, ebx, ecx, edx;
@@ -1070,8 +1082,6 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r) } } -l3_mon_evt_init (r); - r->mon_capable = true;return 0 ;
arch/x86/kernel/cpu/resctrl/rdtgroup.c +8 -1
@@ -4168,9 +4168,15 @@ int __init resctrl_init(void)rdtgroup_setup_default (); + ret =resctrl_mon_resource_init (); +if (ret) +return ret; + ret =sysfs_create_mount_point (fs_kobj,"resctrl" ); -if (ret) +if (ret) { +resctrl_mon_resource_exit ();return ret; + } ret =register_filesystem (&rdt_fs_type);if (ret)
@@ -4203,6 +4209,7 @@ int __init resctrl_init(void) cleanup_mountpoint:sysfs_remove_mount_point (fs_kobj,"resctrl" ); +resctrl_mon_resource_exit ();return ret; }
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 16/38] x86/resctrl: Move monitor init work to a resctrl init call
Hi James,
Quoted 41 lines · James Morse
rdt_get_mon_l3_config()is called from the architecture'sresctrl_arch_late_init(), and initialises both architecture specific fields, such ashw_res->mon_scaleand resctrl filesystem fields by callingdom_data_init().To separate the filesystem and architecture parts of resctrl, this function needs splitting up.
Add
resctrl_mon_resource_init()to do the filesystem specific work, and call it fromresctrl_init(). This runs later, but is still before the filesystem is mounted and the rmid_ptrs[] array can be used.Signed-off-by James Morse <[email protected]>Tested-by Peter Newman <[email protected]>Tested-by Shaopeng Tan <[email protected]>---
Changes since v2:
* Added error handling for the case sysfs files can't be created.--- arch/x86/kernel/cpu/resctrl/internal.h | 1 + arch/x86/kernel/cpu/resctrl/monitor.c | 24 +++++++++++++++++------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 9 ++++++++- 3 files changed, 26 insertions(+), 8 deletions(-)arch/x86/kernel/cpu/resctrl/internal.h +1
-0@@ -542,6 +542,7 @@ int rdtgroup_mondata_show(struct seq_file *m, void *arg);void mon_event_read (struct rmid_read *rr,struct rdt_resource *r,struct rdt_domain *d,struct rdtgroup *rdtgrp,int evtid,int first); +int resctrl_mon_resource_init (void );void mbm_setup_overflow_handler (struct rdt_domain *dom,unsigned long delay_ms,int exclude_cpu);arch/x86/kernel/cpu/resctrl/monitor.c +0
-0
arch/x86/kernel/cpu/resctrl/monitor.c static void l3_mon_evt_init(struct rdt_resource *r) · line 1019
Quoted 4 lines
arch/x86/kernel/cpu/resctrl/monitor.c +1
-0@@ -1016,12 +1016,28 @@ static void l3_mon_evt_init(struct rdt_resource *r)list_add_tail (&mbm_local_event.list, &r->evt_list); } +int resctrl_mon_resource_init (void )
(Lack of an __init is unexpected but I assume it was done since that will be removed in later patch anyway?)
This function needs a big warning to deter anybody from considering this to be the place where any and all monitor related allocations happen. It needs to warn developers that only resources that can only be touched after fs mount may be allocated here.
Reinette
James Morse
reviewerRe: [PATCH v3 16/38] x86/resctrl: Move monitor init work to a resctrl init call
Hi Reinette,
Quoted 13 lines · Reinette Chatre
Quoted 11 lines · James Morse
rdt_get_mon_l3_config()is called from the architecture'sresctrl_arch_late_init(), and initialises both architecture specific fields, such ashw_res->mon_scaleand resctrl filesystem fields by callingdom_data_init().To separate the filesystem and architecture parts of resctrl, this function needs splitting up.
Add
resctrl_mon_resource_init()to do the filesystem specific work, and call it fromresctrl_init(). This runs later, but is still before the filesystem is mounted and the rmid_ptrs[] array can be used.
Quoted 11 lines
arch/x86/kernel/cpu/resctrl/monitor.c static void l3_mon_evt_init(struct rdt_resource *r) · line 1018
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/monitor.c +0
-0@@ -1016,12 +1016,28 @@ static void l3_mon_evt_init(struct rdt_resource *r) list_add_tail(&mbm_local_event.list, &r->evt_list); } +int resctrl_mon_resource_init(void)(Lack of an
__initis unexpected but I assume it was done since that will be removed in later patch anyway?)
Yup - I'll add and remove that if you find it surprising.
Quoted 4 lines
This function needs a big warning to deter anybody from considering this to be the place where any and all monitor related allocations happen. It needs to warn developers that only resources that can only be touched after fs mount may be allocated here.
I'm afraid I don't follow. Can you give an example of the scenario you are worried about?
This is called from resctrl_init(), which is called once the architecture code has done its setup, and reckons resctrl is something that can be supported on this platform. It would be safe for the limbo/overflow callbacks to start ticking after this point - but there is no point if the filesystem isn't mounted yet.
Filesystem mount is triggered through rdt_get_tree(). The filesystem can't be mounted until resctrl_init() goes on to call register_filesystem().
These allocations could be made later (at mount time), but they're allocated once up-front today.
I've added:
/**
* resctrl_mon_resource_init() - Initialise monitoring structures. *
* Allocate and initialise the rmid_ptrs[] used for the limbo and free lists.
* Called once during boot after the struct rdt_resource's have been configured
* but before the filesystem is mounted.
*/
Thanks,
James
Reinette Chatre
maintainerRe: [PATCH v3 16/38] x86/resctrl: Move monitor init work to a resctrl init call
Hi James,
Quoted 31 lines · James Morse
Hi Reinette,
Quoted 13 lines · Reinette Chatre
Quoted 11 lines · James Morse
rdt_get_mon_l3_config()is called from the architecture'sresctrl_arch_late_init(), and initialises both architecture specific fields, such ashw_res->mon_scaleand resctrl filesystem fields by callingdom_data_init().To separate the filesystem and architecture parts of resctrl, this function needs splitting up.
Add
resctrl_mon_resource_init()to do the filesystem specific work, and call it fromresctrl_init(). This runs later, but is still before the filesystem is mounted and the rmid_ptrs[] array can be used.Quoted 11 lines
arch/x86/kernel/cpu/resctrl/monitor.c static void l3_mon_evt_init(struct rdt_resource *r) · line 1018
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/monitor.c +0
-0@@ -1016,12 +1016,28 @@ static void l3_mon_evt_init(struct rdt_resource *r) list_add_tail(&mbm_local_event.list, &r->evt_list); } +int resctrl_mon_resource_init(void)(Lack of an
__initis unexpected but I assume it was done since that will be removed in later patch anyway?)Yup - I'll add and remove that if you find it surprising.
Quoted 6 lines
Quoted 4 lines
This function needs a big warning to deter anybody from considering this to be the place where any and all monitor related allocations happen. It needs to warn developers that only resources that can only be touched after fs mount may be allocated here.
I'm afraid I don't follow. Can you give an example of the scenario you are worried about?
My concern is not a scenario with current code flow but a request for informational comments to prevent future mistakes. Specifically, as I understand the CPU online/offline handlers can run before this function is called. Those handlers do a lot of setup, getting resctrl and the system ready. It can be reasonable that some future action may need to touch a new monitoring structure and with a name like resctrl_mon_resource_init() it seems appropriate to allocate this new monitoring structure there. I am hoping that resctrl_mon_resource_init() will have sufficient comments to deter that.
Quoted 10 lines
This is called from
resctrl_init(), which is called once the architecture code has done its setup, and reckons resctrl is something that can be supported on this platform. It would be safe for the limbo/overflow callbacks to start ticking after this point - but there is no point if the filesystem isn't mounted yet.
Filesystem mount is triggered throughrdt_get_tree(). The filesystem can't be mounted untilresctrl_init()goes on to callregister_filesystem().
These allocations could be made later (at mount time), but they're allocated once up-front today.
Quoted 3 lines
I've added:
/**
*resctrl_mon_resource_init()- Initialise monitoring structures.
How about a more specific "Initialise monitoring structures used after filesystem mount"?
Quoted 4 lines
*
* Allocate and initialise the rmid_ptrs[] used for the limbo and free lists.
* Called once during boot after the struct rdt_resource's have been configured
* but before the filesystem is mounted.
Can there be a warning (please feel free to improve):
"Only for resources used after filesystem mount. For example, do not allocate resources
needed by the CPU online/offline handlers since these handlers may run before this function."
Quoted 6 lines
*/
Thanks,
James
Reinette
James Morse
reviewerRe: [PATCH v3 16/38] x86/resctrl: Move monitor init work to a resctrl init call
Hi Reinette,
Quoted 38 lines · Reinette Chatre
Quoted 30 lines · James Morse
Quoted 13 lines · Reinette Chatre
Quoted 11 lines · James Morse
rdt_get_mon_l3_config()is called from the architecture'sresctrl_arch_late_init(), and initialises both architecture specific fields, such ashw_res->mon_scaleand resctrl filesystem fields by callingdom_data_init().To separate the filesystem and architecture parts of resctrl, this function needs splitting up.
Add
resctrl_mon_resource_init()to do the filesystem specific work, and call it fromresctrl_init(). This runs later, but is still before the filesystem is mounted and the rmid_ptrs[] array can be used.Quoted 12 lines
arch/x86/kernel/cpu/resctrl/monitor.c static void l3_mon_evt_init(struct rdt_resource *r) · line 1018
Quoted 9 lines
arch/x86/kernel/cpu/resctrl/monitor.c +0
-0@@ -1016,12 +1016,28 @@ static void l3_mon_evt_init(struct rdt_resource *r) list_add_tail(&mbm_local_event.list, &r->evt_list); } +int resctrl_mon_resource_init(void)(Lack of an
__initis unexpected but I assume it was done since that will be removed in later patch anyway?)Yup - I'll add and remove that if you find it surprising.
Quoted 6 lines
Quoted 4 lines
This function needs a big warning to deter anybody from considering this to be the place where any and all monitor related allocations happen. It needs to warn developers that only resources that can only be touched after fs mount may be allocated here.
I'm afraid I don't follow. Can you give an example of the scenario you are worried about?
Quoted 8 lines
My concern is not a scenario with current code flow but a request for informational comments to prevent future mistakes. Specifically, as I understand the CPU online/offline handlers can run before this function is called. Those handlers do a lot of setup, getting resctrl and the system ready. It can be reasonable that some future action may need to touch a new monitoring structure and with a name like
resctrl_mon_resource_init()it seems appropriate
to allocate this new monitoring structure there. I am hoping thatresctrl_mon_resource_init()will have sufficient comments to deter that.
Ah, Of course! ... this is about 'global' allocations that don't belong to a specific domain.
I've reworded the comment above the function as:
| * Allocate and initialise global monitor resources that do not belong to a | * specific domain. i.e. the rmid_ptrs[] used for the limbo and free lists. | * Called once during boot after the struct rdt_resource's have been configured | * but before the filesystem is mounted. | * Resctrl's cpuhp callbacks may be called before this point to bring a domain | * online.
and a similar comment above domain_setup_mon_state:
| * Allocate monitor resources that belong to this domain. | * Called when the first CPU of a domain comes online, regardless of whether | * the filesystem is mounted. | * During boot this may be called before global allocations have been made by | * resctrl_mon_resource_init().
Quoted 10 lines
Quoted 10 lines
This is called from
resctrl_init(), which is called once the architecture code has done its setup, and reckons resctrl is something that can be supported on this platform. It would be safe for the limbo/overflow callbacks to start ticking after this point - but there is no point if the filesystem isn't mounted yet.
Filesystem mount is triggered throughrdt_get_tree(). The filesystem can't be mounted untilresctrl_init()goes on to callregister_filesystem().
These allocations could be made later (at mount time), but they're allocated once up-front today.
Quoted 5 lines
Quoted 3 lines
I've added:
/**
*resctrl_mon_resource_init()- Initialise monitoring structures.How about a more specific "Initialise monitoring structures used after filesystem mount"?
Sure, this has become;
| * resctrl_mon_resource_init() - Initialise global monitoring structures used
| * after filesystem mount.
Quoted 9 lines
Quoted 4 lines
*
* Allocate and initialise the rmid_ptrs[] used for the limbo and free lists.
* Called once during boot after the struct rdt_resource's have been configured
* but before the filesystem is mounted.Can there be a warning (please feel free to improve):
"Only for resources used after filesystem mount. For example, do not allocate resources needed by the CPU online/offline handlers since these handlers may run before this function."
Enumerating what not to do feels like the beginning of a never ending story! I think describing these as global/specific-to-a-domain makes it clear what kind of allocation should go here.
Thanks,
James
James Morse
reviewerPATCH v3 17/38 x86/resctrl: Stop using the for_each_*_rdt_resource() walkers
In reply to James MorseThe for_each_*_rdt_resource() helpers walk the architecture's array of structures, using the resctrl visible part as an iterator. These became over-complex when the structures were split into a filesystem and architecture-specific struct. This approach avoided the need to touch every call site.
Once the filesystem parts of resctrl are moved to /fs/, both the architecture's resource array, and the definition of those structures is no longer accessible. To support resctrl, each architecture would have to provide equally complex macros.
Change the resctrl code that uses these to walk through the resource_level enum and check the mon/alloc capable flags instead. Instances in core.c, and resctrl_arch_reset_resources() remain part of x86's architecture specific code.
---
Changes since v1:
* [Whitespace only] Fix bogus whitespace introduced in rdtgroup_create_info_dir().
* [Commit message only] Typo fix:
s/architectures/architecture's/g
--- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 7 +++++- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 28 +++++++++++++++++++---- 2 files changed, 30 insertions(+), 5 deletions(-)
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +6 -1
@@ -840,6 +840,7 @@ bool rdtgroup_cbm_overlaps_pseudo_locked(struct rdt_domain *d, unsigned long cbmbool rdtgroup_pseudo_locked_in_hierarchy (struct rdt_domain *d) {cpumask_var_t cpu_with_psl; +enum resctrl_res_level i;struct rdt_resource *r;struct rdt_domain *d_i;bool ret = false;
@@ -854,7 +855,11 @@ bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d)* First determine which cpus have pseudo-locked regions * associated with them. */ -for_each_alloc_capable_rdt_resource (r) { +for (i =0 ; i <RDT_NUM_RESOURCES ; i++) { + r =resctrl_arch_get_resource (i); +if (!r->alloc_capable) +continue ; +list_for_each_entry (d_i, &r->domains, list) {if (d_i->plr) cpumask_or(cpu_with_psl, cpu_with_psl,
arch/x86/kernel/cpu/resctrl/rdtgroup.c +24 -4
@@ -98,12 +98,17 @@ void rdt_last_cmd_printf(const char *fmt, ...)void rdt_staged_configs_clear (void ) { +enum resctrl_res_level i;struct rdt_resource *r;struct rdt_domain *dom;lockdep_assert_held (&rdtgroup_mutex); -for_each_alloc_capable_rdt_resource (r) { +for (i =0 ; i <RDT_NUM_RESOURCES ; i++) { + r =resctrl_arch_get_resource (i); +if (!r->alloc_capable) +continue ; +list_for_each_entry (dom, &r->domains, list)memset (dom->staged_config,0 ,sizeof (dom->staged_config)); }
@@ -2192,6 +2197,7 @@ static u32 fflags_from_resource(struct rdt_resource *r)static int rdtgroup_create_info_dir (struct kernfs_node *parent_kn) { +enum resctrl_res_level i;struct resctrl_schema *s;struct rdt_resource *r;unsigned long fflags;
@@ -2216,7 +2222,11 @@ static int rdtgroup_create_info_dir(struct kernfs_node *parent_kn) goto out_destroy; } -for_each_mon_capable_rdt_resource (r) { +for (i =0 ; i <RDT_NUM_RESOURCES ; i++) { + r =resctrl_arch_get_resource (i); +if (!r->mon_capable) +continue ; + fflags =fflags_from_resource (r) |RFTYPE_MON_INFO ;sprintf (name,"%s_MON" , r->name); ret =rdtgroup_mkdir_info_resdir (r, name, fflags);
@@ -2637,10 +2647,15 @@ static int schemata_list_add(struct rdt_resource *r, enum resctrl_conf_type typestatic int schemata_list_create (void ) { +enum resctrl_res_level i;struct rdt_resource *r;int ret =0 ; -for_each_alloc_capable_rdt_resource (r) { +for (i =0 ; i <RDT_NUM_RESOURCES ; i++) { + r =resctrl_arch_get_resource (i); +if (!r->alloc_capable) +continue ; +if (resctrl_arch_get_cdp_enabled (r->rid)) { ret =schemata_list_add (r,CDP_CODE );if (ret)
@@ -3181,6 +3196,7 @@ static int mkdir_mondata_all(struct kernfs_node *parent_kn,struct rdtgroup *prgrp, struct kernfs_node **dest_kn) { + enum resctrl_res_level i;struct rdt_resource *r;struct kernfs_node *kn;int ret;
@@ -3199,7 +3215,11 @@ static int mkdir_mondata_all(struct kernfs_node *parent_kn,* Create the subdirectories for each domain. Note that all events * in a domain like L3 are grouped into a resource whose domain is L3 */ -for_each_mon_capable_rdt_resource (r) { +for (i =0 ; i <RDT_NUM_RESOURCES ; i++) { + r =resctrl_arch_get_resource (i); +if (!r->mon_capable) +continue ; + ret =mkdir_mondata_subdir_alldom (kn, r, prgrp);if (ret) goto out_destroy;
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 17/38] x86/resctrl: Stop using the for_each_*_rdt_resource() walkers
Hi James,
Quoted 46 lines · James Morse
The for_each_*_rdt_resource() helpers walk the architecture's array of structures, using the resctrl visible part as an iterator. These became over-complex when the structures were split into a filesystem and architecture-specific struct. This approach avoided the need to touch every call site.
Once the filesystem parts of resctrl are moved to /fs/, both the architecture's resource array, and the definition of those structures is no longer accessible. To support resctrl, each architecture would have to provide equally complex macros.
Change the resctrl code that uses these to walk through the
resource_levelenum and check the mon/alloc capable flags instead. Instances incore.c, andresctrl_arch_reset_resources()remain part of x86's architecture specific code.Co-developed-by Dave Martin <[email protected]>Signed-off-by Dave Martin <[email protected]>Signed-off-by James Morse <[email protected]>Tested-by Peter Newman <[email protected]>Tested-by Shaopeng Tan <[email protected]>---
Changes since v1:
* [Whitespace only] Fix bogus whitespace introduced inrdtgroup_create_info_dir().* [Commit message only] Typo fix:
s/architectures/architecture's/g--- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 7 +++++- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 28 +++++++++++++++++++---- 2 files changed, 30 insertions(+), 5 deletions(-)arch/x86/kernel/cpu/resctrl/pseudo_lock.c +1
-0@@ -840,6 +840,7 @@ bool rdtgroup_cbm_overlaps_pseudo_locked(struct rdt_domain *d, unsigned long cbmbool rdtgroup_pseudo_locked_in_hierarchy (struct rdt_domain *d) {cpumask_var_t cpu_with_psl; +enum resctrl_res_level i;struct rdt_resource *r;struct rdt_domain *d_i;bool ret = false;
arch/x86/kernel/cpu/resctrl/pseudo_lock.c bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d) · line 862
Quoted 9 lines
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +5
-1@@ -854,7 +855,11 @@ bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d)* First determine which cpus have pseudo-locked regions * associated with them. */ -for_each_alloc_capable_rdt_resource (r) { +for (i =0 ; i <RDT_NUM_RESOURCES ; i++) { + r =resctrl_arch_get_resource (i); +if (!r->alloc_capable) +continue ; +
This looks like enough duplicate boilerplate for a new macro. For simplicity the macro could require two arguments with enum resctrl_res_level also provided?
Reinette
James Morse
reviewerRe: [PATCH v3 17/38] x86/resctrl: Stop using the for_each_*_rdt_resource() walkers
Hi Reinette,
Quoted 27 lines · Reinette Chatre
Quoted 26 lines · James Morse
The for_each_*_rdt_resource() helpers walk the architecture's array of structures, using the resctrl visible part as an iterator. These became over-complex when the structures were split into a filesystem and architecture-specific struct. This approach avoided the need to touch every call site.
Once the filesystem parts of resctrl are moved to /fs/, both the architecture's resource array, and the definition of those structures is no longer accessible. To support resctrl, each architecture would have to provide equally complex macros.
Change the resctrl code that uses these to walk through the
resource_levelenum and check the mon/alloc capable flags instead. Instances incore.c, andresctrl_arch_reset_resources()remain part of x86's architecture specific code.arch/x86/kernel/cpu/resctrl/pseudo_lock.c +0
-1@@ -854,7 +855,11 @@ bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d)* First determine which cpus have pseudo-locked regions * associated with them. */ - for_each_alloc_capable_rdt_resource(r) {
Quoted 8 lines
arch/x86/kernel/cpu/resctrl/pseudo_lock.c bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d) · line 862
Quoted 5 lines
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +5
-0+ for (i = 0; i < RDT_NUM_RESOURCES; i++) { + r = resctrl_arch_get_resource(i); + if (!r->alloc_capable) + continue; +This looks like enough duplicate boilerplate for a new macro. For simplicity the macro could require two arguments with enum
resctrl_res_levelalso provided?
I was hoping to escape from these clever macros! If you think this is too much:
- we'd need to come up with another name, as the arch code keeps the existing definition.
- to avoid touching every caller, it needs doing without an explicit iterator variable.
I guess the cleanest thing is to redefine the existing macros to use resctrl_arch_get_resource(). Putting this in include/linxu/resctrl.h at least avoids each architecture needing to define these, or forcing it to use an array.
The result is slightly more readable than the current version:
| #define for_each_rdt_resource(_r) \ | for (_r = resctrl_arch_get_resource(0); \ | _r->rid < RDT_NUM_RESOURCES; \ | _r = resctrl_arch_get_resource(_r->rid + 1))
This leans heavily on resctrl_arch_get_resource() not being able to return NULL, and having to return a dummy resource that is neither alloc nor mon capable. We may need to revisit that if it becomes a burden for the arch code.
Thanks,
James
Reinette Chatre
maintainerRe: [PATCH v3 17/38] x86/resctrl: Stop using the for_each_*_rdt_resource() walkers
Hi James,
Quoted 48 lines · James Morse
Hi Reinette,
Quoted 27 lines · Reinette Chatre
Quoted 26 lines · James Morse
The for_each_*_rdt_resource() helpers walk the architecture's array of structures, using the resctrl visible part as an iterator. These became over-complex when the structures were split into a filesystem and architecture-specific struct. This approach avoided the need to touch every call site.
Once the filesystem parts of resctrl are moved to /fs/, both the architecture's resource array, and the definition of those structures is no longer accessible. To support resctrl, each architecture would have to provide equally complex macros.
Change the resctrl code that uses these to walk through the
resource_levelenum and check the mon/alloc capable flags instead. Instances incore.c, andresctrl_arch_reset_resources()remain part of x86's architecture specific code.arch/x86/kernel/cpu/resctrl/pseudo_lock.c +0
-1@@ -854,7 +855,11 @@ bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d)* First determine which cpus have pseudo-locked regions * associated with them. */ - for_each_alloc_capable_rdt_resource(r) {Quoted 8 lines
arch/x86/kernel/cpu/resctrl/pseudo_lock.c bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d) · line 862
Quoted 5 lines
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +5
-0+ for (i = 0; i < RDT_NUM_RESOURCES; i++) { + r = resctrl_arch_get_resource(i); + if (!r->alloc_capable) + continue; +This looks like enough duplicate boilerplate for a new macro. For simplicity the macro could require two arguments with enum
resctrl_res_levelalso provided?I was hoping to escape from these clever macros! If you think this is too much:
- we'd need to come up with another name, as the arch code keeps the existing definition.
- to avoid touching every caller, it needs doing without an explicit iterator variable.I guess the cleanest thing is to redefine the existing macros to use
resctrl_arch_get_resource(). Putting this ininclude/linxu/resctrl.hat least avoids each architecture needing to define these, or forcing it to use an array.The result is slightly more readable than the current version:
Quoted 8 lines
| #define for_each_rdt_resource(_r) \ | for (_r = resctrl_arch_get_resource(0); \ | _r->rid < RDT_NUM_RESOURCES; \ | _r = resctrl_arch_get_resource(_r->rid + 1))This leans heavily on
resctrl_arch_get_resource()not being able to return NULL, and having to return a dummy resource that is neither alloc nor mon capable. We may need to revisit that if it becomes a burden for the arch code.
Replacing the repetitive four lines of code with a single line seems good to me.
resctrl_arch_get_resource() being able to return NULL is introduced in this series but I am not seeing any handling of a possible NULL value. Not being able to return NULL thus already seems a requirement?
Reinette
James Morse
reviewerRe: [PATCH v3 17/38] x86/resctrl: Stop using the for_each_*_rdt_resource() walkers
Hi Reinette,
Quoted 56 lines · Reinette Chatre
Quoted 46 lines · James Morse
Quoted 27 lines · Reinette Chatre
Quoted 26 lines · James Morse
The for_each_*_rdt_resource() helpers walk the architecture's array of structures, using the resctrl visible part as an iterator. These became over-complex when the structures were split into a filesystem and architecture-specific struct. This approach avoided the need to touch every call site.
Once the filesystem parts of resctrl are moved to /fs/, both the architecture's resource array, and the definition of those structures is no longer accessible. To support resctrl, each architecture would have to provide equally complex macros.
Change the resctrl code that uses these to walk through the
resource_levelenum and check the mon/alloc capable flags instead. Instances incore.c, andresctrl_arch_reset_resources()remain part of x86's architecture specific code.arch/x86/kernel/cpu/resctrl/pseudo_lock.c +0
-1@@ -854,7 +855,11 @@ bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d)* First determine which cpus have pseudo-locked regions * associated with them. */ - for_each_alloc_capable_rdt_resource(r) {Quoted 8 lines
arch/x86/kernel/cpu/resctrl/pseudo_lock.c bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d) · line 862
Quoted 5 lines
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +5
-0+ for (i = 0; i < RDT_NUM_RESOURCES; i++) { + r = resctrl_arch_get_resource(i); + if (!r->alloc_capable) + continue; +This looks like enough duplicate boilerplate for a new macro. For simplicity the macro could require two arguments with enum
resctrl_res_levelalso provided?I was hoping to escape from these clever macros! If you think this is too much:
- we'd need to come up with another name, as the arch code keeps the existing definition.
- to avoid touching every caller, it needs doing without an explicit iterator variable.I guess the cleanest thing is to redefine the existing macros to use
resctrl_arch_get_resource(). Putting this ininclude/linxu/resctrl.hat least avoids each architecture needing to define these, or forcing it to use an array.The result is slightly more readable than the current version:
Quoted 8 lines
| #define for_each_rdt_resource(_r) \ | for (_r = resctrl_arch_get_resource(0); \ | _r->rid < RDT_NUM_RESOURCES; \ | _r = resctrl_arch_get_resource(_r->rid + 1))This leans heavily on
resctrl_arch_get_resource()not being able to return NULL, and having to return a dummy resource that is neither alloc nor mon capable. We may need to revisit that if it becomes a burden for the arch code.
Quoted 5 lines
Replacing the repetitive four lines of code with a single line seems good to me.
resctrl_arch_get_resource()being able to return NULL is introduced in this series but I am not seeing any handling of a possible NULL value. Not being able to return NULL thus already seems a requirement?
It's currently implicit because until this point resctrl has just reached into the rdt_resources_all[] array - and can never get a NULL pointer. Replacing that with a helper needed to preserve the no-NULLs behaviour.
Changing this created too much churn so the resctrl idiom is to check alloc_enabled/mon_enabled to see if the resource actually exists....
If we wanted to change this, that for_each_rdt_resource() would need an index variable as _r could be NULL.
Thanks,
James
James Morse
reviewerPATCH v3 18/38 x86/resctrl: Export the is_mbm_*_enabled() helpers to asm/resctrl.h
In reply to James MorseThe architecture specific parts of resctrl have helpers to hide accesses to the rdt_mon_features bitmap.
Once the filesystem parts of resctrl are moved, these can no longer live in internal.h. Once these are exposed to the wider kernel, they should have a 'resctrl_arch_' prefix, to fit the rest of the arch<->fs interface.
Move and rename the helpers that touch rdt_mon_features directly. is_mbm_event() and is_mbm_enabled() are only called from rdtgroup.c, so can be moved into that file.
--- arch/x86/include/asm/resctrl.h | 17 +++++++++++ arch/x86/kernel/cpu/resctrl/core.c | 4 +-- arch/x86/kernel/cpu/resctrl/internal.h | 27 ----------------- arch/x86/kernel/cpu/resctrl/monitor.c | 18 ++++++------ arch/x86/kernel/cpu/resctrl/rdtgroup.c | 40 +++++++++++++++++--------- 5 files changed, 54 insertions(+), 52 deletions(-)
arch/x86/include/asm/resctrl.h +17 -0
@@ -7,6 +7,7 @@#include <linux/jump_label.h> #include <linux/percpu.h> #include <linux/sched.h> +#include <linux/resctrl_types.h> /* * This value can never be a valid CLOSID, and is used when mapping a
@@ -43,6 +44,7 @@ DECLARE_PER_CPU(struct resctrl_pqr_state, pqr_state);extern bool rdt_alloc_capable;extern bool rdt_mon_capable; +extern unsigned int rdt_mon_features;DECLARE_STATIC_KEY_FALSE (rdt_enable_key);DECLARE_STATIC_KEY_FALSE (rdt_alloc_enable_key);
@@ -82,6 +84,21 @@ static inline void resctrl_arch_disable_mon(void)static_branch_dec_cpuslocked (&rdt_enable_key); } +static inline bool resctrl_arch_is_llc_occupancy_enabled (void ) +{ +return (rdt_mon_features & (1 <<QOS_L3_OCCUP_EVENT_ID )); +} + +static inline bool resctrl_arch_is_mbm_total_enabled (void ) +{ +return (rdt_mon_features & (1 <<QOS_L3_MBM_TOTAL_EVENT_ID )); +} + +static inline bool resctrl_arch_is_mbm_local_enabled (void ) +{ +return (rdt_mon_features & (1 <<QOS_L3_MBM_LOCAL_EVENT_ID )); +} +/* * __resctrl_sched_in() - Writes the task's CLOSid/RMID to IA32_PQR_MSR *
arch/x86/kernel/cpu/resctrl/core.c +2 -2
@@ -465,13 +465,13 @@ static int arch_domain_mbm_alloc(u32 num_rmid, struct rdt_hw_domain *hw_dom) {size_t tsize; -if (is_mbm_total_enabled ()) { +if (resctrl_arch_is_mbm_total_enabled ()) { tsize =sizeof (*hw_dom->arch_mbm_total); hw_dom->arch_mbm_total =kcalloc (num_rmid, tsize,GFP_KERNEL );if (!hw_dom->arch_mbm_total)return -ENOMEM ; } -if (is_mbm_local_enabled ()) { +if (resctrl_arch_is_mbm_local_enabled ()) { tsize =sizeof (*hw_dom->arch_mbm_local); hw_dom->arch_mbm_local =kcalloc (num_rmid, tsize,GFP_KERNEL );if (!hw_dom->arch_mbm_local) {
arch/x86/kernel/cpu/resctrl/internal.h +0 -27
@@ -131,7 +131,6 @@ struct rmid_read {void *arch_mon_ctx; }; -extern unsigned int rdt_mon_features;extern struct list_head resctrl_schema_all;extern bool resctrl_mounted;
@@ -363,32 +362,6 @@ struct msr_param {u32 high; }; -static inline bool is_llc_occupancy_enabled (void ) -{ -return (rdt_mon_features & (1 <<QOS_L3_OCCUP_EVENT_ID )); -} - -static inline bool is_mbm_total_enabled (void ) -{ -return (rdt_mon_features & (1 <<QOS_L3_MBM_TOTAL_EVENT_ID )); -} - -static inline bool is_mbm_local_enabled (void ) -{ -return (rdt_mon_features & (1 <<QOS_L3_MBM_LOCAL_EVENT_ID )); -} - -static inline bool is_mbm_enabled (void ) -{ -return (is_mbm_total_enabled () ||is_mbm_local_enabled ()); -} - -static inline bool is_mbm_event (int e) -{ -return (e >=QOS_L3_MBM_TOTAL_EVENT_ID && - e <=QOS_L3_MBM_LOCAL_EVENT_ID ); -} -/** * struct rdt_hw_resource - arch private attributes of a resctrl resource * @r_resctrl: Attributes of the resource used directly by resctrl.
arch/x86/kernel/cpu/resctrl/monitor.c +9 -9
@@ -252,11 +252,11 @@ void resctrl_arch_reset_rmid_all(struct rdt_resource *r, struct rdt_domain *d) {struct rdt_hw_domain *hw_dom =resctrl_to_arch_dom (d); -if (is_mbm_total_enabled ()) +if (resctrl_arch_is_mbm_total_enabled ())memset (hw_dom->arch_mbm_total,0 ,sizeof (*hw_dom->arch_mbm_total) * r->num_rmid); -if (is_mbm_local_enabled ()) +if (resctrl_arch_is_mbm_local_enabled ())memset (hw_dom->arch_mbm_local,0 ,sizeof (*hw_dom->arch_mbm_local) * r->num_rmid); }
@@ -525,7 +525,7 @@ void free_rmid(u32 closid, u32 rmid) entry =__rmid_entry (idx); -if (is_llc_occupancy_enabled ()) +if (resctrl_arch_is_llc_occupancy_enabled ())add_rmid_to_limbo (entry);else list_add_tail (&entry->list, &rmid_free_lru);
@@ -677,7 +677,7 @@ static void update_mba_bw(struct rdtgroup *rgrp, struct rdt_domain *dom_mbm)struct list_head *head;struct rdtgroup *entry; -if (!is_mbm_local_enabled ()) +if (!resctrl_arch_is_mbm_local_enabled ())return ; r_mba =resctrl_arch_get_resource (RDT_RESOURCE_MBA );
@@ -746,7 +746,7 @@ static void mbm_update(struct rdt_resource *r, struct rdt_domain *d,* This is protected from concurrent reads from user * as both the user and we hold the global mutex. */ -if (is_mbm_total_enabled ()) { +if (resctrl_arch_is_mbm_total_enabled ()) { rr.evtid =QOS_L3_MBM_TOTAL_EVENT_ID ; rr.val =0 ; rr.arch_mon_ctx =resctrl_arch_mon_ctx_alloc (rr.r, rr.evtid);
@@ -760,7 +760,7 @@ static void mbm_update(struct rdt_resource *r, struct rdt_domain *d,resctrl_arch_mon_ctx_free (rr.r, rr.evtid, rr.arch_mon_ctx); } -if (is_mbm_local_enabled ()) { +if (resctrl_arch_is_mbm_local_enabled ()) { rr.evtid =QOS_L3_MBM_LOCAL_EVENT_ID ; rr.val =0 ; rr.arch_mon_ctx =resctrl_arch_mon_ctx_alloc (rr.r, rr.evtid);
@@ -1008,11 +1008,11 @@ static void l3_mon_evt_init(struct rdt_resource *r) {INIT_LIST_HEAD (&r->evt_list); -if (is_llc_occupancy_enabled ()) +if (resctrl_arch_is_llc_occupancy_enabled ())list_add_tail (&llc_occupancy_event.list, &r->evt_list); -if (is_mbm_total_enabled ()) +if (resctrl_arch_is_mbm_total_enabled ())list_add_tail (&mbm_total_event.list, &r->evt_list); -if (is_mbm_local_enabled ()) +if (resctrl_arch_is_mbm_local_enabled ())list_add_tail (&mbm_local_event.list, &r->evt_list); }
arch/x86/kernel/cpu/resctrl/rdtgroup.c +26 -14
@@ -114,6 +114,18 @@ void rdt_staged_configs_clear(void) } } +static bool resctrl_is_mbm_enabled (void ) +{ +return (resctrl_arch_is_mbm_total_enabled () || +resctrl_arch_is_mbm_local_enabled ()); +} + +static bool resctrl_is_mbm_event (int e) +{ +return (e >=QOS_L3_MBM_TOTAL_EVENT_ID && + e <=QOS_L3_MBM_LOCAL_EVENT_ID ); +} +/* * Trivial allocator for CLOSIDs. Since h/w only supports a small number, * we can keep a bitmap of free CLOSIDs in a single integer.
@@ -161,7 +173,7 @@ static int closid_alloc(void)lockdep_assert_held (&rdtgroup_mutex);if (IS_ENABLED (CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID ) && -is_llc_occupancy_enabled ()) { +resctrl_arch_is_llc_occupancy_enabled ()) { cleanest_closid =resctrl_find_cleanest_closid ();if (cleanest_closid <0 )return cleanest_closid;
@@ -2381,7 +2393,7 @@ static bool supports_mba_mbps(void) {struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_MBA ); -return (is_mbm_local_enabled () && +return (resctrl_arch_is_mbm_local_enabled () && r->alloc_capable &&is_mba_linear ()); }
@@ -2760,7 +2772,7 @@ static int rdt_get_tree(struct fs_context *fc)if (resctrl_arch_alloc_capable () ||resctrl_arch_mon_capable ()) resctrl_mounted = true; -if (is_mbm_enabled ()) { +if (resctrl_is_mbm_enabled ()) { r =resctrl_arch_get_resource (RDT_RESOURCE_L3 );list_for_each_entry (dom, &r->domains, list) mbm_setup_overflow_handler(dom, MBM_OVERFLOW_INTERVAL,
@@ -3122,7 +3134,7 @@ static int mkdir_mondata_subdir(struct kernfs_node *parent_kn,if (ret) goto out_destroy; -if (is_mbm_event (mevt->evtid)) +if (resctrl_is_mbm_event (mevt->evtid))mon_event_read (&rr, r, d, prgrp, mevt->evtid, true); }kernfs_activate (kn);
@@ -4024,9 +4036,9 @@ void resctrl_offline_domain(struct rdt_resource *r, struct rdt_domain *d)if (resctrl_mounted &&resctrl_arch_mon_capable ())rmdir_mondata_subdir_allrdtgrp (r, d->id); -if (is_mbm_enabled ()) +if (resctrl_is_mbm_enabled ())cancel_delayed_work (&d->mbm_over); -if (is_llc_occupancy_enabled () &&has_busy_rmid (d)) { +if (resctrl_arch_is_llc_occupancy_enabled () &&has_busy_rmid (d)) {/* * When a package is going down, forcefully * decrement rmid->ebusy. There is no way to know
@@ -4050,12 +4062,12 @@ static int domain_setup_mon_state(struct rdt_resource *r, struct rdt_domain *d)u32 idx_limit =resctrl_arch_system_num_rmid_idx ();size_t tsize; -if (is_llc_occupancy_enabled ()) { +if (resctrl_arch_is_llc_occupancy_enabled ()) { d->rmid_busy_llc =bitmap_zalloc (idx_limit,GFP_KERNEL );if (!d->rmid_busy_llc)return -ENOMEM ; } -if (is_mbm_total_enabled ()) { +if (resctrl_arch_is_mbm_total_enabled ()) { tsize =sizeof (*d->mbm_total); d->mbm_total =kcalloc (idx_limit, tsize,GFP_KERNEL );if (!d->mbm_total) {
@@ -4063,7 +4075,7 @@ static int domain_setup_mon_state(struct rdt_resource *r, struct rdt_domain *d)return -ENOMEM ; } } -if (is_mbm_local_enabled ()) { +if (resctrl_arch_is_mbm_local_enabled ()) { tsize =sizeof (*d->mbm_local); d->mbm_local =kcalloc (idx_limit, tsize,GFP_KERNEL );if (!d->mbm_local) {
@@ -4095,13 +4107,13 @@ int resctrl_online_domain(struct rdt_resource *r, struct rdt_domain *d)if (err) goto out_unlock; -if (is_mbm_enabled ()) { +if (resctrl_is_mbm_enabled ()) {INIT_DELAYED_WORK (&d->mbm_over, mbm_handle_overflow);mbm_setup_overflow_handler (d,MBM_OVERFLOW_INTERVAL ,RESCTRL_PICK_ANY_CPU ); } -if (is_llc_occupancy_enabled ()) +if (resctrl_arch_is_llc_occupancy_enabled ())INIT_DELAYED_WORK (&d->cqm_limbo, cqm_handle_limbo);/*
@@ -4156,12 +4168,12 @@ void resctrl_offline_cpu(unsigned int cpu) d =get_domain_from_cpu (cpu, l3);if (d) { -if (is_mbm_enabled () && cpu == d->mbm_work_cpu) { +if (resctrl_is_mbm_enabled () && cpu == d->mbm_work_cpu) {cancel_delayed_work (&d->mbm_over);mbm_setup_overflow_handler (d,0 , cpu); } -if (is_llc_occupancy_enabled () && cpu == d->cqm_work_cpu && -has_busy_rmid (d)) { +if (resctrl_arch_is_llc_occupancy_enabled () && + cpu == d->cqm_work_cpu &&has_busy_rmid (d)) {cancel_delayed_work (&d->cqm_limbo);cqm_setup_limbo_handler (d,0 , cpu); }
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 19/38 x86/resctrl: Add resctrl_arch_is_evt_configurable() to abstract BMEC
In reply to James MorseWhen BMEC is supported the resctrl event can be configured in a number of ways. This depends on architecture support. rdt_get_mon_l3_config() modifies the struct mon_evt and calls mbm_config_rftype_init() to create the files that allow the configuration.
Splitting this into separate architecture and filesystem parts would require the struct mon_evt and mbm_config_rftype_init() to be exposed.
Instead, add resctrl_arch_is_evt_configurable(), and use this from resctrl_mon_resource_init() to initialise struct mon_evt and call mbm_config_rftype_init(). resctrl_arch_is_evt_configurable() calls rdt_cpu_has() so it doesn't obviously benefit from being inlined. Putting it in core.c will allow rdt_cpu_has() to eventually become static.
resctrl_arch_is_evt_configurable() uses rdt_cpu_has() from resctrl_mon_resource_init(), which isn't marked __init. In addition, MPAM needs to initialise resctrl late. Drop the __init on the relevant functions.
--- arch/x86/kernel/cpu/resctrl/core.c | 19 +++++++++++++++++-- arch/x86/kernel/cpu/resctrl/internal.h | 4 ++-- arch/x86/kernel/cpu/resctrl/monitor.c | 18 +++++++++--------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 2 +- include/linux/resctrl.h | 2 ++ 5 files changed, 31 insertions(+), 14 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +17 -2
@@ -651,7 +651,7 @@ struct rdt_options {bool force_off, force_on; }; -static struct rdt_options rdt_options[]__initdata = { +static struct rdt_options rdt_options[]__ro_after_init = {RDT_OPT (RDT_FLAG_CMT ,"cmt" ,X86_FEATURE_CQM_OCCUP_LLC ),RDT_OPT (RDT_FLAG_MBM_TOTAL ,"mbmtotal" ,X86_FEATURE_CQM_MBM_TOTAL ),RDT_OPT (RDT_FLAG_MBM_LOCAL ,"mbmlocal" ,X86_FEATURE_CQM_MBM_LOCAL ),
@@ -691,7 +691,7 @@ static int __init set_rdt_options(char *str) }__setup ("rdt" , set_rdt_options); -bool __init rdt_cpu_has (int flag) +bool rdt_cpu_has (int flag) {bool ret =boot_cpu_has (flag);struct rdt_options *o;
@@ -711,6 +711,21 @@ bool __init rdt_cpu_has(int flag)return ret; } +bool resctrl_arch_is_evt_configurable (enum resctrl_event_id evt) +{ +if (!rdt_cpu_has (X86_FEATURE_BMEC )) +return false; + +switch (evt) { +case QOS_L3_MBM_TOTAL_EVENT_ID : +return rdt_cpu_has (X86_FEATURE_CQM_MBM_TOTAL ); +case QOS_L3_MBM_LOCAL_EVENT_ID : +return rdt_cpu_has (X86_FEATURE_CQM_MBM_LOCAL ); +default : +return false; + } +} +static __init bool get_mem_config (void ) {struct rdt_hw_resource *hw_res = &rdt_resources_all[RDT_RESOURCE_MBA ];
arch/x86/kernel/cpu/resctrl/internal.h +2 -2
@@ -509,7 +509,7 @@ int alloc_rmid(u32 closid);void free_rmid (u32 closid,u32 rmid);int rdt_get_mon_l3_config (struct rdt_resource *r);void resctrl_mon_resource_exit (void ); -bool __init rdt_cpu_has (int flag); +bool rdt_cpu_has (int flag);void mon_event_count (void *info);int rdtgroup_mondata_show (struct seq_file *m,void *arg); void mon_event_read(struct rmid_read *rr, struct rdt_resource *r,
@@ -529,7 +529,7 @@ bool has_busy_rmid(struct rdt_domain *d);void __check_limbo (struct rdt_domain *d,bool force_free);void rdt_domain_reconfigure_cdp (struct rdt_resource *r);void __init thread_throttle_mode_init (void ); -void __init mbm_config_rftype_init (const char *config); +void mbm_config_rftype_init (const char *config);void rdt_staged_configs_clear (void );bool closid_allocated (unsigned int closid);int resctrl_find_cleanest_closid (void );
arch/x86/kernel/cpu/resctrl/monitor.c +9 -9
@@ -1030,6 +1030,15 @@ int resctrl_mon_resource_init(void)l3_mon_evt_init (r); +if (resctrl_arch_is_evt_configurable (QOS_L3_MBM_TOTAL_EVENT_ID )) { + mbm_total_event.configurable = true; +mbm_config_rftype_init ("mbm_total_bytes_config" ); + } +if (resctrl_arch_is_evt_configurable (QOS_L3_MBM_LOCAL_EVENT_ID )) { + mbm_local_event.configurable = true; +mbm_config_rftype_init ("mbm_local_bytes_config" ); + } +return 0 ; }
@@ -1071,15 +1080,6 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)/* Detect list of bandwidth sources that can be tracked */ cpuid_count (0x80000020 ,3 , &eax, &ebx, &ecx, &edx); hw_res->mbm_cfg_mask = ecx &MAX_EVT_CONFIG_BITS ; - -if (rdt_cpu_has (X86_FEATURE_CQM_MBM_TOTAL )) { - mbm_total_event.configurable = true; -mbm_config_rftype_init ("mbm_total_bytes_config" ); - } -if (rdt_cpu_has (X86_FEATURE_CQM_MBM_LOCAL )) { - mbm_local_event.configurable = true; -mbm_config_rftype_init ("mbm_local_bytes_config" ); - } } r->mon_capable = true;
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1 -1
@@ -2066,7 +2066,7 @@ void __init thread_throttle_mode_init(void) rft->fflags =RFTYPE_CTRL_INFO |RFTYPE_RES_MB ; } -void __init mbm_config_rftype_init (const char *config) +void mbm_config_rftype_init (const char *config) {struct rftype *rft;
include/linux/resctrl.h +2 -0
@@ -262,6 +262,8 @@ u32 resctrl_arch_get_num_closid(struct rdt_resource *r);struct rdt_domain *resctrl_arch_find_domain (struct rdt_resource *r,int id);int resctrl_arch_update_domains (struct rdt_resource *r,u32 closid); +bool resctrl_arch_is_evt_configurable (enum resctrl_event_id evt); +/* * Update the ctrl_val and apply this config right now. * Must be called on one of the domain's CPUs.
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 20/38 x86/resctrl: Change mon_event_config_{read,write}() to be arch helpers
In reply to James Morsemon_event_config_{read,write}() are called via IPI and access model specific registers to do their work.
To support another architecture, this needs abstracting.
Rename mon_event_config_{read,write}() to have a "resctrl_arch_" prefix, and move their struct mon_config_info parameter into <linux/resctrl.h>. This allows another architecture to supply an implementation of these.
As struct mon_config_info is now exposed globally, give it a 'resctrl_' prefix. MPAM systems need access to the domain to do this work, add the resource and domain to struct resctrl_mon_config_info.
---
Changes since v1:
* [Whitespace only] Re-tabbed struct resctrl_mon_config_info in <linux/resctrl.h> to fit the prevailing style.
Non-functional change.
* [Commit message only] Reword to align with the actual naming of the definitions and destination header file.
--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 34 +++++++++++++------------- include/linux/resctrl.h | 9 +++++++ 2 files changed, 26 insertions(+), 17 deletions(-)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +17 -17
@@ -1578,11 +1578,6 @@ static int rdtgroup_size_show(struct kernfs_open_file *of,return ret; } -struct mon_config_info { -u32 evtid; -u32 mon_config; -}; -#define INVALID_CONFIG_INDEX UINT_MAX/**
@@ -1607,9 +1602,9 @@ static inline unsigned int mon_event_config_index_get(u32 evtid) } } -static void mon_event_config_read (void *info) +void resctrl_arch_mon_event_config_read (void *info) { -struct mon_config_info *mon_info = info; +struct resctrl_mon_config_info *mon_info = info;unsigned int index;u64 msrval;
@@ -1624,14 +1619,15 @@ static void mon_event_config_read(void *info) mon_info->mon_config = msrval &MAX_EVT_CONFIG_BITS ; } -static void mondata_config_read (struct rdt_domain *d,struct mon_config_info *mon_info) +static void mondata_config_read (struct resctrl_mon_config_info *mon_info) { -smp_call_function_any (&d->cpu_mask, mon_event_config_read, mon_info,1 ); +smp_call_function_any (&mon_info->d->cpu_mask, + resctrl_arch_mon_event_config_read, mon_info,1 ); }static int mbm_config_show (struct seq_file *s,struct rdt_resource *r,u32 evtid) { -struct mon_config_info mon_info = {0 }; +struct resctrl_mon_config_info mon_info = {0 };struct rdt_domain *dom;bool sep = false;
@@ -1642,9 +1638,11 @@ static int mbm_config_show(struct seq_file *s, struct rdt_resource *r, u32 evtidif (sep)seq_puts (s,";" ); -memset (&mon_info,0 ,sizeof (struct mon_config_info )); +memset (&mon_info,0 ,sizeof (struct resctrl_mon_config_info )); + mon_info.r = r; + mon_info.d = dom; mon_info.evtid = evtid; -mondata_config_read (dom, &mon_info); +mondata_config_read (&mon_info);seq_printf (s,"%d=0x%02x" , dom->id, mon_info.mon_config); sep = true;
@@ -1677,9 +1675,9 @@ static int mbm_local_bytes_config_show(struct kernfs_open_file *of,return 0 ; } -static void mon_event_config_write (void *info) +void resctrl_arch_mon_event_config_write (void *info) { -struct mon_config_info *mon_info = info; +struct resctrl_mon_config_info *mon_info = info;unsigned int index; index =mon_event_config_index_get (mon_info->evtid);
@@ -1693,14 +1691,16 @@ static void mon_event_config_write(void *info)static void mbm_config_write_domain (struct rdt_resource *r,struct rdt_domain *d,u32 evtid,u32 val) { -struct mon_config_info mon_info = {0 }; +struct resctrl_mon_config_info mon_info = {0 };/* * Read the current config value first. If both are the same then * no need to write it again. */ + mon_info.r = r; + mon_info.d = d; mon_info.evtid = evtid; -mondata_config_read (d, &mon_info); +mondata_config_read (&mon_info);if (mon_info.mon_config == val)return ;
@@ -1712,7 +1712,7 @@ static void mbm_config_write_domain(struct rdt_resource *r,* are scoped at the domain level. Writing any of these MSRs * on one CPU is observed by all the CPUs in the domain. */ -smp_call_function_any (&d->cpu_mask, mon_event_config_write, +smp_call_function_any (&d->cpu_mask, resctrl_arch_mon_event_config_write, &mon_info,1 );/*
include/linux/resctrl.h +9 -0
@@ -220,6 +220,13 @@ struct resctrl_cpu_defaults {u32 rmid; }; +struct resctrl_mon_config_info { +struct rdt_resource *r; +struct rdt_domain *d; +u32 evtid; +u32 mon_config; +}; +/** * resctrl_arch_sync_cpu_closid_rmid() - Refresh this CPU's CLOSID and RMID. * Call via IPI.
@@ -263,6 +270,8 @@ struct rdt_domain *resctrl_arch_find_domain(struct rdt_resource *r, int id);int resctrl_arch_update_domains (struct rdt_resource *r,u32 closid);bool resctrl_arch_is_evt_configurable (enum resctrl_event_id evt); +void resctrl_arch_mon_event_config_write (void *info); +void resctrl_arch_mon_event_config_read (void *info);/* * Update the ctrl_val and apply this config right now.
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 20/38] x86/resctrl: Change mon_event_config_{read,write}() to be arch helpers
Hi James,
On 6/14/24 8:00 AM, James Morse wrote:
...
Quoted 19 lines
include/linux/resctrl.h +7
-0@@ -220,6 +220,13 @@ struct resctrl_cpu_defaults {u32 rmid; }; +struct resctrl_mon_config_info { +struct rdt_resource *r; +struct rdt_domain *d; +u32 evtid; +u32 mon_config; +}; +/** * resctrl_arch_sync_cpu_closid_rmid() - Refresh this CPU's CLOSID and RMID. * Call via IPI.
include/linux/resctrl.h struct rdt_domain *resctrl_arch_find_domain(struct rdt_resource *r, int id); · line 275
Quoted 6 lines
include/linux/resctrl.h +2
-0@@ -263,6 +270,8 @@ struct rdt_domain *resctrl_arch_find_domain(struct rdt_resource *r, int id);int resctrl_arch_update_domains (struct rdt_resource *r,u32 closid);bool resctrl_arch_is_evt_configurable (enum resctrl_event_id evt); +void resctrl_arch_mon_event_config_write (void *info); +void resctrl_arch_mon_event_config_read (void *info);
Considering this is the API doc I think it will be useful to have a comment that connects the void pointers in these functions to struct resctrl_mon_config_info.
include/linux/resctrl.h struct rdt_domain *resctrl_arch_find_domain(struct rdt_resource *r, int id); · line 277
Quoted 2 lines
include/linux/resctrl.h +0
-0/* * Update the ctrl_val and apply this config right now.
Reinette
James Morse
reviewerPATCH v3 21/38 x86/resctrl: Move mbm_cfg_mask to struct rdt_resource
In reply to James MorseThe mbm_cfg_mask field lists the bits that user-space can set when configuring an event. This value is output via the last_cmd_status file.
Once the filesystem parts of resctrl are moved to live in /fs/, the struct rdt_hw_resource is inaccessible to the filesystem code. Because this value is output to user-space, it has to be accessible to the filesystem code.
Move it to struct rdt_resource.
---
Change since v1:
* Reword comments to avoid being overly arch-specific.
--- arch/x86/kernel/cpu/resctrl/internal.h | 3 --- arch/x86/kernel/cpu/resctrl/monitor.c | 2 +- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 5 ++--- include/linux/resctrl.h | 3 +++ 4 files changed, 6 insertions(+), 7 deletions(-)
arch/x86/kernel/cpu/resctrl/internal.h +0 -3
@@ -374,8 +374,6 @@ struct msr_param {* @msr_update: Function pointer to update QOS MSRs * @mon_scale: cqm counter * mon_scale = occupancy in bytes * @mbm_width: Monitor width, to detect and correct for overflow. -* @mbm_cfg_mask: Bandwidth sources that can be tracked when Bandwidth -* Monitoring Event Configuration (BMEC) is supported. * @cdp_enabled: CDP state of this resource * * Members of this structure are either private to the architecture
@@ -389,7 +387,6 @@ struct rdt_hw_resource {void (*msr_update)(struct msr_param *m);unsigned int mon_scale;unsigned int mbm_width; -unsigned int mbm_cfg_mask;bool cdp_enabled; };
arch/x86/kernel/cpu/resctrl/monitor.c +1 -1
@@ -1079,7 +1079,7 @@ int __init rdt_get_mon_l3_config(struct rdt_resource *r)/* Detect list of bandwidth sources that can be tracked */ cpuid_count (0x80000020 ,3 , &eax, &ebx, &ecx, &edx); - hw_res->mbm_cfg_mask = ecx &MAX_EVT_CONFIG_BITS ; + r->mbm_cfg_mask = ecx &MAX_EVT_CONFIG_BITS ; } r->mon_capable = true;
arch/x86/kernel/cpu/resctrl/rdtgroup.c +2 -3
@@ -1729,7 +1729,6 @@ static void mbm_config_write_domain(struct rdt_resource *r,static int mon_config_write (struct rdt_resource *r,char *tok,u32 evtid) { -struct rdt_hw_resource *hw_res =resctrl_to_arch_res (r);char *dom_str =NULL , *id_str;unsigned long dom_id, val;struct rdt_domain *d;
@@ -1756,9 +1755,9 @@ static int mon_config_write(struct rdt_resource *r, char *tok, u32 evtid) }/* Value from user cannot be more than the supported set of events */ -if ((val & hw_res->mbm_cfg_mask) != val) { +if ((val & r->mbm_cfg_mask) != val) {rdt_last_cmd_printf ("Invalid event configuration: max valid mask is 0x%02x\n" , - hw_res->mbm_cfg_mask); + r->mbm_cfg_mask);return -EINVAL ; }
include/linux/resctrl.h +3 -0
@@ -167,6 +167,8 @@ enum resctrl_schema_fmt {* @name: Name to use in "schemata" file. * @schema_fmt: Which format string and parser is used for this schema. * @evt_list: List of monitoring events +* @mbm_cfg_mask: Bandwidth sources that can be tracked when Bandwidth +* Monitoring Event Configuration (BMEC) is supported. * @cdp_capable: Is the CDP feature available on this resource */ struct rdt_resource {
@@ -181,6 +183,7 @@ struct rdt_resource {char *name;enum resctrl_schema_fmt schema_fmt;struct list_head evt_list; +unsigned int mbm_cfg_mask;bool cdp_capable; };
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 21/38] x86/resctrl: Move mbm_cfg_mask to struct rdt_resource
Hi James,
Quoted 10 lines · James Morse
The
mbm_cfg_maskfield lists the bits that user-space can set when configuring an event. This value is output via thelast_cmd_statusfile.Once the filesystem parts of resctrl are moved to live in /fs/, the struct
rdt_hw_resourceis inaccessible to the filesystem code. Because this value is output to user-space, it has to be accessible to the filesystem code.Move it to struct
rdt_resource.
Change looks good. Please do note that there is work in progress to consolidate the monitoring related data within a new struct that will impact this change:
https://lore.kernel.org/lkml/8f73c9ec4c9999c262d9297d46a03209a8affe3f.1716552602.git.babu.moger@amd.com/
Reinette
James Morse
reviewerPATCH v3 22/38 x86/resctrl: Add resctrl_arch_ prefix to pseudo lock functions
In reply to James Morseresctrl's pseudo lock has some copy-to-cache and measurement functions that are micro-architecture specific. pseudo_lock_fn() is not at all portable. Label these 'resctrl_arch_' so they stay under /arch/x86.
--- arch/x86/include/asm/resctrl.h | 5 ++++ arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 36 ++++++++++++----------- 2 files changed, 24 insertions(+), 17 deletions(-)
arch/x86/include/asm/resctrl.h +5 -0
@@ -211,6 +211,11 @@ static inline void *resctrl_arch_mon_ctx_alloc(struct rdt_resource *r, int evtidstatic inline void resctrl_arch_mon_ctx_free (struct rdt_resource *r,int evtid,void *ctx) { }; +u64 resctrl_arch_get_prefetch_disable_bits (void ); +int resctrl_arch_pseudo_lock_fn (void *_rdtgrp); +int resctrl_arch_measure_cycles_lat_fn (void *_plr); +int resctrl_arch_measure_l2_residency (void *_plr); +int resctrl_arch_measure_l3_residency (void *_plr);void resctrl_cpu_detect (struct cpuinfo_x86 *c);#else
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +19 -17
@@ -62,7 +62,8 @@ static const struct class pseudo_lock_class = { };/** -* get_prefetch_disable_bits - prefetch disable bits of supported platforms +* resctrl_arch_get_prefetch_disable_bits - prefetch disable bits of supported +* platforms * @void: It takes no parameters. * * Capture the list of platforms that have been validated to support
@@ -76,13 +77,13 @@ static const struct class pseudo_lock_class = {* in the SDM. * * When adding a platform here also add support for its cache events to -* measure_cycles_perf_fn() +* resctrl_arch_measure_l*_residency() * * Return: * If platform is supported, the bits to disable hardware prefetchers, 0 * if platform is not supported. */ -static u64 get_prefetch_disable_bits (void ) +u64 resctrl_arch_get_prefetch_disable_bits (void ) {if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL || boot_cpu_data.x86 !=6 )
@@ -410,7 +411,7 @@ static void pseudo_lock_free(struct rdtgroup *rdtgrp) }/** -* pseudo_lock_fn - Load kernel memory into cache +* resctrl_arch_pseudo_lock_fn - Load kernel memory into cache * @_rdtgrp: resource group to which pseudo-lock region belongs * * This is the core pseudo-locking flow.
@@ -428,7 +429,7 @@ static void pseudo_lock_free(struct rdtgroup *rdtgrp)* * Return: 0. Waiter on waitqueue will be woken on completion. */ -static int pseudo_lock_fn (void *_rdtgrp) +int resctrl_arch_pseudo_lock_fn (void *_rdtgrp) {struct rdtgroup *rdtgrp = _rdtgrp;struct pseudo_lock_region *plr = rdtgrp->plr;
@@ -714,7 +715,7 @@ int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp)* Not knowing the bits to disable prefetching implies that this * platform does not support Cache Pseudo-Locking. */ - prefetch_disable_bits =get_prefetch_disable_bits (); + prefetch_disable_bits =resctrl_arch_get_prefetch_disable_bits ();if (prefetch_disable_bits ==0 ) {rdt_last_cmd_puts ("Pseudo-locking not supported\n" );return -EINVAL ;
@@ -879,7 +880,8 @@ bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d) }/** -* measure_cycles_lat_fn - Measure cycle latency to read pseudo-locked memory +* resctrl_arch_measure_cycles_lat_fn - Measure cycle latency to read +* pseudo-locked memory * @_plr: pseudo-lock region to measure * * There is no deterministic way to test if a memory region is cached. One
@@ -892,7 +894,7 @@ bool rdtgroup_pseudo_locked_in_hierarchy(struct rdt_domain *d)* * Return: 0. Waiter on waitqueue will be woken on completion. */ -static int measure_cycles_lat_fn (void *_plr) +int resctrl_arch_measure_cycles_lat_fn (void *_plr) {struct pseudo_lock_region *plr = _plr;u32 saved_low, saved_high;
@@ -1076,7 +1078,7 @@ static int measure_residency_fn(struct perf_event_attr *miss_attr,return 0 ; } -static int measure_l2_residency (void *_plr) +int resctrl_arch_measure_l2_residency (void *_plr) {struct pseudo_lock_region *plr = _plr;struct residency_counts counts = {0 };
@@ -1114,7 +1116,7 @@ static int measure_l2_residency(void *_plr)return 0 ; } -static int measure_l3_residency (void *_plr) +int resctrl_arch_measure_l3_residency (void *_plr) {struct pseudo_lock_region *plr = _plr;struct residency_counts counts = {0 };
@@ -1212,18 +1214,18 @@ static int pseudo_lock_measure_cycles(struct rdtgroup *rdtgrp, int sel) plr->cpu = cpu;if (sel ==1 ) - thread =kthread_create_on_node (measure_cycles_lat_fn, plr, -cpu_to_node (cpu), + thread =kthread_create_on_node (resctrl_arch_measure_cycles_lat_fn, + plr,cpu_to_node (cpu),"pseudo_lock_measure/%u" , cpu);else if (sel ==2 ) - thread =kthread_create_on_node (measure_l2_residency, plr, -cpu_to_node (cpu), + thread =kthread_create_on_node (resctrl_arch_measure_l2_residency, + plr,cpu_to_node (cpu),"pseudo_lock_measure/%u" , cpu);else if (sel ==3 ) - thread =kthread_create_on_node (measure_l3_residency, plr, -cpu_to_node (cpu), + thread =kthread_create_on_node (resctrl_arch_measure_l3_residency, + plr,cpu_to_node (cpu),"pseudo_lock_measure/%u" , cpu);else
@@ -1322,7 +1324,7 @@ int rdtgroup_pseudo_lock_create(struct rdtgroup *rdtgrp) plr->thread_done =0 ; - thread =kthread_create_on_node (pseudo_lock_fn, rdtgrp, + thread =kthread_create_on_node (resctrl_arch_pseudo_lock_fn, rdtgrp,cpu_to_node (plr->cpu),"pseudo_lock/%u" , plr->cpu);if (IS_ERR (thread)) {
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 23/38 x86/resctrl: Allow an architecture to disable pseudo lock
In reply to James MorsePseudo-lock relies on knowledge of the micro-architecture to disable prefetchers etc.
On arm64 these controls are typically secure only, meaning linux can't access them. Arm's cache-lockdown feature works in a very different way. Resctrl's pseudo-lock isn't going to be used on arm64 platforms.
Add a Kconfig symbol that can be selected by the architecture. This enables or disables building of the pseudo_lock.c file, and replaces the functions with stubs. An additional IS_ENABLED() check is needed in rdtgroup_mode_write() so that attempting to enable pseudo-lock reports an "Unknown or unsupported mode" to user-space via the last_cmd file.
---
Changes since v2:
* Clarified the commit message as to where the error string is printed.
Changes since v1:
* [Commit message only] Typo fix:
s/psuedo/pseudo/g
--- arch/x86/Kconfig | 7 ++++ arch/x86/kernel/cpu/resctrl/Makefile | 5 +-- arch/x86/kernel/cpu/resctrl/internal.h | 48 +++++++++++++++++++++----- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 3 +- 4 files changed, 52 insertions(+), 11 deletions(-)
arch/x86/Kconfig +7 -0
@@ -492,6 +492,7 @@ config X86_CPU_RESCTRL depends on X86 && (CPU_SUP_INTEL || CPU_SUP_AMD) select KERNFS select PROC_CPU_RESCTRL if PROC_FS + select RESCTRL_FS_PSEUDO_LOCK help Enable x86 CPU resource control support.
@@ -508,6 +509,12 @@ config X86_CPU_RESCTRL Say N if unsure. +config RESCTRL_FS_PSEUDO_LOCK + bool + help + Software mechanism to pin data in a cache portion using + micro-architecture specific knowledge. + config X86_FRED bool "Flexible Return and Event Delivery" depends on X86_64
arch/x86/kernel/cpu/resctrl/Makefile +3 -2
@@ -1,4 +1,5 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_X86_CPU_RESCTRL) += core.o rdtgroup.o monitor.o -obj-$(CONFIG_X86_CPU_RESCTRL) += ctrlmondata.o pseudo_lock.o +obj-$(CONFIG_X86_CPU_RESCTRL) += core.o rdtgroup.o monitor.o +obj-$(CONFIG_X86_CPU_RESCTRL) += ctrlmondata.o +obj-$(CONFIG_RESCTRL_FS_PSEUDO_LOCK) += pseudo_lock.o CFLAGS_pseudo_lock.o = -I$(src)
arch/x86/kernel/cpu/resctrl/internal.h +40 -8
58 lines not on this page · raw message
arch/x86/kernel/cpu/resctrl/rdtgroup.c +2 -1
@@ -1450,7 +1450,8 @@ static ssize_t rdtgroup_mode_write(struct kernfs_open_file *of, goto out; } rdtgrp->mode =RDT_MODE_EXCLUSIVE ; - }else if (!strcmp (buf,"pseudo-locksetup" )) { + }else if (IS_ENABLED (CONFIG_RESCTRL_FS_PSEUDO_LOCK ) && + !strcmp (buf,"pseudo-locksetup" )) { ret =rdtgroup_locksetup_enter (rdtgrp);if (ret) goto out;
Signature · 1 line
2.39.2
Re: [PATCH v3 23/38] x86/resctrl: Allow an architecture to disable pseudo lock
arch/x86/kernel/cpu/resctrl/Makefile line 5
Quoted 12 lines · James Morse
arch/x86/kernel/cpu/resctrl/Makefile +3
-2@@ -1,4 +1,5 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_X86_CPU_RESCTRL) += core.o rdtgroup.o monitor.o -obj-$(CONFIG_X86_CPU_RESCTRL) += ctrlmondata.o pseudo_lock.o +obj-$(CONFIG_X86_CPU_RESCTRL) += core.o rdtgroup.o monitor.o +obj-$(CONFIG_X86_CPU_RESCTRL) += ctrlmondata.o +obj-$(CONFIG_RESCTRL_FS_PSEUDO_LOCK) += pseudo_lock.o CFLAGS_pseudo_lock.o = -I$(src)
Now that pseudo_lock.c is only conditionally compiled, the work it's doing to define tracepoints, (that is, #define CREATE_TRACE_POINTS), should be moved to monitor.c which is unconditionally compiled.
And then, the CFLAGS line above should be adjusted to apply to the compilation of monitor.c, that is:
CFLAGS_monitor.o = -I$(src)
Without these changes, compiling without CONFIG_RESCTRL_FS_PSEUDO_LOCK will fail due to undefined tracepoint functions.
-Carl
James Morse
reviewerRe: [PATCH v3 23/38] x86/resctrl: Allow an architecture to disable pseudo lock
Hi Carl,
Quoted 18 lines · Carl Worth
arch/x86/kernel/cpu/resctrl/Makefile line 5
Quoted 12 lines · James Morse
arch/x86/kernel/cpu/resctrl/Makefile +3
-2@@ -1,4 +1,5 @@ # SPDX-License-Identifier: GPL-2.0 -obj-$(CONFIG_X86_CPU_RESCTRL) += core.o rdtgroup.o monitor.o -obj-$(CONFIG_X86_CPU_RESCTRL) += ctrlmondata.o pseudo_lock.o +obj-$(CONFIG_X86_CPU_RESCTRL) += core.o rdtgroup.o monitor.o +obj-$(CONFIG_X86_CPU_RESCTRL) += ctrlmondata.o +obj-$(CONFIG_RESCTRL_FS_PSEUDO_LOCK) += pseudo_lock.o CFLAGS_pseudo_lock.o = -I$(src)Now that
pseudo_lock.cis only conditionally compiled, the work it's doing to define tracepoints, (that is, #defineCREATE_TRACE_POINTS), should be moved tomonitor.cwhich is unconditionally compiled.
Quoted 7 lines
And then, the CFLAGS line above should be adjusted to apply to the compilation of
monitor.c, that is:CFLAGS_monitor.o = -I$(src)
Without these changes, compiling without
CONFIG_RESCTRL_FS_PSEUDO_LOCKwill fail due to undefined tracepoint functions.
Thanks for catching this - I bashed my head against it for a while.
I've split trace.h in two: monitor_trace.h and pseudo_lock_trace.h. When the code gets moved to /fs/ only one of those files moves.
Thanks,
James
James Morse
reviewerPATCH v3 24/38 x86/resctrl: Make prefetch_disable_bits belong to the arch code
In reply to James Morseprefetch_disable_bits is set by rdtgroup_locksetup_enter() from a value provided by the architecture, but is largely read by other architecture helpers.
Instead of exporting this value, make resctrl_arch_get_prefetch_disable_bits() set it so that the other arch-code helpers can use the cached-value.
--- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-)
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +8 -5
@@ -85,6 +85,8 @@ static const struct class pseudo_lock_class = {*/ u64 resctrl_arch_get_prefetch_disable_bits (void ) { + prefetch_disable_bits =0 ; +if (boot_cpu_data.x86_vendor !=X86_VENDOR_INTEL || boot_cpu_data.x86 !=6 )return 0 ;
@@ -100,7 +102,8 @@ u64 resctrl_arch_get_prefetch_disable_bits(void)* 3 DCU IP Prefetcher Disable (R/W) * 63:4 Reserved */ -return 0xF ; + prefetch_disable_bits =0xF ; +break ;case INTEL_ATOM_GOLDMONT :case INTEL_ATOM_GOLDMONT_PLUS :/*
@@ -111,10 +114,11 @@ u64 resctrl_arch_get_prefetch_disable_bits(void)* 2 DCU Hardware Prefetcher Disable (R/W) * 63:3 Reserved */ -return 0x5 ; + prefetch_disable_bits =0x5 ; +break ; } -return 0 ; +return prefetch_disable_bits; }/**
@@ -715,8 +719,7 @@ int rdtgroup_locksetup_enter(struct rdtgroup *rdtgrp)* Not knowing the bits to disable prefetching implies that this * platform does not support Cache Pseudo-Locking. */ - prefetch_disable_bits =resctrl_arch_get_prefetch_disable_bits (); -if (prefetch_disable_bits ==0 ) { +if (resctrl_arch_get_prefetch_disable_bits () ==0 ) {rdt_last_cmd_puts ("Pseudo-locking not supported\n" );return -EINVAL ; }
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 25/38 x86/resctrl: Make resctrl_arch_pseudo_lock_fn() take a plr
In reply to James Morseresctrl_arch_pseudo_lock_fn() has architecture specific behaviour, and takes a struct rdtgroup as an argument.
After the filesystem code moves to /fs/, the definition of struct rdtgroup will not be available to the architecture code.
The only reason resctrl_arch_pseudo_lock_fn() wants the rdtgroup is for the CLOSID. Embed that in the pseudo_lock_region as a closid, and move the definition of struct pseudo_lock_region to resctrl.h.
---
Change since v1:
* [Commit message only] Typo fix:
s/hw_closid/closid/g
--- arch/x86/include/asm/resctrl.h | 2 +- arch/x86/kernel/cpu/resctrl/internal.h | 37 --------------------- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 13 ++++---- include/linux/resctrl.h | 39 +++++++++++++++++++++++ 4 files changed, 47 insertions(+), 44 deletions(-)
arch/x86/include/asm/resctrl.h +1 -1
@@ -212,7 +212,7 @@ static inline void resctrl_arch_mon_ctx_free(struct rdt_resource *r, int evtid, void *ctx) { };u64 resctrl_arch_get_prefetch_disable_bits (void ); -int resctrl_arch_pseudo_lock_fn (void *_rdtgrp); +int resctrl_arch_pseudo_lock_fn (void *_plr);int resctrl_arch_measure_cycles_lat_fn (void *_plr);int resctrl_arch_measure_l2_residency (void *_plr);int resctrl_arch_measure_l3_residency (void *_plr);
arch/x86/kernel/cpu/resctrl/internal.h +0 -37
@@ -183,43 +183,6 @@ struct mongroup {u32 rmid; }; -/** -* struct pseudo_lock_region - pseudo-lock region information -* @s: Resctrl schema for the resource to which this -* pseudo-locked region belongs -* @d: RDT domain to which this pseudo-locked region -* belongs -* @cbm: bitmask of the pseudo-locked region -* @lock_thread_wq: waitqueue used to wait on the pseudo-locking thread -* completion -* @thread_done: variable used by waitqueue to test if pseudo-locking -* thread completed -* @cpu: core associated with the cache on which the setup code -* will be run -* @line_size: size of the cache lines -* @size: size of pseudo-locked region in bytes -* @kmem: the kernel memory associated with pseudo-locked region -* @minor: minor number of character device associated with this -* region -* @debugfs_dir: pointer to this region's directory in the debugfs -* filesystem -* @pm_reqs: Power management QoS requests related to this region -*/ -struct pseudo_lock_region { -struct resctrl_schema *s; -struct rdt_domain *d; -u32 cbm; -wait_queue_head_t lock_thread_wq; -int thread_done; -int cpu; -unsigned int line_size; -unsigned int size; -void *kmem; -unsigned int minor; -struct dentry *debugfs_dir; -struct list_head pm_reqs; -}; -/** * struct rdtgroup - store rdtgroup's data in resctrl file system. * @kn: kernfs node
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +7 -6
@@ -416,7 +416,7 @@ static void pseudo_lock_free(struct rdtgroup *rdtgrp)/** * resctrl_arch_pseudo_lock_fn - Load kernel memory into cache -* @_rdtgrp: resource group to which pseudo-lock region belongs +* @_plr: the pseudo-lock region descriptor * * This is the core pseudo-locking flow. *
@@ -433,10 +433,9 @@ static void pseudo_lock_free(struct rdtgroup *rdtgrp)* * Return: 0. Waiter on waitqueue will be woken on completion. */ -int resctrl_arch_pseudo_lock_fn (void *_rdtgrp) +int resctrl_arch_pseudo_lock_fn (void *_plr) { -struct rdtgroup *rdtgrp = _rdtgrp; -struct pseudo_lock_region *plr = rdtgrp->plr; +struct pseudo_lock_region *plr = _plr;u32 rmid_p, closid_p;unsigned long i;u64 saved_msr;
@@ -496,7 +495,8 @@ int resctrl_arch_pseudo_lock_fn(void *_rdtgrp)* pseudo-locked followed by reading of kernel memory to load it * into the cache. */ -__wrmsr (MSR_IA32_PQR_ASSOC , rmid_p, rdtgrp->closid); +__wrmsr (MSR_IA32_PQR_ASSOC , rmid_p, plr->closid); +/* * Cache was flushed earlier. Now access kernel memory to read it * into cache region associated with just activated plr->closid.
@@ -1327,7 +1327,8 @@ int rdtgroup_pseudo_lock_create(struct rdtgroup *rdtgrp) plr->thread_done =0 ; - thread =kthread_create_on_node (resctrl_arch_pseudo_lock_fn, rdtgrp, + plr->closid = rdtgrp->closid; + thread =kthread_create_on_node (resctrl_arch_pseudo_lock_fn, plr,cpu_to_node (plr->cpu),"pseudo_lock/%u" , plr->cpu);if (IS_ERR (thread)) {
include/linux/resctrl.h +39 -0
@@ -39,6 +39,45 @@ enum resctrl_conf_type {#define CDP_NUM_TYPES (CDP_DATA + 1) +/* +* struct pseudo_lock_region - pseudo-lock region information +* @s: Resctrl schema for the resource to which this +* pseudo-locked region belongs +* @closid: The closid that this pseudo-locked region uses +* @d: RDT domain to which this pseudo-locked region +* belongs +* @cbm: bitmask of the pseudo-locked region +* @lock_thread_wq: waitqueue used to wait on the pseudo-locking thread +* completion +* @thread_done: variable used by waitqueue to test if pseudo-locking +* thread completed +* @cpu: core associated with the cache on which the setup code +* will be run +* @line_size: size of the cache lines +* @size: size of pseudo-locked region in bytes +* @kmem: the kernel memory associated with pseudo-locked region +* @minor: minor number of character device associated with this +* region +* @debugfs_dir: pointer to this region's directory in the debugfs +* filesystem +* @pm_reqs: Power management QoS requests related to this region +*/ +struct pseudo_lock_region { +struct resctrl_schema *s; +u32 closid; +struct rdt_domain *d; +u32 cbm; +wait_queue_head_t lock_thread_wq; +int thread_done; +int cpu; +unsigned int line_size; +unsigned int size; +void *kmem; +unsigned int minor; +struct dentry *debugfs_dir; +struct list_head pm_reqs; +}; +/** * struct resctrl_staged_config - parsed configuration to be applied * @new_ctrl: new ctrl value to be loaded
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 26/38 x86/resctrl: Move thread_throttle_mode_init() to be managed by resctrl
In reply to James Morsethread_throttle_mode_init() is called from the architecture specific code to make the 'thread_throttle_mode' file visible. The architecture specific code has already set the membw.throttle_mode in the rdt_resource.
This doesn't need to be specific to the architecture, the throttle_mode can be used by resctrl to determine if the 'thread_throttle_mode' file should be visible.
Call thread_throttle_mode_init() from resctrl_setup(), check the membw.throttle_mode on the MBA resource. This avoids publishing an extra function between the architecture and filesystem code.
--- arch/x86/kernel/cpu/resctrl/core.c | 1 - arch/x86/kernel/cpu/resctrl/internal.h | 1 - arch/x86/kernel/cpu/resctrl/rdtgroup.c | 9 ++++++++- 3 files changed, 8 insertions(+), 3 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +0 -1
@@ -222,7 +222,6 @@ static bool __get_mem_config_intel(struct rdt_resource *r) r->membw.throttle_mode =THREAD_THROTTLE_PER_THREAD ; else r->membw.throttle_mode = THREAD_THROTTLE_MAX; -thread_throttle_mode_init (); r->alloc_capable = true;
arch/x86/kernel/cpu/resctrl/internal.h +0 -1
@@ -480,7 +480,6 @@ void cqm_handle_limbo(struct work_struct *work);bool has_busy_rmid (struct rdt_domain *d);void __check_limbo (struct rdt_domain *d,bool force_free);void rdt_domain_reconfigure_cdp (struct rdt_resource *r); -void __init thread_throttle_mode_init (void );void mbm_config_rftype_init (const char *config);void rdt_staged_configs_clear (void );bool closid_allocated (unsigned int closid);
arch/x86/kernel/cpu/resctrl/rdtgroup.c +8 -1
@@ -2055,10 +2055,15 @@ static struct rftype *rdtgroup_get_rftype_by_name(const char *name)return NULL ; } -void __init thread_throttle_mode_init (void ) +static void __init thread_throttle_mode_init (void ) { +struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_MBA );struct rftype *rft; +if (!r->alloc_capable || + r->membw.throttle_mode ==THREAD_THROTTLE_UNDEFINED ) +return ; + rft =rdtgroup_get_rftype_by_name ("thread_throttle_mode" );if (!rft)return ;
@@ -4200,6 +4205,8 @@ int __init resctrl_init(void)rdtgroup_setup_default (); +thread_throttle_mode_init (); + ret =resctrl_mon_resource_init ();if (ret)return ret;
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 27/38 x86/resctrl: Move get_config_index() to a header
In reply to James Morseget_config_index() is used by the architecture specific code to map a CLOSID+type pair to an index in the configuration arrays.
MPAM needs to do this too to preserve the ABI to user-space, there is no reason to do it differently.
Move the helper to a header file.
---
Changes since v1:
* Reindent resctrl_get_config_index() as per coding-style.rst rules.
* Remove redundant parentheses from arithmetic in resctrl_get_config_index(), so as to match the original source version of this moved code.
No functional change.
--- arch/x86/kernel/cpu/resctrl/ctrlmondata.c | 19 +++---------------- include/linux/resctrl.h | 15 +++++++++++++++ 2 files changed, 18 insertions(+), 16 deletions(-)
arch/x86/kernel/cpu/resctrl/ctrlmondata.c +3 -16
@@ -283,25 +283,12 @@ static int parse_line(char *line, struct resctrl_schema *s,return -EINVAL ; } -static u32 get_config_index (u32 closid,enum resctrl_conf_type type) -{ -switch (type) { -default : -case CDP_NONE : -return closid; -case CDP_CODE : -return closid *2 +1 ; -case CDP_DATA : -return closid *2 ; - } -} -int resctrl_arch_update_one (struct rdt_resource *r,struct rdt_domain *d,u32 closid,enum resctrl_conf_type t,u32 cfg_val) {struct rdt_hw_resource *hw_res =resctrl_to_arch_res (r);struct rdt_hw_domain *hw_dom =resctrl_to_arch_dom (d); -u32 idx =get_config_index (closid, t); +u32 idx =resctrl_get_config_index (closid, t);struct msr_param msr_param;if (!cpumask_test_cpu (smp_processor_id (), &d->cpu_mask))
@@ -338,7 +325,7 @@ int resctrl_arch_update_domains(struct rdt_resource *r, u32 closid)if (!cfg->have_new_ctrl)continue ; - idx =get_config_index (closid, t); + idx =resctrl_get_config_index (closid, t);if (cfg->new_ctrl == hw_dom->ctrl_val[idx])continue ; hw_dom->ctrl_val[idx] = cfg->new_ctrl;
@@ -458,7 +445,7 @@ u32 resctrl_arch_get_config(struct rdt_resource *r, struct rdt_domain *d, u32 closid, enum resctrl_conf_type type) {struct rdt_hw_domain *hw_dom =resctrl_to_arch_dom (d); -u32 idx =get_config_index (closid, type); +u32 idx =resctrl_get_config_index (closid, type);return hw_dom->ctrl_val[idx]; }
include/linux/resctrl.h +15 -0
@@ -315,6 +315,21 @@ bool resctrl_arch_is_evt_configurable(enum resctrl_event_id evt);void resctrl_arch_mon_event_config_write (void *info);void resctrl_arch_mon_event_config_read (void *info); +/* For use by arch code to remap resctrl's smaller CDP CLOSID range */ +static inline u32 resctrl_get_config_index (u32 closid, +enum resctrl_conf_type type) +{ +switch (type) { +default : +case CDP_NONE : +return closid; +case CDP_CODE : +return closid *2 +1 ; +case CDP_DATA : +return closid *2 ; + } +} +/* * Update the ctrl_val and apply this config right now. * Must be called on one of the domain's CPUs.
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 28/38 x86/resctrl: Claim get_domain_from_cpu() for resctrl
In reply to James Morseget_domain_from_cpu() is a handy helper that both the arch code and resctrl need to use. Rename it resctrl_get_domain_from_cpu() so it gets moved out to /fs, and exported back to the arch code.
--- arch/x86/kernel/cpu/resctrl/core.c | 15 --------------- arch/x86/kernel/cpu/resctrl/internal.h | 1 - arch/x86/kernel/cpu/resctrl/monitor.c | 2 +- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 2 +- include/linux/resctrl.h | 21 +++++++++++++++++++++ 5 files changed, 23 insertions(+), 18 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +0 -15
@@ -343,21 +343,6 @@ static void cat_wrmsr(struct msr_param *m)wrmsrl (hw_res->msr_base + i, hw_dom->ctrl_val[i]); } -struct rdt_domain *get_domain_from_cpu (int cpu,struct rdt_resource *r) -{ -struct rdt_domain *d; - -lockdep_assert_cpus_held (); - -list_for_each_entry (d, &r->domains, list) { -/* Find the domain that contains this CPU */ -if (cpumask_test_cpu (cpu, &d->cpu_mask)) -return d; - } - -return NULL ; -} -u32 resctrl_arch_get_num_closid (struct rdt_resource *r) {return resctrl_to_arch_res (r)->num_closid;
arch/x86/kernel/cpu/resctrl/internal.h +0 -1
@@ -454,7 +454,6 @@ unsigned int rdtgroup_cbm_to_size(struct rdt_resource *r, struct rdt_domain *d, unsigned long cbm);enum rdtgrp_mode rdtgroup_mode_by_closid (int closid);int rdtgroup_tasks_assigned (struct rdtgroup *r); -struct rdt_domain *get_domain_from_cpu (int cpu,struct rdt_resource *r);int closids_supported (void );void closid_free (int closid);int alloc_rmid (u32 closid);
arch/x86/kernel/cpu/resctrl/monitor.c +1 -1
@@ -687,7 +687,7 @@ static void update_mba_bw(struct rdtgroup *rgrp, struct rdt_domain *dom_mbm) idx =resctrl_arch_rmid_idx_encode (closid, rmid); pmbm_data = &dom_mbm->mbm_local[idx]; - dom_mba =get_domain_from_cpu (smp_processor_id (), r_mba); + dom_mba =resctrl_get_domain_from_cpu (smp_processor_id (), r_mba);if (!dom_mba) {pr_warn_once ("Failure to get domain for MBA update\n" );return ;
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1 -1
@@ -4171,7 +4171,7 @@ void resctrl_offline_cpu(unsigned int cpu)if (!l3->mon_capable) goto out_unlock; - d =get_domain_from_cpu (cpu, l3); + d =resctrl_get_domain_from_cpu (cpu, l3);if (d) {if (resctrl_is_mbm_enabled () && cpu == d->mbm_work_cpu) {cancel_delayed_work (&d->mbm_over);
include/linux/resctrl.h +21 -0
@@ -2,6 +2,7 @@#ifndef _RESCTRL_H#define _RESCTRL_H +#include <linux/cpu.h> #include <linux/kernel.h> #include <linux/list.h> #include <linux/pid.h>
@@ -330,6 +331,26 @@ static inline u32 resctrl_get_config_index(u32 closid, } } +/* +* Caller must hold the cpuhp read lock to prevent the struct rdt_domain being +* freed. +*/ +static inline struct rdt_domain * +resctrl_get_domain_from_cpu (int cpu,struct rdt_resource *r) +{ +struct rdt_domain *d; + +lockdep_assert_cpus_held (); + +list_for_each_entry (d, &r->domains, list) { +/* Find the domain that contains this CPU */ +if (cpumask_test_cpu (cpu, &d->cpu_mask)) +return d; + } + +return NULL ; +} +/* * Update the ctrl_val and apply this config right now. * Must be called on one of the domain's CPUs.
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 29/38 x86/resctrl: Describe resctrl's bitmap size assumptions
In reply to James Morseresctrl operates on configuration bitmaps and a bitmap of allocated CLOSID, both are stored in a u32.
MPAM supports configuration/portion bitmaps and PARTIDs larger than will fit in a u32.
Add some preprocessor values that make it clear why MPAM clamps some of these values. This will make it easier to find code related to these values if this resctrl behaviour ever changes.
--- include/linux/resctrl.h | 11 +++++++++++ 1 file changed, 11 insertions(+)
include/linux/resctrl.h +11 -0
@@ -26,6 +26,17 @@ int proc_resctrl_show(struct seq_file *m,/* max value for struct rdt_domain's mbps_val */ #define MBA_MAX_MBPS U32_MAX +/* +* Resctrl uses a u32 as a closid bitmap. The maximum closid is 32. +*/ +#define RESCTRL_MAX_CLOSID 32 + +/* +* Resctrl uses u32 to hold the user-space config. The maximum bitmap size is +* 32. +*/ +#define RESCTRL_MAX_CBM 32 +/** * enum resctrl_conf_type - The type of configuration. * @CDP_NONE: No prioritisation, both code and data are controlled or monitored.
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 30/38 x86/resctrl: Rename resctrl_sched_in() to begin with "resctrl_arch_"
In reply to James Morseresctrl_sched_in() loads the architecture specific CPU MSRs with the CLOSID and RMID values. This function was named before resctrl was split to have architecture specific code, and generic filesystem code.
This function is obviously architecture specific, but does not begin with 'resctrl_arch_', making it the odd one out in the functions an architecture needs to support to enable resctrl.
Rename it for consistency. This is purely cosmetic.
--- arch/x86/include/asm/resctrl.h | 4 ++-- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 12 ++++++------ arch/x86/kernel/process_32.c | 2 +- arch/x86/kernel/process_64.c | 2 +- 4 files changed, 10 insertions(+), 10 deletions(-)
arch/x86/include/asm/resctrl.h +2 -2
@@ -177,7 +177,7 @@ static inline bool resctrl_arch_match_rmid(struct task_struct *tsk, u32 ignored,return READ_ONCE (tsk->rmid) == rmid; } -static inline void resctrl_sched_in (struct task_struct *tsk) +static inline void resctrl_arch_sched_in (struct task_struct *tsk) {if (static_branch_likely (&rdt_enable_key))__resctrl_sched_in (tsk);
@@ -220,7 +220,7 @@ void resctrl_cpu_detect(struct cpuinfo_x86 *c);#else -static inline void resctrl_sched_in (struct task_struct *tsk) {} +static inline void resctrl_arch_sched_in (struct task_struct *tsk) {}static inline void resctrl_cpu_detect (struct cpuinfo_x86 *c) {}#endif /* CONFIG_X86_CPU_RESCTRL */
arch/x86/kernel/cpu/resctrl/rdtgroup.c +6 -6
@@ -359,7 +359,7 @@ static int rdtgroup_cpus_show(struct kernfs_open_file *of, }/* -* This is safe against resctrl_sched_in() called from __switch_to() +* This is safe against resctrl_arch_sched_in() called from __switch_to() * because __switch_to() is executed with interrupts disabled. A local call * from update_closid_rmid() is protected against __switch_to() because * preemption is disabled.
@@ -378,7 +378,7 @@ void resctrl_arch_sync_cpu_closid_rmid(void *info)* executing task might have its own closid selected. Just reuse * the context switch code. */ -resctrl_sched_in (current); +resctrl_arch_sched_in (current); }/*
@@ -603,7 +603,7 @@ static void _update_task_closid_rmid(void *task)* Otherwise, the MSR is updated when the task is scheduled in. */ if (task == current) -resctrl_sched_in (task); +resctrl_arch_sched_in (task); }static void update_task_closid_rmid (struct task_struct *t)
@@ -661,7 +661,7 @@ static int __rdtgroup_move_task(struct task_struct *tsk,* Ensure the task's closid and rmid are written before determining if * the task is current that will decide if it will be interrupted. * This pairs with the full barrier between the rq->curr update and -* resctrl_sched_in() during context switch. +* resctrl_arch_sched_in() during context switch. */ smp_mb ();
@@ -2949,8 +2949,8 @@ static void rdt_move_group_tasks(struct rdtgroup *from, struct rdtgroup *to,/* * Order the closid/rmid stores above before the loads * in task_curr(). This pairs with the full barrier -* between the rq->curr update and resctrl_sched_in() -* during context switch. +* between the rq->curr update and +* resctrl_arch_sched_in() during context switch. */ smp_mb ();
arch/x86/kernel/process_32.c +1 -1
@@ -211,7 +211,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)switch_fpu_finish (next_p);/* Load the Intel cache allocation PQR MSR. */ -resctrl_sched_in (next_p); +resctrl_arch_sched_in (next_p);return prev_p; }
arch/x86/kernel/process_64.c +1 -1
@@ -707,7 +707,7 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) }/* Load the Intel cache allocation PQR MSR. */ -resctrl_sched_in (next_p); +resctrl_arch_sched_in (next_p);return prev_p; }
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 31/38 x86/resctrl: resctrl_exit() teardown resctrl but leave the mount point
In reply to James Morseresctrl_exit() was intended for use when the 'resctrl' module was unloaded. resctrl can't be built as a module, and the kernfs helpers are not exported so this is unlikely to change. MPAM has an error interrupt which indicates the MPAM driver has gone haywire. Should this occur tasks could run with the wrong control values, leading to bad performance for impoartant tasks. The MPAM driver needs a way to tell resctrl that no further configuration should be attempted.
Using resctrl_exit() for this leaves the system in a funny state as resctrl is still mounted, but cannot be un-mounted because the sysfs directory that is typically used has been removed. Dave Martin suggests this may cause systemd trouble in the future as not all filesystems can be unmounted.
Add calls to remove all the files and directories in resctrl, and remove the sysfs_remove_mount_point() call that leaves the system in a funny state. When triggered, this causes all the resctrl files to disappear. resctrl can be unmounted, but not mounted again.
--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1 -1
@@ -4255,9 +4255,9 @@ int __init resctrl_init(void) void __exit resctrl_exit(void) { +rdtgroup_destroy_root ();debugfs_remove_recursive (debugfs_resctrl);unregister_filesystem (&rdt_fs_type); -sysfs_remove_mount_point (fs_kobj,"resctrl" );resctrl_mon_resource_exit (); }
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 31/38] x86/resctrl: resctrl_exit() teardown resctrl but leave the mount point
Hi James,
Quoted 5 lines · James Morse
resctrl_exit()was intended for use when the 'resctrl' module was unloaded. resctrl can't be built as a module, and the kernfs helpers are not exported so this is unlikely to change. MPAM has an error interrupt which indicates the MPAM driver has gone haywire. Should this occur tasks could run with the wrong control values, leading to bad performance for impoartant tasks.
impoartant -> important
Quoted 9 lines
The MPAM driver needs a way to tell resctrl that no further configuration should be attempted.
Using
resctrl_exit()for this leaves the system in a funny state as resctrl is still mounted, but cannot be un-mounted because the sysfs directory that is typically used has been removed. Dave Martin suggests this may cause systemd trouble in the future as not all filesystems can be unmounted.
Quoted 4 lines
Add calls to remove all the files and directories in resctrl, and remove the
sysfs_remove_mount_point()call that leaves the system in a funny state. When triggered, this causes all the resctrl files to disappear. resctrl can be unmounted, but not mounted again.
I am not familiar with these flows so I would like to confirm ... In this scenario the resctrl filesystem will be unregistered, are you saying that it is possible to unmount a filesystem after it has been unregistered?
Reinette
James Morse
reviewerRe: [PATCH v3 31/38] x86/resctrl: resctrl_exit() teardown resctrl but leave the mount point
Hi Reinette,
Quoted 23 lines · Reinette Chatre
Quoted 5 lines · James Morse
resctrl_exit()was intended for use when the 'resctrl' module was unloaded. resctrl can't be built as a module, and the kernfs helpers are not exported so this is unlikely to change. MPAM has an error interrupt which indicates the MPAM driver has gone haywire. Should this occur tasks could run with the wrong control values, leading to bad performance for impoartant tasks.impoartant -> important
Quoted 9 lines
The MPAM driver needs a way to tell resctrl that no further configuration should be attempted.
Using
resctrl_exit()for this leaves the system in a funny state as resctrl is still mounted, but cannot be un-mounted because the sysfs directory that is typically used has been removed. Dave Martin suggests this may cause systemd trouble in the future as not all filesystems can be unmounted.Quoted 4 lines
Add calls to remove all the files and directories in resctrl, and remove the
sysfs_remove_mount_point()call that leaves the system in a funny state. When triggered, this causes all the resctrl files to disappear. resctrl can be unmounted, but not mounted again.
Quoted 4 lines
I am not familiar with these flows so I would like to confirm ... In this scenario the resctrl filesystem will be unregistered, are you saying that it is possible to unmount a filesystem after it has been unregistered?
Counter-intuitively: yes.
The rules are described in fs/filesystems.c: We can access the members of the struct file_system_type if the list lock is held, or a reference is held to the module. This is how /proc/mounts is able to print the filesystem name from struct file_system_type without taking the lock - it holds a reference to any module to prevent the structure from being freed. Because resctrl can't be built as a module, we can say there is always a reference held, and we can never free struct file_system_type.
Thanks,
James
Reinette Chatre
maintainerRe: [PATCH v3 31/38] x86/resctrl: resctrl_exit() teardown resctrl but leave the mount point
Hi James,
Quoted 31 lines · James Morse
Hi Reinette,
Quoted 23 lines · Reinette Chatre
Quoted 5 lines · James Morse
resctrl_exit()was intended for use when the 'resctrl' module was unloaded. resctrl can't be built as a module, and the kernfs helpers are not exported so this is unlikely to change. MPAM has an error interrupt which indicates the MPAM driver has gone haywire. Should this occur tasks could run with the wrong control values, leading to bad performance for impoartant tasks.impoartant -> important
Quoted 9 lines
The MPAM driver needs a way to tell resctrl that no further configuration should be attempted.
Using
resctrl_exit()for this leaves the system in a funny state as resctrl is still mounted, but cannot be un-mounted because the sysfs directory that is typically used has been removed. Dave Martin suggests this may cause systemd trouble in the future as not all filesystems can be unmounted.Quoted 4 lines
Add calls to remove all the files and directories in resctrl, and remove the
sysfs_remove_mount_point()call that leaves the system in a funny state. When triggered, this causes all the resctrl files to disappear. resctrl can be unmounted, but not mounted again.Quoted 4 lines
I am not familiar with these flows so I would like to confirm ... In this scenario the resctrl filesystem will be unregistered, are you saying that it is possible to unmount a filesystem after it has been unregistered?
Quoted 6 lines
Counter-intuitively: yes.
The rules are described in
fs/filesystems.c: We can access the members of the structfile_system_typeif the list lock is held, or a reference is held to the module. This is how/proc/mountsis able to print the filesystem name from structfile_system_typewithout taking the lock - it holds a reference to any module to prevent the structure from being
hmmm ... does this mean I am supposed to find calls to try_module_get() in the flow from mounts_open_common()?
Quoted 2 lines
freed. Because resctrl can't be built as a module, we can say there is always a reference held, and we can never free struct
file_system_type.
unregister_filesystem() continues to be called and as I understand in new MPAM usages will be called during runtime. unregister_filesystem() comments state "Once this function has returned the &struct file_system_type structure may be freed or reused.". Could you please highlight to me what gives the confidence of "we can say there is always a reference held"? Could you please point to me where that reference is obtained that will prevent the structure from being freed?
Thank you
Reinette
James Morse
reviewerRe: [PATCH v3 31/38] x86/resctrl: resctrl_exit() teardown resctrl but leave the mount point
Hi Reinette,
Quoted 32 lines · Reinette Chatre
Quoted 31 lines · James Morse
Quoted 23 lines · Reinette Chatre
Quoted 5 lines · James Morse
resctrl_exit()was intended for use when the 'resctrl' module was unloaded. resctrl can't be built as a module, and the kernfs helpers are not exported so this is unlikely to change. MPAM has an error interrupt which indicates the MPAM driver has gone haywire. Should this occur tasks could run with the wrong control values, leading to bad performance for impoartant tasks.impoartant -> important
Quoted 9 lines
The MPAM driver needs a way to tell resctrl that no further configuration should be attempted.
Using
resctrl_exit()for this leaves the system in a funny state as resctrl is still mounted, but cannot be un-mounted because the sysfs directory that is typically used has been removed. Dave Martin suggests this may cause systemd trouble in the future as not all filesystems can be unmounted.Quoted 4 lines
Add calls to remove all the files and directories in resctrl, and remove the
sysfs_remove_mount_point()call that leaves the system in a funny state. When triggered, this causes all the resctrl files to disappear. resctrl can be unmounted, but not mounted again.Quoted 4 lines
I am not familiar with these flows so I would like to confirm ... In this scenario the resctrl filesystem will be unregistered, are you saying that it is possible to unmount a filesystem after it has been unregistered?
Counter-intuitively: yes.
Quoted 7 lines
Quoted 4 lines
The rules are described in
fs/filesystems.c: We can access the members of the structfile_system_typeif the list lock is held, or a reference is held to the module. This is how/proc/mountsis able to print the filesystem name from structfile_system_typewithout taking the lock - it holds a reference to any module to prevent the structure from beinghmmm ... does this mean I am supposed to find calls to
try_module_get()in the flow frommounts_open_common()?
There may be, but when a filesystem is mounted the code in super.c holds a reference to the filesystem - which translates to a reference on the module/filesystem->owner.
My point was only that its possible to unregister a filesystem while its mounted. The reference counting takes care of this - and is unnecessary in our case.
Quoted 11 lines
Quoted 2 lines
freed. Because resctrl can't be built as a module, we can say there is always a reference held, and we can never free struct
file_system_type.
unregister_filesystem()continues to be called and as I understand in new MPAM usages will be called during runtime.unregister_filesystem()comments state "Once this function has returned
the &structfile_system_typestructure may be freed or reused.". Could you please highlight to me
what gives the confidence of "we can say there is always a reference held"? Could you please point to me where that reference is obtained that will prevent the structure from being freed?
I think we are rat-holing on something that doesn't matter:
* resctrl can't be built as a module - it is always built in.
* rdt_fs_type is therefore part of the kernel data section - it can't be freed.
* likewise the code that is part of resctrl can't be freed either.
Thanks,
James
James Morse
reviewerPATCH v3 32/38 x86/resctrl: Drop __init/__exit on assorted symbols
In reply to James MorseBecause ARM's MPAM controls are probed using MMIO, resctrl can't be initialised until enough CPUs are online to have determined the system-wide supported num_closid. Arm64 also supports 'late onlined secondaries', where only a subset of CPUs are online during boot.
These two combine to mean the MPAM driver may not be able to initialise resctrl until user-space has brought 'enough' CPUs online.
To allow MPAM to initialise resctrl after __init text has been free'd, remove all the __init markings from resctrl.
The existing __exit markings cause these functions to be removed by the linker as it has never been possible to build resctrl as a module. MPAM has an error interrupt which causes the driver to reset and disable itself. Remove the __exit markings to allow the MPAM driver to tear down resctrl when an error occurs.
--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 8 ++++---- include/linux/resctrl.h | 4 ++-- 2 files changed, 6 insertions(+), 6 deletions(-)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +4 -4
@@ -2055,7 +2055,7 @@ static struct rftype *rdtgroup_get_rftype_by_name(const char *name)return NULL ; } -static void __init thread_throttle_mode_init (void ) +static void thread_throttle_mode_init (void ) {struct rdt_resource *r =resctrl_arch_get_resource (RDT_RESOURCE_MBA );struct rftype *rft;
@@ -4003,7 +4003,7 @@ static void rdtgroup_destroy_root(void) rdtgroup_default.kn =NULL ; } -static void __init rdtgroup_setup_default (void ) +static void rdtgroup_setup_default (void ) {mutex_lock (&rdtgroup_mutex);
@@ -4196,7 +4196,7 @@ void resctrl_offline_cpu(unsigned int cpu)* * Return: 0 on success or -errno */ -int __init resctrl_init (void ) +int resctrl_init (void ) {int ret =0 ;
@@ -4253,7 +4253,7 @@ int __init resctrl_init(void)return ret; } -void __exit resctrl_exit(void) +void resctrl_exit (void ) {rdtgroup_destroy_root ();debugfs_remove_recursive (debugfs_resctrl);
include/linux/resctrl.h +2 -2
@@ -452,7 +452,7 @@ void resctrl_arch_reset_rmid_all(struct rdt_resource *r, struct rdt_domain *d);extern unsigned int resctrl_rmid_realloc_threshold;extern unsigned int resctrl_rmid_realloc_limit; -int __init resctrl_init (void ); -void __exit resctrl_exit(void); +int resctrl_init (void ); +void resctrl_exit (void );#endif /* _RESCTRL_H */
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 33/38 x86/resctrl: Move is_mba_sc() out of core.c
In reply to James Morseis_mba_sc() is defined in core.c, but has no callers there. It does not access any architecture private structures.
Move this to rdtgroup.c where the majority of callers are. This makes the move of the filesystem code to /fs/ cleaner.
---
Changes since v2:
* This patch is new.
--- arch/x86/kernel/cpu/resctrl/core.c | 15 --------------- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 15 +++++++++++++++ 2 files changed, 15 insertions(+), 15 deletions(-)
arch/x86/kernel/cpu/resctrl/core.c +0 -15
@@ -160,21 +160,6 @@ static inline void cache_alloc_hsw_probe(void) rdt_alloc_capable = true; } -bool is_mba_sc (struct rdt_resource *r) -{ -if (!r) - r =resctrl_arch_get_resource (RDT_RESOURCE_MBA ); - -/* -* The software controller support is only applicable to MBA resource. -* Make sure to check for resource type. -*/ -if (r->rid !=RDT_RESOURCE_MBA ) -return false; - -return r->membw.mba_sc; -} -/* * rdt_get_mb_table() - get a mapping of bandwidth(b/w) percentage values * exposed to user interface and the h/w understandable delay values.
arch/x86/kernel/cpu/resctrl/rdtgroup.c +15 -0
@@ -1500,6 +1500,21 @@ unsigned int rdtgroup_cbm_to_size(struct rdt_resource *r,return size; } +bool is_mba_sc (struct rdt_resource *r) +{ +if (!r) + r =resctrl_arch_get_resource (RDT_RESOURCE_MBA ); + +/* +* The software controller support is only applicable to MBA resource. +* Make sure to check for resource type. +*/ +if (r->rid !=RDT_RESOURCE_MBA ) +return false; + +return r->membw.mba_sc; +} +/* * rdtgroup_size_show - Display size in bytes of allocated regions *
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 34/38 x86/resctrl: Add end-marker to the resctrl_event_id enum
In reply to James MorseThe resctrl_event_id enum gives names to the counter event numbers on x86. These are used directly by resctrl.
To allow the MPAM driver to keep an array of these the size of the enum needs to be known.
Add a 'num_events' define which can be used to size an array. This isn't a member of the enum to avoid updating switch statements that would otherwise be missing a case.
---
Changes since v2:
* This patch is new.
--- include/linux/resctrl_types.h | 2 ++ 1 file changed, 2 insertions(+)
include/linux/resctrl_types.h +2 -0
@@ -51,4 +51,6 @@ enum resctrl_event_id { QOS_L3_MBM_LOCAL_EVENT_ID = 0x03, }; +#define QOS_NUM_EVENTS (QOS_L3_MBM_LOCAL_EVENT_ID + 1) +#endif /* __LINUX_RESCTRL_TYPES_H */
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 35/38 x86/resctrl: Remove a newline to avoid confusing the code move script
In reply to James MorseThe resctrl filesystem code will shortly be moved to /fs/. This involves splitting all the existing files, with some functions remaining under arch/x86, and others moving to fs/resctrl.
To make this reproducible, a python script does the heavy lif^W copy-and-paste. This involves some clunky parsing of C code.
The parser gets confused by the newline after this #ifdef. Just remove it.
---
This patch, and the post-move cleanup could all be merged together. It's included like this to make the move script easier to work with.
Changes since v2:
* This patch is new.
--- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 1 - 1 file changed, 1 deletion(-)
arch/x86/kernel/cpu/resctrl/rdtgroup.c +0 -1
@@ -880,7 +880,6 @@ static int rdtgroup_rmid_show(struct kernfs_open_file *of, }#ifdef CONFIG_PROC_CPU_RESCTRL -/* * A task can only be part of one resctrl control group and of one monitor * group which is associated to that control group.
Signature · 1 line
2.39.2
James Morse
reviewerPATCH v3 36/38 fs/resctrl: Add boiler plate for external resctrl code
In reply to James MorseAdd Makefile and Kconfig for fs/resctrl. Add ARCH_HAS_CPU_RESCTRL for the common parts of the resctrl interface and make X86_CPU_RESCTRL depend on this.
Adding an include of asm/resctrl.h to linux/resctrl.h allows some of the files to switch over to using this header instead.
---
Changes since v2:
* Dropped KERNFS dependency from arch side Kconfig.
* Added empty trace.h file.
* Merged asm->linux includes from Dave's patch to decouple those patches from this series.
Changes since v1:
* Rename new file psuedo_lock.c to pseudo_lock.c, to match the name of the original file (and to be less surprising).
* [Whitespace only] Under RESCTRL_FS in fs/resctrl/Kconfig, delete alignment space in orphaned select ... if (which has nothing to line up with any more).
* [Whitespace only] Reflow and re-tab Kconfig additions.
--- MAINTAINERS | 1 + arch/Kconfig | 8 +++++ arch/x86/Kconfig | 5 ++-- arch/x86/kernel/cpu/resctrl/internal.h | 3 +- arch/x86/kernel/cpu/resctrl/monitor.c | 2 +- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 2 +- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 2 +- fs/Kconfig | 1 + fs/Makefile | 1 + fs/resctrl/Kconfig | 36 +++++++++++++++++++++++ fs/resctrl/Makefile | 3 ++ fs/resctrl/ctrlmondata.c | 0 fs/resctrl/internal.h | 0 fs/resctrl/monitor.c | 0 fs/resctrl/pseudo_lock.c | 0 fs/resctrl/rdtgroup.c | 0 fs/resctrl/trace.h | 0 include/linux/resctrl.h | 4 +++ 18 files changed, 61 insertions(+), 7 deletions(-) create mode 100644 fs/resctrl/Kconfig create mode 100644 fs/resctrl/Makefile create mode 100644 fs/resctrl/ctrlmondata.c create mode 100644 fs/resctrl/internal.h create mode 100644 fs/resctrl/monitor.c create mode 100644 fs/resctrl/pseudo_lock.c create mode 100644 fs/resctrl/rdtgroup.c create mode 100644 fs/resctrl/trace.h
MAINTAINERS +1 -0
@@ -18859,6 +18859,7 @@ S: Supported F: Documentation/arch/x86/resctrl* F: arch/x86/include/asm/resctrl.h F: arch/x86/kernel/cpu/resctrl/ +F: fs/resctrl/ F: include/linux/resctrl*.h F: tools/testing/selftests/resctrl/
arch/Kconfig +8 -0
@@ -1431,6 +1431,14 @@ config STRICT_MODULE_RWX config ARCH_HAS_PHYS_TO_DMA bool +config ARCH_HAS_CPU_RESCTRL + bool + help + An architecture selects this option to indicate that the necessary + hooks are provided to support the common memory system usage + monitoring and control interfaces provided by the 'resctrl' + filesystem (see RESCTRL_FS). + config HAVE_ARCH_COMPILER_H bool help
arch/x86/Kconfig +3 -2
@@ -490,8 +490,9 @@ config X86_MPPARSE config X86_CPU_RESCTRL bool "x86 CPU resource control support" depends on X86 && (CPU_SUP_INTEL || CPU_SUP_AMD) - select KERNFS - select PROC_CPU_RESCTRL if PROC_FS + depends on MISC_FILESYSTEMS + select ARCH_HAS_CPU_RESCTRL + select RESCTRL_FS select RESCTRL_FS_PSEUDO_LOCK help Enable x86 CPU resource control support.
arch/x86/kernel/cpu/resctrl/internal.h +1 -2
@@ -7,10 +7,9 @@#include <linux/kernfs.h> #include <linux/fs_context.h> #include <linux/jump_label.h> +#include <linux/resctrl.h> #include <linux/tick.h> -#include <asm/resctrl.h> -#define L3_QOS_CDP_ENABLE 0x01ULL#define L2_QOS_CDP_ENABLE 0x01ULL
arch/x86/kernel/cpu/resctrl/monitor.c +1 -1
@@ -17,11 +17,11 @@#include <linux/cpu.h> #include <linux/module.h> +#include <linux/resctrl.h> #include <linux/sizes.h> #include <linux/slab.h> #include <asm/cpu_device_id.h> -#include <asm/resctrl.h> #include "internal.h" #include "trace.h"
arch/x86/kernel/cpu/resctrl/pseudo_lock.c +1 -1
@@ -19,12 +19,12 @@#include <linux/mman.h> #include <linux/perf_event.h> #include <linux/pm_qos.h> +#include <linux/resctrl.h> #include <linux/slab.h> #include <linux/uaccess.h> #include <asm/cacheflush.h> #include <asm/cpu_device_id.h> -#include <asm/resctrl.h> #include <asm/perf_event.h> #include "../../events/perf_event.h" /* For X86_CONFIG() */
arch/x86/kernel/cpu/resctrl/rdtgroup.c +1 -1
@@ -19,6 +19,7 @@#include <linux/fs_parser.h> #include <linux/sysfs.h> #include <linux/kernfs.h> +#include <linux/resctrl.h> #include <linux/seq_buf.h> #include <linux/seq_file.h> #include <linux/sched/signal.h>
@@ -29,7 +30,6 @@#include <uapi/linux/magic.h> -#include <asm/resctrl.h> #include "internal.h" DEFINE_STATIC_KEY_FALSE (rdt_enable_key);
fs/Kconfig +1 -0
@@ -331,6 +331,7 @@ source "fs/omfs/Kconfig" source "fs/hpfs/Kconfig" source "fs/qnx4/Kconfig" source "fs/qnx6/Kconfig" +source "fs/resctrl/Kconfig" source "fs/romfs/Kconfig" source "fs/pstore/Kconfig" source "fs/sysv/Kconfig"
fs/Makefile +1 -0
@@ -129,3 +129,4 @@ obj-$(CONFIG_EFIVAR_FS) += efivarfs/ obj-$(CONFIG_EROFS_FS) += erofs/ obj-$(CONFIG_VBOXSF_FS) += vboxsf/ obj-$(CONFIG_ZONEFS_FS) += zonefs/ +obj-$(CONFIG_RESCTRL_FS) += resctrl/
fs/resctrl/Kconfig new file +36 -0
@@ -0,0 +1,36 @@ +config RESCTRL_FS + bool "CPU Resource Control Filesystem (resctrl)" + depends on ARCH_HAS_CPU_RESCTRL + select KERNFS + select PROC_CPU_RESCTRL if PROC_FS + help + Some architectures provide hardware facilities to group tasks and + monitor and control their usage of memory system resources such as + caches and memory bandwidth. Examples of such facilities include + Intel's Resource Director Technology (Intel(R) RDT) and AMD's + Platform Quality of Service (AMD QoS). + + If your system has the necessary support and you want to be able to + assign tasks to groups and manipulate the associated resource + monitors and controls from userspace, say Y here to get a mountable + 'resctrl' filesystem that lets you do just that. + + If nothing mounts or prods the 'resctrl' filesystem, resource + controls and monitors are left in a quiescent, permissive state. + + If unsure, it is safe to say N. + + See <file:Documentation/arch/x86/resctrl.rst> for more information. + +config RESCTRL_FS_PSEUDO_LOCK + bool + help + Software mechanism to pin data in a cache portion using + micro-architecture specific knowledge. + +config RESCTRL_RMID_DEPENDS_ON_CLOSID + bool + help + Enable by the architecture when the RMID values depend on the CLOSID. + This causes the closid allocator to search for CLOSID with clean + RMID.
fs/resctrl/Makefile new file +3 -0
@@ -0,0 +1,3 @@ +# SPDX-License-Identifier: GPL-2.0 +obj-$(CONFIG_RESCTRL_FS) += rdtgroup.o ctrlmondata.o monitor.o +obj-$(CONFIG_RESCTRL_FS_PSEUDO_LOCK) += pseudo_lock.o
fs/resctrl/ctrlmondata.c new file +0 -0
fs/resctrl/internal.h new file +0 -0
fs/resctrl/monitor.c new file +0 -0
fs/resctrl/pseudo_lock.c new file +0 -0
fs/resctrl/rdtgroup.c new file +0 -0
fs/resctrl/trace.h new file +0 -0
include/linux/resctrl.h +4 -0
@@ -8,6 +8,10 @@#include <linux/pid.h> #include <linux/resctrl_types.h> +#ifdef CONFIG_ARCH_HAS_CPU_RESCTRL +#include <asm/resctrl.h> +#endif +/* CLOSID, RMID value used by the default control group */ #define RESCTRL_RESERVED_CLOSID 0#define RESCTRL_RESERVED_RMID 0
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 36/38] fs/resctrl: Add boiler plate for external resctrl code
Hi James,
Quoted 214 lines · James Morse
Add
MakefileandKconfigfor fs/resctrl. AddARCH_HAS_CPU_RESCTRLfor the common parts of the resctrl interface and makeX86_CPU_RESCTRLdepend on this.Adding an include of
asm/resctrl.htolinux/resctrl.hallows some of the files to switch over to using this header instead.Co-developed-by Dave Martin <[email protected]>Signed-off-by Dave Martin <[email protected]>Signed-off-by James Morse <[email protected]>Tested-by Peter Newman <[email protected]>Tested-by Shaopeng Tan <[email protected]>---
Changes since v2:
* Dropped KERNFS dependency from arch sideKconfig.
* Added emptytrace.hfile.
* Mergedasm->linuxincludes from Dave's patch to decouple those patches from this series.Changes since v1:
* Rename new filepsuedo_lock.ctopseudo_lock.c, to match the name of the original file (and to be less surprising).* [Whitespace only] Under
RESCTRL_FSinfs/resctrl/Kconfig, delete alignment space in orphaned select ... if (which has nothing to line up with any more).* [Whitespace only] Reflow and re-tab
Kconfigadditions.--- MAINTAINERS | 1 + arch/Kconfig | 8 +++++ arch/x86/Kconfig | 5 ++-- arch/x86/kernel/cpu/resctrl/internal.h | 3 +- arch/x86/kernel/cpu/resctrl/monitor.c | 2 +- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 2 +- arch/x86/kernel/cpu/resctrl/rdtgroup.c | 2 +- fs/Kconfig | 1 + fs/Makefile | 1 + fs/resctrl/Kconfig | 36 +++++++++++++++++++++++ fs/resctrl/Makefile | 3 ++ fs/resctrl/ctrlmondata.c | 0 fs/resctrl/internal.h | 0 fs/resctrl/monitor.c | 0 fs/resctrl/pseudo_lock.c | 0 fs/resctrl/rdtgroup.c | 0 fs/resctrl/trace.h | 0 include/linux/resctrl.h | 4 +++ 18 files changed, 61 insertions(+), 7 deletions(-)create mode 100644
fs/resctrl/Kconfig
create mode 100644fs/resctrl/Makefile
create mode 100644fs/resctrl/ctrlmondata.c
create mode 100644fs/resctrl/internal.h
create mode 100644fs/resctrl/monitor.c
create mode 100644fs/resctrl/pseudo_lock.c
create mode 100644fs/resctrl/rdtgroup.c
create mode 100644fs/resctrl/trace.hMAINTAINERS +1
-0@@ -18859,6 +18859,7 @@ S: Supported F: Documentation/arch/x86/resctrl* F: arch/x86/include/asm/resctrl.h F: arch/x86/kernel/cpu/resctrl/ +F: fs/resctrl/ F: include/linux/resctrl*.h F: tools/testing/selftests/resctrl/arch/Kconfig +8
-0@@ -1431,6 +1431,14 @@ config STRICT_MODULE_RWX config ARCH_HAS_PHYS_TO_DMA bool +config ARCH_HAS_CPU_RESCTRL + bool + help + An architecture selects this option to indicate that the necessary + hooks are provided to support the common memory system usage + monitoring and control interfaces provided by the 'resctrl' + filesystem (see RESCTRL_FS). + config HAVE_ARCH_COMPILER_H bool helparch/x86/Kconfig +3
-2@@ -490,8 +490,9 @@ config X86_MPPARSE config X86_CPU_RESCTRL bool "x86 CPU resource control support" depends on X86 && (CPU_SUP_INTEL || CPU_SUP_AMD) - select KERNFS - select PROC_CPU_RESCTRL if PROC_FS + depends on MISC_FILESYSTEMS + select ARCH_HAS_CPU_RESCTRL + select RESCTRL_FS select RESCTRL_FS_PSEUDO_LOCK help Enable x86 CPU resource control support.arch/x86/kernel/cpu/resctrl/internal.h +1
-2@@ -7,10 +7,9 @@#include <linux/kernfs.h> #include <linux/fs_context.h> #include <linux/jump_label.h> +#include <linux/resctrl.h> #include <linux/tick.h> -#include <asm/resctrl.h> -#define L3_QOS_CDP_ENABLE 0x01ULL#define L2_QOS_CDP_ENABLE 0x01ULLarch/x86/kernel/cpu/resctrl/monitor.c +1
-1@@ -17,11 +17,11 @@#include <linux/cpu.h> #include <linux/module.h> +#include <linux/resctrl.h> #include <linux/sizes.h> #include <linux/slab.h> #include <asm/cpu_device_id.h> -#include <asm/resctrl.h> #include "internal.h" #include "trace.h" arch/x86/kernel/cpu/resctrl/pseudo_lock.c +1
-1@@ -19,12 +19,12 @@#include <linux/mman.h> #include <linux/perf_event.h> #include <linux/pm_qos.h> +#include <linux/resctrl.h> #include <linux/slab.h> #include <linux/uaccess.h> #include <asm/cacheflush.h> #include <asm/cpu_device_id.h> -#include <asm/resctrl.h> #include <asm/perf_event.h> #include "../../events/perf_event.h" /* For X86_CONFIG() */ arch/x86/kernel/cpu/resctrl/rdtgroup.c +1
-1@@ -19,6 +19,7 @@#include <linux/fs_parser.h> #include <linux/sysfs.h> #include <linux/kernfs.h> +#include <linux/resctrl.h> #include <linux/seq_buf.h> #include <linux/seq_file.h> #include <linux/sched/signal.h> @@ -29,7 +30,6 @@#include <uapi/linux/magic.h> -#include <asm/resctrl.h> #include "internal.h" DEFINE_STATIC_KEY_FALSE (rdt_enable_key);fs/Kconfig +1
-0@@ -331,6 +331,7 @@ source "fs/omfs/Kconfig" source "fs/hpfs/Kconfig" source "fs/qnx4/Kconfig" source "fs/qnx6/Kconfig" +source "fs/resctrl/Kconfig" source "fs/romfs/Kconfig" source "fs/pstore/Kconfig" source "fs/sysv/Kconfig"fs/Makefile +1
-0@@ -129,3 +129,4 @@ obj-$(CONFIG_EFIVAR_FS) += efivarfs/ obj-$(CONFIG_EROFS_FS) += erofs/ obj-$(CONFIG_VBOXSF_FS) += vboxsf/ obj-$(CONFIG_ZONEFS_FS) += zonefs/ +obj-$(CONFIG_RESCTRL_FS) += resctrl/fs/resctrl/Kconfig new file +12
-0@@ -0,0 +1,36 @@ +config RESCTRL_FS + bool "CPU Resource Control Filesystem (resctrl)" + depends on ARCH_HAS_CPU_RESCTRL + select KERNFS + select PROC_CPU_RESCTRL if PROC_FS + help + Some architectures provide hardware facilities to group tasks and + monitor and control their usage of memory system resources such as + caches and memory bandwidth. Examples of such facilities include + Intel's Resource Director Technology (Intel(R) RDT) and AMD's + Platform Quality of Service (AMD QoS). +
fs/resctrl/Kconfig line 22
Quoted 10 lines
fs/resctrl/Kconfig new file +10
-0+ If your system has the necessary support and you want to be able to + assign tasks to groups and manipulate the associated resource + monitors and controls from userspace, say Y here to get a mountable + 'resctrl' filesystem that lets you do just that. + + If nothing mounts or prods the 'resctrl' filesystem, resource + controls and monitors are left in a quiescent, permissive state. + + If unsure, it is safe to say N. +
Will user ever get opportunity to say "Y" or "N"? It looks to me that RESCTRL_FS will be "forced" on user as it is selected by the arch specific config X86_CPU_RESCTRL and be invisble otherwise because of the dependency on ARCH_HAS_CPU_RESCTRL. The text about when to select "Y" or "N" thus does not look practical to me and it may be helpful to instead provide information about when it is selected? I do not know the customs for this text and if it is intended to document any future usages also.
Reinette
James Morse
reviewerRe: [PATCH v3 36/38] fs/resctrl: Add boiler plate for external resctrl code
Hi Reinette,
Quoted 35 lines · Reinette Chatre
Quoted 34 lines · James Morse
Add
MakefileandKconfigfor fs/resctrl. AddARCH_HAS_CPU_RESCTRLfor the common parts of the resctrl interface and makeX86_CPU_RESCTRLdepend on this.Adding an include of
asm/resctrl.htolinux/resctrl.hallows some of the files to switch over to using this header instead.fs/resctrl/Kconfig new file +20
-0@@ -0,0 +1,36 @@ +config RESCTRL_FS + bool "CPU Resource Control Filesystem (resctrl)" + depends on ARCH_HAS_CPU_RESCTRL + select KERNFS + select PROC_CPU_RESCTRL if PROC_FS + help + Some architectures provide hardware facilities to group tasks and + monitor and control their usage of memory system resources such as + caches and memory bandwidth. Examples of such facilities include + Intel's Resource Director Technology (Intel(R) RDT) and AMD's + Platform Quality of Service (AMD QoS). + + If your system has the necessary support and you want to be able to + assign tasks to groups and manipulate the associated resource + monitors and controls from userspace, say Y here to get a mountable + 'resctrl' filesystem that lets you do just that. + + If nothing mounts or prods the 'resctrl' filesystem, resource + controls and monitors are left in a quiescent, permissive state. +
Quoted 8 lines
fs/resctrl/Kconfig line 22
Quoted 2 lines
fs/resctrl/Kconfig new file +2
-0+ If unsure, it is safe to say N. +Will user ever get opportunity to say "Y" or "N"?
It looks to me thatRESCTRL_FSwill be "forced" on user as it is selected by the arch specific configX86_CPU_RESCTRLand be invisble otherwise because of the dependency onARCH_HAS_CPU_RESCTRL.
I did it like this so that this change is invisible for x86 config files on the principle of 'least noise'. Users can't enable RDT but disable resctrl today.
It isn't actually possible to enable RDT and disable resctrl until after the code has been split from the architecture code.
I have ended up supporting this for MPAM - you can enable the architecture's MPAM code and the driver, but not resctrl. This will eventually be for in-kernel users of resources that resctrl doesn't understand.
Quoted 4 lines
The text about when to select "Y" or "N" thus does
not look practical to me and it may be helpful to instead provide information about when it is selected? I do not know the customs for this text and if it is intended to document any future usages also.
I think Dave wrote this text because its traditional for Kconfig options to say this.
Describing when it is selected gets messy as this varies by architecture, and Kconfig can already tell you this:
| Selected by [y]:
│ - X86_CPU_RESCTRL [=y] && X86 [=y] && (CPU_SUP_INTEL [=y] || CPU_SUP_AMD [=y]) &&MISC_FILESYSTEMS [=y]
I don't think it makes sense for resctrl to be enabled/disabled independently on x86. If you want to support this, we need a few more IS_ENABLED() checks and stubs to make it build. The only reason I can see to do it is to ensure the architecture code is self contained.
I'll reword this as "On architectures where this can be disabled independently, it is safe to say N".
Thanks,
James
Reinette Chatre
maintainerRe: [PATCH v3 36/38] fs/resctrl: Add boiler plate for external resctrl code
Hi James,
Quoted 62 lines · James Morse
Hi Reinette,
Quoted 35 lines · Reinette Chatre
Quoted 34 lines · James Morse
Add
MakefileandKconfigfor fs/resctrl. AddARCH_HAS_CPU_RESCTRLfor the common parts of the resctrl interface and makeX86_CPU_RESCTRLdepend on this.Adding an include of
asm/resctrl.htolinux/resctrl.hallows some of the files to switch over to using this header instead.fs/resctrl/Kconfig new file +20
-0@@ -0,0 +1,36 @@ +config RESCTRL_FS + bool "CPU Resource Control Filesystem (resctrl)" + depends on ARCH_HAS_CPU_RESCTRL + select KERNFS + select PROC_CPU_RESCTRL if PROC_FS + help + Some architectures provide hardware facilities to group tasks and + monitor and control their usage of memory system resources such as + caches and memory bandwidth. Examples of such facilities include + Intel's Resource Director Technology (Intel(R) RDT) and AMD's + Platform Quality of Service (AMD QoS). + + If your system has the necessary support and you want to be able to + assign tasks to groups and manipulate the associated resource + monitors and controls from userspace, say Y here to get a mountable + 'resctrl' filesystem that lets you do just that. + + If nothing mounts or prods the 'resctrl' filesystem, resource + controls and monitors are left in a quiescent, permissive state. +Quoted 8 lines
fs/resctrl/Kconfig line 22
Quoted 2 lines
fs/resctrl/Kconfig new file +2
-0+ If unsure, it is safe to say N. +Will user ever get opportunity to say "Y" or "N"?
It looks to me thatRESCTRL_FSwill be "forced" on user as it is selected by the arch specific configX86_CPU_RESCTRLand be invisble otherwise because of the dependency onARCH_HAS_CPU_RESCTRL.I did it like this so that this change is invisible for x86 config files on the principle of 'least noise'. Users can't enable RDT but disable resctrl today.
It isn't actually possible to enable RDT and disable resctrl until after the code has been split from the architecture code.I have ended up supporting this for MPAM - you can enable the architecture's MPAM code and the driver, but not resctrl. This will eventually be for in-kernel users of resources that resctrl doesn't understand.
Quoted 4 lines
The text about when to select "Y" or "N" thus does
not look practical to me and it may be helpful to instead provide information about when it is selected? I do not know the customs for this text and if it is intended to document any future usages also.
Quoted 7 lines
I think Dave wrote this text because its traditional for
Kconfigoptions to say this.Describing when it is selected gets messy as this varies by architecture, and
Kconfigcan already tell you this:
| Selected by [y]:
│ -X86_CPU_RESCTRL[=y] && X86 [=y] && (CPU_SUP_INTEL[=y] ||CPU_SUP_AMD[=y]) &&MISC_FILESYSTEMS[=y]
Right.
Quoted 2 lines
I don't think it makes sense for resctrl to be enabled/disabled independently on x86.
I was not asking for resctrl to be enabled/disabled independently on x86. I commented on this patch that adds text to guide user for options that the user is never able to select.
Quoted 1 line
If you want to support this, we need a few more
IS_ENABLED()checks and stubs to make it
I did not intend to suggest this at all.
Quoted 5 lines
build. The only reason I can see to do it is to ensure the architecture code is self contained.
I'll reword this as "On architectures where this can be disabled independently, it is safe to say N".
ok
Reinette
James Morse
reviewerPATCH v3 37/38 x86/resctrl: Move the filesystem bits to headers visible to fs/resctrl
In reply to James MorseOnce the filesystem parts of resctrl move to fs/resctrl, it cannot rely on definitions in x86's internal.h.
Move definitions in internal.h that need to be shared between the filesystem and architecture code to header files that fs/resctrl can include.
Doing this separately means the filesystem code only moves between files of the same name, instead of having these changes mixed in too.
----
Changes since v2:
* Dropped the rfflags and some other defines from being moved.
Changes since v1:
* Revert apparently unintentional duplication of a couple of variable declarations in <linux/resctrl.h>.
No functional change.
--- arch/x86/include/asm/resctrl.h | 3 +++ arch/x86/kernel/cpu/resctrl/core.c | 5 +++++ arch/x86/kernel/cpu/resctrl/internal.h | 9 --------- include/linux/resctrl_types.h | 3 +++ 4 files changed, 11 insertions(+), 9 deletions(-)
arch/x86/include/asm/resctrl.h +3 -0
@@ -218,6 +218,9 @@ int resctrl_arch_measure_l2_residency(void *_plr);int resctrl_arch_measure_l3_residency (void *_plr);void resctrl_cpu_detect (struct cpuinfo_x86 *c); +bool resctrl_arch_get_cdp_enabled (enum resctrl_res_level l); +int resctrl_arch_set_cdp_enabled (enum resctrl_res_level l,bool enable); +#else static inline void resctrl_arch_sched_in (struct task_struct *tsk) {}
arch/x86/kernel/cpu/resctrl/core.c +5 -0
@@ -283,6 +283,11 @@ static void rdt_get_cdp_l2_config(void)rdt_get_cdp_config (RDT_RESOURCE_L2 ); } +bool resctrl_arch_get_cdp_enabled (enum resctrl_res_level l) +{ +return rdt_resources_all[l].cdp_enabled; +} +static void mba_wrmsr_amd (struct msr_param *m) {struct rdt_hw_resource *hw_res =resctrl_to_arch_res (m->res);
arch/x86/kernel/cpu/resctrl/internal.h +0 -9
@@ -17,8 +17,6 @@#define CQM_LIMBOCHECK_INTERVAL 1000#define MBM_CNTR_WIDTH_BASE 24 -#define MBM_OVERFLOW_INTERVAL 1000 -#define MAX_MBA_BW 100u#define MBA_IS_LINEAR 0x4#define MBM_CNTR_WIDTH_OFFSET_AMD 20
@@ -371,13 +369,6 @@ static inline struct rdt_resource *resctrl_inc(struct rdt_resource *res)return &hw_res->r_resctrl; } -static inline bool resctrl_arch_get_cdp_enabled (enum resctrl_res_level l) -{ -return rdt_resources_all[l].cdp_enabled; -} - -int resctrl_arch_set_cdp_enabled (enum resctrl_res_level l,bool enable); -/* * To return the common struct rdt_resource, which is contained in struct * rdt_hw_resource, walk the resctrl member of struct rdt_hw_resource.
include/linux/resctrl_types.h +3 -0
@@ -7,6 +7,9 @@#ifndef __LINUX_RESCTRL_TYPES_H#define __LINUX_RESCTRL_TYPES_H +#define MAX_MBA_BW 100u +#define MBM_OVERFLOW_INTERVAL 1000 +/* Reads to Local DRAM Memory */ #define READS_TO_LOCAL_MEM BIT(0)
Signature · 1 line
2.39.2
Reinette Chatre
maintainerRe: [PATCH v3 37/38] x86/resctrl: Move the filesystem bits to headers visible to fs/resctrl
Hi James,
Quoted 10 lines · James Morse
Once the filesystem parts of resctrl move to fs/resctrl, it cannot rely on definitions in x86's
internal.h.Move definitions in
internal.hthat need to be shared between the filesystem and architecture code to header files that fs/resctrl can include.Doing this separately means the filesystem code only moves between files of the same name, instead of having these changes mixed in too.
Quoted 6 lines
Co-developed-by Dave Martin <[email protected]>Signed-off-by Dave Martin <[email protected]>Signed-off-by James Morse <[email protected]>Tested-by Peter Newman <[email protected]>Tested-by Shaopeng Tan <[email protected]>----
"----" -> "---" to prevent following text from being included in changelog.
Reinette
James Morse
reviewerPATCH v3 38/38 x86/resctrl: Add python script to move resctrl code to /fs/resctrl
In reply to James MorseTo support more than one architecture resctrl needs to move from arch/x86 to live under fs. Moving all the code breaks any series on the mailing list, so needs scheduling carefully.
Maintaining the patch that moves all this code has proved labour intensive. It's also near-impossible to review that no inadvertent changes have crept in.
To solve these problems, temporarily add a hacky python program that lists all the functions that should move, and those that should stay.
No attempt to parse C code is made, this thing tries to name 'blocks' based on hueristics about the kernel coding style. It's fragile, but good enough for its single use here.
This causes the original files to be regenerated, which will add newlines that are not present in the original file.
I don't suggested this gets merged.
---
Changes since v2:
* This patch is new.
--- resctrl_copy_pasta.py | 766 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 766 insertions(+) create mode 100644 resctrl_copy_pasta.py
resctrl_copy_pasta.py new file +766 -0
@@ -0,0 +1,766 @@ +#!/usr/bin/python +import sys; +import os; +import re; + +############ + +SRC_DIR = "arch/x86/kernel/cpu/resctrl"; +DST_DIR = "fs/resctrl"; + +resctrl_files = [ + "ctrlmondata.c", + "internal.h", + "monitor.c", + "pseudo_lock.c", + "rdtgroup.c", + "trace.h", +]; + +functions_to_keep = [ + # common + "pr_fmt", + + # core.c + "domain_list_lock", + "resctrl_arch_late_init", + "resctrl_arch_exit", + "resctrl_cpu_detect", + "rdt_cpu_has", + "resctrl_arch_is_evt_configurable", + "get_mem_config", + "get_slow_mem_config", + "get_rdt_alloc_resources", + "get_rdt_mon_resources", + "__check_quirks_intel", + "check_quirks", + "get_rdt_resources", + "rdt_init_res_defs_intel", + "rdt_init_res_defs_amd", + "rdt_init_res_defs", + "resctrl_cpu_detect", + "resctrl_arch_late_init", + "resctrl_arch_exit", + "setup_default_ctrlval", + "domain_free", + "domain_setup_ctrlval", + "arch_domain_mbm_alloc", + "domain_add_cpu", + "domain_remove_cpu", + "clear_closid_rmid", + "resctrl_arch_online_cpu", + "resctrl_arch_offline_cpu", + "resctrl_arch_find_domain", + "resctrl_arch_get_num_closid", + "rdt_ctrl_update", + "domain_init", + "resctrl_arch_get_resource", + "cache_alloc_hsw_probe", + "rdt_get_mb_table", + "__get_mem_config_intel", + "__rdt_get_mem_config_amd", + "rdt_get_cache_alloc_cfg", + "rdt_get_cdp_config", + "rdt_get_cdp_l3_config", + "rdt_get_cdp_l2_config", + "resctrl_arch_get_cdp_enabled", + "set_rdt_options", + "pqr_state", + "rdt_resources_all", + "delay_bw_map", + "rdt_options", + "cat_wrmsr", + "mba_wrmsr_amd", + "mba_wrmsr_intel", + "anonymous-enum", + "rdt_find_domain", + "rdt_alloc_capable", + "rdt_online", + "RDT_OPT", + + # ctrlmon.c + "apply_config", + "resctrl_arch_update_one", + "resctrl_arch_update_domains", + "resctrl_arch_get_config", + + # internal.h + "L3_QOS_CDP_ENABLE", + "L2_QOS_CDP_ENABLE", + "MBM_CNTR_WIDTH_OFFSET_AMD", + "arch_mbm_state", + "rdt_hw_domain", + "resctrl_to_arch_dom", + "msr_param", + "rdt_hw_resource", + "resctrl_to_arch_res", + "rdt_resources_all", + "resctrl_inc", + "for_each_rdt_resource", + "for_each_capable_rdt_resource", + "for_each_alloc_capable_rdt_resource", + "for_each_mon_capable_rdt_resource", + "cpuid_0x10_1_eax", + "cpuid_0x10_3_eax", + "cpuid_0x10_x_ecx", + "cpuid_0x10_x_edx", + "rdt_ctrl_update", + "rdt_get_mon_l3_config", + "rdt_cpu_has", + "intel_rdt_mbm_apply_quirk", + "rdt_domain_reconfigure_cdp", + + # monitor.c + "rdt_mon_capable", + "rdt_mon_features", + "CF", + "mbm_cf_table", + "mbm_cf_rmidthreshold", + "mbm_cf", + "get_corrected_mbm_count", + "__rmid_read", + "get_arch_mbm_state", + "resctrl_arch_reset_rmid", + "resctrl_arch_reset_rmid_all", + "mbm_overflow_count", + "resctrl_arch_rmid_read", + "rdt_get_mon_l3_config", + "intel_rdt_mbm_apply_quirk", + + # pseudo_lock.c + "prefetch_disable_bits", + "resctrl_arch_get_prefetch_disable_bits", + "resctrl_arch_pseudo_lock_fn", + "resctrl_arch_measure_cycles_lat_fn", + "perf_miss_attr", + "perf_hit_attr", + "residency_counts", + "measure_residency_fn", + "resctrl_arch_measure_l2_residency", + "resctrl_arch_measure_l3_residency", + + # rdtgroup.c + "rdt_enable_key", + "rdt_mon_enable_key", + "rdt_alloc_enable_key", + "resctrl_arch_sync_cpu_closid_rmid", + "INVALID_CONFIG_INDEX", + "mon_event_config_index_get", + "resctrl_arch_mon_event_config_read", + "resctrl_arch_mon_event_config_write", + "l3_qos_cfg_update", + "l2_qos_cfg_update", + "set_cache_qos_cfg", + "rdt_domain_reconfigure_cdp", + "cdp_enable", + "cdp_disable", + "resctrl_arch_set_cdp_enabled", + "reset_all_ctrls", + "resctrl_arch_reset_resources", + + # trace.h + "TRACE_SYSTEM", + "pseudo_lock_mem_latency", + "pseudo_lock_l2", + "pseudo_lock_l3", +]; + +functions_to_move = [ + # common + "pr_fmt", + + # ctrlmon.c + "rdt_parse_data", + "(ctrlval_parser_t)", + "bw_validate", + "parse_bw", + "cbm_validate", + "parse_cbm", + "get_parser", + "parse_line", + "rdtgroup_parse_resource", + "rdtgroup_schemata_write", + "show_doms", + "rdtgroup_schemata_show", + "smp_mon_event_count", + "mon_event_read", + "rdtgroup_mondata_show", + + # internal.h + "MBM_OVERFLOW_INTERVAL", + "CQM_LIMBOCHECK_INTERVAL", + "cpumask_any_housekeeping", + "rdt_fs_context", + "rdt_fc2context", + "mon_evt", + "mon_data_bits", + "rmid_read", + "resctrl_schema_all", + "resctrl_mounted", + "rdt_group_type", + "rdtgrp_mode", + "mongroup", + "rdtgroup", + "RFTYPE_FLAGS_CPUS_LIST", + "rdt_all_groups", + "rftype", + "mbm_state", + "is_mba_sc", + + # monitor.c + "rmid_entry", + "rmid_free_lru", + "closid_num_dirty_rmid", + "rmid_limbo_count", + "rmid_ptrs", + "resctrl_rmid_realloc_threshold", + "resctrl_rmid_realloc_limit", + "__rmid_entry", + "limbo_release_entry", + "__check_limbo", + "has_busy_rmid", + "resctrl_find_free_rmid", + "resctrl_find_cleanest_closid", + "alloc_rmid", + "add_rmid_to_limbo", + "free_rmid", + "get_mbm_state", + "__mon_event_count", + "mbm_bw_count", + "mon_event_count", + "update_mba_bw", + "mbm_update", + "cqm_handle_limbo", + "cqm_setup_limbo_handler", + "mbm_handle_overflow", + "mbm_setup_overflow_handler", + "dom_data_init", + "dom_data_exit", + "llc_occupancy_event", + "mbm_total_event", + "mbm_local_event", + "l3_mon_evt_init", + "resctrl_mon_resource_init", + "resctrl_mon_resource_exit", + + # pseudo_lock.c + "pseudo_lock_major", + "pseudo_lock_minor_avail", + "pseudo_lock_devnode", + "pseudo_lock_class", + "pseudo_lock_minor_get", + "pseudo_lock_minor_release", + "region_find_by_minor", + "pseudo_lock_pm_req", + "pseudo_lock_cstates_relax", + "pseudo_lock_cstates_constrain", + "pseudo_lock_region_clear", + "pseudo_lock_region_init", + "pseudo_lock_init", + "pseudo_lock_region_alloc", + "pseudo_lock_free", + "rdtgroup_monitor_in_progress", + "rdtgroup_locksetup_user_restrict", + "rdtgroup_locksetup_user_restore", + "rdtgroup_locksetup_enter", + "rdtgroup_locksetup_exit", + "rdtgroup_cbm_overlaps_pseudo_locked", + "rdtgroup_pseudo_locked_in_hierarchy", + "pseudo_lock_measure_cycles", + "pseudo_lock_measure_trigger", + "pseudo_measure_fops", + "rdtgroup_pseudo_lock_create", + "rdtgroup_pseudo_lock_remove", + "pseudo_lock_dev_open", + "pseudo_lock_dev_release", + "pseudo_lock_dev_mremap", + "pseudo_mmap_ops", + "pseudo_lock_dev_mmap", + "pseudo_lock_dev_fops", + "rdt_pseudo_lock_init", + "rdt_pseudo_lock_release", + + # rdtgroup.c + "rdtgroup_mutex", + "rdt_root", + "rdtgroup_default", + "rdt_all_groups", + "resctrl_schema_all", + "resctrl_mounted", + "kn_info", + "kn_mongrp", + "kn_mondata", + "max_name_width", + "last_cmd_status", + "last_cmd_status_buf", + "rdtgroup_setup_root", + "rdtgroup_destroy_root", + "debugfs_resctrl", + "resctrl_debug", + "rdt_last_cmd_clear", + "rdt_last_cmd_puts", + "rdt_last_cmd_printf", + "rdt_staged_configs_clear", + "resctrl_is_mbm_enabled", + "resctrl_is_mbm_event", + "closid_free_map", + "closid_free_map_len", + "closids_supported", + "closid_init", + "closid_alloc", + "closid_free", + "closid_allocated", + "rdtgroup_mode_by_closid", + "rdt_mode_str", + "rdtgroup_mode_str", + "rdtgroup_kn_set_ugid", + "rdtgroup_add_file", + "rdtgroup_seqfile_show", + "rdtgroup_file_write", + "rdtgroup_kf_single_ops", + "kf_mondata_ops", + "is_cpu_list", + "rdtgroup_cpus_show", + "update_closid_rmid", + "cpus_mon_write", + "cpumask_rdtgrp_clear", + "cpus_ctrl_write", + "rdtgroup_cpus_write", + "rdtgroup_remove", + "_update_task_closid_rmid", + "update_task_closid_rmid", + "task_in_rdtgroup", + "__rdtgroup_move_task", + "is_closid_match", + "is_rmid_match", + "rdtgroup_tasks_assigned", + "rdtgroup_task_write_permission", + "rdtgroup_move_task", + "rdtgroup_tasks_write", + "show_rdt_tasks", + "rdtgroup_tasks_show", + "rdtgroup_closid_show", + "rdtgroup_rmid_show", + "proc_resctrl_show", + "rdt_last_cmd_status_show", + "rdt_num_closids_show", + "rdt_default_ctrl_show", + "rdt_min_cbm_bits_show", + "rdt_shareable_bits_show", + "rdt_bit_usage_show", + "rdt_min_bw_show", + "rdt_num_rmids_show", + "rdt_mon_features_show", + "rdt_bw_gran_show", + "rdt_delay_linear_show", + "max_threshold_occ_show", + "rdt_thread_throttle_mode_show", + "max_threshold_occ_write", + "rdtgroup_mode_show", + "resctrl_peer_type", + "rdt_has_sparse_bitmasks_show", + "__rdtgroup_cbm_overlaps", + "rdtgroup_cbm_overlaps", + "rdtgroup_mode_test_exclusive", + "rdtgroup_mode_write", + "rdtgroup_cbm_to_size", + "rdtgroup_size_show", + "mondata_config_read", + "mbm_config_show", + "mbm_total_bytes_config_show", + "mbm_local_bytes_config_show", + "mbm_config_write_domain", + "mon_config_write", + "mbm_total_bytes_config_write", + "mbm_local_bytes_config_write", + "res_common_files", + "rdtgroup_add_files", + "rdtgroup_get_rftype_by_name", + "thread_throttle_mode_init", + "mbm_config_rftype_init", + "rdtgroup_kn_mode_restrict", + "rdtgroup_kn_mode_restore", + "rdtgroup_mkdir_info_resdir", + "fflags_from_resource", + "rdtgroup_create_info_dir", + "mongroup_create_dir", + "is_mba_linear", + "mba_sc_domain_allocate", + "mba_sc_domain_destroy", + "supports_mba_mbps", + "set_mba_sc", + "kernfs_to_rdtgroup", + "rdtgroup_kn_get", + "rdtgroup_kn_put", + "rdtgroup_kn_lock_live", + "rdtgroup_kn_unlock", + "rdt_disable_ctx", + "rdt_enable_ctx", + "schemata_list_add", + "schemata_list_create", + "schemata_list_destroy", + "rdt_get_tree", + "rdt_param", + "rdt_fs_parameters", + "rdt_parse_param", + "rdt_fs_context_free", + "rdt_fs_context_ops", + "rdt_init_fs_context", + "rdt_move_group_tasks", + "free_all_child_rdtgrp", + "rmdir_all_sub", + "rdt_kill_sb", + "rdt_fs_type", + "mon_addfile", + "rmdir_mondata_subdir_allrdtgrp", + "mkdir_mondata_subdir", + "mkdir_mondata_subdir_allrdtgrp", + "mkdir_mondata_subdir_alldom", + "mkdir_mondata_all", + "cbm_ensure_valid", + "__init_one_rdt_domain", + "rdtgroup_init_cat", + "rdtgroup_init_mba", + "rdtgroup_init_alloc", + "mkdir_rdt_prepare_rmid_alloc", + "mkdir_rdt_prepare_rmid_free", + "mkdir_rdt_prepare", + "mkdir_rdt_prepare_clean", + "rdtgroup_mkdir_mon", + "rdtgroup_mkdir_ctrl_mon", + "is_mon_groups", + "rdtgroup_mkdir", + "rdtgroup_rmdir_mon", + "rdtgroup_ctrl_remove", + "rdtgroup_rmdir_ctrl", + "rdtgroup_rmdir", + "mongrp_reparent", + "rdtgroup_rename", + "rdtgroup_show_options", + "rdtgroup_kf_syscall_ops", + "rdtgroup_setup_root", + "rdtgroup_destroy_root", + "rdtgroup_setup_default", + "domain_destroy_mon_state", + "resctrl_offline_domain", + "domain_setup_mon_state", + "resctrl_online_domain", + "resctrl_online_cpu", + "clear_childcpus", + "resctrl_offline_cpu", + "resctrl_init", + "resctrl_exit", + + # trace.h + "TRACE_SYSTEM", + "mon_llc_occupancy_limbo", +]; + +############ + +builtin_non_functions = ["__setup", "__exitcall", "__printf"]; +builtin_one_arg_macros = ["LIST_HEAD", "DEFINE_MUTEX", "DEFINE_STATIC_KEY_FALSE"]; +types = ["bool", "char", "int", "u32", "long", "u64"]; + +def get_array_name(line): + tok = re.search(r'([^\s]+?)\[\]', line) + if (tok is None): + return None; + return tok.group(1); + + +def get_struct_name(line): + tok = re.search(r'struct ([^\s]+?) {', line) + if (tok is None): + return None; + return tok.group(1); + +def get_enum_name(line): + tok = re.search(r'enum ([^\s]+?) {', line) + if (tok is None): + return None; + return tok.group(1); + +def get_union_name(line): + tok = re.search(r'union ([^\s]+?) {', line) + if (tok is None): + return None; + return tok.group(1); + + +def get_macro_name(line): + tok = re.search(r'#define[\s]+([^\s]+?)\(', line) + if (tok): + return tok.group(1); + + tok = re.search(r'#define[\s]+([^\s]+?)[\s]+[^\s]+?\n', line) + if (tok): + return tok.group(1); + + return None; + + +def get_macro_target(line): + tok = re.search(r'[^\s]+?\(([^\s]+?)\);\n', line) + if (tok): + return tok.group(1); + + return None; + + +# Things like 'bool my_bool;' +def get_object_name(line): + # remove things that don't change the meaning of the name + if line.startswith("static "): + line = line[len("static "):]; + if line.startswith("extern "): + line = line[len("extern "):]; + if line.startswith("unsigned "): + line = line[len("unsigned "):]; + + # Note the trailing semicolon.. + tok = re.search(r'([^\s]+)\s[\*]*([^\s\[\],;]+)', line) + if tok: + if tok.group(1) in types: + return tok.group(2); + + tok = re.search(r'struct\s[^\s]+\s[\*]*([^\s;]+)', line) + if tok: + return tok.group(1); + + tok = re.search(r'enum\s[^\s]+\s([^\s;]+)', line) + if tok: + return tok.group(1); + + return None; + + +# Is there a name for this block of code? +# +# Function names are the token before '(' ... assuming there is only one '('. +# This also handles structs and arrays, +def get_block_name(line): + # remove things that don't change the meaning of the name + if (" __read_mostly" in line): + line = line.replace(" __read_mostly", ""); + if (" __initconst" in line): + line = line.replace(" __initconst", ""); + + if line == "enum {\n": + return "anonymous-enum"; + if (line.startswith("#define ")): + return get_macro_name(line); + + if ("=" in line): + tok = re.search(r'[\*]*([^\s\[\]]+?)[\s\[\]]*=', line) + else: + tok = re.search(r'[\*]*([^\s]+?)\(.+?', line) + + if (tok is None): + if ("[]" in line): + return get_array_name(line); + if (line.startswith("struct") and line.endswith("{\n")): + return get_struct_name(line); + if (line.startswith("enum") and line.endswith("{\n")): + return get_enum_name(line); + if (line.startswith("union") and line.endswith("{\n")): + return get_union_name(line); + if (line.endswith(";\n") and '(' not in line): + return get_object_name(line); + if (line.endswith("= {\n") and '(' not in line): + return get_object_name(line); + return None; + + func_name = tok.group(1); + if (func_name in builtin_one_arg_macros): + tok = re.search(r'[^\(]+\(([^\s]+?)\)', line) + if (tok is None): + return None; + return tok.group(1); + elif (func_name == "DEFINE_PER_CPU"): + tok = re.search(r'DEFINE_PER_CPU\(.+?, ([^\s]+?)\)', line) + if (tok is None): + return None; + return tok.group(1); + elif (func_name == "TRACE_EVENT"): + tok = re.search(r'TRACE_EVENT\((.+?),', line) + if (tok is None): + return None; + return tok.group(1); + elif (func_name == "late_initcall"): + return get_macro_target(line); + else: + return func_name; + +def output_function_body(body, file): + # Mandatory whitespace between blocks + if os.lseek(file.fileno(), 0, os.SEEK_CUR) > 0: + file.write("\n".encode()); + + for out_line in body: + file.write(out_line.encode()); + +# Where should we put this block of code? +def output_function(name, body, files): + output = False; + (new_src, new_dst) = files; + + if (len(body)) == 0: + return; + + # Output to both files + if (name is None): + output_function_body(body, new_src); + output_function_body(body, new_dst); + output = True; + if (name in functions_to_keep): + output_function_body(body, new_src); + output = True; + if (name in functions_to_move): + output_function_body(body, new_dst); + output = True; + + if not output: + print("Missing function name: "+name); + #print(body); + +def reset_parser(): + global function_name; + global define_name; + global function_body; + global in_define; + + function_name = None; + define_name = None; + function_body = []; + in_define = False; + +############ + +for file in resctrl_files: + function_name = None; + # function_names take priority over defines, this is only used when + # no function_name was found + define_name = None; + function_body = []; + # Nothing clever - this is just to detect newlines between functions + in_function = False; + in_define = False; + + src_path = SRC_DIR + "/" + str(file); + if (not os.path.isfile(src_path)): + continue; + dst_path = DST_DIR + "/" + str(file); + + orig_file = open(src_path, "r"); + lines = orig_file.readlines(); + + # Now unlink the original file, so it can be re-created with new + # contents. + try: + os.unlink(src_path); + except Exception as err: + print("Failed to unlink source file: {err}"); + sys.exit(1); + + # non-buffering is so we can snoop the fd offset to avoid trailing newlines + new_src = open(src_path, "wb", buffering=0); + new_dst = open(dst_path, "wb", buffering=0); + + for line in lines: + # Empty lines outside a function - reset the function tracking + if (line == "\n" and not in_function): + if function_name is None and define_name is not None: + function_name = define_name; + output_function(function_name, function_body, (new_src, new_dst)); + reset_parser(); + + # Function prototypes are a funny C thing - reset the function tracking + elif (line[0].isspace() and not in_function and line.endswith(");\n")): + function_body += [line]; + output_function(function_name, function_body, (new_src, new_dst)); + reset_parser(); + + # Lines that begin with whitespace are part of the current function. + elif (line[0].isspace()): + function_body += [line]; + + # Next, try to find the kind of line that contains a function name + + # Ignore lines with comment markers, braces + elif (line.startswith("/*")): + function_body += [line]; + elif (line.startswith("*/")): + function_body += [line]; + elif (line.startswith("//")): + function_body += [line]; + elif (line == "{\n"): + function_body += [line]; + in_function = True; + elif (line == "}\n"): + function_body += [line]; + in_function = False; + elif (line == "};\n"): + function_body += [line]; + in_function = False; + + elif (line.startswith("#include")): + function_body += [line]; + elif (line.startswith("#if ")): + function_body += [line]; + elif (line.startswith("#ifdef ")): + function_body += [line]; + elif (line.startswith("#ifndef ")): + function_body += [line]; + elif (line.startswith("#else")): + function_body += [line]; + elif (line.startswith("#endif")): + function_body += [line]; + elif (line.startswith("#undef ")): + function_body += [line]; + elif (line.startswith("#define")): + function_body += [line]; + define_name = get_block_name(line); + if line.endswith("\\\n"): + in_define = True; + elif in_define and line.endswith("\\\n"): + function_body += [line]; + + # goto was always a crime + elif (' ' not in line and line.endswith(":\n")): + function_body += [line]; + + # Try and parse a function/array name + + # Things like late_initcall() aren't function names, but belong to + # the previous function. + elif (get_block_name(line) in builtin_non_functions): + function_body += [line]; + + # Start a new block if we can get a block name for this line + elif (get_block_name(line) != None and function_name is None): + _name = get_block_name(line); + + if (line.endswith("{\n")): + in_function = True; + + # Is this a function prototype? Output it now + if (line.endswith(";\n")): + function_body += [line]; + output_function(_name, function_body, (new_src, new_dst)); + reset_parser(); + else: + function_name = _name; + function_body += [line]; + + # Failed to parse a function name ... did it get split up? + elif (line.startswith("static")): + function_body += [line]; + + else: + print("Unknown: '" + line + "'"); + + # Output whatever is left in the buffer + output_function(function_name, function_body, (new_src, new_dst)); + + orig_file.close();
Signature · 1 line
2.39.2
Quoted 6 lines · James Morse
This is the final series that allows other architectures to implement resctrl. The final patch to move the code has been ommited, but can be generated using the python script at the end of the series.
The final move is a bit of a monster. I don't expect that to get merged as part of this series - we should wait for it to make less impact on other series.
Thanks, again, James.
As with previous versions, I've tested this code (along with additional MPAM code from you and other code we've written), to test MPAM functionality on an Ampere implementation.
I replied to the in the series which introduces CONFIG_RESCTRL_FS_PSEUDO_LOCK to point out how that commit will actually break compilation if that option is not selected, (and I described the minor change needed to fix that).
With that fixed, for the series:
-Carl
James Morse
reviewerHi Carl,
Quoted 14 lines · Carl Worth
Quoted 6 lines · James Morse
This is the final series that allows other architectures to implement resctrl. The final patch to move the code has been ommited, but can be generated using the python script at the end of the series.
The final move is a bit of a monster. I don't expect that to get merged as part of this series - we should wait for it to make less impact on other series.Thanks, again, James.
As with previous versions, I've tested this code (along with additional MPAM code from you and other code we've written), to test MPAM functionality on an Ampere implementation.
Quoted 8 lines
I replied to the in the series which introduces
CONFIG_RESCTRL_FS_PSEUDO_LOCKto point out how that commit will actually break compilation if that option is not selected, (and I described the minor change needed to fix that).With that fixed, for the series:
Tested-by Carl Worth <[email protected]> # arm64
Great - thanks!
(I assume you didn't test the python script that generates the move-to-fs patch)
Thanks,
James