- 16 Feb, 2024 15 commits
-
-
James Morse authored
Depending on the number of monitors available, Arm's MPAM may need to allocate a monitor prior to reading the counter value. Allocating a contended resource may involve sleeping. __check_limbo() and mon_event_count() each make multiple calls to resctrl_arch_rmid_read(), to avoid extra work on contended systems, the allocation should be valid for multiple invocations of resctrl_arch_rmid_read(). The memory or hardware allocated is not specific to a domain. Add arch hooks for this allocation, which need calling before resctrl_arch_rmid_read(). The allocated monitor is passed to resctrl_arch_rmid_read(), then freed again afterwards. The helper can be called on any CPU, and can sleep. Signed-off-by:
James Morse <james.morse@arm.com> Signed-off-by:
Borislav Petkov (AMD) <bp@alien8.de> Reviewed-by:
Shaopeng Tan <tan.shaopeng@fujitsu.com> Reviewed-by:
Reinette Chatre <reinette.chatre@intel.com> Reviewed-by:
Babu Moger <babu.moger@amd.com> Tested-by:
Peter Newman <peternewman@google.com> Tested-by:
-
James Morse authored
MPAM's cache occupancy counters can take a little while to settle once the monitor has been configured. The maximum settling time is described to the driver via a firmware table. The value could be large enough that it makes sense to sleep. To avoid exposing this to resctrl, it should be hidden behind MPAM's resctrl_arch_rmid_read(). resctrl_arch_rmid_read() may be called via IPI meaning it is unable to sleep. In this case, it should return an error if it needs to sleep. This will only affect MPAM platforms where the cache occupancy counter isn't available immediately, nohz_full is in use, and there are no housekeeping CPUs in the necessary domain. There are three callers of resctrl_arch_rmid_read(): __mon_event_count() and __check_limbo() are both called from a non-migrateable context. mon_event_read() invokes __mon_event_count() using smp_call_on_cpu(), which adds work to the target CPUs workqueue. rdtgroup_mutex() is held, meaning this cannot race with the resctrl cpuhp callback. __check_limbo() is invoked via schedule_delayed_work_on() also adds work to a per-cpu workqueue. The remaining call is add_rmid_to_limbo() which is called in response to a user-space syscall that frees an RMID. This opportunistically reads the LLC occupancy counter on the current domain to see if the RMID is over the dirty threshold. This has to disable preemption to avoid reading the wrong domain's value. Disabling preemption here prevents resctrl_arch_rmid_read() from sleeping. add_rmid_to_limbo() walks each domain, but only reads the counter on one domain. If the system has more than one domain, the RMID will always be added to the limbo list. If the RMIDs usage was not over the threshold, it will be removed from the list when __check_limbo() runs. Make this the default behaviour. Free RMIDs are always added to the limbo list for each domain. The user visible effect of this is that a clean RMID is not available for re-allocation immediately after 'rmdir()' completes. This behaviour was never portable as it never happened on a machine with multiple domains. Removing this path allows resctrl_arch_rmid_read() to sleep if its called with interrupts unmasked. Document this is the expected behaviour, and add a might_sleep() annotation to catch changes that won't work on arm64. Signed-off-by:
-
James Morse authored
Intel is blessed with an abundance of monitors, one per RMID, that can be read from any CPU in the domain. MPAMs monitors reside in the MMIO MSC, the number implemented is up to the manufacturer. This means when there are fewer monitors than needed, they need to be allocated and freed. MPAM's CSU monitors are used to back the 'llc_occupancy' monitor file. The CSU counter is allowed to return 'not ready' for a small number of micro-seconds after programming. To allow one CSU hardware monitor to be used for multiple control or monitor groups, the CPU accessing the monitor needs to be able to block when configuring and reading the counter. Worse, the domain may be broken up into slices, and the MMIO accesses for each slice may need performing from different CPUs. These two details mean MPAMs monitor code needs to be able to sleep, and IPI another CPU in the domain to read from a resource that has been sliced. mon_event_read() already invokes mon_event_count() via IPI, which means this isn't possible. On systems using nohz-full, some CPUs need to be interrupted to run kernel work as they otherwise stay in user-space running realtime workloads. Interrupting these CPUs should be avoided, and scheduling work on them may never complete. Change mon_event_read() to pick a housekeeping CPU, (one that is not using nohz_full) and schedule mon_event_count() and wait. If all the CPUs in a domain are using nohz-full, then an IPI is used as the fallback. This function is only used in response to a user-space filesystem request (not the timing sensitive overflow code). This allows MPAM to hide the slice behaviour from resctrl, and to keep the monitor-allocation in monitor.c. When the IPI fallback is used on machines where MPAM needs to make an access on multiple CPUs, the counter read will always fail. Signed-off-by:
-
James Morse authored
The limbo and overflow code picks a CPU to use from the domain's list of online CPUs. Work is then scheduled on these CPUs to maintain the limbo list and any counters that may overflow. cpumask_any() may pick a CPU that is marked nohz_full, which will either penalise the work that CPU was dedicated to, or delay the processing of limbo list or counters that may overflow. Perhaps indefinitely. Delaying the overflow handling will skew the bandwidth values calculated by mba_sc, which expects to be called once a second. Add cpumask_any_housekeeping() as a replacement for cpumask_any() that prefers housekeeping CPUs. This helper will still return a nohz_full CPU if that is the only option. The CPU to use is re-evaluated each time the limbo/overflow work runs. This ensures the work will move off a nohz_full CPU once a housekeeping CPU is available. Signed-off-by:
-
James Morse authored
When switching tasks, the CLOSID and RMID that the new task should use are stored in struct task_struct. For x86 the CLOSID known by resctrl, the value in task_struct, and the value written to the CPU register are all the same thing. MPAM's CPU interface has two different PARTIDs - one for data accesses the other for instruction fetch. Storing resctrl's CLOSID value in struct task_struct implies the arch code knows whether resctrl is using CDP. Move the matching and setting of the struct task_struct properties to use helpers. This allows arm64 to store the hardware format of the register, instead of having to convert it each time. __rdtgroup_move_task()s use of READ_ONCE()/WRITE_ONCE() ensures torn values aren't seen as another CPU may schedule the task being moved while the value is being changed. MPAM has an additional corner-case here as the PMG bits extend the PARTID space. If the scheduler sees a new-CLOSID but old-RMID, the task will dirty an RMID that the limbo code is not watching causing an inaccurate count. x86's RMID are independent values, so the limbo code will still be watching the old-RMID in this circumstance. To avoid this, arm64 needs both the CLOSID/RMID WRITE_ONCE()d together. Both values must be provided together. Because MPAM's RMID values are not unique, the CLOSID must be provided when matching the RMID. Signed-off-by:
-
James Morse authored
MPAM's PMG bits extend its PARTID space, meaning the same PMG value can be used for different control groups. This means once a CLOSID is allocated, all its monitoring ids may still be dirty, and held in limbo. Instead of allocating the first free CLOSID, on architectures where CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID is enabled, search closid_num_dirty_rmid[] to find the cleanest CLOSID. The CLOSID found is returned to closid_alloc() for the free list to be updated. Signed-off-by:
-
James Morse authored
The resctrl CLOSID allocator uses a single 32bit word to track which CLOSID are free. The setting and clearing of bits is open coded. Convert the existing open coded bit manipulations of closid_free_map to use __set_bit() and friends. These don't need to be atomic as this list is protected by the mutex. Signed-off-by:
-
James Morse authored
MPAM's PMG bits extend its PARTID space, meaning the same PMG value can be used for different control groups. This means once a CLOSID is allocated, all its monitoring ids may still be dirty, and held in limbo. Keep track of the number of RMID held in limbo each CLOSID has. This will allow a future helper to find the 'cleanest' CLOSID when allocating. The array is only needed when CONFIG_RESCTRL_RMID_DEPENDS_ON_CLOSID is defined. This will never be the case on x86. Signed-off-by:
-
James Morse authored
MPAMs RMID values are not unique unless the CLOSID is considered as well. alloc_rmid() expects the RMID to be an independent number. Pass the CLOSID in to alloc_rmid(). Use this to compare indexes when allocating. If the CLOSID is not relevant to the index, this ends up comparing the free RMID with itself, and the first free entry will be used. With MPAM the CLOSID is included in the index, so this becomes a walk of the free RMID entries, until one that matches the supplied CLOSID is found. Signed-off-by:
-
James Morse authored
x86 systems identify traffic using the CLOSID and RMID. The CLOSID is used to lookup the control policy, the RMID is used for monitoring. For x86 these are independent numbers. Arm's MPAM has equivalent features PARTID and PMG, where the PARTID is used to lookup the control policy. The PMG in contrast is a small number of bits that are used to subdivide PARTID when monitoring. The cache-occupancy monitors require the PARTID to be specified when monitoring. This means MPAM's PMG field is not unique. There are multiple PMG-0, one per allocated CLOSID/PARTID. If PMG is treated as equivalent to RMID, it cannot be allocated as an independent number. Bitmaps like rmid_busy_llc need to be sized by the number of unique entries for this resource. Treat the combined CLOSID and RMID as an index, and provide architecture helpers to pack and unpack an index. This makes the MPAM values unique. The domain's rmid_busy_llc and rmid_ptrs[] are then sized by index, as are domain mbm_local[] and mbm_total[]. x86 can ignore the CLOSID field when packing and unpacking an index, and report as many indexes as RMID. Signed-off-by:
-
James Morse authored
x86's RMID are independent of the CLOSID. An RMID can be allocated, used and freed without considering the CLOSID. MPAM's equivalent feature is PMG, which is not an independent number, it extends the CLOSID/PARTID space. For MPAM, only PMG-bits worth of 'RMID' can be allocated for a single CLOSID. i.e. if there is 1 bit of PMG space, then each CLOSID can have two monitor groups. To allow resctrl to disambiguate RMID values for different CLOSID, everything in resctrl that keeps an RMID value needs to know the CLOSID too. This will always be ignored on x86. Signed-off-by:
Xin Hao <xhao@linux.alibaba.com> Reviewed-by:
-
James Morse authored
RMIDs are allocated for each monitor or control group directory, because each of these needs its own RMID. For control groups, rdtgroup_mkdir_ctrl_mon() later goes on to allocate the CLOSID. MPAM's equivalent of RMID is not an independent number, so can't be allocated until the CLOSID is known. An RMID allocation for one CLOSID may fail, whereas another may succeed depending on how many monitor groups a control group has. The RMID allocation needs to move to be after the CLOSID has been allocated. Move the RMID allocation out of mkdir_rdt_prepare() to occur in its caller, after the mkdir_rdt_prepare() call. This allows the RMID allocator to know the CLOSID. Signed-off-by:
-
James Morse authored
When monitoring is supported, each monitor and control group is allocated an RMID. For control groups, rdtgroup_mkdir_ctrl_mon() later goes on to allocate the CLOSID. MPAM's equivalent of RMID are not an independent number, so can't be allocated until the CLOSID is known. An RMID allocation for one CLOSID may fail, whereas another may succeed depending on how many monitor groups a control group has. The RMID allocation needs to move to be after the CLOSID has been allocated. Move the RMID allocation and mondata dir creation to a helper. Signed-off-by:
Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Reviewed-by:
-
James Morse authored
rmid_ptrs[] is allocated from dom_data_init() but never free()d. While the exit text ends up in the linker script's DISCARD section, the direction of travel is for resctrl to be/have loadable modules. Add resctrl_put_mon_l3_config() to cleanup any memory allocated by rdt_get_mon_l3_config(). There is no reason to backport this to a stable kernel. Signed-off-by:
-
James Morse authored
tick_nohz_full_mask lists the CPUs that are nohz_full. This is only needed when CONFIG_NO_HZ_FULL is defined. tick_nohz_full_cpu() allows a specific CPU to be tested against the mask, and evaluates to false when CONFIG_NO_HZ_FULL is not defined. The resctrl code needs to pick a CPU to run some work on, a new helper prefers housekeeping CPUs by examining the tick_nohz_full_mask. Hiding the declaration behind #ifdef CONFIG_NO_HZ_FULL forces all the users to be behind an #ifdef too. Move the tick_nohz_full_mask declaration, this lets callers drop the #ifdef, and guard access to tick_nohz_full_mask with IS_ENABLED() or something like tick_nohz_full_cpu(). The definition does not need to be moved as any callers should be removed at compile time unless CONFIG_NO_HZ_FULL is defined. Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Acked-by: Reinette Chatre <reinette.chatre@intel.com> # for resctrl dependency Tested-by:
-
- 24 Jan, 2024 2 commits
-
-
Babu Moger authored
The kernel test robot reported the following warning after commit 54e35eb8 ("x86/resctrl: Read supported bandwidth sources from CPUID"). even though the issue is present even in the original commit 92bd5a13 ("x86/resctrl: Add interface to write mbm_total_bytes_config") which added this function. The reported warning is: $ make C=1 CHECK=scripts/coccicheck arch/x86/kernel/cpu/resctrl/rdtgroup.o ... arch/x86/kernel/cpu/resctrl/rdtgroup.c:1621:5-8: Unneeded variable: "ret". Return "0" on line 1655 Remove the local variable 'ret'. [ bp: Massage commit message, make mbm_config_write_domain() void. ] Fixes: 92bd5a13 ("x86/resctrl: Add interface to write mbm_total_bytes_config") Reported-by:
kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202401241810.jbd8Ipa1-lkp@intel.com/Signed-off-by:
-
Tony Luck authored
The mba_MBps feedback loop increases throttling when a group is using more bandwidth than the target set by the user in the schemata file, and decreases throttling when below target. To avoid possibly stepping throttling up and down on every poll a flag "delta_comp" is set whenever throttling is changed to indicate that the actual change in bandwidth should be recorded on the next poll in "delta_bw". Throttling is only reduced if the current bandwidth plus delta_bw is below the user target. This algorithm works well if the workload has steady bandwidth needs. But it can go badly wrong if the workload moves to a different phase just as the throttling level changed. E.g. if the workload becomes essentially idle right as throttling level is increased, the value calculated for delta_bw will be more or less the old bandwidth level. If the workload then resumes, Linux may never reduce throttling because current bandwidth plus delta_bw is above the target set by the user. Implement a simpler heuristic by assuming that in the worst case the currently measured bandwidth is being controlled by the current level of throttling. Compute how much it may increase if throttling is relaxed to the next higher level. If that is still below the user target, then it is ok to reduce the amount of throttling. Fixes: ba0f26d8 ("x86/intel_rdt/mba_sc: Prepare for feedback loop") Reported-by:
Xiaochen Shen <xiaochen.shen@intel.com> Signed-off-by:
Tony Luck <tony.luck@intel.com> Signed-off-by:
-
- 23 Jan, 2024 2 commits
-
-
Babu Moger authored
If the BMEC (Bandwidth Monitoring Event Configuration) feature is supported, the bandwidth events can be configured. The maximum supported bandwidth bitmask can be read from CPUID: CPUID_Fn80000020_ECX_x03 [Platform QoS Monitoring Bandwidth Event Configuration] Bits Description 31:7 Reserved 6:0 Identifies the bandwidth sources that can be tracked. While at it, move the mask checking to mon_config_write() before iterating over all the domains. Also, print the valid bitmask when the user tries to configure invalid event configuration value. The CPUID details are documented in the Processor Programming Reference (PPR) Vol 1.1 for AMD Family 19h Model 11h B1 - 55901 Rev 0.25 in the Link tag. Fixes: dc2a3e85 ("x86/resctrl: Add interface to read mbm_total_bytes_config") Signed-off-by:
-
Babu Moger authored
The QOS Memory Bandwidth Enforcement Limit is reported by CPUID_Fn80000020_EAX_x01 and CPUID_Fn80000020_EAX_x02: Bits Description 31:0 BW_LEN: Size of the QOS Memory Bandwidth Enforcement Limit. Newer processors can support higher bandwidth limit than the current hard-coded value. Remove latter and detect using CPUID instead. Also, update the register variables eax and edx to match the AMD CPUID definition. The CPUID details are documented in the Processor Programming Reference (PPR) Vol 1.1 for AMD Family 19h Model 11h B1 - 55901 Rev 0.25 in the Link tag below. Fixes: 4d05bf71 ("x86/resctrl: Introduce AMD QOS feature") Signed-off-by:
-
- 22 Jan, 2024 1 commit
-
-
Tony Luck authored
In a "W=1" build gcc throws a warning: arch/x86/kernel/cpu/resctrl/core.c: In function ‘cache_alloc_hsw_probe’: arch/x86/kernel/cpu/resctrl/core.c:139:16: warning: variable ‘h’ set but not used Switch from wrmsr_safe() to wrmsrl_safe(), and from rdmsr() to rdmsrl() using a single u64 argument for the MSR value instead of the pair of u32 for the high and low halves. Signed-off-by:
-
- 21 Jan, 2024 20 commits
-
-
Linus Torvalds authored
-
https://evilpiepirate.org/git/bcachefsLinus Torvalds authored
Pull more bcachefs updates from Kent Overstreet: "Some fixes, Some refactoring, some minor features: - Assorted prep work for disk space accounting rewrite - BTREE_TRIGGER_ATOMIC: after combining our trigger callbacks, this makes our trigger context more explicit - A few fixes to avoid excessive transaction restarts on multithreaded workloads: fstests (in addition to ktest tests) are now checking slowpath counters, and that's shaking out a few bugs - Assorted tracepoint improvements - Starting to break up bcachefs_format.h and move on disk types so they're with the code they belong to; this will make room to start documenting the on disk format better. - A few minor fixes" * tag 'bcachefs-2024-01-21' of https://evilpiepirate.org/git/bcachefs: (46 commits) bcachefs: Improve inode_to_text() bcachefs: logged_ops_format.h bcachefs: reflink_format.h bcachefs; extents_format.h bcachefs: ec_format.h bcachefs: subvolume_format.h bcachefs: snapshot_format.h bcachefs: alloc_background_format.h bcachefs: xattr_format.h bcachefs: dirent_format.h bcachefs: inode_format.h bcachefs; quota_format.h bcachefs: sb-counters_format.h bcachefs: counters.c -> sb-counters.c bcachefs: comment bch_subvolume bcachefs: bch_snapshot::btime bcachefs: add missing __GFP_NOWARN bcachefs: opts->compression can now also be applied in the background bcachefs: Prep work for variable size btree node buffers bcachefs: grab s_umount only if snapshotting ... -
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull timer updates from Thomas Gleixner: "Updates for time and clocksources: - A fix for the idle and iowait time accounting vs CPU hotplug. The time is reset on CPU hotplug which makes the accumulated systemwide time jump backwards. - Assorted fixes and improvements for clocksource/event drivers" * tag 'timers-core-2024-01-21' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: tick-sched: Fix idle and iowait sleeptime accounting vs CPU hotplug clocksource/drivers/ep93xx: Fix error handling during probe clocksource/drivers/cadence-ttc: Fix some kernel-doc warnings clocksource/drivers/timer-ti-dm: Fix make W=n kerneldoc warnings clocksource/timer-riscv: Add riscv_clock_shutdown callback dt-bindings: timer: Add StarFive JH8100 clint dt-bindings: timer: thead,c900-aclint-mtimer: separate mtime and mtimecmp regs -
git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linuxLinus Torvalds authored
Pull powerpc fixes from Aneesh Kumar: - Increase default stack size to 32KB for Book3S Thanks to Michael Ellerman. * tag 'powerpc-6.8-2' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: powerpc/64s: Increase default stack size to 32KB
-
Kent Overstreet authored
Add line breaks - inode_to_text() is now much easier to read. Signed-off-by:Kent Overstreet <kent.overstreet@linux.dev>
-
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
bcachefs_format.h has gotten too big; let's do some organizing. Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Signed-off-by: -
Kent Overstreet authored
Add a field to bch_snapshot for creation time; this will be important when we start exposing the snapshot tree to userspace. Signed-off-by:
-
