Add a helper to translate from the item list head to the
xfs_extent_free_item structure and use it so shorten assignments
and avoid the need for extra local variables.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
All callers of xfs_perag_intent_get have a fsbno and need boilerplate
code to turn that into an agno. Just pass the fsbno to
xfs_perag_intent_get and look up the agno there.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Convert the boolean to skip discard on free into a proper flags field so
that we can add more flags in the next patch.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Pass the incore EFI structure to the tracepoints instead of open-coding
the argument passing. This cleans up the call sites a bit.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Currently, the XFS_SB_CRC_OFF macro uses the incore superblock struct
(xfs_sb) to compute the address of sb_crc within the ondisk superblock
struct (xfs_dsb). This is a landmine if we ever change the layout of
the incore superblock (as we're about to do), so redefine the macro
to use xfs_dsb to compute the layout of xfs_dsb.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Get rid of the largely pointless xfs_cross_rename and xfs_finish_rename
now that we've refactored its parent.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Move the directory entry update hook code to xfs_dir2 so that it is
mostly consolidated with the higher level directory functions. Retain
the exports so that online fsck can still send notifications through the
hooks.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Create a new libxfs function to rename two directory entries. The
upcoming metadata directory feature will need this to replace a metadata
inode directory entry.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Create a new libxfs function to exchange two directory entries.
The upcoming metadata directory feature will need this to replace a
metadata inode directory entry.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Create a new libxfs function to remove a (name, inode) entry from a
directory. The upcoming metadata directory feature will need this to
create a metadata directory tree.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Create a libxfs helper function that marks an inode free on disk.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Create a new libxfs function to link an existing inode into a directory.
The upcoming metadata directory feature will need this to create a
metadata directory tree.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Create a new libxfs function to link a newly created inode into a
directory. The upcoming metadata directory feature will need this to
create a metadata directory tree.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
INIT_XATTRS is overloaded here -- it's set during the creat process when
we think that we're immediately going to set some ACL xattrs to save
time. However, it's also used by the parent pointers code to enable the
attr fork in preparation to receive ppptr xattrs. This results in
xfs_has_parent() branches scattered around the codebase to turn on
INIT_XATTRS.
Linkable files are created far more commonly than unlinkable temporary
files or directory tree roots, so we should centralize this logic in
xfs_inode_init. For the three callers that don't want parent pointers
(online repiar tempfiles, unlinkable tempfiles, rootdir creation) we
provide an UNLINKABLE flag to skip attr fork initialization.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Create a helper that calls dqalloc to allocate and grab a reference to
dquots for the user, group, and project ids listed in an icreate
structure. This simplifies the creat-related dqalloc callsites
scattered around the code base.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Move the initialization of the xfs_icreate_args structure out of
xfs_create and xfs_create_tempfile into their callers so that we can set
the new inode's attributes in one place and pass that through instead of
open coding the collection of attributes all over the code.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Move all the code that initializes a new inode's attributes from the
icreate_args structure and the parent directory into libxfs.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
There are two parts to initializing a newly allocated inode: setting up
the incore structures, and initializing the new inode core based on the
parent inode and the current user's environment. The initialization
code is not specific to the kernel, so we would like to share that with
userspace by hoisting it to libxfs. Therefore, split xfs_icreate into
separate functions to prepare for the next few patches.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Use xfs_trans_ichgtime to set the inode times when allocating an inode,
instead of open-coding them here.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Enable xfs_trans_ichgtime to change the inode access time so that we can
use this function to set inode times when allocating inodes instead of
open-coding it.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Callers that want to create an inode currently pass all possible file
attribute values for the new inode into xfs_init_new_inode as ten
separate parameters. This causes two code maintenance issues: first, we
have large multi-line call sites which programmers must read carefully
to make sure they did not accidentally invert a value. Second, all
three file id parameters must be passed separately to the quota
functions; any discrepancy results in quota count errors.
Clean this up by creating a new icreate_args structure to hold all this
information, some helpers to initialize them properly, and make the
callers pass this structure through to the creation function, whose name
we shorten to xfs_icreate. This eliminates the issues, enables us to
keep the inode init code in sync with userspace via libxfs, and is
needed for future metadata directory tree management.
(A subsequent cleanup will also fix the quota alloc calls.)
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Hoist the inode flag conversion functions into libxfs so that we can
keep them in sync. Do this by creating a new xfs_inode_util.c file in
libxfs.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Move the extent size helpers to xfs_bmap.c in libxfs since they're used
there already.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Move these inode predicate functions to xfs_inode.[ch] since they're not
reflink functions.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
I noticed that callers of xfs_qm_vop_dqalloc use the following code to
compute the anticipated uid of the new file:
mapped_fsuid(idmap, &init_user_ns);
whereas the VFS uses a slightly different computation for actually
assigning i_uid:
mapped_fsuid(idmap, i_user_ns(inode));
Technically, these are not the same things. According to Christian
Brauner, the only time that inode->i_sb->s_user_ns != &init_user_ns is
when the filesystem was mounted in a new mount namespace by an
unpriviledged user. XFS does not allow this, which is why we've never
seen bug reports about quotas being incorrect or the uid checks in
xfs_qm_vop_create_dqattach tripping debug assertions.
However, this /is/ a logic bomb, so let's make the code consistent.
Link: https://lore.kernel.org/linux-fsdevel/20240617-weitblick-gefertigt-4a41f37119fa@brauner/
Fixes: c14329d39f ("fs: port fs{g,u}id helpers to mnt_idmap")
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
generic/388 has an annoying tendency to fail like this during log
recovery:
XFS (sda4): Unmounting Filesystem 435fe39b-82b6-46ef-be56-819499585130
XFS (sda4): Mounting V5 Filesystem 435fe39b-82b6-46ef-be56-819499585130
XFS (sda4): Starting recovery (logdev: internal)
00000000: 49 4e 81 b6 03 02 00 00 00 00 00 07 00 00 00 07 IN..............
00000010: 00 00 00 01 00 00 00 00 00 00 00 00 00 00 00 10 ................
00000020: 35 9a 8b c1 3e 6e 81 00 35 9a 8b c1 3f dc b7 00 5...>n..5...?...
00000030: 35 9a 8b c1 3f dc b7 00 00 00 00 00 00 3c 86 4f 5...?........<.O
00000040: 00 00 00 00 00 00 02 f3 00 00 00 00 00 00 00 00 ................
00000050: 00 00 1f 01 00 00 00 00 00 00 00 02 b2 74 c9 0b .............t..
00000060: ff ff ff ff d7 45 73 10 00 00 00 00 00 00 00 2d .....Es........-
00000070: 00 00 07 92 00 01 fe 30 00 00 00 00 00 00 00 1a .......0........
00000080: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00000090: 35 9a 8b c1 3b 55 0c 00 00 00 00 00 04 27 b2 d1 5...;U.......'..
000000a0: 43 5f e3 9b 82 b6 46 ef be 56 81 94 99 58 51 30 C_....F..V...XQ0
XFS (sda4): Internal error Bad dinode after recovery at line 539 of file fs/xfs/xfs_inode_item_recover.c. Caller xlog_recover_items_pass2+0x4e/0xc0 [xfs]
CPU: 0 PID: 2189311 Comm: mount Not tainted 6.9.0-rc4-djwx #rc4
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS ?-20171121_152543-x86-ol7-builder-01.us.oracle.com-4.el7.1 04/01/2014
Call Trace:
<TASK>
dump_stack_lvl+0x4f/0x60
xfs_corruption_error+0x90/0xa0
xlog_recover_inode_commit_pass2+0x5f1/0xb00
xlog_recover_items_pass2+0x4e/0xc0
xlog_recover_commit_trans+0x2db/0x350
xlog_recovery_process_trans+0xab/0xe0
xlog_recover_process_data+0xa7/0x130
xlog_do_recovery_pass+0x398/0x840
xlog_do_log_recovery+0x62/0xc0
xlog_do_recover+0x34/0x1d0
xlog_recover+0xe9/0x1a0
xfs_log_mount+0xff/0x260
xfs_mountfs+0x5d9/0xb60
xfs_fs_fill_super+0x76b/0xa30
get_tree_bdev+0x124/0x1d0
vfs_get_tree+0x17/0xa0
path_mount+0x72b/0xa90
__x64_sys_mount+0x112/0x150
do_syscall_64+0x49/0x100
entry_SYSCALL_64_after_hwframe+0x4b/0x53
</TASK>
XFS (sda4): Corruption detected. Unmount and run xfs_repair
XFS (sda4): Metadata corruption detected at xfs_dinode_verify.part.0+0x739/0x920 [xfs], inode 0x427b2d1
XFS (sda4): Filesystem has been shut down due to log error (0x2).
XFS (sda4): Please unmount the filesystem and rectify the problem(s).
XFS (sda4): log mount/recovery failed: error -117
XFS (sda4): log mount failed
This inode log item recovery failing the dinode verifier after
replaying the contents of the inode log item into the ondisk inode.
Looking back into what the kernel was doing at the time of the fs
shutdown, a thread was in the middle of running a series of
transactions, each of which committed changes to the inode.
At some point in the middle of that chain, an invalid (at least
according to the verifier) change was committed. Had the filesystem not
shut down in the middle of the chain, a subsequent transaction would
have corrected the invalid state and nobody would have noticed. But
that's not what happened here. Instead, the invalid inode state was
committed to the ondisk log, so log recovery tripped over it.
The actual defect here was an overzealous inode verifier, which was
fixed in a separate patch. This patch adds some transaction precommit
functions for CONFIG_XFS_DEBUG=y mode so that we can detect these kinds
of transient errors at transaction commit time, where it's much easier
to find the root cause.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
The acpi_cst_latency_cmp() comparison function currently used for
sorting C-state latencies does not satisfy transitivity, causing
incorrect sorting results.
Specifically, if there are two valid acpi_processor_cx elements A and B
and one invalid element C, it may occur that A < B, A = C, and B = C.
Sorting algorithms assume that if A < B and A = C, then C < B, leading
to incorrect ordering.
Given the small size of the array (<=8), we replace the library sort
function with a simple insertion sort that properly ignores invalid
elements and sorts valid ones based on latency. This change ensures
correct ordering of the C-state latencies.
Fixes: 65ea8f2c6e ("ACPI: processor idle: Fix up C-state latency if not ordered")
Reported-by: Julian Sikorski <belegdol@gmail.com>
Closes: https://lore.kernel.org/lkml/70674dc7-5586-4183-8953-8095567e73df@gmail.com
Signed-off-by: Kuan-Wei Chiu <visitorckw@gmail.com>
Tested-by: Julian Sikorski <belegdol@gmail.com>
Cc: All applicable <stable@vger.kernel.org>
Link: https://patch.msgid.link/20240701205639.117194-1-visitorckw@gmail.com
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
With Sub-NUMA Cluster (SNC) mode enabled, the scope of monitoring resources
is per-NODE instead of per-L3 cache. Backwards compatibility is maintained
by providing files in the mon_L3_XX directories that sum event counts
for all SNC nodes sharing an L3 cache.
New files provide per-SNC node event counts.
Users should be aware that SNC mode also affects the amount of L3 cache
available for allocation within each SNC node.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-20-tony.luck@intel.com
There isn't a simple hardware bit that indicates whether a CPU is running in
Sub-NUMA Cluster (SNC) mode. Infer the state by comparing the number of CPUs
sharing the L3 cache with CPU0 to the number of CPUs in the same NUMA node as
CPU0.
Add the missing definition of pr_fmt() to monitor.c. This wasn't noticed
before as there are only "can't happen" console messages from this file.
[ bp: Massage commit message. ]
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-19-tony.luck@intel.com
Hardware has two RMID configuration options for SNC systems. The default
mode divides RMID counters between SNC nodes. E.g. with 200 RMIDs and
two SNC nodes per L3 cache RMIDs 0..99 are used on node 0, and 100..199
on node 1. This isn't compatible with Linux resctrl usage. On this
example system a process using RMID 5 would only update monitor counters
while running on SNC node 0.
The other mode is "RMID Sharing Mode". This is enabled by clearing bit
0 of the RMID_SNC_CONFIG (0xCA0) model specific register. In this mode
the number of logical RMIDs is the number of physical RMIDs (from CPUID
leaf 0xF) divided by the number of SNC nodes per L3 cache instance. A
process can use the same RMID across different SNC nodes.
See the "Intel Resource Director Technology Architecture Specification"
for additional details.
When SNC is enabled, update the MSR when a monitor domain is marked
online. Technically this is overkill. It only needs to be done once
per L3 cache instance rather than per SNC domain. But there is no harm
in doing it more than once, and this is not in a critical path.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Link: https://lore.kernel.org/r/20240702173820.90368-3-tony.luck@intel.com
Legacy resctrl monitor files must provide the sum of event values across
all Sub-NUMA Cluster (SNC) domains that share an L3 cache instance.
There are now two cases:
1) A specific domain is provided in struct rmid_read
This is either a non-SNC system, or the request is to read data
from just one SNC node.
2) Domain pointer is NULL. In this case the cacheinfo field in struct
rmid_read indicates that all SNC nodes that share that L3 cache
instance should have the event read and return the sum of all
values.
Update the CPU sanity check. The existing check that an event is read
from a CPU in the requested domain still applies when reading a single
domain. But when summing across domains a more relaxed check that the
current CPU is in the scope of the L3 cache instance is appropriate
since the MSRs to read events are scoped at L3 cache level.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-17-tony.luck@intel.com
mon_event_read() fills out most fields of the struct rmid_read that is passed
via an smp_call*() function to a CPU that is part of the correct domain to
read the monitor counters.
With Sub-NUMA Cluster (SNC) mode there are now two cases to handle:
1) Reading a file that returns a value for a single domain.
+ Choose the CPU to execute from the domain cpu_mask
2) Reading a file that must sum across domains sharing an L3 cache
instance.
+ Indicate to called code that a sum is needed by passing a NULL
rdt_mon_domain pointer.
+ Choose the CPU from the L3 shared_cpu_map.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-16-tony.luck@intel.com
In SNC mode, there are multiple subdirectories in each L3 level monitor
directory (one for each SNC node). If all the CPUs in an SNC node are taken
offline, just remove the SNC directory for that node. In non-SNC mode, or when
the last SNC node directory is removed, remove the L3 monitor directory.
Add a helper function to avoid duplicated code.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Link: https://lore.kernel.org/r/20240702173820.90368-2-tony.luck@intel.com
When SNC mode is enabled, create subdirectories and files to monitor at the SNC
node granularity. Legacy behavior is preserved by tagging the monitor files at
the L3 granularity with the "sum" attribute. When the user reads these files
the kernel will read monitor data from all SNC nodes that share the same L3
cache instance and return the aggregated value to the user.
Note that the "domid" field for files that must sum across SNC domains has the
L3 cache instance id, while non-summing files use the domain id.
The "sum" files do not need to make a call to mon_event_read() to initialize
the MBM counters. This will be handled by initializing the individual SNC nodes
that share the L3.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-14-tony.luck@intel.com
When Sub-NUMA Cluster (SNC) mode is enabled, the legacy monitor reporting files
must report the sum of the data from all of the SNC nodes that share the L3
cache that is referenced by the monitor file.
Resctrl squeezes all the attributes of these files into 32 bits so they can be
stored in the "priv" field of struct kernfs_node.
Currently, only three monitor events are defined by enum resctrl_event_id so
reducing it from 8 bits to 7 bits still provides more than enough space to
represent all the known event types.
But note that this choice was arbitrary. The "rid" field is also far wider
than needed for the current number of resource id types. This structure is
purely internal to resctrl, no ABI issues with modifying it. Subsequent changes
may rearrange the allocation of bits between each of the fields as needed.
Give the bit to a new "sum" field that indicates that reading this file must
sum across SNC nodes. This bit also indicates that the domid field is the id of
an L3 cache (instead of a domain id) to find which domains must be summed.
Fix up other issues in the kerneldoc description for mon_data_bits.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-13-tony.luck@intel.com
In Sub-NUMA Cluster (SNC) mode Linux must create the monitor
files in the original "mon_L3_XX" directories and also in each
of the "mon_sub_L3_YY" directories.
Refactor mkdir_mondata_subdir() to move the creation of monitoring files
into a helper function to avoid the need to duplicate code later.
No functional change.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-12-tony.luck@intel.com
New semantics rely on some struct rmid_read members having NULL values to
distinguish between the SNC and non-SNC scenarios. resctrl can thus no longer
rely on this struct not being initialized properly.
Initialize all on-stack declarations of struct rmid_read:
rdtgroup_mondata_show()
mbm_update()
mkdir_mondata_subdir()
to ensure that garbage values from the stack are not passed down to other
functions.
[ bp: Massage commit message. ]
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-11-tony.luck@intel.com
When a user reads a monitor file rdtgroup_mondata_show() calls mon_event_read()
to package up all the required details into an rmid_read structure which is
passed across the smp_call*() infrastructure to code that will read data from
hardware and return the value (or error status) in the rmid_read structure.
Sub-NUMA Cluster (SNC) mode adds files with new semantics. These require the
smp_call-ed code to sum event data from all domains that share an L3 cache.
Add a pointer to the L3 "cacheinfo" structure to struct rmid_read for the data
collection routines to use to pick the domains to be summed.
[ Reinette: the rmid_read structure has become complex enough so document each
of its fields and provide the kerneldoc documentation for struct rmid_read. ]
Co-developed-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-10-tony.luck@intel.com
When SNC is enabled, monitoring data is collected at the SNC node granularity,
but must be reported at L3-cache granularity for backwards compatibility in
addition to reporting at the node level.
Add a "ci" field to the rdt_mon_domain structure to save the cache information
about the enclosing L3 cache for the domain. This provides:
1) The cache id which is needed to compose the name of the legacy monitoring
directory, and to determine which domains should be summed to provide
L3-scoped data.
2) The shared_cpu_map which is needed to determine which CPUs can be used to
read the RMID counters with the MSR interface.
This is the first step to an eventual goal of monitor reporting files like this
(for a system with two SNC nodes per L3):
$ cd /sys/fs/resctrl/mon_data
$ tree mon_L3_00
mon_L3_00 <- 00 here is L3 cache id
├── llc_occupancy \ These files provide legacy support
├── mbm_local_bytes > for non-SNC aware monitor apps
├── mbm_total_bytes / that expect data at L3 cache level
├── mon_sub_L3_00 <- 00 here is SNC node id
│ ├── llc_occupancy \ These files are finer grained
│ ├── mbm_local_bytes > data from each SNC node
│ └── mbm_total_bytes /
└── mon_sub_L3_01
├── llc_occupancy \
├── mbm_local_bytes > As above, but for node 1.
└── mbm_total_bytes /
[ bp: Massage commit message. ]
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-9-tony.luck@intel.com
When SNC is enabled there is a mismatch between the MBA control function
which operates at L3 cache scope and the MBM monitor functions which
measure memory bandwidth on each SNC node.
Block use of the mba_MBps when scopes for MBA/MBM do not match.
Improve user diagnostics by adding invalfc() message when mba_MBps
is not supported.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-8-tony.luck@intel.com
Intel Sub-NUMA Cluster (SNC) is a feature that subdivides the CPU cores
and memory controllers on a socket into two or more groups. These are
presented to the operating system as NUMA nodes.
This may enable some workloads to have slightly lower latency to memory
as the memory controller(s) in an SNC node are electrically closer to the
CPU cores on that SNC node. This cost may be offset by lower bandwidth
since the memory accesses for each core can only be interleaved between
the memory controllers on the same SNC node.
Resctrl monitoring on an Intel system depends upon attaching RMIDs to tasks
to track L3 cache occupancy and memory bandwidth. There is an MSR that
controls how the RMIDs are shared between SNC nodes.
The default mode divides them numerically. E.g. when there are two SNC
nodes on a socket the lower number half of the RMIDs are given to the
first node, the remainder to the second node. This would be difficult
to use with the Linux resctrl interface as specific RMID values assigned
to resctrl groups are not visible to users.
RMID sharing mode divides the physical RMIDs evenly between SNC nodes
but uses a logical RMID in the IA32_PQR_ASSOC MSR. For example a system
with 200 physical RMIDs (as enumerated by CPUID leaf 0xF) that has two
SNC nodes per L3 cache instance would have 100 logical RMIDs available
for Linux to use. A task running on SNC node 0 with RMID 5 would
accumulate LLC occupancy and MBM bandwidth data in physical RMID 5.
Another task using RMID 5, but running on SNC node 1 would accumulate
data in physical RMID 105.
Even with this renumbering SNC mode requires several changes in resctrl
behavior for correct operation.
Add a static global to arch/x86/kernel/cpu/resctrl/monitor.c to indicate
how many SNC domains share an L3 cache instance. Initialize this to
"1". Runtime detection of SNC mode will adjust this value.
Update all places to take appropriate action when SNC mode is enabled:
1) The number of logical RMIDs per L3 cache available for use is the
number of physical RMIDs divided by the number of SNC nodes.
2) Likewise the "mon_scale" value must be divided by the number of SNC
nodes.
3) Add a function to convert from logical RMID values (assigned to
tasks and loaded into the IA32_PQR_ASSOC MSR on context switch)
to physical RMID values to load into IA32_QM_EVTSEL MSR when
reading counters on each SNC node.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-7-tony.luck@intel.com
Currently supported resctrl features are all domain scoped the same as the
scope of the L2 or L3 caches.
Add RESCTRL_L3_NODE as a new option for features that are scoped at the
same granularity as NUMA nodes. This is needed for Intel's Sub-NUMA
Cluster (SNC) feature where monitoring features are divided between
nodes that share an L3 cache.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-6-tony.luck@intel.com
The same rdt_domain structure is used for both control and monitor
functions. But this results in wasted memory as some of the fields are
only used by control functions, while most are only used for monitor
functions.
Split into separate rdt_ctrl_domain and rdt_mon_domain structures with
just the fields required for control and monitoring respectively.
Similar split of the rdt_hw_domain structure into rdt_hw_ctrl_domain
and rdt_hw_mon_domain.
Signed-off-by: Tony Luck <tony.luck@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Reinette Chatre <reinette.chatre@intel.com>
Tested-by: Babu Moger <babu.moger@amd.com>
Link: https://lore.kernel.org/r/20240628215619.76401-5-tony.luck@intel.com
