At region creation time the next region-id is atomically cached so that
there is predictability of region device names. If that region is
destroyed and then a new one is created the region id increments. That
ends up looking like a memory leak, or is otherwise surprising that
identifiers roll forward even after destroying all previously created
regions.
Try to reuse rather than free old region ids at region release time.
While this fixes a cosmetic issue, the needlessly advancing memory
region-id gives the appearance of a memory leak, hence the "Fixes" tag,
but no "Cc: stable" tag.
Cc: Ben Widawsky <bwidawsk@kernel.org>
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Fixes: 779dd20cfb ("cxl/region: Add region creation support")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Link: https://lore.kernel.org/r/166752186062.947915.13200195701224993317.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
When programming port decode targets, the algorithm wants to ensure that
two devices are compatible to be programmed as peers beneath a given
port. A compatible peer is a target that shares the same dport, and
where that target's interleave position also routes it to the same
dport. Compatibility is determined by the device's interleave position
being >= to distance. For example, if a given dport can only map every
Nth position then positions less than N away from the last target
programmed are incompatible.
The @distance for the host-bridge's cxl_port in a simple dual-ported
host-bridge configuration with 2 direct-attached devices is 1, i.e. An
x2 region divided by 2 dports to reach 2 region targets.
An x4 region under an x2 host-bridge would need 2 intervening switches
where the @distance at the host bridge level is 2 (x4 region divided by
2 switches to reach 4 devices).
However, the distance between peers underneath a single ported
host-bridge is always zero because there is no limit to the number of
devices that can be mapped. In other words, there are no decoders to
program in a passthrough, all descendants are mapped and distance only
starts matters for the intervening descendant ports of the passthrough
port.
Add tracking for the number of dports mapped to a port, and use that to
detect the passthrough case for calculating @distance.
Cc: <stable@vger.kernel.org>
Reported-by: Bobo WL <lmw.bobo@gmail.com>
Reported-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: http://lore.kernel.org/r/20221010172057.00001559@huawei.com
Fixes: 27b3f8d138 ("cxl/region: Program target lists")
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Link: https://lore.kernel.org/r/166752185440.947915.6617495912508299445.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
When a cxl_nvdimm object goes through a ->remove() event (device
physically removed, nvdimm-bridge disabled, or nvdimm device disabled),
then any associated regions must also be disabled. As highlighted by the
cxl-create-region.sh test [1], a single device may host multiple
regions, but the driver was only tracking one region at a time. This
leads to a situation where only the last enabled region per nvdimm
device is cleaned up properly. Other regions are leaked, and this also
causes cxl_memdev reference leaks.
Fix the tracking by allowing cxl_nvdimm objects to track multiple region
associations.
Cc: <stable@vger.kernel.org>
Link: https://github.com/pmem/ndctl/blob/main/test/cxl-create-region.sh [1]
Reported-by: Vishal Verma <vishal.l.verma@intel.com>
Fixes: 04ad63f086 ("cxl/region: Introduce cxl_pmem_region objects")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Link: https://lore.kernel.org/r/166752183647.947915.2045230911503793901.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
When a region is deleted any targets that have been previously assigned
to that region hold references to it. Trigger those references to
drop by detaching all targets at unregister_region() time.
Otherwise that region object will leak as userspace has lost the ability
to detach targets once region sysfs is torn down.
Cc: <stable@vger.kernel.org>
Fixes: b9686e8c8e ("cxl/region: Enable the assignment of endpoint decoders to regions")
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Link: https://lore.kernel.org/r/166752183055.947915.17681995648556534844.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Some regions may not have any address space allocated. Skip them when
validating HPA order otherwise a crash like the following may result:
devm_cxl_add_region: cxl_acpi cxl_acpi.0: decoder3.4: created region9
BUG: kernel NULL pointer dereference, address: 0000000000000000
[..]
RIP: 0010:store_targetN+0x655/0x1740 [cxl_core]
[..]
Call Trace:
<TASK>
kernfs_fop_write_iter+0x144/0x200
vfs_write+0x24a/0x4d0
ksys_write+0x69/0xf0
do_syscall_64+0x3a/0x90
store_targetN+0x655/0x1740:
alloc_region_ref at drivers/cxl/core/region.c:676
(inlined by) cxl_port_attach_region at drivers/cxl/core/region.c:850
(inlined by) cxl_region_attach at drivers/cxl/core/region.c:1290
(inlined by) attach_target at drivers/cxl/core/region.c:1410
(inlined by) store_targetN at drivers/cxl/core/region.c:1453
Cc: <stable@vger.kernel.org>
Fixes: 384e624bb2 ("cxl/region: Attach endpoint decoders")
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Link: https://lore.kernel.org/r/166752182461.947915.497032805239915067.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
When an intermediate port's decoders have been exhausted by existing
regions, and creating a new region with the port in question in it's
hierarchical path is attempted, cxl_port_attach_region() fails to find a
port decoder (as would be expected), and drops into the failure / cleanup
path.
However, during cleanup of the region reference, a sanity check attempts
to dereference the decoder, which in the above case didn't exist. This
causes a NULL pointer dereference BUG.
To fix this, refactor the decoder allocation and de-allocation into
helper routines, and in this 'free' routine, check that the decoder,
@cxld, is valid before attempting any operations on it.
Cc: <stable@vger.kernel.org>
Suggested-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Dave Jiang <dave.jiang@intel.com>
Fixes: 384e624bb2 ("cxl/region: Attach endpoint decoders")
Link: https://lore.kernel.org/r/20221101074100.1732003-1-vishal.l.verma@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Writes to the device must include an offset and size as defined in
CXL 2.0 8.2.9.5.2.4 Set LSA (Opcode 4103h)
Fixes tag is non obvious as this code has been through several
reworks and variable names + wasn't in use until the addition
of the region code.
Due to a bug in QEMU CXL emulation this overrun resulted in QEMU
crashing.
Reported-by: Bobo WL <lmw.bobo@gmail.com>
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Fixes: 60b8f17215 ("cxl/pmem: Translate NVDIMM label commands to CXL label commands")
Link: https://lore.kernel.org/r/20220815154044.24733-3-Jonathan.Cameron@huawei.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Not all decoders have a commit callback.
The CXL specification allows a host bridge with a single root port to
have no explicit HDM decoders. Currently the region driver assumes there
are none. As such the CXL core creates a special pass through decoder
instance without a commit callback.
Prior to this patch, the ->commit() callback was called unconditionally.
Thus a configuration with 1 Host Bridge, 1 Root Port, 1 switch with
multiple downstream ports below which there are multiple CXL type 3
devices results in a situation where committing the region causes a null
pointer dereference.
Reported-by: Bobo WL <lmw.bobo@gmail.com>
Fixes: 176baefb2e ("cxl/hdm: Commit decoder state to hardware")
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Link: https://lore.kernel.org/r/20220818164210.2084-1-Jonathan.Cameron@huawei.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
A bug in the LSA code resulted in transfers slightly larger
than the mailbox size. Let us make it easier to catch similar
issues in future by adding a low level check.
Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220815154044.24733-2-Jonathan.Cameron@huawei.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Vishal notes that when attempting to define a second pmem region on a
device the DPA allocation fails with a message of the form:
decoder11.1: failed to reserve skipped space
Recall that the skip setting is used when there is a pmem allocation in
the presence of free ram DPA space. The first pmem allocation skips over
the free ram and subsequent pmem allocations do not require a skip. The
bug is that a skip is still attempted and the DPA reservation code
flags the double skip allocation conflict.
Fixes: cf880423b6 ("cxl/hdm: Add support for allocating DPA to an endpoint decoder")
Reported-by: Vishal Verma <vishal.l.verma@intel.com>
Tested-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Link: https://lore.kernel.org/r/165973754730.1558392.15466392461645857658.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The endpoint decode granularity must be <= the window granularity
otherwise capacity in the endpoints is lost in the decode. Consider an
attempt to have a region granularity of 512 with 4 devices within a
window that maps 2 host bridges at a granularity of 256 bytes:
HPA DPA Offset HB Port EP
0x0 0x0 0 0 0
0x100 0x0 1 0 2
0x200 0x100 0 0 0
0x300 0x100 1 0 2
0x400 0x200 0 1 1
0x500 0x200 1 1 3
0x600 0x300 0 1 1
0x700 0x300 1 1 3
0x800 0x400 0 0 0
0x900 0x400 1 0 2
0xA00 0x500 0 0 0
0xB00 0x500 1 0 2
Notice how endpoint0 maps HPA 0x0 and 0x200 correctly, but then at HPA
0x800 it results in DPA 0x200-0x400 on being skipped.
Fix this by restricing the region granularity to be equal to the window
granularity resulting in the following for a x4 region under a x2 window
at a granularity of 256.
HPA DPA Offset HB Port EP
0x0 0x0 0 0 0
0x100 0x0 1 0 2
0x200 0x0 0 1 1
0x300 0x0 1 1 3
0x400 0x100 0 0 0
0x500 0x100 1 0 2
0x600 0x100 0 1 1
0x700 0x100 1 1 3
Not that it ever made practical sense to support region granularity >
window granularity. The window rotates host bridges causing endpoints to
never see a consecutive stream of requests at the desired granularity
without breaks to issue cycles to the other host bridge.
Fixes: 80d10a6cee ("cxl/region: Add interleave geometry attributes")
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Link: https://lore.kernel.org/r/165973127171.1526540.9923273539049172976.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In cases where the decode fans out as it traverses downstream, the
interleave granularity needs to increment to identify the port selector
bits out of the remaining address bits. For example, recall that with an
x2 parent port intereleave (IW == 1), the downstream decode for children
of those ports will either see address bit IG+8 always set, or address
bit IG+8 always clear. So if the child port needs to select a downstream
port it can only use address bits starting at IG+9 (where IG and IW are
the CXL encoded values for interleave granularity (ilog2(ig) - 8) and
ways (ilog2(iw))).
When the parent port interleave is x1 no such masking occurs and the
child port can maintain the granularity that was routed to the parent
port.
Reported-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Link: https://lore.kernel.org/r/165973126583.1526540.657948655360009242.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
A recent bug fix added the setup of the endpoint decoder interleave
geometry settings to cxl_region_attach(). Move the HPA setup there as
well to keep all endpoint decoder parameter setting in a central
location.
For symmetry, move endpoint HPA teardown to cxl_region_detach(), and for
switches move HPA setup / teardown to cxl_port_{setup,reset}_targets().
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Link: https://lore.kernel.org/r/165973126020.1526540.14701949254436069807.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Jonathan notes:
"Curiously interleave ways = 1 for the EPs which is obviously wrong"
...while testing the latest CXL development branch on QEMU.
It turns out the region creation process failed to program the endpoint
decoders. This was missed because the default settings of x1 at 4K
intereleave still results in the region appearing to function. Jonathan
caught the bug by reverse mapping the translations that need to happen
for the QEMU support.
Link: https://lore.kernel.org/r/62e95fdf9f6e2_30440294e4@dwillia2-xfh.jf.intel.com.notmuch
Fixes: 384e624bb2 ("cxl/region: Attach endpoint decoders")
Reported-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165951146336.967013.11160153960900111443.stgit@dwillia2-xfh.jf.intel.com
Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Sphinx reported undescribed parameters in cxl_region_params struct:
./drivers/cxl/cxl.h:376: warning: Function parameter or member 'targets' not described in 'cxl_region_params'
./drivers/cxl/cxl.h:376: warning: Function parameter or member 'nr_targets' not described in 'cxl_region_params'
Describe these members.
Fixes: b9686e8c8e ("cxl/region: Enable the assignment of endpoint decoders to regions")
Signed-off-by: Bagas Sanjaya <bagasdotme@gmail.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220804075448.98241-3-bagasdotme@gmail.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Sphinx reported indentation warnings:
Documentation/driver-api/cxl/memory-devices:457: ./drivers/cxl/core/region.c:732: WARNING: Unexpected indentation.
Documentation/driver-api/cxl/memory-devices:457: ./drivers/cxl/core/region.c:733: WARNING: Block quote ends without a blank line; unexpected unindent.
Documentation/driver-api/cxl/memory-devices:457: ./drivers/cxl/core/region.c:735: WARNING: Unexpected indentation.
These warnings above are due to missing blank line padding in the nested list
in kernel-doc comment for cxl_rr_ep_add().
Add the paddings to fix the warnings.
Fixes: 384e624bb2 ("cxl/region: Attach endpoint decoders")
Signed-off-by: Bagas Sanjaya <bagasdotme@gmail.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220804075448.98241-2-bagasdotme@gmail.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The nvdimm_pmem_region_create() function returns NULL on error. It does
not return error pointers.
Fixes: 04ad63f086 ("cxl/region: Introduce cxl_pmem_region objects")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/Yuo65lq2WtfdGJ0X@kili
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Dan reports:
The error handling in cxl_port_attach_region() looks like it might
have a similar bug. The cxl_rr->nr_targets++; might want a --.
That function is more complicated.
Indeed cxl_rr->nr_targets leaks when cxl_rr_ep_add() fails, but that
flow is not clear. Fix the bug and the clarity by separating the 'new'
region-reference case from the 'extend' region-reference case. This also
moves the host-physical-address (HPA) validation, that the HPA of a new
region being accounted to the port is greater than the HPA of all other
regions associated with the port, to alloc_region_ref().
Introduce @nr_targets_inc to track when the error exit path needs to
clean up cxl_rr->nr_targets.
Fixes: 384e624bb2 ("cxl/region: Attach endpoint decoders")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Link: http://lore.kernel.org/r/165939482134.252363.1915691883146696327.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
0day robot reports:
drivers/cxl/core/region.c:196 cxl_region_decode_commit() error: uninitialized symbol 'rc'.
The re-checking of loop termination conditions to determine "success"
makes it hard to see that @rc is initialized in all cases. Remove those
to make it explicit that @rc reflects a commit error and that the rest
of logic is concerned with unwinding committed decoders.
This change potentially results in cxl_region_decode_reset() being
called with @count == 0 where it was not called before, but
cxl_region_decode_reset() treats that as a nop.
Fixes: 176baefb2e ("cxl/hdm: Commit decoder state to hardware")
Reported-by: kernel test robot <lkp@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: http://lore.kernel.org/r/165951148105.967013.14191992449932268431.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In preparation for a patch that validates that the region ways setting
is compatible with the granularity setting, the initial granularity
setting needs to start at zero to indicate "unset".
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Alison Schofield <alison.schofield@intel.com>
Link: https://lore.kernel.org/r/165853777484.2430596.3423921169034844397.stgit@dwillia2-xfh.jf.intel.com
[djbw: fix up unused variable]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
After adding support for emulating platform firmware established DPA
reservations, the cxl-topology.sh [1] unit test started crashing with
the following signature:
general protection fault, probably for non-canonical address 0x6b6b6b6b6b6b6bc3: 0000 [#1] PREEMPT SMP
[..]
RIP: 0010:to_cxl_port+0x8/0x60 [cxl_core]
[..]
Call Trace:
<TASK>
__cxl_dpa_release+0x1b/0xd0 [cxl_core]
cxl_dpa_release+0x1d/0x30 [cxl_core]
release_nodes+0x63/0x90
devres_release_all+0x88/0xc0
...i.e. a use after free of a 'struct cxl_endpoint_decoder' object. This
results from the ordering of init_hdm_decoder() before add_hdm_decoder()
where, at release time, the decoder is unregistered and released before
the DPA reservation.
Fix this by extending the life of the object until all DPA reservations
have been released which also preserves platform decoder settings being
settled by the time the decoder is published in sysfs (KOBJ_ADD time).
Note that the @len == 0 case in __cxl_dpa_reserve() is avoided in
practice as this function is only called for committed decoders and new
non-zero DPA allocations.
Link: https://github.com/pmem/ndctl/blob/pending/test/cxl-topology.sh [1]
Fixes: 9c57cde0dc ("cxl/hdm: Enumerate allocated DPA")
Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Link: https://lore.kernel.org/r/165896020625.3546860.12390103413706292760.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The kernel enforces that region granularity is >= to the top-level
interleave-granularity for the given CXL window. However, when the CXL
window interleave is x1, i.e. non-interleaved at the host bridge level,
then the specified granularity does not matter. Override the window
specified granularity to the CXL minimum so that any valid region
granularity is >= to the root granularity.
Reported-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Alison Schofield <alison.schofield@intel.com>
Link: https://lore.kernel.org/r/165853776917.2430596.16823264262010844458.stgit@dwillia2-xfh.jf.intel.com
[djbw: add CXL_DECODER_MIN_GRANULARITY per vishal]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
For switch and endpoint decoders the relationship of decoders to regions
is 1:1. However, for root decoders the relationship is 1:N. Also,
regions are already children of root decoders, so the 1:N relationship
is observed by walking the following glob:
/sys/bus/cxl/devices/$decoder/region*
Hide the vestigial 'region' attribute for root decoders.
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Reviewed-by: Alison Schofield <alison.schofield@intel.com>
Link: https://lore.kernel.org/r/165853776328.2430596.4647259305040072751.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In support of CXL unit tests in the ndctl project, arrange for the
cxl_acpi driver to load in response to the registration of cxl_test
devices.
Reported-by: Dave Jiang <dave.jiang@intel.com>
Tested-by: Dave Jiang <dave.jiang@intel.com>
Reviewed-by: Vishal Verma <vishal.l.verma@intel.com>
Link: https://lore.kernel.org/r/165853775783.2430596.13637998086505316619.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The ++ needs a match -- on the clean up path. If the p->nr_targets
value gets to be more than 16 it leads to uninitialized data in
cxl_port_setup_targets().
drivers/cxl/core/region.c:995 cxl_port_setup_targets() error: uninitialized symbol 'eiw'.
Fixes: 27b3f8d138 ("cxl/region: Program target lists")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/YuepCvUAoCtdpcoO@kili
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The "ways" variable comes from the user. The ways_to_cxl() function
has an upper bound but it doesn't check for negatives. Make
the "ways" variable an unsigned int to fix this bug.
Fixes: 80d10a6cee ("cxl/region: Add interleave geometry attributes")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/Yueo3NV2hFCXx1iV@kili
[djbw: fixup interleave_ways_store() to only accept unsigned input]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
This should check "p->res" instead of "res" (which is uninitialized).
Fixes: 23a22cd1c9 ("cxl/region: Allocate HPA capacity to regions")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/Yueor88I/DkVSOtL@kili
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The LIBNVDIMM subsystem is a platform agnostic representation of system
NVDIMM / persistent memory resources. To date, the CXL subsystem's
interaction with LIBNVDIMM has been to register an nvdimm-bridge device
and cxl_nvdimm objects to proxy CXL capabilities into existing LIBNVDIMM
subsystem mechanics.
With regions the approach is the same. Create a new cxl_pmem_region
object to proxy CXL region details into a LIBNVDIMM definition. With
this enabling LIBNVDIMM can partition CXL persistent memory regions with
legacy namespace labels. A follow-on patch will add CXL region label and
CXL namespace label support to persist region configurations across
driver reload / system-reset events.
Co-developed-by: Ben Widawsky <bwidawsk@kernel.org>
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784340111.1758207.3036498385188290968.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Be careful to only disable cxl_pmem objects related to a given
cxl_nvdimm_bridge. Otherwise, offline_nvdimm_bus() reaches across CXL
domains and disables more than is expected.
Fixes: 21083f5152 ("cxl/pmem: Register 'pmem' / cxl_nvdimm devices")
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784339569.1758207.1557084545278004577.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The CXL region driver is responsible for routing fully formed CXL
regions to one of libnvdimm, for persistent memory regions, device-dax
for volatile memory regions, or just act as an enumeration placeholder
if the region was setup and configuration locked by platform firmware.
In the platform-firmware-setup case the expectation is that region is
already accounted in the system memory map, i.e. already enabled as
"System RAM".
For now, just attach to CXL regions in the CXL_CONFIG_COMMIT state, and
take no further action.
Given this driver is just a small / simple router, include it in the
core rather than its own module.
Co-developed-by: Ben Widawsky <bwidawsk@kernel.org>
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220624041950.559155-18-dan.j.williams@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
After all the soft validation of the region has completed, convey the
region configuration to hardware while being careful to commit decoders
in specification mandated order. In addition to programming the endpoint
decoder base-address, interleave ways and granularity, the switch
decoder target lists are also established.
While the kernel can enforce spec-mandated commit order, it can not
enforce spec-mandated reset order. For example, the kernel can't stop
someone from removing an endpoint device that is occupying decoderN in a
switch decoder where decoderN+1 is also committed. To reset decoderN,
decoderN+1 must be torn down first. That "tear down the world"
implementation is saved for a follow-on patch.
Callback operations are provided for the 'commit' and 'reset'
operations. While those callbacks may prove useful for CXL accelerators
(Type-2 devices with memory) the primary motivation is to enable a
simple way for cxl_test to intercept those operations.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784338418.1758207.14659830845389904356.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Once the region's interleave geometry (ways, granularity, size) is
established and all the endpoint decoder targets are assigned, the next
phase is to program all the intermediate decoders. Specifically, each
CXL switch in the path between the endpoint and its CXL host-bridge
(including the logical switch internal to the host-bridge) needs to have
its decoders programmed and the target list order assigned.
The difficulty in this implementation lies in determining which endpoint
decoder ordering combinations are valid. Consider the cxl_test case of 2
host bridges, each of those host-bridges attached to 2 switches, and
each of those switches attached to 2 endpoints for a potential 8-way
interleave. The x2 interleave at the host-bridge level requires that all
even numbered endpoint decoder positions be located on the "left" hand
side of the topology tree, and the odd numbered positions on the other.
The endpoints that are peers on the same switch need to have a position
that can be routed with a dedicated address bit per-endpoint. See
check_last_peer() for the details.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784337827.1758207.132121746122685208.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
CXL regions (interleave sets) are made up of a set of memory devices
where each device maps a portion of the interleave with one of its
decoders (see CXL 2.0 8.2.5.12 CXL HDM Decoder Capability Structure).
As endpoint decoders are identified by a provisioning tool they can be
added to a region provided the region interleave properties are set
(way, granularity, HPA) and DPA has been assigned to the decoder.
The attach event triggers several validation checks, for example:
- is the DPA sized appropriately for the region
- is the decoder reachable via the host-bridges identified by the
region's root decoder
- is the device already active in a different region position slot
- are there already regions with a higher HPA active on a given port
(per CXL 2.0 8.2.5.12.20 Committing Decoder Programming)
...and the attach event affords an opportunity to collect data and
resources relevant to later programming the target lists in switch
decoders, for example:
- allocate a decoder at each cxl_port in the decode chain
- for a given switch port, how many the region's endpoints are hosted
through the port
- how many unique targets (next hops) does a port need to map to reach
those endpoints
The act of reconciling this information and deploying it to the decoder
configuration is saved for a follow-on patch.
Co-developed-by: Ben Widawsky <bwidawsk@kernel.org>
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784337277.1758207.4108508181328528703.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The ACPI CXL Fixed Memory Window Structure (CFMWS) defines multiple
methods to determine which host bridge provides access to a given
endpoint relative to that device's position in the interleave. The
"Interleave Arithmetic" defines either a "standard modulo" /
round-random algorithm, or "xormap" based algorithm which can be defined
as a non-linear transform. Given that there are already more options
beyond "standard modulo" and that "xormap" may turn out to be ACPI CXL
specific, provide a callback for the region provisioning code to map
endpoint positions back to expected host bridge id (cxl_dport target).
For now just support the simple modulo math case and save the xormap for
a follow-on change.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220624041950.559155-14-dan.j.williams@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The region provisioning process involves allocating DPA to a set of
endpoint decoders, and HPA plus the region geometry to a region device.
Then the decoder is assigned to the region. At this point several
validation steps can be performed to validate that the decoder is
suitable to participate in the region.
Co-developed-by: Ben Widawsky <bwidawsk@kernel.org>
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/r/165784336184.1758207.16403282029203949622.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
After a region's interleave parameters (ways and granularity) are set,
add a way for regions to allocate HPA (host physical address space) from
the free capacity in their parent root-decoder. The allocator for this
capacity reuses the 'struct resource' based allocator used for
CONFIG_DEVICE_PRIVATE.
Once the tuple of "ways, granularity, [uuid], and size" is set the
region configuration transitions to the CXL_CONFIG_INTERLEAVE_ACTIVE
state which is a precursor to allowing endpoint decoders to be added to
a region.
Co-developed-by: Ben Widawsky <bwidawsk@kernel.org>
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784335630.1758207.420216490941955417.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Add ABI to allow the number of devices that comprise a region to be
set as well as the interleave granularity for the region.
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
[djbw: reword changelog]
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220624041950.559155-11-dan.j.williams@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The process of provisioning a region involves triggering the creation of
a new region object, pouring in the configuration, and then binding that
configured object to the region driver to start its operation. For
persistent memory regions the CXL specification mandates that it
identified by a uuid. Add an ABI for userspace to specify a region's
uuid.
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
[djbw: simplify locking]
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784334465.1758207.8224025435884752570.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
CXL 2.0 allows for dynamic provisioning of new memory regions (system
physical address resources like "System RAM" and "Persistent Memory").
Whereas DDR and PMEM resources are conveyed statically at boot, CXL
allows for assembling and instantiating new regions from the available
capacity of CXL memory expanders in the system.
Sysfs with an "echo $region_name > $create_region_attribute" interface
is chosen as the mechanism to initiate the provisioning process. This
was chosen over ioctl() and netlink() to keep the configuration
interface entirely in a pseudo-fs interface, and it was chosen over
configfs since, aside from this one creation event, the interface is
read-mostly. I.e. configfs supports cases where an object is designed to
be provisioned each boot, like an iSCSI storage target, and CXL region
creation is mostly for PMEM regions which are created usually once
per-lifetime of a server instance. This is an improvement over nvdimm
that pre-created "seed" devices that tended to confuse users looking to
determine which devices are active and which are idle.
Recall that the major change that CXL brings over previous persistent
memory architectures is the ability to dynamically define new regions.
Compare that to drivers like 'nfit' where the region configuration is
statically defined by platform firmware.
Regions are created as a child of a root decoder that encompasses an
address space with constraints. When created through sysfs, the root
decoder is explicit. When created from an LSA's region structure a root
decoder will possibly need to be inferred by the driver.
Upon region creation through sysfs, a vacant region is created with a
unique name. Regions have a number of attributes that must be configured
before the region can be bound to the driver where HDM decoder program
is completed.
An example of creating a new region:
- Allocate a new region name:
region=$(cat /sys/bus/cxl/devices/decoder0.0/create_pmem_region)
- Create a new region by name:
while
region=$(cat /sys/bus/cxl/devices/decoder0.0/create_pmem_region)
! echo $region > /sys/bus/cxl/devices/decoder0.0/create_pmem_region
do true; done
- Region now exists in sysfs:
stat -t /sys/bus/cxl/devices/decoder0.0/$region
- Delete the region, and name:
echo $region > /sys/bus/cxl/devices/decoder0.0/delete_region
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784333909.1758207.794374602146306032.stgit@dwillia2-xfh.jf.intel.com
[djbw: simplify locking, reword changelog]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The port scanning algorithm in devm_cxl_enumerate_ports() walks up the
topology and adds cxl_port objects starting from the root down to the
endpoint. When those ports are initially created they know all their
dports, but they do not know the downstream cxl_port instance that
represents the next descendant in the topology. Rework create_endpoint()
into devm_cxl_add_endpoint() that enumerates the downstream cxl_port
topology into each port's 'struct cxl_ep' record for each endpoint it
that the port is an ancestor.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220624041950.559155-7-dan.j.williams@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The region provisioning flow involves selecting interleave ways +
granularity settings for a region, and then programming the decoder
topology to meet those constraints, if possible. For example, root
decoders set the minimum interleave ways + granularity for any hosted
regions.
Given decoder programming is not atomic and collisions can occur between
multiple requesting regions userspace will be responsible for conflict
resolution and it needs these attributes to make those decisions.
Signed-off-by: Ben Widawsky <bwidawsk@kernel.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784332235.1758207.7185062713652694607.stgit@dwillia2-xfh.jf.intel.com
[djbw: reword changelog, make read-only, add sysfs ABI documentaion]
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Reduce the complexity and the overhead of walking the topology to
determine endpoint connectivity to root decoder interleave
configurations.
Note that cxl_detach_ep(), after it determines that the last @ep has
departed and decides to delete the port, now needs to walk the dport
array with the device_lock() held to remove entries. Previously
list_splice_init() could be used atomically delete all dport entries at
once and then perform entry tear down outside the lock. There is no
list_splice_init() equivalent for the xarray.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784331647.1758207.6345820282285119339.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
In preparation for region provisioning that needs to walk the topology
by endpoints, use an xarray to record endpoint interest in a given port.
In addition to being more space and time efficient it also reduces the
complexity of the implementation by moving locking internal to the
xarray implementation. It also allows for a single cxl_ep reference to
be recorded in multiple xarrays.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220624041950.559155-2-dan.j.williams@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
At the time that cxl_port instances are being created, cache the dport
from the parent port that points to this new child port. This will be
useful for region provisioning when walking the tree to calculate
decoder targets, and saves rewalking the dport list after the fact to
build this information.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/20220624041950.559155-1-dan.j.williams@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Recall that the primary role of the cxl_mem driver is to probe if the
given endpoint is connected to a CXL port topology. In that process it
walks its device ancestry to its PCI root port. If that root port is
also a CXL root port then the probe process adds cxl_port object
instances at switch in the path between to the root and the endpoint. As
those cxl_port instances are added, or if a previous enumeration
attempt already created the port, a 'struct cxl_ep' instance is
registered with that port to track the endpoints interested in that
port.
At the time the cxl_ep is registered the downstream egress path from the
port to the endpoint is known. Take the opportunity to record that
information as it will be needed for dynamic programming of decoder
targets during region provisioning.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784329944.1758207.15203961796832072116.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The region provisioning flow will roughly follow a sequence of:
1/ Allocate DPA to a set of decoders
2/ Allocate HPA to a region
3/ Associate decoders with a region and validate that the DPA allocations
and topologies match the parameters of the region.
For now, this change (step 1) arranges for DPA capacity to be allocated
and deleted from non-committed decoders based on the decoder's mode /
partition selection. Capacity is allocated from the lowest DPA in the
partition and any 'pmem' allocation blocks out all remaining ram
capacity in its 'skip' setting. DPA allocations are enforced in decoder
instance order. I.e. decoder N + 1 always starts at a higher DPA than
instance N, and deleting allocations must proceed from the
highest-instance allocated decoder to the lowest.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784329399.1758207.16732038126938632700.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
The CXL specification enforces that endpoint decoders are committed in
hw instance id order. In preparation for adding dynamic DPA allocation,
record the hw instance id in endpoint decoders, and enforce allocations
to occur in hw instance id order.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165784328827.1758207.9627538529944559954.stgit@dwillia2-xfh.jf.intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Recall that the Device Physical Address (DPA) space of a CXL Memory
Expander is potentially partitioned into a volatile and persistent
portion. A decoder maps a Host Physical Address (HPA) range to a DPA
range and that translation depends on the value of all previous (lower
instance number) decoders before the current one.
In preparation for allowing dynamic provisioning of regions, decoders
need an ABI to indicate which DPA partition a decoder targets. This ABI
needs to be prepared for the possibility that some other agent committed
and locked a decoder that spans the partition boundary.
Add 'decoderX.Y/mode' to endpoint decoders that indicates which
partition 'ram' / 'pmem' the decoder targets, or 'mixed' if the decoder
currently spans the partition boundary.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Link: https://lore.kernel.org/r/165603881967.551046.6007594190951596439.stgit@dwillia2-xfh
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Dan Williams
Yu Zhe
Vishal Verma
Jonathan Cameron
Bagas Sanjaya
Dan Carpenter
Ben Widawsky