Jonathan Cameron wrote:
> On Fri, 16 Aug 2024 09:44:33 -0500
> Ira Weiny wrote:
>
> > cxl_test provides a good way to ensure quick smoke and regression
> > testing. The complexity of Dynamic Capacity (DC) extent processing as
> > well as the complexity of the new sparse DAX regions can mostly be
Jonathan Cameron wrote:
> On Fri, 16 Aug 2024 09:44:32 -0500
> Ira Weiny wrote:
>
> > The test event logs were created as static arrays as an easy way to mock
> > events. Dynamic Capacity Device (DCD) test support requires events be
> > generated dynamically when extents are created or destroyed
On Fri, Aug 16, 2024 at 09:44:12AM -0500, Ira Weiny wrote:
> The event buffer does not need to be allocated if something has failed in
> setting up event irq's.
>
> In prep for adjusting event configuration for DCD events move the buffer
> allocation to the end of the event configuration.
>
> Rev
On Fri, Aug 16, 2024 at 09:44:31AM -0500, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> CXL rev 3.1 section 8.2.9.2.1 adds the Dynamic Capacity Event Records.
> User space can use trace events for debugging of DC capacity changes.
>
> Add DC trace points to the trace log.
>
> Signed-off-
On Fri, Aug 16, 2024 at 09:44:30AM -0500, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic capacity device extents may be left in an accepted state on a
> device due to an unexpected host crash. In this case it is expected
> that the creation of a new region on top of a DC partition
On Fri, Aug 16, 2024 at 09:44:26AM -0500, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> A dynamic capacity device (DCD) sends events to signal the host for
> changes in the availability of Dynamic Capacity (DC) memory. These
> events contain extents describing a DPA range and meta data fo
On 8/16/2024 10:44 PM, Ira Weiny wrote:
> cxl_dpa_to_region() finds the region from a tuple.
> The search involves finding the device endpoint decoder as well.
>
> Dynamic capacity extent processing uses the endpoint decoder HPA
> information to calculate the HPA offset. In addition, well behaved
On 8/16/2024 10:44 PM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic Capacity Devices (DCD) support extent change notifications
> through the event log mechanism. The interrupt mailbox commands were
> extended in CXL 3.1 to support these notifications. Firmware can't
> configure D
On 8/16/2024 10:44 PM, Ira Weiny wrote:
> Dynamic Capacity Devices (DCD) require event interrupts to process
> memory addition or removal. BIOS may have control over non-DCD event
> processing. DCD interrupt configuration needs to be separate from
> memory event interrupt configuration.
>
> Facto
On 8/16/2024 10:44 PM, Ira Weiny wrote:
> Dynamic Capacity Devices (DCD) require event interrupts to process
> memory addition or removal. BIOS may have control over non-DCD event
> processing. DCD interrupt configuration needs to be separate from
> memory event interrupt configuration.
>
> Split
On 8/16/2024 10:44 PM, Ira Weiny wrote:
> create_pmem_region_store() and create_ram_region_store() are identical
> with the exception of the region mode. With the addition of DC region
> mode this would end up being 3 copies of the same code.
>
> Refactor create_pmem_region_store() and create_ram_
On 8/16/2024 10:44 PM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> One or more decoders each pointing to a Dynamic Capacity (DC) partition
> form a CXL software region. The region mode reflects composition of
> that entire software region. Decoder mode reflects a specific DC
> partition
On 8/16/2024 10:44 PM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Until now region modes and decoder modes were equivalent in that both
> modes were either PMEM or RAM. The addition of Dynamic
> Capacity partitions defines up to 8 DC partitions per device.
>
> The region mode is thus no
On 8/16/2024 10:44 PM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Per the CXL 3.1 specification software must check the Command Effects
> Log (CEL) for dynamic capacity command support.
>
> Detect support for the DCD commands while reading the CEL, including:
>
> Get DC Config
>
On 8/16/2024 10:44 PM, Ira Weiny wrote:
> The event buffer does not need to be allocated if something has failed in
> setting up event irq's.
>
> In prep for adjusting event configuration for DCD events move the buffer
> allocation to the end of the event configuration.
>
> Reviewed-by: Davidlohr B
On 8/16/2024 10:44 PM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> A dynamic capacity device (DCD) sends events to signal the host for
> changes in the availability of Dynamic Capacity (DC) memory. These
> events contain extents describing a DPA range and meta data for memory
> to be add
On 8/16/2024 10:44 PM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic Capacity CXL regions must allow memory to be added or removed
> dynamically. In addition to the quantity of memory available the
> location of the memory within a DC partition is dynamic based on the
> extents off
> > > +int cxl_rm_extent(struct cxl_memdev_state *mds, struct cxl_extent
> > > *extent)
> > > +{
> > > + u64 start_dpa = le64_to_cpu(extent->start_dpa);
> > > + struct cxl_memdev *cxlmd = mds->cxlds.cxlmd;
> > > + struct cxl_endpoint_decoder *cxled;
> > > + struct range hpa_range, dpa_range;
> > >
Jonathan Cameron wrote:
> On Fri, 16 Aug 2024 09:44:29 -0500
> ira.we...@intel.com wrote:
>
> > From: Navneet Singh
> >
[snip]
> > diff --git a/drivers/cxl/core/extent.c b/drivers/cxl/core/extent.c
> > index d7d526a51e2b..103b0bec3a4a 100644
> > --- a/drivers/cxl/core/extent.c
> > +++ b/driver
Jonathan Cameron wrote:
> On Fri, 16 Aug 2024 09:44:28 -0500
> Ira Weiny wrote:
>
> > Dynamic Capacity regions must limit dev dax resources to those areas
> > which have extents backing real memory. Such DAX regions are dubbed
> > 'sparse' regions. In order to manage where memory is available f
Jonathan Cameron wrote:
> On Fri, 16 Aug 2024 09:44:26 -0500
> ira.we...@intel.com wrote:
>
> > From: Navneet Singh
> >
[snip]
> > +static int match_contains(struct device *dev, void *data)
> > +{
> > + struct region_extent *region_extent = to_region_extent(dev);
> > + struct match_data *m
On Fri, Aug 16, 2024 at 09:44:27AM -0500, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Extent information can be helpful to the user to coordinate memory usage
> with the external orchestrator and FM.
>
> Expose the details of region extents by creating the following
> sysfs entries.
>
On Tue, Aug 27, 2024 at 11:12:47PM -0500, Ira Weiny wrote:
> Andy Shevchenko wrote:
> > On Mon, Aug 26, 2024 at 04:17:52PM -0500, Ira Weiny wrote:
> > > Andy Shevchenko wrote:
> > > > On Mon, Aug 26, 2024 at 03:23:50PM +0200, Petr Mladek wrote:
> > > > > On Thu 2024-08-22 21:10:25, Andy Shevchenko
Andy Shevchenko wrote:
> On Mon, Aug 26, 2024 at 04:17:52PM -0500, Ira Weiny wrote:
> > Andy Shevchenko wrote:
> > > On Mon, Aug 26, 2024 at 03:23:50PM +0200, Petr Mladek wrote:
> > > > On Thu 2024-08-22 21:10:25, Andy Shevchenko wrote:
> > > > > On Thu, Aug 22, 2024 at 12:53:32PM -0500, Ira Weiny
Petr Mladek wrote:
> On Mon 2024-08-26 16:17:52, Ira Weiny wrote:
> > Andy Shevchenko wrote:
> > > On Mon, Aug 26, 2024 at 03:23:50PM +0200, Petr Mladek wrote:
> > > > On Thu 2024-08-22 21:10:25, Andy Shevchenko wrote:
> > > > > On Thu, Aug 22, 2024 at 12:53:32PM -0500, Ira Weiny wrote:
> > > > > >
On Tue, Aug 27, 2024 at 01:08:29PM +0100, Jonathan Cameron wrote:
> On Fri, 23 Aug 2024 14:32:32 -0700
> Fan Ni wrote:
>
> > On Fri, Aug 16, 2024 at 09:44:26AM -0500, ira.we...@intel.com wrote:
> > > From: Navneet Singh
> > >
> > > A dynamic capacity device (DCD) sends events to signal the host
On Fri, 16 Aug 2024 09:44:33 -0500
Ira Weiny wrote:
> cxl_test provides a good way to ensure quick smoke and regression
> testing. The complexity of Dynamic Capacity (DC) extent processing as
> well as the complexity of the new sparse DAX regions can mostly be
> tested through cxl_test. This in
On Fri, 16 Aug 2024 09:44:32 -0500
Ira Weiny wrote:
> The test event logs were created as static arrays as an easy way to mock
> events. Dynamic Capacity Device (DCD) test support requires events be
> generated dynamically when extents are created or destroyed.
>
> Modify the event log storage
On Fri, 16 Aug 2024 09:44:31 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> CXL rev 3.1 section 8.2.9.2.1 adds the Dynamic Capacity Event Records.
> User space can use trace events for debugging of DC capacity changes.
>
> Add DC trace points to the trace log.
>
> Signed-off-by: Na
On Fri, 16 Aug 2024 09:44:30 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic capacity device extents may be left in an accepted state on a
> device due to an unexpected host crash. In this case it is expected
> that the creation of a new region on top of a DC partition can re
On Fri, 16 Aug 2024 09:44:29 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> DAX regions which map dynamic capacity partitions require that memory be
> allowed to come and go. Recall sparse regions were created for this
> purpose. Now that extents can be realized within DAX regions
On Fri, 16 Aug 2024 09:44:28 -0500
Ira Weiny wrote:
> Dynamic Capacity regions must limit dev dax resources to those areas
> which have extents backing real memory. Such DAX regions are dubbed
> 'sparse' regions. In order to manage where memory is available four
> alternatives were considered:
On Tue, Aug 27, 2024 at 09:43:32AM +0200, Petr Mladek wrote:
> On Mon 2024-08-26 16:17:52, Ira Weiny wrote:
> > Andy Shevchenko wrote:
...
> But I could live with all other variants, except for %pn mentioned below.
I believe %r is also no go as we most likely get a complier warning.
...
> > Am
On Fri, 16 Aug 2024 09:44:26 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> A dynamic capacity device (DCD) sends events to signal the host for
> changes in the availability of Dynamic Capacity (DC) memory. These
> events contain extents describing a DPA range and meta data for memo
On Mon, Aug 26, 2024 at 04:17:52PM -0500, Ira Weiny wrote:
> Andy Shevchenko wrote:
> > On Mon, Aug 26, 2024 at 03:23:50PM +0200, Petr Mladek wrote:
> > > On Thu 2024-08-22 21:10:25, Andy Shevchenko wrote:
> > > > On Thu, Aug 22, 2024 at 12:53:32PM -0500, Ira Weiny wrote:
> > > > > Petr Mladek wrot
On Fri, 23 Aug 2024 14:32:32 -0700
Fan Ni wrote:
> On Fri, Aug 16, 2024 at 09:44:26AM -0500, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
> > A dynamic capacity device (DCD) sends events to signal the host for
> > changes in the availability of Dynamic Capacity (DC) memory. These
> >
On Mon 2024-08-26 16:17:52, Ira Weiny wrote:
> Andy Shevchenko wrote:
> > On Mon, Aug 26, 2024 at 03:23:50PM +0200, Petr Mladek wrote:
> > > On Thu 2024-08-22 21:10:25, Andy Shevchenko wrote:
> > > > On Thu, Aug 22, 2024 at 12:53:32PM -0500, Ira Weiny wrote:
> > > > > Petr Mladek wrote:
> > > > > >
Andy Shevchenko wrote:
> On Mon, Aug 26, 2024 at 03:23:50PM +0200, Petr Mladek wrote:
> > On Thu 2024-08-22 21:10:25, Andy Shevchenko wrote:
> > > On Thu, Aug 22, 2024 at 12:53:32PM -0500, Ira Weiny wrote:
> > > > Petr Mladek wrote:
> > > > > On Fri 2024-08-16 09:44:10, Ira Weiny wrote:
>
> ...
>
Dave Jiang wrote:
>
>
> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
> > CXL rev 3.1 section 8.2.9.2.1 adds the Dynamic Capacity Event Records.
> > User space can use trace events for debugging of DC capacity changes.
> >
> > Add DC trace points to the trace log.
On Mon, Aug 26, 2024 at 03:23:50PM +0200, Petr Mladek wrote:
> On Thu 2024-08-22 21:10:25, Andy Shevchenko wrote:
> > On Thu, Aug 22, 2024 at 12:53:32PM -0500, Ira Weiny wrote:
> > > Petr Mladek wrote:
> > > > On Fri 2024-08-16 09:44:10, Ira Weiny wrote:
...
> > > > > + %par[range 0x6
On Mon, Aug 26, 2024 at 03:17:26PM +0200, Petr Mladek wrote:
> On Thu 2024-08-22 12:53:32, Ira Weiny wrote:
> > Petr Mladek wrote:
> > > On Fri 2024-08-16 09:44:10, Ira Weiny wrote:
[...]
> > > > + static const struct printf_spec range_spec = {
> > > > + .base = 16,
> > > > +
On Thu 2024-08-22 21:10:25, Andy Shevchenko wrote:
> On Thu, Aug 22, 2024 at 12:53:32PM -0500, Ira Weiny wrote:
> > Petr Mladek wrote:
> > > On Fri 2024-08-16 09:44:10, Ira Weiny wrote:
>
> ...
>
> > > > + %par[range 0x6000-0x6fff] or
> > >
> > > It seems that it is always 64-b
On Thu 2024-08-22 12:53:32, Ira Weiny wrote:
> Petr Mladek wrote:
> > On Fri 2024-08-16 09:44:10, Ira Weiny wrote:
> > > The use of struct range in the CXL subsystem is growing. In particular,
> > > the addition of Dynamic Capacity devices uses struct range in a number
> > > of places which are re
On Fri, Aug 16, 2024 at 09:44:26AM -0500, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> A dynamic capacity device (DCD) sends events to signal the host for
> changes in the availability of Dynamic Capacity (DC) memory. These
> events contain extents describing a DPA range and meta data fo
Dave Jiang wrote:
>
>
> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
> > Dynamic capacity device extents may be left in an accepted state on a
> > device due to an unexpected host crash. In this case it is expected
> > that the creation of a new region on top of a
On Fri, 16 Aug 2024 09:44:27 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Extent information can be helpful to the user to coordinate memory usage
> with the external orchestrator and FM.
>
> Expose the details of region extents by creating the following
> sysfs entries.
>
>
On Thu, 22 Aug 2024 21:58:02 -0500
Ira Weiny wrote:
> Dave Jiang wrote:
> >
> >
> > On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > > From: Navneet Singh
> > >
> > > Extent information can be helpful to the user to coordinate memory usage
> > > with the external orchestrator and FM.
> >
On Fri, 16 Aug 2024 09:44:25 -0500
Ira Weiny wrote:
> cxl_dpa_to_region() finds the region from a tuple.
> The search involves finding the device endpoint decoder as well.
>
> Dynamic capacity extent processing uses the endpoint decoder HPA
> information to calculate the HPA offset. In additio
On Fri, 16 Aug 2024 09:44:24 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic Capacity Devices (DCD) support extent change notifications
> through the event log mechanism. The interrupt mailbox commands were
> extended in CXL 3.1 to support these notifications. Firmware can't
On Fri, 16 Aug 2024 09:44:22 -0500
Ira Weiny wrote:
> Dynamic Capacity Devices (DCD) require event interrupts to process
> memory addition or removal. BIOS may have control over non-DCD event
> processing. DCD interrupt configuration needs to be separate from
> memory event interrupt configurat
On Fri, 16 Aug 2024 09:44:21 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic Capacity CXL regions must allow memory to be added or removed
> dynamically. In addition to the quantity of memory available the
> location of the memory within a DC partition is dynamic based on the
On Fri, 16 Aug 2024 09:44:20 -0500
Ira Weiny wrote:
> create_pmem_region_store() and create_ram_region_store() are identical
> with the exception of the region mode. With the addition of DC region
> mode this would end up being 3 copies of the same code.
>
> Refactor create_pmem_region_store()
On Thu, 22 Aug 2024 21:28:41 -0500
Ira Weiny wrote:
> Dave Jiang wrote:
> >
> >
> > On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > > From: Navneet Singh
> > >
> > > To properly configure CXL regions on Dynamic Capacity Devices (DCD),
> > > user space will need to know the details of the
On Fri, 16 Aug 2024 09:44:18 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Endpoint decoder mode is used to represent the partition the decoder
> points to such as ram or pmem.
>
> Expand the mode to allow a decoder to point to a specific DC partition
> (Region).
>
> Signed-off-by
On Fri, 16 Aug 2024 09:44:17 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> To support Dynamic Capacity Devices (DCD) endpoint decoders will need to
> map DC partitions (regions). In addition to assigning the size of the
> DC partition, the decoder must assign any skip value from th
On Fri, 16 Aug 2024 09:44:15 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Until now region modes and decoder modes were equivalent in that both
> modes were either PMEM or RAM. The addition of Dynamic
> Capacity partitions defines up to 8 DC partitions per device.
>
> The region
On Fri, 16 Aug 2024 09:44:14 -0500
ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Devices which optionally support Dynamic Capacity (DC) are configured
> via mailbox commands. CXL 3.1 requires the host to issue the Get DC
> Configuration command in order to properly configure DCDs. Witho
On Fri, 16 Aug 2024 09:44:11 -0500
Ira Weiny wrote:
> The device DAX structure is being enhanced to track additional DCD
> information.
>
> The current range tuple was not fully documented. Document it prior to
> adding information for DC.
>
> Suggested-by: Jonathan Cameron
> Signed-off-by: I
On 8/22/24 7:28 PM, Ira Weiny wrote:
> Dave Jiang wrote:
>>
>>
>> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
>>> From: Navneet Singh
>>>
>>> To properly configure CXL regions on Dynamic Capacity Devices (DCD),
>>> user space will need to know the details of the DC partitions available.
>>>
Dave Jiang wrote:
>
>
> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
[snip]
> > diff --git a/drivers/cxl/core/extent.c b/drivers/cxl/core/extent.c
> > index d7d526a51e2b..103b0bec3a4a 100644
> > --- a/drivers/cxl/core/extent.c
> > +++ b/drivers/cxl/core/extent.c
Dave Jiang wrote:
>
>
> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
> > Extent information can be helpful to the user to coordinate memory usage
> > with the external orchestrator and FM.
> >
> > Expose the details of region extents by creating the following
> >
Dave Jiang wrote:
>
>
> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
[snip]
> >
> > Process DCD events and create region devices.
> >
> > Signed-off-by: Navneet Singh
> > Co-developed-by: Ira Weiny
> > Signed-off-by: Ira Weiny
>
> A few nits below, but in g
Dave Jiang wrote:
>
>
> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
> > To properly configure CXL regions on Dynamic Capacity Devices (DCD),
> > user space will need to know the details of the DC partitions available.
> >
> > Expose dynamic capacity capabilities
Dave Jiang wrote:
>
>
> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
[snip]
> > +static int dc_mode_to_region_index(enum cxl_decoder_mode mode)
> > +{
> > + return mode - CXL_DECODER_DC0;
> > +}
> > +
> > +static int cxl_request_skip(struct cxl_endpoint_decoder
Fan Ni wrote:
> On Fri, Aug 16, 2024 at 02:45:47PM -0700, Dave Jiang wrote:
[snip]
> > > +
> > > +/**
> > > + * cxl_dev_dynamic_capacity_identify() - Reads the dynamic capacity
> > > + *information from the device.
> > > + * @mds: The memory device state
>
Fan Ni wrote:
> On Fri, Aug 16, 2024 at 02:45:47PM -0700, Dave Jiang wrote:
> >
> > > +
> > > +/**
> > > + * cxl_dev_dynamic_capacity_identify() - Reads the dynamic capacity
> > > + *information from the device.
> > > + * @mds: The memory device state
> > >
On Fri, Aug 16, 2024 at 09:44:23AM -0500, Ira Weiny wrote:
> Dynamic Capacity Devices (DCD) require event interrupts to process
> memory addition or removal. BIOS may have control over non-DCD event
> processing. DCD interrupt configuration needs to be separate from
> memory event interrupt confi
On Fri, Aug 16, 2024 at 09:44:22AM -0500, Ira Weiny wrote:
> Dynamic Capacity Devices (DCD) require event interrupts to process
> memory addition or removal. BIOS may have control over non-DCD event
> processing. DCD interrupt configuration needs to be separate from
> memory event interrupt confi
On Fri, Aug 16, 2024 at 09:44:20AM -0500, Ira Weiny wrote:
> create_pmem_region_store() and create_ram_region_store() are identical
> with the exception of the region mode. With the addition of DC region
> mode this would end up being 3 copies of the same code.
>
> Refactor create_pmem_region_sto
On Fri, Aug 16, 2024 at 09:44:21AM -0500, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic Capacity CXL regions must allow memory to be added or removed
> dynamically. In addition to the quantity of memory available the
> location of the memory within a DC partition is dynamic based
On Thu, Aug 22, 2024 at 12:53:32PM -0500, Ira Weiny wrote:
> Petr Mladek wrote:
> > On Fri 2024-08-16 09:44:10, Ira Weiny wrote:
...
> > > + %par[range 0x6000-0x6fff] or
> >
> > It seems that it is always 64-bit. It prints:
> >
> > struct range {
> > u64 start;
> > u64 e
Petr Mladek wrote:
> On Fri 2024-08-16 09:44:10, Ira Weiny wrote:
> > The use of struct range in the CXL subsystem is growing. In particular,
> > the addition of Dynamic Capacity devices uses struct range in a number
> > of places which are reported in debug and error messages.
> >
> > To wit req
On 8/16/24 7:44 AM, Ira Weiny wrote:
> The test event logs were created as static arrays as an easy way to mock
> events. Dynamic Capacity Device (DCD) test support requires events be
> generated dynamically when extents are created or destroyed.
>
> Modify the event log storage to be dynamica
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> CXL rev 3.1 section 8.2.9.2.1 adds the Dynamic Capacity Event Records.
> User space can use trace events for debugging of DC capacity changes.
>
> Add DC trace points to the trace log.
>
> Signed-off-by: Navneet Singh
On Fri, Aug 16, 2024 at 02:45:47PM -0700, Dave Jiang wrote:
>
>
> On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> > From: Navneet Singh
> >
> > Devices which optionally support Dynamic Capacity (DC) are configured
> > via mailbox commands. CXL 3.1 requires the host to issue the Get DC
> > Con
On Fri 2024-08-16 09:44:10, Ira Weiny wrote:
> The use of struct range in the CXL subsystem is growing. In particular,
> the addition of Dynamic Capacity devices uses struct range in a number
> of places which are reported in debug and error messages.
>
> To wit requiring the printing of the star
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic capacity device extents may be left in an accepted state on a
> device due to an unexpected host crash. In this case it is expected
> that the creation of a new region on top of a DC partition can read
> those ex
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> DAX regions which map dynamic capacity partitions require that memory be
> allowed to come and go. Recall sparse regions were created for this
> purpose. Now that extents can be realized within DAX regions the DAX
> reg
On 8/16/24 7:44 AM, Ira Weiny wrote:
> Dynamic Capacity regions must limit dev dax resources to those areas
> which have extents backing real memory. Such DAX regions are dubbed
> 'sparse' regions. In order to manage where memory is available four
> alternatives were considered:
>
> 1) Create
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Extent information can be helpful to the user to coordinate memory usage
> with the external orchestrator and FM.
>
> Expose the details of region extents by creating the following
> sysfs entries.
>
> /sys/bus/
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> A dynamic capacity device (DCD) sends events to signal the host for
> changes in the availability of Dynamic Capacity (DC) memory. These
> events contain extents describing a DPA range and meta data for memory
> to be ad
On 8/16/24 7:44 AM, Ira Weiny wrote:
> cxl_dpa_to_region() finds the region from a tuple.
> The search involves finding the device endpoint decoder as well.
>
> Dynamic capacity extent processing uses the endpoint decoder HPA
> information to calculate the HPA offset. In addition, well behave
…
> +++ b/drivers/cxl/core/extent.c
> @@ -271,20 +271,67 @@ static void calc_hpa_range(struct cxl_endpoint_decoder
> *cxled,
> hpa_range->end = hpa_range->start + range_len(dpa_range) - 1;
> }
>
> +static int cxlr_notify_extent(struct cxl_region *cxlr, enum dc_event event,
> +
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic Capacity Devices (DCD) support extent change notifications
> through the event log mechanism. The interrupt mailbox commands were
> extended in CXL 3.1 to support these notifications. Firmware can't
> configure
On 8/16/24 7:44 AM, Ira Weiny wrote:
> Dynamic Capacity Devices (DCD) require event interrupts to process
> memory addition or removal. BIOS may have control over non-DCD event
> processing. DCD interrupt configuration needs to be separate from
> memory event interrupt configuration.
>
> Spli
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Dynamic Capacity CXL regions must allow memory to be added or removed
> dynamically. In addition to the quantity of memory available the
> location of the memory within a DC partition is dynamic based on the
> extents of
On 8/16/24 7:44 AM, Ira Weiny wrote:
> create_pmem_region_store() and create_ram_region_store() are identical
> with the exception of the region mode. With the addition of DC region
> mode this would end up being 3 copies of the same code.
>
> Refactor create_pmem_region_store() and create_ram
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> To properly configure CXL regions on Dynamic Capacity Devices (DCD),
> user space will need to know the details of the DC partitions available.
>
> Expose dynamic capacity capabilities through sysfs.
>
> Signed-off-by:
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Endpoint decoder mode is used to represent the partition the decoder
> points to such as ram or pmem.
>
> Expand the mode to allow a decoder to point to a specific DC partition
> (Region).
>
> Signed-off-by: Navneet Sin
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> To support Dynamic Capacity Devices (DCD) endpoint decoders will need to
> map DC partitions (regions). In addition to assigning the size of the
> DC partition, the decoder must assign any skip value from the previous
>
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> One or more decoders each pointing to a Dynamic Capacity (DC) partition
> form a CXL software region. The region mode reflects composition of
> that entire software region. Decoder mode reflects a specific DC
> partitio
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Until now region modes and decoder modes were equivalent in that both
> modes were either PMEM or RAM. The addition of Dynamic
> Capacity partitions defines up to 8 DC partitions per device.
>
> The region mode is thus
On 8/16/24 7:44 AM, ira.we...@intel.com wrote:
> From: Navneet Singh
>
> Devices which optionally support Dynamic Capacity (DC) are configured
> via mailbox commands. CXL 3.1 requires the host to issue the Get DC
> Configuration command in order to properly configure DCDs. Without the
> Get
On 8/16/24 7:44 AM, Ira Weiny wrote:
> The device DAX structure is being enhanced to track additional DCD
> information.
>
> The current range tuple was not fully documented. Document it prior to
> adding information for DC.
>
> Suggested-by: Jonathan Cameron
> Signed-off-by: Ira Weiny
Rev
On Wed, Aug 14, 2024 at 09:01:42PM +0200, Thorsten Blum wrote:
> On 14. Aug 2024, at 20:02, Nathan Chancellor wrote:
> > On Mon, Aug 12, 2024 at 02:03:44PM +0200, Thorsten Blum wrote:
> >> On 12. Aug 2024, at 12:54, Johannes Thumshirn
> >> wrote:
> >>> On 12.08.24 12:37, Thorsten Blum wrote:
> >
On 14. Aug 2024, at 20:02, Nathan Chancellor wrote:
> On Mon, Aug 12, 2024 at 02:03:44PM +0200, Thorsten Blum wrote:
>> On 12. Aug 2024, at 12:54, Johannes Thumshirn
>> wrote:
>>> On 12.08.24 12:37, Thorsten Blum wrote:
Add the __counted_by compiler attribute to the flexible array member
>>
On Mon, Aug 12, 2024 at 02:03:44PM +0200, Thorsten Blum wrote:
> On 12. Aug 2024, at 12:54, Johannes Thumshirn
> wrote:
> > On 12.08.24 12:37, Thorsten Blum wrote:
> >> Add the __counted_by compiler attribute to the flexible array member
> >> stripes to improve access bounds-checking via CONFIG_U
On Tue, Aug 13, 2024 at 12:53:15PM +0200, Thorsten Blum wrote:
> Add the __counted_by compiler attribute to the flexible array member
> name to improve access bounds-checking via CONFIG_UBSAN_BOUNDS and
> CONFIG_FORTIFY_SOURCE.
>
> Signed-off-by: Thorsten Blum
Thanks,
Reviewed-by: David Sterba
On 12. Aug 2024, at 12:54, Johannes Thumshirn
wrote:
> On 12.08.24 12:37, Thorsten Blum wrote:
>> Add the __counted_by compiler attribute to the flexible array member
>> stripes to improve access bounds-checking via CONFIG_UBSAN_BOUNDS and
>> CONFIG_FORTIFY_SOURCE.
>>
>> Signed-off-by: Thorsten
On Mon, Aug 12, 2024 at 12:36:20PM +0200, Thorsten Blum wrote:
> Add the __counted_by compiler attribute to the flexible array member
> stripes to improve access bounds-checking via CONFIG_UBSAN_BOUNDS and
> CONFIG_FORTIFY_SOURCE.
>
> Signed-off-by: Thorsten Blum
Reviewed-by: David Sterba
1 - 100 of 14303 matches
Mail list logo
