From: "Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
To: "David E. Box" <david.e.box@linux.intel.com>
Cc: thomas.hellstrom@linux.intel.com, rodrigo.vivi@intel.com,
irenic.rajneesh@gmail.com, srinivas.pandruvada@linux.intel.com,
intel-xe@lists.freedesktop.org, dri-devel@lists.freedesktop.org,
xi.pardee@linux.intel.com, Hans de Goede <hansg@kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
platform-driver-x86@vger.kernel.org
Subject: Re: [PATCH 10/22] platform/x86/intel/pmt: Move header decode into common helper
Date: Tue, 17 Mar 2026 17:42:30 +0200 (EET) [thread overview]
Message-ID: <b429177a-178e-f72d-c591-4f1eaa9d8fea@linux.intel.com> (raw)
In-Reply-To: <20260313015202.3660072-11-david.e.box@linux.intel.com>
On Thu, 12 Mar 2026, David E. Box wrote:
> Unify PMT discovery table parsing by moving header decode logic into the
> class driver. A new helper, pmt_read_header(), now fills in the standard
> header fields from the discovery table, replacing the per-namespace
> pmt_header_decode callbacks in telemetry and crashlog.
>
> This centralizes the discovery table bit-field definitions in class.h,
> removes duplicate decode code from telemetry and crashlog, and prepares the
> PMT class for additional discovery sources.
>
> Signed-off-by: David E. Box <david.e.box@linux.intel.com>
> ---
> drivers/platform/x86/intel/pmt/class.c | 38 +++++++++++++++-------
> drivers/platform/x86/intel/pmt/class.h | 15 +++++++--
> drivers/platform/x86/intel/pmt/crashlog.c | 17 ----------
> drivers/platform/x86/intel/pmt/telemetry.c | 26 ---------------
> 4 files changed, 40 insertions(+), 56 deletions(-)
>
> diff --git a/drivers/platform/x86/intel/pmt/class.c b/drivers/platform/x86/intel/pmt/class.c
> index 9b315334a69b..a4d6ffed2fed 100644
> --- a/drivers/platform/x86/intel/pmt/class.c
> +++ b/drivers/platform/x86/intel/pmt/class.c
> @@ -8,6 +8,7 @@
> * Author: "Alexander Duyck" <alexander.h.duyck@linux.intel.com>
> */
>
> +#include <linux/bitfield.h>
> #include <linux/kernel.h>
> #include <linux/log2.h>
> #include <linux/intel_vsec.h>
> @@ -368,26 +369,41 @@ static int intel_pmt_dev_register(struct intel_pmt_entry *entry,
> return ret;
> }
>
> -int intel_pmt_dev_create(struct intel_pmt_entry *entry, struct intel_pmt_namespace *ns,
> - struct intel_vsec_device *intel_vsec_dev, int idx)
> +static int pmt_read_header(struct intel_vsec_device *ivdev, int idx,
> + struct intel_pmt_entry *entry)
> {
> - struct device *dev = &intel_vsec_dev->auxdev.dev;
> - struct resource *disc_res;
> - int ret;
> + struct intel_pmt_header *header = &entry->header;
> + struct device *dev = &ivdev->auxdev.dev;
> + u64 headers[2];
>
> - disc_res = &intel_vsec_dev->resource[idx];
> -
> - entry->disc_table = devm_ioremap_resource(dev, disc_res);
> + entry->disc_table = devm_ioremap_resource(dev, &ivdev->resource[idx]);
> if (IS_ERR(entry->disc_table))
> return PTR_ERR(entry->disc_table);
>
> + memcpy_fromio(headers, entry->disc_table, 2 * sizeof(u64));
> +
> + header->access_type = FIELD_GET(PMT_ACCESS_TYPE, headers[0]);
> + header->telem_type = FIELD_GET(PMT_TELEM_TYPE, headers[0]);
> + /* Size is measured in DWORDS, but accessor returns bytes */
> + header->size = PMT_GET_SIZE(FIELD_GET(PMT_SIZE, headers[0]));
IMO, the FIELD_GET() could be done within the macro as the macro itself
is PMT SIZE specific.
Perhaps name it such that the name indicates unit as you currently end up
using comments to explain that.
> + header->guid = FIELD_GET(PMT_GUID32, headers[0]);
> + header->base_offset = FIELD_GET(PMT_BASE_OFFSET, headers[1]);
> +
> + return 0;
> +}
> +
> +int intel_pmt_dev_create(struct intel_pmt_entry *entry, struct intel_pmt_namespace *ns,
> + struct intel_vsec_device *intel_vsec_dev, int idx)
> +{
> + int ret;
> +
> if (ns->pmt_pre_decode) {
> ret = ns->pmt_pre_decode(intel_vsec_dev, entry);
> if (ret)
> return ret;
> }
>
> - ret = ns->pmt_header_decode(entry, dev);
> + ret = pmt_read_header(intel_vsec_dev, idx, entry);
> if (ret)
> return ret;
>
> @@ -397,11 +413,11 @@ int intel_pmt_dev_create(struct intel_pmt_entry *entry, struct intel_pmt_namespa
> return ret;
> }
>
> - ret = intel_pmt_populate_entry(entry, intel_vsec_dev, disc_res);
> + ret = intel_pmt_populate_entry(entry, intel_vsec_dev, &intel_vsec_dev->resource[idx]);
> if (ret)
> return ret;
>
> - return intel_pmt_dev_register(entry, ns, dev);
> + return intel_pmt_dev_register(entry, ns, &intel_vsec_dev->auxdev.dev);
> }
> EXPORT_SYMBOL_NS_GPL(intel_pmt_dev_create, "INTEL_PMT");
>
> diff --git a/drivers/platform/x86/intel/pmt/class.h b/drivers/platform/x86/intel/pmt/class.h
> index 8a0db0ef58c1..06f90c7ce6b3 100644
> --- a/drivers/platform/x86/intel/pmt/class.h
> +++ b/drivers/platform/x86/intel/pmt/class.h
> @@ -11,6 +11,19 @@
>
> #include "telemetry.h"
>
> +/* PMT Discovery Table DWORD 1 */
> +#define PMT_ACCESS_TYPE GENMASK_ULL(3, 0)
> +#define PMT_TELEM_TYPE GENMASK_ULL(7, 4)
> +#define PMT_SIZE GENMASK_ULL(27, 12)
> +#define PMT_GUID32 GENMASK_ULL(63, 32)
> +
> +/* PMT Discovery Table DWORD 2 */
> +#define PMT_BASE_OFFSET GENMASK_ULL(31, 0)
> +#define PMT_TELE_ID GENMASK_ULL(63, 32)
> +
> +/* Get size bytes from DWORDs */
> +#define PMT_GET_SIZE(v) ((v) << 2)
Use (v) * sizeof(u32) or something along those lines.
> +
> /* PMT access types */
> #define ACCESS_BARID 2
> #define ACCESS_LOCAL 3
> @@ -61,8 +74,6 @@ struct intel_pmt_entry {
> struct intel_pmt_namespace {
> const char *name;
> struct xarray *xa;
> - int (*pmt_header_decode)(struct intel_pmt_entry *entry,
> - struct device *dev);
> int (*pmt_pre_decode)(struct intel_vsec_device *ivdev,
> struct intel_pmt_entry *entry);
> int (*pmt_post_decode)(struct intel_vsec_device *ivdev,
> diff --git a/drivers/platform/x86/intel/pmt/crashlog.c b/drivers/platform/x86/intel/pmt/crashlog.c
> index f936daf99e4d..67795e2cb1ed 100644
> --- a/drivers/platform/x86/intel/pmt/crashlog.c
> +++ b/drivers/platform/x86/intel/pmt/crashlog.c
> @@ -516,28 +516,11 @@ static int pmt_crashlog_pre_decode(struct intel_vsec_device *ivdev,
> return 0;
> }
>
> -static int pmt_crashlog_header_decode(struct intel_pmt_entry *entry,
> - struct device *dev)
> -{
> - void __iomem *disc_table = entry->disc_table;
> - struct intel_pmt_header *header = &entry->header;
> -
> - header->access_type = GET_ACCESS(readl(disc_table));
> - header->guid = readl(disc_table + GUID_OFFSET);
> - header->base_offset = readl(disc_table + BASE_OFFSET);
> -
> - /* Size is measured in DWORDS, but accessor returns bytes */
> - header->size = GET_SIZE(readl(disc_table + SIZE_OFFSET));
The macros become unused after this but were not removed?
Overall, this change is nice consolidation. :-)
--
i.
> -
> - return 0;
> -}
> -
> static DEFINE_XARRAY_ALLOC(crashlog_array);
> static struct intel_pmt_namespace pmt_crashlog_ns = {
> .name = "crashlog",
> .xa = &crashlog_array,
> .pmt_pre_decode = pmt_crashlog_pre_decode,
> - .pmt_header_decode = pmt_crashlog_header_decode,
> };
>
> /*
> diff --git a/drivers/platform/x86/intel/pmt/telemetry.c b/drivers/platform/x86/intel/pmt/telemetry.c
> index d22f633638be..80773e3c3efa 100644
> --- a/drivers/platform/x86/intel/pmt/telemetry.c
> +++ b/drivers/platform/x86/intel/pmt/telemetry.c
> @@ -27,14 +27,6 @@
>
> #include "class.h"
>
> -#define TELEM_SIZE_OFFSET 0x0
> -#define TELEM_GUID_OFFSET 0x4
> -#define TELEM_BASE_OFFSET 0x8
> -#define TELEM_ACCESS(v) ((v) & GENMASK(3, 0))
> -#define TELEM_TYPE(v) (((v) & GENMASK(7, 4)) >> 4)
> -/* size is in bytes */
> -#define TELEM_SIZE(v) (((v) & GENMASK(27, 12)) >> 10)
> -
> /* Used by client hardware to identify a fixed telemetry entry*/
> #define TELEM_CLIENT_FIXED_BLOCK_GUID 0x10000000
>
> @@ -69,23 +61,6 @@ static bool pmt_telem_region_overlaps(struct device *dev, u32 guid, u32 type)
> return false;
> }
>
> -static int pmt_telem_header_decode(struct intel_pmt_entry *entry,
> - struct device *dev)
> -{
> - void __iomem *disc_table = entry->disc_table;
> - struct intel_pmt_header *header = &entry->header;
> -
> - header->access_type = TELEM_ACCESS(readl(disc_table));
> - header->guid = readl(disc_table + TELEM_GUID_OFFSET);
> - header->base_offset = readl(disc_table + TELEM_BASE_OFFSET);
> -
> - /* Size is measured in DWORDS, but accessor returns bytes */
> - header->size = TELEM_SIZE(readl(disc_table));
> - header->telem_type = TELEM_TYPE(readl(entry->disc_table));
> -
> - return 0;
> -}
> -
> static int pmt_telem_post_decode(struct intel_vsec_device *ivdev,
> struct intel_pmt_entry *entry)
> {
> @@ -135,7 +110,6 @@ static DEFINE_XARRAY_ALLOC(telem_array);
> static struct intel_pmt_namespace pmt_telem_ns = {
> .name = "telem",
> .xa = &telem_array,
> - .pmt_header_decode = pmt_telem_header_decode,
> .pmt_post_decode = pmt_telem_post_decode,
> .pmt_add_endpoint = pmt_telem_add_endpoint,
> };
>
next prev parent reply other threads:[~2026-03-17 15:42 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-13 1:51 [PATCH 00/22] platform/x86/intel: Add ACPI PMT discovery support and enable NVL PMC telemetry David E. Box
2026-03-13 1:51 ` [PATCH 01/22] platform/x86/intel/vsec: Refactor base_addr handling David E. Box
2026-03-13 1:51 ` [PATCH 02/22] platform/x86/intel/vsec: Make driver_data info const David E. Box
2026-03-13 1:51 ` [PATCH 03/22] platform/x86/intel/vsec: Decouple add/link helpers from PCI David E. Box
2026-03-13 1:51 ` [PATCH 04/22] platform/x86/intel/vsec: Switch exported helpers from pci_dev to device David E. Box
2026-03-13 1:51 ` [PATCH 05/22] platform/x86/intel/vsec: Return real error codes from registration path David E. Box
2026-03-13 1:51 ` [PATCH 06/22] platform/x86/intel/vsec: Plumb ACPI PMT discovery tables through vsec David E. Box
2026-03-17 16:54 ` Ilpo Järvinen
2026-03-13 1:51 ` [PATCH 07/22] platform/x86/intel/pmt: Add pre/post decode hooks around header parsing David E. Box
2026-03-13 1:51 ` [PATCH 08/22] platform/x86/intel/pmt/crashlog: Split init into pre-decode David E. Box
2026-03-13 1:51 ` [PATCH 09/22] platform/x86/intel/pmt/telemetry: Move overlap check to post-decode hook David E. Box
2026-03-13 1:51 ` [PATCH 10/22] platform/x86/intel/pmt: Move header decode into common helper David E. Box
2026-03-17 15:42 ` Ilpo Järvinen [this message]
2026-03-13 1:51 ` [PATCH 11/22] platform/x86/intel/pmt: Pass discovery index instead of resource David E. Box
2026-03-13 1:51 ` [PATCH 12/22] platform/x86/intel/pmt: Unify header fetch and add ACPI source David E. Box
2026-03-17 15:57 ` Ilpo Järvinen
2026-03-13 1:51 ` [PATCH 13/22] platform/x86/intel/pmc: Add PMC SSRAM Kconfig description David E. Box
2026-03-13 1:51 ` [PATCH 14/22] platform/x86/intel/pmc: Add ACPI PWRM telemetry driver for Nova Lake S David E. Box
2026-03-17 16:24 ` Ilpo Järvinen
2026-03-13 1:51 ` [PATCH 15/22] platform/x86/intel/pmc/ssram: Rename probe and PCI ID table for consistency David E. Box
2026-03-17 16:26 ` Ilpo Järvinen
2026-03-13 1:51 ` [PATCH 16/22] platform/x86/intel/pmc/ssram: Use fixed-size static pmc array David E. Box
2026-03-13 1:51 ` [PATCH 17/22] platform/x86/intel/pmc/ssram: Refactor DEVID/PWRMBASE extraction into helper David E. Box
2026-03-17 16:33 ` Ilpo Järvinen
2026-03-13 1:51 ` [PATCH 18/22] platform/x86/intel/pmc/ssram: Add PCI platform data David E. Box
2026-03-17 16:35 ` Ilpo Järvinen
2026-03-13 1:51 ` [PATCH 19/22] platform/x86/intel/pmc/ssram: Refactor memory barrier for reentrant probe David E. Box
2026-03-17 16:40 ` Ilpo Järvinen
2026-03-13 1:51 ` [PATCH 20/22] platform/x86/intel/pmc/ssram: Add ACPI discovery scaffolding David E. Box
2026-03-17 16:45 ` Ilpo Järvinen
2026-03-13 1:52 ` [PATCH 21/22] platform/x86/intel/pmc/ssram: Make PMT registration optional David E. Box
2026-03-17 16:48 ` Ilpo Järvinen
2026-03-13 1:52 ` [PATCH 22/22] platform/x86/intel/pmc: Add NVL PCI IDs for SSRAM telemetry discovery David E. Box
2026-03-13 2:02 ` ✗ CI.checkpatch: warning for platform/x86/intel: Add ACPI PMT discovery support and enable NVL PMC telemetry Patchwork
2026-03-13 2:04 ` ✓ CI.KUnit: success " Patchwork
2026-03-13 2:19 ` ✗ CI.checksparse: warning " Patchwork
2026-03-13 2:39 ` ✓ Xe.CI.BAT: success " Patchwork
2026-03-13 15:03 ` [PATCH 00/22] " srinivas pandruvada
2026-03-14 4:07 ` ✓ Xe.CI.FULL: success for " Patchwork
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=b429177a-178e-f72d-c591-4f1eaa9d8fea@linux.intel.com \
--to=ilpo.jarvinen@linux.intel.com \
--cc=david.e.box@linux.intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=hansg@kernel.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=irenic.rajneesh@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=platform-driver-x86@vger.kernel.org \
--cc=rodrigo.vivi@intel.com \
--cc=srinivas.pandruvada@linux.intel.com \
--cc=thomas.hellstrom@linux.intel.com \
--cc=xi.pardee@linux.intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.