From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: "Michael J. Ruhl" <michael.j.ruhl@intel.com>
Cc: <platform-driver-x86@vger.kernel.org>,
<intel-xe@lists.freedesktop.org>, <hansg@kernel.org>,
<ilpo.jarvinen@linux.intel.com>, <matthew.brost@intel.com>,
<thomas.hellstrom@linux.intel.com>, <airlied@gmail.com>,
<simona@ffwll.ch>, <david.e.box@linux.intel.com>,
<anoop.c.vijay@intel.com>, <badal.nilawar@intel.com>,
<matthew.d.roper@intel.com>, <james.ausmus@intel.com>
Subject: Re: [PATCH v19 14/19] drm/xe/vsec: Crescent Island PMT callbacks
Date: Thu, 19 Nov 2026 16:05:05 -0500 [thread overview]
Message-ID: <av9kgfXlujk2SeT-@intel.com> (raw)
In-Reply-To: <20261119155438.40762-35-michael.j.ruhl@intel.com>
On Thu, Nov 19, 2026 at 07:54:52AM -0800, Michael J. Ruhl wrote:
> CRI PMT support requires callbacks to access the discovery data and
> control areas. Access is a common MMIO area that requires an index to
> be set before access is allowed.
>
> Introduce the necessary callbacks to get the discovery, status and
> control information for CRI PMT usage.
>
> Add the glue logic to register the CRI PMT functionality.
>
> Signed-off-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
> ---
> drivers/gpu/drm/xe/xe_vsec.c | 178 ++++++++++++++++++++++++++++++++++-
> 1 file changed, 174 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_vsec.c b/drivers/gpu/drm/xe/xe_vsec.c
> index a814d7151eea..58eb85fb6a9f 100644
> --- a/drivers/gpu/drm/xe/xe_vsec.c
> +++ b/drivers/gpu/drm/xe/xe_vsec.c
> @@ -4,6 +4,7 @@
> #include <linux/bits.h>
> #include <linux/cleanup.h>
> #include <linux/errno.h>
> +#include <linux/log2.h>
> #include <linux/intel_vsec.h>
> #include <linux/module.h>
> #include <linux/mutex.h>
> @@ -344,10 +345,15 @@ int xe_pmt_telem_read(struct device *dev, u32 guid, u64 *data, loff_t user_offse
> }
>
> scoped_guard(mutex, &xe->pmt.lock) {
> + u32 device_id = FIELD_GET(GUID_DEVICE_ID, guid);
> +
> /* set SoC re-mapper index register based on GUID memory region */
> xe->soc_remapper.set_telem_region(xe, mem_region);
>
> memcpy_fromio(data, telem_addr, count);
> +
> + if (device_id == CRI_DEVICE_ID)
> + xe->soc_remapper.set_telem_region(xe, CRI_IDX_TELEM_DISCOVERY);
> }
>
> xe_pm_runtime_put(xe);
> @@ -358,20 +364,178 @@ int xe_pmt_telem_read(struct device *dev, u32 guid, u64 *data, loff_t user_offse
> return ret == 0 ? count : ret;
> }
>
> -static struct pmt_callbacks xe_pmt_cb = {
> +/**
> + * xe_pmt_discovery_read - Given a device and an index, read discovery data into a buffer
> + * @dev: valid Xe device
> + * @cap_id: PMT type
> + * @element: element to access
> + * @disc_data: buffer for discovery data
> + * @count: bytes to read
count doesn't exist below...
> + *
> + * This is a callback API. I.e this can be accessed external to the XE driver
> + * (PMT driver scope). Because of this, DRM hotplug needs to be verified
> + * (drm_dev_enter()).
> + *
> + * Return:
> + * * -ENODEV device is not available (hotplug)
> + * * -EINVAL for unsupported device or PMT feature
> + * * 0 on success
> + */
> +static int xe_pmt_discovery_read(struct device *dev, unsigned long cap_id, u32 element,
> + u32 *disc_data)
> +{
> + struct xe_device *xe = kdev_to_xe_device(dev);
> + void __iomem *disc_addr = xe->mmio.regs;
> + struct intel_vsec_header *headers;
> + int ret = 0;
> + u32 offset;
> + int idx;
> +
> + if (!drm_dev_enter(&xe->drm, &idx))
> + return -ENODEV;
> +
> + if (xe->info.platform != XE_CRESCENTISLAND) {
> + ret = -EINVAL;
> + goto dev_exit;
> + }
> +
> + switch (cap_id) {
> + case ilog2(VSEC_CAP_TELEMETRY):
> + headers = &cri_telemetry;
> + break;
> +
> + case ilog2(VSEC_CAP_CRASHLOG):
> + headers = &cri_crashlog;
> + break;
> +
> + default:
> + ret = -EINVAL;
> + goto dev_exit;
> + }
> +
> + offset = element * (headers->entry_size * sizeof(u32));
> + disc_addr += headers->offset + offset;
> +
> + scoped_guard(mutex, &xe->pmt.lock) {
> + xe->soc_remapper.set_telem_region(xe, CRI_IDX_TELEM_DISCOVERY);
> + memcpy_fromio(disc_data, disc_addr, PMT_DISC_DWORDS * sizeof(u32));
> + }
we probably need rpm get/put here, no?!
> +
> +dev_exit:
> + drm_dev_exit(idx);
> +
> + return ret;
> +}
> +
> +/**
> + * xe_pmt_read_reg() - read a crashlog register
> + * @dev: the xe device that registered the callback
> + * @guid: PMT guid of the crashlog instance
> + * @reg: data read from the PMT data structure
> + * @offset: which data to read from the PMT data structure
> + *
> + * Read the requested PMT register based on the pcie device and guid. The
> + * supported struct is the Crashlog Type1 Version2.
> + *
> + * Currently this is for CRI only.
> + *
> + * Return:
> + * * -ENODEV device is not available (hotplug)
> + * * -EINVAL invalid guid
> + * * 0 success
> + */
> +static int xe_pmt_read_reg(struct device *dev, u32 guid, u32 *reg, u32 offset)
> +{
> + struct xe_device *xe = kdev_to_xe_device(dev);
> + void __iomem *disc_addr = xe->mmio.regs + xe->pmt.base_offset;
> + int ret = 0;
> + u32 inst;
> + int idx;
> +
> + if (FIELD_GET(GUID_DEVICE_ID, guid) != CRI_DEVICE_ID ||
> + FIELD_GET(GUID_CAP_TYPE, guid) != CRASHLOG)
> + return -EINVAL;
> +
> + if (!drm_dev_enter(&xe->drm, &idx))
> + return -ENODEV;
> +
> + inst = FIELD_GET(GUID_RECORD_ID, guid) == PUNIT ?
> + CRI_CRASHLOG_PUNIT_DISC_OFFSET : CRI_CRASHLOG_OOBMSM_DISC_OFFSET;
> + disc_addr += inst + offset;
> +
> + xe_pm_runtime_get(xe);
> +
> + scoped_guard(mutex, &xe->pmt.lock) {
> + xe->soc_remapper.set_telem_region(xe, CRI_IDX_TELEM_DISCOVERY);
> +
> + *reg = readl(disc_addr);
> + }
> +
> + xe_pm_runtime_put(xe);
> +
> + drm_dev_exit(idx);
> +
> + return ret;
> +}
> +
> +static int xe_pmt_write_reg(struct device *dev, u32 guid, u32 reg, u32 offset)
> +{
> + struct xe_device *xe = kdev_to_xe_device(dev);
> + void __iomem *disc_addr = xe->mmio.regs + xe->pmt.base_offset;
> + int ret = 0;
> + u32 inst;
> + int idx;
> +
> + if (FIELD_GET(GUID_DEVICE_ID, guid) != CRI_DEVICE_ID ||
> + FIELD_GET(GUID_CAP_TYPE, guid) != CRASHLOG)
> + return -EINVAL;
> +
> + if (!drm_dev_enter(&xe->drm, &idx))
> + return -ENODEV;
> +
> + inst = FIELD_GET(GUID_RECORD_ID, guid) == PUNIT ?
> + CRI_CRASHLOG_PUNIT_DISC_OFFSET : CRI_CRASHLOG_OOBMSM_DISC_OFFSET;
> + disc_addr += inst + offset;
> +
> + xe_pm_runtime_get(xe);
> +
> + scoped_guard(mutex, &xe->pmt.lock) {
> + xe->soc_remapper.set_telem_region(xe, CRI_IDX_TELEM_DISCOVERY);
> +
> + writel(reg, disc_addr);
> + }
> +
> + xe_pm_runtime_put(xe);
> +
> + drm_dev_exit(idx);
> +
> + return ret;
> +}
> +
> +static struct pmt_callbacks xe_bmg_pmt_cb = {
> .read_telem = xe_pmt_telem_read,
> };
>
> +static struct pmt_callbacks xe_cri_pmt_cb = {
> + .read_telem = xe_pmt_telem_read,
> + .read_discovery = xe_pmt_discovery_read,
> + .read_reg = xe_pmt_read_reg,
> + .write_reg = xe_pmt_write_reg,
> +};
> +
> static const int vsec_platforms[] = {
> [XE_BATTLEMAGE] = XE_VSEC_BMG,
> + [XE_CRESCENTISLAND] = XE_VSEC_CRI,
> };
>
> static enum xe_vsec get_platform_info(struct xe_device *xe)
> {
> - if (xe->info.platform > XE_BATTLEMAGE)
> + if (xe->info.platform > XE_CRESCENTISLAND)
> return XE_VSEC_UNKNOWN;
>
> - if (IS_SRIOV_VF(xe) && xe->info.platform == XE_BATTLEMAGE)
> + if (IS_SRIOV_VF(xe) &&
> + (xe->info.platform == XE_BATTLEMAGE ||
> + xe->info.platform == XE_CRESCENTISLAND))
> return XE_VSEC_UNKNOWN;
>
> return vsec_platforms[xe->info.platform];
> @@ -399,8 +563,14 @@ void xe_vsec_init(struct xe_device *xe)
> switch (platform) {
> case XE_VSEC_BMG:
> xe->pmt.base_offset = BMG_TELEMETRY_OFFSET;
> - info->priv_data = &xe_pmt_cb;
> + info->priv_data = &xe_bmg_pmt_cb;
> break;
> +
> + case XE_VSEC_CRI:
> + xe->pmt.base_offset = CRI_PMT_OFFSET;
> + info->priv_data = &xe_cri_pmt_cb;
> + break;
> +
> default:
> break;
> }
> --
> 2.43.0
>
next prev parent reply other threads:[~2026-11-19 21:06 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-19 15:54 [PATCH v19 00/19] Crescent Island PMT support Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 01/19] platform/x86/intel/pmt: Introduce a callback for discovery data Michael J. Ruhl
2026-11-19 20:06 ` David E. Box
2026-11-19 15:54 ` [PATCH v19 02/19] platform/x86/intel/pmt: Use the pre-populated header info for crashlog Michael J. Ruhl
2026-11-19 20:13 ` David E. Box
2026-11-19 15:54 ` [PATCH v19 03/19] drm/xe/vsec: VSEC requires soc-remapper Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 04/19] drm/xe/vsec: Do not register BMG PMT for VF Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 05/19] drm/xe/vsec: Correct locking order Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 06/19] drm/xe/vsec: Use correct pm state get Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 07/19] drm/xe/vsec: Add DOC text for VSEC Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 08/19] drm/xe/vsec: Support possible hotplug exit Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 09/19] drm/xe/vsec: Refactor BattleMage PMT defines Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 10/19] drm/xe/vsec: Update VSEC probe order Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 11/19] drm/xe/vsec: Add base_offset to allow for more flexibilty Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 12/19] drm/xe/vsec: Support Crescent Island PMT Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 13/19] drm/xe/vsec: Crescent Island PMT decode Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 14/19] drm/xe/vsec: Crescent Island PMT callbacks Michael J. Ruhl
2026-11-19 21:05 ` Rodrigo Vivi [this message]
2026-11-19 21:41 ` Ruhl, Michael J
2026-11-19 21:55 ` Vivi, Rodrigo
2026-11-19 15:54 ` [PATCH v19 15/19] drm/xe/vsec: Support late bind fw information Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 16/19] drm/xe/vsec: Add PMT GUID internal access Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 17/19] drm/xe/vsec: Update PMT " Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 18/19] drm/xe/vsec: Refactor platform check Michael J. Ruhl
2026-11-19 15:54 ` [PATCH v19 19/19] drm/xe/vsec: Support survivability mode for CRI Michael J. Ruhl
2026-11-19 20:11 ` ✗ CI.checkpatch: warning for Crescent Island PMT support (rev20) Patchwork
2026-11-19 20:13 ` ✓ CI.KUnit: success " Patchwork
2026-11-19 20:55 ` ✓ Xe.CI.BAT: " 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=av9kgfXlujk2SeT-@intel.com \
--to=rodrigo.vivi@intel.com \
--cc=airlied@gmail.com \
--cc=anoop.c.vijay@intel.com \
--cc=badal.nilawar@intel.com \
--cc=david.e.box@linux.intel.com \
--cc=hansg@kernel.org \
--cc=ilpo.jarvinen@linux.intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=james.ausmus@intel.com \
--cc=matthew.brost@intel.com \
--cc=matthew.d.roper@intel.com \
--cc=michael.j.ruhl@intel.com \
--cc=platform-driver-x86@vger.kernel.org \
--cc=simona@ffwll.ch \
--cc=thomas.hellstrom@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox