public inbox for igt-dev@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Xin Wang <x.wang@intel.com>
To: igt-dev@lists.freedesktop.org
Cc: alex.zuo@intel.com, shuicheng.lin@intel.com,
	stuart.summers@intel.com, kamil.konieczny@intel.com,
	matthew.d.roper@intel.com, ravi.kumar.vodapalli@intel.com,
	Xin Wang <x.wang@intel.com>
Subject: [PATCH v9 2/2] lib/intel_device_info: Query runtime xe device graphics versions
Date: Wed, 17 Dec 2025 04:34:50 +0000	[thread overview]
Message-ID: <20251217043450.241761-3-x.wang@intel.com> (raw)
In-Reply-To: <20251217043450.241761-1-x.wang@intel.com>

For platforms with graphics_ver >= 20, query the runtime xe device ver
instead of relying solely on hardcoded values from the PCI device table.
This enables accurate IP minor version (graphics_rel) detection for
platforms like Xe2 where different steppings have different IP versions.

Implementation details:
- Use weak symbol linkage for xe_ipver_cache_lookup() to handle static
  library compilation (libigt_chipset.a, libigt_device_scan.a) without
  xe_query.c dependencies which are used for i915 tools (i915_perf and
  intel_gpu_top)
- Provide a weak stub that returns NULL when xe_query is not linked
- For Gen20+ platforms, prefer runtime xe device versions over static data
- Fall back to PCI table if xe device info is unavailable
- Reset cache on query failure to allow retry

Remove hardcoded graphics_rel from static table entries for xe devices
as they will be populated at runtime from GMD_ID.

This unifies device info handling between i915 and xe drivers, enabling:
- Platform-specific workarounds based on accurate IP versions
- Consistent device info API across both drivers

Signed-off-by: Xin Wang <x.wang@intel.com>
---
 lib/intel_device_info.c | 26 +++++++++++++++++++++++---
 1 file changed, 23 insertions(+), 3 deletions(-)

diff --git a/lib/intel_device_info.c b/lib/intel_device_info.c
index 20147479a..3bc20ab68 100644
--- a/lib/intel_device_info.c
+++ b/lib/intel_device_info.c
@@ -3,6 +3,16 @@
 #include "i915_pciids_local.h"
 
 #include <strings.h> /* ffs() */
+#include <stddef.h>
+#include <string.h>
+
+/* Weak symbol stub - will be overridden if xe_query.c is linked */
+struct xe_device_ipver *xe_ipver_cache_lookup(uint32_t devid) __attribute__((weak));
+
+struct xe_device_ipver *xe_ipver_cache_lookup(uint32_t devid)
+{
+	return NULL;
+}
 
 static const struct intel_device_info intel_generic_info = {
 	.graphics_ver = 0,
@@ -505,7 +515,6 @@ static const struct intel_device_info intel_pontevecchio_info = {
 
 static const struct intel_device_info intel_lunarlake_info = {
 	.graphics_ver = 20,
-	.graphics_rel = 4,
 	.display_ver = 20,
 	.has_4tile = true,
 	.has_flatccs = true,
@@ -517,7 +526,6 @@ static const struct intel_device_info intel_lunarlake_info = {
 
 static const struct intel_device_info intel_battlemage_info = {
 	.graphics_ver = 20,
-	.graphics_rel = 1,
 	.display_ver = 14,
 	.has_4tile = true,
 	.has_flatccs = true,
@@ -529,7 +537,6 @@ static const struct intel_device_info intel_battlemage_info = {
 
 static const struct intel_device_info intel_pantherlake_info = {
 	.graphics_ver = 30,
-	.graphics_rel = 0,
 	.display_ver = 30,
 	.has_4tile = true,
 	.has_flatccs = true,
@@ -700,6 +707,8 @@ const struct intel_device_info *intel_get_device_info(uint16_t devid)
 {
 	static __thread const struct intel_device_info *cache = &intel_generic_info;
 	static __thread uint16_t cached_devid;
+	static __thread struct intel_device_info xe_dev_info;
+	struct xe_device_ipver *ipver;
 	int i;
 
 	if (cached_devid == devid)
@@ -714,6 +723,17 @@ const struct intel_device_info *intel_get_device_info(uint16_t devid)
 	cached_devid = devid;
 	cache = (void *)intel_device_match[i].match_data;
 
+	if (cache->graphics_ver >= 20) {
+		ipver = xe_ipver_cache_lookup(devid);
+		if (ipver && ipver->devid == devid) {
+			memcpy(&xe_dev_info, cache, sizeof(struct intel_device_info));
+			xe_dev_info.graphics_ver = ipver->graphics_ver;
+			xe_dev_info.graphics_rel = ipver->graphics_rel;
+			cache = &xe_dev_info;
+		} else {
+			cached_devid = 0;
+		}
+	}
 out:
 	return cache;
 }
-- 
2.43.0


  parent reply	other threads:[~2025-12-17  4:35 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-12-17  4:34 [PATCH v9 0/2] lib/intel_device_info: get the xe .graphics_rel from GMD_ID Xin Wang
2025-12-17  4:34 ` [PATCH v9 1/2] lib/xe/xe_query: Get runtime xe device graphics version " Xin Wang
2025-12-17  7:08   ` Vodapalli, Ravi Kumar
2025-12-17  4:34 ` Xin Wang [this message]
2025-12-17  7:21   ` [PATCH v9 2/2] lib/intel_device_info: Query runtime xe device graphics versions Vodapalli, Ravi Kumar
2025-12-17  6:19 ` ✓ i915.CI.BAT: success for lib/intel_device_info: get the xe .graphics_rel from GMD_ID Patchwork
2025-12-17  7:02 ` ✓ Xe.CI.BAT: " Patchwork
2025-12-17  8:02 ` ✗ i915.CI.Full: failure " Patchwork
2025-12-18  4:50 ` ✗ Xe.CI.Full: " Patchwork
2025-12-22 16:16 ` ✗ Xe.CI.BAT: failure for lib/intel_device_info: get the xe .graphics_rel from GMD_ID (rev3) Patchwork
2025-12-22 16:19 ` ✓ i915.CI.BAT: success " Patchwork
2025-12-22 18:34 ` ✓ Xe.CI.Full: " Patchwork
2025-12-23 19:58 ` ✓ Xe.CI.BAT: success for lib/intel_device_info: get the xe .graphics_rel from GMD_ID (rev4) Patchwork
2025-12-23 20:11 ` ✓ i915.CI.BAT: " Patchwork
2025-12-23 23:05 ` ✗ i915.CI.Full: failure for lib/intel_device_info: get the xe .graphics_rel from GMD_ID (rev3) Patchwork
2025-12-24  6:10 ` ✓ Xe.CI.Full: success for lib/intel_device_info: get the xe .graphics_rel from GMD_ID (rev4) Patchwork
2025-12-25  0:39 ` ✗ i915.CI.Full: failure " Patchwork
2026-01-02 19:23 ` ✓ Xe.CI.BAT: success for lib/intel_device_info: get the xe .graphics_rel from GMD_ID (rev5) Patchwork
2026-01-02 19:37 ` ✗ i915.CI.BAT: failure " Patchwork
2026-01-02 20:17 ` ✗ Xe.CI.Full: " 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=20251217043450.241761-3-x.wang@intel.com \
    --to=x.wang@intel.com \
    --cc=alex.zuo@intel.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=kamil.konieczny@intel.com \
    --cc=matthew.d.roper@intel.com \
    --cc=ravi.kumar.vodapalli@intel.com \
    --cc=shuicheng.lin@intel.com \
    --cc=stuart.summers@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