From: Jeevan B <jeevan.b@intel.com>
To: igt-dev@lists.freedesktop.org
Cc: ramanaidu.naladala@intel.com, jani.nikula@intel.com,
Jeevan B <jeevan.b@intel.com>
Subject: [PATCH i-g-t 1/2] lib/igt_fb: Add IGT_FORMAT_MOD_PREFERRED helper modifier
Date: Mon, 18 May 2026 14:16:30 +0530 [thread overview]
Message-ID: <20260518084631.1085515-2-jeevan.b@intel.com> (raw)
In-Reply-To: <20260518084631.1085515-1-jeevan.b@intel.com>
Add IGT_FORMAT_MOD_PREFERRED as an internal FB modifier sentinel so tests
can request preferred tiling without hardcoding driver rules.
Resolve it in igt_fb_resolve_modifier() (never pass it to kernel) by
checking support for the requested format in this order:
X_TILED -> 4_TILED -> LINEAR.
v2:
- Switch from gen/driver policy to capability-based selection
- Drop Y_TILED preference
Signed-off-by: Jeevan B <jeevan.b@intel.com>
---
lib/igt_fb.c | 38 +++++++++++++++++++++++++++++++++++++-
lib/igt_fb.h | 18 ++++++++++++++++++
2 files changed, 55 insertions(+), 1 deletion(-)
diff --git a/lib/igt_fb.c b/lib/igt_fb.c
index fa9220953..1b0e3f358 100644
--- a/lib/igt_fb.c
+++ b/lib/igt_fb.c
@@ -790,6 +790,40 @@ static int fb_num_planes(const struct igt_fb *fb)
return num_planes;
}
+static uint64_t igt_fb_resolve_modifier(int fd, uint32_t format,
+ uint64_t modifier)
+{
+ static const uint64_t preferred_modifiers[] = {
+ I915_FORMAT_MOD_X_TILED,
+ I915_FORMAT_MOD_4_TILED,
+ DRM_FORMAT_MOD_LINEAR,
+ };
+ igt_display_t display;
+ uint64_t resolved = DRM_FORMAT_MOD_LINEAR;
+ int i;
+
+ if (modifier != IGT_FORMAT_MOD_PREFERRED)
+ return modifier;
+
+ igt_display_require(&display, fd);
+
+ for (i = 0; i < ARRAY_SIZE(preferred_modifiers); i++) {
+ if (preferred_modifiers[i] == DRM_FORMAT_MOD_LINEAR ||
+ igt_display_has_format_mod(&display, format,
+ preferred_modifiers[i])) {
+ resolved = preferred_modifiers[i];
+ break;
+ }
+ }
+
+ igt_display_fini(&display);
+
+ igt_debug("IGT_FORMAT_MOD_PREFERRED resolved to %s\n",
+ igt_fb_modifier_name(resolved));
+
+ return resolved;
+}
+
void igt_init_fb(struct igt_fb *fb, int fd, int width, int height,
uint32_t drm_format, uint64_t modifier,
enum igt_color_encoding color_encoding,
@@ -803,7 +837,7 @@ void igt_init_fb(struct igt_fb *fb, int fd, int width, int height,
fb->width = width;
fb->height = height;
- fb->modifier = modifier;
+ fb->modifier = igt_fb_resolve_modifier(fd, drm_format, modifier);
fb->drm_format = drm_format;
fb->fd = fd;
fb->num_planes = fb_num_planes(fb);
@@ -5266,6 +5300,8 @@ void igt_format_array_fill(uint32_t **formats_array, unsigned int *count,
const char *igt_fb_modifier_name(uint64_t modifier)
{
switch (modifier) {
+ case IGT_FORMAT_MOD_PREFERRED:
+ return "preferred";
case DRM_FORMAT_MOD_LINEAR:
return "linear";
case I915_FORMAT_MOD_X_TILED:
diff --git a/lib/igt_fb.h b/lib/igt_fb.h
index 8e5907dab..cd0923270 100644
--- a/lib/igt_fb.h
+++ b/lib/igt_fb.h
@@ -50,6 +50,24 @@ typedef struct _igt_crc igt_crc_t;
*/
#define IGT_FORMAT_FLOAT fourcc_code('I', 'G', 'F', 'x')
+/**
+ * IGT_FORMAT_MOD_PREFERRED:
+ *
+ * Modifier used by tests to request the best tiling format for the
+ * current device. The actual modifier is selected inside igt_init_fb()
+ * based on device capabilities, so tests don't need to handle driver or
+ * generation-specific details.
+ *
+ * Default selection rules (capability-based):
+ * - If X_TILED is supported for the requested format, use X_TILED
+ * - Else if 4_TILED is supported for the requested format, use 4_TILED
+ * - Else use LINEAR
+ *
+ * Note: This is a special placeholder value and must never be sent to
+ * the kernel. Modifier selection is logged in IGT debug output.
+ */
+#define IGT_FORMAT_MOD_PREFERRED UINT64_C(0xfffffffffffffffe)
+
#define IGT_FORMAT_FMT "%c%c%c%c(0x%08x)"
#define IGT_FORMAT_ARGS(f) ((f) >> 0) & 0xff, ((f) >> 8) & 0xff, \
((f) >> 16) & 0xff, ((f) >> 24) & 0xff, (f)
--
2.43.0
next prev parent reply other threads:[~2026-05-18 8:48 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-18 8:46 [PATCH i-g-t 0/2] Add preferred FB modifier helper Jeevan B
2026-05-18 8:46 ` Jeevan B [this message]
2026-05-18 8:46 ` [PATCH i-g-t 2/2] tests/kms_cursor_legacy: Use preferred modifier for non-cursor framebuffers Jeevan B
2026-05-19 0:30 ` ✓ Xe.CI.BAT: success for Add preferred FB modifier helper Patchwork
2026-05-19 0:52 ` ✓ i915.CI.BAT: " Patchwork
2026-05-19 7:05 ` ✗ Xe.CI.FULL: failure " Patchwork
2026-05-19 20:06 ` ✓ i915.CI.Full: success " Patchwork
-- strict thread matches above, loose matches on Subject: below --
2026-04-13 16:50 [PATCH i-g-t 0/2] RFC: " Jeevan B
2026-04-13 16:50 ` [PATCH i-g-t 1/2] lib/igt_fb: Add IGT_FORMAT_MOD_PREFERRED helper modifier Jeevan B
2026-04-13 18:06 ` Ville Syrjälä
2026-04-16 13:37 ` Maarten Lankhorst
2026-04-16 13:54 ` Ville Syrjälä
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=20260518084631.1085515-2-jeevan.b@intel.com \
--to=jeevan.b@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=jani.nikula@intel.com \
--cc=ramanaidu.naladala@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.