From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: <intel-xe@lists.freedesktop.org>,
Stuart Summers <stuart.summers@intel.com>,
Matt Roper <matthew.d.roper@intel.com>,
"Riana Tauro" <riana.tauro@intel.com>,
Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>,
Tvrtko Ursulin <tursulin@ursulin.net>,
Raag Jadav <raag.jadav@intel.com>
Subject: Re: [PATCH v5 5/7] drm/xe/lrc: Allow INDIRECT_CTX for more engine classes
Date: Wed, 17 Sep 2025 15:52:24 -0400 [thread overview]
Message-ID: <aMsReM70YRm0TtoY@intel.com> (raw)
In-Reply-To: <20250916-wa-bb-cmds-v5-5-306bddbc15da@intel.com>
On Tue, Sep 16, 2025 at 02:15:42PM -0700, Lucas De Marchi wrote:
> Currently it's only allowed for render and compute. Going forward we
> want to enable it for more engine classes. Let the XE_LRC_FLAG_INDIRECT_CTX
> flag (and thus gt_engine_needs_indirect_ctx()) be the deciding factor
> for its availability.
>
> While at it, add the missing const to rcs_funcs array. Since
> CTX_INDIRECT_CTX_OFFSET_DEFAULT already matches the HW default and
> gt_engine_needs_indirect_ctx() only ever enables it for rcs/ccs, there
> is no change in behavior, it's only preparation for future use case.
>
thanks for the offline explanation and pointers
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
> ---
> drivers/gpu/drm/xe/regs/xe_lrc_layout.h | 3 ---
> drivers/gpu/drm/xe/xe_lrc.c | 14 ++++++++++----
> 2 files changed, 10 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/regs/xe_lrc_layout.h b/drivers/gpu/drm/xe/regs/xe_lrc_layout.h
> index 1b101edb838bf..b5eff383902c5 100644
> --- a/drivers/gpu/drm/xe/regs/xe_lrc_layout.h
> +++ b/drivers/gpu/drm/xe/regs/xe_lrc_layout.h
> @@ -40,7 +40,4 @@
> #define INDIRECT_CTX_RING_START_UDW (0x08 + 1)
> #define INDIRECT_CTX_RING_CTL (0x0a + 1)
>
> -#define CTX_INDIRECT_CTX_OFFSET_MASK REG_GENMASK(15, 6)
> -#define CTX_INDIRECT_CTX_OFFSET_DEFAULT REG_FIELD_PREP(CTX_INDIRECT_CTX_OFFSET_MASK, 0xd)
> -
> #endif
> diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c
> index c706585611d55..0ab99c210d882 100644
> --- a/drivers/gpu/drm/xe/xe_lrc.c
> +++ b/drivers/gpu/drm/xe/xe_lrc.c
> @@ -1281,9 +1281,11 @@ static int setup_wa_bb(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
> static int
> setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
> {
> - static struct bo_setup rcs_funcs[] = {
> + static const struct bo_setup rcs_funcs[] = {
> { .setup = setup_timestamp_wa },
> };
> + static const struct bo_setup xcs_funcs[] = {
> + };
> struct bo_setup_state state = {
> .lrc = lrc,
> .hwe = hwe,
> @@ -1300,6 +1302,9 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
> hwe->class == XE_ENGINE_CLASS_COMPUTE) {
> state.funcs = rcs_funcs;
> state.num_funcs = ARRAY_SIZE(rcs_funcs);
> + } else {
> + state.funcs = xcs_funcs;
> + state.num_funcs = ARRAY_SIZE(xcs_funcs);
> }
>
> if (xe_gt_WARN_ON(lrc->gt, !state.funcs))
> @@ -1326,14 +1331,15 @@ setup_indirect_ctx(struct xe_lrc *lrc, struct xe_hw_engine *hwe)
> finish_bo(&state);
> kfree(state.buffer);
>
> + /*
> + * Enable INDIRECT_CTX leaving INDIRECT_CTX_OFFSET at its default: it
> + * varies per engine class, but the default is good enough
> + */
> xe_lrc_write_ctx_reg(lrc,
> CTX_CS_INDIRECT_CTX,
> (xe_bo_ggtt_addr(lrc->bo) + state.offset) |
> /* Size in CLs. */
> (state.written * sizeof(u32) / 64));
> - xe_lrc_write_ctx_reg(lrc,
> - CTX_CS_INDIRECT_CTX_OFFSET,
> - CTX_INDIRECT_CTX_OFFSET_DEFAULT);
>
> return 0;
> }
>
> --
> 2.50.1
>
next prev parent reply other threads:[~2025-09-17 19:52 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-09-16 21:15 [PATCH v5 0/7] drm/xe: Add user commands to WA BB via configfs Lucas De Marchi
2025-09-16 21:15 ` [PATCH v5 1/7] drm/xe/configfs: Extract function to parse engine Lucas De Marchi
2025-09-16 21:15 ` [PATCH v5 2/7] drm/xe/configfs: Allow to select by class only Lucas De Marchi
2025-09-16 21:15 ` [PATCH v5 3/7] drm/xe/lrc: Allow to add user commands on context switch Lucas De Marchi
2025-09-17 16:32 ` Rodrigo Vivi
2025-09-16 21:15 ` [PATCH v5 4/7] drm/xe/configfs: Add post context restore bb Lucas De Marchi
2025-09-24 20:31 ` Kees Bakker
2025-09-24 21:34 ` Lucas De Marchi
2025-09-16 21:15 ` [PATCH v5 5/7] drm/xe/lrc: Allow INDIRECT_CTX for more engine classes Lucas De Marchi
2025-09-17 19:52 ` Rodrigo Vivi [this message]
2025-09-16 21:15 ` [PATCH v5 6/7] drm/xe/lrc: Allow to add user commands mid context switch Lucas De Marchi
2025-09-17 19:53 ` Rodrigo Vivi
2025-09-16 21:15 ` [PATCH v5 7/7] drm/xe/configfs: Add mid context restore bb Lucas De Marchi
2025-09-17 19:53 ` Rodrigo Vivi
2025-09-16 21:21 ` ✗ CI.checkpatch: warning for drm/xe: Add user commands to WA BB via configfs Patchwork
2025-09-16 21:23 ` ✓ CI.KUnit: success " Patchwork
2025-09-16 21:58 ` ✓ Xe.CI.BAT: " Patchwork
2025-09-17 1:22 ` ✗ Xe.CI.Full: failure " Patchwork
2025-09-17 6:44 ` [PATCH v5 0/7] " Raag Jadav
2025-09-18 21:47 ` Lucas De Marchi
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=aMsReM70YRm0TtoY@intel.com \
--to=rodrigo.vivi@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=lucas.demarchi@intel.com \
--cc=matthew.d.roper@intel.com \
--cc=raag.jadav@intel.com \
--cc=riana.tauro@intel.com \
--cc=stuart.summers@intel.com \
--cc=tursulin@ursulin.net \
--cc=umesh.nerlige.ramappa@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