From: Matthew Brost <matthew.brost@intel.com>
To: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: <intel-xe@lists.freedesktop.org>, Matt Roper <matthew.d.roper@intel.com>
Subject: Re: [RFC PATCH 2/5] drm/xe: Add MI_COPY_MEM_MEM GPU instruction definitions
Date: Fri, 7 Jun 2024 15:22:58 +0000 [thread overview]
Message-ID: <ZmMl0rCBHnw8gPZz@DUT025-TGLU.fm.intel.com> (raw)
In-Reply-To: <2dfb69a3-d74f-4988-8643-b9801b343161@intel.com>
On Fri, Jun 07, 2024 at 01:04:27PM +0200, Michal Wajdeczko wrote:
>
>
> On 07.06.2024 08:52, Matthew Brost wrote:
> > MI_COPY_MEM_MEM GPU instructions are used to copy ctx timestamp from a
> > LRC registers to another location at the beginning of every jobs
> > execution. Add MI_COPY_MEM_MEM GPU instruction definitions.
> >
> > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > ---
> > drivers/gpu/drm/xe/instructions/xe_mi_commands.h | 4 ++++
> > 1 file changed, 4 insertions(+)
> >
> > diff --git a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
> > index c74ceb550dce..f3deabb18ce4 100644
> > --- a/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
> > +++ b/drivers/gpu/drm/xe/instructions/xe_mi_commands.h
> > @@ -56,6 +56,10 @@
> > #define MI_FLUSH_IMM_QW REG_FIELD_PREP(MI_FLUSH_DW_LEN_DW, 5 - 2)
> > #define MI_FLUSH_DW_USE_GTT REG_BIT(2)
> >
> > +#define MI_COPY_MEM_MEM (__MI_INSTR(0x2e) | 3)
>
> all instruction definitions shall be in the opcode order, so move this
> one with 0x2e below 0x29
>
Thanks - Will fix.
> and maybe instead of plain "3" better to use magic XE_INSTR_NUM_DW(5) ?
>
I figured there was macro for this but wasn't immediately clear what to
use. TBH I don't love using a macro like this as it is quite clear in
bspec bits 7:0 the only valid value for this instruction is 3. Now in
our headers we have level of direction... I guess using the macro makes
more sense in variable length instructions but this is a fixed
instruction. Anyways can change if using the macro is the consensus.
> btw, IMO this latter macro would be better if defined as:
>
> #define XE_INSTR_DW_LEN(dw) REG_FIELD_PREP(GENMASK(7, 0), (dw))
> #define XE_INSTR_SIZE(size) XE_INSTR_DW_LEN((size) - 2)
>
> so for fixed instr length we could use XE_INSTR_DW_LEN(3) as in bspec
>
> > +#define MI_COPY_MEM_MEM_SRC_GGTT REG_BIT(22)
> > +#define MI_COPY_MEM_MEM_DST_GGTT REG_BIT(21)
> > +
>
> there should be just extra 2 spaces, not tab + spc + spc
Noticed that in PW link that alignment looked goofy. Will fix.
Matt
>
> > #define MI_LOAD_REGISTER_MEM (__MI_INSTR(0x29) | XE_INSTR_NUM_DW(4))
> > #define MI_LRM_USE_GGTT REG_BIT(22)
> >
next prev parent reply other threads:[~2024-06-07 15:24 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-07 6:52 [RFC PATCH 0/5] Only timeout jobs if they run longer than timeout period Matthew Brost
2024-06-07 6:52 ` [RFC PATCH 1/5] drm/xe: Add LRC ctx timestamp support functions Matthew Brost
2024-06-07 7:10 ` Jani Nikula
2024-06-07 15:16 ` Matthew Brost
2024-06-07 6:52 ` [RFC PATCH 2/5] drm/xe: Add MI_COPY_MEM_MEM GPU instruction definitions Matthew Brost
2024-06-07 11:04 ` Michal Wajdeczko
2024-06-07 15:22 ` Matthew Brost [this message]
2024-06-07 6:52 ` [RFC PATCH 3/5] drm/xe: Emit ctx timestamp copy in ring ops Matthew Brost
2024-06-07 6:52 ` [RFC PATCH 4/5] drm/xe: Add ctx timestamp to LRC snapshot Matthew Brost
2024-06-07 6:52 ` [RFC PATCH 5/5] drm/xe: Sample ctx timestamp to determine if jobs have timed out Matthew Brost
2024-06-07 6:56 ` ✓ CI.Patch_applied: success for Only timeout jobs if they run longer than timeout period Patchwork
2024-06-07 6:56 ` ✗ CI.checkpatch: warning " Patchwork
2024-06-07 6:57 ` ✓ CI.KUnit: success " Patchwork
2024-06-07 7:09 ` ✓ CI.Build: " Patchwork
2024-06-07 7:11 ` ✓ CI.Hooks: " Patchwork
2024-06-07 7:13 ` ✓ CI.checksparse: " Patchwork
2024-06-07 15:27 ` ✗ CI.FULL: failure " 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=ZmMl0rCBHnw8gPZz@DUT025-TGLU.fm.intel.com \
--to=matthew.brost@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=matthew.d.roper@intel.com \
--cc=michal.wajdeczko@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