diff mbox

drm/i915/cnl: Do not add an extra page for precaution in the Gen10 LRC size

Message ID 1506976313-1144-1-git-send-email-oscar.mateo@intel.com (mailing list archive)
State New, archived
Headers show

Commit Message

oscar.mateo@intel.com Oct. 2, 2017, 8:31 p.m. UTC
BSpec indicates exactly 16750 DWORDs (17 pages), plus one page for PPHWSP.

Suggested-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Fixes: 7fd0b1a ("drm/i915/cnl: Add Gen10 LRC size")
Signed-off-by: Oscar Mateo <oscar.mateo@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Daniele Ceraolo Spurio <daniele.ceraolospurio@intel.com>
Cc: Ben Widawsky <ben@bwidawsk.net>
---
 drivers/gpu/drm/i915/intel_engine_cs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Joonas Lahtinen Oct. 3, 2017, 12:25 p.m. UTC | #1
On Mon, 2017-10-02 at 13:31 -0700, Oscar Mateo wrote:
> BSpec indicates exactly 16750 DWORDs (17 pages), plus one page for PPHWSP.

Assuming this is;

Bspec: 1383

I'm actually getting 16801 DWORDs by copy-pasting to spreadsheet and
summing up. Are we looking from different places?

Both lead to the same amount of pages, but I'd like the commit message
to be correct for easier double-checking :) With commit message
corrected and the Bspec: tag added (or more correct Bspec: tag
specified), this is;

Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

Regards, Joonas
oscar.mateo@intel.com Oct. 3, 2017, 3:46 p.m. UTC | #2
On 10/03/2017 05:25 AM, Joonas Lahtinen wrote:
> On Mon, 2017-10-02 at 13:31 -0700, Oscar Mateo wrote:
>> BSpec indicates exactly 16750 DWORDs (17 pages), plus one page for PPHWSP.
> Assuming this is;
>
> Bspec: 1383
>
> I'm actually getting 16801 DWORDs by copy-pasting to spreadsheet and
> summing up. Are we looking from different places?

No, we are looking at the same BSpec page, but you need to filter by 
"CNL" (that will remove some fields). Even then, I'm now getting 16752 
instead of 16750 (maybe they have updated some tags?).
I'll resend with the BSpec tag, the new count and your RB, if that's OK.

>
> Both lead to the same amount of pages, but I'd like the commit message
> to be correct for easier double-checking :) With commit message
> corrected and the Bspec: tag added (or more correct Bspec: tag
> specified), this is;
>
> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
>
> Regards, Joonas
Joonas Lahtinen Oct. 4, 2017, 10:44 a.m. UTC | #3
On Tue, 2017-10-03 at 08:46 -0700, Oscar Mateo wrote:
> 
> On 10/03/2017 05:25 AM, Joonas Lahtinen wrote:
> > On Mon, 2017-10-02 at 13:31 -0700, Oscar Mateo wrote:
> > > BSpec indicates exactly 16750 DWORDs (17 pages), plus one page for PPHWSP.
> > 
> > Assuming this is;
> > 
> > Bspec: 1383
> > 
> > I'm actually getting 16801 DWORDs by copy-pasting to spreadsheet and
> > summing up. Are we looking from different places?
> 
> No, we are looking at the same BSpec page, but you need to filter by 
> "CNL" (that will remove some fields). Even then, I'm now getting 16752 
> instead of 16750 (maybe they have updated some tags?).
> I'll resend with the BSpec tag, the new count and your RB, if that's OK.

Yep, that's fine. Do include a note in the commit message that there
are exclusions of rows in the BSpec table, so when looking back, one is
reminded that copy and paste is not enough :)

Regards, Joonas
diff mbox

Patch

diff --git a/drivers/gpu/drm/i915/intel_engine_cs.c b/drivers/gpu/drm/i915/intel_engine_cs.c
index a28e2a8..febbb91 100644
--- a/drivers/gpu/drm/i915/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/intel_engine_cs.c
@@ -39,7 +39,7 @@ 
 
 #define GEN8_LR_CONTEXT_RENDER_SIZE	(20 * PAGE_SIZE)
 #define GEN9_LR_CONTEXT_RENDER_SIZE	(22 * PAGE_SIZE)
-#define GEN10_LR_CONTEXT_RENDER_SIZE	(19 * PAGE_SIZE)
+#define GEN10_LR_CONTEXT_RENDER_SIZE	(18 * PAGE_SIZE)
 
 #define GEN8_LR_CONTEXT_OTHER_SIZE	( 2 * PAGE_SIZE)