On Wed, 2015-06-10 at 11:37 +0100, Chris Wilson wrote:
> On Wed, Jun 10, 2015 at 09:12:16AM +0100, Peter Antoine wrote:
> > This change adds the programming of the MOCS registers to the gen 9+
> > platforms. This change set programs the MOCS register values to a set
> > of values that are defined to be optimal.
> >
> > It creates a fixed register set that is programmed across the different
> > engines so that all engines have the same table. This is done as the
> > main RCS context only holds the registers for itself and the shared
> > L3 values. By trying to keep the registers consistent across the
> > different engines it should make the programming for the registers
> > consistent.
> >
> > v2:
> > -'static const' for private data structures and style changes.(Matt Turner)
> > v3:
> > - Make the tables "slightly" more readable. (Damien Lespiau)
> > - Updated tables fix performance regression.
> >
> >
> > Signed-off-by: Peter Antoine <peter.antoine@xxxxxxxxx>
> > ---
> > drivers/gpu/drm/i915/Makefile | 3 +-
> > drivers/gpu/drm/i915/i915_reg.h | 9 ++
> > drivers/gpu/drm/i915/intel_lrc.c | 68 ++++++++++
> > drivers/gpu/drm/i915/intel_mocs.c | 252 ++++++++++++++++++++++++++++++++++++++
> > drivers/gpu/drm/i915/intel_mocs.h | 119 ++++++++++++++++++
> > 5 files changed, 450 insertions(+), 1 deletion(-)
> > create mode 100644 drivers/gpu/drm/i915/intel_mocs.c
> > create mode 100644 drivers/gpu/drm/i915/intel_mocs.h
> >
> > diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile
> > index b7ddf48..cd7b910 100644
> > --- a/drivers/gpu/drm/i915/Makefile
> > +++ b/drivers/gpu/drm/i915/Makefile
> > @@ -36,7 +36,8 @@ i915-y += i915_cmd_parser.o \
> > i915_trace_points.o \
> > intel_lrc.o \
> > intel_ringbuffer.o \
> > - intel_uncore.o
> > + intel_uncore.o \
> > + intel_mocs.o
>
> Please keep it alphabetical.
done.
>
> > # autogenerated null render state
> > i915-y += intel_renderstate_gen6.o \
> > diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
> > index 7213224..3a435b5 100644
> > --- a/drivers/gpu/drm/i915/i915_reg.h
> > +++ b/drivers/gpu/drm/i915/i915_reg.h
> > @@ -7829,4 +7829,13 @@ enum skl_disp_power_wells {
> > #define _PALETTE_A (dev_priv->info.display_mmio_offset + 0xa000)
> > #define _PALETTE_B (dev_priv->info.display_mmio_offset + 0xa800)
> >
> > +/* MOCS (Memory Object Control State) registers */
> > +#define GEN9_LNCFCMOCS0 (0xB020) /* L3 Cache Control base */
> > +
> > +#define GEN9_GFX_MOCS_0 (0xc800) /* Graphics MOCS base register*/
> > +#define GEN9_MFX0_MOCS_0 (0xc900) /* Media 0 MOCS base register*/
> > +#define GEN9_MFX1_MOCS_0 (0xcA00) /* Media 1 MOCS base register*/
> > +#define GEN9_VEBOX_MOCS_0 (0xcB00) /* Video MOCS base register*/
> > +#define GEN9_BLT_MOCS_0 (0xcc00) /* Blitter MOCS base register*/
> > +
> > #endif /* _I915_REG_H_ */
> > diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
> > index 9f5485d..c875569 100644
> > --- a/drivers/gpu/drm/i915/intel_lrc.c
> > +++ b/drivers/gpu/drm/i915/intel_lrc.c
> > @@ -135,6 +135,7 @@
> > #include <drm/drmP.h>
> > #include <drm/i915_drm.h>
> > #include "i915_drv.h"
> > +#include "intel_mocs.h"
> >
> > #define GEN9_LR_CONTEXT_RENDER_SIZE (22 * PAGE_SIZE)
> > #define GEN8_LR_CONTEXT_RENDER_SIZE (20 * PAGE_SIZE)
> > @@ -1370,6 +1371,67 @@ out:
> > return ret;
> > }
> >
> > +/*
> > + * i915_gem_program_mocs() - program the MOCS register.
> > + *
> > + * ring: The ring that the programming batch will be run in.
> > + * ctx: The intel_context to be used.
> > + *
> > + * This function will emit a batch buffer with the values required for
> > + * programming the MOCS register values for all the currenly supported
> > + * rings.
> > + *
> > + * Return: 0 on success, otherwise the error status.
> > + */
> > +static int i915_gem_program_mocs(struct intel_engine_cs *ring,
> > + struct intel_context *ctx)
>
> Odd function name. gen8_init_mocs_tables()
Name fixed. gen9_*
>
> > +{
> > + int ret = 0;
> > +
> > + struct drm_i915_mocs_table t;
> > + struct drm_device *dev = ring->dev;
> > + struct intel_ringbuffer *ringbuf = ctx->engine[ring->id].ringbuf;
> > +
> > + if (get_mocs_settings(dev, &t)) {
>
> Worse, this function is being pulled in from another file so needs a
> proper function naem. Just return the table pointer.
>
function calls back were they belong.
> > + u32 table_size;
> > +
> > + /*
> > + * OK. For each supported ring:
> > + * table_size * 2 dwords for each control_value
> > + * plus table/2 dwords for l3cc values.
> > + *
> > + * Plus 1 for the load command and 1 for the NOOP per ring
> > + * and the l3cc programming.
> > + */
> > + table_size = GEN9_NUM_MOCS_RINGS * ((2 * t.size) + 2) +
> > + t.size + 2;
> > + ret = intel_logical_ring_begin(ringbuf, ctx, table_size);
> > + if (ret) {
> > + DRM_ERROR("intel_logical_ring_begin failed %d\n", ret);
>
> Error? Does this help the user in any way, especially as not enabling
> mocs is not critical?
No. Changed to DEBUG.
>
> > + return ret;
> > + }
> > +
> > + /* program the control registers */
> > + emit_mocs_control_table(ringbuf, &t, GEN9_GFX_MOCS_0);
> > + emit_mocs_control_table(ringbuf, &t, GEN9_MFX0_MOCS_0);
> > + emit_mocs_control_table(ringbuf, &t, GEN9_MFX1_MOCS_0);
> > + emit_mocs_control_table(ringbuf, &t, GEN9_VEBOX_MOCS_0);
> > + emit_mocs_control_table(ringbuf, &t, GEN9_BLT_MOCS_0);
>
> So why are programming other rings on RCS without explicit inter-engine
> ordering?
Basically the same as the workarounds (and the null context) set these
up when the the RCS is loaded. As the other engines DO NOT have these
registers in their context it is done here. ALSO, as the l3cc registers
are indexed by the same index value (when using the MOCS) as part of the
caching, they need to match so this is why they are done at the same
time.
I have changed the comments to make this a little clearer.
>
> > +
> > + /* now program the l3cc registers */
> > + emit_mocs_l3cc_table(ringbuf, &t);
> > +
> > + intel_logical_ring_advance(ringbuf);
> > +
> > + DRM_INFO("MOCS: Table set in Context\n");
> > + } else {
> > + DRM_INFO("MOCS: Table Not supported on platform\n");
>
> Info? Just debug, these are not helpful user messages.
Change to DEBUG.
>
> > + }
> > +
> > + return ret;
> > +}
> > +
> > +
> > static int gen8_init_rcs_context(struct intel_engine_cs *ring,
> > struct intel_context *ctx)
> > {
> > @@ -1379,6 +1441,12 @@ static int gen8_init_rcs_context(struct intel_engine_cs *ring,
> > if (ret)
> > return ret;
> >
> > + /*
> > + * Failing to program the MOCS is non-fatal.The system will not
> > + * run at peak performance. So generate a warning and carry on.
> > + */
> > + WARN_ON(i915_gem_program_mocs(ring, ctx) != 0);
>
> Warn? Just NOTICE since it is not a useful user error message and a
> non-critical error to boot.
Changed to DRM_ERROR on fail.
>
> > return intel_lr_context_render_state_init(ring, ctx);
> > }
> >
> > diff --git a/drivers/gpu/drm/i915/intel_mocs.c b/drivers/gpu/drm/i915/intel_mocs.c
> > new file mode 100644
> > index 0000000..841932a
> > --- /dev/null
> > +++ b/drivers/gpu/drm/i915/intel_mocs.c
> > @@ -0,0 +1,252 @@
> > +/*
> > + * Copyright (c) 2015 Intel Corporation
> > + *
> > + * Permission is hereby granted, free of charge, to any person obtaining a
> > + * copy of this software and associated documentation files (the "Software"),
> > + * to deal in the Software without restriction, including without limitation
> > + * the rights to use, copy, modify, merge, publish, distribute, sublicense,
> > + * and/or sell copies of the Software, and to permit persons to whom the
> > + * Software is furnished to do so, subject to the following conditions: *
> > + * The above copyright notice and this permission notice (including the next
> > + * paragraph) shall be included in all copies or substantial portions of the
> > + * Software.
> > + *
> > + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
> > + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
> > + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
> > + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
> > + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
> > + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
> > + * SOFTWARE.
> > + *
> > + * Authors:
> > + * Peter Antoine <peter.antoine@xxxxxxxxx>
>
> We don't encourage Authors: since git does a better job.
Removed. Was in the file header that I looked at.
>
> > + */
> > +
> > +#include "intel_mocs.h"
> > +#include "intel_lrc.h"
> > +#include "intel_ringbuffer.h"
> > +
> > +/*
> > + * MOCS tables
> > + *
> > + * These are the MOCS tables that are programmed across all the rings.
> > + * The control value is programmed to all the rings that support the
> > + * MOCS registers. While the l3cc_values are only programmed to the
> > + * LNCFCMOCS0 - LNCFCMOCS32 registers.
> > + *
> > + * NOTE: These tables MUST start with being uncached {0,0} and the
> > + * the length MUST be less than 63 as the last two registers are
> > + * reserved by the hardware.
> > + */
> > +struct drm_i915_mocs_entry skylake_mocs_table[] = {
>
> static const ...
done.
>
> > + /* {0x00000009, 0x0010} */
> > + {(MOCS_CACHEABILITY(EDRAM_UC) | MOCS_TGT_CACHE(LLC_ELLC) |
> > + MOCS_LRUM(0) | MOCS_AOM(0) | MOCS_LECC_ESC(0) | MOCS_SCC(0) |
> > + MOC_PFM(0) | MOCS_SCF(0)),
> > + (MOCS_ESC(0) | MOCS_SCC(0) | MOCS_L3_CACHEABILITY(L3_UC))},
>
> Whitespace and alignment should be treated as our friends and not
> enemies.
>
> #define SKYLAKE_MOCS(cache, ellc, lrum, l3) { \
> (MOCS_CACHEABILITY(cache) | \
> MOCS_TGT_CACHE(ellc) | \
> MOCS_LRUM(lrum) | \
> MOCS_AOM(0) | \
> MOCS_LECC_ESC(0) | \
> MOCS_SCC(0) | \
> MOC_PFM(0) | \
> MOCS_SCF(0))), \
> (MOCS_ESC(0) | \
> MOSC_SCC(0) | \
> MOCS_L3_CACHEABILITY(l3)) \
> }
>
> would help clarify these even further, or at the very least being
> consistent in layout of the tables would improve readiblity and being
> able to cross-check.
Made table more consistent.
>
> > +
> > +/**
> > + * get_mocs_settings
> > + *
> > + * This function will return the values of the MOCS table that needs to
> > + * be programmed for the platform. It will return the values that need
> > + * to be programmed and if they need to be programmed.
> > + *
> > + * If the return values is false then the registers do not need programming.
> > + */
> > +bool get_mocs_settings(struct drm_device *dev,
> > + struct drm_i915_mocs_table *table) {
>
> const struct drm_i915_mocs_table *
> gen8_lookup_mocs_table(struct drm_i915_private *i915);
>
> > +/**
> > + * emit_mocs_control_table() - emit the mocs control table
> > + * @ringbuf: DRM device.
> > + * @table: The values to program into the control regs.
> > + * @reg_base: The base for the Engine that needs to be programmed.
> > + *
> > + * This function simply emits a MI_LOAD_REGISTER_IMM command for the
> > + * given table starting at the given address.
> > + *
> > + * Return: Nothing.
> > + */
> > +void emit_mocs_control_table(struct intel_ringbuffer *ringbuf,
> > + struct drm_i915_mocs_table *table,
> > + u32 reg_base)
>
> gen8_emit_mocs_table(...)
> etc
functions internal again.
>
> > +/* Defines for the tables (XXX_MOCS_0 - XXX_MOCS_63) */
> > +#define MOCS_CACHEABILITY(value) ((value & 0x03) << 0)
>
> value & mask? These macros should only be feed enums so the masking of
> the input is superfluous and indicative of a programming bug.
Superfluous yes, but it lets the coder know the layout of the field, but
may hide a programming bug but does not indicate one (other coding
standards (for safe systems) that I have used require this as it reduces
the impact of coding errors).
I have removed them.
> -Chris
>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
