Kill resource streamer
Subsystems (3)
cover Kill resource streamer
First patch is makes sense for ICL since it doesn't have resource streamer. It has been a feature that was never properly tested/used so also kill it on second patch - this is a tentative/rfc: see commit message.
Lucas De Marchi (2):
drm/i915/icl: move has_resource_streamer to GEN11_FEATURES drm/i915: kill resource streamer
drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 8 ++------ drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - include/uapi/drm/i915_drm.h | 1 + tools/include/uapi/drm/i915_drm.h | 1 + 10 files changed, 8 insertions(+), 31 deletions(-)
Signature · 6 lines
2.17.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 1/2 drm/i915/icl: move has_resource_streamer to GEN11_FEATURES
Resource streamer has been removed on GEN11 so move it to the FEATURES macro.
--- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 2 +- drivers/gpu/drm/i915/i915_pci.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/i915_gem_execbuffer.c +1 -1
@@ -2223,7 +2223,7 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (args->flags &I915_EXEC_RESOURCE_STREAMER ) {if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell, Gen8 and above\n" ); +DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" );return -EINVAL ; }if (eb.engine->id !=RCS ) {
drivers/gpu/drm/i915/i915_pci.c +1 -1
@@ -593,13 +593,13 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr = 0, \ + .has_resource_streamer = 0, \ .has_logical_ring_elsq = 1static const struct intel_device_info intel_icelake_11_info = {GEN11_FEATURES ,PLATFORM (INTEL_ICELAKE ), .is_alpha_support =1 , - .has_resource_streamer =0 , .ring_mask =RENDER_RING |BLT_RING |VEBOX_RING |BSD_RING |BSD3_RING , };
Signature · 6 lines
2.17.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 1/2] drm/i915/icl: move has_resource_streamer to GEN11_FEATURES
Quoted 4 lines · Lucas De Marchi
Resource streamer has been removed on GEN11 so move it to the FEATURES macro.
Signed-off-by Lucas De Marchi <[email protected]>
Daniele
Quoted 28 lines
--- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 2 +- drivers/gpu/drm/i915/i915_pci.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)drivers/gpu/drm/i915/i915_gem_execbuffer.c +1
-1@@ -2223,7 +2223,7 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (args->flags &I915_EXEC_RESOURCE_STREAMER ) {if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell, Gen8 and above\n" ); +DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" );return -EINVAL ; }if (eb.engine->id !=RCS ) {drivers/gpu/drm/i915/i915_pci.c +1
-0@@ -593,13 +593,13 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr = 0, \ + .has_resource_streamer = 0, \
drivers/gpu/drm/i915/i915_pci.c static const struct intel_device_info intel_cannonlake_info = { · line 605
Quoted 10 lines
drivers/gpu/drm/i915/i915_pci.c +0
-1.has_logical_ring_elsq = 1 static const struct intel_device_info intel_icelake_11_info = {GEN11_FEATURES ,PLATFORM (INTEL_ICELAKE ), .is_alpha_support =1 , - .has_resource_streamer =0 , .ring_mask =RENDER_RING |BLT_RING |VEBOX_RING |BSD_RING |BSD3_RING , };
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 2/2 drm/i915: kill resource streamer
In reply to Lucas De MarchiAfter disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep CTX_CTRL_RS_CTX_ENABLE on some platforms - Daniele mentioned there's possible performance regression in gem_latency if this bit is not set. On this patch, I'm just removing it in order to get proper numbers.
--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 8 ++------ drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - include/uapi/drm/i915_drm.h | 1 + tools/include/uapi/drm/i915_drm.h | 1 + 10 files changed, 8 insertions(+), 31 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +1 -1
@@ -370,7 +370,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU : value =HAS_POOLED_EU (dev_priv);
drivers/gpu/drm/i915/i915_drv.h +0 -2
@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submission()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)->info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define INTEL_PCH_DEVICE_ID_MASK 0xff80
drivers/gpu/drm/i915/i915_gem_execbuffer.c +2 -13
@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" ); -return -EINVAL ; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } +if (args->flags &I915_EXEC_RESOURCE_STREAMER ) +DRM_DEBUG ("RS is removed from all Gens: ignoring.\n" );if (args->flags &I915_EXEC_FENCE_IN ) { in_fence =sync_file_get_fence (lower_32_bits (args->rsvd2));
drivers/gpu/drm/i915/i915_pci.c +0 -4
@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p =0 /* RC6p removed-by HSW */ , \ .has_runtime_pm = 1
@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1, .has_gmch_display = 1,
@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \ .has_logical_ring_contexts = 1, \
@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr = 0, \ - .has_resource_streamer = 0, \ .has_logical_ring_elsq = 1static const struct intel_device_info intel_icelake_11_info = {
drivers/gpu/drm/i915/intel_device_info.h +0 -1
@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \
drivers/gpu/drm/i915/intel_lrc.c +2 -6
@@ -2138,8 +2138,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER :0 ); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset); *cs++ =upper_32_bits (offset);
@@ -2657,10 +2656,7 @@ static void execlists_init_reg_state(u32 *regs,CTX_REG (regs,CTX_CONTEXT_CONTROL ,RING_CONTEXT_CONTROL (engine),_MASKED_BIT_DISABLE (CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | -CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 ))); +CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );
drivers/gpu/drm/i915/intel_ringbuffer.c +1 -3
@@ -1979,9 +1979,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ =MI_BATCH_BUFFER_START | (dispatch_flags &I915_DISPATCH_SECURE ? -0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW ) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); +0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW );/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);
drivers/gpu/drm/i915/intel_ringbuffer.h +0 -1
@@ -472,7 +472,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;
include/uapi/drm/i915_drm.h +1 -0
@@ -449,6 +449,7 @@ typedef struct drm_i915_irq_wait {#define I915_PARAM_SUBSLICE_TOTAL 33#define I915_PARAM_EU_TOTAL 34#define I915_PARAM_HAS_GPU_RESET 35 +/* Deprecated: do not use */ #define I915_PARAM_HAS_RESOURCE_STREAMER 36#define I915_PARAM_HAS_EXEC_SOFTPIN 37#define I915_PARAM_HAS_POOLED_EU 38
tools/include/uapi/drm/i915_drm.h +1 -0
@@ -449,6 +449,7 @@ typedef struct drm_i915_irq_wait {#define I915_PARAM_SUBSLICE_TOTAL 33#define I915_PARAM_EU_TOTAL 34#define I915_PARAM_HAS_GPU_RESET 35 +/* Deprecated: do not use */ #define I915_PARAM_HAS_RESOURCE_STREAMER 36#define I915_PARAM_HAS_EXEC_SOFTPIN 37#define I915_PARAM_HAS_POOLED_EU 38
Signature · 6 lines
2.17.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 2/2] drm/i915: kill resource streamer
Quoting Lucas De Marchi (2018-07-10 01:06:58)
Quoted 13 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +0
-5@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" ); -return -EINVAL ; - }
drivers/gpu/drm/i915/i915_gem_execbuffer.c i915_gem_do_execbuffer(struct drm_device *dev, · line 2225
Quoted 10 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-8-if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } +if (args->flags &I915_EXEC_RESOURCE_STREAMER ) +DRM_DEBUG ("RS is removed from all Gens: ignoring.\n" );
return -EINVAL;
If userspace set up its batch on the assumption that RS is enabled, it will fail.
drivers/gpu/drm/i915/i915_gem_execbuffer.c static void execlists_init_reg_state(u32 *regs, · line 2661
Quoted 6 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +0
-2@@ -2657,10 +2656,7 @@ static void execlists_init_reg_state(u32 *regs, CTX_REG(regs, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(engine), _MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | -CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | - _MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH |
Bring back INHIBIT_SYNC_CTX_SWITCH.
include/uapi/drm/i915_drm.h typedef struct drm_i915_irq_wait { · line 452
Quoted 9 lines
include/uapi/drm/i915_drm.h +1
-0@@ -449,6 +449,7 @@ typedef struct drm_i915_irq_wait {#define I915_PARAM_SUBSLICE_TOTAL 33#define I915_PARAM_EU_TOTAL 34#define I915_PARAM_HAS_GPU_RESET 35 +/* Deprecated: do not use */
Don't bother. As written it implies all subsequent param are deprecated. Userspace knows simply by the query return.
include/uapi/drm/i915_drm.h typedef struct drm_i915_irq_wait { · line 455
Quoted 3 lines
include/uapi/drm/i915_drm.h +0
-0#define I915_PARAM_HAS_RESOURCE_STREAMER 36#define I915_PARAM_HAS_EXEC_SOFTPIN 37#define I915_PARAM_HAS_POOLED_EU 38
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 2/2] drm/i915: kill resource streamer
Quoted 28 lines · Chris Wilson
Quoting Lucas De Marchi (2018-07-10 01:06:58)
Quoted 13 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +0
-5@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" ); -return -EINVAL ; - }drivers/gpu/drm/i915/i915_gem_execbuffer.c i915_gem_do_execbuffer(struct drm_device *dev, · line 2225
Quoted 10 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-8-if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } +if (args->flags &I915_EXEC_RESOURCE_STREAMER ) +DRM_DEBUG ("RS is removed from all Gens: ignoring.\n" );return
-EINVAL;
Quoted 2 lines
If userspace set up its batch on the assumption that RS is enabled, it will fail.
ok,
Quoted 9 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c static void execlists_init_reg_state(u32 *regs, · line 2661
Quoted 7 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +0
-2@@ -2657,10 +2656,7 @@ static void execlists_init_reg_state(u32 *regs, CTX_REG(regs, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(engine), _MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | -CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | - _MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH |Bring back
INHIBIT_SYNC_CTX_SWITCH.
ugh, thanks for catching that.
Quoted 13 lines
include/uapi/drm/i915_drm.h typedef struct drm_i915_irq_wait { · line 452
Quoted 10 lines
include/uapi/drm/i915_drm.h +1
-0@@ -449,6 +449,7 @@ typedef struct drm_i915_irq_wait {#define I915_PARAM_SUBSLICE_TOTAL 33#define I915_PARAM_EU_TOTAL 34#define I915_PARAM_HAS_GPU_RESET 35 +/* Deprecated: do not use */ Don't bother. As written it implies all subsequent param are deprecated. Userspace knows simply by the query return.
ok.
I'll respin a new version.
Lucas De Marchi
Quoted 8 lines
include/uapi/drm/i915_drm.h typedef struct drm_i915_irq_wait { · line 455
Quoted 4 lines
include/uapi/drm/i915_drm.h +0
-0#define I915_PARAM_HAS_RESOURCE_STREAMER 36#define I915_PARAM_HAS_EXEC_SOFTPIN 37#define I915_PARAM_HAS_POOLED_EU 38_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH v2 1/1 drm/i915: kill resource streamer
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep CTX_CTRL_RS_CTX_ENABLE on some platforms - Daniele mentioned there's possible performance regression in gem_latency if this bit is not set. On this patch, I'm just removing it in order to get proper numbers.
v2: - re-add the inadvertent removal of CTX_CTRL_INHIBIT_SYN_CTX_SWITCH
- don't bother trying to document removed params on uapi header:
applications should know that from the query.
(from Chris)
--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 7 ++----- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - 8 files changed, 6 insertions(+), 30 deletions(-)
drivers/gpu/drm/i915/i915_drv.c +1 -1
@@ -373,7 +373,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU : value =HAS_POOLED_EU (dev_priv);
drivers/gpu/drm/i915/i915_drv.h +0 -2
@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submission()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)->info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define INTEL_PCH_DEVICE_ID_MASK 0xff80
drivers/gpu/drm/i915/i915_gem_execbuffer.c +2 -13
@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" ); -return -EINVAL ; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } +if (args->flags &I915_EXEC_RESOURCE_STREAMER ) +return -EINVAL ;if (args->flags &I915_EXEC_FENCE_IN ) { in_fence =sync_file_get_fence (lower_32_bits (args->rsvd2));
drivers/gpu/drm/i915/i915_pci.c +0 -4
@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p =0 /* RC6p removed-by HSW */ , \ .has_runtime_pm = 1
@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1, .has_gmch_display = 1,
@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \ .has_logical_ring_contexts = 1, \
@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr = 0, \ - .has_resource_streamer = 0, \ .has_logical_ring_elsq = 1static const struct intel_device_info intel_icelake_11_info = {
drivers/gpu/drm/i915/intel_device_info.h +0 -1
@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \
drivers/gpu/drm/i915/intel_lrc.c +2 -5
@@ -2127,8 +2127,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER :0 ); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset); *cs++ =upper_32_bits (offset);
@@ -2647,9 +2646,7 @@ static void execlists_init_reg_state(u32 *regs,CTX_REG (regs,CTX_CONTEXT_CONTROL ,RING_CONTEXT_CONTROL (engine),_MASKED_BIT_DISABLE (CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 ))); +_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH ));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );
drivers/gpu/drm/i915/intel_ringbuffer.c +1 -3
@@ -1977,9 +1977,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ =MI_BATCH_BUFFER_START | (dispatch_flags &I915_DISPATCH_SECURE ? -0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW ) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); +0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW );/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);
drivers/gpu/drm/i915/intel_ringbuffer.h +0 -1
@@ -467,7 +467,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;
Signature · 6 lines
2.17.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v2] drm/i915: kill resource streamer
Quoted 9 lines · Lucas De Marchi
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep
CTX_CTRL_RS_CTX_ENABLEon some platforms - Daniele mentioned there's possible performance regression ingem_latencyif this bit is not set. On this patch, I'm just removing it in order to get proper numbers.
I've re-run gem_latency on the same machine I saw the regression on with and without this patch applied. I've tried both with 2 and 10 contexts and the regression I saw with a similar patch a few months back seems to not be happening. The average numbers with the patch applied came out very slightly worse (10 iterations of 60s, ~1% extra latency average) but I'm assuming that's within the noise (what I saw at the time was ~20% latency increase).
Daniele
Quoted 164 lines
v2: - re-add the inadvertent removal of
CTX_CTRL_INHIBIT_SYN_CTX_SWITCH
- don't bother trying to document removed params on uapi header:
applications should know that from the query.
(from Chris)Signed-off-by Lucas De Marchi <[email protected]>--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 7 ++----- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - 8 files changed, 6 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/i915_drv.c +1
-1@@ -373,7 +373,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU : value =HAS_POOLED_EU (dev_priv);drivers/gpu/drm/i915/i915_drv.h +0
-2@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submission()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)->info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define INTEL_PCH_DEVICE_ID_MASK 0xff80drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-13@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" ); -return -EINVAL ; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } +if (args->flags &I915_EXEC_RESOURCE_STREAMER ) +return -EINVAL ;if (args->flags &I915_EXEC_FENCE_IN ) { in_fence =sync_file_get_fence (lower_32_bits (args->rsvd2));drivers/gpu/drm/i915/i915_pci.c +0
-4@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p =0 /* RC6p removed-by HSW */ , \ .has_runtime_pm = 1@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1, .has_gmch_display = 1,@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \ .has_logical_ring_contexts = 1, \@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr = 0, \ - .has_resource_streamer = 0, \ .has_logical_ring_elsq = 1static const struct intel_device_info intel_icelake_11_info = {drivers/gpu/drm/i915/intel_device_info.h +0
-1@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \drivers/gpu/drm/i915/intel_lrc.c +2
-5@@ -2127,8 +2127,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER :0 ); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset); *cs++ =upper_32_bits (offset);@@ -2647,9 +2646,7 @@ static void execlists_init_reg_state(u32 *regs,CTX_REG (regs,CTX_CONTEXT_CONTROL ,RING_CONTEXT_CONTROL (engine),_MASKED_BIT_DISABLE (CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 ))); +_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH ));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );drivers/gpu/drm/i915/intel_ringbuffer.c +1
-3@@ -1977,9 +1977,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ =MI_BATCH_BUFFER_START | (dispatch_flags &I915_DISPATCH_SECURE ? -0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW ) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); +0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW );/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);drivers/gpu/drm/i915/intel_ringbuffer.h +0
-0
drivers/gpu/drm/i915/intel_ringbuffer.h struct intel_engine_cs { · line 472
Quoted 7 lines
drivers/gpu/drm/i915/intel_ringbuffer.h +0
-1@@ -467,7 +467,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Rodrigo Vivi
maintainerRe: [PATCH v2] drm/i915: kill resource streamer
In reply to Lucas De MarchiQuoted 9 lines · Lucas De Marchi
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep
CTX_CTRL_RS_CTX_ENABLEon some platforms - Daniele mentioned there's possible performance regression ingem_latencyif this bit is not set. On this patch, I'm just removing it in order to get proper numbers.
Since I see below 3 occurrences of
- .has_resource_streamer = 1
I feel this commit message lacks of the reason why we are removing this features for the platforms that already had it.
Quoted 177 lines
v2: - re-add the inadvertent removal of
CTX_CTRL_INHIBIT_SYN_CTX_SWITCH
- don't bother trying to document removed params on uapi header:
applications should know that from the query.
(from Chris)Signed-off-by Lucas De Marchi <[email protected]>--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 7 ++----- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - 8 files changed, 6 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/i915_drv.c +1
-1@@ -373,7 +373,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU : value =HAS_POOLED_EU (dev_priv);drivers/gpu/drm/i915/i915_drv.h +0
-2@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submission()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)->info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define INTEL_PCH_DEVICE_ID_MASK 0xff80drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-13@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" ); -return -EINVAL ; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } +if (args->flags &I915_EXEC_RESOURCE_STREAMER ) +return -EINVAL ;if (args->flags &I915_EXEC_FENCE_IN ) { in_fence =sync_file_get_fence (lower_32_bits (args->rsvd2));drivers/gpu/drm/i915/i915_pci.c +0
-4@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p =0 /* RC6p removed-by HSW */ , \ .has_runtime_pm = 1@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1, .has_gmch_display = 1,@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \ .has_logical_ring_contexts = 1, \@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr = 0, \ - .has_resource_streamer = 0, \ .has_logical_ring_elsq = 1static const struct intel_device_info intel_icelake_11_info = {drivers/gpu/drm/i915/intel_device_info.h +0
-1@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \drivers/gpu/drm/i915/intel_lrc.c +2
-5@@ -2127,8 +2127,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER :0 ); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset); *cs++ =upper_32_bits (offset);@@ -2647,9 +2646,7 @@ static void execlists_init_reg_state(u32 *regs,CTX_REG (regs,CTX_CONTEXT_CONTROL ,RING_CONTEXT_CONTROL (engine),_MASKED_BIT_DISABLE (CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 ))); +_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH ));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );drivers/gpu/drm/i915/intel_ringbuffer.c +1
-3@@ -1977,9 +1977,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ =MI_BATCH_BUFFER_START | (dispatch_flags &I915_DISPATCH_SECURE ? -0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW ) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); +0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW );/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);drivers/gpu/drm/i915/intel_ringbuffer.h +0
-1@@ -467,7 +467,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;Signature · 6 lines
2.17.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v2] drm/i915: kill resource streamer
Quoted 14 lines · Rodrigo Vivi
Quoted 9 lines · Lucas De Marchi
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep
CTX_CTRL_RS_CTX_ENABLEon some platforms - Daniele mentioned there's possible performance regression ingem_latencyif this bit is not set. On this patch, I'm just removing it in order to get proper numbers.Since I see below 3 occurrences of
(quoted diff) +0
-1- .has_resource_streamer = 1
Quoted 2 lines
I feel this commit message lacks of the reason why we are removing this features for the platforms that already had it.
/me confused. Isn't the first paragraph in the commit message exactly that?
Lucas De Marchi
Quoted 192 lines
Quoted 192 lines
v2: - re-add the inadvertent removal of
CTX_CTRL_INHIBIT_SYN_CTX_SWITCH
- don't bother trying to document removed params on uapi header:
applications should know that from the query.
(from Chris)Signed-off-by Lucas De Marchi <[email protected]>--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 7 ++----- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - 8 files changed, 6 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/i915_drv.c +1
-1@@ -373,7 +373,7 @@ static int i915_getparam_ioctl(struct drm_device *dev,void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU :drivers/gpu/drm/i915/i915_drv.c +0
-0value =HAS_POOLED_EU (dev_priv);drivers/gpu/drm/i915/i915_drv.h +0
-2@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submis sion()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)- >info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define
INTEL_PCH_DEVICE_ID_MASK0xff80drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-13@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { - DRM_DEBUG("RS is only allowed for Haswell and Gen8 - Gen10\n"); - return -EINVAL; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } + if (args->flags & I915_EXEC_RESOURCE_STREAMER) + return -EINVAL;if (args->flags &I915_EXEC_FENCE_IN ) {
in_fence= sync_file_get_fence(lower_32_bits(args-Quoted 1 line
rsvd2));
drivers/gpu/drm/i915/i915_pci.c +0
-1@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p = 0 /* RC6p removed-by HSW */, \drivers/gpu/drm/i915/i915_pci.c +0
-1.has_runtime_pm = 1@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1,drivers/gpu/drm/i915/i915_pci.c +0
-1.has_gmch_display = 1,@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \drivers/gpu/drm/i915/i915_pci.c +0
-1.has_logical_ring_contexts = 1, \@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr =0 , \ - .has_resource_streamer =0 , \ .has_logical_ring_elsq = 1static const struct
intel_device_infointel_icelake_11_info= {drivers/gpu/drm/i915/intel_device_info.h +0
-1@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \drivers/gpu/drm/i915/intel_lrc.c +1
-2@@ -2127,8 +2127,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER : 0); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset);drivers/gpu/drm/i915/intel_lrc.c +1
-3*cs++ =upper_32_bits (offset);@@ -2647,9 +2646,7 @@ static void execlists_init_reg_state(u32 *regs,CTX_REG (regs,CTX_CONTEXT_CONTROL ,RING_CONTEXT_CONTROL (engine),_MASKED_BIT_DISABLE (CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 ))); +_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH ));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );drivers/gpu/drm/i915/intel_ringbuffer.c +0
-0@@ -1977,9 +1977,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ = MI_BATCH_BUFFER_START | (dispatch_flags &
I915_DISPATCH_SECURE?drivers/gpu/drm/i915/intel_ringbuffer.c +1
-3- 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); + 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW);/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);drivers/gpu/drm/i915/intel_ringbuffer.h +0
-1@@ -467,7 +467,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;
Quoted 7 lines
Quoted 7 lines
Signature · 6 lines
2.17.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Rodrigo Vivi
maintainerRe: [PATCH v2] drm/i915: kill resource streamer
Quoted 12 lines · Lucas
Quoted 11 lines · Rodrigo Vivi
Quoted 9 lines · Lucas De Marchi
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep
CTX_CTRL_RS_CTX_ENABLEon some platforms - Daniele mentioned there's possible performance regression ingem_latencyif this bit is not set. On this patch, I'm just removing it in order to get proper numbers.
Quoted 7 lines
Quoted 5 lines
Since I see below 3 occurrences of
(quoted diff) +0
-1- .has_resource_streamer = 1I feel this commit message lacks of the reason why we are removing this features for the platforms that already had it.
/me confused. Isn't the first paragraph in the commit message exactly that?
/me confused now... For me the first paragraph tells that after removing from ICL mesa might want to use it on ICL.
second paragraph tells that you are removing the support but keeping the uapi.
what I couldn't find is the explanation why exactly you are removing this feature. Is this useless? No body using it? It is buggy? It is deprecated? why?
Quoted 198 lines
Lucas De Marchi
Quoted 195 lines
Quoted 195 lines
v2: - re-add the inadvertent removal of
CTX_CTRL_INHIBIT_SYN_CTX_SWITCH
- don't bother trying to document removed params on uapi header:
applications should know that from the query.
(from Chris)Signed-off-by Lucas De Marchi <[email protected]>--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 7 ++----- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - 8 files changed, 6 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/i915_drv.c +1
-1@@ -373,7 +373,7 @@ static int i915_getparam_ioctl(struct drm_device *dev,void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU :drivers/gpu/drm/i915/i915_drv.c +0
-0value =HAS_POOLED_EU (dev_priv);drivers/gpu/drm/i915/i915_drv.h +0
-2@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submis sion()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)- >info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define
INTEL_PCH_DEVICE_ID_MASK0xff80drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-13@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { - DRM_DEBUG("RS is only allowed for Haswell and Gen8 - Gen10\n"); - return -EINVAL; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } + if (args->flags & I915_EXEC_RESOURCE_STREAMER) + return -EINVAL;if (args->flags &I915_EXEC_FENCE_IN ) {
in_fence= sync_file_get_fence(lower_32_bits(args-Quoted 1 line
rsvd2));
drivers/gpu/drm/i915/i915_pci.c +0
-1@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p = 0 /* RC6p removed-by HSW */, \drivers/gpu/drm/i915/i915_pci.c +0
-1.has_runtime_pm = 1@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1,drivers/gpu/drm/i915/i915_pci.c +0
-1.has_gmch_display = 1,@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \drivers/gpu/drm/i915/i915_pci.c +0
-1.has_logical_ring_contexts = 1, \@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr =0 , \ - .has_resource_streamer =0 , \ .has_logical_ring_elsq = 1static const struct
intel_device_infointel_icelake_11_info= {drivers/gpu/drm/i915/intel_device_info.h +0
-1@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \drivers/gpu/drm/i915/intel_lrc.c +1
-2@@ -2127,8 +2127,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER : 0); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset);drivers/gpu/drm/i915/intel_lrc.c +1
-3*cs++ =upper_32_bits (offset);@@ -2647,9 +2646,7 @@ static void execlists_init_reg_state(u32 *regs,CTX_REG (regs,CTX_CONTEXT_CONTROL ,RING_CONTEXT_CONTROL (engine),_MASKED_BIT_DISABLE (CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 ))); +_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH ));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );drivers/gpu/drm/i915/intel_ringbuffer.c +0
-0@@ -1977,9 +1977,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ = MI_BATCH_BUFFER_START | (dispatch_flags &
I915_DISPATCH_SECURE?drivers/gpu/drm/i915/intel_ringbuffer.c +1
-3- 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); + 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW);/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);drivers/gpu/drm/i915/intel_ringbuffer.h +0
-1@@ -467,7 +467,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;Signature · 2 lines
-- 2.17.1
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v2] drm/i915: kill resource streamer
Quoted 21 lines · Rodrigo Vivi
Quoted 12 lines · Lucas
Quoted 11 lines · Rodrigo Vivi
Quoted 9 lines · Lucas De Marchi
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep
CTX_CTRL_RS_CTX_ENABLEon some platforms - Daniele mentioned there's possible performance regression ingem_latencyif this bit is not set. On this patch, I'm just removing it in order to get proper numbers.Quoted 7 lines
Quoted 5 lines
Since I see below 3 occurrences of
(quoted diff) +0
-1- .has_resource_streamer = 1I feel this commit message lacks of the reason why we are removing this features for the platforms that already had it.
/me confused. Isn't the first paragraph in the commit message exactly that?
Quoted 7 lines
/me confused now... For me the first paragraph tells that after removing from ICL mesa might want to use it on ICL.
second paragraph tells that you are removing the support but keeping the uapi.
what I couldn't find is the explanation why exactly you are removing this feature. Is this useless? No body using it? It is buggy? It is deprecated? why?
I guess a small change should fix the meaning of the first para:
"... there have been some experimental patches for mesa to use it years ago, but nothing ever landed or shipped because there was no performance improvement."
These experimental patches long stopped existing :-)
Also note that the uapi doesn't stay around, only the #defines (we can't ever reuse them for anything else than RS, so need to keep them reserved).
Cheers, Daniel
Quoted 207 lines
Quoted 207 lines
Lucas De Marchi
Quoted 192 lines
Quoted 192 lines
v2: - re-add the inadvertent removal of
CTX_CTRL_INHIBIT_SYN_CTX_SWITCH
- don't bother trying to document removed params on uapi header:
applications should know that from the query.
(from Chris)Signed-off-by Lucas De Marchi <[email protected]>--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 7 ++----- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - 8 files changed, 6 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/i915_drv.c +1
-1@@ -373,7 +373,7 @@ static int i915_getparam_ioctl(struct drm_device *dev,void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU :drivers/gpu/drm/i915/i915_drv.c +0
-0value =HAS_POOLED_EU (dev_priv);drivers/gpu/drm/i915/i915_drv.h +0
-2@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submis sion()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)- >info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define
INTEL_PCH_DEVICE_ID_MASK0xff80drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-13@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { - DRM_DEBUG("RS is only allowed for Haswell and Gen8 - Gen10\n"); - return -EINVAL; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } + if (args->flags & I915_EXEC_RESOURCE_STREAMER) + return -EINVAL;if (args->flags &I915_EXEC_FENCE_IN ) {
in_fence= sync_file_get_fence(lower_32_bits(args-Quoted 1 line
rsvd2));
drivers/gpu/drm/i915/i915_pci.c +0
-1@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p = 0 /* RC6p removed-by HSW */, \drivers/gpu/drm/i915/i915_pci.c +0
-1.has_runtime_pm = 1@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1,drivers/gpu/drm/i915/i915_pci.c +0
-1.has_gmch_display = 1,@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \drivers/gpu/drm/i915/i915_pci.c +0
-1.has_logical_ring_contexts = 1, \@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr =0 , \ - .has_resource_streamer =0 , \ .has_logical_ring_elsq = 1static const struct
intel_device_infointel_icelake_11_info= {drivers/gpu/drm/i915/intel_device_info.h +0
-1@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \drivers/gpu/drm/i915/intel_lrc.c +1
-2@@ -2127,8 +2127,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER : 0); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset);drivers/gpu/drm/i915/intel_lrc.c +1
-3*cs++ =upper_32_bits (offset);@@ -2647,9 +2646,7 @@ static void execlists_init_reg_state(u32 *regs,CTX_REG (regs,CTX_CONTEXT_CONTROL ,RING_CONTEXT_CONTROL (engine),_MASKED_BIT_DISABLE (CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 ))); +_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH ));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );drivers/gpu/drm/i915/intel_ringbuffer.c +0
-0@@ -1977,9 +1977,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ = MI_BATCH_BUFFER_START | (dispatch_flags &
I915_DISPATCH_SECURE?drivers/gpu/drm/i915/intel_ringbuffer.c +1
-3- 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); + 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW);/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);drivers/gpu/drm/i915/intel_ringbuffer.h +0
-1@@ -467,7 +467,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;Quoted 7 lines
Quoted 7 lines
Signature · 6 lines
2.17.1 _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v2] drm/i915: kill resource streamer
Quoted 17 lines · Daniel Vetter
Quoted 16 lines · Rodrigo Vivi
Quoted 15 lines · Lucas
Quoted 14 lines · Rodrigo Vivi
Quoted 9 lines · Lucas De Marchi
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep
CTX_CTRL_RS_CTX_ENABLEon some platforms - Daniele mentioned there's possible performance regression ingem_latencyif this bit is not set. On this patch, I'm just removing it in order to get proper numbers.Since I see below 3 occurrences of
(quoted diff) +0
-1- .has_resource_streamer = 1
Quoted 7 lines
Quoted 7 lines
Quoted 4 lines
Quoted 2 lines
I feel this commit message lacks of the reason why we are removing this features for the platforms that already had it.
/me confused. Isn't the first paragraph in the commit message exactly that?
/me confused now... For me the first paragraph tells that after removing from ICL mesa might want to use it on ICL.
nah, the *hw* block doesn't exist on ICL, a patch to use it there would be garbage, not experimental :)
Quoted 2 lines
Quoted 2 lines
second paragraph tells that you are removing the support but keeping the uapi.
by "use it" I was referring to the feature itself.
Quoted 9 lines
Quoted 3 lines
what I couldn't find is the explanation why exactly you are removing this feature. Is this useless? No body using it? It is buggy? It is deprecated? why?
I guess a small change should fix the meaning of the first para:
"... there have been some experimental patches for mesa to use it years ago, but nothing ever landed or shipped because there was no performance improvement."
yep, it's better. And s/it/RS/ would help clear out the confusion on my badly worded paragraph.
Quoted 5 lines
These experimental patches long stopped existing :-)
Also note that the uapi doesn't stay around, only the #defines (we can't ever reuse them for anything else than RS, so need to keep them reserved).
I think I got it right on this one saying "keeping the uapi *defines* around for compatibility.
Lucas De Marchi
Quoted 202 lines
Cheers, Daniel
Quoted 199 lines
Quoted 199 lines
Lucas De Marchi
Quoted 195 lines
Quoted 195 lines
v2: - re-add the inadvertent removal of
CTX_CTRL_INHIBIT_SYN_CTX_SWITCH
- don't bother trying to document removed params on uapi header:
applications should know that from the query.
(from Chris)Signed-off-by Lucas De Marchi <[email protected]>--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 7 ++----- drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - 8 files changed, 6 insertions(+), 30 deletions(-)drivers/gpu/drm/i915/i915_drv.c +1
-1@@ -373,7 +373,7 @@ static int i915_getparam_ioctl(struct drm_device *dev,void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU :drivers/gpu/drm/i915/i915_drv.c +0
-0value =HAS_POOLED_EU (dev_priv);drivers/gpu/drm/i915/i915_drv.h +0
-2@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submis sion()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)- >info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define
INTEL_PCH_DEVICE_ID_MASK0xff80drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-13@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { - DRM_DEBUG("RS is only allowed for Haswell and Gen8 - Gen10\n"); - return -EINVAL; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } + if (args->flags & I915_EXEC_RESOURCE_STREAMER) + return -EINVAL;if (args->flags &I915_EXEC_FENCE_IN ) {
in_fence= sync_file_get_fence(lower_32_bits(args-Quoted 1 line
rsvd2));
drivers/gpu/drm/i915/i915_pci.c +0
-1@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p = 0 /* RC6p removed-by HSW */, \drivers/gpu/drm/i915/i915_pci.c +0
-1.has_runtime_pm = 1@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1,drivers/gpu/drm/i915/i915_pci.c +0
-1.has_gmch_display = 1,@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \drivers/gpu/drm/i915/i915_pci.c +0
-1.has_logical_ring_contexts = 1, \@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr =0 , \ - .has_resource_streamer =0 , \ .has_logical_ring_elsq = 1static const struct
intel_device_infointel_icelake_11_info= {drivers/gpu/drm/i915/intel_device_info.h +0
-1@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \drivers/gpu/drm/i915/intel_lrc.c +1
-2@@ -2127,8 +2127,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER : 0); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset);drivers/gpu/drm/i915/intel_lrc.c +1
-3*cs++ =upper_32_bits (offset);@@ -2647,9 +2646,7 @@ static void execlists_init_reg_state(u32 *regs,CTX_REG (regs,CTX_CONTEXT_CONTROL ,RING_CONTEXT_CONTROL (engine),_MASKED_BIT_DISABLE (CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 ))); +_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH ));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );drivers/gpu/drm/i915/intel_ringbuffer.c +0
-0@@ -1977,9 +1977,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ = MI_BATCH_BUFFER_START | (dispatch_flags &
I915_DISPATCH_SECURE?drivers/gpu/drm/i915/intel_ringbuffer.c +1
-3- 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); + 0 : MI_BATCH_PPGTT_HSW | MI_BATCH_NON_SECURE_HSW);/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);drivers/gpu/drm/i915/intel_ringbuffer.h +0
-1@@ -467,7 +467,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;Signature · 2 lines
-- 2.17.1
Quoted 13 lines
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfxSignature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 2/2] drm/i915: kill resource streamer
In reply to Lucas De MarchiQuoted 63 lines · Lucas De Marchi
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep
CTX_CTRL_RS_CTX_ENABLEon some platforms - Daniele mentioned there's possible performance regression ingem_latencyif this bit is not set. On this patch, I'm just removing it in order to get proper numbers.Signed-off-by Lucas De Marchi <[email protected]>--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 8 ++------ drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - include/uapi/drm/i915_drm.h | 1 + tools/include/uapi/drm/i915_drm.h | 1 + 10 files changed, 8 insertions(+), 31 deletions(-)drivers/gpu/drm/i915/i915_drv.c +1
-1@@ -370,7 +370,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU : value =HAS_POOLED_EU (dev_priv);drivers/gpu/drm/i915/i915_drv.h +0
-2@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submission()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)->info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define INTEL_PCH_DEVICE_ID_MASK 0xff80drivers/gpu/drm/i915/i915_gem_execbuffer.c +0
-5@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" ); -return -EINVAL ; - }
drivers/gpu/drm/i915/i915_gem_execbuffer.c i915_gem_do_execbuffer(struct drm_device *dev, · line 2225
Quoted 10 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-8-if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ; - } - - eb.batch_flags |=I915_DISPATCH_RS ; - } +if (args->flags &I915_EXEC_RESOURCE_STREAMER ) +DRM_DEBUG ("RS is removed from all Gens: ignoring.\n" );
Still missing the return -EINVAL here. Needs an igt update I think.
Quoted 66 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +0
-0if (args->flags &I915_EXEC_FENCE_IN ) { in_fence =sync_file_get_fence (lower_32_bits (args->rsvd2));drivers/gpu/drm/i915/i915_pci.c +0
-4@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p =0 /* RC6p removed-by HSW */ , \ .has_runtime_pm = 1@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1, .has_gmch_display = 1,@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \ .has_logical_ring_contexts = 1, \@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr = 0, \ - .has_resource_streamer = 0, \ .has_logical_ring_elsq = 1static const struct intel_device_info intel_icelake_11_info = {drivers/gpu/drm/i915/intel_device_info.h +0
-1@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \drivers/gpu/drm/i915/intel_lrc.c +1
-2@@ -2138,8 +2138,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER :0 ); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset); *cs++ =upper_32_bits (offset);
drivers/gpu/drm/i915/intel_lrc.c static void execlists_init_reg_state(u32 *regs, · line 2663
Quoted 7 lines
drivers/gpu/drm/i915/intel_lrc.c +0
-4@@ -2657,10 +2656,7 @@ static void execlists_init_reg_state(u32 *regs, CTX_REG(regs, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(engine), _MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | -CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 )));
I'd hard-code this to disable the RS instead of removing it outright.
With these 2 nits addressed and the commit message augmented as I propose in my other reply:
Quoted 58 lines
drivers/gpu/drm/i915/intel_lrc.c +1
-0+ CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );drivers/gpu/drm/i915/intel_ringbuffer.c +1
-3@@ -1979,9 +1979,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ =MI_BATCH_BUFFER_START | (dispatch_flags &I915_DISPATCH_SECURE ? -0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW ) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); +0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW );/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);drivers/gpu/drm/i915/intel_ringbuffer.h +0
-1@@ -472,7 +472,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;include/uapi/drm/i915_drm.h +1
-0@@ -449,6 +449,7 @@ typedef struct drm_i915_irq_wait {#define I915_PARAM_SUBSLICE_TOTAL 33#define I915_PARAM_EU_TOTAL 34#define I915_PARAM_HAS_GPU_RESET 35 +/* Deprecated: do not use */ #define I915_PARAM_HAS_RESOURCE_STREAMER 36#define I915_PARAM_HAS_EXEC_SOFTPIN 37#define I915_PARAM_HAS_POOLED_EU 38tools/include/uapi/drm/i915_drm.h +1
-0@@ -449,6 +449,7 @@ typedef struct drm_i915_irq_wait {#define I915_PARAM_SUBSLICE_TOTAL 33#define I915_PARAM_EU_TOTAL 34#define I915_PARAM_HAS_GPU_RESET 35 +/* Deprecated: do not use */ #define I915_PARAM_HAS_RESOURCE_STREAMER 36#define I915_PARAM_HAS_EXEC_SOFTPIN 37#define I915_PARAM_HAS_POOLED_EU 38Signature · 1 line
2.17.1
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 2/2] drm/i915: kill resource streamer
Quoted 68 lines · Daniel Vetter
Quoted 67 lines · Lucas De Marchi
After disabling resource streamer on ICL (due to it actually not existing there), I got feedback that there have been some experimental patches for mesa to use it, but nothing ever landed nor shipped.
This is a tentative to remove it from kernel keeping the uapi defines around for compatibility. We may need to keep
CTX_CTRL_RS_CTX_ENABLEon some platforms - Daniele mentioned there's possible performance regression ingem_latencyif this bit is not set. On this patch, I'm just removing it in order to get proper numbers.Signed-off-by Lucas De Marchi <[email protected]>--- drivers/gpu/drm/i915/i915_drv.c | 2 +- drivers/gpu/drm/i915/i915_drv.h | 2 -- drivers/gpu/drm/i915/i915_gem_execbuffer.c | 15 ++------------- drivers/gpu/drm/i915/i915_pci.c | 4 ---- drivers/gpu/drm/i915/intel_device_info.h | 1 - drivers/gpu/drm/i915/intel_lrc.c | 8 ++------ drivers/gpu/drm/i915/intel_ringbuffer.c | 4 +--- drivers/gpu/drm/i915/intel_ringbuffer.h | 1 - include/uapi/drm/i915_drm.h | 1 + tools/include/uapi/drm/i915_drm.h | 1 + 10 files changed, 8 insertions(+), 31 deletions(-)drivers/gpu/drm/i915/i915_drv.c +1
-1@@ -370,7 +370,7 @@ static int i915_getparam_ioctl(struct drm_device *dev, void *data, value =2 ;break ;case I915_PARAM_HAS_RESOURCE_STREAMER : - value =HAS_RESOURCE_STREAMER (dev_priv); + value =0 ;break ;case I915_PARAM_HAS_POOLED_EU : value =HAS_POOLED_EU (dev_priv);drivers/gpu/drm/i915/i915_drv.h +0
-2@@ -2609,8 +2609,6 @@ intel_info(const struct drm_i915_private *dev_priv)#define USES_GUC_SUBMISSION (dev_priv) intel_uc_is_using_guc_submission()#define USES_HUC (dev_priv) intel_uc_is_using_huc() -#define HAS_RESOURCE_STREAMER (dev_priv) ((dev_priv)->info.has_resource_streamer) -#define HAS_POOLED_EU (dev_priv) ((dev_priv)->info.has_pooled_eu)#define INTEL_PCH_DEVICE_ID_MASK 0xff80drivers/gpu/drm/i915/i915_gem_execbuffer.c +0
-9@@ -2221,19 +2221,8 @@ i915_gem_do_execbuffer(struct drm_device *dev,if (!eb.engine)return -EINVAL ; -if (args->flags &I915_EXEC_RESOURCE_STREAMER ) { -if (!HAS_RESOURCE_STREAMER (eb.i915)) { -DRM_DEBUG ("RS is only allowed for Haswell and Gen8 - Gen10\n" ); -return -EINVAL ; - } -if (eb.engine->id !=RCS ) { -DRM_DEBUG ("RS is not available on %s\n" , - eb.engine->name); -return -EINVAL ;
Quoted 8 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c i915_gem_do_execbuffer(struct drm_device *dev, · line 2225
Quoted 6 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +2
-4- } - - eb.batch_flags |=I915_DISPATCH_RS ; - } +if (args->flags &I915_EXEC_RESOURCE_STREAMER ) +DRM_DEBUG ("RS is removed from all Gens: ignoring.\n" );Still missing the return
-EINVALhere. Needs an igt update I think.
that's because you are replying on v1 rather than v2. See the updated patch down the thread.
Quoted 68 lines
Quoted 68 lines
drivers/gpu/drm/i915/i915_gem_execbuffer.c +0
-0if (args->flags &I915_EXEC_FENCE_IN ) { in_fence =sync_file_get_fence (lower_32_bits (args->rsvd2));drivers/gpu/drm/i915/i915_pci.c +0
-4@@ -360,7 +360,6 @@ static const struct intel_device_info intel_valleyview_info = { .has_ddi = 1, \ .has_fpga_dbg = 1, \ .has_psr = 1, \ - .has_resource_streamer = 1, \ .has_dp_mst = 1, \ .has_rc6p =0 /* RC6p removed-by HSW */ , \ .has_runtime_pm = 1@@ -433,7 +432,6 @@ static const struct intel_device_info intel_cherryview_info = { .ring_mask = RENDER_RING | BSD_RING | BLT_RING | VEBOX_RING, .has_64bit_reloc = 1, .has_runtime_pm = 1, - .has_resource_streamer = 1, .has_rc6 = 1, .has_logical_ring_contexts = 1, .has_gmch_display = 1,@@ -506,7 +504,6 @@ static const struct intel_device_info intel_skylake_gt4_info = { .has_runtime_pm = 1, \ .has_pooled_eu = 0, \ .has_csr = 1, \ - .has_resource_streamer = 1, \ .has_rc6 = 1, \ .has_dp_mst = 1, \ .has_logical_ring_contexts = 1, \@@ -593,7 +590,6 @@ static const struct intel_device_info intel_cannonlake_info = { GEN(11), \ .ddb_size = 2048, \ .has_csr = 0, \ - .has_resource_streamer = 0, \ .has_logical_ring_elsq = 1static const struct intel_device_info intel_icelake_11_info = {drivers/gpu/drm/i915/intel_device_info.h +0
-1@@ -103,7 +103,6 @@ enum intel_platform {func (has_psr); \func (has_rc6); \func (has_rc6p); \ -func (has_resource_streamer); \func (has_runtime_pm); \func (has_snoop); \func (unfenced_needs_alignment); \drivers/gpu/drm/i915/intel_lrc.c +1
-2@@ -2138,8 +2138,7 @@ static int gen8_emit_bb_start(struct i915_request *rq,/* FIXME(BDW): Address space and security selectors. */ *cs++ =MI_BATCH_BUFFER_START_GEN8 | - (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )) | - (flags &I915_DISPATCH_RS ?MI_BATCH_RESOURCE_STREAMER :0 ); + (flags &I915_DISPATCH_SECURE ?0 :BIT (8 )); *cs++ =lower_32_bits (offset); *cs++ =upper_32_bits (offset);@@ -2657,10 +2656,7 @@ static void execlists_init_reg_state(u32 *regs,
Quoted 8 lines
drivers/gpu/drm/i915/intel_lrc.c static void execlists_init_reg_state(u32 *regs, · line 2663
Quoted 6 lines
drivers/gpu/drm/i915/intel_lrc.c +0
-4CTX_REG(regs, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(engine), _MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | -CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT ) | -_MASKED_BIT_ENABLE (CTX_CTRL_INHIBIT_SYN_CTX_SWITCH | - (HAS_RESOURCE_STREAMER (dev_priv) ? -CTX_CTRL_RS_CTX_ENABLE :0 )));I'd hard-code this to disable the RS instead of removing it outright.
_MASKED_BIT_DISABLE(CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT | CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT | CTX_CTRL_RS_CTX_ENABLE) | _MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH);
right?
Quoted 5 lines
With these 2 nits addressed and the commit message augmented as I propose in my other reply:
Reviewed-by Daniel Vetter <[email protected]>
thanks
Lucas De Marchi
Quoted 52 lines
Quoted 52 lines
drivers/gpu/drm/i915/intel_lrc.c +1
-0+ CTX_CTRL_ENGINE_CTX_SAVE_INHIBIT));CTX_REG (regs,CTX_RING_HEAD ,RING_HEAD (base),0 );CTX_REG (regs,CTX_RING_TAIL ,RING_TAIL (base),0 );CTX_REG (regs,CTX_RING_BUFFER_START ,RING_START (base),0 );drivers/gpu/drm/i915/intel_ringbuffer.c +1
-3@@ -1979,9 +1979,7 @@ hsw_emit_bb_start(struct i915_request *rq,return PTR_ERR (cs); *cs++ =MI_BATCH_BUFFER_START | (dispatch_flags &I915_DISPATCH_SECURE ? -0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW ) | - (dispatch_flags &I915_DISPATCH_RS ? -MI_BATCH_RESOURCE_STREAMER :0 ); +0 :MI_BATCH_PPGTT_HSW |MI_BATCH_NON_SECURE_HSW );/* bit0-7 is the length on GEN6+ */ *cs++ = offset;intel_ring_advance (rq, cs);drivers/gpu/drm/i915/intel_ringbuffer.h +0
-1@@ -472,7 +472,6 @@ struct intel_engine_cs { unsigned int dispatch_flags);#define I915_DISPATCH_SECURE BIT(0)#define I915_DISPATCH_PINNED BIT(1) -#define I915_DISPATCH_RS BIT(2)void (*emit_breadcrumb)(struct i915_request *rq,u32 *cs);int emit_breadcrumb_sz;include/uapi/drm/i915_drm.h +1
-0@@ -449,6 +449,7 @@ typedef struct drm_i915_irq_wait {#define I915_PARAM_SUBSLICE_TOTAL 33#define I915_PARAM_EU_TOTAL 34#define I915_PARAM_HAS_GPU_RESET 35 +/* Deprecated: do not use */ #define I915_PARAM_HAS_RESOURCE_STREAMER 36#define I915_PARAM_HAS_EXEC_SOFTPIN 37#define I915_PARAM_HAS_POOLED_EU 38tools/include/uapi/drm/i915_drm.h +0
-0@@ -449,6 +449,7 @@ typedef struct drm_i915_irq_wait {#define I915_PARAM_SUBSLICE_TOTAL 33#define I915_PARAM_EU_TOTAL 34
Quoted 13 lines
Quoted 8 lines
tools/include/uapi/drm/i915_drm.h +1
-0#define I915_PARAM_HAS_GPU_RESET 35 +/* Deprecated: do not use */ #define I915_PARAM_HAS_RESOURCE_STREAMER 36#define I915_PARAM_HAS_EXEC_SOFTPIN 37#define I915_PARAM_HAS_POOLED_EU 38Signature · 1 line
2.17.1Signature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI✗ Fi.CI.SPARSE: warning for Kill resource streamer
In reply to Lucas De Marchi== Series Details ==
== Series Details ==
Series: Kill resource streamer
URL : https://patchwork.freedesktop.org/series/46224/
State : warning
== Summary ==
$ dim sparse origin/drm-tip Commit: drm/i915/icl: move has_resource_streamer to GEN11_FEATURES Okay!
Commit: drm/i915: kill resource streamer -drivers/gpu/drm/i915/selftests/../i915_drv.h:3652:16: warning: expression using sizeof(void) +drivers/gpu/drm/i915/selftests/../i915_drv.h:3650:16: warning: expression using sizeof(void)
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI✓ Fi.CI.BAT: success for Kill resource streamer
In reply to Lucas De Marchi== Series Details ==
== Series Details ==
Series: Kill resource streamer
URL : https://patchwork.freedesktop.org/series/46224/
State : success
== Summary ==
= CI Bug Log - changes from CI_DRM_4458 -> Patchwork_9599 =
== Summary - SUCCESS ==
No regressions found.
External URL: https://patchwork.freedesktop.org/api/1.0/series/46224/revisions/1/mbox/
== Known issues ==
Here are the changes found in Patchwork_9599 that come from known issues:
=== IGT changes ===
==== Possible fixes ====
igt@gem_exec_suspend@basic-s4-devices:
fi-kbl-7500u: DMESG-WARN (fdo#107139, fdo#105128) -> PASS
igt@kms_flip@basic-flip-vs-dpms:
fi-skl-6700hq: DMESG-WARN (fdo#105998) -> PASS fdo#105128 https://bugs.freedesktop.org/show_bug.cgi?id=105128
fdo#105998 https://bugs.freedesktop.org/show_bug.cgi?id=105998
fdo#107139 https://bugs.freedesktop.org/show_bug.cgi?id=107139
== Participating hosts (45 -> 41) ==
Additional (1): fi-skl-guc
Missing (5): fi-ctg-p8600 fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-hsw-4200u
== Build changes ==
* Linux: CI_DRM_4458 -> Patchwork_9599
CI_DRM_4458: dbed4c29cc0cd0225aae60c57ac2f4a738bb5a17 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4544: 764160f214cd916ddb79408b9f28ac0ad2df40e0 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_9599: 526eac8de54b23e00a6293cb7ff185a2b1530156 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
526eac8de54b drm/i915: kill resource streamera10cc1118b3e drm/i915/icl: move has_resource_streamer to GEN11_FEATURES
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_9599/issues.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI✗ Fi.CI.IGT: failure for Kill resource streamer
In reply to Lucas De Marchi== Series Details ==
== Series Details ==
Series: Kill resource streamer
URL : https://patchwork.freedesktop.org/series/46224/
State : failure
== Summary ==
= CI Bug Log - changes from CI_DRM_4458_full -> Patchwork_9599_full =
== Summary - FAILURE ==
Serious unknown changes coming with Patchwork_9599_full absolutely need to be verified manually.
If you think the reported changes have nothing to do with the changes introduced in Patchwork_9599_full, please notify your bug team to allow them to document this new failure mode, which will reduce false positives in CI.
== Possible new issues ==
Here are the unknown changes that may have been introduced in Patchwork_9599_full:
=== IGT changes ===
==== Possible regressions ====
igt@gem_exec_params@rs-invalid-gen:
shard-snb: PASS -> FAIL
igt@gem_exec_params@rs-invalid-on-vebox-ring:
shard-hsw: PASS -> FAIL +2
igt@perf_pmu@event-wait-rcs0:
shard-glk: PASS -> FAIL +11
igt@perf_pmu@semaphore-wait-idle-bcs0:
shard-kbl: PASS -> FAIL +13
igt@perf_pmu@semaphore-wait-vcs0:
shard-apl: PASS -> FAIL +11
==== Warnings ====
igt@gem_exec_schedule@deep-vebox:
shard-kbl: SKIP -> PASS +2
igt@gem_mocs_settings@mocs-rc6-bsd1:
shard-kbl: PASS -> SKIP== Known issues ==
Here are the changes found in Patchwork_9599_full that come from known issues:
=== IGT changes ===
==== Issues hit ====
igt@drv_suspend@shrink:
shard-glk: PASS -> INCOMPLETE (fdo#103359, k.org#198133)
igt@kms_vblank@pipe-c-ts-continuation-suspend:
shard-hsw: PASS -> FAIL (fdo#104894)
igt@perf@blocking:
shard-hsw: PASS -> FAIL (fdo#102252)
==== Possible fixes ====
igt@gem_exec_big:
shard-hsw: INCOMPLETE (fdo#103540) -> PASS
igt@kms_flip@2x-dpms-vs-vblank-race-interruptible:
shard-glk: FAIL (fdo#103060) -> PASS
igt@kms_flip_tiling@flip-y-tiled:
shard-glk: FAIL (fdo#107161) -> PASS
igt@kms_setmode@basic:
shard-kbl: FAIL (fdo#99912) -> PASS fdo#102252 https://bugs.freedesktop.org/show_bug.cgi?id=102252
fdo#103060 https://bugs.freedesktop.org/show_bug.cgi?id=103060
fdo#103359 https://bugs.freedesktop.org/show_bug.cgi?id=103359
fdo#103540 https://bugs.freedesktop.org/show_bug.cgi?id=103540
fdo#104894 https://bugs.freedesktop.org/show_bug.cgi?id=104894
fdo#107161 https://bugs.freedesktop.org/show_bug.cgi?id=107161
fdo#99912 https://bugs.freedesktop.org/show_bug.cgi?id=99912
k.org#198133 https://bugzilla.kernel.org/show_bug.cgi?id=198133
== Participating hosts (5 -> 5) ==
No changes in participating hosts
== Build changes ==
* Linux: CI_DRM_4458 -> Patchwork_9599
CI_DRM_4458: dbed4c29cc0cd0225aae60c57ac2f4a738bb5a17 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4544: 764160f214cd916ddb79408b9f28ac0ad2df40e0 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_9599: 526eac8de54b23e00a6293cb7ff185a2b1530156 @ git://anongit.freedesktop.org/gfx-ci/linux
piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_9599/shards.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI✗ Fi.CI.SPARSE: warning for Kill resource streamer (rev2)
In reply to Lucas De Marchi== Series Details ==
== Series Details ==
Series: Kill resource streamer (rev2)
URL : https://patchwork.freedesktop.org/series/46224/
State : warning
== Summary ==
$ dim sparse origin/drm-tip Commit: drm/i915/icl: move has_resource_streamer to GEN11_FEATURES Okay!
Commit: drm/i915: kill resource streamer -drivers/gpu/drm/i915/selftests/../i915_drv.h:3655:16: warning: expression using sizeof(void) +drivers/gpu/drm/i915/selftests/../i915_drv.h:3653:16: warning: expression using sizeof(void)
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
Patchwork
CI✓ Fi.CI.BAT: success for Kill resource streamer (rev2)
In reply to Lucas De Marchi== Series Details ==
== Series Details ==
Series: Kill resource streamer (rev2)
URL : https://patchwork.freedesktop.org/series/46224/
State : success
== Summary ==
= CI Bug Log - changes from CI_DRM_4479 -> Patchwork_9640 =
== Summary - SUCCESS ==
No regressions found.
External URL: https://patchwork.freedesktop.org/api/1.0/series/46224/revisions/2/mbox/
== Possible new issues ==
Here are the unknown changes that may have been introduced in Patchwork_9640:
=== IGT changes ===
==== Possible regressions ====
igt@kms_pipe_crc_basic@suspend-read-crc-pipe-c:
{fi-cfl-8109u}: PASS -> INCOMPLETE== Known issues ==
Here are the changes found in Patchwork_9640 that come from known issues:
=== IGT changes ===
==== Possible fixes ====
igt@drv_module_reload@basic-reload:
fi-ilk-650: DMESG-WARN (fdo#106387) -> PASS +2
igt@kms_chamelium@hdmi-hpd-fast:
fi-kbl-7500u: FAIL (fdo#103841, fdo#102672) -> SKIP
igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b:
fi-snb-2520m: INCOMPLETE (fdo#103713) -> PASS
==== Warnings ====
igt@gem_exec_suspend@basic-s4-devices:
{fi-kbl-8809g}: INCOMPLETE (fdo#107139) -> DMESG-WARN (fdo#107139) {name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
fdo#102672 https://bugs.freedesktop.org/show_bug.cgi?id=102672
fdo#103713 https://bugs.freedesktop.org/show_bug.cgi?id=103713
fdo#103841 https://bugs.freedesktop.org/show_bug.cgi?id=103841
fdo#106387 https://bugs.freedesktop.org/show_bug.cgi?id=106387
fdo#107139 https://bugs.freedesktop.org/show_bug.cgi?id=107139
== Participating hosts (46 -> 42) ==
Additional (1): fi-byt-j1900
Missing (5): fi-ctg-p8600 fi-ilk-m540 fi-byt-squawks fi-bsw-cyan fi-hsw-4200u
== Build changes ==
* Linux: CI_DRM_4479 -> Patchwork_9640
CI_DRM_4479: c9c54a1c37adefd414d51ed919aab3f94794ee35 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_4553: 9d25d62bb6ff62694e5226b02d79075eab304402 @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_9640: 2997ade936fa14aa75840c30d62237891853f586 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
2997ade936fa drm/i915: kill resource streamera668fe24a019 drm/i915/icl: move has_resource_streamer to GEN11_FEATURES
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_9640/issues.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx