libdrm: Move intel_atomic.h to libdrm core for sharing.
PATCH 1/2 libdrm: Move intel_atomic.h to libdrm core for sharing.
intel_atomic.h includes very usefull atomic operations for lock free parrallel access of variables. Moving these to core libdrm for code sharing with radeon.
--- Makefile.am | 2 +- configure.ac | 2 +- intel/intel_atomic.h | 56 +----------------------------- xf86atomic.h | 93 ++++++++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 96 insertions(+), 57 deletions(-) create mode 100644 xf86atomic.h
Makefile.am +1 -1
@@ -59,7 +59,7 @@ libdrm_la_SOURCES = \ libdrm_lists.h libdrmincludedir = ${includedir} -libdrminclude_HEADERS = xf86drm.h xf86drmMode.h +libdrminclude_HEADERS = xf86drm.h xf86drmMode.h xf86atomic.h EXTRA_DIST = libdrm.pc.in include/drm/*
configure.ac +1 -1
@@ -198,7 +198,7 @@ if test "x$INTEL" != "xno"; then ]) if test "x$drm_cv_atomic_primitives" = xIntel; then - AC_DEFINE(HAVE_INTEL_ATOMIC_PRIMITIVES, 1, + AC_DEFINE(HAVE_LIBDRM_ATOMIC_PRIMITIVES, 1, [Enable if your compiler supports the Intel __sync_* atomic primitives]) fi if test "x$drm_cv_atomic_primitives" = "xlibatomic-ops"; then
intel/intel_atomic.h +1 -55
61 lines not on this page · raw message
xf86atomic.h new file +93 -0
93 lines not on this page · raw message
Signature · 1 line
1.6.3.3
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel
PATCH 2/2 libdrm_radeon: Optimize reloc writing to do less looping.
Bit has table will be first checked from BO if we can quarentee this BO is not in this cs already.
To quarentee that there is no other cs with same id number of CS that can have id is limited to 32. Adding and remocing reference in bo is done with atomic operations to allow parallel access to a bo from multiple contexts.
This optimization decreases cs_write_reloc share of torcs profiling from 4.3% to 2.6%.
--- radeon/radeon_bo_gem.c | 1 + radeon/radeon_cs.c | 6 ++ radeon/radeon_cs.h | 2 +- radeon/radeon_cs_gem.c | 133 ++++++++++++++++++++++++++++++++++++----------- radeon/radeon_cs_int.h | 1 + 5 files changed, 111 insertions(+), 32 deletions(-)
radeon/radeon_bo_gem.c +1 -0
@@ -80,6 +80,7 @@ static struct radeon_bo *bo_open(struct radeon_bo_manager *bom, bo->base.domains = domains; bo->base.flags = flags; bo->base.ptr =NULL ; + bo->base.referenced_in_cs =0 ; bo->map_count =0 ;if (handle) {struct drm_gem_open open_arg;
radeon/radeon_cs.c +6 -0
@@ -88,3 +88,9 @@ void radeon_cs_space_set_flush(struct radeon_cs *cs, void (*fn)(void *), void *d csi->space_flush_fn = fn; csi->space_flush_data = data; } + +uint32_t radeon_cs_get_id (struct radeon_cs *cs) +{ +struct radeon_cs_int *csi = (struct radeon_cs_int *)cs; +return csi->id; +}
radeon/radeon_cs.h +1 -1
@@ -85,7 +85,7 @@ extern int radeon_cs_write_reloc(struct radeon_cs *cs,uint32_t read_domain, uint32_t write_domain, uint32_t flags); - +extern uint32_t radeon_cs_get_id (struct radeon_cs *cs);/* * add a persistent BO to the list * a persistent BO is one that will be referenced across flushes,
radeon/radeon_cs_gem.c +102 -31
182 lines not on this page · raw message
radeon/radeon_cs_int.h +1 -0
@@ -28,6 +28,7 @@ struct radeon_cs_int {int bo_count;void (*space_flush_fn)(void *);void *space_flush_data; +uint32_t id; };/* cs functions */
Signature · 10 lines
1.6.3.3 ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev --
Re: [PATCH 2/2] libdrm_radeon: Optimize reloc writing to do less looping.
Quoted 11 lines · Pauli Nieminen
Bit has table will be first checked from BO if we can quarentee this BO is not in this cs already.
To quarentee that there is no other cs with same id number of CS that can have id is limited to 32. Adding and remocing reference in bo is done with atomic operations to allow parallel access to a bo from multiple contexts.
This optimization decreases
cs_write_relocshare of torcs profiling from 4.3% to 2.6%.Signed-off-by Pauli Nieminen <[email protected]>
[...]
Quoted 13 lines
radeon/radeon_cs_gem.c +5
-0@@ -68,6 +69,66 @@ struct cs_gem {struct radeon_bo_int **relocs_bo; }; + +#if !defined(__GNUC__) || __GNUC__ <4 || (__GNUC__ ==4 && __GNUC_MINOR__ <2 ) +/* no built in sync support in compiler define place holders */ +uint32_t __sync_add_and_fetch (uint32_t *a,uint32_t val) +{
radeon/radeon_cs_gem.c struct cs_gem { · line 86
Quoted 10 lines
radeon/radeon_cs_gem.c +10
-0+ *a += val; + return val; +} + +uint32_t __sync_add_and_fetch (uint32_t *a,uint32_t val) +{ + *a -= val; +return val; +} +#endif
This doesn't look like it could build... presumably the latter should be called __sync_sub_and_fetch()?
Do these stand any chance of working properly in circumstances where atomicity is actually important though?
Signature · 14 lines
Earthling Michel Dänzer | http://www.vmware.com Libre software enthusiast | Debian, X and DRI developer ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev -- _______________________________________________ Dri-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [PATCH 2/2] libdrm_radeon: Optimize reloc writing to do less looping.
2010/3/10 Michel Dänzer <[email protected]>:
Quoted 33 lines · Pauli Nieminen
Quoted 11 lines
Bit has table will be first checked from BO if we can quarentee this BO is not in this cs already.
To quarentee that there is no other cs with same id number of CS that can have id is limited to 32. Adding and remocing reference in bo is done with atomic operations to allow parallel access to a bo from multiple contexts.
This optimization decreases
cs_write_relocshare of torcs profiling from 4.3% to 2.6%.Signed-off-by Pauli Nieminen <[email protected]>[...]
Quoted 18 lines
radeon/radeon_cs_gem.c +10
-0@@ -68,6 +69,66 @@ struct cs_gem { struct radeon_bo_int **relocs_bo; }; + +#if !defined(__GNUC__) || __GNUC__ <4 || (__GNUC__ ==4 && __GNUC_MINOR__ <2 ) +/* no built in sync support in compiler define place holders */ +uint32_t __sync_add_and_fetch (uint32_t *a,uint32_t val) +{ + *a += val; + return val; +} + +uint32_t __sync_add_and_fetch (uint32_t *a,uint32_t val)
Quoted 9 lines
radeon/radeon_cs_gem.c struct cs_gem { · line 86
Quoted 5 lines
radeon/radeon_cs_gem.c +5
-0+{ + *a -= val; + return val; +} +#endif This doesn't look like it could build... presumably the latter should be called
__sync_sub_and_fetch()?
sorry .wrong patch coming from somewhere :/
Quoted 8 lines
Do these stand any chance of working properly in circumstances where atomicity is actually important though?
--
Earthling Michel Dänzer | http://www.vmware.com
Libre software enthusiast | Debian, X and DRI developer
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
Quoted 130 lines · Pauli Nieminen
intel_atomic.hincludes very usefull atomic operations for lock free parrallel access of variables. Moving these to core libdrm for code sharing with radeon.Signed-off-by Pauli Nieminen <[email protected]>---
Makefile.am | 2 +-
configure.ac | 2 +-
intel/intel_atomic.h | 56 +-----------------------------xf86atomic.h| 93 ++++++++++++++++++++++++++++++++++++++++++++++++++4 files changed, 96 insertions(+), 57 deletions(-)create mode 100644
xf86atomic.hMakefile.am +0
-0@@ -59,7 +59,7 @@ libdrm_la_SOURCES = \libdrm_lists.h
libdrmincludedir = ${includedir}
Makefile.am +1
-1-libdrminclude_HEADERS = xf86drm.h xf86drmMode.h +libdrminclude_HEADERS = xf86drm.h xf86drmMode.h xf86atomic.hEXTRA_DIST = libdrm.pc.in
include/drm/*configure.ac +0
-0@@ -198,7 +198,7 @@ if test "x$INTEL" != "xno"; then ])if test "x$drm_cv_atomic_primitives" = xIntel; then
- AC_DEFINE(HAVE_INTEL_ATOMIC_PRIMITIVES, 1,
+ AC_DEFINE(HAVE_LIBDRM_ATOMIC_PRIMITIVES, 1,
[Enable if your compiler supports the Intel __sync_* atomic primitives]) fi
if test "x$drm_cv_atomic_primitives" = "xlibatomic-ops"; thenintel/intel_atomic.h +1
-55@@ -34,60 +34,6 @@ #ifndef INTEL_ATOMICS_H #define INTEL_ATOMICS_H -#ifdef HAVE_CONFIG_H -#include "config.h" -#endif - -#if HAVE_INTEL_ATOMIC_PRIMITIVES - -#define HAS_ATOMIC_OPS 1 - -typedef struct { - int atomic; -}atomic_t ; - -# define atomic_read (x) ((x)->atomic) -# define atomic_set (x, val) ((x)->atomic = (val)) -# define atomic_inc (x) ((void) __sync_fetch_and_add (&(x)->atomic, 1)) -# define atomic_dec_and_test (x) (__sync_fetch_and_add (&(x)->atomic, -1) == 1) -# define atomic_cmpxchg (x, oldv, newv) __sync_val_compare_and_swap (&(x)->atomic, oldv, newv) - -#endif - -#if HAVE_LIB_ATOMIC_OPS -#include <atomic_ops.h> - -#define HAS_ATOMIC_OPS 1 - -typedef struct { - AO_t atomic; -}atomic_t ; - -# define atomic_read (x) AO_load_full(&(x)->atomic) -# define atomic_set (x, val) AO_store_full(&(x)->atomic, (val)) -# define atomic_inc (x) ((void) AO_fetch_and_add1_full(&(x)->atomic)) -# define atomic_dec_and_test (x) (AO_fetch_and_sub1_full(&(x)->atomic) == 1) -# define atomic_cmpxchg (x, oldv, newv) AO_compare_and_swap_full(&(x)->atomic, oldv, newv) - -#endif - -#if defined(__sun) && !defined(HAS_ATOMIC_OPS) /* Solaris & OpenSolaris */ - -#include <sys/atomic.h> -#define HAS_ATOMIC_OPS 1 - -typedef struct {uint_t atomic; }atomic_t ; - -# define atomic_read (x) (int) ((x)->atomic) -# define atomic_set (x, val) ((x)->atomic = (uint_t)(val)) -# define atomic_inc (x) (atomic_inc_uint (&(x)->atomic)) -# define atomic_dec_and_test (x) (atomic_dec_uint_nv(&(x)->atomic) == 1) -# define atomic_cmpxchg (x, oldv, newv) atomic_cas_uint (&(x)->atomic, oldv, newv) - -#endif - -#if !HAS_ATOMIC_OPS -#error libdrm-intel requires atomic operations, please define them for your CPU/compiler. -#endif +#include <xf86atomic.h> #endifxf86atomic.h new file +19
-0@@ -0,0 +1,93 @@ +/* +* Copyright © 2009 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
xf86atomic.h line 29
Quoted 10 lines
xf86atomic.h new file +10
-0+* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS +* IN THE SOFTWARE. +* +* Authors: +* Chris Wilson <[email protected]> +* +*/ + +/** +* @file intel_atomics.h
This needs to be updated.
Quoted 52 lines
xf86atomic.h new file +52
-0+* +* Private definitions for atomic operations +*/ + +#ifndef LIBDRM_ATOMICS_H +#define LIBDRM_ATOMICS_H + +#ifdef HAVE_CONFIG_H +#include "config.h" +#endif + +#if HAVE_LIBDRM_ATOMIC_PRIMITIVES + +#define HAS_ATOMIC_OPS 1 + +typedef struct { + int atomic; +}atomic_t ; + +# define atomic_read (x) ((x)->atomic) +# define atomic_set (x, val) ((x)->atomic = (val)) +# define atomic_inc (x) ((void) __sync_fetch_and_add (&(x)->atomic, 1)) +# define atomic_dec_and_test (x) (__sync_fetch_and_add (&(x)->atomic, -1) == 1) +# define atomic_cmpxchg (x, oldv, newv) __sync_val_compare_and_swap (&(x)->atomic, oldv, newv) + +#endif + +#if HAVE_LIB_ATOMIC_OPS +#include <atomic_ops.h> + +#define HAS_ATOMIC_OPS 1 + +typedef struct { + AO_t atomic; +}atomic_t ; + +# define atomic_read (x) AO_load_full(&(x)->atomic) +# define atomic_set (x, val) AO_store_full(&(x)->atomic, (val)) +# define atomic_inc (x) ((void) AO_fetch_and_add1_full(&(x)->atomic)) +# define atomic_dec_and_test (x) (AO_fetch_and_sub1_full(&(x)->atomic) == 1) +# define atomic_cmpxchg (x, oldv, newv) AO_compare_and_swap_full(&(x)->atomic, oldv, newv) + +#endif + +#if defined(__sun) && !defined(HAS_ATOMIC_OPS) /* Solaris & OpenSolaris */ + +#include <sys/atomic.h> +#define HAS_ATOMIC_OPS 1 + +typedef struct {uint_t atomic; }atomic_t ; + +# define atomic_read (x) (int) ((x)->atomic)
xf86atomic.h line 91
Quoted 10 lines
xf86atomic.h new file +10
-0+# define atomic_set (x, val) ((x)->atomic = (uint_t)(val)) +# define atomic_inc (x) (atomic_inc_uint (&(x)->atomic)) +# define atomic_dec_and_test (x) (atomic_dec_uint_nv(&(x)->atomic) == 1) +# define atomic_cmpxchg (x, oldv, newv) atomic_cas_uint (&(x)->atomic, oldv, newv) + +#endif + +#if !HAS_ATOMIC_OPS +#error libdrm-intel requires atomic operations, please define them for your CPU/compiler. +#endif
This needs to be updated/fixed.
Quoted 4 lines
xf86atomic.h new file +2
-1+ +#endif --1.6.3.3
Matt
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
Quoted 21 lines · Pauli Nieminen
Quoted 20 lines
intel_atomic.hincludes very usefull atomic operations for lock free parrallel access of variables. Moving these to core libdrm for code sharing with radeon.Signed-off-by Pauli Nieminen <[email protected]>---
Makefile.am | 2 +-
configure.ac | 2 +-
intel/intel_atomic.h | 56 +-----------------------------xf86atomic.h| 93 ++++++++++++++++++++++++++++++++++++++++++++++++++4 files changed, 96 insertions(+), 57 deletions(-)create mode 100644
xf86atomic.hMakefile.am +0
-0@@ -59,7 +59,7 @@ libdrm_la_SOURCES = \libdrm_lists.h
Quoted 4 lines
Makefile.am libdrm_la_SOURCES = \ · line 62
Quoted 4 lines
libdrmincludedir = ${includedir}
Makefile.am +1
-1-libdrminclude_HEADERS = xf86drm.h xf86drmMode.h +libdrminclude_HEADERS = xf86drm.h xf86drmMode.h xf86atomic.h
Should this really get installed? I'm not sure this should be public libdrm interface.
Cheers,
Julien
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
Quoted 29 lines · Julien Cristau
Quoted 21 lines · Pauli Nieminen
Quoted 20 lines
intel_atomic.hincludes very usefull atomic operations for lock free parrallel access of variables. Moving these to core libdrm for code sharing with radeon.Signed-off-by Pauli Nieminen <[email protected]>---
Makefile.am | 2 +-
configure.ac | 2 +-
intel/intel_atomic.h | 56 +-----------------------------xf86atomic.h| 93 ++++++++++++++++++++++++++++++++++++++++++++++++++4 files changed, 96 insertions(+), 57 deletions(-)create mode 100644
xf86atomic.hMakefile.am +0
-0@@ -59,7 +59,7 @@ libdrm_la_SOURCES = \libdrm_lists.h
Quoted 4 lines
Makefile.am libdrm_la_SOURCES = \ · line 62
Quoted 4 lines
libdrmincludedir = ${includedir}
Makefile.am +1
-1-libdrminclude_HEADERS = xf86drm.h xf86drmMode.h +libdrminclude_HEADERS = xf86drm.h xf86drmMode.h xf86atomic.hShould this really get installed? I'm not sure this should be public libdrm interface.
Quoted 3 lines
Cheers,
Julien
Inlte had it installed. And my patch adds it to radeon so that mesa will reference it indirectly tough radeon_bo_int.h. Maybe dependency in mesa to radeon_bo_int.h is bad idea in first place.
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
Quoted 9 lines · Pauli Nieminen
Quoted 6 lines · Julien Cristau
Should this really get installed? �I'm not sure this should be public libdrm interface.
Cheers,
JulienIntel had it installed.
My bad, change it to noinst_HEADERS please.
Signature · 1 line
Chris Wilson, Intel Open Source Technology Centre
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel
Quoted 3 lines · Pauli Nieminen
intel_atomic.hincludes very usefull atomic operations for lock free parrallel access of variables. Moving these to core libdrm for code sharing with radeon.
s/xf86/libdrm/ but other than that, cool.
------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel