drm/stm: drv: Improve data transfers
Subsystems (4)
PATCH v2 1/1 drm/stm: drv: Improve data transfers
To optimize data transfers, align pitch on 128 bytes & height on 4 bytes. This optimization is not applicable on hw without MMU.
---
Changes in v2: Rename stm_dumb_create() to stm_gem_cma_dumb_create() and
move CONFIG_MMU inside the function following comments from Benjamin Gaignard.
drivers/gpu/drm/stm/drv.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)
drivers/gpu/drm/stm/drv.c +19 -1
@@ -31,6 +31,24 @@ static const struct drm_mode_config_funcs drv_mode_config_funcs = { .atomic_commit = drm_atomic_helper_commit, }; +static int stm_gem_cma_dumb_create (struct drm_file *file, +struct drm_device *dev, +struct drm_mode_create_dumb *args) +{ +#ifdef CONFIG_MMU +unsigned int min_pitch =DIV_ROUND_UP (args->width * args->bpp,8 ); + +/* +* in order to optimize data transfer, pitch is aligned on +* 128 bytes, height is aligned on 4 bytes +*/ + args->pitch =roundup (min_pitch,128 ); + args->height =roundup (args->height,4 ); +#endif + +return drm_gem_cma_dumb_create_internal (file, dev, args); +} +DEFINE_DRM_GEM_CMA_FOPS (drv_driver_fops);static struct drm_driver drv_driver = {
@@ -44,7 +62,7 @@ static struct drm_driver drv_driver = { .minor = 0, .patchlevel = 0, .fops = &drv_driver_fops, - .dumb_create = drm_gem_cma_dumb_create, + .dumb_create = stm_gem_cma_dumb_create, .prime_handle_to_fd = drm_gem_prime_handle_to_fd, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, .gem_free_object_unlocked = drm_gem_cma_free_object,
Signature · 6 lines
2.15.1 _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
2018-01-30 11:42 GMT+01:00 Philippe Cornu <[email protected]>:
Quoted 6 lines
To optimize data transfers, align pitch on 128 bytes & height on 4 bytes. This optimization is not applicable on hw without MMU.
Signed-off-by Yannick Fertre <[email protected]>Signed-off-by Vincent Abriou <[email protected]>Signed-off-by Philippe Cornu <[email protected]>
Applied on drm-misc-next.
Regards,
Benjamin
Quoted 47 lines
---
Changes in v2: Renamestm_dumb_create()tostm_gem_cma_dumb_create()and
moveCONFIG_MMUinside the function following comments from Benjamin Gaignard.drivers/gpu/drm/stm/drv.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)drivers/gpu/drm/stm/drv.c +19
-2@@ -31,6 +31,24 @@ static const struct drm_mode_config_funcs drv_mode_config_funcs = { .atomic_commit = drm_atomic_helper_commit, }; +static int stm_gem_cma_dumb_create (struct drm_file *file, +struct drm_device *dev, +struct drm_mode_create_dumb *args) +{ +#ifdef CONFIG_MMU +unsigned int min_pitch =DIV_ROUND_UP (args->width * args->bpp,8 ); + +/* +* in order to optimize data transfer, pitch is aligned on +* 128 bytes, height is aligned on 4 bytes +*/ + args->pitch =roundup (min_pitch,128 ); + args->height =roundup (args->height,4 ); +#endif + +return drm_gem_cma_dumb_create_internal (file, dev, args); +} +DEFINE_DRM_GEM_CMA_FOPS (drv_driver_fops);static struct drm_driver drv_driver = {@@ -44,7 +62,7 @@ static struct drm_driver drv_driver = { .minor = 0, .patchlevel = 0, .fops = &drv_driver_fops, - .dumb_create = drm_gem_cma_dumb_create, + .dumb_create = stm_gem_cma_dumb_create, .prime_handle_to_fd = drm_gem_prime_handle_to_fd, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, .gem_free_object_unlocked = drm_gem_cma_free_object, --
Quoted 2 lines
2.15.1
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
Hi Benjamin,
Quoted 9 lines · On Tuesday, 30 January 2018 12:51:25 EET Benjamin Gaignard wrote:
2018-01-30 11:42 GMT+01:00 Philippe Cornu <[email protected]>:
Quoted 6 lines
To optimize data transfers, align pitch on 128 bytes & height on 4 bytes. This optimization is not applicable on hw without MMU.
Signed-off-by Yannick Fertre <[email protected]>Signed-off-by Vincent Abriou <[email protected]>Signed-off-by Philippe Cornu <[email protected]>Applied on drm-misc-next.
A 9 minutes review window is pretty small... I've been told before that I should then review v1 instead of complaining about short review times for v2, but in this case v2 introduces a bug that wasn't present in v1.
Quoted 57 lines
Quoted 57 lines
---
Changes in v2: Renamestm_dumb_create()tostm_gem_cma_dumb_create()andmove
CONFIG_MMUinside the function following comments from Benjamin Gaignard.drivers/gpu/drm/stm/drv.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)drivers/gpu/drm/stm/drv.c +0
-0@@ -31,6 +31,24 @@ static const struct drm_mode_config_funcs
drv_mode_config_funcs= {>
.atomic_commit =drm_atomic_helper_commit,};
drivers/gpu/drm/stm/drv.c +19
-1+static int stm_gem_cma_dumb_create (struct drm_file *file, +struct drm_device *dev, +struct drm_mode_create_dumb *args) +{ +#ifdef CONFIG_MMU +unsigned int min_pitch =DIV_ROUND_UP (args->width * args->bpp,8 ); + +/* +* in order to optimize data transfer, pitch is aligned on +* 128 bytes, height is aligned on 4 bytes +*/ + args->pitch =roundup (min_pitch,128 ); + args->height =roundup (args->height,4 ); +#endif + +return drm_gem_cma_dumb_create_internal (file, dev, args); +} +DEFINE_DRM_GEM_CMA_FOPS (drv_driver_fops);static struct drm_driver drv_driver = {@@ -44,7 +62,7 @@ static struct drm_driver drv_driver = { .minor = 0, .patchlevel = 0, .fops = &drv_driver_fops, - .dumb_create = drm_gem_cma_dumb_create, + .dumb_create = stm_gem_cma_dumb_create,.prime_handle_to_fd =
drm_gem_prime_handle_to_fd, .prime_fd_to_handle =drm_gem_prime_fd_to_handle, .gem_free_object_unlocked =drm_gem_cma_free_object,--
2.15.1
Signature · 8 lines
Regards, Laurent Pinchart _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Quoted 11 lines · Laurent Pinchart
Hi Benjamin,
Quoted 8 lines · On Tuesday, 30 January 2018 12:51:25 EET Benjamin Gaignard wrote:
2018-01-30 11:42 GMT+01:00 Philippe Cornu <[email protected]>:
Quoted 6 lines
To optimize data transfers, align pitch on 128 bytes & height on 4 bytes. This optimization is not applicable on hw without MMU.
Signed-off-by Yannick Fertre <[email protected]>Signed-off-by Vincent Abriou <[email protected]>Signed-off-by Philippe Cornu <[email protected]>
Quoted 5 lines
Quoted 1 line
Applied on drm-misc-next.
A 9 minutes review window is pretty small... I've been told before that I should then review v1 instead of complaining about short review times for v2, but in this case v2 introduces a bug that wasn't present in v1.
IMO no patch ever needs to be applied this fast. Give more eyeballs a chance.
BR,
Jani.
Quoted 58 lines
Quoted 58 lines
Quoted 58 lines
---
Changes in v2: Renamestm_dumb_create()tostm_gem_cma_dumb_create()andmove
CONFIG_MMUinside the function following comments from Benjamin Gaignard.drivers/gpu/drm/stm/drv.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)drivers/gpu/drm/stm/drv.c +0
-0@@ -31,6 +31,24 @@ static const struct drm_mode_config_funcs
drv_mode_config_funcs= {>
.atomic_commit =drm_atomic_helper_commit,};
drivers/gpu/drm/stm/drv.c +19
-1+static int stm_gem_cma_dumb_create (struct drm_file *file, +struct drm_device *dev, +struct drm_mode_create_dumb *args) +{ +#ifdef CONFIG_MMU +unsigned int min_pitch =DIV_ROUND_UP (args->width * args->bpp,8 ); + +/* +* in order to optimize data transfer, pitch is aligned on +* 128 bytes, height is aligned on 4 bytes +*/ + args->pitch =roundup (min_pitch,128 ); + args->height =roundup (args->height,4 ); +#endif + +return drm_gem_cma_dumb_create_internal (file, dev, args); +} +DEFINE_DRM_GEM_CMA_FOPS (drv_driver_fops);static struct drm_driver drv_driver = {@@ -44,7 +62,7 @@ static struct drm_driver drv_driver = { .minor = 0, .patchlevel = 0, .fops = &drv_driver_fops, - .dumb_create = drm_gem_cma_dumb_create, + .dumb_create = stm_gem_cma_dumb_create,.prime_handle_to_fd =
drm_gem_prime_handle_to_fd, .prime_fd_to_handle =drm_gem_prime_fd_to_handle, .gem_free_object_unlocked =drm_gem_cma_free_object,--
2.15.1
Signature · 5 lines
Jani Nikula, Intel Open Source Technology Center _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
Hi Philippe,
Thank you for the patch.
Quoted 28 lines · On Tuesday, 30 January 2018 12:42:00 EET Philippe Cornu wrote:
To optimize data transfers, align pitch on 128 bytes & height on 4 bytes. This optimization is not applicable on hw without MMU.
Signed-off-by Yannick Fertre <[email protected]>Signed-off-by Vincent Abriou <[email protected]>Signed-off-by Philippe Cornu <[email protected]>---
Changes in v2: Renamestm_dumb_create()tostm_gem_cma_dumb_create()and
moveCONFIG_MMUinside the function following comments from Benjamin Gaignard.drivers/gpu/drm/stm/drv.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)drivers/gpu/drm/stm/drv.c +6
-0@@ -31,6 +31,24 @@ static const struct drm_mode_config_funcs drv_mode_config_funcs = { .atomic_commit = drm_atomic_helper_commit, }; +static int stm_gem_cma_dumb_create (struct drm_file *file, +struct drm_device *dev, +struct drm_mode_create_dumb *args) +{ +#ifdef CONFIG_MMU +unsigned int min_pitch =DIV_ROUND_UP (args->width * args->bpp,8 );
drivers/gpu/drm/stm/drv.c static const struct drm_mode_config_funcs · line 49
Quoted 10 lines
drivers/gpu/drm/stm/drv.c +10
-0+ +/* +* in order to optimize data transfer, pitch is aligned on +* 128 bytes, height is aligned on 4 bytes +*/ + args->pitch =roundup (min_pitch,128 ); + args->height =roundup (args->height,4 ); +#endif + +return drm_gem_cma_dumb_create_internal (file, dev, args);
In the !CONFIG_MMU case you now bypass the pitch and size calculations performed by drm_gem_cma_dumb_create(), allowing userspace to allocate arbitrarily large buffers. Is that intentional ?
Quoted 14 lines
drivers/gpu/drm/stm/drv.c +3
-1+} +DEFINE_DRM_GEM_CMA_FOPS (drv_driver_fops);static struct drm_driver drv_driver = {@@ -44,7 +62,7 @@ static struct drm_driver drv_driver = { .minor = 0, .patchlevel = 0, .fops = &drv_driver_fops, - .dumb_create = drm_gem_cma_dumb_create, + .dumb_create = stm_gem_cma_dumb_create, .prime_handle_to_fd = drm_gem_prime_handle_to_fd, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, .gem_free_object_unlocked = drm_gem_cma_free_object,
Signature · 8 lines
Regards, Laurent Pinchart _______________________________________________ dri-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/dri-devel
2018-01-30 15:58 GMT+01:00 Laurent Pinchart <[email protected]>:
Quoted 39 lines
Hi Philippe,
Thank you for the patch.
Quoted 34 lines · On Tuesday, 30 January 2018 12:42:00 EET Philippe Cornu wrote:
To optimize data transfers, align pitch on 128 bytes & height on 4 bytes. This optimization is not applicable on hw without MMU.
Signed-off-by Yannick Fertre <[email protected]>Signed-off-by Vincent Abriou <[email protected]>Signed-off-by Philippe Cornu <[email protected]>---
Changes in v2: Renamestm_dumb_create()tostm_gem_cma_dumb_create()and
moveCONFIG_MMUinside the function following comments from Benjamin Gaignard.drivers/gpu/drm/stm/drv.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)drivers/gpu/drm/stm/drv.c +12
-0@@ -31,6 +31,24 @@ static const struct drm_mode_config_funcs drv_mode_config_funcs = { .atomic_commit = drm_atomic_helper_commit, }; +static int stm_gem_cma_dumb_create (struct drm_file *file, +struct drm_device *dev, +struct drm_mode_create_dumb *args) +{ +#ifdef CONFIG_MMU +unsigned int min_pitch =DIV_ROUND_UP (args->width * args->bpp,8 ); + +/* +* in order to optimize data transfer, pitch is aligned on +* 128 bytes, height is aligned on 4 bytes +*/ + args->pitch =roundup (min_pitch,128 );
Quoted 8 lines
drivers/gpu/drm/stm/drv.c static const struct drm_mode_config_funcs drv_mode_config_funcs = { · line 49
Quoted 4 lines
drivers/gpu/drm/stm/drv.c +4
-0+ args->height =roundup (args->height,4 ); +#endif + +return drm_gem_cma_dumb_create_internal (file, dev, args);In the !CONFIG_MMU case you now bypass the pitch and size calculations performed by
drm_gem_cma_dumb_create(), allowing userspace to allocate arbitrarily large buffers. Is that intentional ?
My bad, I will send a fix
Quoted 21 lines
Quoted 7 lines
drivers/gpu/drm/stm/drv.c +2
-0+} +DEFINE_DRM_GEM_CMA_FOPS (drv_driver_fops);static struct drm_driver drv_driver = {drivers/gpu/drm/stm/drv.c static struct drm_driver drv_driver = { · line 68
Quoted 8 lines
drivers/gpu/drm/stm/drv.c +1
-1@@ -44,7 +62,7 @@ static struct drm_driver drv_driver = { .minor = 0, .patchlevel = 0, .fops = &drv_driver_fops, - .dumb_create = drm_gem_cma_dumb_create, + .dumb_create = stm_gem_cma_dumb_create, .prime_handle_to_fd = drm_gem_prime_handle_to_fd, .prime_fd_to_handle = drm_gem_prime_fd_to_handle, .gem_free_object_unlocked = drm_gem_cma_free_object,--
Regards,Laurent Pinchart
Hi Benjamin,
Quoted 47 lines · On Tuesday, 30 January 2018 17:08:48 EET Benjamin Gaignard wrote:
2018-01-30 15:58 GMT+01:00 Laurent Pinchart:
Quoted 37 lines · On Tuesday, 30 January 2018 12:42:00 EET Philippe Cornu wrote:
Quoted 36 lines
To optimize data transfers, align pitch on 128 bytes & height on 4 bytes. This optimization is not applicable on hw without MMU.
Signed-off-by Yannick Fertre <[email protected]>Signed-off-by Vincent Abriou <[email protected]>Signed-off-by Philippe Cornu <[email protected]>---
Changes in v2: Renamestm_dumb_create()tostm_gem_cma_dumb_create()andmove
CONFIG_MMUinside the function following comments from Benjamin Gaignard.drivers/gpu/drm/stm/drv.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-)drivers/gpu/drm/stm/drv.c +12
-0@@ -31,6 +31,24 @@ static const struct drm_mode_config_funcs drv_mode_config_funcs = { .atomic_commit = drm_atomic_helper_commit, }; +static int stm_gem_cma_dumb_create (struct drm_file *file, +struct drm_device *dev, +struct drm_mode_create_dumb *args) +{ +#ifdef CONFIG_MMU +unsigned int min_pitch =DIV_ROUND_UP (args->width * args->bpp,8 ); + +/* +* in order to optimize data transfer, pitch is aligned on +* 128 bytes, height is aligned on 4 bytes +*/ + args->pitch =roundup (min_pitch,128 );Quoted 8 lines
drivers/gpu/drm/stm/drv.c static const struct drm_mode_config_funcs drv_mode_config_funcs = { · line 49
Quoted 4 lines
drivers/gpu/drm/stm/drv.c +4
-0+ args->height =roundup (args->height,4 ); +#endif + +return drm_gem_cma_dumb_create_internal (file, dev, args);In the !CONFIG_MMU case you now bypass the pitch and size calculations performed by
drm_gem_cma_dumb_create(), allowing userspace to allocate arbitrarily large buffers. Is that intentional ?
Quoted 1 line
My bad, I will send a fix
Thank you.
Quoted 19 lines
Quoted 19 lines
Quoted 19 lines
drivers/gpu/drm/stm/drv.c +2
-0+} +DEFINE_DRM_GEM_CMA_FOPS (drv_driver_fops);static struct
drm_driverdrv_driver= {drivers/gpu/drm/stm/drv.c +1
-1@@ -44,7 +62,7 @@ static struct drm_driver drv_driver = { .minor = 0, .patchlevel = 0, .fops = &drv_driver_fops, - .dumb_create = drm_gem_cma_dumb_create, + .dumb_create = stm_gem_cma_dumb_create,.prime_handle_to_fd =
drm_gem_prime_handle_to_fd, .prime_fd_to_handle =drm_gem_prime_fd_to_handle, .gem_free_object_unlocked =drm_gem_cma_free_object,
Signature · 3 lines
Regards, Laurent Pinchart