public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Daniel Vetter <daniel@ffwll.ch>
To: Chris Wilson <chris@chris-wilson.co.uk>
Cc: intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Subject: Re: [PATCH v4 05/38] drm: Add a simple generator of random permutations
Date: Tue, 27 Dec 2016 12:33:50 +0100	[thread overview]
Message-ID: <20161227113350.GH19172@dvetter-linux.ger.corp.intel.com> (raw)
In-Reply-To: <20161222083641.2691-6-chris@chris-wilson.co.uk>

On Thu, Dec 22, 2016 at 08:36:08AM +0000, Chris Wilson wrote:
> When testing, we want a random but yet reproducible order in which to
> process elements. Here we create an array which is a random (using the
> Tausworthe PRNG) permutation of the order in which to execute.
> 
> Note these are simple helpers intended to be merged upstream in lib/

Is that happening? I mean including some kernel-doc and all that fun?
-Daniel

> 
> v2: Tidier code by David Herrmann
> v3: Add reminder that this code is intended to be temporary, with at
> least the bulk of the prandom changes going to lib/
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> Cc: David Herrmann <dh.herrmann@gmail.com>
> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> ---
>  drivers/gpu/drm/Kconfig          |  4 ++++
>  drivers/gpu/drm/Makefile         |  1 +
>  drivers/gpu/drm/lib/drm_random.c | 41 ++++++++++++++++++++++++++++++++++++++++
>  drivers/gpu/drm/lib/drm_random.h | 25 ++++++++++++++++++++++++
>  4 files changed, 71 insertions(+)
>  create mode 100644 drivers/gpu/drm/lib/drm_random.c
>  create mode 100644 drivers/gpu/drm/lib/drm_random.h
> 
> diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
> index ebfe8404c25f..45a1c7468e88 100644
> --- a/drivers/gpu/drm/Kconfig
> +++ b/drivers/gpu/drm/Kconfig
> @@ -321,3 +321,7 @@ config DRM_SAVAGE
>  	  chipset. If M is selected the module will be called savage.
>  
>  endif # DRM_LEGACY
> +
> +config DRM_LIB_RANDOM
> +	bool
> +	default n
> diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
> index b9ae4280de9d..6bb416360ae4 100644
> --- a/drivers/gpu/drm/Makefile
> +++ b/drivers/gpu/drm/Makefile
> @@ -18,6 +18,7 @@ drm-y       :=	drm_auth.o drm_bufs.o drm_cache.o \
>  		drm_plane.o drm_color_mgmt.o drm_print.o \
>  		drm_dumb_buffers.o drm_mode_config.o
>  
> +drm-$(CONFIG_DRM_LIB_RANDOM) += lib/drm_random.o
>  drm-$(CONFIG_COMPAT) += drm_ioc32.o
>  drm-$(CONFIG_DRM_GEM_CMA_HELPER) += drm_gem_cma_helper.o
>  drm-$(CONFIG_PCI) += ati_pcigart.o
> diff --git a/drivers/gpu/drm/lib/drm_random.c b/drivers/gpu/drm/lib/drm_random.c
> new file mode 100644
> index 000000000000..7b12a68c3b54
> --- /dev/null
> +++ b/drivers/gpu/drm/lib/drm_random.c
> @@ -0,0 +1,41 @@
> +#include <linux/bitops.h>
> +#include <linux/kernel.h>
> +#include <linux/random.h>
> +#include <linux/slab.h>
> +#include <linux/types.h>
> +
> +#include "drm_random.h"
> +
> +static inline u32 drm_prandom_u32_max_state(u32 ep_ro, struct rnd_state *state)
> +{
> +	return upper_32_bits((u64)prandom_u32_state(state) * ep_ro);
> +}
> +
> +void drm_random_reorder(unsigned int *order, unsigned int count,
> +			struct rnd_state *state)
> +{
> +	unsigned int i, j;
> +
> +	for (i = 0; i < count; ++i) {
> +		BUILD_BUG_ON(sizeof(unsigned int) > sizeof(u32));
> +		j = drm_prandom_u32_max_state(count, state);
> +		swap(order[i], order[j]);
> +	}
> +}
> +EXPORT_SYMBOL(drm_random_reorder);
> +
> +unsigned int *drm_random_order(unsigned int count, struct rnd_state *state)
> +{
> +	unsigned int *order, i;
> +
> +	order = kmalloc_array(count, sizeof(*order), GFP_TEMPORARY);
> +	if (!order)
> +		return order;
> +
> +	for (i = 0; i < count; i++)
> +		order[i] = i;
> +
> +	drm_random_reorder(order, count, state);
> +	return order;
> +}
> +EXPORT_SYMBOL(drm_random_order);
> diff --git a/drivers/gpu/drm/lib/drm_random.h b/drivers/gpu/drm/lib/drm_random.h
> new file mode 100644
> index 000000000000..a78644bea7f9
> --- /dev/null
> +++ b/drivers/gpu/drm/lib/drm_random.h
> @@ -0,0 +1,25 @@
> +#ifndef __DRM_RANDOM_H__
> +#define __DRM_RANDOM_H__
> +
> +/* This is a temporary home for a couple of utility functions that should
> + * be transposed to lib/ at the earliest convenience.
> + */
> +
> +#include <linux/random.h>
> +
> +#define DRM_RND_STATE_INITIALIZER(seed__) ({				\
> +	struct rnd_state state__;					\
> +	prandom_seed_state(&state__, (seed__));				\
> +	state__;							\
> +})
> +
> +#define DRM_RND_STATE(name__, seed__) \
> +	struct rnd_state name__ = DRM_RND_STATE_INITIALIZER(seed__)
> +
> +unsigned int *drm_random_order(unsigned int count,
> +			       struct rnd_state *state);
> +void drm_random_reorder(unsigned int *order,
> +			unsigned int count,
> +			struct rnd_state *state);
> +
> +#endif /* !__DRM_RANDOM_H__ */
> -- 
> 2.11.0
> 
> _______________________________________________
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/dri-devel

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2016-12-27 11:33 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-22  8:36 drm_mm fixes, take 4? Chris Wilson
2016-12-22  8:36 ` [PATCH v4 01/38] drm/i915: Use the MRU stack search after evicting Chris Wilson
2016-12-27 11:30   ` Daniel Vetter
2016-12-22  8:36 ` [PATCH v4 02/38] drm: Use drm_mm_nodes() as shorthand for the list of nodes under struct drm_mm Chris Wilson
2016-12-22  8:36 ` [PATCH v4 03/38] drm: Compile time enabling for asserts in drm_mm Chris Wilson
2016-12-22  8:36 ` [PATCH v4 04/38] lib: Add a simple prime number generator Chris Wilson
2016-12-22  9:52   ` Joonas Lahtinen
2016-12-22 10:00     ` [Intel-gfx] " Chris Wilson
2016-12-22 14:45   ` [PATCH v10] " Chris Wilson
2016-12-27 11:31     ` Daniel Vetter
2016-12-22  8:36 ` [PATCH v4 05/38] drm: Add a simple generator of random permutations Chris Wilson
2016-12-27 11:33   ` Daniel Vetter [this message]
2016-12-22  8:36 ` [PATCH v4 06/38] drm: Add some kselftests for the DRM range manager (struct drm_mm) Chris Wilson
2016-12-27 11:36   ` Daniel Vetter
2016-12-22  8:36 ` [PATCH v4 07/38] drm: kselftest for drm_mm_init() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 08/38] drm: kselftest for drm_mm_debug() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 09/38] drm: kselftest for drm_mm_reserve_node() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 10/38] drm: kselftest for drm_mm_insert_node() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 11/38] drm: kselftest for drm_mm_replace_node() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 12/38] drm: kselftest for drm_mm_insert_node_in_range() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 13/38] drm: kselftest for drm_mm and alignment Chris Wilson
2016-12-22  8:36 ` [PATCH v4 14/38] drm: kselftest for drm_mm and eviction Chris Wilson
2016-12-22  8:36 ` [PATCH v4 15/38] drm: kselftest for drm_mm and range restricted eviction Chris Wilson
2016-12-22  8:36 ` [PATCH v4 16/38] drm: kselftest for drm_mm and top-down allocation Chris Wilson
2016-12-22  8:36 ` [PATCH v4 17/38] drm: kselftest for drm_mm and color adjustment Chris Wilson
2016-12-22  8:36 ` [PATCH v4 18/38] drm: kselftest for drm_mm and color eviction Chris Wilson
2016-12-22  8:36 ` [PATCH v4 19/38] drm: kselftest for drm_mm and restricted " Chris Wilson
2016-12-22  8:36 ` [PATCH v4 20/38] drm/i915: Build DRM range manager selftests for CI Chris Wilson
2016-12-27 13:03   ` Daniel Vetter
2016-12-22  8:36 ` [PATCH v4 21/38] drm: Promote drm_mm alignment to u64 Chris Wilson
2016-12-22  8:36 ` [PATCH v4 22/38] drm: Fix kerneldoc for drm_mm_scan_remove_block() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 23/38] drm: Detect overflow in drm_mm_reserve_node() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 24/38] drm: Simplify drm_mm_clean() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 25/38] drm: Add asserts to catch overflow in drm_mm_init() and drm_mm_init_scan() Chris Wilson
2016-12-27 13:12   ` Daniel Vetter
2016-12-22  8:36 ` [PATCH v4 26/38] drm: Extract struct drm_mm_scan from struct drm_mm Chris Wilson
2016-12-27 15:48   ` Daniel Vetter
2016-12-22  8:36 ` [PATCH v4 27/38] drm: Rename prev_node to hole in drm_mm_scan_add_block() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 28/38] drm: Unconditionally do the range check " Chris Wilson
2016-12-22  8:36 ` [PATCH v4 29/38] drm: Fix application of color vs range restriction when scanning drm_mm Chris Wilson
2016-12-22  8:36 ` [PATCH v4 30/38] drm: Compute tight evictions for drm_mm_scan Chris Wilson
2016-12-28 13:01   ` [Intel-gfx] " Daniel Vetter
2016-12-28 14:36     ` Chris Wilson
2016-12-22  8:36 ` [PATCH v4 31/38] drm: Optimise power-of-two alignments in drm_mm_scan_add_block() Chris Wilson
2016-12-22  8:36 ` [PATCH v4 32/38] drm: Simplify drm_mm scan-list manipulation Chris Wilson
2016-12-22  8:36 ` [PATCH v4 33/38] drm: Apply tight eviction scanning to color_adjust Chris Wilson
2016-12-22  8:36 ` [PATCH v4 34/38] drm: Wrap drm_mm_node.hole_follows Chris Wilson
2016-12-28 13:02   ` Daniel Vetter
2016-12-28 13:31     ` Chris Wilson
2016-12-28 14:31       ` Daniel Vetter
2016-12-28 18:47         ` Chris Wilson
2016-12-22  8:36 ` [PATCH v4 35/38] drm: Apply range restriction after color adjustment when allocation Chris Wilson
2016-12-22  8:36 ` [PATCH v4 36/38] drm: Use drm_mm_insert_node_in_range_generic() for everyone Chris Wilson
2016-12-22  8:36 ` [PATCH v4 37/38] drm: Improve drm_mm search (and fix topdown allocation) with rbtrees Chris Wilson
2016-12-28 11:08   ` Chris Wilson
2016-12-28 13:48   ` Daniel Vetter
2016-12-28 14:34     ` Daniel Vetter
2016-12-22  8:36 ` [PATCH v4 38/38] drm: kselftest for drm_mm and bottom-up allocation Chris Wilson
2016-12-22  9:15 ` ✗ Fi.CI.BAT: warning for series starting with [v4,01/38] drm/i915: Use the MRU stack search after evicting Patchwork
2016-12-22  9:47   ` Imre Deak
2016-12-22 20:53 ` ✓ Fi.CI.BAT: success for series starting with [v4,01/38] drm/i915: Use the MRU stack search after evicting (rev2) Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20161227113350.GH19172@dvetter-linux.ger.corp.intel.com \
    --to=daniel@ffwll.ch \
    --cc=chris@chris-wilson.co.uk \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-gfx@lists.freedesktop.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox