linux/drivers/gpu/drm/i915/i915_gem_tiling.c
Daniel Vetter f691e2f4ce drm/i915: swizzling support for snb/ivb
We have to do this manually. Somebody had a Great Idea.

I've measured speed-ups just a few percent above the noise level
(below 5% for the best case), but no slowdows. Chris Wilson measured
quite a bit more (10-20% above the usual snb variance) on a more
recent and better tuned version of sna, but also recorded a few
slow-downs on benchmarks know for uglier amounts of snb-induced
variance.

v2: Incorporate Ben Widawsky's preliminary review comments and
elaborate a bit about the performance impact in the changelog.

v3: Add a comment as to why we don't need to check the 3rd memory
channel.

v4: Fixup whitespace.

Acked-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: Ben Widawsky <ben@bwidawsk.net>
Reviewed-by: Eric Anholt <eric@anholt.net>
Signed-Off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2012-02-08 23:16:24 +01:00

499 lines
15 KiB
C

/*
* Copyright © 2008 Intel Corporation
*
* Permission is hereby granted, free of charge, to any person obtaining a
* copy of this software and associated documentation files (the "Software"),
* to deal in the Software without restriction, including without limitation
* the rights to use, copy, modify, merge, publish, distribute, sublicense,
* and/or sell copies of the Software, and to permit persons to whom the
* Software is furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice (including the next
* paragraph) shall be included in all copies or substantial portions of the
* Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*
* Authors:
* Eric Anholt <eric@anholt.net>
*
*/
#include "linux/string.h"
#include "linux/bitops.h"
#include "drmP.h"
#include "drm.h"
#include "i915_drm.h"
#include "i915_drv.h"
/** @file i915_gem_tiling.c
*
* Support for managing tiling state of buffer objects.
*
* The idea behind tiling is to increase cache hit rates by rearranging
* pixel data so that a group of pixel accesses are in the same cacheline.
* Performance improvement from doing this on the back/depth buffer are on
* the order of 30%.
*
* Intel architectures make this somewhat more complicated, though, by
* adjustments made to addressing of data when the memory is in interleaved
* mode (matched pairs of DIMMS) to improve memory bandwidth.
* For interleaved memory, the CPU sends every sequential 64 bytes
* to an alternate memory channel so it can get the bandwidth from both.
*
* The GPU also rearranges its accesses for increased bandwidth to interleaved
* memory, and it matches what the CPU does for non-tiled. However, when tiled
* it does it a little differently, since one walks addresses not just in the
* X direction but also Y. So, along with alternating channels when bit
* 6 of the address flips, it also alternates when other bits flip -- Bits 9
* (every 512 bytes, an X tile scanline) and 10 (every two X tile scanlines)
* are common to both the 915 and 965-class hardware.
*
* The CPU also sometimes XORs in higher bits as well, to improve
* bandwidth doing strided access like we do so frequently in graphics. This
* is called "Channel XOR Randomization" in the MCH documentation. The result
* is that the CPU is XORing in either bit 11 or bit 17 to bit 6 of its address
* decode.
*
* All of this bit 6 XORing has an effect on our memory management,
* as we need to make sure that the 3d driver can correctly address object
* contents.
*
* If we don't have interleaved memory, all tiling is safe and no swizzling is
* required.
*
* When bit 17 is XORed in, we simply refuse to tile at all. Bit
* 17 is not just a page offset, so as we page an objet out and back in,
* individual pages in it will have different bit 17 addresses, resulting in
* each 64 bytes being swapped with its neighbor!
*
* Otherwise, if interleaved, we have to tell the 3d driver what the address
* swizzling it needs to do is, since it's writing with the CPU to the pages
* (bit 6 and potentially bit 11 XORed in), and the GPU is reading from the
* pages (bit 6, 9, and 10 XORed in), resulting in a cumulative bit swizzling
* required by the CPU of XORing in bit 6, 9, 10, and potentially 11, in order
* to match what the GPU expects.
*/
/**
* Detects bit 6 swizzling of address lookup between IGD access and CPU
* access through main memory.
*/
void
i915_gem_detect_bit_6_swizzle(struct drm_device *dev)
{
drm_i915_private_t *dev_priv = dev->dev_private;
uint32_t swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN;
uint32_t swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN;
if (INTEL_INFO(dev)->gen >= 6) {
uint32_t dimm_c0, dimm_c1;
dimm_c0 = I915_READ(MAD_DIMM_C0);
dimm_c1 = I915_READ(MAD_DIMM_C1);
dimm_c0 &= MAD_DIMM_A_SIZE_MASK | MAD_DIMM_B_SIZE_MASK;
dimm_c1 &= MAD_DIMM_A_SIZE_MASK | MAD_DIMM_B_SIZE_MASK;
/* Enable swizzling when the channels are populated with
* identically sized dimms. We don't need to check the 3rd
* channel because no cpu with gpu attached ships in that
* configuration. Also, swizzling only makes sense for 2
* channels anyway. */
if (dimm_c0 == dimm_c1) {
swizzle_x = I915_BIT_6_SWIZZLE_9_10;
swizzle_y = I915_BIT_6_SWIZZLE_9;
} else {
swizzle_x = I915_BIT_6_SWIZZLE_NONE;
swizzle_y = I915_BIT_6_SWIZZLE_NONE;
}
} else if (IS_GEN5(dev)) {
/* On Ironlake whatever DRAM config, GPU always do
* same swizzling setup.
*/
swizzle_x = I915_BIT_6_SWIZZLE_9_10;
swizzle_y = I915_BIT_6_SWIZZLE_9;
} else if (IS_GEN2(dev)) {
/* As far as we know, the 865 doesn't have these bit 6
* swizzling issues.
*/
swizzle_x = I915_BIT_6_SWIZZLE_NONE;
swizzle_y = I915_BIT_6_SWIZZLE_NONE;
} else if (IS_MOBILE(dev) || (IS_GEN3(dev) && !IS_G33(dev))) {
uint32_t dcc;
/* On 9xx chipsets, channel interleave by the CPU is
* determined by DCC. For single-channel, neither the CPU
* nor the GPU do swizzling. For dual channel interleaved,
* the GPU's interleave is bit 9 and 10 for X tiled, and bit
* 9 for Y tiled. The CPU's interleave is independent, and
* can be based on either bit 11 (haven't seen this yet) or
* bit 17 (common).
*/
dcc = I915_READ(DCC);
switch (dcc & DCC_ADDRESSING_MODE_MASK) {
case DCC_ADDRESSING_MODE_SINGLE_CHANNEL:
case DCC_ADDRESSING_MODE_DUAL_CHANNEL_ASYMMETRIC:
swizzle_x = I915_BIT_6_SWIZZLE_NONE;
swizzle_y = I915_BIT_6_SWIZZLE_NONE;
break;
case DCC_ADDRESSING_MODE_DUAL_CHANNEL_INTERLEAVED:
if (dcc & DCC_CHANNEL_XOR_DISABLE) {
/* This is the base swizzling by the GPU for
* tiled buffers.
*/
swizzle_x = I915_BIT_6_SWIZZLE_9_10;
swizzle_y = I915_BIT_6_SWIZZLE_9;
} else if ((dcc & DCC_CHANNEL_XOR_BIT_17) == 0) {
/* Bit 11 swizzling by the CPU in addition. */
swizzle_x = I915_BIT_6_SWIZZLE_9_10_11;
swizzle_y = I915_BIT_6_SWIZZLE_9_11;
} else {
/* Bit 17 swizzling by the CPU in addition. */
swizzle_x = I915_BIT_6_SWIZZLE_9_10_17;
swizzle_y = I915_BIT_6_SWIZZLE_9_17;
}
break;
}
if (dcc == 0xffffffff) {
DRM_ERROR("Couldn't read from MCHBAR. "
"Disabling tiling.\n");
swizzle_x = I915_BIT_6_SWIZZLE_UNKNOWN;
swizzle_y = I915_BIT_6_SWIZZLE_UNKNOWN;
}
} else {
/* The 965, G33, and newer, have a very flexible memory
* configuration. It will enable dual-channel mode
* (interleaving) on as much memory as it can, and the GPU
* will additionally sometimes enable different bit 6
* swizzling for tiled objects from the CPU.
*
* Here's what I found on the G965:
* slot fill memory size swizzling
* 0A 0B 1A 1B 1-ch 2-ch
* 512 0 0 0 512 0 O
* 512 0 512 0 16 1008 X
* 512 0 0 512 16 1008 X
* 0 512 0 512 16 1008 X
* 1024 1024 1024 0 2048 1024 O
*
* We could probably detect this based on either the DRB
* matching, which was the case for the swizzling required in
* the table above, or from the 1-ch value being less than
* the minimum size of a rank.
*/
if (I915_READ16(C0DRB3) != I915_READ16(C1DRB3)) {
swizzle_x = I915_BIT_6_SWIZZLE_NONE;
swizzle_y = I915_BIT_6_SWIZZLE_NONE;
} else {
swizzle_x = I915_BIT_6_SWIZZLE_9_10;
swizzle_y = I915_BIT_6_SWIZZLE_9;
}
}
dev_priv->mm.bit_6_swizzle_x = swizzle_x;
dev_priv->mm.bit_6_swizzle_y = swizzle_y;
}
/* Check pitch constriants for all chips & tiling formats */
static bool
i915_tiling_ok(struct drm_device *dev, int stride, int size, int tiling_mode)
{
int tile_width;
/* Linear is always fine */
if (tiling_mode == I915_TILING_NONE)
return true;
if (IS_GEN2(dev) ||
(tiling_mode == I915_TILING_Y && HAS_128_BYTE_Y_TILING(dev)))
tile_width = 128;
else
tile_width = 512;
/* check maximum stride & object size */
if (INTEL_INFO(dev)->gen >= 4) {
/* i965 stores the end address of the gtt mapping in the fence
* reg, so dont bother to check the size */
if (stride / 128 > I965_FENCE_MAX_PITCH_VAL)
return false;
} else {
if (stride > 8192)
return false;
if (IS_GEN3(dev)) {
if (size > I830_FENCE_MAX_SIZE_VAL << 20)
return false;
} else {
if (size > I830_FENCE_MAX_SIZE_VAL << 19)
return false;
}
}
/* 965+ just needs multiples of tile width */
if (INTEL_INFO(dev)->gen >= 4) {
if (stride & (tile_width - 1))
return false;
return true;
}
/* Pre-965 needs power of two tile widths */
if (stride < tile_width)
return false;
if (stride & (stride - 1))
return false;
return true;
}
/* Is the current GTT allocation valid for the change in tiling? */
static bool
i915_gem_object_fence_ok(struct drm_i915_gem_object *obj, int tiling_mode)
{
u32 size;
if (tiling_mode == I915_TILING_NONE)
return true;
if (INTEL_INFO(obj->base.dev)->gen >= 4)
return true;
if (INTEL_INFO(obj->base.dev)->gen == 3) {
if (obj->gtt_offset & ~I915_FENCE_START_MASK)
return false;
} else {
if (obj->gtt_offset & ~I830_FENCE_START_MASK)
return false;
}
/*
* Previous chips need to be aligned to the size of the smallest
* fence register that can contain the object.
*/
if (INTEL_INFO(obj->base.dev)->gen == 3)
size = 1024*1024;
else
size = 512*1024;
while (size < obj->base.size)
size <<= 1;
if (obj->gtt_space->size != size)
return false;
if (obj->gtt_offset & (size - 1))
return false;
return true;
}
/**
* Sets the tiling mode of an object, returning the required swizzling of
* bit 6 of addresses in the object.
*/
int
i915_gem_set_tiling(struct drm_device *dev, void *data,
struct drm_file *file)
{
struct drm_i915_gem_set_tiling *args = data;
drm_i915_private_t *dev_priv = dev->dev_private;
struct drm_i915_gem_object *obj;
int ret = 0;
obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
if (&obj->base == NULL)
return -ENOENT;
if (!i915_tiling_ok(dev,
args->stride, obj->base.size, args->tiling_mode)) {
drm_gem_object_unreference_unlocked(&obj->base);
return -EINVAL;
}
if (obj->pin_count) {
drm_gem_object_unreference_unlocked(&obj->base);
return -EBUSY;
}
if (args->tiling_mode == I915_TILING_NONE) {
args->swizzle_mode = I915_BIT_6_SWIZZLE_NONE;
args->stride = 0;
} else {
if (args->tiling_mode == I915_TILING_X)
args->swizzle_mode = dev_priv->mm.bit_6_swizzle_x;
else
args->swizzle_mode = dev_priv->mm.bit_6_swizzle_y;
/* Hide bit 17 swizzling from the user. This prevents old Mesa
* from aborting the application on sw fallbacks to bit 17,
* and we use the pread/pwrite bit17 paths to swizzle for it.
* If there was a user that was relying on the swizzle
* information for drm_intel_bo_map()ed reads/writes this would
* break it, but we don't have any of those.
*/
if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_17)
args->swizzle_mode = I915_BIT_6_SWIZZLE_9;
if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_10_17)
args->swizzle_mode = I915_BIT_6_SWIZZLE_9_10;
/* If we can't handle the swizzling, make it untiled. */
if (args->swizzle_mode == I915_BIT_6_SWIZZLE_UNKNOWN) {
args->tiling_mode = I915_TILING_NONE;
args->swizzle_mode = I915_BIT_6_SWIZZLE_NONE;
args->stride = 0;
}
}
mutex_lock(&dev->struct_mutex);
if (args->tiling_mode != obj->tiling_mode ||
args->stride != obj->stride) {
/* We need to rebind the object if its current allocation
* no longer meets the alignment restrictions for its new
* tiling mode. Otherwise we can just leave it alone, but
* need to ensure that any fence register is cleared.
*/
i915_gem_release_mmap(obj);
obj->map_and_fenceable =
obj->gtt_space == NULL ||
(obj->gtt_offset + obj->base.size <= dev_priv->mm.gtt_mappable_end &&
i915_gem_object_fence_ok(obj, args->tiling_mode));
/* Rebind if we need a change of alignment */
if (!obj->map_and_fenceable) {
u32 unfenced_alignment =
i915_gem_get_unfenced_gtt_alignment(dev,
obj->base.size,
args->tiling_mode);
if (obj->gtt_offset & (unfenced_alignment - 1))
ret = i915_gem_object_unbind(obj);
}
if (ret == 0) {
obj->tiling_changed = true;
obj->tiling_mode = args->tiling_mode;
obj->stride = args->stride;
}
}
/* we have to maintain this existing ABI... */
args->stride = obj->stride;
args->tiling_mode = obj->tiling_mode;
drm_gem_object_unreference(&obj->base);
mutex_unlock(&dev->struct_mutex);
return ret;
}
/**
* Returns the current tiling mode and required bit 6 swizzling for the object.
*/
int
i915_gem_get_tiling(struct drm_device *dev, void *data,
struct drm_file *file)
{
struct drm_i915_gem_get_tiling *args = data;
drm_i915_private_t *dev_priv = dev->dev_private;
struct drm_i915_gem_object *obj;
obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle));
if (&obj->base == NULL)
return -ENOENT;
mutex_lock(&dev->struct_mutex);
args->tiling_mode = obj->tiling_mode;
switch (obj->tiling_mode) {
case I915_TILING_X:
args->swizzle_mode = dev_priv->mm.bit_6_swizzle_x;
break;
case I915_TILING_Y:
args->swizzle_mode = dev_priv->mm.bit_6_swizzle_y;
break;
case I915_TILING_NONE:
args->swizzle_mode = I915_BIT_6_SWIZZLE_NONE;
break;
default:
DRM_ERROR("unknown tiling mode\n");
}
/* Hide bit 17 from the user -- see comment in i915_gem_set_tiling */
if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_17)
args->swizzle_mode = I915_BIT_6_SWIZZLE_9;
if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_10_17)
args->swizzle_mode = I915_BIT_6_SWIZZLE_9_10;
drm_gem_object_unreference(&obj->base);
mutex_unlock(&dev->struct_mutex);
return 0;
}
/**
* Swap every 64 bytes of this page around, to account for it having a new
* bit 17 of its physical address and therefore being interpreted differently
* by the GPU.
*/
static void
i915_gem_swizzle_page(struct page *page)
{
char temp[64];
char *vaddr;
int i;
vaddr = kmap(page);
for (i = 0; i < PAGE_SIZE; i += 128) {
memcpy(temp, &vaddr[i], 64);
memcpy(&vaddr[i], &vaddr[i + 64], 64);
memcpy(&vaddr[i + 64], temp, 64);
}
kunmap(page);
}
void
i915_gem_object_do_bit_17_swizzle(struct drm_i915_gem_object *obj)
{
int page_count = obj->base.size >> PAGE_SHIFT;
int i;
if (obj->bit_17 == NULL)
return;
for (i = 0; i < page_count; i++) {
char new_bit_17 = page_to_phys(obj->pages[i]) >> 17;
if ((new_bit_17 & 0x1) !=
(test_bit(i, obj->bit_17) != 0)) {
i915_gem_swizzle_page(obj->pages[i]);
set_page_dirty(obj->pages[i]);
}
}
}
void
i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj)
{
int page_count = obj->base.size >> PAGE_SHIFT;
int i;
if (obj->bit_17 == NULL) {
obj->bit_17 = kmalloc(BITS_TO_LONGS(page_count) *
sizeof(long), GFP_KERNEL);
if (obj->bit_17 == NULL) {
DRM_ERROR("Failed to allocate memory for bit 17 "
"record\n");
return;
}
}
for (i = 0; i < page_count; i++) {
if (page_to_phys(obj->pages[i]) & (1 << 17))
__set_bit(i, obj->bit_17);
else
__clear_bit(i, obj->bit_17);
}
}