drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
/*
|
|
|
|
* Copyright © 2014 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.
|
|
|
|
*
|
|
|
|
* Please try to maintain the following order within this file unless it makes
|
|
|
|
* sense to do otherwise. From top to bottom:
|
|
|
|
* 1. typedefs
|
|
|
|
* 2. #defines, and macros
|
|
|
|
* 3. structure definitions
|
|
|
|
* 4. function prototypes
|
|
|
|
*
|
|
|
|
* Within each section, please try to order by generation in ascending order,
|
|
|
|
* from top to bottom (ie. gen6 on the top, gen8 on the bottom).
|
|
|
|
*/
|
|
|
|
|
|
|
|
#ifndef __I915_GEM_GTT_H__
|
|
|
|
#define __I915_GEM_GTT_H__
|
|
|
|
|
2016-04-28 15:56:39 +07:00
|
|
|
#include <linux/io-mapping.h>
|
2019-06-21 01:37:05 +07:00
|
|
|
#include <linux/kref.h>
|
2016-11-11 17:43:54 +07:00
|
|
|
#include <linux/mm.h>
|
2017-02-15 15:43:40 +07:00
|
|
|
#include <linux/pagevec.h>
|
2019-06-21 01:37:05 +07:00
|
|
|
#include <linux/workqueue.h>
|
|
|
|
|
|
|
|
#include <drm/drm_mm.h>
|
2016-04-28 15:56:39 +07:00
|
|
|
|
2019-04-25 00:48:39 +07:00
|
|
|
#include "gt/intel_reset.h"
|
2019-06-13 14:32:54 +07:00
|
|
|
#include "i915_gem_fence_reg.h"
|
2018-02-21 16:56:36 +07:00
|
|
|
#include "i915_request.h"
|
2019-05-28 16:29:50 +07:00
|
|
|
#include "i915_scatterlist.h"
|
2017-02-15 15:43:40 +07:00
|
|
|
#include "i915_selftest.h"
|
2019-06-21 14:08:10 +07:00
|
|
|
#include "gt/intel_timeline.h"
|
2016-08-04 13:52:44 +07:00
|
|
|
|
2018-10-25 16:18:22 +07:00
|
|
|
#define I915_GTT_PAGE_SIZE_4K BIT_ULL(12)
|
|
|
|
#define I915_GTT_PAGE_SIZE_64K BIT_ULL(16)
|
|
|
|
#define I915_GTT_PAGE_SIZE_2M BIT_ULL(21)
|
2017-10-07 05:18:16 +07:00
|
|
|
|
|
|
|
#define I915_GTT_PAGE_SIZE I915_GTT_PAGE_SIZE_4K
|
|
|
|
#define I915_GTT_MAX_PAGE_SIZE I915_GTT_PAGE_SIZE_2M
|
|
|
|
|
2018-10-25 16:18:23 +07:00
|
|
|
#define I915_GTT_PAGE_MASK -I915_GTT_PAGE_SIZE
|
|
|
|
|
2017-01-10 21:47:34 +07:00
|
|
|
#define I915_GTT_MIN_ALIGNMENT I915_GTT_PAGE_SIZE
|
|
|
|
|
2016-08-18 23:17:00 +07:00
|
|
|
#define I915_FENCE_REG_NONE -1
|
|
|
|
#define I915_MAX_NUM_FENCES 32
|
|
|
|
/* 32 fences + sign bit for FENCE_REG_NONE */
|
|
|
|
#define I915_MAX_NUM_FENCE_BITS 6
|
|
|
|
|
2014-08-06 20:04:47 +07:00
|
|
|
struct drm_i915_file_private;
|
2019-05-28 16:29:42 +07:00
|
|
|
struct drm_i915_gem_object;
|
2018-06-07 22:40:46 +07:00
|
|
|
struct i915_vma;
|
2019-06-21 14:07:51 +07:00
|
|
|
struct intel_gt;
|
2014-08-06 20:04:47 +07:00
|
|
|
|
2017-02-15 15:43:57 +07:00
|
|
|
typedef u32 gen6_pte_t;
|
|
|
|
typedef u64 gen8_pte_t;
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2018-06-05 22:37:58 +07:00
|
|
|
#define ggtt_total_entries(ggtt) ((ggtt)->vm.total >> PAGE_SHIFT)
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
|
|
|
/* gen6-hsw has bit 11-4 for physical addr bit 39-32 */
|
|
|
|
#define GEN6_GTT_ADDR_ENCODE(addr) ((addr) | (((addr) >> 28) & 0xff0))
|
|
|
|
#define GEN6_PTE_ADDR_ENCODE(addr) GEN6_GTT_ADDR_ENCODE(addr)
|
|
|
|
#define GEN6_PDE_ADDR_ENCODE(addr) GEN6_GTT_ADDR_ENCODE(addr)
|
|
|
|
#define GEN6_PTE_CACHE_LLC (2 << 1)
|
|
|
|
#define GEN6_PTE_UNCACHED (1 << 1)
|
|
|
|
#define GEN6_PTE_VALID (1 << 0)
|
|
|
|
|
2017-02-15 15:43:46 +07:00
|
|
|
#define I915_PTES(pte_len) ((unsigned int)(PAGE_SIZE / (pte_len)))
|
2015-03-16 23:00:54 +07:00
|
|
|
#define I915_PTE_MASK(pte_len) (I915_PTES(pte_len) - 1)
|
|
|
|
#define I915_PDES 512
|
|
|
|
#define I915_PDE_MASK (I915_PDES - 1)
|
2015-03-16 23:00:56 +07:00
|
|
|
#define NUM_PTE(pde_shift) (1 << (pde_shift - PAGE_SHIFT))
|
2015-03-16 23:00:54 +07:00
|
|
|
|
|
|
|
#define GEN6_PTES I915_PTES(sizeof(gen6_pte_t))
|
|
|
|
#define GEN6_PD_SIZE (I915_PDES * PAGE_SIZE)
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
#define GEN6_PD_ALIGN (PAGE_SIZE * 16)
|
2015-03-16 23:00:56 +07:00
|
|
|
#define GEN6_PDE_SHIFT 22
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
#define GEN6_PDE_VALID (1 << 0)
|
|
|
|
|
|
|
|
#define GEN7_PTE_CACHE_L3_LLC (3 << 1)
|
|
|
|
|
|
|
|
#define BYT_PTE_SNOOPED_BY_CPU_CACHES (1 << 2)
|
|
|
|
#define BYT_PTE_WRITEABLE (1 << 1)
|
|
|
|
|
|
|
|
/* Cacheability Control is a 4-bit value. The low three bits are stored in bits
|
|
|
|
* 3:1 of the PTE, while the fourth bit is stored in bit 11 of the PTE.
|
|
|
|
*/
|
|
|
|
#define HSW_CACHEABILITY_CONTROL(bits) ((((bits) & 0x7) << 1) | \
|
|
|
|
(((bits) & 0x8) << (11 - 3)))
|
|
|
|
#define HSW_WB_LLC_AGE3 HSW_CACHEABILITY_CONTROL(0x2)
|
|
|
|
#define HSW_WB_LLC_AGE0 HSW_CACHEABILITY_CONTROL(0x3)
|
|
|
|
#define HSW_WB_ELLC_LLC_AGE3 HSW_CACHEABILITY_CONTROL(0x8)
|
|
|
|
#define HSW_WB_ELLC_LLC_AGE0 HSW_CACHEABILITY_CONTROL(0xb)
|
|
|
|
#define HSW_WT_ELLC_LLC_AGE3 HSW_CACHEABILITY_CONTROL(0x7)
|
|
|
|
#define HSW_WT_ELLC_LLC_AGE0 HSW_CACHEABILITY_CONTROL(0x6)
|
|
|
|
#define HSW_PTE_UNCACHED (0)
|
|
|
|
#define HSW_GTT_ADDR_ENCODE(addr) ((addr) | (((addr) >> 28) & 0x7f0))
|
|
|
|
#define HSW_PTE_ADDR_ENCODE(addr) HSW_GTT_ADDR_ENCODE(addr)
|
|
|
|
|
2019-07-12 18:27:25 +07:00
|
|
|
/*
|
|
|
|
* GEN8 32b style address is defined as a 3 level page table:
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
* 31:30 | 29:21 | 20:12 | 11:0
|
|
|
|
* PDPE | PDE | PTE | offset
|
|
|
|
* The difference as compared to normal x86 3 level page table is the PDPEs are
|
|
|
|
* programmed via register.
|
2019-07-12 18:27:25 +07:00
|
|
|
*
|
|
|
|
* GEN8 48b style address is defined as a 4 level page table:
|
2015-08-03 15:52:01 +07:00
|
|
|
* 47:39 | 38:30 | 29:21 | 20:12 | 11:0
|
|
|
|
* PML4E | PDPE | PDE | PTE | offset
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
*/
|
2019-07-12 18:27:25 +07:00
|
|
|
#define GEN8_3LVL_PDPES 4
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2017-09-14 19:39:41 +07:00
|
|
|
#define PPAT_UNCACHED (_PAGE_PWT | _PAGE_PCD)
|
|
|
|
#define PPAT_CACHED_PDE 0 /* WB LLC */
|
|
|
|
#define PPAT_CACHED _PAGE_PAT /* WB LLCeLLC */
|
|
|
|
#define PPAT_DISPLAY_ELLC _PAGE_PCD /* WT eLLC */
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2014-04-09 17:28:01 +07:00
|
|
|
#define CHV_PPAT_SNOOP (1<<6)
|
2017-09-08 23:11:30 +07:00
|
|
|
#define GEN8_PPAT_AGE(x) ((x)<<4)
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
#define GEN8_PPAT_LLCeLLC (3<<2)
|
|
|
|
#define GEN8_PPAT_LLCELLC (2<<2)
|
|
|
|
#define GEN8_PPAT_LLC (1<<2)
|
|
|
|
#define GEN8_PPAT_WB (3<<0)
|
|
|
|
#define GEN8_PPAT_WT (2<<0)
|
|
|
|
#define GEN8_PPAT_WC (1<<0)
|
|
|
|
#define GEN8_PPAT_UC (0<<0)
|
|
|
|
#define GEN8_PPAT_ELLC_OVERRIDE (0<<2)
|
2017-02-15 15:43:57 +07:00
|
|
|
#define GEN8_PPAT(i, x) ((u64)(x) << ((i) * 8))
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2017-10-07 05:18:26 +07:00
|
|
|
#define GEN8_PDE_IPS_64K BIT(11)
|
2017-10-07 05:18:24 +07:00
|
|
|
#define GEN8_PDE_PS_2M BIT(7)
|
|
|
|
|
2019-08-30 03:19:19 +07:00
|
|
|
#define for_each_sgt_daddr(__dp, __iter, __sgt) \
|
|
|
|
__for_each_sgt_daddr(__dp, __iter, __sgt, I915_GTT_PAGE_SIZE)
|
2016-11-11 17:43:54 +07:00
|
|
|
|
2019-05-09 19:21:52 +07:00
|
|
|
struct intel_remapped_plane_info {
|
|
|
|
/* in gtt pages */
|
|
|
|
unsigned int width, height, stride, offset;
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
struct intel_remapped_info {
|
|
|
|
struct intel_remapped_plane_info plane[2];
|
|
|
|
unsigned int unused_mbz;
|
|
|
|
} __packed;
|
|
|
|
|
2015-03-23 18:10:36 +07:00
|
|
|
struct intel_rotation_info {
|
2019-05-09 19:21:52 +07:00
|
|
|
struct intel_remapped_plane_info plane[2];
|
2017-01-14 07:28:22 +07:00
|
|
|
} __packed;
|
|
|
|
|
2017-01-14 07:28:21 +07:00
|
|
|
struct intel_partial_info {
|
|
|
|
u64 offset;
|
|
|
|
unsigned int size;
|
2017-01-14 07:28:22 +07:00
|
|
|
} __packed;
|
|
|
|
|
2017-01-14 07:28:23 +07:00
|
|
|
enum i915_ggtt_view_type {
|
|
|
|
I915_GGTT_VIEW_NORMAL = 0,
|
|
|
|
I915_GGTT_VIEW_ROTATED = sizeof(struct intel_rotation_info),
|
|
|
|
I915_GGTT_VIEW_PARTIAL = sizeof(struct intel_partial_info),
|
2019-05-09 19:21:52 +07:00
|
|
|
I915_GGTT_VIEW_REMAPPED = sizeof(struct intel_remapped_info),
|
2017-01-14 07:28:23 +07:00
|
|
|
};
|
|
|
|
|
2018-08-28 20:37:23 +07:00
|
|
|
static inline void assert_i915_gem_gtt_types(void)
|
2017-01-14 07:28:23 +07:00
|
|
|
{
|
2018-08-28 20:37:23 +07:00
|
|
|
BUILD_BUG_ON(sizeof(struct intel_rotation_info) != 8*sizeof(unsigned int));
|
|
|
|
BUILD_BUG_ON(sizeof(struct intel_partial_info) != sizeof(u64) + sizeof(unsigned int));
|
2019-05-09 19:21:52 +07:00
|
|
|
BUILD_BUG_ON(sizeof(struct intel_remapped_info) != 9*sizeof(unsigned int));
|
|
|
|
|
|
|
|
/* Check that rotation/remapped shares offsets for simplicity */
|
|
|
|
BUILD_BUG_ON(offsetof(struct intel_remapped_info, plane[0]) !=
|
|
|
|
offsetof(struct intel_rotation_info, plane[0]));
|
|
|
|
BUILD_BUG_ON(offsetofend(struct intel_remapped_info, plane[1]) !=
|
|
|
|
offsetofend(struct intel_rotation_info, plane[1]));
|
2018-08-28 20:37:23 +07:00
|
|
|
|
2017-01-14 07:28:23 +07:00
|
|
|
/* As we encode the size of each branch inside the union into its type,
|
|
|
|
* we have to be careful that each branch has a unique size.
|
|
|
|
*/
|
|
|
|
switch ((enum i915_ggtt_view_type)0) {
|
|
|
|
case I915_GGTT_VIEW_NORMAL:
|
|
|
|
case I915_GGTT_VIEW_PARTIAL:
|
|
|
|
case I915_GGTT_VIEW_ROTATED:
|
2019-05-09 19:21:52 +07:00
|
|
|
case I915_GGTT_VIEW_REMAPPED:
|
2017-01-14 07:28:23 +07:00
|
|
|
/* gcc complains if these are identical cases */
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-12-11 00:27:58 +07:00
|
|
|
struct i915_ggtt_view {
|
|
|
|
enum i915_ggtt_view_type type;
|
2015-05-06 18:35:38 +07:00
|
|
|
union {
|
2017-01-14 07:28:23 +07:00
|
|
|
/* Members need to contain no holes/padding */
|
2017-01-14 07:28:21 +07:00
|
|
|
struct intel_partial_info partial;
|
2016-01-21 02:05:22 +07:00
|
|
|
struct intel_rotation_info rotated;
|
2019-05-09 19:21:52 +07:00
|
|
|
struct intel_remapped_info remapped;
|
2017-01-14 07:28:25 +07:00
|
|
|
};
|
2014-12-11 00:27:58 +07:00
|
|
|
};
|
|
|
|
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
enum i915_cache_level;
|
2014-12-11 00:27:58 +07:00
|
|
|
|
2016-11-11 17:43:54 +07:00
|
|
|
struct i915_vma;
|
2016-08-15 16:49:07 +07:00
|
|
|
|
2015-06-25 22:35:07 +07:00
|
|
|
struct i915_page_dma {
|
2015-02-24 23:22:34 +07:00
|
|
|
struct page *page;
|
2015-06-25 22:35:07 +07:00
|
|
|
union {
|
|
|
|
dma_addr_t daddr;
|
|
|
|
|
|
|
|
/* For gen6/gen7 only. This is the offset in the GGTT
|
|
|
|
* where the page directory entries for PPGTT begin
|
|
|
|
*/
|
2017-02-15 15:43:57 +07:00
|
|
|
u32 ggtt_offset;
|
2015-06-25 22:35:07 +07:00
|
|
|
};
|
|
|
|
};
|
|
|
|
|
2019-07-12 16:43:26 +07:00
|
|
|
struct i915_page_scratch {
|
|
|
|
struct i915_page_dma base;
|
|
|
|
u64 encode;
|
|
|
|
};
|
|
|
|
|
2015-06-25 22:35:07 +07:00
|
|
|
struct i915_page_table {
|
|
|
|
struct i915_page_dma base;
|
2019-06-14 23:43:42 +07:00
|
|
|
atomic_t used;
|
2015-02-24 23:22:34 +07:00
|
|
|
};
|
|
|
|
|
2015-04-08 18:13:23 +07:00
|
|
|
struct i915_page_directory {
|
2019-07-12 16:43:22 +07:00
|
|
|
struct i915_page_table pt;
|
2019-06-04 22:38:30 +07:00
|
|
|
spinlock_t lock;
|
2019-06-14 23:43:42 +07:00
|
|
|
void *entry[512];
|
2015-08-03 15:52:01 +07:00
|
|
|
};
|
|
|
|
|
2019-07-12 14:58:18 +07:00
|
|
|
#define __px_choose_expr(x, type, expr, other) \
|
|
|
|
__builtin_choose_expr( \
|
|
|
|
__builtin_types_compatible_p(typeof(x), type) || \
|
|
|
|
__builtin_types_compatible_p(typeof(x), const type), \
|
|
|
|
({ type __x = (type)(x); expr; }), \
|
|
|
|
other)
|
|
|
|
|
|
|
|
#define px_base(px) \
|
|
|
|
__px_choose_expr(px, struct i915_page_dma *, __x, \
|
2019-07-12 16:43:26 +07:00
|
|
|
__px_choose_expr(px, struct i915_page_scratch *, &__x->base, \
|
2019-07-12 14:58:18 +07:00
|
|
|
__px_choose_expr(px, struct i915_page_table *, &__x->base, \
|
2019-07-12 16:43:22 +07:00
|
|
|
__px_choose_expr(px, struct i915_page_directory *, &__x->pt.base, \
|
2019-07-12 16:43:26 +07:00
|
|
|
(void)0))))
|
2019-07-12 14:58:18 +07:00
|
|
|
#define px_dma(px) (px_base(px)->daddr)
|
|
|
|
|
2019-07-12 16:43:22 +07:00
|
|
|
#define px_pt(px) \
|
|
|
|
__px_choose_expr(px, struct i915_page_table *, __x, \
|
|
|
|
__px_choose_expr(px, struct i915_page_directory *, &__x->pt, \
|
|
|
|
(void)0))
|
|
|
|
#define px_used(px) (&px_pt(px)->used)
|
|
|
|
|
2018-06-07 22:40:46 +07:00
|
|
|
struct i915_vma_ops {
|
|
|
|
/* Map an object into an address space with the given cache flags. */
|
|
|
|
int (*bind_vma)(struct i915_vma *vma,
|
|
|
|
enum i915_cache_level cache_level,
|
|
|
|
u32 flags);
|
|
|
|
/*
|
|
|
|
* Unmap an object from an address space. This usually consists of
|
|
|
|
* setting the valid PTE entries to a reserved scratch page.
|
|
|
|
*/
|
|
|
|
void (*unbind_vma)(struct i915_vma *vma);
|
|
|
|
|
|
|
|
int (*set_pages)(struct i915_vma *vma);
|
|
|
|
void (*clear_pages)(struct i915_vma *vma);
|
|
|
|
};
|
|
|
|
|
2018-07-05 01:55:18 +07:00
|
|
|
struct pagestash {
|
|
|
|
spinlock_t lock;
|
|
|
|
struct pagevec pvec;
|
|
|
|
};
|
|
|
|
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
struct i915_address_space {
|
2019-06-11 16:12:37 +07:00
|
|
|
struct kref ref;
|
2019-06-21 01:37:05 +07:00
|
|
|
struct rcu_work rcu;
|
2019-06-11 16:12:37 +07:00
|
|
|
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
struct drm_mm mm;
|
2019-06-21 14:07:59 +07:00
|
|
|
struct intel_gt *gt;
|
2016-11-29 16:50:08 +07:00
|
|
|
struct drm_i915_private *i915;
|
2017-02-15 15:43:40 +07:00
|
|
|
struct device *dma;
|
2016-08-04 13:52:25 +07:00
|
|
|
/* Every address space belongs to a struct file - except for the global
|
|
|
|
* GTT that is owned by the driver (and so @file is set to NULL). In
|
|
|
|
* principle, no information should leak from one context to another
|
|
|
|
* (or between files/processes etc) unless explicitly shared by the
|
|
|
|
* owner. Tracking the owner is important in order to free up per-file
|
|
|
|
* objects along with the file, to aide resource tracking, and to
|
|
|
|
* assign blame.
|
|
|
|
*/
|
|
|
|
struct drm_i915_file_private *file;
|
2015-06-25 22:35:05 +07:00
|
|
|
u64 total; /* size addr space maps (ex. 2GB for ggtt) */
|
2017-05-31 09:35:52 +07:00
|
|
|
u64 reserved; /* size addr space reserved */
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2019-10-04 20:39:57 +07:00
|
|
|
unsigned int bind_async_flags;
|
|
|
|
|
drm/i915: Pull i915_vma_pin under the vm->mutex
Replace the struct_mutex requirement for pinning the i915_vma with the
local vm->mutex instead. Note that the vm->mutex is tainted by the
shrinker (we require unbinding from inside fs-reclaim) and so we cannot
allocate while holding that mutex. Instead we have to preallocate
workers to do allocate and apply the PTE updates after we have we
reserved their slot in the drm_mm (using fences to order the PTE writes
with the GPU work and with later unbind).
In adding the asynchronous vma binding, one subtle requirement is to
avoid coupling the binding fence into the backing object->resv. That is
the asynchronous binding only applies to the vma timeline itself and not
to the pages as that is a more global timeline (the binding of one vma
does not need to be ordered with another vma, nor does the implicit GEM
fencing depend on a vma, only on writes to the backing store). Keeping
the vma binding distinct from the backing store timelines is verified by
a number of async gem_exec_fence and gem_exec_schedule tests. The way we
do this is quite simple, we keep the fence for the vma binding separate
and only wait on it as required, and never add it to the obj->resv
itself.
Another consequence in reducing the locking around the vma is the
destruction of the vma is no longer globally serialised by struct_mutex.
A natural solution would be to add a kref to i915_vma, but that requires
decoupling the reference cycles, possibly by introducing a new
i915_mm_pages object that is own by both obj->mm and vma->pages.
However, we have not taken that route due to the overshadowing lmem/ttm
discussions, and instead play a series of complicated games with
trylocks to (hopefully) ensure that only one destruction path is called!
v2: Add some commentary, and some helpers to reduce patch churn.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191004134015.13204-4-chris@chris-wilson.co.uk
2019-10-04 20:39:58 +07:00
|
|
|
/*
|
|
|
|
* Each active user context has its own address space (in full-ppgtt).
|
|
|
|
* Since the vm may be shared between multiple contexts, we count how
|
|
|
|
* many contexts keep us "open". Once open hits zero, we are closed
|
|
|
|
* and do not allow any new attachments, and proceed to shutdown our
|
|
|
|
* vma and page directories.
|
|
|
|
*/
|
|
|
|
atomic_t open;
|
2016-08-04 13:52:46 +07:00
|
|
|
|
2018-07-11 14:36:02 +07:00
|
|
|
struct mutex mutex; /* protects vma and our lists */
|
2019-01-15 04:59:56 +07:00
|
|
|
#define VM_CLASS_GGTT 0
|
|
|
|
#define VM_CLASS_PPGTT 1
|
2018-07-11 14:36:02 +07:00
|
|
|
|
2019-07-12 16:43:26 +07:00
|
|
|
struct i915_page_scratch scratch[4];
|
|
|
|
unsigned int scratch_order;
|
2019-07-12 16:43:24 +07:00
|
|
|
unsigned int top;
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
|
|
|
/**
|
2019-01-28 17:23:52 +07:00
|
|
|
* List of vma currently bound.
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
*/
|
2019-01-28 17:23:52 +07:00
|
|
|
struct list_head bound_list;
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2018-07-05 01:55:18 +07:00
|
|
|
struct pagestash free_pages;
|
2018-07-13 01:53:11 +07:00
|
|
|
|
2018-08-31 21:36:43 +07:00
|
|
|
/* Global GTT */
|
|
|
|
bool is_ggtt:1;
|
|
|
|
|
2018-07-13 01:53:11 +07:00
|
|
|
/* Some systems require uncached updates of the page directories */
|
|
|
|
bool pt_kmap_wc:1;
|
|
|
|
|
|
|
|
/* Some systems support read-only mappings for GGTT and/or PPGTT */
|
|
|
|
bool has_read_only:1;
|
2017-02-15 15:43:40 +07:00
|
|
|
|
2018-10-30 01:27:20 +07:00
|
|
|
u64 (*pte_encode)(dma_addr_t addr,
|
|
|
|
enum i915_cache_level level,
|
|
|
|
u32 flags); /* Create a valid PTE */
|
2015-04-14 22:35:15 +07:00
|
|
|
#define PTE_READ_ONLY (1<<0)
|
2018-10-30 01:27:20 +07:00
|
|
|
|
2015-03-16 23:00:56 +07:00
|
|
|
int (*allocate_va_range)(struct i915_address_space *vm,
|
2017-02-15 15:43:57 +07:00
|
|
|
u64 start, u64 length);
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
void (*clear_range)(struct i915_address_space *vm,
|
2017-02-15 15:43:57 +07:00
|
|
|
u64 start, u64 length);
|
2016-06-10 15:52:59 +07:00
|
|
|
void (*insert_page)(struct i915_address_space *vm,
|
|
|
|
dma_addr_t addr,
|
2017-02-15 15:43:57 +07:00
|
|
|
u64 offset,
|
2016-06-10 15:52:59 +07:00
|
|
|
enum i915_cache_level cache_level,
|
|
|
|
u32 flags);
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
void (*insert_entries)(struct i915_address_space *vm,
|
2017-06-22 16:58:36 +07:00
|
|
|
struct i915_vma *vma,
|
2017-02-15 15:43:57 +07:00
|
|
|
enum i915_cache_level cache_level,
|
|
|
|
u32 flags);
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
void (*cleanup)(struct i915_address_space *vm);
|
2018-06-07 22:40:46 +07:00
|
|
|
|
|
|
|
struct i915_vma_ops vma_ops;
|
2017-02-15 15:43:40 +07:00
|
|
|
|
|
|
|
I915_SELFTEST_DECLARE(struct fault_attr fault_attr);
|
2018-05-11 16:51:40 +07:00
|
|
|
I915_SELFTEST_DECLARE(bool scrub_64K);
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
};
|
|
|
|
|
2018-08-31 21:36:43 +07:00
|
|
|
#define i915_is_ggtt(vm) ((vm)->is_ggtt)
|
2016-02-26 18:03:20 +07:00
|
|
|
|
2017-02-28 22:28:07 +07:00
|
|
|
static inline bool
|
2019-03-15 05:38:38 +07:00
|
|
|
i915_vm_is_4lvl(const struct i915_address_space *vm)
|
2017-02-28 22:28:07 +07:00
|
|
|
{
|
|
|
|
return (vm->total - 1) >> 32;
|
|
|
|
}
|
|
|
|
|
2017-10-07 05:18:26 +07:00
|
|
|
static inline bool
|
|
|
|
i915_vm_has_scratch_64K(struct i915_address_space *vm)
|
|
|
|
{
|
2019-03-05 20:54:27 +07:00
|
|
|
return vm->scratch_order == get_order(I915_GTT_PAGE_SIZE_64K);
|
2017-10-07 05:18:26 +07:00
|
|
|
}
|
|
|
|
|
2019-09-09 19:40:52 +07:00
|
|
|
static inline bool
|
|
|
|
i915_vm_has_cache_coloring(struct i915_address_space *vm)
|
|
|
|
{
|
|
|
|
return i915_is_ggtt(vm) && vm->mm.color_adjust;
|
|
|
|
}
|
|
|
|
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
/* The Graphics Translation Table is the way in which GEN hardware translates a
|
|
|
|
* Graphics Virtual Address into a Physical Address. In addition to the normal
|
|
|
|
* collateral associated with any va->pa translations GEN hardware also has a
|
|
|
|
* portion of the GTT which can be mapped by the CPU and remain both coherent
|
|
|
|
* and correct (in cases like swizzling). That region is referred to as GMADR in
|
|
|
|
* the spec.
|
|
|
|
*/
|
2016-03-18 15:42:57 +07:00
|
|
|
struct i915_ggtt {
|
2018-06-05 22:37:58 +07:00
|
|
|
struct i915_address_space vm;
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2017-12-11 22:18:20 +07:00
|
|
|
struct io_mapping iomap; /* Mapping to our CPU mappable region */
|
|
|
|
struct resource gmadr; /* GMADR resource */
|
2017-12-11 22:18:22 +07:00
|
|
|
resource_size_t mappable_end; /* End offset that we can CPU map */
|
2017-01-06 22:20:11 +07:00
|
|
|
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
/** "Graphics Stolen Memory" holds the global PTEs */
|
|
|
|
void __iomem *gsm;
|
2019-06-21 14:07:58 +07:00
|
|
|
void (*invalidate)(struct i915_ggtt *ggtt);
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2019-07-30 21:32:08 +07:00
|
|
|
/** PPGTT used for aliasing the PPGTT with the GTT */
|
|
|
|
struct i915_ppgtt *alias;
|
|
|
|
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
bool do_idle_maps;
|
|
|
|
|
|
|
|
int mtrr;
|
2016-10-12 16:05:20 +07:00
|
|
|
|
2019-10-16 21:32:34 +07:00
|
|
|
/** Bit 6 swizzling required for X tiling */
|
|
|
|
u32 bit_6_swizzle_x;
|
|
|
|
/** Bit 6 swizzling required for Y tiling */
|
|
|
|
u32 bit_6_swizzle_y;
|
|
|
|
|
2018-07-27 21:11:45 +07:00
|
|
|
u32 pin_bias;
|
|
|
|
|
2019-06-13 14:32:54 +07:00
|
|
|
unsigned int num_fences;
|
|
|
|
struct i915_fence_reg fence_regs[I915_MAX_NUM_FENCES];
|
|
|
|
struct list_head fence_list;
|
|
|
|
|
|
|
|
/** List of all objects in gtt_space, currently mmaped by userspace.
|
|
|
|
* All objects within this list must also be on bound_list.
|
|
|
|
*/
|
|
|
|
struct list_head userfault_list;
|
|
|
|
|
|
|
|
/* Manual runtime pm autosuspend delay for user GGTT mmaps */
|
|
|
|
struct intel_wakeref_auto userfault_wakeref;
|
|
|
|
|
2016-10-12 16:05:20 +07:00
|
|
|
struct drm_mm_node error_capture;
|
2019-04-20 06:00:12 +07:00
|
|
|
struct drm_mm_node uc_fw;
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
};
|
|
|
|
|
2019-06-11 16:12:38 +07:00
|
|
|
struct i915_ppgtt {
|
2018-06-05 22:37:58 +07:00
|
|
|
struct i915_address_space vm;
|
2018-06-12 15:18:14 +07:00
|
|
|
|
2019-06-14 23:43:42 +07:00
|
|
|
struct i915_page_directory *pd;
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
};
|
|
|
|
|
2019-06-11 16:12:38 +07:00
|
|
|
struct gen6_ppgtt {
|
|
|
|
struct i915_ppgtt base;
|
2018-06-12 15:18:14 +07:00
|
|
|
|
2019-11-30 03:13:28 +07:00
|
|
|
struct mutex flush;
|
2018-06-12 19:04:46 +07:00
|
|
|
struct i915_vma *vma;
|
2018-06-12 15:18:14 +07:00
|
|
|
gen6_pte_t __iomem *pd_addr;
|
2018-06-14 16:41:03 +07:00
|
|
|
|
2019-09-13 13:42:00 +07:00
|
|
|
atomic_t pin_count;
|
|
|
|
struct mutex pin_mutex;
|
|
|
|
|
2018-06-14 20:43:15 +07:00
|
|
|
bool scan_for_unused_pt;
|
2018-06-12 15:18:14 +07:00
|
|
|
};
|
|
|
|
|
2019-06-11 16:12:38 +07:00
|
|
|
#define __to_gen6_ppgtt(base) container_of(base, struct gen6_ppgtt, base)
|
2018-06-12 15:18:14 +07:00
|
|
|
|
2019-06-11 16:12:38 +07:00
|
|
|
static inline struct gen6_ppgtt *to_gen6_ppgtt(struct i915_ppgtt *base)
|
2018-06-12 15:18:14 +07:00
|
|
|
{
|
2019-06-11 16:12:38 +07:00
|
|
|
BUILD_BUG_ON(offsetof(struct gen6_ppgtt, base));
|
2018-06-12 15:18:14 +07:00
|
|
|
return __to_gen6_ppgtt(base);
|
|
|
|
}
|
|
|
|
|
2016-06-25 01:37:46 +07:00
|
|
|
/*
|
|
|
|
* gen6_for_each_pde() iterates over every pde from start until start+length.
|
|
|
|
* If start and start+length are not perfectly divisible, the macro will round
|
|
|
|
* down and up as needed. Start=0 and length=2G effectively iterates over
|
|
|
|
* every PDE in the system. The macro modifies ALL its parameters except 'pd',
|
|
|
|
* so each of the other parameters should preferably be a simple variable, or
|
|
|
|
* at most an lvalue with no side-effects!
|
2015-03-16 23:00:56 +07:00
|
|
|
*/
|
2016-06-25 01:37:46 +07:00
|
|
|
#define gen6_for_each_pde(pt, pd, start, length, iter) \
|
|
|
|
for (iter = gen6_pde_index(start); \
|
|
|
|
length > 0 && iter < I915_PDES && \
|
2019-06-14 23:43:42 +07:00
|
|
|
(pt = i915_pt_entry(pd, iter), true); \
|
2016-06-25 01:37:46 +07:00
|
|
|
({ u32 temp = ALIGN(start+1, 1 << GEN6_PDE_SHIFT); \
|
|
|
|
temp = min(temp - start, length); \
|
|
|
|
start += temp, length -= temp; }), ++iter)
|
|
|
|
|
|
|
|
#define gen6_for_all_pdes(pt, pd, iter) \
|
|
|
|
for (iter = 0; \
|
|
|
|
iter < I915_PDES && \
|
2019-06-14 23:43:42 +07:00
|
|
|
(pt = i915_pt_entry(pd, iter), true); \
|
2016-06-25 01:37:46 +07:00
|
|
|
++iter)
|
2015-04-08 18:13:30 +07:00
|
|
|
|
2017-02-15 15:43:57 +07:00
|
|
|
static inline u32 i915_pte_index(u64 address, unsigned int pde_shift)
|
2015-03-16 23:00:56 +07:00
|
|
|
{
|
2017-02-15 15:43:57 +07:00
|
|
|
const u32 mask = NUM_PTE(pde_shift) - 1;
|
2015-03-16 23:00:56 +07:00
|
|
|
|
|
|
|
return (address >> PAGE_SHIFT) & mask;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Helper to counts the number of PTEs within the given length. This count
|
|
|
|
* does not cross a page table boundary, so the max value would be
|
|
|
|
* GEN6_PTES for GEN6, and GEN8_PTES for GEN8.
|
|
|
|
*/
|
2017-02-15 15:43:57 +07:00
|
|
|
static inline u32 i915_pte_count(u64 addr, u64 length, unsigned int pde_shift)
|
2015-03-16 23:00:56 +07:00
|
|
|
{
|
2017-02-15 15:43:57 +07:00
|
|
|
const u64 mask = ~((1ULL << pde_shift) - 1);
|
|
|
|
u64 end;
|
2015-03-16 23:00:56 +07:00
|
|
|
|
2018-06-15 01:42:18 +07:00
|
|
|
GEM_BUG_ON(length == 0);
|
|
|
|
GEM_BUG_ON(offset_in_page(addr | length));
|
2015-03-16 23:00:56 +07:00
|
|
|
|
|
|
|
end = addr + length;
|
|
|
|
|
|
|
|
if ((addr & mask) != (end & mask))
|
|
|
|
return NUM_PTE(pde_shift) - i915_pte_index(addr, pde_shift);
|
|
|
|
|
|
|
|
return i915_pte_index(end, pde_shift) - i915_pte_index(addr, pde_shift);
|
|
|
|
}
|
|
|
|
|
2017-02-15 15:43:57 +07:00
|
|
|
static inline u32 i915_pde_index(u64 addr, u32 shift)
|
2015-03-16 23:00:56 +07:00
|
|
|
{
|
|
|
|
return (addr >> shift) & I915_PDE_MASK;
|
|
|
|
}
|
|
|
|
|
2017-02-15 15:43:57 +07:00
|
|
|
static inline u32 gen6_pte_index(u32 addr)
|
2015-03-16 23:00:56 +07:00
|
|
|
{
|
|
|
|
return i915_pte_index(addr, GEN6_PDE_SHIFT);
|
|
|
|
}
|
|
|
|
|
2017-02-15 15:43:57 +07:00
|
|
|
static inline u32 gen6_pte_count(u32 addr, u32 length)
|
2015-03-16 23:00:56 +07:00
|
|
|
{
|
|
|
|
return i915_pte_count(addr, length, GEN6_PDE_SHIFT);
|
|
|
|
}
|
|
|
|
|
2017-02-15 15:43:57 +07:00
|
|
|
static inline u32 gen6_pde_index(u32 addr)
|
2015-03-16 23:00:56 +07:00
|
|
|
{
|
|
|
|
return i915_pde_index(addr, GEN6_PDE_SHIFT);
|
|
|
|
}
|
|
|
|
|
2019-06-14 23:43:42 +07:00
|
|
|
static inline struct i915_page_table *
|
|
|
|
i915_pt_entry(const struct i915_page_directory * const pd,
|
|
|
|
const unsigned short n)
|
|
|
|
{
|
|
|
|
return pd->entry[n];
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline struct i915_page_directory *
|
|
|
|
i915_pd_entry(const struct i915_page_directory * const pdp,
|
|
|
|
const unsigned short n)
|
|
|
|
{
|
|
|
|
return pdp->entry[n];
|
|
|
|
}
|
|
|
|
|
2015-06-25 22:35:06 +07:00
|
|
|
static inline dma_addr_t
|
2019-06-11 16:12:38 +07:00
|
|
|
i915_page_dir_dma_addr(const struct i915_ppgtt *ppgtt, const unsigned int n)
|
2015-06-25 22:35:06 +07:00
|
|
|
{
|
2019-07-12 16:43:22 +07:00
|
|
|
struct i915_page_dma *pt = ppgtt->pd->entry[n];
|
2019-06-14 23:43:42 +07:00
|
|
|
|
2019-07-12 16:43:27 +07:00
|
|
|
return px_dma(pt ?: px_base(&ppgtt->vm.scratch[ppgtt->vm.top]));
|
2015-06-25 22:35:06 +07:00
|
|
|
}
|
|
|
|
|
2016-11-11 17:43:54 +07:00
|
|
|
static inline struct i915_ggtt *
|
|
|
|
i915_vm_to_ggtt(struct i915_address_space *vm)
|
|
|
|
{
|
2019-06-11 16:12:37 +07:00
|
|
|
BUILD_BUG_ON(offsetof(struct i915_ggtt, vm));
|
2016-11-11 17:43:54 +07:00
|
|
|
GEM_BUG_ON(!i915_is_ggtt(vm));
|
2018-06-05 22:37:58 +07:00
|
|
|
return container_of(vm, struct i915_ggtt, vm);
|
2016-11-11 17:43:54 +07:00
|
|
|
}
|
|
|
|
|
2019-06-11 16:12:38 +07:00
|
|
|
static inline struct i915_ppgtt *
|
2019-06-11 16:12:37 +07:00
|
|
|
i915_vm_to_ppgtt(struct i915_address_space *vm)
|
|
|
|
{
|
2019-06-11 16:12:38 +07:00
|
|
|
BUILD_BUG_ON(offsetof(struct i915_ppgtt, vm));
|
2019-06-11 16:12:37 +07:00
|
|
|
GEM_BUG_ON(i915_is_ggtt(vm));
|
2019-06-11 16:12:38 +07:00
|
|
|
return container_of(vm, struct i915_ppgtt, vm);
|
2019-06-11 16:12:37 +07:00
|
|
|
}
|
|
|
|
|
2016-08-04 13:52:22 +07:00
|
|
|
int i915_ggtt_probe_hw(struct drm_i915_private *dev_priv);
|
|
|
|
int i915_ggtt_init_hw(struct drm_i915_private *dev_priv);
|
|
|
|
int i915_ggtt_enable_hw(struct drm_i915_private *dev_priv);
|
2019-07-13 17:00:14 +07:00
|
|
|
void i915_ggtt_enable_guc(struct i915_ggtt *ggtt);
|
|
|
|
void i915_ggtt_disable_guc(struct i915_ggtt *ggtt);
|
2019-06-21 14:08:05 +07:00
|
|
|
int i915_init_ggtt(struct drm_i915_private *dev_priv);
|
2019-07-12 18:24:28 +07:00
|
|
|
void i915_ggtt_driver_release(struct drm_i915_private *dev_priv);
|
2014-08-06 20:04:45 +07:00
|
|
|
|
2019-10-29 16:58:50 +07:00
|
|
|
static inline bool i915_ggtt_has_aperture(const struct i915_ggtt *ggtt)
|
|
|
|
{
|
|
|
|
return ggtt->mappable_end > 0;
|
|
|
|
}
|
|
|
|
|
2019-06-21 14:07:51 +07:00
|
|
|
int i915_ppgtt_init_hw(struct intel_gt *gt);
|
2019-03-21 21:07:08 +07:00
|
|
|
|
2019-06-11 16:12:38 +07:00
|
|
|
struct i915_ppgtt *i915_ppgtt_create(struct drm_i915_private *dev_priv);
|
2019-03-21 21:07:08 +07:00
|
|
|
|
2019-06-11 16:12:37 +07:00
|
|
|
static inline struct i915_address_space *
|
|
|
|
i915_vm_get(struct i915_address_space *vm)
|
2014-08-06 20:04:45 +07:00
|
|
|
{
|
2019-06-11 16:12:37 +07:00
|
|
|
kref_get(&vm->ref);
|
|
|
|
return vm;
|
2014-08-06 20:04:45 +07:00
|
|
|
}
|
2019-03-21 21:07:08 +07:00
|
|
|
|
2019-06-11 16:12:37 +07:00
|
|
|
void i915_vm_release(struct kref *kref);
|
|
|
|
|
|
|
|
static inline void i915_vm_put(struct i915_address_space *vm)
|
2014-08-06 20:04:45 +07:00
|
|
|
{
|
2019-06-11 16:12:37 +07:00
|
|
|
kref_put(&vm->ref, i915_vm_release);
|
2014-08-06 20:04:45 +07:00
|
|
|
}
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
drm/i915: Pull i915_vma_pin under the vm->mutex
Replace the struct_mutex requirement for pinning the i915_vma with the
local vm->mutex instead. Note that the vm->mutex is tainted by the
shrinker (we require unbinding from inside fs-reclaim) and so we cannot
allocate while holding that mutex. Instead we have to preallocate
workers to do allocate and apply the PTE updates after we have we
reserved their slot in the drm_mm (using fences to order the PTE writes
with the GPU work and with later unbind).
In adding the asynchronous vma binding, one subtle requirement is to
avoid coupling the binding fence into the backing object->resv. That is
the asynchronous binding only applies to the vma timeline itself and not
to the pages as that is a more global timeline (the binding of one vma
does not need to be ordered with another vma, nor does the implicit GEM
fencing depend on a vma, only on writes to the backing store). Keeping
the vma binding distinct from the backing store timelines is verified by
a number of async gem_exec_fence and gem_exec_schedule tests. The way we
do this is quite simple, we keep the fence for the vma binding separate
and only wait on it as required, and never add it to the obj->resv
itself.
Another consequence in reducing the locking around the vma is the
destruction of the vma is no longer globally serialised by struct_mutex.
A natural solution would be to add a kref to i915_vma, but that requires
decoupling the reference cycles, possibly by introducing a new
i915_mm_pages object that is own by both obj->mm and vma->pages.
However, we have not taken that route due to the overshadowing lmem/ttm
discussions, and instead play a series of complicated games with
trylocks to (hopefully) ensure that only one destruction path is called!
v2: Add some commentary, and some helpers to reduce patch churn.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191004134015.13204-4-chris@chris-wilson.co.uk
2019-10-04 20:39:58 +07:00
|
|
|
static inline struct i915_address_space *
|
|
|
|
i915_vm_open(struct i915_address_space *vm)
|
|
|
|
{
|
|
|
|
GEM_BUG_ON(!atomic_read(&vm->open));
|
|
|
|
atomic_inc(&vm->open);
|
|
|
|
return i915_vm_get(vm);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline bool
|
|
|
|
i915_vm_tryopen(struct i915_address_space *vm)
|
|
|
|
{
|
|
|
|
if (atomic_add_unless(&vm->open, 1, 0))
|
|
|
|
return i915_vm_get(vm);
|
|
|
|
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
void __i915_vm_close(struct i915_address_space *vm);
|
|
|
|
|
|
|
|
static inline void
|
|
|
|
i915_vm_close(struct i915_address_space *vm)
|
|
|
|
{
|
|
|
|
GEM_BUG_ON(!atomic_read(&vm->open));
|
|
|
|
if (atomic_dec_and_test(&vm->open))
|
|
|
|
__i915_vm_close(vm);
|
|
|
|
|
|
|
|
i915_vm_put(vm);
|
|
|
|
}
|
|
|
|
|
2019-06-11 16:12:38 +07:00
|
|
|
int gen6_ppgtt_pin(struct i915_ppgtt *base);
|
|
|
|
void gen6_ppgtt_unpin(struct i915_ppgtt *base);
|
|
|
|
void gen6_ppgtt_unpin_all(struct i915_ppgtt *base);
|
2018-06-14 16:41:03 +07:00
|
|
|
|
2016-11-16 15:55:34 +07:00
|
|
|
void i915_gem_suspend_gtt_mappings(struct drm_i915_private *dev_priv);
|
|
|
|
void i915_gem_restore_gtt_mappings(struct drm_i915_private *dev_priv);
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2016-10-28 19:58:36 +07:00
|
|
|
int __must_check i915_gem_gtt_prepare_pages(struct drm_i915_gem_object *obj,
|
|
|
|
struct sg_table *pages);
|
|
|
|
void i915_gem_gtt_finish_pages(struct drm_i915_gem_object *obj,
|
|
|
|
struct sg_table *pages);
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
|
2017-01-11 18:23:11 +07:00
|
|
|
int i915_gem_gtt_reserve(struct i915_address_space *vm,
|
|
|
|
struct drm_mm_node *node,
|
|
|
|
u64 size, u64 offset, unsigned long color,
|
|
|
|
unsigned int flags);
|
|
|
|
|
2017-01-11 18:23:10 +07:00
|
|
|
int i915_gem_gtt_insert(struct i915_address_space *vm,
|
|
|
|
struct drm_mm_node *node,
|
|
|
|
u64 size, u64 alignment, unsigned long color,
|
|
|
|
u64 start, u64 end, unsigned int flags);
|
|
|
|
|
2016-08-04 22:32:31 +07:00
|
|
|
/* Flags used by pin/bind&friends. */
|
2019-08-21 19:32:34 +07:00
|
|
|
#define PIN_NOEVICT BIT_ULL(0)
|
|
|
|
#define PIN_NOSEARCH BIT_ULL(1)
|
|
|
|
#define PIN_NONBLOCK BIT_ULL(2)
|
drm/i915: Enlarge vma->pin_count
Previously we only accommodated having a vma pinned by a small number of
users, with the maximum being pinned for use by the display engine. As
such, we used a small bitfield only large enough to allow the vma to
be pinned twice (for back/front buffers) in each scanout plane. Keeping
the maximum permissible pin_count small allows us to quickly catch a
potential leak. However, as we want to split a 4096B page into 64
different cachelines and pin each cacheline for use by a different
timeline, we will exceed the current maximum permissible vma->pin_count
and so time has come to enlarge it.
Whilst we are here, try to pull together the similar bits:
Address/layout specification:
- bias, mappable, zone_4g: address limit specifiers
- fixed: address override, limits still apply though
- high: not strictly an address limit, but an address direction to search
Search controls:
- nonblock, nonfault, noevict
v2: Rewrite the guideline comment on bit consumption.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: John Harrison <john.C.Harrison@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190128181812.22804-2-chris@chris-wilson.co.uk
2019-01-29 01:18:08 +07:00
|
|
|
#define PIN_MAPPABLE BIT_ULL(3)
|
|
|
|
#define PIN_ZONE_4G BIT_ULL(4)
|
|
|
|
#define PIN_HIGH BIT_ULL(5)
|
|
|
|
#define PIN_OFFSET_BIAS BIT_ULL(6)
|
|
|
|
#define PIN_OFFSET_FIXED BIT_ULL(7)
|
|
|
|
|
drm/i915: Pull i915_vma_pin under the vm->mutex
Replace the struct_mutex requirement for pinning the i915_vma with the
local vm->mutex instead. Note that the vm->mutex is tainted by the
shrinker (we require unbinding from inside fs-reclaim) and so we cannot
allocate while holding that mutex. Instead we have to preallocate
workers to do allocate and apply the PTE updates after we have we
reserved their slot in the drm_mm (using fences to order the PTE writes
with the GPU work and with later unbind).
In adding the asynchronous vma binding, one subtle requirement is to
avoid coupling the binding fence into the backing object->resv. That is
the asynchronous binding only applies to the vma timeline itself and not
to the pages as that is a more global timeline (the binding of one vma
does not need to be ordered with another vma, nor does the implicit GEM
fencing depend on a vma, only on writes to the backing store). Keeping
the vma binding distinct from the backing store timelines is verified by
a number of async gem_exec_fence and gem_exec_schedule tests. The way we
do this is quite simple, we keep the fence for the vma binding separate
and only wait on it as required, and never add it to the obj->resv
itself.
Another consequence in reducing the locking around the vma is the
destruction of the vma is no longer globally serialised by struct_mutex.
A natural solution would be to add a kref to i915_vma, but that requires
decoupling the reference cycles, possibly by introducing a new
i915_mm_pages object that is own by both obj->mm and vma->pages.
However, we have not taken that route due to the overshadowing lmem/ttm
discussions, and instead play a series of complicated games with
trylocks to (hopefully) ensure that only one destruction path is called!
v2: Add some commentary, and some helpers to reduce patch churn.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191004134015.13204-4-chris@chris-wilson.co.uk
2019-10-04 20:39:58 +07:00
|
|
|
#define PIN_UPDATE BIT_ULL(9)
|
|
|
|
#define PIN_GLOBAL BIT_ULL(10) /* I915_VMA_GLOBAL_BIND */
|
|
|
|
#define PIN_USER BIT_ULL(11) /* I915_VMA_LOCAL_BIND */
|
drm/i915: Enlarge vma->pin_count
Previously we only accommodated having a vma pinned by a small number of
users, with the maximum being pinned for use by the display engine. As
such, we used a small bitfield only large enough to allow the vma to
be pinned twice (for back/front buffers) in each scanout plane. Keeping
the maximum permissible pin_count small allows us to quickly catch a
potential leak. However, as we want to split a 4096B page into 64
different cachelines and pin each cacheline for use by a different
timeline, we will exceed the current maximum permissible vma->pin_count
and so time has come to enlarge it.
Whilst we are here, try to pull together the similar bits:
Address/layout specification:
- bias, mappable, zone_4g: address limit specifiers
- fixed: address override, limits still apply though
- high: not strictly an address limit, but an address direction to search
Search controls:
- nonblock, nonfault, noevict
v2: Rewrite the guideline comment on bit consumption.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reviewed-by: John Harrison <john.C.Harrison@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190128181812.22804-2-chris@chris-wilson.co.uk
2019-01-29 01:18:08 +07:00
|
|
|
|
2017-01-10 21:47:34 +07:00
|
|
|
#define PIN_OFFSET_MASK (-I915_GTT_PAGE_SIZE)
|
2016-08-04 22:32:31 +07:00
|
|
|
|
drm/i915: Split out GTT specific header file
This file contains all necessary defines, prototypes and typesdefs for
manipulating GEN graphics address translation (this does not include the
legacy AGP driver)
Reiterating the comment in the header,
"Please try to maintain the following order within this file unless it
makes sense to do otherwise. From top to bottom:
1. typedefs
2. #defines, and macros
3. structure definitions
4. function prototypes
Within each section, please try to order by generation in ascending
order, from top to bottom (ie. GEN6 on the top, GEN8 on the bottom)."
I've made some minor cleanups, and fixed a couple of typos while here -
but there should be no functional changes.
The purpose of the patch is to reduce clutter in our main header file,
making room for new growth, and make documentation of our interfaces
easier by splitting things out.
With a little more work, like making i915_gtt a pointer, we could
potentially completely isolate this header from i915_drv.h. At the
moment however, I don't think it's worth the effort.
Personally, I would have liked to put the PTE encoding functions in this
file too, but I didn't want to rock the boat too much.
A similar patch has been in use on my machine for some time. This exact
patch though has only been compile tested.
Signed-off-by: Ben Widawsky <ben@bwidawsk.net>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-03-23 12:47:21 +07:00
|
|
|
#endif
|