Finish gen6/7 ppgtt dynamic page allocations
Subsystems (2)
cover Finish gen6/7 ppgtt dynamic page allocations
The first 2 patches are fixes from the previous patchset, reported by static analysis tools, while the last 2 patches complete the required work for gen6/7.
I've also started changing the authorship of the patches as suggested by Daniel.
Michel Thierry (5):
drm/i915: Prevent out of range pt in gen6_for_each_pde drm/i915: i915_dma_map_single returns positive error codes drm/i915: Remove unnecessary gen6_ppgtt_unmap_pages drm/i915: Finish gen6/7 dynamic page table allocation drm/i915: Add dynamic page trace events
drivers/gpu/drm/i915/i915_gem.c | 2 + drivers/gpu/drm/i915/i915_gem_gtt.c | 164 +++++++++++++++++++++++++++++------- drivers/gpu/drm/i915/i915_gem_gtt.h | 8 +- drivers/gpu/drm/i915/i915_trace.h | 99 ++++++++++++++++++++++ 4 files changed, 239 insertions(+), 34 deletions(-)
Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 1/5 drm/i915: Prevent out of range pt in gen6_for_each_pde
Found by static analysis tool, this was harmless as the pt was not used out of scope though.
Introduced by commit 678d96fbb3b5995a2fdff2bca5e1ab4a40b7e968 ("drm/i915: Track GEN6 page table usage").
--- drivers/gpu/drm/i915/i915_gem_gtt.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.h +3 -3
@@ -327,9 +327,9 @@ struct i915_hw_ppgtt {* XXX: temp is not actually needed, but it saves doing the ALIGN operation. */ #define gen6_for_each_pde (pt, pd, start, length, temp, iter) \ - for (iter = gen6_pde_index(start), pt = (pd)->page_table[iter]; \ - length > 0 && iter < I915_PDES; \ - pt = (pd)->page_table[++iter], \ + for (iter = gen6_pde_index(start); \ + pt = (pd)->page_table[iter], length > 0 && iter < I915_PDES; \ + iter++, \ temp = ALIGN(start+1, 1 << GEN6_PDE_SHIFT) - start, \ temp = min_t(unsigned, temp, length), \ start += temp, length -= temp)
Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 2/5 drm/i915: i915_dma_map_single returns positive error codes
In reply to Michel Thierryi915_dma_map_single relies on dma_mapping_error, which returns positive error codes. Found by static checker.
Introduced by commit 678d96fbb3b5995a2fdff2bca5e1ab4a40b7e968 ("drm/i915: Track GEN6 page table usage").
--- drivers/gpu/drm/i915/i915_gem_gtt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +1 -1
@@ -348,7 +348,7 @@ static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev) goto fail_page; ret =i915_dma_map_single (pt, dev); -if (ret) +if (ret !=0 ) goto fail_dma;return pt;
Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 2/5] drm/i915: i915_dma_map_single returns positive error codes
Quoted 6 lines · Michel Thierry
i915_dma_map_singlerelies ondma_mapping_error, which returns positive error codes. Found by static checker.Introduced by commit
678d96fbb3b5995a2fdff2bca5e1ab4a40b7e968("drm/i915: Track GEN6 page table usage").
Please give me reporter credit so I can justify my paycheck to my boss.
Quoted 12 lines
Cc Dan Carpenter <[email protected]>Cc Mika Kuoppala <[email protected]>Signed-off-by Michel Thierry <[email protected]>--- drivers/gpu/drm/i915/i915_gem_gtt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/gpu/drm/i915/i915_gem_gtt.c +0
-0
drivers/gpu/drm/i915/i915_gem_gtt.c static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev) · line 351
Quoted 5 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +1
-1@@ -348,7 +348,7 @@ static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev) goto fail_page; ret =i915_dma_map_single (pt, dev); -if (ret) +if (ret !=0 )
Heh. What is this? :P You just added a double negative. Just change to return negative error codes on failure.
regards,
dan carpenter
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 2/5] drm/i915: i915_dma_map_single returns positive error codes
Quoted 7 lines · Dan Carpenter
Quoted 7 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +0
-0@@ -348,7 +348,7 @@ static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev) goto fail_page;
Quoted 6 lines
drivers/gpu/drm/i915/i915_gem_gtt.c static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev) · line 351
Quoted 3 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +1
-1ret =i915_dma_map_single (pt, dev); -if (ret) +if (ret !=0 )Heh. What is this? :P You just added a double negative. Just change to return negative error codes on failure.
Oops. I left out a word. That should say "Just change i915_dma_map_page_single() to negative error codes on failure."
Currently, it returns 0 on success and 1 on failure which is non-standard.
regards,
dan carpenter
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 2/5] drm/i915: i915_dma_map_single returns positive error codes
Quoted 13 lines · Dan Carpenter
Quoted 7 lines · Dan Carpenter
Quoted 7 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +0
-0@@ -348,7 +348,7 @@ static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev) goto fail_page;Quoted 5 lines
drivers/gpu/drm/i915/i915_gem_gtt.c static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev) · line 351
Quoted 3 lines
drivers/gpu/drm/i915/i915_gem_gtt.c +1
-1ret =i915_dma_map_single (pt, dev); -if (ret) +if (ret !=0 )Heh. What is this? :P You just added a double negative. Just change to return negative error codes on failure.
Quoted 5 lines
Oops. I left out a word. That should say "Just change
i915_dma_map_page_single()to negative error codes on failure."Currently, it returns 0 on success and 1 on failure which is non-standard.
Ok, the function is returning the value straight from dma_mapping_error(),
I'll change i915_dma_map_page_single() to return -ENOMEM when dma_mapping_error doesn't return 0.
Thanks,
-Michel
Quoted 3 lines
regards,
dan carpenter
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH v2 1/1 drm/i915: Fix i915_dma_map_single positive error code
In reply to Michel Thierryi915_dma_map_single relies on dma_mapping_error, which returns positive error codes. Found by static checker.
Introduced by commit 678d96fbb3b5995a2fdff2bca5e1ab4a40b7e968 ("drm/i915: Track GEN6 page table usage").
v2: Return negative error code and rename commit title. (Dan)
--- drivers/gpu/drm/i915/i915_gem_gtt.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +4 -1
@@ -311,7 +311,10 @@ static inline int i915_dma_map_page_single(struct page *page,struct device *device = &dev->pdev->dev; *daddr =dma_map_page (device, page,0 ,4096 ,PCI_DMA_BIDIRECTIONAL ); -return dma_mapping_error (device, *daddr); +if (dma_mapping_error (device, *daddr)) +return -ENOMEM ; + +return 0 ; } static void unmap_and_free_pt(struct i915_page_table_entry *pt,
Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v2] drm/i915: Fix i915_dma_map_single positive error code
Quoted 10 lines · Michel Thierry
i915_dma_map_singlerelies ondma_mapping_error, which returns positive error codes. Found by static checker.Introduced by commit
678d96fbb3b5995a2fdff2bca5e1ab4a40b7e968("drm/i915: Track GEN6 page table usage").v2: Return negative error code and rename commit title. (Dan)
Cc Dan Carpenter <[email protected]>Cc Mika Kuoppala <[email protected]>
Still missing the reported-by. Giving credit where it's due is important, especially when they work for other companies than Intel. -Daniel
Quoted 28 lines
Signed-off-by Michel Thierry <[email protected]>--- drivers/gpu/drm/i915/i915_gem_gtt.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)drivers/gpu/drm/i915/i915_gem_gtt.c +4
-1@@ -311,7 +311,10 @@ static inline int i915_dma_map_page_single(struct page *page,struct device *device = &dev->pdev->dev; *daddr =dma_map_page (device, page,0 ,4096 ,PCI_DMA_BIDIRECTIONAL ); -return dma_mapping_error (device, *daddr); +if (dma_mapping_error (device, *daddr)) +return -ENOMEM ; + +return 0 ; } static void unmap_and_free_pt(struct i915_page_table_entry *pt,Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v2] drm/i915: Fix i915_dma_map_single positive error code
In reply to Michel ThierryTask id: 6041 -------------------------------------Summary-------------------------------------
Platform Delta drm-intel-nightly Series Applied
PNV -1 269/269 268/269
ILK 303/303 303/303 SNB 304/304 304/304 IVB -1 338/338 337/338 BYT 287/287 287/287 HSW 362/362 362/362 BDW 310/310 310/310 -------------------------------------Detailed------------------------------------- Platform Test drm-intel-nightly Series Applied PNV igt@gen3_render_linear_blits FAIL(1)PASS(2) FAIL(2) *IVB igt@gem_storedw_batches_loop@normal PASS(2) DMESG_WARN(1)PASS(1)
Note: You need to pay more attention to line start with '*'
_______________________________________________
Intel-gfx mailing list [email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH v3 1/1 drm/i915: Fix i915_dma_map_single positive error code
In reply to Michel Thierryi915_dma_map_single relies on dma_mapping_error, which returns positive error codes. Found by static checker.
Introduced by commit 678d96fbb3b5995a2fdff2bca5e1ab4a40b7e968 ("drm/i915: Track GEN6 page table usage").
v2: Return negative error code and renamed commit title. (Dan)
v3: Missing reported-by tag (Daniel)
--- drivers/gpu/drm/i915/i915_gem_gtt.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +4 -1
@@ -311,7 +311,10 @@ static inline int i915_dma_map_page_single(struct page *page,struct device *device = &dev->pdev->dev; *daddr =dma_map_page (device, page,0 ,4096 ,PCI_DMA_BIDIRECTIONAL ); -return dma_mapping_error (device, *daddr); +if (dma_mapping_error (device, *daddr)) +return -ENOMEM ; + +return 0 ; } static void unmap_and_free_pt(struct i915_page_table_entry *pt,
Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH v3] drm/i915: Fix i915_dma_map_single positive error code
Task id: 6041 -------------------------------------Summary-------------------------------------
Platform Delta drm-intel-nightly Series Applied
PNV -1 269/269 268/269
ILK 303/303 303/303 SNB 304/304 304/304 IVB -1 338/338 337/338 BYT 287/287 287/287 HSW 362/362 362/362 BDW 310/310 310/310 -------------------------------------Detailed------------------------------------- Platform Test drm-intel-nightly Series Applied PNV igt@gen3_render_linear_blits FAIL(1)PASS(2) FAIL(2) *IVB igt@gem_storedw_batches_loop@normal PASS(2) DMESG_WARN(1)PASS(1)
Note: You need to pay more attention to line start with '*'
_______________________________________________
Intel-gfx mailing list [email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 3/5 drm/i915: Remove unnecessary gen6_ppgtt_unmap_pages
In reply to Michel ThierryWe are already unmapping them in gen6_ppgtt_free. This function became redundant since commit 06fda602dbca9c59d87db7da71192e4b54c9f5ff ("drm/i915: Create page table allocators").
--- drivers/gpu/drm/i915/i915_gem_gtt.c | 11 ----------- 1 file changed, 11 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +0 -11
@@ -1152,16 +1152,6 @@ static void gen6_ppgtt_insert_entries(struct i915_address_space *vm,kunmap_atomic (pt_vaddr); } -static void gen6_ppgtt_unmap_pages (struct i915_hw_ppgtt *ppgtt) -{ -int i; - -for (i =0 ; i < ppgtt->num_pd_entries; i++) -pci_unmap_page (ppgtt->base.dev->pdev, - ppgtt->pd.page_table[i]->daddr, -4096 ,PCI_DMA_BIDIRECTIONAL ); -} -/* PDE TLBs are a pain invalidate pre GEN8. It requires a context reload. If we * are switching between contexts with the same LRCA, we also must do a force * restore.
@@ -1212,7 +1202,6 @@ static void gen6_ppgtt_cleanup(struct i915_address_space *vm)drm_mm_remove_node (&ppgtt->node); -gen6_ppgtt_unmap_pages (ppgtt);gen6_ppgtt_free (ppgtt); }
Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 4/5 drm/i915: Finish gen6/7 dynamic page table allocation
In reply to Michel ThierryThis patch continues on the idea from "Track GEN6 page table usage". From here on, in the steady state, PDEs are all pointing to the scratch page table (as recommended in the spec). When an object is allocated in the VA range, the code will determine if we need to allocate a page for the page table. Similarly when the object is destroyed, we will remove, and free the page table pointing the PDE back to the scratch page.
Following patches will work to unify the code a bit as we bring in GEN8 support. GEN6 and GEN8 are different enough that I had a hard time to get to this point with as much common code as I do.
The aliasing PPGTT must pre-allocate all of the page tables. There are a few reasons for this. Two trivial ones: aliasing ppgtt goes through the ggtt paths, so it's hard to maintain, we currently do not restore the default context (assuming the previous force reload is indeed necessary). Most importantly though, the only way (it seems from empirical evidence) to invalidate the CS TLBs on non-render ring is to either use ring sync (which requires actually stopping the rings in order to synchronize when the sync completes vs. where you are in execution), or to reload DCLV. Since without full PPGTT we do not ever reload the DCLV register, there is no good way to achieve this. The simplest solution is just to not support dynamic page table creation/destruction in the aliasing PPGTT.
We could always reload DCLV, but this seems like quite a bit of excess overhead only to save at most 2MB-4k of memory for the aliasing PPGTT page tables.
v2: Make the page table bitmap declared inside the function (Chris) Simplify the way scratching address space works.
Move the alloc/teardown tracepoints up a level in the call stack so that both all implementations get the trace.
v3: Updated trace event to spit out a name
v4: Aliasing ppgtt is now initialized differently (in setup global gtt)
v5: Rebase to latest code. Also removed unnecessary aliasing ppgtt check for trace, as it is no longer possible after the PPGTT cleanup patch series of a couple of months ago (Daniel).
v6: Implement changes from code review (Daniel):
- allocate/teardown_va_range calls added.
- Add a scratch page allocation helper (only need the address).
- Move trace events to a new patch.
- Use updated mark_tlbs_dirty.
- Moved pt preallocation for aliasing ppgtt into gen6_ppgtt_init.
v7: teardown_va_range removed (Daniel).
In init, gen6_ppgtt_clear_range call is only needed for aliasing ppgtt.
v8: Rebase after s/page_tables/page_table/.
v9: Remove unnecessary scratch flag in page_table struct, future patches can just compare against ppgtt->scratch_pt, and alloc_pt_scratch becomes redundant. Initialize scratch_pt and pt. (Mika)
v10: Clean up aliasing ppgtt init error path and prevent leaking the ppgtt obj when init fails. (Mika)
Updated commit author. (Daniel)
--- drivers/gpu/drm/i915/i915_gem_gtt.c | 146 +++++++++++++++++++++++++++++++----- drivers/gpu/drm/i915/i915_gem_gtt.h | 2 + 2 files changed, 129 insertions(+), 19 deletions(-)
drivers/gpu/drm/i915/i915_gem_gtt.c +127 -19
@@ -343,7 +343,7 @@ static struct i915_page_table_entry *alloc_pt_single(struct drm_device *dev)if (!pt->used_ptes) goto fail_bitmap; - pt->page =alloc_page (GFP_KERNEL | __GFP_ZERO); + pt->page =alloc_page (GFP_KERNEL );if (!pt->page) goto fail_page;
@@ -378,7 +378,7 @@ fail_bitmap:* Return: 0 if allocation succeeded. */ static int alloc_pt_range (struct i915_page_directory_entry *pd,uint16_t pde,size_t count, -struct drm_device *dev) +struct drm_device *dev) {int i, ret;
@@ -1162,13 +1162,70 @@ static inline void mark_tlbs_dirty(struct i915_hw_ppgtt *ppgtt) ppgtt->pd_dirty_rings =INTEL_INFO (ppgtt->base.dev)->ring_mask; } +static void gen6_initialize_pt (struct i915_address_space *vm, +struct i915_page_table_entry *pt) +{ +gen6_pte_t *pt_vaddr, scratch_pte; +int i; + +WARN_ON (vm->scratch.addr ==0 ); + + scratch_pte = vm->pte_encode (vm->scratch.addr, +I915_CACHE_LLC , true,0 ); + + pt_vaddr =kmap_atomic (pt->page); + +for (i =0 ; i <GEN6_PTES ; i++) + pt_vaddr[i] = scratch_pte; + +kunmap_atomic (pt_vaddr); +} +static int gen6_alloc_va_range (struct i915_address_space *vm,uint64_t start,uint64_t length) { +DECLARE_BITMAP (new_page_tables,I915_PDES ); +struct drm_device *dev = vm->dev; +struct drm_i915_private *dev_priv = dev->dev_private;struct i915_hw_ppgtt *ppgtt = container_of(vm, struct i915_hw_ppgtt, base);struct i915_page_table_entry *pt; +const uint32_t start_save = start, length_save = length;uint32_t pde, temp; +int ret; + +WARN_ON (upper_32_bits (start)); + +bitmap_zero (new_page_tables,I915_PDES ); + +/* The allocation is done in two stages so that we can bail out with +* minimal amount of pain. The first stage finds new page tables that +* need allocation. The second stage marks use ptes within the page +* tables. +*/ +gen6_for_each_pde (pt, &ppgtt->pd, start, length, temp, pde) { +if (pt != ppgtt->scratch_pt) { +WARN_ON (bitmap_empty (pt->used_ptes,GEN6_PTES )); +continue ; + } + +/* We've already allocated a page table */ +WARN_ON (!bitmap_empty (pt->used_ptes,GEN6_PTES )); + + pt =alloc_pt_single (dev); +if (IS_ERR (pt)) { + ret =PTR_ERR (pt); + goto unwind_out; + } + +gen6_initialize_pt (vm, pt); + + ppgtt->pd.page_table[pde] = pt; +set_bit (pde, new_page_tables); + } + + start = start_save; + length = length_save;gen6_for_each_pde (pt, &ppgtt->pd, start, length, temp, pde) {DECLARE_BITMAP (tmp_bitmap,GEN6_PTES );
@@ -1177,21 +1234,46 @@ static int gen6_alloc_va_range(struct i915_address_space *vm,bitmap_set (tmp_bitmap,gen6_pte_index (start),gen6_pte_count (start, length)); -bitmap_or (pt->used_ptes, pt->used_ptes, tmp_bitmap, +if (test_and_clear_bit (pde, new_page_tables)) +gen6_write_pde (&ppgtt->pd, pde, pt); + +bitmap_or (pt->used_ptes, tmp_bitmap, pt->used_ptes,GEN6_PTES ); } +WARN_ON (!bitmap_empty (new_page_tables,I915_PDES )); + +/* Make sure write is complete before other code can use this page +* table. Also require for WC mapped PTEs */ +readl (dev_priv->gtt.gsm); +mark_tlbs_dirty (ppgtt);return 0 ; + +unwind_out: +for_each_set_bit (pde, new_page_tables,I915_PDES ) { +struct i915_page_table_entry *pt = ppgtt->pd.page_table[pde]; + + ppgtt->pd.page_table[pde] = ppgtt->scratch_pt; +unmap_and_free_pt (pt, vm->dev); + } + +mark_tlbs_dirty (ppgtt); +return ret; }static void gen6_ppgtt_free (struct i915_hw_ppgtt *ppgtt) {int i; -for (i =0 ; i < ppgtt->num_pd_entries; i++) -unmap_and_free_pt (ppgtt->pd.page_table[i], ppgtt->base.dev); +for (i =0 ; i < ppgtt->num_pd_entries; i++) { +struct i915_page_table_entry *pt = ppgtt->pd.page_table[i]; + +if (pt != ppgtt->scratch_pt) +unmap_and_free_pt (ppgtt->pd.page_table[i], ppgtt->base.dev); + } +unmap_and_free_pt (ppgtt->scratch_pt, ppgtt->base.dev);unmap_and_free_pd (&ppgtt->pd); }
@@ -1217,6 +1299,12 @@ static int gen6_ppgtt_allocate_page_directories(struct i915_hw_ppgtt *ppgtt)* size. We allocate at the top of the GTT to avoid fragmentation. */ BUG_ON (!drm_mm_initialized (&dev_priv->gtt.base.mm)); + ppgtt->scratch_pt =alloc_pt_single (ppgtt->base.dev); +if (IS_ERR (ppgtt->scratch_pt)) +return PTR_ERR (ppgtt->scratch_pt); + +gen6_initialize_pt (&ppgtt->base, ppgtt->scratch_pt); + alloc: ret = drm_mm_insert_node_in_range_generic(&dev_priv->gtt.base.mm, &ppgtt->node, GEN6_PD_SIZE,
@@ -1247,6 +1335,7 @@ alloc:return 0 ; err_out: +unmap_and_free_pt (ppgtt->scratch_pt, ppgtt->base.dev);return ret; }
@@ -1258,18 +1347,20 @@ static int gen6_ppgtt_alloc(struct i915_hw_ppgtt *ppgtt)if (ret)return ret; - ret =alloc_pt_range (&ppgtt->pd,0 , ppgtt->num_pd_entries, - ppgtt->base.dev); +return 0 ; +} -if (ret) { -drm_mm_remove_node (&ppgtt->node); -return ret; - } +static void gen6_scratch_va_range (struct i915_hw_ppgtt *ppgtt, +uint64_t start,uint64_t length) +{ +struct i915_page_table_entry *unused; +uint32_t pde, temp; -return 0 ; +gen6_for_each_pde (unused, &ppgtt->pd, start, length, temp, pde) + ppgtt->pd.page_table[pde] = ppgtt->scratch_pt; } -static int gen6_ppgtt_init (struct i915_hw_ppgtt *ppgtt) +static int gen6_ppgtt_init (struct i915_hw_ppgtt *ppgtt,bool aliasing) {struct drm_device *dev = ppgtt->base.dev;struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1292,6 +1383,17 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)if (ret)return ret; +if (aliasing) { +/* preallocate all pts */ + ret =alloc_pt_range (&ppgtt->pd,0 , ppgtt->num_pd_entries, + ppgtt->base.dev); + +if (ret) { +gen6_ppgtt_cleanup (&ppgtt->base); +return ret; + } + } + ppgtt->base.allocate_va_range = gen6_alloc_va_range; ppgtt->base.clear_range = gen6_ppgtt_clear_range; ppgtt->base.insert_entries = gen6_ppgtt_insert_entries;
@@ -1306,7 +1408,10 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt) ppgtt->pd_addr = (gen6_pte_t __iomem *)dev_priv->gtt.gsm + ppgtt->pd.pd_offset /sizeof (gen6_pte_t); - ppgtt->base.clear_range (&ppgtt->base,0 , ppgtt->base.total, true); +if (aliasing) + ppgtt->base.clear_range (&ppgtt->base,0 , ppgtt->base.total, true); +else +gen6_scratch_va_range (ppgtt,0 , ppgtt->base.total);gen6_write_page_range (dev_priv, &ppgtt->pd,0 , ppgtt->base.total);
@@ -1320,7 +1425,8 @@ static int gen6_ppgtt_init(struct i915_hw_ppgtt *ppgtt)return 0 ; } -static int __hw_ppgtt_init (struct drm_device *dev,struct i915_hw_ppgtt *ppgtt) +static int __hw_ppgtt_init (struct drm_device *dev,struct i915_hw_ppgtt *ppgtt, +bool aliasing) {struct drm_i915_private *dev_priv = dev->dev_private;
@@ -1328,7 +1434,7 @@ static int __hw_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt) ppgtt->base.scratch = dev_priv->gtt.base.scratch;if (INTEL_INFO (dev)->gen <8 ) -return gen6_ppgtt_init (ppgtt); +return gen6_ppgtt_init (ppgtt, aliasing);else return gen8_ppgtt_init (ppgtt, dev_priv->gtt.base.total); }
@@ -1337,7 +1443,7 @@ int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)struct drm_i915_private *dev_priv = dev->dev_private;int ret =0 ; - ret =__hw_ppgtt_init (dev, ppgtt); + ret =__hw_ppgtt_init (dev, ppgtt, false);if (ret ==0 ) {kref_init (&ppgtt->ref); drm_mm_init(&ppgtt->base.mm, ppgtt->base.start,
@@ -1972,9 +2078,11 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,if (!ppgtt)return -ENOMEM ; - ret =__hw_ppgtt_init (dev, ppgtt); -if (ret !=0 ) + ret =__hw_ppgtt_init (dev, ppgtt, true); +if (ret) { +kfree (ppgtt);return ret; + } dev_priv->mm.aliasing_ppgtt = ppgtt; }
drivers/gpu/drm/i915/i915_gem_gtt.h +2 -0
@@ -308,6 +308,8 @@ struct i915_hw_ppgtt {struct i915_page_directory_entry pd; }; +struct i915_page_table_entry *scratch_pt; +struct drm_i915_file_private *file_priv;gen6_pte_t __iomem *pd_addr;
Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
PATCH 5/5 drm/i915: Add dynamic page trace events
In reply to Michel ThierryTraces for page directories and tables allocation and map.
v2: Removed references to teardown.
v3: bitmap_scnprintf has been deprecated.
v4: Replace bitmap_scnprintf with scnprintf correctly, and get right range lengths. (Mika)
--- drivers/gpu/drm/i915/i915_gem.c | 2 + drivers/gpu/drm/i915/i915_gem_gtt.c | 5 ++ drivers/gpu/drm/i915/i915_trace.h | 99 +++++++++++++++++++++++++++++++++++++ 3 files changed, 106 insertions(+)
drivers/gpu/drm/i915/i915_gem.c +2 -0
@@ -3538,6 +3538,8 @@ search_free:/* allocate before insert / bind */ if (vma->vm->allocate_va_range) { +trace_i915_va_alloc (vma->vm, vma->node.start, vma->node.size, +VM_TO_TRACE_NAME (vma->vm)); ret = vma->vm->allocate_va_range (vma->vm, vma->node.start, vma->node.size);
drivers/gpu/drm/i915/i915_gem_gtt.c +5 -0
@@ -1222,6 +1222,7 @@ static int gen6_alloc_va_range(struct i915_address_space *vm, ppgtt->pd.page_table[pde] = pt;set_bit (pde, new_page_tables); +trace_i915_page_table_entry_alloc (vm, pde, start,GEN6_PDE_SHIFT ); } start = start_save;
@@ -1237,6 +1238,10 @@ static int gen6_alloc_va_range(struct i915_address_space *vm,if (test_and_clear_bit (pde, new_page_tables))gen6_write_pde (&ppgtt->pd, pde, pt); +trace_i915_page_table_entry_map (vm, pde, pt, +gen6_pte_index (start), +gen6_pte_count (start, length), +GEN6_PTES );bitmap_or (pt->used_ptes, tmp_bitmap, pt->used_ptes,GEN6_PTES ); }
drivers/gpu/drm/i915/i915_trace.h +99 -0
105 lines not on this page · raw message
Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Re: [PATCH 5/5] drm/i915: Add dynamic page trace events
Task id: 6040 -------------------------------------Summary-------------------------------------
Platform Delta drm-intel-nightly Series Applied
PNV -3 269/269 266/269 ILK -1 303/303 302/303 SNB 304/304 304/304 IVB -1 338/338 337/338 BYT 287/287 287/287 HSW 362/362 362/362 BDW 310/310 310/310 -------------------------------------Detailed------------------------------------- Platform Test drm-intel-nightly Series Applied PNV igt@gem_userptr_blits@coherency-sync CRASH(1)PASS(2) CRASH(1)PASS(1) *PNV igt@gen3_render_linear_blits PASS(2) FAIL(1)PASS(1) PNV igt@gem_userptr_blits@minor-unsync-interruptible DMESG_WARN(1)PASS(1) DMESG_WARN(2) *ILK igt@gem_seqno_wrap PASS(3) DMESG_WARN(1)PASS(1) *IVB igt@gem_storedw_batches_loop@secure-dispatch PASS(2) DMESG_WARN(1)PASS(1)
Note: You need to pay more attention to line start with '*'
_______________________________________________
Intel-gfx mailing list [email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 12 lines · Michel Thierry
The first 2 patches are fixes from the previous patchset, reported by static analysis tools, while the last 2 patches complete the required work for gen6/7.
I've also started changing the authorship of the patches as suggested by Daniel.
Michel Thierry (5):
drm/i915: Prevent out of range pt ingen6_for_each_pdedrm/i915:i915_dma_map_singlereturns positive error codes drm/i915: Remove unnecessarygen6_ppgtt_unmap_pagesdrm/i915: Finish gen6/7 dynamic page table allocation drm/i915: Add dynamic page trace events
Daniel,
We have tossed the Finish gen6/7 one around in irc quite a lot.
Patches 1, 2v3, 3, 4, 5:
Quoted 13 lines
drivers/gpu/drm/i915/i915_gem.c | 2 + drivers/gpu/drm/i915/i915_gem_gtt.c | 164 +++++++++++++++++++++++++++++------- drivers/gpu/drm/i915/i915_gem_gtt.h | 8 +- drivers/gpu/drm/i915/i915_trace.h | 99 ++++++++++++++++++++++ 4 files changed, 239 insertions(+), 34 deletions(-)Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 15 lines · Mika Kuoppala
Quoted 12 lines · Michel Thierry
The first 2 patches are fixes from the previous patchset, reported by static analysis tools, while the last 2 patches complete the required work for gen6/7.
I've also started changing the authorship of the patches as suggested by Daniel.
Michel Thierry (5):
drm/i915: Prevent out of range pt ingen6_for_each_pdedrm/i915:i915_dma_map_singlereturns positive error codes drm/i915: Remove unnecessarygen6_ppgtt_unmap_pagesdrm/i915: Finish gen6/7 dynamic page table allocation drm/i915: Add dynamic page trace events
Quoted 6 lines
Daniel,
We have tossed the Finish gen6/7 one around in irc quite a lot.
Patches 1, 2v3, 3, 4, 5:
Reviewed-by Mika Kuoppala <[email protected]>
All merged, thanks for patches&review.
btw since you're having all these discussions: Are you taking notes for some nice kerneldoc polish on top of all these patches?
Thanks, Daniel
Quoted 18 lines
Quoted 14 lines
drivers/gpu/drm/i915/i915_gem.c | 2 + drivers/gpu/drm/i915/i915_gem_gtt.c | 164 +++++++++++++++++++++++++++++------- drivers/gpu/drm/i915/i915_gem_gtt.h | 8 +- drivers/gpu/drm/i915/i915_trace.h | 99 ++++++++++++++++++++++ 4 files changed, 239 insertions(+), 34 deletions(-)Signature · 6 lines
2.1.1 _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Signature · 7 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list [email protected] http://lists.freedesktop.org/mailman/listinfo/intel-gfx
Quoted 20 lines · Daniel Vetter
Quoted 19 lines · Mika Kuoppala
Quoted 12 lines · Michel Thierry
The first 2 patches are fixes from the previous patchset, reported by static analysis tools, while the last 2 patches complete the required work for gen6/7.
I've also started changing the authorship of the patches as suggested by Daniel.
Michel Thierry (5):
drm/i915: Prevent out of range pt ingen6_for_each_pdedrm/i915:i915_dma_map_singlereturns positive error codes drm/i915: Remove unnecessarygen6_ppgtt_unmap_pagesdrm/i915: Finish gen6/7 dynamic page table allocation drm/i915: Add dynamic page trace eventsDaniel,
We have tossed the Finish gen6/7 one around in irc quite a lot.
Quoted 8 lines
Quoted 2 lines
Patches 1, 2v3, 3, 4, 5:
Reviewed-by Mika Kuoppala <[email protected]>All merged, thanks for patches&review.
btw since you're having all these discussions: Are you taking notes for some nice kerneldoc polish on top of all these patches?
I have something saved yes. However mostly they are irc logs where I am chasing my own tail in a strong coffee induced stupor.
But after the gen8 parts, including 48bit have landed, we should see if any naming/cleanups are needed and after that do the kerneldoc polish.
I feel like I have overused my naming rebase quota already at this stage.
- Mika
Quoted 11 lines
Thanks, Daniel
Quoted 10 lines
Quoted 10 lines
drivers/gpu/drm/i915/i915_gem.c | 2 + drivers/gpu/drm/i915/i915_gem_gtt.c | 164 +++++++++++++++++++++++++++++------- drivers/gpu/drm/i915/i915_gem_gtt.h | 8 +- drivers/gpu/drm/i915/i915_trace.h | 99 ++++++++++++++++++++++ 4 files changed, 239 insertions(+), 34 deletions(-)Signature · 1 line
2.1.1
Quoted 13 lines
Quoted 8 lines
Quoted 4 lines
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfxSignature · 3 lines
Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/intel-gfx