From: "Goel, Akash" <akash.goel@intel.com>
To: Michel Thierry <michel.thierry@intel.com>,
intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH v3 09/17] drm/i915/gen8: Add 4 level support in insert_entries and clear_range
Date: Tue, 7 Jul 2015 18:21:51 +0530 [thread overview]
Message-ID: <559BCB67.1020806@intel.com> (raw)
In-Reply-To: <1435764453-11954-10-git-send-email-michel.thierry@intel.com>
On 7/1/2015 8:57 PM, Michel Thierry wrote:
> When 48b is enabled, gen8_ppgtt_insert_entries needs to read the Page Map
> Level 4 (PML4), before it selects which Page Directory Pointer (PDP)
> it will write to.
>
> Similarly, gen8_ppgtt_clear_range needs to get the correct PDP/PD range.
>
> This patch was inspired by Ben's "Depend exclusively on map and
> unmap_vma".
>
> v2: Rebase after s/page_tables/page_table/.
> v3: Remove unnecessary pdpe loop in gen8_ppgtt_clear_range_4lvl and use
> clamp_pdp in gen8_ppgtt_insert_entries (Akash).
> v4: Merge gen8_ppgtt_clear_range_4lvl into gen8_ppgtt_clear_range to
> maintain symmetry with gen8_ppgtt_insert_entries (Akash).
> v5: Do not mix pages and bytes in insert_entries (Akash).
> v6: Prevent overflow in sg_nents << PAGE_SHIFT, when inserting 4GB at
> once.
> v7: Rebase after Mika's ppgtt cleanup / scratch merge patch series.
> Use gen8_px_index functions, and remove unnecessary number of pages
> parameter in insert_pte_entries.
>
> Cc: Akash Goel <akash.goel@intel.com>
> Signed-off-by: Michel Thierry <michel.thierry@intel.com>
> ---
> drivers/gpu/drm/i915/i915_gem_gtt.c | 51 ++++++++++++++++++++++++++++---------
> drivers/gpu/drm/i915/i915_gem_gtt.h | 11 ++++++++
> 2 files changed, 50 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
> index 67d02b9..d16fbce 100644
> --- a/drivers/gpu/drm/i915/i915_gem_gtt.c
> +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
> @@ -712,9 +712,9 @@ static void gen8_ppgtt_clear_pte_range(struct i915_address_space *vm,
> struct i915_hw_ppgtt *ppgtt =
> container_of(vm, struct i915_hw_ppgtt, base);
> gen8_pte_t *pt_vaddr;
> - unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
> - unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
> - unsigned pte = start >> GEN8_PTE_SHIFT & GEN8_PTE_MASK;
> + unsigned pdpe = gen8_pdpe_index(start);
> + unsigned pde = gen8_pde_index(start);
> + unsigned pte = gen8_pte_index(start);
> unsigned num_entries = length >> PAGE_SHIFT;
> unsigned last_pte, i;
>
> @@ -763,12 +763,24 @@ static void gen8_ppgtt_clear_range(struct i915_address_space *vm,
> {
> struct i915_hw_ppgtt *ppgtt =
> container_of(vm, struct i915_hw_ppgtt, base);
> - struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
> -
> gen8_pte_t scratch_pte = gen8_pte_encode(px_dma(vm->scratch_page),
> I915_CACHE_LLC, use_scratch);
>
> - gen8_ppgtt_clear_pte_range(vm, pdp, start, length, scratch_pte);
> + if (!USES_FULL_48BIT_PPGTT(vm->dev)) {
> + gen8_ppgtt_clear_pte_range(vm, &ppgtt->pdp, start, length,
> + scratch_pte);
> + } else {
> + uint64_t templ4, pml4e;
> + struct i915_page_directory_pointer *pdp;
> +
> + gen8_for_each_pml4e(pdp, &ppgtt->pml4, start, length, templ4, pml4e) {
> + uint64_t pdp_len = gen8_clamp_pdp(start, length);
> + uint64_t pdp_start = start;
> +
> + gen8_ppgtt_clear_pte_range(vm, pdp, pdp_start, pdp_len,
> + scratch_pte);
> + }
> + }
> }
>
> static void
> @@ -781,9 +793,9 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
> struct i915_hw_ppgtt *ppgtt =
> container_of(vm, struct i915_hw_ppgtt, base);
> gen8_pte_t *pt_vaddr;
> - unsigned pdpe = start >> GEN8_PDPE_SHIFT & GEN8_PDPE_MASK;
> - unsigned pde = start >> GEN8_PDE_SHIFT & GEN8_PDE_MASK;
> - unsigned pte = start >> GEN8_PTE_SHIFT & GEN8_PTE_MASK;
> + unsigned pdpe = gen8_pdpe_index(start);
> + unsigned pde = gen8_pde_index(start);
> + unsigned pte = gen8_pte_index(start);
>
> pt_vaddr = NULL;
>
> @@ -801,7 +813,8 @@ gen8_ppgtt_insert_pte_entries(struct i915_address_space *vm,
> kunmap_px(ppgtt, pt_vaddr);
> pt_vaddr = NULL;
> if (++pde == I915_PDES) {
> - pdpe++;
> + if (++pdpe == I915_PDPES_PER_PDP(vm->dev))
> + break;
Can the same pdpe check (for Page directory pointer boundary) be added
in the gen8_ppgtt_clear_pte_range function also, to make it consistent
with gen8_ppgtt_insert_pte_entries and this will also obviate the need
for gen8_clamp_pdp macro.
> pde = 0;
> }
> pte = 0;
> @@ -820,11 +833,25 @@ static void gen8_ppgtt_insert_entries(struct i915_address_space *vm,
> {
> struct i915_hw_ppgtt *ppgtt =
> container_of(vm, struct i915_hw_ppgtt, base);
> - struct i915_page_directory_pointer *pdp = &ppgtt->pdp; /* FIXME: 48b */
> struct sg_page_iter sg_iter;
>
> __sg_page_iter_start(&sg_iter, pages->sgl, sg_nents(pages->sgl), 0);
> - gen8_ppgtt_insert_pte_entries(vm, pdp, &sg_iter, start, cache_level);
> +
> + if (!USES_FULL_48BIT_PPGTT(vm->dev)) {
> + gen8_ppgtt_insert_pte_entries(vm, &ppgtt->pdp, &sg_iter, start,
> + cache_level);
> + } else {
> + struct i915_page_directory_pointer *pdp;
> + uint64_t templ4, pml4e;
> + uint64_t length = (uint64_t)sg_nents(pages->sgl) << PAGE_SHIFT;
> +
> + gen8_for_each_pml4e(pdp, &ppgtt->pml4, start, length, templ4, pml4e) {
> + uint64_t pdp_start = start;
> +
Isn't the 'pdp_start' dispensable here ? ‘start’ can be used directly.
> + gen8_ppgtt_insert_pte_entries(vm, pdp, &sg_iter,
> + pdp_start, cache_level);
> + }
> + }
> }
>
> static void gen8_free_page_tables(struct drm_device *dev,
> diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
> index fb939fb..fd61325 100644
> --- a/drivers/gpu/drm/i915/i915_gem_gtt.h
> +++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
> @@ -478,6 +478,17 @@ static inline uint32_t gen6_pde_index(uint32_t addr)
> #define gen8_for_each_pdpe(pd, pdp, start, length, temp, iter) \
> gen8_for_each_pdpe_e(pd, pdp, start, length, temp, iter, I915_PDPES_PER_PDP(dev))
>
> +/* Clamp length to the next page_directory pointer boundary */
> +static inline uint64_t gen8_clamp_pdp(uint64_t start, uint64_t length)
> +{
> + uint64_t next_pdp = ALIGN(start + 1, 1ULL << GEN8_PML4E_SHIFT);
> +
> + if (next_pdp > (start + length))
> + return length;
> +
> + return next_pdp - start;
> +}
> +
> static inline uint32_t gen8_pte_index(uint64_t address)
> {
> return i915_pte_index(address, GEN8_PDE_SHIFT);
>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2015-07-07 12:51 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-10 16:46 [PATCH v2 00/18] 48-bit PPGTT Michel Thierry
2015-06-10 16:46 ` [PATCH v2 01/18] drm/i915/lrc: Update PDPx registers with lri commands Michel Thierry
2015-06-11 18:04 ` Mika Kuoppala
2015-06-22 9:18 ` Michel Thierry
2015-06-26 12:46 ` [PATCH v3] " Michel Thierry
2015-06-26 14:45 ` Mika Kuoppala
2015-06-10 16:46 ` [PATCH v2 02/18] drm/i915/gtt: Switch gen8_free_page_tables params Michel Thierry
2015-06-11 18:05 ` Mika Kuoppala
2015-06-26 16:38 ` Daniel Vetter
2015-06-10 16:46 ` [PATCH v2 03/18] drm/i915: Remove unnecessary gen8_clamp_pd Michel Thierry
2015-06-10 16:46 ` [PATCH v2 04/18] drm/i915/gen8: Make pdp allocation more dynamic Michel Thierry
2015-06-10 16:46 ` [PATCH v2 05/18] drm/i915/gen8: Abstract PDP usage Michel Thierry
2015-06-10 16:46 ` [PATCH v2 06/18] drm/i915/gen8: Add dynamic page trace events Michel Thierry
2015-06-10 16:46 ` [PATCH v2 07/18] drm/i915/gen8: implement alloc/free for 4lvl Michel Thierry
2015-06-10 16:46 ` [PATCH v2 08/18] drm/i915/gen8: Add 4 level switching infrastructure and lrc support Michel Thierry
2015-06-10 16:46 ` [PATCH v2 09/18] drm/i915/gen8: Generalize PTE writing for GEN8 PPGTT Michel Thierry
2015-06-10 16:46 ` [PATCH v2 10/18] drm/i915/gen8: Pass sg_iter through pte inserts Michel Thierry
2015-06-10 16:46 ` [PATCH v2 11/18] drm/i915/gen8: Add 4 level support in insert_entries and clear_range Michel Thierry
2015-06-10 16:46 ` [PATCH v2 12/18] drm/i915/gen8: Initialize PDPs Michel Thierry
2015-06-10 16:46 ` [PATCH v2 13/18] drm/i915: Expand error state's address width to 64b Michel Thierry
2015-06-10 16:46 ` [PATCH v2 14/18] drm/i915/gen8: Add ppgtt info and debug_dump Michel Thierry
2015-06-10 16:46 ` [PATCH v2 15/18] drm/i915: object size needs to be u64 Michel Thierry
2015-06-10 16:46 ` [PATCH v2 16/18] drm/i915: Check against correct user_size limit in 48b ppgtt mode Michel Thierry
2015-06-10 17:57 ` Chris Wilson
2015-06-10 16:46 ` [PATCH v2 17/18] drm/i915: Wa32bitGeneralStateOffset & Wa32bitInstructionBaseOffset Michel Thierry
2015-06-10 18:09 ` Chris Wilson
2015-06-17 12:49 ` Daniel Vetter
2015-06-17 12:53 ` Chris Wilson
2015-06-17 15:03 ` Daniel Vetter
2015-06-17 17:37 ` Chris Wilson
2015-06-18 6:45 ` Daniel Vetter
2015-06-18 7:03 ` Chris Wilson
2015-06-18 7:11 ` Daniel Vetter
2015-06-18 7:34 ` Chris Wilson
2015-06-23 12:21 ` [PATCH v3] " Michel Thierry
2015-06-23 13:22 ` Chris Wilson
2015-06-10 16:46 ` [PATCH v2 18/18] drm/i915/gen8: Flip the 48b switch Michel Thierry
2015-06-10 16:46 ` [PATCH v2] tests/gem_ppgtt: Check Wa32bitOffsets workarounds Michel Thierry
2015-07-01 15:27 ` [PATCH v3 00/17] 48-bit PPGTT Michel Thierry
2015-07-01 15:27 ` [PATCH v3 01/17] drm/i915: Remove unnecessary gen8_clamp_pd Michel Thierry
2015-07-01 15:27 ` [PATCH v3 02/17] drm/i915/gen8: Make pdp allocation more dynamic Michel Thierry
2015-07-07 12:36 ` Goel, Akash
2015-07-07 12:56 ` Michel Thierry
2015-07-01 15:27 ` [PATCH v3 03/17] drm/i915/gen8: Abstract PDP usage Michel Thierry
2015-07-07 12:43 ` Goel, Akash
2015-07-07 13:35 ` Michel Thierry
2015-07-01 15:27 ` [PATCH v3 04/17] drm/i915/gen8: Add dynamic page trace events Michel Thierry
2015-07-01 15:27 ` [PATCH v3 05/17] drm/i915/gen8: implement alloc/free for 4lvl Michel Thierry
2015-07-07 12:48 ` Goel, Akash
2015-07-07 13:40 ` Michel Thierry
2015-07-01 15:27 ` [PATCH v3 06/17] drm/i915/gen8: Add 4 level switching infrastructure and lrc support Michel Thierry
2015-07-01 15:27 ` [PATCH v3 07/17] drm/i915/gen8: Generalize PTE writing for GEN8 PPGTT Michel Thierry
2015-07-01 15:27 ` [PATCH v3 08/17] drm/i915/gen8: Pass sg_iter through pte inserts Michel Thierry
2015-07-01 15:27 ` [PATCH v3 09/17] drm/i915/gen8: Add 4 level support in insert_entries and clear_range Michel Thierry
2015-07-07 12:51 ` Goel, Akash [this message]
2015-07-07 13:42 ` Michel Thierry
2015-07-01 15:27 ` [PATCH v3 10/17] drm/i915/gen8: Initialize PDPs Michel Thierry
2015-07-01 15:27 ` [PATCH v3 11/17] drm/i915: Expand error state's address width to 64b Michel Thierry
2015-07-07 12:53 ` Goel, Akash
2015-07-07 13:50 ` Michel Thierry
2015-07-01 15:27 ` [PATCH v3 12/17] drm/i915/gen8: Add ppgtt info and debug_dump Michel Thierry
2015-07-07 12:56 ` Goel, Akash
2015-07-07 13:51 ` Michel Thierry
2015-07-01 15:27 ` [PATCH v3 13/17] drm/i915: object size needs to be u64 Michel Thierry
2015-07-01 15:27 ` [PATCH v3 14/17] drm/i915: batch_obj vm offset must " Michel Thierry
2015-07-01 16:07 ` John Harrison
2015-07-01 15:27 ` [PATCH v3 15/17] drm/i915/userptr: Kill user_size limit check Michel Thierry
2015-07-01 15:31 ` Chris Wilson
2015-07-01 15:27 ` [PATCH v3 16/17] drm/i915: Wa32bitGeneralStateOffset & Wa32bitInstructionBaseOffset Michel Thierry
2015-07-01 15:43 ` Chris Wilson
2015-07-01 15:54 ` Michel Thierry
2015-07-01 16:02 ` [PATCH v5] " Michel Thierry
2015-07-01 15:27 ` [PATCH v3 17/17] drm/i915/gen8: Flip the 48b switch Michel Thierry
2015-07-01 15:38 ` [PATCH v3 00/17] 48-bit PPGTT Daniel Vetter
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=559BCB67.1020806@intel.com \
--to=akash.goel@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=michel.thierry@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox