lkml.org 
[lkml]   [2022]   [Aug]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH v3 3/3] drm/i915/gt: document TLB cache invalidation functions
Date
Add a description for the TLB cache invalidation algorithm and for
the related kAPI functions.

Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
---

To avoid mailbombing on a large number of people, only mailing lists were C/C on the cover.
See [PATCH v3 0/3] at: https://lore.kernel.org/all/cover.1659598090.git.mchehab@kernel.org/

Documentation/gpu/i915.rst | 7 ++
drivers/gpu/drm/i915/gt/intel_tlb.c | 25 ++++++++
drivers/gpu/drm/i915/gt/intel_tlb.h | 99 +++++++++++++++++++++++++++++
3 files changed, 131 insertions(+)

diff --git a/Documentation/gpu/i915.rst b/Documentation/gpu/i915.rst
index 4e59db1cfb00..46911fdd79e8 100644
--- a/Documentation/gpu/i915.rst
+++ b/Documentation/gpu/i915.rst
@@ -58,6 +58,13 @@ Intel GVT-g Host Support(vGPU device model)
.. kernel-doc:: drivers/gpu/drm/i915/intel_gvt.c
:internal:

+TLB cache invalidation
+----------------------
+
+.. kernel-doc:: drivers/gpu/drm/i915/gt/intel_tlb.h
+
+.. kernel-doc:: drivers/gpu/drm/i915/gt/intel_tlb.c
+
Workarounds
-----------

diff --git a/drivers/gpu/drm/i915/gt/intel_tlb.c b/drivers/gpu/drm/i915/gt/intel_tlb.c
index af8cae979489..16b918ffe824 100644
--- a/drivers/gpu/drm/i915/gt/intel_tlb.c
+++ b/drivers/gpu/drm/i915/gt/intel_tlb.c
@@ -145,6 +145,18 @@ static void mmio_invalidate_full(struct intel_gt *gt)
intel_uncore_forcewake_put_delayed(uncore, FORCEWAKE_ALL);
}

+/**
+ * intel_gt_invalidate_tlb_full - do full TLB cache invalidation
+ * @gt: GT structure
+ * @seqno: sequence number
+ *
+ * Do a full TLB cache invalidation if the @seqno is bigger than the last
+ * full TLB cache invalidation.
+ *
+ * Note:
+ * The TLB cache invalidation logic depends on GEN-specific registers.
+ * It currently supports MMIO-based TLB flush for GEN8 to GEN12.
+ */
void intel_gt_invalidate_tlb_full(struct intel_gt *gt, u32 seqno)
{
intel_wakeref_t wakeref;
@@ -171,12 +183,25 @@ void intel_gt_invalidate_tlb_full(struct intel_gt *gt, u32 seqno)
}
}

+/**
+ * intel_gt_init_tlb - initialize TLB-specific vars
+ * @gt: GT structure
+ *
+ * TLB cache invalidation logic internally uses some resources that require
+ * initialization. Should be called before doing any TLB cache invalidation.
+ */
void intel_gt_init_tlb(struct intel_gt *gt)
{
mutex_init(&gt->tlb.invalidate_lock);
seqcount_mutex_init(&gt->tlb.seqno, &gt->tlb.invalidate_lock);
}

+/**
+ * intel_gt_fini_tlb - free TLB-specific vars
+ * @gt: GT structure
+ *
+ * Frees any resources needed by TLB cache invalidation logic.
+ */
void intel_gt_fini_tlb(struct intel_gt *gt)
{
mutex_destroy(&gt->tlb.invalidate_lock);
diff --git a/drivers/gpu/drm/i915/gt/intel_tlb.h b/drivers/gpu/drm/i915/gt/intel_tlb.h
index 46ce25bf5afe..2838c051f872 100644
--- a/drivers/gpu/drm/i915/gt/intel_tlb.h
+++ b/drivers/gpu/drm/i915/gt/intel_tlb.h
@@ -11,16 +11,115 @@

#include "intel_gt_types.h"

+/**
+ * DOC: TLB cache invalidation logic
+ *
+ * The way the current algorithm works is that a struct drm_i915_gem_object can
+ * be created on any order. At unbind/evict time, the object is warranted that
+ * it won't be used anymore. So, a sequence number provided by
+ * intel_gt_next_invalidate_tlb_full() is stored on it. This can happen either
+ * at __vma_put_pages() - for VMA sync unbind, or at ppgtt_unbind_vma() - for
+ * VMA async VMA bind.
+ *
+ * At __i915_gem_object_unset_pages(), intel_gt_invalidate_tlb_full() is called,
+ * where it checks if the sequence number of the object was already invalidated
+ * or not. If not, it flushes the TLB and increments the sequence number::
+ *
+ * void intel_gt_invalidate_tlb_full(struct intel_gt *gt, u32 seqno)
+ * {
+ * ...
+ * with_intel_gt_pm_if_awake(gt, wakeref) {
+ * mutex_lock(&gt->tlb.invalidate_lock);
+ * if (tlb_seqno_passed(gt, seqno))
+ * goto unlock;
+ *
+ * // Some code to do TLB invalidation
+ * ...
+ *
+ * write_seqcount_invalidate(&gt->tlb.seqno); // increment seqno
+ * mutex_lock(&gt->tlb.invalidate_lock);
+ * }
+ *
+ * So, let's say the current seqno is 2 and 3 new objects were created,
+ * on this order::
+ *
+ * obj1
+ * obj2
+ * obj3
+ *
+ * They can be unbind/evict on a different order. At unbind/evict time,
+ * the mm.tlb will be stamped with the sequence number, using the number
+ * from the last TLB flush, plus 1.
+ *
+ * Different threads may be used on unbind/evict and/or unset pages.
+ * As the logic at intel_gt_invalidate_tlb_full() is protected by a mutex,
+ * for simplicity, let's consider just two threads:
+ *
+ * +-------------------+-------------------------+---------------------------------+
+ * | sequence number | Thread 0 | Thread 1 +
+ * +===================+=========================+=================================+
+ * | seqno=2 | | |
+ * | +-------------------------+---------------------------------+
+ * | | unbind/evict obj3. | |
+ * | | | |
+ * | | obj3.mm.tlb = seqno | 1 | |
+ * | | // obj3.mm.tlb = 3 | |
+ * | +-------------------------+---------------------------------+
+ * | | unbind/evict obj1. | |
+ * | | | |
+ * | | obj1.mm.tlb = seqno | 1 | |
+ * | | // obj1.mm.tlb = 3 | |
+ * | +-------------------------+---------------------------------+
+ * | | | __i915_gem_object_unset_pages() |
+ * | | | called for obj3 => TLB flush |
+ * | | | invalidating both obj1 and obj2.|
+ * | | | |
+ * | | | seqno += 2 |
+ * +-------------------+-------------------------+---------------------------------+
+ * | seqno=4 | | |
+ * | +-------------------------+---------------------------------+
+ * | | unbind/evict obj2. | |
+ * | | | |
+ * | | obj2.mm.tlb = seqno | 1 | |
+ * | | // obj2.mm.tlb = 5 | |
+ * | +-------------------------+---------------------------------+
+ * | | | __i915_gem_object_unset_pages() |
+ * | | | called for obj1, don't flush |
+ * | | | as past flush invalidated obj1. |
+ * | +-------------------------+---------------------------------+
+ * | | | __i915_gem_object_unset_pages() |
+ * | | | called for obj2 => TLB flush. |
+ * | | | invalidating obj2. |
+ * | | | |
+ * | | | seqno += 2 |
+ * +-------------------+-------------------------+---------------------------------+
+ * | seqno=6 | | |
+ * +-------------------+-------------------------+---------------------------------+
+ */
+
void intel_gt_invalidate_tlb_full(struct intel_gt *gt, u32 seqno);

void intel_gt_init_tlb(struct intel_gt *gt);
void intel_gt_fini_tlb(struct intel_gt *gt);

+/**
+ * intel_gt_tlb_seqno - Returns the current TLB invlidation sequence number
+ * @gt: GT structure
+ *
+ * There's no need to lock while calling it, as seqprop_sequence is thread-safe
+ */
static inline u32 intel_gt_tlb_seqno(const struct intel_gt *gt)
{
return seqprop_sequence(&gt->tlb.seqno);
}

+/**
+ * intel_gt_next_invalidate_tlb_full - Returns the next TLB full invalidation
+ * sequence number
+ * @gt: GT structure
+ *
+ * There's no need to lock while calling it, as seqprop_sequence is thread-safe
+ */
static inline u32 intel_gt_next_invalidate_tlb_full(const struct intel_gt *gt)
{
return intel_gt_tlb_seqno(gt) | 1;
--
2.37.1
\
 
 \ /
  Last update: 2022-08-04 09:38    [W:0.309 / U:0.032 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site