Commit 95d1deb0 authored by Rob Clark's avatar Rob Clark

drm/msm/gem: Add fenced vma unpin

With userspace allocated iova (next patch), we can have a race condition
where userspace observes the fence completion and deletes the vma before
retire_submit() gets around to unpinning the vma.  To handle this, add a
fenced unpin which drops the refcount but tracks the fence, and update
msm_gem_vma_inuse() to check any previously unsignaled fences.

v2: Fix inuse underflow (duplicate unpin)
v3: Fix msm_job_run() vs submit_cleanup() race condition
Signed-off-by: default avatarRob Clark <robdclark@chromium.org>
Link: https://lore.kernel.org/r/20220411215849.297838-10-robdclark@gmail.comSigned-off-by: default avatarRob Clark <robdclark@chromium.org>
parent 27674c66
...@@ -15,6 +15,7 @@ msm_fence_context_alloc(struct drm_device *dev, volatile uint32_t *fenceptr, ...@@ -15,6 +15,7 @@ msm_fence_context_alloc(struct drm_device *dev, volatile uint32_t *fenceptr,
const char *name) const char *name)
{ {
struct msm_fence_context *fctx; struct msm_fence_context *fctx;
static int index = 0;
fctx = kzalloc(sizeof(*fctx), GFP_KERNEL); fctx = kzalloc(sizeof(*fctx), GFP_KERNEL);
if (!fctx) if (!fctx)
...@@ -23,6 +24,7 @@ msm_fence_context_alloc(struct drm_device *dev, volatile uint32_t *fenceptr, ...@@ -23,6 +24,7 @@ msm_fence_context_alloc(struct drm_device *dev, volatile uint32_t *fenceptr,
fctx->dev = dev; fctx->dev = dev;
strncpy(fctx->name, name, sizeof(fctx->name)); strncpy(fctx->name, name, sizeof(fctx->name));
fctx->context = dma_fence_context_alloc(1); fctx->context = dma_fence_context_alloc(1);
fctx->index = index++;
fctx->fenceptr = fenceptr; fctx->fenceptr = fenceptr;
spin_lock_init(&fctx->spinlock); spin_lock_init(&fctx->spinlock);
...@@ -34,7 +36,7 @@ void msm_fence_context_free(struct msm_fence_context *fctx) ...@@ -34,7 +36,7 @@ void msm_fence_context_free(struct msm_fence_context *fctx)
kfree(fctx); kfree(fctx);
} }
static inline bool fence_completed(struct msm_fence_context *fctx, uint32_t fence) bool msm_fence_completed(struct msm_fence_context *fctx, uint32_t fence)
{ {
/* /*
* Note: Check completed_fence first, as fenceptr is in a write-combine * Note: Check completed_fence first, as fenceptr is in a write-combine
...@@ -76,7 +78,7 @@ static const char *msm_fence_get_timeline_name(struct dma_fence *fence) ...@@ -76,7 +78,7 @@ static const char *msm_fence_get_timeline_name(struct dma_fence *fence)
static bool msm_fence_signaled(struct dma_fence *fence) static bool msm_fence_signaled(struct dma_fence *fence)
{ {
struct msm_fence *f = to_msm_fence(fence); struct msm_fence *f = to_msm_fence(fence);
return fence_completed(f->fctx, f->base.seqno); return msm_fence_completed(f->fctx, f->base.seqno);
} }
static const struct dma_fence_ops msm_fence_ops = { static const struct dma_fence_ops msm_fence_ops = {
......
...@@ -21,6 +21,8 @@ struct msm_fence_context { ...@@ -21,6 +21,8 @@ struct msm_fence_context {
char name[32]; char name[32];
/** context: see dma_fence_context_alloc() */ /** context: see dma_fence_context_alloc() */
unsigned context; unsigned context;
/** index: similar to context, but local to msm_fence_context's */
unsigned index;
/** /**
* last_fence: * last_fence:
...@@ -56,6 +58,7 @@ struct msm_fence_context * msm_fence_context_alloc(struct drm_device *dev, ...@@ -56,6 +58,7 @@ struct msm_fence_context * msm_fence_context_alloc(struct drm_device *dev,
volatile uint32_t *fenceptr, const char *name); volatile uint32_t *fenceptr, const char *name);
void msm_fence_context_free(struct msm_fence_context *fctx); void msm_fence_context_free(struct msm_fence_context *fctx);
bool msm_fence_completed(struct msm_fence_context *fctx, uint32_t fence);
void msm_update_fence(struct msm_fence_context *fctx, uint32_t fence); void msm_update_fence(struct msm_fence_context *fctx, uint32_t fence);
struct dma_fence * msm_fence_alloc(struct msm_fence_context *fctx); struct dma_fence * msm_fence_alloc(struct msm_fence_context *fctx);
......
...@@ -445,7 +445,7 @@ void msm_gem_unpin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vm ...@@ -445,7 +445,7 @@ void msm_gem_unpin_vma_locked(struct drm_gem_object *obj, struct msm_gem_vma *vm
GEM_WARN_ON(!msm_gem_is_locked(obj)); GEM_WARN_ON(!msm_gem_is_locked(obj));
msm_gem_unmap_vma(vma->aspace, vma); msm_gem_unpin_vma(vma);
msm_obj->pin_count--; msm_obj->pin_count--;
GEM_WARN_ON(msm_obj->pin_count < 0); GEM_WARN_ON(msm_obj->pin_count < 0);
......
...@@ -49,6 +49,8 @@ struct msm_gem_address_space * ...@@ -49,6 +49,8 @@ struct msm_gem_address_space *
msm_gem_address_space_create(struct msm_mmu *mmu, const char *name, msm_gem_address_space_create(struct msm_mmu *mmu, const char *name,
u64 va_start, u64 size); u64 va_start, u64 size);
struct msm_fence_context;
struct msm_gem_vma { struct msm_gem_vma {
struct drm_mm_node node; struct drm_mm_node node;
uint64_t iova; uint64_t iova;
...@@ -56,6 +58,9 @@ struct msm_gem_vma { ...@@ -56,6 +58,9 @@ struct msm_gem_vma {
struct list_head list; /* node in msm_gem_object::vmas */ struct list_head list; /* node in msm_gem_object::vmas */
bool mapped; bool mapped;
int inuse; int inuse;
uint32_t fence_mask;
uint32_t fence[MSM_GPU_MAX_RINGS];
struct msm_fence_context *fctx[MSM_GPU_MAX_RINGS];
}; };
int msm_gem_init_vma(struct msm_gem_address_space *aspace, int msm_gem_init_vma(struct msm_gem_address_space *aspace,
...@@ -64,8 +69,8 @@ int msm_gem_init_vma(struct msm_gem_address_space *aspace, ...@@ -64,8 +69,8 @@ int msm_gem_init_vma(struct msm_gem_address_space *aspace,
bool msm_gem_vma_inuse(struct msm_gem_vma *vma); bool msm_gem_vma_inuse(struct msm_gem_vma *vma);
void msm_gem_purge_vma(struct msm_gem_address_space *aspace, void msm_gem_purge_vma(struct msm_gem_address_space *aspace,
struct msm_gem_vma *vma); struct msm_gem_vma *vma);
void msm_gem_unmap_vma(struct msm_gem_address_space *aspace, void msm_gem_unpin_vma(struct msm_gem_vma *vma);
struct msm_gem_vma *vma); void msm_gem_unpin_vma_fenced(struct msm_gem_vma *vma, struct msm_fence_context *fctx);
int msm_gem_map_vma(struct msm_gem_address_space *aspace, int msm_gem_map_vma(struct msm_gem_address_space *aspace,
struct msm_gem_vma *vma, int prot, struct msm_gem_vma *vma, int prot,
struct sg_table *sgt, int size); struct sg_table *sgt, int size);
...@@ -363,6 +368,11 @@ struct msm_gem_submit { ...@@ -363,6 +368,11 @@ struct msm_gem_submit {
struct drm_msm_gem_submit_reloc *relocs; struct drm_msm_gem_submit_reloc *relocs;
} *cmd; /* array of size nr_cmds */ } *cmd; /* array of size nr_cmds */
struct { struct {
/* make sure these don't conflict w/ MSM_SUBMIT_BO_x */
#define BO_VALID 0x8000 /* is current addr in cmdstream correct/valid? */
#define BO_LOCKED 0x4000 /* obj lock is held */
#define BO_ACTIVE 0x2000 /* active refcnt is held */
#define BO_PINNED 0x1000 /* obj is pinned and on active list */
uint32_t flags; uint32_t flags;
union { union {
struct msm_gem_object *obj; struct msm_gem_object *obj;
......
...@@ -21,12 +21,6 @@ ...@@ -21,12 +21,6 @@
* Cmdstream submission: * Cmdstream submission:
*/ */
/* make sure these don't conflict w/ MSM_SUBMIT_BO_x */
#define BO_VALID 0x8000 /* is current addr in cmdstream correct/valid? */
#define BO_LOCKED 0x4000 /* obj lock is held */
#define BO_ACTIVE 0x2000 /* active refcnt is held */
#define BO_PINNED 0x1000 /* obj is pinned and on active list */
static struct msm_gem_submit *submit_create(struct drm_device *dev, static struct msm_gem_submit *submit_create(struct drm_device *dev,
struct msm_gpu *gpu, struct msm_gpu *gpu,
struct msm_gpu_submitqueue *queue, uint32_t nr_bos, struct msm_gpu_submitqueue *queue, uint32_t nr_bos,
...@@ -231,6 +225,13 @@ static void submit_cleanup_bo(struct msm_gem_submit *submit, int i, ...@@ -231,6 +225,13 @@ static void submit_cleanup_bo(struct msm_gem_submit *submit, int i,
struct drm_gem_object *obj = &submit->bos[i].obj->base; struct drm_gem_object *obj = &submit->bos[i].obj->base;
unsigned flags = submit->bos[i].flags & cleanup_flags; unsigned flags = submit->bos[i].flags & cleanup_flags;
/*
* Clear flags bit before dropping lock, so that the msm_job_run()
* path isn't racing with submit_cleanup() (ie. the read/modify/
* write is protected by the obj lock in all paths)
*/
submit->bos[i].flags &= ~cleanup_flags;
if (flags & BO_PINNED) if (flags & BO_PINNED)
msm_gem_unpin_vma_locked(obj, submit->bos[i].vma); msm_gem_unpin_vma_locked(obj, submit->bos[i].vma);
...@@ -239,8 +240,6 @@ static void submit_cleanup_bo(struct msm_gem_submit *submit, int i, ...@@ -239,8 +240,6 @@ static void submit_cleanup_bo(struct msm_gem_submit *submit, int i,
if (flags & BO_LOCKED) if (flags & BO_LOCKED)
dma_resv_unlock(obj->resv); dma_resv_unlock(obj->resv);
submit->bos[i].flags &= ~cleanup_flags;
} }
static void submit_unlock_unpin_bo(struct msm_gem_submit *submit, int i) static void submit_unlock_unpin_bo(struct msm_gem_submit *submit, int i)
......
...@@ -5,6 +5,7 @@ ...@@ -5,6 +5,7 @@
*/ */
#include "msm_drv.h" #include "msm_drv.h"
#include "msm_fence.h"
#include "msm_gem.h" #include "msm_gem.h"
#include "msm_mmu.h" #include "msm_mmu.h"
...@@ -39,7 +40,19 @@ msm_gem_address_space_get(struct msm_gem_address_space *aspace) ...@@ -39,7 +40,19 @@ msm_gem_address_space_get(struct msm_gem_address_space *aspace)
bool msm_gem_vma_inuse(struct msm_gem_vma *vma) bool msm_gem_vma_inuse(struct msm_gem_vma *vma)
{ {
return !!vma->inuse; if (vma->inuse > 0)
return true;
while (vma->fence_mask) {
unsigned idx = ffs(vma->fence_mask) - 1;
if (!msm_fence_completed(vma->fctx[idx], vma->fence[idx]))
return true;
vma->fence_mask &= ~BIT(idx);
}
return false;
} }
/* Actually unmap memory for the vma */ /* Actually unmap memory for the vma */
...@@ -63,13 +76,24 @@ void msm_gem_purge_vma(struct msm_gem_address_space *aspace, ...@@ -63,13 +76,24 @@ void msm_gem_purge_vma(struct msm_gem_address_space *aspace,
} }
/* Remove reference counts for the mapping */ /* Remove reference counts for the mapping */
void msm_gem_unmap_vma(struct msm_gem_address_space *aspace, void msm_gem_unpin_vma(struct msm_gem_vma *vma)
struct msm_gem_vma *vma)
{ {
if (GEM_WARN_ON(!vma->inuse))
return;
if (!GEM_WARN_ON(!vma->iova)) if (!GEM_WARN_ON(!vma->iova))
vma->inuse--; vma->inuse--;
} }
/* Replace pin reference with fence: */
void msm_gem_unpin_vma_fenced(struct msm_gem_vma *vma, struct msm_fence_context *fctx)
{
vma->fctx[fctx->index] = fctx;
vma->fence[fctx->index] = fctx->last_fence;
vma->fence_mask |= BIT(fctx->index);
msm_gem_unpin_vma(vma);
}
/* Map and pin vma: */
int int
msm_gem_map_vma(struct msm_gem_address_space *aspace, msm_gem_map_vma(struct msm_gem_address_space *aspace,
struct msm_gem_vma *vma, int prot, struct msm_gem_vma *vma, int prot,
......
...@@ -14,9 +14,20 @@ module_param(num_hw_submissions, uint, 0600); ...@@ -14,9 +14,20 @@ module_param(num_hw_submissions, uint, 0600);
static struct dma_fence *msm_job_run(struct drm_sched_job *job) static struct dma_fence *msm_job_run(struct drm_sched_job *job)
{ {
struct msm_gem_submit *submit = to_msm_submit(job); struct msm_gem_submit *submit = to_msm_submit(job);
struct msm_fence_context *fctx = submit->ring->fctx;
struct msm_gpu *gpu = submit->gpu; struct msm_gpu *gpu = submit->gpu;
int i;
submit->hw_fence = msm_fence_alloc(submit->ring->fctx); submit->hw_fence = msm_fence_alloc(fctx);
for (i = 0; i < submit->nr_bos; i++) {
struct drm_gem_object *obj = &submit->bos[i].obj->base;
msm_gem_lock(obj);
msm_gem_unpin_vma_fenced(submit->bos[i].vma, fctx);
submit->bos[i].flags &= ~BO_PINNED;
msm_gem_unlock(obj);
}
pm_runtime_get_sync(&gpu->pdev->dev); pm_runtime_get_sync(&gpu->pdev->dev);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment