Re: [PATCH v12 09/13] vduse: take out allocations from vduse_dev_alloc_coherent

["Michael S. Tsirkin" <mst@redhat.com>]

On Thu, Jan 15, 2026 at 04:22:49PM +0800, Jason Wang wrote:
> On Thu, Jan 15, 2026 at 2:49 AM Eugenio Pérez <eperezma@redhat.com> wrote:
> >
> > The function vduse_dev_alloc_coherent will be called under rwlock in
> > next patches.  Make it out of the lock to avoid increasing its fail
> > rate.
> >
> > Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
> > ---
> > v12:
> > * Avoid free_pages_exact(NULL, size) in case vduse_domain_alloc_coherent
> >   fails (MST).
> >
> > v11: Remove duplicated call to free_pages_exact (Jason).
> > ---
> >  drivers/vdpa/vdpa_user/iova_domain.c | 24 +++++++-----------------
> >  drivers/vdpa/vdpa_user/iova_domain.h |  5 ++---
> >  drivers/vdpa/vdpa_user/vduse_dev.c   | 17 +++++++++++------
> >  3 files changed, 20 insertions(+), 26 deletions(-)
> >
> > diff --git a/drivers/vdpa/vdpa_user/iova_domain.c b/drivers/vdpa/vdpa_user/iova_domain.c
> > index 309cd5a039d1..0a9f668467a8 100644
> > --- a/drivers/vdpa/vdpa_user/iova_domain.c
> > +++ b/drivers/vdpa/vdpa_user/iova_domain.c
> > @@ -493,17 +493,15 @@ void vduse_domain_unmap_page(struct vduse_iova_domain *domain,
> >         vduse_domain_free_iova(iovad, dma_addr, size);
> >  }
> >
> > -void *vduse_domain_alloc_coherent(struct vduse_iova_domain *domain,
> > -                                 size_t size, dma_addr_t *dma_addr,
> > -                                 gfp_t flag)
> > +dma_addr_t vduse_domain_alloc_coherent(struct vduse_iova_domain *domain,
> > +                                      size_t size, void *orig)
> >  {
> >         struct iova_domain *iovad = &domain->consistent_iovad;
> >         unsigned long limit = domain->iova_limit;
> >         dma_addr_t iova = vduse_domain_alloc_iova(iovad, size, limit);
> > -       void *orig = alloc_pages_exact(size, flag);
> >
> > -       if (!iova || !orig)
> > -               goto err;
> > +       if (!iova)
> > +               return DMA_MAPPING_ERROR;
> >
> >         spin_lock(&domain->iotlb_lock);
> >         if (vduse_iotlb_add_range(domain, (u64)iova, (u64)iova + size - 1,
> > @@ -514,17 +512,12 @@ void *vduse_domain_alloc_coherent(struct vduse_iova_domain *domain,
> >         }
> >         spin_unlock(&domain->iotlb_lock);
> >
> > -       *dma_addr = iova;
> > +       return iova;
> >
> > -       return orig;
> >  err:
> > -       *dma_addr = DMA_MAPPING_ERROR;
> > -       if (orig)
> > -               free_pages_exact(orig, size);
> > -       if (iova)
> > -               vduse_domain_free_iova(iovad, iova, size);
> > +       vduse_domain_free_iova(iovad, iova, size);
> >
> > -       return NULL;
> > +       return DMA_MAPPING_ERROR;
> >  }
> >
> >  void vduse_domain_free_coherent(struct vduse_iova_domain *domain, size_t size,
> > @@ -533,7 +526,6 @@ void vduse_domain_free_coherent(struct vduse_iova_domain *domain, size_t size,
> >         struct iova_domain *iovad = &domain->consistent_iovad;
> >         struct vhost_iotlb_map *map;
> >         struct vdpa_map_file *map_file;
> > -       phys_addr_t pa;
> >
> >         spin_lock(&domain->iotlb_lock);
> >         map = vhost_iotlb_itree_first(domain->iotlb, (u64)dma_addr,
> > @@ -545,12 +537,10 @@ void vduse_domain_free_coherent(struct vduse_iova_domain *domain, size_t size,
> >         map_file = (struct vdpa_map_file *)map->opaque;
> >         fput(map_file->file);
> >         kfree(map_file);
> > -       pa = map->addr;
> >         vhost_iotlb_map_free(domain->iotlb, map);
> >         spin_unlock(&domain->iotlb_lock);
> >
> >         vduse_domain_free_iova(iovad, dma_addr, size);
> > -       free_pages_exact(phys_to_virt(pa), size);
> >  }
> >
> >  static vm_fault_t vduse_domain_mmap_fault(struct vm_fault *vmf)
> > diff --git a/drivers/vdpa/vdpa_user/iova_domain.h b/drivers/vdpa/vdpa_user/iova_domain.h
> > index 081f06c52cdc..e50e55d1396f 100644
> > --- a/drivers/vdpa/vdpa_user/iova_domain.h
> > +++ b/drivers/vdpa/vdpa_user/iova_domain.h
> > @@ -65,9 +65,8 @@ void vduse_domain_unmap_page(struct vduse_iova_domain *domain,
> >                              dma_addr_t dma_addr, size_t size,
> >                              enum dma_data_direction dir, unsigned long attrs);
> >
> > -void *vduse_domain_alloc_coherent(struct vduse_iova_domain *domain,
> > -                                 size_t size, dma_addr_t *dma_addr,
> > -                                 gfp_t flag);
> > +dma_addr_t vduse_domain_alloc_coherent(struct vduse_iova_domain *domain,
> > +                                      size_t size, void *orig);
> >
> >  void vduse_domain_free_coherent(struct vduse_iova_domain *domain, size_t size,
> >                                 dma_addr_t dma_addr, unsigned long attrs);
> > diff --git a/drivers/vdpa/vdpa_user/vduse_dev.c b/drivers/vdpa/vdpa_user/vduse_dev.c
> > index 0e3cf5128ad0..6dba1f3224d9 100644
> > --- a/drivers/vdpa/vdpa_user/vduse_dev.c
> > +++ b/drivers/vdpa/vdpa_user/vduse_dev.c
> > @@ -916,23 +916,27 @@ static void *vduse_dev_alloc_coherent(union virtio_map token, size_t size,
> >  {
> >         struct vduse_dev *vdev;
> >         struct vduse_iova_domain *domain;
> > -       unsigned long iova;
> >         void *addr;
> >
> >         *dma_addr = DMA_MAPPING_ERROR;
> >         if (!token.group)
> >                 return NULL;
> >
> > -       vdev = token.group->dev;
> > -       domain = vdev->domain;
> > -       addr = vduse_domain_alloc_coherent(domain, size,
> > -                                          (dma_addr_t *)&iova, flag);
> > +       addr = alloc_pages_exact(size, flag);
> >         if (!addr)
> >                 return NULL;
> >
> > -       *dma_addr = (dma_addr_t)iova;
> > +       vdev = token.group->dev;
> > +       domain = vdev->domain;
> > +       *dma_addr = vduse_domain_alloc_coherent(domain, size, addr);
> > +       if (*dma_addr == DMA_MAPPING_ERROR)
> > +               goto err;
> >
> >         return addr;
> > +
> > +err:
> > +       free_pages_exact(addr, size);
> > +       return NULL;
> >  }
> 
> I think the code tries to hide the iova domain specific things (e.g
> mappings and DMA_MAPPING_ERROR) from vduse core. Could we find a way
> to stick to that?
> 
> Thanks


what do you suggest more spefically though?

> >
> >  static void vduse_dev_free_coherent(union virtio_map token, size_t size,
> > @@ -949,6 +953,7 @@ static void vduse_dev_free_coherent(union virtio_map token, size_t size,
> >         domain = vdev->domain;
> >
> >         vduse_domain_free_coherent(domain, size, dma_addr, attrs);
> > +       free_pages_exact(vaddr, size);
> >  }
> >
> >  static bool vduse_dev_need_sync(union virtio_map token, dma_addr_t dma_addr)
> > --
> > 2.52.0
> >