search for: nouveau_dmem_init

In nouveau_dmem_init(), a number of struct nouveau_dmem_chunk are allocated and put on the dmem->chunk_empty list. Then in nouveau_dmem_pages_alloc(), a nouveau_dmem_chunk is removed from the list and GPU memory is allocated. However, the nouveau_dmem_chunk is never removed from the chunk_empty list nor placed on th...

...c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c index aa9fec80492d..28624dd0c920 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -623,7 +623,7 @@ nouveau_dmem_init(struct nouveau_drm *drm) */ drm->dmem->devmem = hmm_devmem_add(&nouveau_dmem_devmem_ops, device, size); - if (drm->dmem->devmem == NULL) { + if (IS_ERR(drm->dmem->devmem)) { kfree(drm->dmem); drm->dmem = NULL; return; -- 2.17.1

...----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c index aa9fec80492d..35b6e83ead8a 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -593,6 +593,11 @@ nouveau_dmem_init(struct nouveau_drm *drm) unsigned long i, size; int ret; + mutex_init(&drm->dmem->mutex); + INIT_LIST_HEAD(&drm->dmem->chunk_free); + INIT_LIST_HEAD(&drm->dmem->chunk_full); + INIT_LIST_HEAD(&drm->dmem->chunk_empty); + /* This only make sense on PASCA...

From: Yonatan Maman <Ymaman at Nvidia.com> This patch series aims to enable Peer-to-Peer (P2P) DMA access in GPU-centric applications that utilize RDMA and private device pages. This enhancement is crucial for minimizing data transfer overhead by allowing the GPU to directly expose device private page data to devices such as NICs, eliminating the need to traverse system RAM, which is the

When acting on device private mappings a driver needs to know if the device (or other entity in case of kvmppc) actually owns this private mapping. This series adds an owner field and converts the migrate_vma code over to check it. I looked into doing the same for hmm_range_fault, but as far as I can tell that code has never been wired up to actually work for device private memory, so instead of

On Tue, Apr 21, 2020 at 04:11:07PM -0700, Ralph Campbell wrote: > In nouveau_dmem_init(), a number of struct nouveau_dmem_chunk are allocated > and put on the dmem->chunk_empty list. Then in nouveau_dmem_pages_alloc(), > a nouveau_dmem_chunk is removed from the list and GPU memory is allocated. > However, the nouveau_dmem_chunk is never removed from the chunk_empty > l...

..._NO_NODE); if (IS_ERR(addr)) { ret = PTR_ERR(addr); diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouveau/nouveau_dmem.c index 0ad5d87b5a8e..a4682272586e 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -526,6 +526,7 @@ nouveau_dmem_init(struct nouveau_drm *drm) drm->dmem->pagemap.type = MEMORY_DEVICE_PRIVATE; drm->dmem->pagemap.res = *res; drm->dmem->pagemap.ops = &nouveau_dmem_pagemap_ops; + drm->dmem->pagemap.owner = drm->dev; if (IS_ERR(devm_memremap_pages(device, &drm->dmem->pag...

...? page_mapping+0x98/0x110 [ 1295.009511] migrate_vma+0xa74/0x1090 [ 1295.013186] ? move_to_new_page+0x480/0x480 [ 1295.017400] ? __kmalloc+0x153/0x300 [ 1295.021052] ? nouveau_dmem_migrate_vma+0xd8/0x1e0 [nouveau] [ 1295.026796] nouveau_dmem_migrate_vma+0x157/0x1e0 [nouveau] [ 1295.032466] ? nouveau_dmem_init+0x490/0x490 [nouveau] [ 1295.037612] ? vmacache_find+0xc2/0x110 [ 1295.041537] nouveau_svmm_bind+0x1b4/0x210 [nouveau] [ 1295.046583] ? nouveau_svm_fault+0x13e0/0x13e0 [nouveau] [ 1295.051912] drm_ioctl_kernel+0x14d/0x1a0 [ 1295.055930] ? drm_setversion+0x330/0x330 [ 1295.059971] drm_ioctl+0x...

...? page_mapping+0x98/0x110 [ 1295.009511] migrate_vma+0xa74/0x1090 [ 1295.013186] ? move_to_new_page+0x480/0x480 [ 1295.017400] ? __kmalloc+0x153/0x300 [ 1295.021052] ? nouveau_dmem_migrate_vma+0xd8/0x1e0 [nouveau] [ 1295.026796] nouveau_dmem_migrate_vma+0x157/0x1e0 [nouveau] [ 1295.032466] ? nouveau_dmem_init+0x490/0x490 [nouveau] [ 1295.037612] ? vmacache_find+0xc2/0x110 [ 1295.041537] nouveau_svmm_bind+0x1b4/0x210 [nouveau] [ 1295.046583] ? nouveau_svm_fault+0x13e0/0x13e0 [nouveau] [ 1295.051912] drm_ioctl_kernel+0x14d/0x1a0 [ 1295.055930] ? drm_setversion+0x330/0x330 [ 1295.059971] drm_ioctl+0x...

.../nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -176,6 +176,7 @@ static vm_fault_t nouveau_dmem_migrate_to_ram(struct vm_fault *vmf) .end = vmf->address + PAGE_SIZE, .src = &src, .dst = &dst, + .dev_private_owner = drm->dev, }; /* @@ -526,6 +527,7 @@ nouveau_dmem_init(struct nouveau_drm *drm) drm->dmem->pagemap.type = MEMORY_DEVICE_PRIVATE; drm->dmem->pagemap.res = *res; drm->dmem->pagemap.ops = &nouveau_dmem_pagemap_ops; + drm->dmem->pagemap.owner = drm->dev; if (IS_ERR(devm_memremap_pages(device, &drm->dmem->pag...

...x_unlock(&svmm->mutex); args->dst[0] = migrate_pfn(page_to_pfn(dpage)) | MIGRATE_PFN_LOCKED; return 0; error_dma_unmap: + mutex_unlock(&svmm->mutex); dma_unmap_page(dev, *dma_addr, PAGE_SIZE, DMA_BIDIRECTIONAL); error_free_page: __free_page(dpage); @@ -531,7 +537,8 @@ nouveau_dmem_init(struct nouveau_drm *drm) } static unsigned long nouveau_dmem_migrate_copy_one(struct nouveau_drm *drm, - unsigned long src, dma_addr_t *dma_addr, u64 *pfn) + struct nouveau_svmm *svmm, unsigned long src, + dma_addr_t *dma_addr, u64 *pfn) { struct device *dev = drm->dev->dev; struc...

...dmem.h @@ -38,8 +38,6 @@ int nouveau_dmem_migrate_vma(struct nouveau_drm *drm, unsigned long start, unsigned long end); -void nouveau_dmem_convert_pfn(struct nouveau_drm *drm, - struct hmm_range *range); #else /* IS_ENABLED(CONFIG_DRM_NOUVEAU_SVM) */ static inline void nouveau_dmem_init(struct nouveau_drm *drm) {} static inline void nouveau_dmem_fini(struct nouveau_drm *drm) {} diff --git a/drivers/gpu/drm/nouveau/nouveau_svm.c b/drivers/gpu/drm/nouveau/nouveau_svm.c index df9bf1fd1bc0..7e0376dca137 100644 --- a/drivers/gpu/drm/nouveau/nouveau_svm.c +++ b/drivers/gpu/drm/nouveau/...

...a_struct *vma; - struct nouveau_drm *drm; - struct nouveau_fence *fence; - unsigned long npages; - dma_addr_t *dma; - unsigned long dma_nr; -}; - static unsigned long nouveau_dmem_page_addr(struct page *page) { struct nouveau_dmem_chunk *chunk = page->zone_device_data; @@ -569,131 +558,67 @@ nouveau_dmem_init(struct nouveau_drm *drm) drm->dmem = NULL; } -static void -nouveau_dmem_migrate_alloc_and_copy(struct vm_area_struct *vma, - const unsigned long *src_pfns, - unsigned long *dst_pfns, - unsigned long start, - unsigned long end, - struct nouveau_migrate *mig...

...? page_mapping+0x98/0x110 [ 1295.009511] migrate_vma+0xa74/0x1090 [ 1295.013186] ? move_to_new_page+0x480/0x480 [ 1295.017400] ? __kmalloc+0x153/0x300 [ 1295.021052] ? nouveau_dmem_migrate_vma+0xd8/0x1e0 [nouveau] [ 1295.026796] nouveau_dmem_migrate_vma+0x157/0x1e0 [nouveau] [ 1295.032466] ? nouveau_dmem_init+0x490/0x490 [nouveau] [ 1295.037612] ? vmacache_find+0xc2/0x110 [ 1295.041537] nouveau_svmm_bind+0x1b4/0x210 [nouveau] [ 1295.046583] ? nouveau_svm_fault+0x13e0/0x13e0 [nouveau] [ 1295.051912] drm_ioctl_kernel+0x14d/0x1a0 [ 1295.055930] ? drm_setversion+0x330/0x330 [ 1295.059971] drm_ioctl+0x...

...au_bo.h" +#include "nouveau_svm.h" #include <nvif/class.h> #include <nvif/object.h> #include <nvif/if500b.h> #include <nvif/if900b.h> +#include <nvif/if000c.h> #include <linux/sched/mm.h> #include <linux/hmm.h> @@ -560,11 +562,12 @@ nouveau_dmem_init(struct nouveau_drm *drm) } static unsigned long nouveau_dmem_migrate_copy_one(struct nouveau_drm *drm, - struct vm_area_struct *vma, unsigned long addr, - unsigned long src, dma_addr_t *dma_addr) + struct vm_area_struct *vma, unsigned long src, + dma_addr_t *dma_addr, u64 *pfn) { struct...

In nouveau_dmem_pages_alloc(), the drm->dmem->mutex is unlocked before calling nouveau_dmem_chunk_alloc(). Reacquire the lock before continuing to the next page. Signed-off-by: Ralph Campbell <rcampbell at nvidia.com> --- I found this while testing Jason Gunthorpe's hmm tree but this is independant of those changes. I guess it could go through David Airlie's tree for nouveau

Hi, just for your information and maybe for some help: with 5.1rc1 and SVM enabled i see the following backtrace [1] when the nouveau card (reverse prime) goes to sleep, for now i have papered over with [2] which leaves me with userspace hangs. Any pointers where to look for the actual culprit? PS: Card is: nouveau 0000:01:00.0: NVIDIA GP106 (136000a1) Greetings, Tobias [1]: BUG: unable

...> > #include <nvif/class.h> > #include <nvif/object.h> > #include <nvif/if500b.h> > #include <nvif/if900b.h> > +#include <nvif/if000c.h> > > #include <linux/sched/mm.h> > #include <linux/hmm.h> > @@ -560,11 +562,12 @@ nouveau_dmem_init(struct nouveau_drm *drm) > } > > static unsigned long nouveau_dmem_migrate_copy_one(struct nouveau_drm *drm, > - struct vm_area_struct *vma, unsigned long addr, > - unsigned long src, dma_addr_t *dma_addr) > + struct vm_area_struct *vma, unsigned long src, > + dma_addr_...

Hi Dan, Jérôme and Jason, below is a series that cleans up the dev_pagemap interface so that it is more easily usable, which removes the need to wrap it in hmm and thus allowing to kill a lot of code Note: this series is on top of the rdma/hmm branch + the dev_pagemap releas fix series from Dan that went into 5.2-rc5. Git tree: git://git.infradead.org/users/hch/misc.git

...au_bo.h" +#include "nouveau_svm.h" #include <nvif/class.h> #include <nvif/object.h> #include <nvif/if500b.h> #include <nvif/if900b.h> +#include <nvif/if000c.h> #include <linux/sched/mm.h> #include <linux/hmm.h> @@ -558,10 +560,11 @@ nouveau_dmem_init(struct nouveau_drm *drm) } static unsigned long nouveau_dmem_migrate_copy_one(struct nouveau_drm *drm, - unsigned long src, dma_addr_t *dma_addr) + unsigned long src, dma_addr_t *dma_addr, u64 *pfn) { struct device *dev = drm->dev->dev; struct page *dpage, *spage; + unsigned long p...