dev.dpdk.org archive mirror
 help / color / mirror / Atom feed
From: David Christensen <drc@linux.vnet.ibm.com>
To: "Takeshi Yoshimura <tyos@jp.ibm.com>; Anatoly Burakov"
	<anatoly.burakov@intel.com>,
	dev@dpdk.org
Cc: drc@ibm.com, pradeep@us.ibm.com
Subject: Re: [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le
Date: Thu, 13 Jun 2019 10:37:00 -0700	[thread overview]
Message-ID: <509fdc91-997d-6cae-491b-418365b4efd5@linux.vnet.ibm.com> (raw)
In-Reply-To: <20190613022239.6946-1-tyos@jp.ibm.com>


Adding the vfio maintainer on the To: line.

On 6/12/19 7:22 PM, Takeshi Yoshimura wrote:
> In ppc64le, expanding DMA areas always fail because we cannot remove
> a DMA window. As a result, we cannot allocate more than one memseg in
> ppc64le. This is because vfio_spapr_dma_mem_map() doesn't unmap all
> the mapped DMA before removing the window. This patch fixes this
> incorrect behavior.
> 
> I added a global variable to track current window size since we do
> not have better ways to get exact size of it than doing so. sPAPR
> IOMMU seems not to provide any ways to get window size with ioctl
> interfaces. rte_memseg_walk*() is currently used to calculate window
> size, but it walks memsegs that are marked as used, not mapped. So,
> we need to determine if a given memseg is mapped or not, otherwise
> the ioctl reports errors due to attempting to unregister memory
> addresses that are not registered. The global variable is excluded
> in non-ppc64le binaries.
> 
> Similar problems happen in user maps. We need to avoid attempting to
> unmap the address that is given as the function's parameter. The
> compaction of user maps prevents us from passing correct length for
> unmapping DMA at the window recreation. So, I removed it in ppc64le.
> 
> I also fixed the order of ioctl for unregister and unmap. The ioctl
> for unregister sometimes report device busy errors due to the
> existence of mapped area.

I count at least three different changes happening in this commit.  Can 
you break it up into a multi-part patchset that targets each change 
individually?  It would be best if you break out the PPC64 changes 
separately from the changes that affect all architectures.

> 
> Signed-off-by: Takeshi Yoshimura <tyos@jp.ibm.com>
> ---
>   lib/librte_eal/linux/eal/eal_vfio.c | 154 +++++++++++++++++++---------
>   1 file changed, 103 insertions(+), 51 deletions(-)
> 
> diff --git a/lib/librte_eal/linux/eal/eal_vfio.c b/lib/librte_eal/linux/eal/eal_vfio.c
> index f16c5c3c0..c1b275b56 100644
> --- a/lib/librte_eal/linux/eal/eal_vfio.c
> +++ b/lib/librte_eal/linux/eal/eal_vfio.c
> @@ -93,6 +93,7 @@ is_null_map(const struct user_mem_map *map)
>   	return map->addr == 0 && map->iova == 0 && map->len == 0;
>   }
> 
> +#ifndef RTE_ARCH_PPC_64
>   /* we may need to merge user mem maps together in case of user mapping/unmapping
>    * chunks of memory, so we'll need a comparator function to sort segments.
>    */
> @@ -126,6 +127,7 @@ user_mem_map_cmp(const void *a, const void *b)
> 
>   	return 0;
>   }
> +#endif
> 
>   /* adjust user map entry. this may result in shortening of existing map, or in
>    * splitting existing map in two pieces.
> @@ -162,6 +164,7 @@ adjust_map(struct user_mem_map *src, struct user_mem_map *end,
>   	}
>   }
> 
> +#ifndef RTE_ARCH_PPC_64
>   /* try merging two maps into one, return 1 if succeeded */
>   static int
>   merge_map(struct user_mem_map *left, struct user_mem_map *right)
> @@ -177,6 +180,7 @@ merge_map(struct user_mem_map *left, struct user_mem_map *right)
> 
>   	return 1;
>   }
> +#endif
> 
>   static struct user_mem_map *
>   find_user_mem_map(struct user_mem_maps *user_mem_maps, uint64_t addr,
> @@ -211,6 +215,16 @@ find_user_mem_map(struct user_mem_maps *user_mem_maps, uint64_t addr,
>   	return NULL;
>   }
> 
> +#ifdef RTE_ARCH_PPC_64
> +/* Recreation of DMA window requires unregistering DMA memory.
> + * Compaction confuses the logic and causes false reports in the recreation.
> + * For now, we do not compact user maps in ppc64le.
> + */
> +static void
> +compact_user_maps(__rte_unused struct user_mem_maps *user_mem_maps)
> +{
> +}
> +#else
>   /* this will sort all user maps, and merge/compact any adjacent maps */
>   static void
>   compact_user_maps(struct user_mem_maps *user_mem_maps)
> @@ -256,6 +270,7 @@ compact_user_maps(struct user_mem_maps *user_mem_maps)
>   		user_mem_maps->n_maps = cur_idx;
>   	}
>   }
> +#endif
> 
>   static int
>   vfio_open_group_fd(int iommu_group_num)
> @@ -1306,6 +1321,7 @@ vfio_type1_dma_map(int vfio_container_fd)
>   	return rte_memseg_walk(type1_map, &vfio_container_fd);
>   }

The changes below starting with this #ifdef hide a lot of code on 
x86/ARM.  Was that your intent?  Does x86/ARM still work without it?

> +#ifdef RTE_ARCH_PPC_64
>   static int
>   vfio_spapr_dma_do_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
>   		uint64_t len, int do_map)
> @@ -1357,14 +1373,6 @@ vfio_spapr_dma_do_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
>   		}
> 
>   	} else {
> -		ret = ioctl(vfio_container_fd,
> -				VFIO_IOMMU_SPAPR_UNREGISTER_MEMORY, &reg);
> -		if (ret) {
> -			RTE_LOG(ERR, EAL, "  cannot unregister vaddr for IOMMU, error %i (%s)\n",
> -					errno, strerror(errno));
> -			return -1;
> -		}
> -
>   		memset(&dma_unmap, 0, sizeof(dma_unmap));
>   		dma_unmap.argsz = sizeof(struct vfio_iommu_type1_dma_unmap);
>   		dma_unmap.size = len;
> @@ -1377,24 +1385,50 @@ vfio_spapr_dma_do_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
>   					errno, strerror(errno));
>   			return -1;
>   		}
> +
> +		ret = ioctl(vfio_container_fd,
> +				VFIO_IOMMU_SPAPR_UNREGISTER_MEMORY, &reg);
> +		if (ret) {
> +			RTE_LOG(ERR, EAL, "  cannot unregister vaddr for IOMMU, error %i (%s)\n",
> +					errno, strerror(errno));
> +			return -1;
> +		}
>   	}
> 
>   	return 0;
>   }
> 
> +struct spapr_remap_walk_param {
> +	int vfio_container_fd;
> +	uint64_t window_size;
> +};
> +
>   static int
>   vfio_spapr_map_walk(const struct rte_memseg_list *msl,
>   		const struct rte_memseg *ms, void *arg)
>   {
> -	int *vfio_container_fd = arg;
> +	struct spapr_remap_walk_param *p = arg;
> 
> -	if (msl->external)
> +	if (msl->external || ms->iova >= p->window_size)
>   		return 0;
> 
> -	return vfio_spapr_dma_do_map(*vfio_container_fd, ms->addr_64, ms->iova,
> +	return vfio_spapr_dma_do_map(p->vfio_container_fd, ms->addr_64, ms->iova,
>   			ms->len, 1);
>   }
> 
> +static int
> +vfio_spapr_unmap_walk(const struct rte_memseg_list *msl,
> +		const struct rte_memseg *ms, void *arg)
> +{
> +	struct spapr_remap_walk_param *p = arg;
> +
> +	if (msl->external || ms->iova >= p->window_size)
> +		return 0;
> +
> +	return vfio_spapr_dma_do_map(p->vfio_container_fd, ms->addr_64, ms->iova,
> +			ms->len, 0);
> +}
> +
>   struct spapr_walk_param {
>   	uint64_t window_size;
>   	uint64_t hugepage_sz;
> @@ -1481,14 +1515,13 @@ vfio_spapr_create_new_dma_window(int vfio_container_fd,
>   	return 0;
>   }
> 
> +static struct vfio_iommu_spapr_tce_create prev_create;

Not a fan of global variables.  Also, you're using the value of this 
uninitialized variable in the first invocation of vfio_spapr_dma_mem_map().

> +
>   static int
>   vfio_spapr_dma_mem_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
>   		uint64_t len, int do_map)
>   {
> -	struct spapr_walk_param param;
> -	struct vfio_iommu_spapr_tce_create create = {
> -		.argsz = sizeof(create),
> -	};
> +	struct vfio_iommu_spapr_tce_create create;
>   	struct vfio_config *vfio_cfg;
>   	struct user_mem_maps *user_mem_maps;
>   	int i, ret = 0;
> @@ -1502,43 +1535,59 @@ vfio_spapr_dma_mem_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
>   	user_mem_maps = &vfio_cfg->mem_maps;
>   	rte_spinlock_recursive_lock(&user_mem_maps->lock);
> 
> -	/* check if window size needs to be adjusted */
> -	memset(&param, 0, sizeof(param));
> -
> -	/* we're inside a callback so use thread-unsafe version */
> -	if (rte_memseg_walk_thread_unsafe(vfio_spapr_window_size_walk,
> -				&param) < 0) {
> -		RTE_LOG(ERR, EAL, "Could not get window size\n");
> -		ret = -1;
> -		goto out;
> -	}
> +	memcpy(&create, &prev_create, sizeof(create));
> 
>   	/* also check user maps */
>   	for (i = 0; i < user_mem_maps->n_maps; i++) {
> -		uint64_t max = user_mem_maps->maps[i].iova +
> -				user_mem_maps->maps[i].len;
> -		create.window_size = RTE_MAX(create.window_size, max);
> +		struct user_mem_map *map = &user_mem_maps->maps[i];
> +
> +		if (vaddr == map->addr && len == map->len)
> +			continue;
> +		create.window_size = RTE_MAX(create.window_size, map->iova + map->len);
>   	}
> 
>   	/* sPAPR requires window size to be a power of 2 */
> -	create.window_size = rte_align64pow2(param.window_size);
> -	create.page_shift = __builtin_ctzll(param.hugepage_sz);
> -	create.levels = 1;
> +	create.window_size = rte_align64pow2(create.window_size);
> 
>   	if (do_map) {
> -		void *addr;
>   		/* re-create window and remap the entire memory */
> -		if (iova > create.window_size) {
> +		if (iova + len > create.window_size) {
> +			struct spapr_remap_walk_param param = {
> +				.vfio_container_fd = vfio_container_fd,
> +			    .window_size = create.window_size,
> +			};
> +
> +			/* we're inside a callback, so use thread-unsafe version
> +			 */
> +			rte_memseg_walk_thread_unsafe(vfio_spapr_unmap_walk,
> +					&param);
> +			/* destruct all user maps */
> +			for (i = 0; i < user_mem_maps->n_maps; i++) {
> +				struct user_mem_map *map =
> +						&user_mem_maps->maps[i];
> +				if (vaddr == map->addr && len == map->len)
> +					continue;
> +				if (vfio_spapr_dma_do_map(vfio_container_fd,
> +						map->addr, map->iova, map->len,
> +						0)) {
> +					RTE_LOG(ERR, EAL, "Could not destruct user DMA maps\n");
> +					ret = -1;
> +					goto out;
> +				}
> +			}
> +
> +			create.window_size = rte_align64pow2(iova + len);
>   			if (vfio_spapr_create_new_dma_window(vfio_container_fd,
>   					&create) < 0) {
>   				RTE_LOG(ERR, EAL, "Could not create new DMA window\n");
>   				ret = -1;
>   				goto out;
>   			}
> +			memcpy(&prev_create, &create, sizeof(create));
>   			/* we're inside a callback, so use thread-unsafe version
>   			 */
>   			if (rte_memseg_walk_thread_unsafe(vfio_spapr_map_walk,
> -					&vfio_container_fd) < 0) {
> +					&param) < 0) {
>   				RTE_LOG(ERR, EAL, "Could not recreate DMA maps\n");
>   				ret = -1;
>   				goto out;
> @@ -1547,6 +1596,8 @@ vfio_spapr_dma_mem_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
>   			for (i = 0; i < user_mem_maps->n_maps; i++) {
>   				struct user_mem_map *map =
>   						&user_mem_maps->maps[i];
> +				if (vaddr == map->addr && len == map->len)
> +					continue;
>   				if (vfio_spapr_dma_do_map(vfio_container_fd,
>   						map->addr, map->iova, map->len,
>   						1)) {
> @@ -1556,23 +1607,8 @@ vfio_spapr_dma_mem_map(int vfio_container_fd, uint64_t vaddr, uint64_t iova,
>   				}
>   			}
>   		}
> -
> -		/* now that we've remapped all of the memory that was present
> -		 * before, map the segment that we were requested to map.
> -		 *
> -		 * however, if we were called by the callback, the memory we
> -		 * were called with was already in the memseg list, so previous
> -		 * mapping should've mapped that segment already.
> -		 *
> -		 * virt2memseg_list is a relatively cheap check, so use that. if
> -		 * memory is within any memseg list, it's a memseg, so it's
> -		 * already mapped.
> -		 */
> -		addr = (void *)(uintptr_t)vaddr;
> -		if (rte_mem_virt2memseg_list(addr) == NULL &&
> -				vfio_spapr_dma_do_map(vfio_container_fd,
> -					vaddr, iova, len, 1) < 0) {
> -			RTE_LOG(ERR, EAL, "Could not map segment\n");
> +		if (vfio_spapr_dma_do_map(vfio_container_fd, vaddr, iova, len, 1)) {
> +			RTE_LOG(ERR, EAL, "Failed to map DMA\n");
>   			ret = -1;
>   			goto out;
>   		}
> @@ -1613,6 +1649,7 @@ vfio_spapr_dma_map(int vfio_container_fd)
>   		RTE_LOG(ERR, EAL, "Could not create new DMA window\n");
>   		return -1;
>   	}
> +	memcpy(&prev_create, &create, sizeof(create));
> 
>   	/* map all DPDK segments for DMA. use 1:1 PA to IOVA mapping */
>   	if (rte_memseg_walk(vfio_spapr_map_walk, &vfio_container_fd) < 0)
> @@ -1620,6 +1657,21 @@ vfio_spapr_dma_map(int vfio_container_fd)
> 
>   	return 0;
>   }
> +#else
> +static int
> +vfio_spapr_dma_mem_map(int __rte_unused vfio_container_fd,
> +			uint64_t __rte_unused vaddr,
> +			uint64_t __rte_unused iova, uint64_t __rte_unused len,
> +			int __rte_unused do_map)
> +{
> +	return 0;
> +}
> +static int
> +vfio_spapr_dma_map(int __rte_unused vfio_container_fd)
> +{
> +	return 0;
> +}
> +#endif
> 
>   static int
>   vfio_noiommu_dma_map(int __rte_unused vfio_container_fd)
> 


  reply	other threads:[~2019-06-13 17:37 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-12  6:33 [dpdk-dev] [PATCH] vfio: fix expanding DMA area in ppc64le Takeshi Yoshimura
2019-06-12 14:06 ` Aaron Conole
2019-06-13  2:22 ` Takeshi Yoshimura
2019-06-13 17:37   ` David Christensen [this message]
2019-06-14  7:34   ` David Marchand
2019-06-14  7:49   ` [dpdk-dev] [PATCH v2] " Takeshi Yoshimura
2019-07-13  1:15     ` [dpdk-dev] [PATCH v3] " Takeshi Yoshimura
2019-07-16  0:20       ` David Christensen
2019-07-16 10:56         ` Thomas Monjalon
2019-06-18  2:37   ` [dpdk-dev] [PATCH] " Mo, YufengX
2019-06-18  2:39   ` Mo, YufengX
2019-06-26  9:43   ` Burakov, Anatoly
2019-06-28 11:38   ` Takeshi T Yoshimura
2019-06-28 13:47     ` Burakov, Anatoly
2019-06-28 14:04       ` Burakov, Anatoly
2019-06-13  2:30 ` Takeshi T Yoshimura

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=509fdc91-997d-6cae-491b-418365b4efd5@linux.vnet.ibm.com \
    --to=drc@linux.vnet.ibm.com \
    --cc=anatoly.burakov@intel.com \
    --cc=dev@dpdk.org \
    --cc=drc@ibm.com \
    --cc=pradeep@us.ibm.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;
as well as URLs for NNTP newsgroup(s).