[RFC v2 20/21] nvme-pci: use new dma API
Leon Romanovsky
leon at kernel.org
Thu Sep 12 04:15:55 PDT 2024
From: Leon Romanovsky <leonro at nvidia.com>
This demonstrates how the new DMA API can fit into the NVMe driver and
replace the old DMA APIs.
As this is an RFC, I expect more robust error handling, optimizations,
and in-depth testing for the final version once we agree on DMA API
architecture.
Following is the performance comparision for existing DMA API case
with sg_table and with dma_map, once we have agreement on the new DMA
API design I intend to get similar profiling numbers for new DMA API.
sgl (sg_table + old dma API ) vs no_sgl (iod_dma_map + new DMA API) :-
block size IOPS (k) Average of 3
4K
--------------------------------------------------------------
sg-list-fio-perf.bs-4k-1.fio: 68.6
sg-list-fio-perf.bs-4k-2.fio: 68 68.36
sg-list-fio-perf.bs-4k-3.fio: 68.5
no-sg-list-fio-perf.bs-4k-1.fio: 68.7
no-sg-list-fio-perf.bs-4k-2.fio: 68.5 68.43
no-sg-list-fio-perf.bs-4k-3.fio: 68.1
% Change default vs new DMA API = +0.0975%
8K
--------------------------------------------------------------
sg-list-fio-perf.bs-8k-1.fio: 67
sg-list-fio-perf.bs-8k-2.fio: 67.1 67.03
sg-list-fio-perf.bs-8k-3.fio: 67
no-sg-list-fio-perf.bs-8k-1.fio: 66.7
no-sg-list-fio-perf.bs-8k-2.fio: 66.7 66.7
no-sg-list-fio-perf.bs-8k-3.fio: 66.7
% Change default vs new DMA API = +0.4993%
16K
--------------------------------------------------------------
sg-list-fio-perf.bs-16k-1.fio: 63.8
sg-list-fio-perf.bs-16k-2.fio: 63.4 63.5
sg-list-fio-perf.bs-16k-3.fio: 63.3
no-sg-list-fio-perf.bs-16k-1.fio: 63.5
no-sg-list-fio-perf.bs-16k-2.fio: 63.4 63.33
no-sg-list-fio-perf.bs-16k-3.fio: 63.1
% Change default vs new DMA API = -0.2632%
32K
--------------------------------------------------------------
sg-list-fio-perf.bs-32k-1.fio: 59.3
sg-list-fio-perf.bs-32k-2.fio: 59.3 59.36
sg-list-fio-perf.bs-32k-3.fio: 59.5
no-sg-list-fio-perf.bs-32k-1.fio: 59.5
no-sg-list-fio-perf.bs-32k-2.fio: 59.6 59.43
no-sg-list-fio-perf.bs-32k-3.fio: 59.2
% Change default vs new DMA API = +0.1122%
64K
--------------------------------------------------------------
sg-list-fio-perf.bs-64k-1.fio: 53.7
sg-list-fio-perf.bs-64k-2.fio: 53.4 53.56
sg-list-fio-perf.bs-64k-3.fio: 53.6
no-sg-list-fio-perf.bs-64k-1.fio: 53.5
no-sg-list-fio-perf.bs-64k-2.fio: 53.8 53.63
no-sg-list-fio-perf.bs-64k-3.fio: 53.6
% Change default vs new DMA API = +0.1246%
128K
--------------------------------------------------------------
sg-list-fio-perf/bs-128k-1.fio: 48
sg-list-fio-perf/bs-128k-2.fio: 46.4 47.13
sg-list-fio-perf/bs-128k-3.fio: 47
no-sg-list-fio-perf/bs-128k-1.fio: 46.6
no-sg-list-fio-perf/bs-128k-2.fio: 47 46.9
no-sg-list-fio-perf/bs-128k-3.fio: 47.1
% Change default vs new DMA API = −0.495%
256K
--------------------------------------------------------------
sg-list-fio-perf/bs-256k-1.fio: 37
sg-list-fio-perf/bs-256k-2.fio: 41 39.93
sg-list-fio-perf/bs-256k-3.fio: 41.8
no-sg-list-fio-perf/bs-256k-1.fio: 37.5
no-sg-list-fio-perf/bs-256k-2.fio: 41.4 40.5
no-sg-list-fio-perf/bs-256k-3.fio: 42.6
% Change default vs new DMA API = +1.42%
512K
--------------------------------------------------------------
sg-list-fio-perf/bs-512k-1.fio: 28.5
sg-list-fio-perf/bs-512k-2.fio: 28.2 28.4
sg-list-fio-perf/bs-512k-3.fio: 28.5
no-sg-list-fio-perf/bs-512k-1.fio: 28.7
no-sg-list-fio-perf/bs-512k-2.fio: 28.6 28.7
no-sg-list-fio-perf/bs-512k-3.fio: 28.8
% Change default vs new DMA API = +1.06%
Signed-off-by: Chaitanya Kulkarni <kch at nvidia.com>
Signed-off-by: Leon Romanovsky <leonro at nvidia.com>
---
drivers/nvme/host/pci.c | 354 ++++++++++++++++++++++++----------------
1 file changed, 215 insertions(+), 139 deletions(-)
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 2b236b1d209e..881cbf2c0cac 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -221,6 +221,12 @@ union nvme_descriptor {
__le64 *prp_list;
};
+/* TODO: move to common header */
+struct dma_entry {
+ dma_addr_t addr;
+ unsigned int len;
+};
+
/*
* The nvme_iod describes the data in an I/O.
*
@@ -234,9 +240,11 @@ struct nvme_iod {
u8 nr_dmas;
s8 nr_allocations; /* PRP list pool allocations. 0 means small
pool in use */
+ struct dma_iova_state state;
+ struct dma_entry dma;
+ struct dma_entry *map;
dma_addr_t first_dma;
dma_addr_t meta_dma;
- struct sg_table sgt;
union nvme_descriptor list[NVME_MAX_NR_ALLOCATIONS];
};
@@ -540,10 +548,9 @@ static void nvme_free_prps(struct nvme_dev *dev, struct request *req)
static void nvme_unmap_data(struct nvme_dev *dev, struct request *req)
{
struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
-
- WARN_ON_ONCE(!iod->sgt.nents);
-
- dma_unmap_sgtable(dev->dev, &iod->sgt, rq_dma_dir(req), 0);
+ struct req_iterator iter;
+ struct bio_vec bv;
+ int cnt = 0;
if (iod->nr_allocations == 0)
dma_pool_free(dev->prp_small_pool, iod->list[0].sg_list,
@@ -553,20 +560,17 @@ static void nvme_unmap_data(struct nvme_dev *dev, struct request *req)
iod->first_dma);
else
nvme_free_prps(dev, req);
- mempool_free(iod->sgt.sgl, dev->iod_mempool);
-}
-static void nvme_print_sgl(struct scatterlist *sgl, int nents)
-{
- int i;
- struct scatterlist *sg;
-
- for_each_sg(sgl, sg, nents, i) {
- dma_addr_t phys = sg_phys(sg);
- pr_warn("sg[%d] phys_addr:%pad offset:%d length:%d "
- "dma_address:%pad dma_length:%d\n",
- i, &phys, sg->offset, sg->length, &sg_dma_address(sg),
- sg_dma_len(sg));
+ if (iod->map) {
+ rq_for_each_bvec(bv, req, iter) {
+ dma_unmap_page(dev->dev, iod->map[cnt].addr,
+ iod->map[cnt].len, rq_dma_dir(req));
+ cnt++;
+ }
+ kfree(iod->map);
+ } else {
+ dma_unlink_range(&iod->state);
+ dma_free_iova(&iod->state);
}
}
@@ -574,97 +578,63 @@ static blk_status_t nvme_pci_setup_prps(struct nvme_dev *dev,
struct request *req, struct nvme_rw_command *cmnd)
{
struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
- struct dma_pool *pool;
- int length = blk_rq_payload_bytes(req);
- struct scatterlist *sg = iod->sgt.sgl;
- int dma_len = sg_dma_len(sg);
- u64 dma_addr = sg_dma_address(sg);
- int offset = dma_addr & (NVME_CTRL_PAGE_SIZE - 1);
- __le64 *prp_list;
- dma_addr_t prp_dma;
- int nprps, i;
-
- length -= (NVME_CTRL_PAGE_SIZE - offset);
- if (length <= 0) {
- iod->first_dma = 0;
- goto done;
- }
-
- dma_len -= (NVME_CTRL_PAGE_SIZE - offset);
- if (dma_len) {
- dma_addr += (NVME_CTRL_PAGE_SIZE - offset);
- } else {
- sg = sg_next(sg);
- dma_addr = sg_dma_address(sg);
- dma_len = sg_dma_len(sg);
- }
+ __le64 *prp_list = iod->list[0].prp_list;
+ int i = 0, idx = 0;
+ struct bio_vec bv;
+ struct req_iterator iter;
+ dma_addr_t offset = 0;
- if (length <= NVME_CTRL_PAGE_SIZE) {
- iod->first_dma = dma_addr;
- goto done;
+ if (iod->nr_dmas <= 2) {
+ i = iod->nr_dmas;
+ /* We can use the inline PRP/SG list */
+ goto set_addr;
}
- nprps = DIV_ROUND_UP(length, NVME_CTRL_PAGE_SIZE);
- if (nprps <= (256 / 8)) {
- pool = dev->prp_small_pool;
- iod->nr_allocations = 0;
- } else {
- pool = dev->prp_page_pool;
- iod->nr_allocations = 1;
- }
+ rq_for_each_bvec(bv, req, iter) {
+ dma_addr_t addr;
- prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma);
- if (!prp_list) {
- iod->nr_allocations = -1;
- return BLK_STS_RESOURCE;
- }
- iod->list[0].prp_list = prp_list;
- iod->first_dma = prp_dma;
- i = 0;
- for (;;) {
- if (i == NVME_CTRL_PAGE_SIZE >> 3) {
- __le64 *old_prp_list = prp_list;
- prp_list = dma_pool_alloc(pool, GFP_ATOMIC, &prp_dma);
- if (!prp_list)
- goto free_prps;
- iod->list[iod->nr_allocations++].prp_list = prp_list;
- prp_list[0] = old_prp_list[i - 1];
- old_prp_list[i - 1] = cpu_to_le64(prp_dma);
- i = 1;
+ if (iod->map)
+ offset = 0;
+
+ while (offset < bv.bv_len) {
+ if (iod->map)
+ addr = iod->map[i].addr;
+ else
+ addr = iod->dma.addr;
+
+ prp_list[idx] = cpu_to_le64(addr + offset);
+ offset += NVME_CTRL_PAGE_SIZE;
+ idx++;
}
- prp_list[i++] = cpu_to_le64(dma_addr);
- dma_len -= NVME_CTRL_PAGE_SIZE;
- dma_addr += NVME_CTRL_PAGE_SIZE;
- length -= NVME_CTRL_PAGE_SIZE;
- if (length <= 0)
- break;
- if (dma_len > 0)
- continue;
- if (unlikely(dma_len < 0))
- goto bad_sgl;
- sg = sg_next(sg);
- dma_addr = sg_dma_address(sg);
- dma_len = sg_dma_len(sg);
- }
-done:
- cmnd->dptr.prp1 = cpu_to_le64(sg_dma_address(iod->sgt.sgl));
- cmnd->dptr.prp2 = cpu_to_le64(iod->first_dma);
+ i++;
+ }
+
+set_addr:
+ if (iod->map)
+ cmnd->dptr.prp1 = cpu_to_le64(iod->map[0].addr);
+ else
+ cmnd->dptr.prp1 = cpu_to_le64(iod->dma.addr);
+ if (idx == 1 && i == 1)
+ cmnd->dptr.prp2 = 0;
+ else if (idx == 2 && i == 2)
+ if (iod->map)
+ cmnd->dptr.prp2 =
+ cpu_to_le64((iod->map[0].addr + NVME_CTRL_PAGE_SIZE) &
+ ~(NVME_CTRL_PAGE_SIZE - 1));
+ else
+ cmnd->dptr.prp2 =
+ cpu_to_le64((iod->dma.addr + NVME_CTRL_PAGE_SIZE) &
+ ~(NVME_CTRL_PAGE_SIZE - 1));
+ else
+ cmnd->dptr.prp2 = cpu_to_le64(iod->first_dma);
return BLK_STS_OK;
-free_prps:
- nvme_free_prps(dev, req);
- return BLK_STS_RESOURCE;
-bad_sgl:
- WARN(DO_ONCE(nvme_print_sgl, iod->sgt.sgl, iod->sgt.nents),
- "Invalid SGL for payload:%d nents:%d\n",
- blk_rq_payload_bytes(req), iod->sgt.nents);
- return BLK_STS_IOERR;
}
-static void nvme_pci_sgl_set_data(struct nvme_sgl_desc *sge,
- struct scatterlist *sg)
+static void nvme_pci_sgl_set_data(struct nvme_sgl_desc *sge, dma_addr_t addr,
+ int len)
{
- sge->addr = cpu_to_le64(sg_dma_address(sg));
- sge->length = cpu_to_le32(sg_dma_len(sg));
+ sge->addr = cpu_to_le64(addr);
+ sge->length = cpu_to_le32(len);
sge->type = NVME_SGL_FMT_DATA_DESC << 4;
}
@@ -680,17 +650,77 @@ static blk_status_t nvme_pci_setup_sgls(struct nvme_dev *dev,
struct request *req, struct nvme_rw_command *cmd)
{
struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
- struct dma_pool *pool;
- struct nvme_sgl_desc *sg_list;
- struct scatterlist *sg = iod->sgt.sgl;
- unsigned int entries = iod->sgt.nents;
- dma_addr_t sgl_dma;
- int i = 0;
+ struct nvme_sgl_desc *sg_list = iod->list[0].sg_list;
+ struct bio_vec bv = req_bvec(req);
+ struct req_iterator iter;
+ int i = 0, idx = 0;
+ dma_addr_t offset = 0;
/* setting the transfer type as SGL */
cmd->flags = NVME_CMD_SGL_METABUF;
- if (entries <= (256 / sizeof(struct nvme_sgl_desc))) {
+ if (iod->nr_dmas <= 1)
+ /* We can use the inline PRP/SG list */
+ goto set_addr;
+
+ rq_for_each_bvec(bv, req, iter) {
+ dma_addr_t addr;
+
+ if (iod->map)
+ offset = 0;
+
+ while (offset < bv.bv_len) {
+ if (iod->map)
+ addr = iod->map[i].addr;
+ else
+ addr = iod->dma.addr;
+
+ nvme_pci_sgl_set_data(&sg_list[idx], addr + offset,
+ bv.bv_len);
+ offset += NVME_CTRL_PAGE_SIZE;
+ idx++;
+ }
+ i++;
+ }
+
+set_addr:
+ nvme_pci_sgl_set_seg(&cmd->dptr.sgl, iod->first_dma,
+ blk_rq_nr_phys_segments(req));
+ return BLK_STS_OK;
+}
+
+static void nvme_pci_free_pool(struct nvme_dev *dev, struct request *req)
+{
+ struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
+
+ if (iod->nr_allocations == 0)
+ dma_pool_free(dev->prp_small_pool, iod->list[0].sg_list,
+ iod->first_dma);
+ else if (iod->nr_allocations == 1)
+ dma_pool_free(dev->prp_page_pool, iod->list[0].sg_list,
+ iod->first_dma);
+ else
+ nvme_free_prps(dev, req);
+}
+
+static blk_status_t nvme_pci_setup_pool(struct nvme_dev *dev,
+ struct request *req, bool is_sgl)
+{
+ struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
+ struct dma_pool *pool;
+ size_t entry_sz;
+ dma_addr_t addr;
+ u8 entries;
+ void *list;
+
+ if (iod->nr_dmas <= 2)
+ /* Do nothing, we can use the inline PRP/SG list */
+ return BLK_STS_OK;
+
+ /* First DMA address goes to prp1 anyway */
+ entries = iod->nr_dmas - 1;
+ entry_sz = (is_sgl) ? sizeof(struct nvme_sgl_desc) : sizeof(__le64);
+ if (entries <= (256 / entry_sz)) {
pool = dev->prp_small_pool;
iod->nr_allocations = 0;
} else {
@@ -698,21 +728,20 @@ static blk_status_t nvme_pci_setup_sgls(struct nvme_dev *dev,
iod->nr_allocations = 1;
}
- sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &sgl_dma);
- if (!sg_list) {
+ /* TBD: allocate mulitple pools and chain them */
+ WARN_ON(entries > 512);
+
+ list = dma_pool_alloc(pool, GFP_ATOMIC, &addr);
+ if (!list) {
iod->nr_allocations = -1;
return BLK_STS_RESOURCE;
}
- iod->list[0].sg_list = sg_list;
- iod->first_dma = sgl_dma;
-
- nvme_pci_sgl_set_seg(&cmd->dptr.sgl, sgl_dma, entries);
- do {
- nvme_pci_sgl_set_data(&sg_list[i++], sg);
- sg = sg_next(sg);
- } while (--entries > 0);
-
+ if (is_sgl)
+ iod->list[0].sg_list = list;
+ else
+ iod->list[0].prp_list = list;
+ iod->first_dma = addr;
return BLK_STS_OK;
}
@@ -721,36 +750,84 @@ static blk_status_t nvme_map_data(struct nvme_dev *dev, struct request *req,
{
struct nvme_iod *iod = blk_mq_rq_to_pdu(req);
blk_status_t ret = BLK_STS_RESOURCE;
- int rc;
-
- iod->sgt.sgl = mempool_alloc(dev->iod_mempool, GFP_ATOMIC);
- if (!iod->sgt.sgl)
+ unsigned short n_segments = blk_rq_nr_phys_segments(req);
+ struct bio_vec bv = req_bvec(req);
+ struct req_iterator iter;
+ dma_addr_t dma_addr;
+ int rc, cnt = 0;
+ bool is_sgl;
+
+ dma_init_iova_state(&iod->state, dev->dev, rq_dma_dir(req));
+ dma_set_iova_state(&iod->state, bv.bv_page, bv.bv_len);
+
+ rc = dma_start_range(&iod->state);
+ if (rc)
return BLK_STS_RESOURCE;
- sg_init_table(iod->sgt.sgl, blk_rq_nr_phys_segments(req));
- iod->sgt.orig_nents = blk_rq_map_sg(req->q, req, iod->sgt.sgl);
- if (!iod->sgt.orig_nents)
- goto out_free_sg;
- rc = dma_map_sgtable(dev->dev, &iod->sgt, rq_dma_dir(req),
- DMA_ATTR_NO_WARN);
- if (rc) {
- if (rc == -EREMOTEIO)
- ret = BLK_STS_TARGET;
- goto out_free_sg;
+ iod->dma.len = 0;
+ iod->dma.addr = 0;
+
+ if (dma_can_use_iova(&iod->state)) {
+ iod->map = NULL;
+ rc = dma_alloc_iova_unaligned(&iod->state, bvec_phys(&bv),
+ blk_rq_payload_bytes(req));
+ if (rc)
+ return BLK_STS_RESOURCE;
+
+ rq_for_each_bvec(bv, req, iter) {
+ dma_addr = dma_link_range(&iod->state, bvec_phys(&bv),
+ bv.bv_len);
+ if (dma_mapping_error(dev->dev, dma_addr))
+ goto out_free;
+
+ if (!iod->dma.addr)
+ iod->dma.addr = dma_addr;
+ }
+ WARN_ON(blk_rq_payload_bytes(req) != iod->state.range_size);
+ } else {
+ iod->map = kmalloc_array(n_segments, sizeof(*iod->map),
+ GFP_ATOMIC);
+ if (!iod->map)
+ return BLK_STS_RESOURCE;
+
+ rq_for_each_bvec(bv, req, iter) {
+ dma_addr = dma_map_bvec(dev->dev, &bv, rq_dma_dir(req), 0);
+ if (dma_mapping_error(dev->dev, dma_addr))
+ goto out_free;
+
+ iod->map[cnt].addr = dma_addr;
+ iod->map[cnt].len = bv.bv_len;
+ cnt++;
+ }
}
+ dma_end_range(&iod->state);
- if (nvme_pci_use_sgls(dev, req, iod->sgt.nents))
+ is_sgl = nvme_pci_use_sgls(dev, req, n_segments);
+ ret = nvme_pci_setup_pool(dev, req, is_sgl);
+ if (ret != BLK_STS_OK)
+ goto out_free;
+
+ if (is_sgl)
ret = nvme_pci_setup_sgls(dev, req, &cmnd->rw);
else
ret = nvme_pci_setup_prps(dev, req, &cmnd->rw);
if (ret != BLK_STS_OK)
- goto out_unmap_sg;
+ goto out_free_pool;
+
return BLK_STS_OK;
-out_unmap_sg:
- dma_unmap_sgtable(dev->dev, &iod->sgt, rq_dma_dir(req), 0);
-out_free_sg:
- mempool_free(iod->sgt.sgl, dev->iod_mempool);
+out_free_pool:
+ nvme_pci_free_pool(dev, req);
+out_free:
+ if (iod->map) {
+ while (cnt--)
+ dma_unmap_page(dev->dev, iod->map[cnt].addr,
+ iod->map[cnt].len, rq_dma_dir(req));
+ kfree(iod->map);
+ } else {
+ dma_unlink_range(&iod->state);
+ dma_free_iova(&iod->state);
+ }
return ret;
}
@@ -791,7 +868,6 @@ static blk_status_t nvme_prep_rq(struct nvme_dev *dev, struct request *req)
iod->aborted = false;
iod->nr_allocations = -1;
- iod->sgt.nents = 0;
ret = nvme_setup_cmd(req->q->queuedata, req);
if (ret)
--
2.46.0
More information about the Linux-nvme
mailing list