Add MSI doorbell support to reduce latency between PCI host and EP.
Before this change:
ping 169.254.172.137
64 bytes from 169.254.172.137: icmp_seq=1 ttl=64 time=0.575 ms
64 bytes from 169.254.172.137: icmp_seq=2 ttl=64 time=1.80 ms
64 bytes from 169.254.172.137: icmp_seq=3 ttl=64 time=8.19 ms
64 bytes from 169.254.172.137: icmp_seq=4 ttl=64 time=2.00 ms
After this change:
ping 169.254.144.71
64 bytes from 169.254.144.71: icmp_seq=1 ttl=64 time=0.215 ms
64 bytes from 169.254.144.71: icmp_seq=2 ttl=64 time=0.456 ms
64 bytes from 169.254.144.71: icmp_seq=3 ttl=64 time=0.448 ms
Change u64 db to atomic_64 because difference doorbell may happen at the
same time.
Signed-off-by: Frank Li <Frank.Li@xxxxxxx>
---
drivers/pci/endpoint/functions/pci-epf-vntb.c | 153 +++++++++++++++++++++++---
1 file changed, 136 insertions(+), 17 deletions(-)
diff --git a/drivers/pci/endpoint/functions/pci-epf-vntb.c b/drivers/pci/endpoint/functions/pci-epf-vntb.c
index 83e9ab10f9c4fc2b485d5463faa2172500f12999..1c586205835fe9c7c5352e74819bccb4ece84438 100644
--- a/drivers/pci/endpoint/functions/pci-epf-vntb.c
+++ b/drivers/pci/endpoint/functions/pci-epf-vntb.c
@@ -36,11 +36,13 @@
* PCIe Root Port PCI EP
*/
+#include <linux/atomic.h>
#include <linux/delay.h>
#include <linux/io.h>
#include <linux/module.h>
#include <linux/slab.h>
+#include <linux/pci-ep-msi.h>
#include <linux/pci-epc.h>
#include <linux/pci-epf.h>
#include <linux/ntb.h>
@@ -126,12 +128,13 @@ struct epf_ntb {
u32 db_count;
u32 spad_count;
u64 mws_size[MAX_MW];
- u64 db;
+ atomic64_t db;
u32 vbus_number;
u16 vntb_pid;
u16 vntb_vid;
bool linkup;
+ bool msi_doorbell;
u32 spad_size;
enum pci_barno epf_ntb_bar[VNTB_BAR_NUM];
@@ -258,9 +261,9 @@ static void epf_ntb_cmd_handler(struct work_struct *work)
ntb = container_of(work, struct epf_ntb, cmd_handler.work);
- for (i = 1; i < ntb->db_count; i++) {
+ for (i = 1; i < ntb->db_count && !ntb->msi_doorbell; i++) {
if (ntb->epf_db[i]) {
- ntb->db |= 1 << (i - 1);
+ atomic64_or(1 << (i - 1), &ntb->db);
ntb_db_event(&ntb->ntb, i);
ntb->epf_db[i] = 0;
}
@@ -319,7 +322,24 @@ static void epf_ntb_cmd_handler(struct work_struct *work)
reset_handler:
queue_delayed_work(kpcintb_workqueue, &ntb->cmd_handler,
- msecs_to_jiffies(5));
+ ntb->msi_doorbell ? msecs_to_jiffies(500) : msecs_to_jiffies(5));
+}
+
+static irqreturn_t epf_ntb_doorbell_handler(int irq, void *data)
+{
+ struct epf_ntb *ntb = data;
+ int i = 0;
+
+ for (i = 1; i < ntb->db_count; i++)
+ if (irq == ntb->epf->db_msg[i].virq) {
+ atomic64_or(1 << (i - 1), &ntb->db);
+ ntb_db_event(&ntb->ntb, i);
+ }
+
+ if (irq == ntb->epf->db_msg[0].virq)
+ queue_delayed_work(kpcintb_workqueue, &ntb->cmd_handler, 0);
+
+ return IRQ_HANDLED;
}
/**
@@ -500,6 +520,90 @@ static int epf_ntb_configure_interrupt(struct epf_ntb *ntb)
return 0;
}
+static int epf_ntb_db_bar_init_msi_doorbell(struct epf_ntb *ntb,
+ struct pci_epf_bar *db_bar,
+ const struct pci_epc_features *epc_features,
+ enum pci_barno barno)
+{
+ struct pci_epf *epf = ntb->epf;
+ dma_addr_t low, high;
+ struct msi_msg *msg;
+ size_t sz;
+ int ret;
+ int i;
+
+ ret = pci_epf_alloc_doorbell(epf, ntb->db_count);
+ if (ret)
+ return ret;
+
+ for (i = 0; i < ntb->db_count; i++) {
+ ret = request_irq(epf->db_msg[i].virq, epf_ntb_doorbell_handler,
+ 0, "vntb_db", ntb);
+
+ if (ret) {
+ dev_err(&epf->dev,
+ "Failed to request doorbell IRQ: %d\n",
+ epf->db_msg[i].virq);
+ goto err_request_irq;
+ }
+ }
+
+ msg = &epf->db_msg[0].msg;
+
+ high = 0;
+ low = (u64)msg->address_hi << 32 | msg->address_lo;
+
+ for (i = 0; i < ntb->db_count; i++) {
+ struct msi_msg *msg = &epf->db_msg[i].msg;
+ dma_addr_t addr = (u64)msg->address_hi << 32 | msg->address_lo;
+
+ low = min(low, addr);
+ high = max(high, addr);
+ }
+
+ sz = high - low + sizeof(u32);
+
+ ret = pci_epf_set_inbound_space(epf, sz, barno,
+ epc_features, 0, false, low);
+
+ ret = pci_epc_set_bar(ntb->epf->epc, ntb->epf->func_no, ntb->epf->vfunc_no, db_bar);
+ if (ret) {
+ dev_err(&epf->dev, "Doorbell BAR set failed\n");
+ goto err_request_irq;
+ }
+
+ for (i = 0; i < ntb->db_count; i++) {
+ struct msi_msg *msg = &epf->db_msg[i].msg;
+ dma_addr_t addr;
+ size_t offset;
+
+ ret = pci_epf_align_inbound_addr(epf, db_bar->barno,
+ ((u64)msg->address_hi << 32) | msg->address_lo,
+ &addr, &offset);
+
+ if (ret) {
+ ntb->msi_doorbell = false;
+ goto err_request_irq;
+ }
+
+ ntb->reg->db_data[i] = msg->data;
+ ntb->reg->db_offset[i] = offset;
+ }
+
+ ntb->reg->db_entry_size = 0;
+
+ ntb->msi_doorbell = true;
+
+ return 0;
+
+err_request_irq:
+ for (i--; i >= 0; i--)
+ free_irq(epf->db_msg[i].virq, ntb);
+
+ pci_epf_free_doorbell(ntb->epf);
+ return ret;
+}
+
/**
* epf_ntb_db_bar_init() - Configure Doorbell window BARs
* @ntb: NTB device that facilitates communication between HOST and VHOST
@@ -520,22 +624,27 @@ static int epf_ntb_db_bar_init(struct epf_ntb *ntb)
ntb->epf->func_no,
ntb->epf->vfunc_no);
barno = ntb->epf_ntb_bar[BAR_DB];
-
- mw_addr = pci_epf_alloc_space(ntb->epf, size, barno, epc_features, 0);
- if (!mw_addr) {
- dev_err(dev, "Failed to allocate OB address\n");
- return -ENOMEM;
- }
-
- ntb->epf_db = mw_addr;
-
epf_bar = &ntb->epf->bar[barno];
- ret = pci_epc_set_bar(ntb->epf->epc, ntb->epf->func_no, ntb->epf->vfunc_no, epf_bar);
+ ret = epf_ntb_db_bar_init_msi_doorbell(ntb, epf_bar, epc_features, barno);
if (ret) {
- dev_err(dev, "Doorbell BAR set failed\n");
+ /* fall back to polling mode */
+ mw_addr = pci_epf_alloc_space(ntb->epf, size, barno, epc_features, 0);
+ if (!mw_addr) {
+ dev_err(dev, "Failed to allocate OB address\n");
+ return -ENOMEM;
+ }
+
+ ntb->epf_db = mw_addr;
+
+ ret = pci_epc_set_bar(ntb->epf->epc, ntb->epf->func_no,
+ ntb->epf->vfunc_no, epf_bar);
+ if (ret) {
+ dev_err(dev, "Doorbell BAR set failed\n");
goto err_alloc_peer_mem;
+ }
}
+
return ret;
err_alloc_peer_mem:
@@ -554,6 +663,16 @@ static void epf_ntb_db_bar_clear(struct epf_ntb *ntb)
{
enum pci_barno barno;
+ if (ntb->msi_doorbell) {
+ int i;
+
+ for (i = 0; i < ntb->db_count; i++)
+ free_irq(ntb->epf->db_msg[i].virq, ntb);
+ }
+
+ if (ntb->epf->db_msg)
+ pci_epf_free_doorbell(ntb->epf);
+
barno = ntb->epf_ntb_bar[BAR_DB];
pci_epf_free_space(ntb->epf, ntb->epf_db, barno, 0);
pci_epc_clear_bar(ntb->epf->epc,
@@ -1268,7 +1387,7 @@ static u64 vntb_epf_db_read(struct ntb_dev *ndev)
{
struct epf_ntb *ntb = ntb_ndev(ndev);
- return ntb->db;
+ return atomic64_read(&ntb->db);
}
static int vntb_epf_mw_get_align(struct ntb_dev *ndev, int pidx, int idx,
@@ -1308,7 +1427,7 @@ static int vntb_epf_db_clear(struct ntb_dev *ndev, u64 db_bits)
{
struct epf_ntb *ntb = ntb_ndev(ndev);
- ntb->db &= ~db_bits;
+ atomic64_and(~db_bits, &ntb->db);
return 0;
}
--
2.34.1
