Instead of relying on the vendor specific implementations to send the
PME_Turn_Off message, let's introduce a generic way of sending the message
using the MSG TLP.
This is achieved by reserving a region for MSG TLP of size
'pci->region_align', at the end of the first IORESOURCE_MEM window of the
host bridge. And then sending the PME_Turn_Off message during system
suspend with the help of iATU.
The reason for reserving the MSG TLP region at the end of the
IORESOURCE_MEM is to avoid generating holes in between. Because, when the
region is allocated using allocate_resource(), memory will be allocated
from the start of the window. Later, if memory gets allocated for an
endpoint of size bigger than 'region_align', there will be a hole between
MSG TLP region and endpoint memory.
It should be noted that this generic implementation is optional for the
glue drivers and can be overridden by a custom 'pme_turn_off' callback.
Signed-off-by: Frank Li <Frank.Li@xxxxxxx>
---
drivers/pci/controller/dwc/pcie-designware-host.c | 102 ++++++++++++++++++++--
drivers/pci/controller/dwc/pcie-designware.h | 3 +
2 files changed, 100 insertions(+), 5 deletions(-)
diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c b/drivers/pci/controller/dwc/pcie-designware-host.c
index 3a9cb4be22ab2..f57805b21ef53 100644
--- a/drivers/pci/controller/dwc/pcie-designware-host.c
+++ b/drivers/pci/controller/dwc/pcie-designware-host.c
@@ -398,6 +398,34 @@ static int dw_pcie_msi_host_init(struct dw_pcie_rp *pp)
return 0;
}
+static void dw_pcie_host_request_msg_tlp_res(struct dw_pcie_rp *pp)
+{
+ struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+ struct resource_entry *win;
+ struct resource *res;
+
+ win = resource_list_first_type(&pp->bridge->windows, IORESOURCE_MEM);
+ if (win) {
+ res = devm_kzalloc(pci->dev, sizeof(*res), GFP_KERNEL);
+ if (!res)
+ return;
+
+ /*
+ * Allocate MSG TLP region of size 'region_align' at the end of
+ * the host bridge window.
+ */
+ res->start = win->res->end - pci->region_align + 1;
+ res->end = win->res->end;
+ res->name = "msg";
+ res->flags = win->res->flags | IORESOURCE_BUSY;
+
+ if (!request_resource(win->res, res))
+ pp->msg_res = res;
+ else
+ devm_kfree(pci->dev, res);
+ }
+}
+
int dw_pcie_host_init(struct dw_pcie_rp *pp)
{
struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
@@ -484,6 +512,18 @@ int dw_pcie_host_init(struct dw_pcie_rp *pp)
dw_pcie_iatu_detect(pci);
+ /*
+ * Allocate the resource for MSG TLP before programming the iATU
+ * outbound window in dw_pcie_setup_rc(). Since the allocation depends
+ * on the value of 'region_align', this has to be done after
+ * dw_pcie_iatu_detect().
+ *
+ * Glue drivers need to set 'use_atu_msg' before dw_pcie_host_init() to
+ * make use of the generic MSG TLP implementation.
+ */
+ if (pp->use_atu_msg)
+ dw_pcie_host_request_msg_tlp_res(pp);
+
ret = dw_pcie_edma_detect(pci);
if (ret)
goto err_free_msi;
@@ -541,6 +581,11 @@ void dw_pcie_host_deinit(struct dw_pcie_rp *pp)
dw_pcie_edma_remove(pci);
+ if (pp->msg_res) {
+ release_resource(pp->msg_res);
+ devm_kfree(pci->dev, pp->msg_res);
+ }
+
if (pp->has_msi_ctrl)
dw_pcie_free_msi(pp);
@@ -700,7 +745,13 @@ static int dw_pcie_iatu_setup(struct dw_pcie_rp *pp)
atu.type = PCIE_ATU_TYPE_MEM;
atu.cpu_addr = entry->res->start;
atu.pci_addr = entry->res->start - entry->offset;
- atu.size = resource_size(entry->res);
+
+ /* Adjust iATU size if MSG TLP region was allocated before */
+ if (pp->msg_res && pp->msg_res->parent == entry->res)
+ atu.size = resource_size(entry->res) -
+ resource_size(pp->msg_res);
+ else
+ atu.size = resource_size(entry->res);
ret = dw_pcie_prog_outbound_atu(pci, &atu);
if (ret) {
@@ -733,6 +784,8 @@ static int dw_pcie_iatu_setup(struct dw_pcie_rp *pp)
dev_warn(pci->dev, "Ranges exceed outbound iATU size (%d)\n",
pci->num_ob_windows);
+ pp->msg_atu_index = i;
+
i = 0;
resource_list_for_each_entry(entry, &pp->bridge->dma_ranges) {
if (resource_type(entry->res) != IORESOURCE_MEM)
@@ -838,11 +891,47 @@ int dw_pcie_setup_rc(struct dw_pcie_rp *pp)
}
EXPORT_SYMBOL_GPL(dw_pcie_setup_rc);
+static int dw_pcie_pme_turn_off(struct dw_pcie *pci)
+{
+ struct dw_pcie_ob_atu_cfg atu = { 0 };
+ void __iomem *mem;
+ int ret;
+
+ if (pci->num_ob_windows <= pci->pp.msg_atu_index)
+ return -ENOSPC;
+
+ if (!pci->pp.msg_res)
+ return -ENOSPC;
+
+ atu.code = PCIE_MSG_CODE_PME_TURN_OFF;
+ atu.routing = PCIE_MSG_TYPE_R_BC;
+ atu.type = PCIE_ATU_TYPE_MSG;
+ atu.size = resource_size(pci->pp.msg_res);
+ atu.index = pci->pp.msg_atu_index;
+
+ atu.cpu_addr = pci->pp.msg_res->start;
+
+ ret = dw_pcie_prog_outbound_atu(pci, &atu);
+ if (ret)
+ return ret;
+
+ mem = ioremap(atu.cpu_addr, pci->region_align);
+ if (!mem)
+ return -ENOMEM;
+
+ /* A dummy write is converted to a Msg TLP */
+ writel(0, mem);
+
+ iounmap(mem);
+
+ return 0;
+}
+
int dw_pcie_suspend_noirq(struct dw_pcie *pci)
{
u8 offset = dw_pcie_find_capability(pci, PCI_CAP_ID_EXP);
u32 val;
- int ret;
+ int ret = 0;
/*
* If L1SS is supported, then do not put the link into L2 as some
@@ -854,10 +943,13 @@ int dw_pcie_suspend_noirq(struct dw_pcie *pci)
if (dw_pcie_get_ltssm(pci) <= DW_PCIE_LTSSM_DETECT_ACT)
return 0;
- if (!pci->pp.ops->pme_turn_off)
- return 0;
+ if (pci->pp.ops->pme_turn_off)
+ pci->pp.ops->pme_turn_off(&pci->pp);
+ else
+ ret = dw_pcie_pme_turn_off(pci);
- pci->pp.ops->pme_turn_off(&pci->pp);
+ if (ret)
+ return ret;
ret = read_poll_timeout(dw_pcie_get_ltssm, val, val == DW_PCIE_LTSSM_L2_IDLE,
PCIE_PME_TO_L2_TIMEOUT_US/10,
diff --git a/drivers/pci/controller/dwc/pcie-designware.h b/drivers/pci/controller/dwc/pcie-designware.h
index 703b50bc5e0f1..dca5de4c6e877 100644
--- a/drivers/pci/controller/dwc/pcie-designware.h
+++ b/drivers/pci/controller/dwc/pcie-designware.h
@@ -341,6 +341,9 @@ struct dw_pcie_rp {
struct pci_host_bridge *bridge;
raw_spinlock_t lock;
DECLARE_BITMAP(msi_irq_in_use, MAX_MSI_IRQS);
+ bool use_atu_msg;
+ int msg_atu_index;
+ struct resource *msg_res;
};
struct dw_pcie_ep_ops {
--
2.34.1
