[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <1234463775.1863.134.camel@lb-tlvb-eliezer>
Date: Thu, 12 Feb 2009 20:36:15 +0200
From: "Eilon Greenstein" <eilong@...adcom.com>
To: "David Miller" <davem@...emloft.net>
cc: "netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: [PATCH 03/41]bnx2x: MSI support
Subject: [PATCH 03/41]bnx2x: MSI support
Enabling MSI on top of MSI-X and INTA. Also changing the module parameter to
allow choosing INTA or MSI even when MSI-X is available. The default status
block should not be reversed for endianity. Since MSI can issue
re-configuration, the interrupt disable function now requires mmiowb
Signed-off-by: Eilon Greenstein <eilong@...adcom.com>
---
drivers/net/bnx2x.h | 5 +-
drivers/net/bnx2x_hsi.h | 6 --
drivers/net/bnx2x_main.c | 155 ++++++++++++++++++++++++++++++++--------------
drivers/net/bnx2x_reg.h | 20 ++++++
4 files changed, 132 insertions(+), 54 deletions(-)
diff --git a/drivers/net/bnx2x.h b/drivers/net/bnx2x.h
index 7edad1f..153cc95 100644
--- a/drivers/net/bnx2x.h
+++ b/drivers/net/bnx2x.h
@@ -740,6 +740,9 @@ struct bnx2x {
atomic_t intr_sem;
struct msix_entry msix_table[MAX_CONTEXT+1];
+#define INT_MODE_INTx 1
+#define INT_MODE_MSI 2
+#define INT_MODE_MSIX 3
int tx_ring_size;
@@ -794,7 +797,7 @@ struct bnx2x {
#define NO_WOL_FLAG 8
#define USING_DAC_FLAG 0x10
#define USING_MSIX_FLAG 0x20
-#define ASF_ENABLE_FLAG 0x40
+#define USING_MSI_FLAG 0x40
#define TPA_ENABLE_FLAG 0x80
#define NO_MCP_FLAG 0x100
#define BP_NOMCP(bp) (bp->flags & NO_MCP_FLAG)
diff --git a/drivers/net/bnx2x_hsi.h b/drivers/net/bnx2x_hsi.h
index c416823..25751a2 100644
--- a/drivers/net/bnx2x_hsi.h
+++ b/drivers/net/bnx2x_hsi.h
@@ -1224,15 +1224,9 @@ struct host_func_stats {
struct atten_def_status_block {
u32 attn_bits;
u32 attn_bits_ack;
-#if defined(__BIG_ENDIAN)
- u16 attn_bits_index;
- u8 reserved0;
- u8 status_block_id;
-#elif defined(__LITTLE_ENDIAN)
u8 status_block_id;
u8 reserved0;
u16 attn_bits_index;
-#endif
u32 reserved1;
};
diff --git a/drivers/net/bnx2x_main.c b/drivers/net/bnx2x_main.c
index 60762f7..db8506d 100644
--- a/drivers/net/bnx2x_main.c
+++ b/drivers/net/bnx2x_main.c
@@ -77,17 +77,19 @@ static int multi_mode = 1;
module_param(multi_mode, int, 0);
static int disable_tpa;
-static int use_inta;
static int poll;
static int debug;
static int load_count[3]; /* 0-common, 1-port0, 2-port1 */
module_param(disable_tpa, int, 0);
-module_param(use_inta, int, 0);
+
+static int int_mode;
+module_param(int_mode, int, 0);
+MODULE_PARM_DESC(int_mode, " Force interrupt mode (1 INT#x; 2 MSI)");
+
module_param(poll, int, 0);
module_param(debug, int, 0);
MODULE_PARM_DESC(disable_tpa, "disable the TPA (LRO) feature");
-MODULE_PARM_DESC(use_inta, "use INT#A instead of MSI-X");
MODULE_PARM_DESC(poll, "use polling (for debug)");
MODULE_PARM_DESC(debug, "default debug msglevel");
@@ -588,34 +590,41 @@ static void bnx2x_int_enable(struct bnx2x *bp)
u32 addr = port ? HC_REG_CONFIG_1 : HC_REG_CONFIG_0;
u32 val = REG_RD(bp, addr);
int msix = (bp->flags & USING_MSIX_FLAG) ? 1 : 0;
+ int msi = (bp->flags & USING_MSI_FLAG) ? 1 : 0;
if (msix) {
- val &= ~HC_CONFIG_0_REG_SINGLE_ISR_EN_0;
+ val &= ~(HC_CONFIG_0_REG_SINGLE_ISR_EN_0 |
+ HC_CONFIG_0_REG_INT_LINE_EN_0);
val |= (HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0 |
HC_CONFIG_0_REG_ATTN_BIT_EN_0);
+ } else if (msi) {
+ val &= ~HC_CONFIG_0_REG_INT_LINE_EN_0;
+ val |= (HC_CONFIG_0_REG_SINGLE_ISR_EN_0 |
+ HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0 |
+ HC_CONFIG_0_REG_ATTN_BIT_EN_0);
} else {
val |= (HC_CONFIG_0_REG_SINGLE_ISR_EN_0 |
HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0 |
HC_CONFIG_0_REG_INT_LINE_EN_0 |
HC_CONFIG_0_REG_ATTN_BIT_EN_0);
- DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x) MSI-X %d\n",
- val, port, addr, msix);
+ DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x)\n",
+ val, port, addr);
REG_WR(bp, addr, val);
val &= ~HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0;
}
- DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x) MSI-X %d\n",
- val, port, addr, msix);
+ DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x) mode %s\n",
+ val, port, addr, (msix ? "MSI-X" : (msi ? "MSI" : "INTx")));
REG_WR(bp, addr, val);
if (CHIP_IS_E1H(bp)) {
/* init leading/trailing edge */
if (IS_E1HMF(bp)) {
- val = (0xfe0f | (1 << (BP_E1HVN(bp) + 4)));
+ val = (0xee0f | (1 << (BP_E1HVN(bp) + 4)));
if (bp->port.pmf)
/* enable nig attention */
val |= 0x0100;
@@ -641,6 +650,9 @@ static void bnx2x_int_disable(struct bnx2x *bp)
DP(NETIF_MSG_INTR, "write %x to HC %d (addr 0x%x)\n",
val, port, addr);
+ /* flush all outstanding writes */
+ mmiowb();
+
REG_WR(bp, addr, val);
if (REG_RD(bp, addr) != val)
BNX2X_ERR("BUG! proper val not read from IGU!\n");
@@ -649,7 +661,7 @@ static void bnx2x_int_disable(struct bnx2x *bp)
static void bnx2x_int_disable_sync(struct bnx2x *bp, int disable_hw)
{
int msix = (bp->flags & USING_MSIX_FLAG) ? 1 : 0;
- int i;
+ int i, offset;
/* disable interrupt handling */
atomic_inc(&bp->intr_sem);
@@ -659,11 +671,10 @@ static void bnx2x_int_disable_sync(struct bnx2x *bp, int disable_hw)
/* make sure all ISRs are done */
if (msix) {
+ synchronize_irq(bp->msix_table[0].vector);
+ offset = 1;
for_each_queue(bp, i)
- synchronize_irq(bp->msix_table[i].vector);
-
- /* one more for the Slow Path IRQ */
- synchronize_irq(bp->msix_table[i].vector);
+ synchronize_irq(bp->msix_table[i + offset].vector);
} else
synchronize_irq(bp->pdev->irq);
@@ -5198,6 +5209,8 @@ static int bnx2x_init_common(struct bnx2x *bp)
REG_WR(bp, PXP2_REG_RQ_SRC_ENDIAN_M, 1);
REG_WR(bp, PXP2_REG_RQ_CDU_ENDIAN_M, 1);
REG_WR(bp, PXP2_REG_RQ_DBG_ENDIAN_M, 1);
+ /* make sure this value is 0 */
+ REG_WR(bp, PXP2_REG_RQ_HC_ENDIAN_M, 0);
/* REG_WR(bp, PXP2_REG_RD_PBF_SWAP_MODE, 1); */
REG_WR(bp, PXP2_REG_RD_QM_SWAP_MODE, 1);
@@ -5648,10 +5661,17 @@ static int bnx2x_init_func(struct bnx2x *bp)
{
int port = BP_PORT(bp);
int func = BP_FUNC(bp);
+ u32 addr, val;
int i;
DP(BNX2X_MSG_MCP, "starting func init func %x\n", func);
+ /* set MSI reconfigure capability */
+ addr = (port ? HC_REG_CONFIG_1 : HC_REG_CONFIG_0);
+ val = REG_RD(bp, addr);
+ val |= HC_CONFIG_0_REG_MSI_ATTN_EN_0;
+ REG_WR(bp, addr, val);
+
i = FUNC_ILT_BASE(func);
bnx2x_ilt_wr(bp, i, bnx2x_sp_mapping(bp, context));
@@ -6053,10 +6073,6 @@ static void bnx2x_free_msix_irqs(struct bnx2x *bp)
"state %x\n", i, bp->msix_table[i + offset].vector,
bnx2x_fp(bp, i, state));
- if (bnx2x_fp(bp, i, state) != BNX2X_FP_STATE_CLOSED)
- BNX2X_ERR("IRQ of fp #%d being freed while "
- "state != closed\n", i);
-
free_irq(bp->msix_table[i + offset].vector, &bp->fp[i]);
}
}
@@ -6068,21 +6084,25 @@ static void bnx2x_free_irq(struct bnx2x *bp)
pci_disable_msix(bp->pdev);
bp->flags &= ~USING_MSIX_FLAG;
+ } else if (bp->flags & USING_MSI_FLAG) {
+ free_irq(bp->pdev->irq, bp->dev);
+ pci_disable_msi(bp->pdev);
+ bp->flags &= ~USING_MSI_FLAG;
+
} else
free_irq(bp->pdev->irq, bp->dev);
}
static int bnx2x_enable_msix(struct bnx2x *bp)
{
- int i, rc, offset;
+ int i, rc, offset = 1;
+ int igu_vec = 0;
- bp->msix_table[0].entry = 0;
- offset = 1;
- DP(NETIF_MSG_IFUP, "msix_table[0].entry = 0 (slowpath)\n");
+ bp->msix_table[0].entry = igu_vec;
+ DP(NETIF_MSG_IFUP, "msix_table[0].entry = %d (slowpath)\n", igu_vec);
for_each_queue(bp, i) {
- int igu_vec = offset + i + BP_L_ID(bp);
-
+ igu_vec = BP_L_ID(bp) + offset + i;
bp->msix_table[i + offset].entry = igu_vec;
DP(NETIF_MSG_IFUP, "msix_table[%d].entry = %d "
"(fastpath #%u)\n", i + offset, igu_vec, i);
@@ -6091,9 +6111,10 @@ static int bnx2x_enable_msix(struct bnx2x *bp)
rc = pci_enable_msix(bp->pdev, &bp->msix_table[0],
BNX2X_NUM_QUEUES(bp) + offset);
if (rc) {
- DP(NETIF_MSG_IFUP, "MSI-X is not attainable\n");
- return -1;
+ DP(NETIF_MSG_IFUP, "MSI-X is not attainable rc %d\n", rc);
+ return rc;
}
+
bp->flags |= USING_MSIX_FLAG;
return 0;
@@ -6140,11 +6161,31 @@ static int bnx2x_req_msix_irqs(struct bnx2x *bp)
return 0;
}
+static int bnx2x_enable_msi(struct bnx2x *bp)
+{
+ int rc;
+
+ rc = pci_enable_msi(bp->pdev);
+ if (rc) {
+ DP(NETIF_MSG_IFUP, "MSI is not attainable\n");
+ return -1;
+ }
+ bp->flags |= USING_MSI_FLAG;
+
+ return 0;
+}
+
static int bnx2x_req_irq(struct bnx2x *bp)
{
+ unsigned long flags;
int rc;
- rc = request_irq(bp->pdev->irq, bnx2x_interrupt, IRQF_SHARED,
+ if (bp->flags & USING_MSI_FLAG)
+ flags = 0;
+ else
+ flags = IRQF_SHARED;
+
+ rc = request_irq(bp->pdev->irq, bnx2x_interrupt, flags,
bp->dev->name, bp->dev);
if (!rc)
bnx2x_fp(bp, 0, state) = BNX2X_FP_STATE_IRQ;
@@ -6365,28 +6406,23 @@ static int bnx2x_setup_multi(struct bnx2x *bp, int index)
}
static int bnx2x_poll(struct napi_struct *napi, int budget);
-static void bnx2x_set_rx_mode(struct net_device *dev);
-/* must be called with rtnl_lock */
-static int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
+static void bnx2x_set_int_mode(struct bnx2x *bp)
{
- u32 load_code;
- int i, rc = 0;
int num_queues;
-#ifdef BNX2X_STOP_ON_ERROR
- if (unlikely(bp->panic))
- return -EPERM;
-#endif
-
- bp->state = BNX2X_STATE_OPENING_WAIT4_LOAD;
- if (use_inta) {
+ switch (int_mode) {
+ case INT_MODE_INTx:
+ case INT_MODE_MSI:
num_queues = 1;
bp->num_rx_queues = num_queues;
bp->num_tx_queues = num_queues;
DP(NETIF_MSG_IFUP,
"set number of queues to %d\n", num_queues);
- } else {
+ break;
+
+ case INT_MODE_MSIX:
+ default:
if (bp->multi_mode == ETH_RSS_MODE_REGULAR)
num_queues = min_t(u32, num_online_cpus(),
BNX2X_MAX_QUEUES(bp));
@@ -6401,8 +6437,7 @@ static int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
* so try to enable MSI-X with the requested number of fp's
* and fallback to MSI or legacy INTx with one fp
*/
- rc = bnx2x_enable_msix(bp);
- if (rc) {
+ if (bnx2x_enable_msix(bp)) {
/* failed to enable MSI-X */
num_queues = 1;
bp->num_rx_queues = num_queues;
@@ -6412,8 +6447,27 @@ static int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
"enable MSI-X set number of "
"queues to %d\n", num_queues);
}
+ break;
}
bp->dev->real_num_tx_queues = bp->num_tx_queues;
+}
+
+static void bnx2x_set_rx_mode(struct net_device *dev);
+
+/* must be called with rtnl_lock */
+static int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
+{
+ u32 load_code;
+ int i, rc = 0;
+#ifdef BNX2X_STOP_ON_ERROR
+ DP(NETIF_MSG_IFUP, "enter load_mode %d\n", load_mode);
+ if (unlikely(bp->panic))
+ return -EPERM;
+#endif
+
+ bp->state = BNX2X_STATE_OPENING_WAIT4_LOAD;
+
+ bnx2x_set_int_mode(bp);
if (bnx2x_alloc_mem(bp))
return -ENOMEM;
@@ -6445,14 +6499,22 @@ static int bnx2x_nic_load(struct bnx2x *bp, int load_mode)
pci_disable_msix(bp->pdev);
goto load_error1;
}
- printk(KERN_INFO PFX "%s: using MSI-X\n", bp->dev->name);
} else {
+ if ((rc != -ENOMEM) && (int_mode != INT_MODE_INTx))
+ bnx2x_enable_msi(bp);
bnx2x_ack_int(bp);
rc = bnx2x_req_irq(bp);
if (rc) {
BNX2X_ERR("IRQ request failed rc %d, aborting\n", rc);
+ if (bp->flags & USING_MSI_FLAG)
+ pci_disable_msi(bp->pdev);
goto load_error1;
}
+ if (bp->flags & USING_MSI_FLAG) {
+ bp->dev->irq = bp->pdev->irq;
+ printk(KERN_INFO PFX "%s: using MSI IRQ %d\n",
+ bp->dev->name, bp->pdev->irq);
+ }
}
/* Send LOAD_REQUEST command to MCP
@@ -6689,8 +6751,6 @@ static void bnx2x_reset_func(struct bnx2x *bp)
REG_WR(bp, HC_REG_LEADING_EDGE_0 + port*8, 0);
REG_WR(bp, HC_REG_TRAILING_EDGE_0 + port*8, 0);
- REG_WR(bp, HC_REG_CONFIG_0 + port*4, 0x1000);
-
/* Clear ILT */
base = FUNC_ILT_BASE(func);
for (i = base; i < base + ILT_PER_FUNC; i++)
@@ -7636,9 +7696,10 @@ static int __devinit bnx2x_init_bp(struct bnx2x *bp)
"MCP disabled, must load devices in order!\n");
/* Set multi queue mode */
- if ((multi_mode != ETH_RSS_MODE_DISABLED) && (!use_inta)) {
+ if ((multi_mode != ETH_RSS_MODE_DISABLED) &&
+ ((int_mode == INT_MODE_INTx) || (int_mode == INT_MODE_MSI))) {
printk(KERN_ERR PFX
- "Multi disabled since INTA is requested\n");
+ "Multi disabled since int_mode requested is not MSI-X\n");
multi_mode = ETH_RSS_MODE_DISABLED;
}
bp->multi_mode = multi_mode;
diff --git a/drivers/net/bnx2x_reg.h b/drivers/net/bnx2x_reg.h
index fc957fa..0be77c5 100644
--- a/drivers/net/bnx2x_reg.h
+++ b/drivers/net/bnx2x_reg.h
@@ -745,6 +745,7 @@
#define DORQ_REG_SHRT_CMHEAD 0x170054
#define HC_CONFIG_0_REG_ATTN_BIT_EN_0 (0x1<<4)
#define HC_CONFIG_0_REG_INT_LINE_EN_0 (0x1<<3)
+#define HC_CONFIG_0_REG_MSI_ATTN_EN_0 (0x1<<7)
#define HC_CONFIG_0_REG_MSI_MSIX_INT_EN_0 (0x1<<2)
#define HC_CONFIG_0_REG_SINGLE_ISR_EN_0 (0x1<<1)
#define HC_REG_AGG_INT_0 0x108050
@@ -5359,9 +5360,28 @@
#define PCICFG_PM_CSR_STATE (0x3<<0)
#define PCICFG_PM_CSR_PME_ENABLE (1<<8)
#define PCICFG_PM_CSR_PME_STATUS (1<<15)
+#define PCICFG_MSI_CAP_ID 0x58
+#define PCICFG_MSI_CONTROL_ENABLE (0x1<<16)
+#define PCICFG_MSI_CONTROL_MCAP (0x7<<17)
+#define PCICFG_MSI_CONTROL_MENA (0x7<<20)
+#define PCICFG_MSI_CONTROL_64_BIT_ADDR_CAP (0x1<<23)
+#define PCICFG_MSI_CONTROL_MSI_PVMASK_CAPABLE (0x1<<24)
#define PCICFG_GRC_ADDRESS 0x78
#define PCICFG_GRC_DATA 0x80
+#define PCICFG_MSIX_CAP_ID 0xa0
+#define PCICFG_MSIX_CONTROL_TABLE_SIZE (0x7ff<<16)
+#define PCICFG_MSIX_CONTROL_RESERVED (0x7<<27)
+#define PCICFG_MSIX_CONTROL_FUNC_MASK (0x1<<30)
+#define PCICFG_MSIX_CONTROL_MSIX_ENABLE (0x1<<31)
+
#define PCICFG_DEVICE_CONTROL 0xb4
+#define PCICFG_DEVICE_STATUS 0xb6
+#define PCICFG_DEVICE_STATUS_CORR_ERR_DET (1<<0)
+#define PCICFG_DEVICE_STATUS_NON_FATAL_ERR_DET (1<<1)
+#define PCICFG_DEVICE_STATUS_FATAL_ERR_DET (1<<2)
+#define PCICFG_DEVICE_STATUS_UNSUP_REQ_DET (1<<3)
+#define PCICFG_DEVICE_STATUS_AUX_PWR_DET (1<<4)
+#define PCICFG_DEVICE_STATUS_NO_PEND (1<<5)
#define PCICFG_LINK_CONTROL 0xbc
--
1.5.4.3
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists