linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH AUTOSEL 4.14 02/25] qed: Wait for MCP halt and resume commands to take place
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 03/25] qed: Prevent a possible deadlock during driver load and unload Sasha Levin
                   ` (22 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: Tomer Tayar, Ariel Elior, David S . Miller, Sasha Levin

From: Tomer Tayar <Tomer.Tayar@cavium.com>

[ Upstream commit 76271809f49056f079e202bf6513d17b0d6dd34d ]

Successive iterations of halting and resuming the management chip (MCP)
might fail, since currently the driver doesn't wait for these operations to
actually take place.
This patch prevents the driver from moving forward before the operations
are reflected in the state register.

Signed-off-by: Tomer Tayar <Tomer.Tayar@cavium.com>
Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/net/ethernet/qlogic/qed/qed_mcp.c     | 46 +++++++++++++++----
 .../net/ethernet/qlogic/qed/qed_reg_addr.h    |  1 +
 2 files changed, 39 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 9348d367cfdf..5748bcb26d40 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -1998,31 +1998,61 @@ qed_mcp_send_drv_version(struct qed_hwfn *p_hwfn,
 	return rc;
 }
 
+/* A maximal 100 msec waiting time for the MCP to halt */
+#define QED_MCP_HALT_SLEEP_MS		10
+#define QED_MCP_HALT_MAX_RETRIES	10
+
 int qed_mcp_halt(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 {
-	u32 resp = 0, param = 0;
+	u32 resp = 0, param = 0, cpu_state, cnt = 0;
 	int rc;
 
 	rc = qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_MCP_HALT, 0, &resp,
 			 &param);
-	if (rc)
+	if (rc) {
 		DP_ERR(p_hwfn, "MCP response failure, aborting\n");
+		return rc;
+	}
 
-	return rc;
+	do {
+		msleep(QED_MCP_HALT_SLEEP_MS);
+		cpu_state = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_STATE);
+		if (cpu_state & MCP_REG_CPU_STATE_SOFT_HALTED)
+			break;
+	} while (++cnt < QED_MCP_HALT_MAX_RETRIES);
+
+	if (cnt == QED_MCP_HALT_MAX_RETRIES) {
+		DP_NOTICE(p_hwfn,
+			  "Failed to halt the MCP [CPU_MODE = 0x%08x, CPU_STATE = 0x%08x]\n",
+			  qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_MODE), cpu_state);
+		return -EBUSY;
+	}
+
+	return 0;
 }
 
+#define QED_MCP_RESUME_SLEEP_MS	10
+
 int qed_mcp_resume(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 {
-	u32 value, cpu_mode;
+	u32 cpu_mode, cpu_state;
 
 	qed_wr(p_hwfn, p_ptt, MCP_REG_CPU_STATE, 0xffffffff);
 
-	value = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_MODE);
-	value &= ~MCP_REG_CPU_MODE_SOFT_HALT;
-	qed_wr(p_hwfn, p_ptt, MCP_REG_CPU_MODE, value);
 	cpu_mode = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_MODE);
+	cpu_mode &= ~MCP_REG_CPU_MODE_SOFT_HALT;
+	qed_wr(p_hwfn, p_ptt, MCP_REG_CPU_MODE, cpu_mode);
+	msleep(QED_MCP_RESUME_SLEEP_MS);
+	cpu_state = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_STATE);
 
-	return (cpu_mode & MCP_REG_CPU_MODE_SOFT_HALT) ? -EAGAIN : 0;
+	if (cpu_state & MCP_REG_CPU_STATE_SOFT_HALTED) {
+		DP_NOTICE(p_hwfn,
+			  "Failed to resume the MCP [CPU_MODE = 0x%08x, CPU_STATE = 0x%08x]\n",
+			  cpu_mode, cpu_state);
+		return -EBUSY;
+	}
+
+	return 0;
 }
 
 int qed_mcp_ov_update_current_config(struct qed_hwfn *p_hwfn,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
index 0cdb4337b3a0..887cfb16b53e 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
@@ -554,6 +554,7 @@
 	0
 #define MCP_REG_CPU_STATE \
 	0xe05004UL
+#define MCP_REG_CPU_STATE_SOFT_HALTED	(0x1UL << 10)
 #define MCP_REG_CPU_EVENT_MASK \
 	0xe05008UL
 #define PGLUE_B_REG_PF_BAR0_SIZE \
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem
@ 2018-09-20  2:48 Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 02/25] qed: Wait for MCP halt and resume commands to take place Sasha Levin
                   ` (23 more replies)
  0 siblings, 24 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: Tomer Tayar, Ariel Elior, David S . Miller, Sasha Levin

From: Tomer Tayar <Tomer.Tayar@cavium.com>

[ Upstream commit f00d25f3154b676fcea4502a25b94bd7f142ca74 ]

The MFW might be reset and re-update its shared memory.
Upon the detection of such a reset the driver rereads this memory, but it
has to wait till the data is valid.
This patch adds the missing wait for a data ready indication.

Signed-off-by: Tomer Tayar <Tomer.Tayar@cavium.com>
Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/net/ethernet/qlogic/qed/qed_mcp.c | 50 +++++++++++++++++++----
 1 file changed, 41 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 3c469355f5a4..9348d367cfdf 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -182,18 +182,57 @@ int qed_mcp_free(struct qed_hwfn *p_hwfn)
 	return 0;
 }
 
+/* Maximum of 1 sec to wait for the SHMEM ready indication */
+#define QED_MCP_SHMEM_RDY_MAX_RETRIES	20
+#define QED_MCP_SHMEM_RDY_ITER_MS	50
+
 static int qed_load_mcp_offsets(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 {
 	struct qed_mcp_info *p_info = p_hwfn->mcp_info;
+	u8 cnt = QED_MCP_SHMEM_RDY_MAX_RETRIES;
+	u8 msec = QED_MCP_SHMEM_RDY_ITER_MS;
 	u32 drv_mb_offsize, mfw_mb_offsize;
 	u32 mcp_pf_id = MCP_PF_ID(p_hwfn);
 
 	p_info->public_base = qed_rd(p_hwfn, p_ptt, MISC_REG_SHARED_MEM_ADDR);
-	if (!p_info->public_base)
-		return 0;
+	if (!p_info->public_base) {
+		DP_NOTICE(p_hwfn,
+			  "The address of the MCP scratch-pad is not configured\n");
+		return -EINVAL;
+	}
 
 	p_info->public_base |= GRCBASE_MCP;
 
+	/* Get the MFW MB address and number of supported messages */
+	mfw_mb_offsize = qed_rd(p_hwfn, p_ptt,
+				SECTION_OFFSIZE_ADDR(p_info->public_base,
+						     PUBLIC_MFW_MB));
+	p_info->mfw_mb_addr = SECTION_ADDR(mfw_mb_offsize, mcp_pf_id);
+	p_info->mfw_mb_length = (u16)qed_rd(p_hwfn, p_ptt,
+					    p_info->mfw_mb_addr +
+					    offsetof(struct public_mfw_mb,
+						     sup_msgs));
+
+	/* The driver can notify that there was an MCP reset, and might read the
+	 * SHMEM values before the MFW has completed initializing them.
+	 * To avoid this, the "sup_msgs" field in the MFW mailbox is used as a
+	 * data ready indication.
+	 */
+	while (!p_info->mfw_mb_length && --cnt) {
+		msleep(msec);
+		p_info->mfw_mb_length =
+			(u16)qed_rd(p_hwfn, p_ptt,
+				    p_info->mfw_mb_addr +
+				    offsetof(struct public_mfw_mb, sup_msgs));
+	}
+
+	if (!cnt) {
+		DP_NOTICE(p_hwfn,
+			  "Failed to get the SHMEM ready notification after %d msec\n",
+			  QED_MCP_SHMEM_RDY_MAX_RETRIES * msec);
+		return -EBUSY;
+	}
+
 	/* Calculate the driver and MFW mailbox address */
 	drv_mb_offsize = qed_rd(p_hwfn, p_ptt,
 				SECTION_OFFSIZE_ADDR(p_info->public_base,
@@ -203,13 +242,6 @@ static int qed_load_mcp_offsets(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 		   "drv_mb_offsiz = 0x%x, drv_mb_addr = 0x%x mcp_pf_id = 0x%x\n",
 		   drv_mb_offsize, p_info->drv_mb_addr, mcp_pf_id);
 
-	/* Set the MFW MB address */
-	mfw_mb_offsize = qed_rd(p_hwfn, p_ptt,
-				SECTION_OFFSIZE_ADDR(p_info->public_base,
-						     PUBLIC_MFW_MB));
-	p_info->mfw_mb_addr = SECTION_ADDR(mfw_mb_offsize, mcp_pf_id);
-	p_info->mfw_mb_length =	(u16)qed_rd(p_hwfn, p_ptt, p_info->mfw_mb_addr);
-
 	/* Get the current driver mailbox sequence before sending
 	 * the first command
 	 */
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 03/25] qed: Prevent a possible deadlock during driver load and unload
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 02/25] qed: Wait for MCP halt and resume commands to take place Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 05/25] thermal: of-thermal: disable passive polling when thermal zone is disabled Sasha Levin
                   ` (21 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: Tomer Tayar, Ariel Elior, David S . Miller, Sasha Levin

From: Tomer Tayar <Tomer.Tayar@cavium.com>

[ Upstream commit eaa50fc59e5841910987e90b0438b2643041f508 ]

The MFW manages an internal lock to prevent concurrent hardware
(de)initialization of different PFs.
This, together with the busy-waiting for the MFW's responses for commands,
might lead to a deadlock during concurrent load or unload of PFs.
This patch adds the option to sleep within the busy-waiting, and uses it
for the (un)load requests (which are not sent from an interrupt context) to
prevent the possible deadlock.

Signed-off-by: Tomer Tayar <Tomer.Tayar@cavium.com>
Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/net/ethernet/qlogic/qed/qed_mcp.c | 43 ++++++++++++++++-------
 drivers/net/ethernet/qlogic/qed/qed_mcp.h | 21 ++++++-----
 2 files changed, 44 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index 5748bcb26d40..a1e1092fa3ce 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -47,7 +47,7 @@
 #include "qed_reg_addr.h"
 #include "qed_sriov.h"
 
-#define CHIP_MCP_RESP_ITER_US 10
+#define QED_MCP_RESP_ITER_US	10
 
 #define QED_DRV_MB_MAX_RETRIES	(500 * 1000)	/* Account for 5 sec */
 #define QED_MCP_RESET_RETRIES	(50 * 1000)	/* Account for 500 msec */
@@ -316,7 +316,7 @@ static void qed_mcp_reread_offsets(struct qed_hwfn *p_hwfn,
 
 int qed_mcp_reset(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 {
-	u32 org_mcp_reset_seq, seq, delay = CHIP_MCP_RESP_ITER_US, cnt = 0;
+	u32 org_mcp_reset_seq, seq, delay = QED_MCP_RESP_ITER_US, cnt = 0;
 	int rc = 0;
 
 	/* Ensure that only a single thread is accessing the mailbox */
@@ -448,10 +448,10 @@ static int
 _qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 		       struct qed_ptt *p_ptt,
 		       struct qed_mcp_mb_params *p_mb_params,
-		       u32 max_retries, u32 delay)
+		       u32 max_retries, u32 usecs)
 {
+	u32 cnt = 0, msecs = DIV_ROUND_UP(usecs, 1000);
 	struct qed_mcp_cmd_elem *p_cmd_elem;
-	u32 cnt = 0;
 	u16 seq_num;
 	int rc = 0;
 
@@ -474,7 +474,11 @@ _qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 			goto err;
 
 		spin_unlock_bh(&p_hwfn->mcp_info->cmd_lock);
-		udelay(delay);
+
+		if (QED_MB_FLAGS_IS_SET(p_mb_params, CAN_SLEEP))
+			msleep(msecs);
+		else
+			udelay(usecs);
 	} while (++cnt < max_retries);
 
 	if (cnt >= max_retries) {
@@ -503,7 +507,11 @@ _qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 		 * The spinlock stays locked until the list element is removed.
 		 */
 
-		udelay(delay);
+		if (QED_MB_FLAGS_IS_SET(p_mb_params, CAN_SLEEP))
+			msleep(msecs);
+		else
+			udelay(usecs);
+
 		spin_lock_bh(&p_hwfn->mcp_info->cmd_lock);
 
 		if (p_cmd_elem->b_is_completed)
@@ -538,7 +546,7 @@ _qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 		   "MFW mailbox: response 0x%08x param 0x%08x [after %d.%03d ms]\n",
 		   p_mb_params->mcp_resp,
 		   p_mb_params->mcp_param,
-		   (cnt * delay) / 1000, (cnt * delay) % 1000);
+		   (cnt * usecs) / 1000, (cnt * usecs) % 1000);
 
 	/* Clear the sequence number from the MFW response */
 	p_mb_params->mcp_resp &= FW_MSG_CODE_MASK;
@@ -556,7 +564,7 @@ static int qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 {
 	size_t union_data_size = sizeof(union drv_union_data);
 	u32 max_retries = QED_DRV_MB_MAX_RETRIES;
-	u32 delay = CHIP_MCP_RESP_ITER_US;
+	u32 usecs = QED_MCP_RESP_ITER_US;
 
 	/* MCP not initialized */
 	if (!qed_mcp_is_init(p_hwfn)) {
@@ -573,8 +581,13 @@ static int qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 		return -EINVAL;
 	}
 
+	if (QED_MB_FLAGS_IS_SET(p_mb_params, CAN_SLEEP)) {
+		max_retries = DIV_ROUND_UP(max_retries, 1000);
+		usecs *= 1000;
+	}
+
 	return _qed_mcp_cmd_and_union(p_hwfn, p_ptt, p_mb_params, max_retries,
-				      delay);
+				      usecs);
 }
 
 int qed_mcp_cmd(struct qed_hwfn *p_hwfn,
@@ -763,6 +776,7 @@ __qed_mcp_load_req(struct qed_hwfn *p_hwfn,
 	mb_params.data_src_size = sizeof(load_req);
 	mb_params.p_data_dst = &load_rsp;
 	mb_params.data_dst_size = sizeof(load_rsp);
+	mb_params.flags = QED_MB_FLAG_CAN_SLEEP;
 
 	DP_VERBOSE(p_hwfn, QED_MSG_SP,
 		   "Load Request: param 0x%08x [init_hw %d, drv_type %d, hsi_ver %d, pda 0x%04x]\n",
@@ -984,7 +998,8 @@ int qed_mcp_load_req(struct qed_hwfn *p_hwfn,
 
 int qed_mcp_unload_req(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 {
-	u32 wol_param, mcp_resp, mcp_param;
+	struct qed_mcp_mb_params mb_params;
+	u32 wol_param;
 
 	switch (p_hwfn->cdev->wol_config) {
 	case QED_OV_WOL_DISABLED:
@@ -1002,8 +1017,12 @@ int qed_mcp_unload_req(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 		wol_param = DRV_MB_PARAM_UNLOAD_WOL_MCP;
 	}
 
-	return qed_mcp_cmd(p_hwfn, p_ptt, DRV_MSG_CODE_UNLOAD_REQ, wol_param,
-			   &mcp_resp, &mcp_param);
+	memset(&mb_params, 0, sizeof(mb_params));
+	mb_params.cmd = DRV_MSG_CODE_UNLOAD_REQ;
+	mb_params.param = wol_param;
+	mb_params.flags = QED_MB_FLAG_CAN_SLEEP;
+
+	return qed_mcp_cmd_and_union(p_hwfn, p_ptt, &mb_params);
 }
 
 int qed_mcp_unload_done(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
index c7ec2395d1ce..e5d0483f3c44 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
@@ -565,14 +565,19 @@ struct qed_mcp_info {
 };
 
 struct qed_mcp_mb_params {
-	u32			cmd;
-	u32			param;
-	void			*p_data_src;
-	u8			data_src_size;
-	void			*p_data_dst;
-	u8			data_dst_size;
-	u32			mcp_resp;
-	u32			mcp_param;
+	u32 cmd;
+	u32 param;
+	void *p_data_src;
+	void *p_data_dst;
+	u8 data_src_size;
+	u8 data_dst_size;
+	u32 mcp_resp;
+	u32 mcp_param;
+	u32 flags;
+#define QED_MB_FLAG_CAN_SLEEP	(0x1 << 0)
+#define QED_MB_FLAGS_IS_SET(params, flag) \
+	({ typeof(params) __params = (params); \
+	   (__params && (__params->flags & QED_MB_FLAG_ ## flag)); })
 };
 
 /**
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 04/25] qed: Avoid sending mailbox commands when MFW is not responsive
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (2 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 05/25] thermal: of-thermal: disable passive polling when thermal zone is disabled Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 06/25] isofs: reject hardware sector size > 2048 bytes Sasha Levin
                   ` (19 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: Tomer Tayar, Ariel Elior, David S . Miller, Sasha Levin

From: Tomer Tayar <Tomer.Tayar@cavium.com>

[ Upstream commit b310974e041913231b6e3d5d475d4df55c312301 ]

Keep sending mailbox commands to the MFW when it is not responsive ends up
with a redundant amount of timeout expiries.
This patch prints the MCP status on the first command which is not
responded, and blocks the following commands.
Since the (un)load request commands might be not responded due to other
PFs, the patch also adds the option to skip the blocking upon a failure.

Signed-off-by: Tomer Tayar <Tomer.Tayar@cavium.com>
Signed-off-by: Ariel Elior <Ariel.Elior@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/net/ethernet/qlogic/qed/qed_mcp.c     | 52 ++++++++++++++++++-
 drivers/net/ethernet/qlogic/qed/qed_mcp.h     |  6 ++-
 .../net/ethernet/qlogic/qed/qed_reg_addr.h    |  1 +
 3 files changed, 56 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index a1e1092fa3ce..7938abe9a301 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -319,6 +319,12 @@ int qed_mcp_reset(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 	u32 org_mcp_reset_seq, seq, delay = QED_MCP_RESP_ITER_US, cnt = 0;
 	int rc = 0;
 
+	if (p_hwfn->mcp_info->b_block_cmd) {
+		DP_NOTICE(p_hwfn,
+			  "The MFW is not responsive. Avoid sending MCP_RESET mailbox command.\n");
+		return -EBUSY;
+	}
+
 	/* Ensure that only a single thread is accessing the mailbox */
 	spin_lock_bh(&p_hwfn->mcp_info->cmd_lock);
 
@@ -444,6 +450,33 @@ static void __qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 		   (p_mb_params->cmd | seq_num), p_mb_params->param);
 }
 
+static void qed_mcp_cmd_set_blocking(struct qed_hwfn *p_hwfn, bool block_cmd)
+{
+	p_hwfn->mcp_info->b_block_cmd = block_cmd;
+
+	DP_INFO(p_hwfn, "%s sending of mailbox commands to the MFW\n",
+		block_cmd ? "Block" : "Unblock");
+}
+
+static void qed_mcp_print_cpu_info(struct qed_hwfn *p_hwfn,
+				   struct qed_ptt *p_ptt)
+{
+	u32 cpu_mode, cpu_state, cpu_pc_0, cpu_pc_1, cpu_pc_2;
+	u32 delay = QED_MCP_RESP_ITER_US;
+
+	cpu_mode = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_MODE);
+	cpu_state = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_STATE);
+	cpu_pc_0 = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_PROGRAM_COUNTER);
+	udelay(delay);
+	cpu_pc_1 = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_PROGRAM_COUNTER);
+	udelay(delay);
+	cpu_pc_2 = qed_rd(p_hwfn, p_ptt, MCP_REG_CPU_PROGRAM_COUNTER);
+
+	DP_NOTICE(p_hwfn,
+		  "MCP CPU info: mode 0x%08x, state 0x%08x, pc {0x%08x, 0x%08x, 0x%08x}\n",
+		  cpu_mode, cpu_state, cpu_pc_0, cpu_pc_1, cpu_pc_2);
+}
+
 static int
 _qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 		       struct qed_ptt *p_ptt,
@@ -530,11 +563,15 @@ _qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 		DP_NOTICE(p_hwfn,
 			  "The MFW failed to respond to command 0x%08x [param 0x%08x].\n",
 			  p_mb_params->cmd, p_mb_params->param);
+		qed_mcp_print_cpu_info(p_hwfn, p_ptt);
 
 		spin_lock_bh(&p_hwfn->mcp_info->cmd_lock);
 		qed_mcp_cmd_del_elem(p_hwfn, p_cmd_elem);
 		spin_unlock_bh(&p_hwfn->mcp_info->cmd_lock);
 
+		if (!QED_MB_FLAGS_IS_SET(p_mb_params, AVOID_BLOCK))
+			qed_mcp_cmd_set_blocking(p_hwfn, true);
+
 		return -EAGAIN;
 	}
 
@@ -572,6 +609,13 @@ static int qed_mcp_cmd_and_union(struct qed_hwfn *p_hwfn,
 		return -EBUSY;
 	}
 
+	if (p_hwfn->mcp_info->b_block_cmd) {
+		DP_NOTICE(p_hwfn,
+			  "The MFW is not responsive. Avoid sending mailbox command 0x%08x [param 0x%08x].\n",
+			  p_mb_params->cmd, p_mb_params->param);
+		return -EBUSY;
+	}
+
 	if (p_mb_params->data_src_size > union_data_size ||
 	    p_mb_params->data_dst_size > union_data_size) {
 		DP_ERR(p_hwfn,
@@ -776,7 +820,7 @@ __qed_mcp_load_req(struct qed_hwfn *p_hwfn,
 	mb_params.data_src_size = sizeof(load_req);
 	mb_params.p_data_dst = &load_rsp;
 	mb_params.data_dst_size = sizeof(load_rsp);
-	mb_params.flags = QED_MB_FLAG_CAN_SLEEP;
+	mb_params.flags = QED_MB_FLAG_CAN_SLEEP | QED_MB_FLAG_AVOID_BLOCK;
 
 	DP_VERBOSE(p_hwfn, QED_MSG_SP,
 		   "Load Request: param 0x%08x [init_hw %d, drv_type %d, hsi_ver %d, pda 0x%04x]\n",
@@ -1020,7 +1064,7 @@ int qed_mcp_unload_req(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 	memset(&mb_params, 0, sizeof(mb_params));
 	mb_params.cmd = DRV_MSG_CODE_UNLOAD_REQ;
 	mb_params.param = wol_param;
-	mb_params.flags = QED_MB_FLAG_CAN_SLEEP;
+	mb_params.flags = QED_MB_FLAG_CAN_SLEEP | QED_MB_FLAG_AVOID_BLOCK;
 
 	return qed_mcp_cmd_and_union(p_hwfn, p_ptt, &mb_params);
 }
@@ -2047,6 +2091,8 @@ int qed_mcp_halt(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 		return -EBUSY;
 	}
 
+	qed_mcp_cmd_set_blocking(p_hwfn, true);
+
 	return 0;
 }
 
@@ -2071,6 +2117,8 @@ int qed_mcp_resume(struct qed_hwfn *p_hwfn, struct qed_ptt *p_ptt)
 		return -EBUSY;
 	}
 
+	qed_mcp_cmd_set_blocking(p_hwfn, false);
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.h b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
index e5d0483f3c44..f1fe5e3427ea 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.h
@@ -540,11 +540,14 @@ struct qed_mcp_info {
 	 */
 	spinlock_t				cmd_lock;
 
+	/* Flag to indicate whether sending a MFW mailbox command is blocked */
+	bool					b_block_cmd;
+
 	/* Spinlock used for syncing SW link-changes and link-changes
 	 * originating from attention context.
 	 */
 	spinlock_t				link_lock;
-	bool					block_mb_sending;
+
 	u32					public_base;
 	u32					drv_mb_addr;
 	u32					mfw_mb_addr;
@@ -575,6 +578,7 @@ struct qed_mcp_mb_params {
 	u32 mcp_param;
 	u32 flags;
 #define QED_MB_FLAG_CAN_SLEEP	(0x1 << 0)
+#define QED_MB_FLAG_AVOID_BLOCK	(0x1 << 1)
 #define QED_MB_FLAGS_IS_SET(params, flag) \
 	({ typeof(params) __params = (params); \
 	   (__params && (__params->flags & QED_MB_FLAG_ ## flag)); })
diff --git a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
index 887cfb16b53e..d1201bb2d4bb 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_reg_addr.h
@@ -557,6 +557,7 @@
 #define MCP_REG_CPU_STATE_SOFT_HALTED	(0x1UL << 10)
 #define MCP_REG_CPU_EVENT_MASK \
 	0xe05008UL
+#define MCP_REG_CPU_PROGRAM_COUNTER	0xe0501cUL
 #define PGLUE_B_REG_PF_BAR0_SIZE \
 	0x2aae60UL
 #define PGLUE_B_REG_PF_BAR1_SIZE \
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 05/25] thermal: of-thermal: disable passive polling when thermal zone is disabled
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 02/25] qed: Wait for MCP halt and resume commands to take place Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 03/25] qed: Prevent a possible deadlock during driver load and unload Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 04/25] qed: Avoid sending mailbox commands when MFW is not responsive Sasha Levin
                   ` (20 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Anson Huang, Eduardo Valentin, Sasha Levin

From: Anson Huang <Anson.Huang@nxp.com>

[ Upstream commit 152395fd03d4ce1e535a75cdbf58105e50587611 ]

When thermal zone is in passive mode, disabling its mode from
sysfs is NOT taking effect at all, it is still polling the
temperature of the disabled thermal zone and handling all thermal
trips, it makes user confused. The disabling operation should
disable the thermal zone behavior completely, for both active and
passive mode, this patch clears the passive_delay when thermal
zone is disabled and restores it when it is enabled.

Signed-off-by: Anson Huang <Anson.Huang@nxp.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/thermal/of-thermal.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/thermal/of-thermal.c b/drivers/thermal/of-thermal.c
index d04ec3b9e5ff..8a70b57d129c 100644
--- a/drivers/thermal/of-thermal.c
+++ b/drivers/thermal/of-thermal.c
@@ -278,10 +278,13 @@ static int of_thermal_set_mode(struct thermal_zone_device *tz,
 
 	mutex_lock(&tz->lock);
 
-	if (mode == THERMAL_DEVICE_ENABLED)
+	if (mode == THERMAL_DEVICE_ENABLED) {
 		tz->polling_delay = data->polling_delay;
-	else
+		tz->passive_delay = data->passive_delay;
+	} else {
 		tz->polling_delay = 0;
+		tz->passive_delay = 0;
+	}
 
 	mutex_unlock(&tz->lock);
 
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 06/25] isofs: reject hardware sector size > 2048 bytes
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (3 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 04/25] qed: Avoid sending mailbox commands when MFW is not responsive Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 07/25] tls: possible hang when do_tcp_sendpages hits sndbuf is full case Sasha Levin
                   ` (18 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Eric Sandeen, Jan Kara, Sasha Levin

From: Eric Sandeen <sandeen@redhat.com>

[ Upstream commit 09a4e0be5826aa66c4ce9954841f110ffe63ef4f ]

The largest block size supported by isofs is ISOFS_BLOCK_SIZE (2048), but
isofs_fill_super calls sb_min_blocksize and sets the blocksize to the
device's logical block size if it's larger than what we ended up with after
option parsing.

If for some reason we try to mount a hard 4k device as an isofs filesystem,
we'll set opt.blocksize to 4096, and when we try to read the superblock
we found via:

        block = iso_blknum << (ISOFS_BLOCK_BITS - s->s_blocksize_bits)

with s_blocksize_bits greater than ISOFS_BLOCK_BITS, we'll have a negative
shift and the bread will fail somewhat cryptically:

  isofs_fill_super: bread failed, dev=sda, iso_blknum=17, block=-2147483648

It seems best to just catch and clearly reject mounts of such a device.

Reported-by: Bryan Gurney <bgurney@redhat.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 fs/isofs/inode.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c
index ed4edcd2bc56..a4994e25e19e 100644
--- a/fs/isofs/inode.c
+++ b/fs/isofs/inode.c
@@ -24,6 +24,7 @@
 #include <linux/mpage.h>
 #include <linux/user_namespace.h>
 #include <linux/seq_file.h>
+#include <linux/blkdev.h>
 
 #include "isofs.h"
 #include "zisofs.h"
@@ -653,6 +654,12 @@ static int isofs_fill_super(struct super_block *s, void *data, int silent)
 	/*
 	 * What if bugger tells us to go beyond page size?
 	 */
+	if (bdev_logical_block_size(s->s_bdev) > 2048) {
+		printk(KERN_WARNING
+		       "ISOFS: unsupported/invalid hardware sector size %d\n",
+			bdev_logical_block_size(s->s_bdev));
+		goto out_freesbi;
+	}
 	opt.blocksize = sb_min_blocksize(s, opt.blocksize);
 
 	sbi->s_high_sierra = 0; /* default is iso9660 */
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 07/25] tls: possible hang when do_tcp_sendpages hits sndbuf is full case
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (4 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 06/25] isofs: reject hardware sector size > 2048 bytes Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 08/25] bpf: sockmap: write_space events need to be passed to TCP handler Sasha Levin
                   ` (17 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: John Fastabend, Daniel Borkmann, Sasha Levin

From: John Fastabend <john.fastabend@gmail.com>

[ Upstream commit 67db7cd249e71f64346f481b629724376d063e08 ]

Currently, the lower protocols sk_write_space handler is not called if
TLS is sending a scatterlist via  tls_push_sg. However, normally
tls_push_sg calls do_tcp_sendpage, which may be under memory pressure,
that in turn may trigger a wait via sk_wait_event. Typically, this
happens when the in-flight bytes exceed the sdnbuf size. In the normal
case when enough ACKs are received sk_write_space() will be called and
the sk_wait_event will be woken up allowing it to send more data
and/or return to the user.

But, in the TLS case because the sk_write_space() handler does not
wake up the events the above send will wait until the sndtimeo is
exceeded. By default this is MAX_SCHEDULE_TIMEOUT so it look like a
hang to the user (especially this impatient user). To fix this pass
the sk_write_space event to the lower layers sk_write_space event
which in the TCP case will wake any pending events.

I observed the above while integrating sockmap and ktls. It
initially appeared as test_sockmap (modified to use ktls) occasionally
hanging. To reliably reproduce this reduce the sndbuf size and stress
the tls layer by sending many 1B sends. This results in every byte
needing a header and each byte individually being sent to the crypto
layer.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Acked-by: Dave Watson <davejwatson@fb.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 net/tls/tls_main.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c
index 055b9992d8c7..fa011318ffe4 100644
--- a/net/tls/tls_main.c
+++ b/net/tls/tls_main.c
@@ -195,9 +195,14 @@ static void tls_write_space(struct sock *sk)
 {
 	struct tls_context *ctx = tls_get_ctx(sk);
 
-	/* We are already sending pages, ignore notification */
-	if (ctx->in_tcp_sendpages)
+	/* If in_tcp_sendpages call lower protocol write space handler
+	 * to ensure we wake up any waiting operations there. For example
+	 * if do_tcp_sendpages where to call sk_wait_event.
+	 */
+	if (ctx->in_tcp_sendpages) {
+		ctx->sk_write_space(sk);
 		return;
+	}
 
 	if (!sk->sk_write_pending && tls_is_pending_closed_record(ctx)) {
 		gfp_t sk_allocation = sk->sk_allocation;
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 08/25] bpf: sockmap: write_space events need to be passed to TCP handler
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (5 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 07/25] tls: possible hang when do_tcp_sendpages hits sndbuf is full case Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 09/25] net: hns: fix length and page_offset overflow when CONFIG_ARM64_64K_PAGES Sasha Levin
                   ` (16 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: John Fastabend, Daniel Borkmann, Sasha Levin

From: John Fastabend <john.fastabend@gmail.com>

[ Upstream commit 9b2e0388bec8ec5427403e23faff3b58dd1c3200 ]

When sockmap code is using the stream parser it also handles the write
space events in order to handle the case where (a) verdict redirects
skb to another socket and (b) the sockmap then sends the skb but due
to memory constraints (or other EAGAIN errors) needs to do a retry.

But the initial code missed a third case where the
skb_send_sock_locked() triggers an sk_wait_event(). A typically case
would be when sndbuf size is exceeded. If this happens because we
do not pass the write_space event to the lower layers we never wake
up the event and it will wait for sndtimeo. Which as noted in ktls
fix may be rather large and look like a hang to the user.

To reproduce the best test is to reduce the sndbuf size and send
1B data chunks to stress the memory handling. To fix this pass the
event from the upper layer to the lower layer.

Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 kernel/bpf/sockmap.c | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/kernel/bpf/sockmap.c b/kernel/bpf/sockmap.c
index 53a4787c08d8..20eaddfa691c 100644
--- a/kernel/bpf/sockmap.c
+++ b/kernel/bpf/sockmap.c
@@ -313,12 +313,15 @@ static void smap_tx_work(struct work_struct *w)
 static void smap_write_space(struct sock *sk)
 {
 	struct smap_psock *psock;
+	void (*write_space)(struct sock *sk);
 
 	rcu_read_lock();
 	psock = smap_psock_sk(sk);
 	if (likely(psock && test_bit(SMAP_TX_RUNNING, &psock->state)))
 		schedule_work(&psock->tx_work);
+	write_space = psock->save_write_space;
 	rcu_read_unlock();
+	write_space(sk);
 }
 
 static void smap_stop_sock(struct smap_psock *psock, struct sock *sk)
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 09/25] net: hns: fix length and page_offset overflow when CONFIG_ARM64_64K_PAGES
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (6 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 08/25] bpf: sockmap: write_space events need to be passed to TCP handler Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 10/25] net: hns: fix skb->truesize underestimation Sasha Levin
                   ` (15 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: Huazhong Tan, Salil Mehta, David S . Miller, Sasha Levin

From: Huazhong Tan <tanhuazhong@huawei.com>

[ Upstream commit 3ed614dce3ca9912d22be215ff0f11104b69fe62 ]

When enable the config item "CONFIG_ARM64_64K_PAGES", the size of PAGE_SIZE
is 65536(64K). But the  type of length and page_offset are u16, they will
overflow. So change them to u32.

Fixes: 6fe6611ff275 ("net: add Hisilicon Network Subsystem hnae framework support")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Salil Mehta <salil.mehta@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/net/ethernet/hisilicon/hns/hnae.h | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/hisilicon/hns/hnae.h b/drivers/net/ethernet/hisilicon/hns/hnae.h
index fa5b30f547f6..cad52bd331f7 100644
--- a/drivers/net/ethernet/hisilicon/hns/hnae.h
+++ b/drivers/net/ethernet/hisilicon/hns/hnae.h
@@ -220,10 +220,10 @@ struct hnae_desc_cb {
 
 	/* priv data for the desc, e.g. skb when use with ip stack*/
 	void *priv;
-	u16 page_offset;
-	u16 reuse_flag;
+	u32 page_offset;
+	u32 length;     /* length of the buffer */
 
-	u16 length;     /* length of the buffer */
+	u16 reuse_flag;
 
        /* desc type, used by the ring user to mark the type of the priv data */
 	u16 type;
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 10/25] net: hns: fix skb->truesize underestimation
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (7 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 09/25] net: hns: fix length and page_offset overflow when CONFIG_ARM64_64K_PAGES Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 11/25] e1000: check on netif_running() before calling e1000_up() Sasha Levin
                   ` (14 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: Huazhong Tan, Salil Mehta, David S . Miller, Sasha Levin

From: Huazhong Tan <tanhuazhong@huawei.com>

[ Upstream commit b1ccd4c0ab6ef499f47dd84ed4920502a7147bba ]

skb->truesize is not meant to be tracking amount of used bytes in a skb,
but amount of reserved/consumed bytes in memory.

For instance, if we use a single byte in last page fragment, we have to
account the full size of the fragment.

So skb_add_rx_frag needs to calculate the length of the entire buffer into
turesize.

Fixes: 9cbe9fd5214e ("net: hns: optimize XGE capability by reducing cpu usage")
Signed-off-by: Huazhong tan <tanhuazhong@huawei.com>
Signed-off-by: Salil Mehta <salil.mehta@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/net/ethernet/hisilicon/hns/hns_enet.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
index e77192683dba..25a9732afc84 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
@@ -530,7 +530,7 @@ static void hns_nic_reuse_page(struct sk_buff *skb, int i,
 	}
 
 	skb_add_rx_frag(skb, i, desc_cb->priv, desc_cb->page_offset + pull_len,
-			size - pull_len, truesize - pull_len);
+			size - pull_len, truesize);
 
 	 /* avoid re-using remote pages,flag default unreuse */
 	if (unlikely(page_to_nid(desc_cb->priv) != numa_node_id()))
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 11/25] e1000: check on netif_running() before calling e1000_up()
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (8 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 10/25] net: hns: fix skb->truesize underestimation Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 12/25] e1000: ensure to free old tx/rx rings in set_ringparam() Sasha Levin
                   ` (13 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Bo Chen, Jeff Kirsher, Sasha Levin

From: Bo Chen <chenbo@pdx.edu>

[ Upstream commit cf1acec008f8d7761aa3fd7c4bca7e17b2d2512d ]

When the device is not up, the call to 'e1000_up()' from the error handling path
of 'e1000_set_ringparam()' causes a kernel oops with a null-pointer
dereference. The null-pointer dereference is triggered in function
'e1000_alloc_rx_buffers()' at line 'buffer_info = &rx_ring->buffer_info[i]'.

This bug was reported by COD, a tool for testing kernel module binaries I am
building. This bug was also detected by KFI from Dr. Kai Cong.

This patch fixes the bug by checking on 'netif_running()' before calling
'e1000_up()' in 'e1000_set_ringparam()'.

Signed-off-by: Bo Chen <chenbo@pdx.edu>
Acked-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/net/ethernet/intel/e1000/e1000_ethtool.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
index 3b3983a1ffbb..d44d3643677d 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
@@ -664,7 +664,8 @@ static int e1000_set_ringparam(struct net_device *netdev,
 err_alloc_rx:
 	kfree(txdr);
 err_alloc_tx:
-	e1000_up(adapter);
+	if (netif_running(adapter->netdev))
+		e1000_up(adapter);
 err_setup:
 	clear_bit(__E1000_RESETTING, &adapter->flags);
 	return err;
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 12/25] e1000: ensure to free old tx/rx rings in set_ringparam()
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (9 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 11/25] e1000: check on netif_running() before calling e1000_up() Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 13/25] crypto: cavium/nitrox - fix for command corruption in queue full case with backlog submissions Sasha Levin
                   ` (12 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Bo Chen, Jeff Kirsher, Sasha Levin

From: Bo Chen <chenbo@pdx.edu>

[ Upstream commit ee400a3f1bfe7004a3e14b81c38ccc5583c26295 ]

In 'e1000_set_ringparam()', the tx_ring and rx_ring are updated with new value
and the old tx/rx rings are freed only when the device is up. There are resource
leaks on old tx/rx rings when the device is not up. This bug is reported by COD,
a tool for testing kernel module binaries I am building.

This patch fixes the bug by always calling 'kfree()' on old tx/rx rings in
'e1000_set_ringparam()'.

Signed-off-by: Bo Chen <chenbo@pdx.edu>
Reviewed-by: Alexander Duyck <alexander.h.duyck@intel.com>
Tested-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/net/ethernet/intel/e1000/e1000_ethtool.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
index d44d3643677d..10df2d60c181 100644
--- a/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
+++ b/drivers/net/ethernet/intel/e1000/e1000_ethtool.c
@@ -644,14 +644,14 @@ static int e1000_set_ringparam(struct net_device *netdev,
 		adapter->tx_ring = tx_old;
 		e1000_free_all_rx_resources(adapter);
 		e1000_free_all_tx_resources(adapter);
-		kfree(tx_old);
-		kfree(rx_old);
 		adapter->rx_ring = rxdr;
 		adapter->tx_ring = txdr;
 		err = e1000_up(adapter);
 		if (err)
 			goto err_setup;
 	}
+	kfree(tx_old);
+	kfree(rx_old);
 
 	clear_bit(__E1000_RESETTING, &adapter->flags);
 	return 0;
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 13/25] crypto: cavium/nitrox - fix for command corruption in queue full case with backlog submissions.
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (10 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 12/25] e1000: ensure to free old tx/rx rings in set_ringparam() Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 14/25] hwmon: (ina2xx) fix sysfs shunt resistor read access Sasha Levin
                   ` (11 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Srikanth Jampala, Herbert Xu, Sasha Levin

From: Srikanth Jampala <Jampala.Srikanth@cavium.com>

[ Upstream commit 3d7c82060d1fe65bde4023aac41a0b1bd7718e07 ]

Earlier used to post the current command without checking queue full
     after backlog submissions. So, post the current command only after
     confirming the space in queue after backlog submissions.

     Maintain host write index instead of reading device registers
     to get the next free slot to post the command.

     Return -ENOSPC in queue full case.

Signed-off-by: Srikanth Jampala <Jampala.Srikanth@cavium.com>
Reviewed-by: Gadam Sreerama <sgadam@cavium.com>
Tested-by: Jha, Chandan <Chandan.Jha@cavium.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/crypto/cavium/nitrox/nitrox_dev.h    |  3 +-
 drivers/crypto/cavium/nitrox/nitrox_lib.c    |  1 +
 drivers/crypto/cavium/nitrox/nitrox_reqmgr.c | 57 +++++++++++---------
 3 files changed, 35 insertions(+), 26 deletions(-)

diff --git a/drivers/crypto/cavium/nitrox/nitrox_dev.h b/drivers/crypto/cavium/nitrox/nitrox_dev.h
index 9a476bb6d4c7..af596455b420 100644
--- a/drivers/crypto/cavium/nitrox/nitrox_dev.h
+++ b/drivers/crypto/cavium/nitrox/nitrox_dev.h
@@ -35,6 +35,7 @@ struct nitrox_cmdq {
 	/* requests in backlog queues */
 	atomic_t backlog_count;
 
+	int write_idx;
 	/* command size 32B/64B */
 	u8 instr_size;
 	u8 qno;
@@ -87,7 +88,7 @@ struct nitrox_bh {
 	struct bh_data *slc;
 };
 
-/* NITROX-5 driver state */
+/* NITROX-V driver state */
 #define NITROX_UCODE_LOADED	0
 #define NITROX_READY		1
 
diff --git a/drivers/crypto/cavium/nitrox/nitrox_lib.c b/drivers/crypto/cavium/nitrox/nitrox_lib.c
index 4fdc921ba611..9906c0086647 100644
--- a/drivers/crypto/cavium/nitrox/nitrox_lib.c
+++ b/drivers/crypto/cavium/nitrox/nitrox_lib.c
@@ -36,6 +36,7 @@ static int cmdq_common_init(struct nitrox_cmdq *cmdq)
 	cmdq->head = PTR_ALIGN(cmdq->head_unaligned, PKT_IN_ALIGN);
 	cmdq->dma = PTR_ALIGN(cmdq->dma_unaligned, PKT_IN_ALIGN);
 	cmdq->qsize = (qsize + PKT_IN_ALIGN);
+	cmdq->write_idx = 0;
 
 	spin_lock_init(&cmdq->response_lock);
 	spin_lock_init(&cmdq->cmdq_lock);
diff --git a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c
index 4addc238a6ef..4adf28176a4e 100644
--- a/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c
+++ b/drivers/crypto/cavium/nitrox/nitrox_reqmgr.c
@@ -43,6 +43,16 @@
  *   Invalid flag options in AES-CCM IV.
  */
 
+static inline int incr_index(int index, int count, int max)
+{
+	if ((index + count) >= max)
+		index = index + count - max;
+	else
+		index += count;
+
+	return index;
+}
+
 /**
  * dma_free_sglist - unmap and free the sg lists.
  * @ndev: N5 device
@@ -427,30 +437,29 @@ static void post_se_instr(struct nitrox_softreq *sr,
 			  struct nitrox_cmdq *cmdq)
 {
 	struct nitrox_device *ndev = sr->ndev;
-	union nps_pkt_in_instr_baoff_dbell pkt_in_baoff_dbell;
-	u64 offset;
+	int idx;
 	u8 *ent;
 
 	spin_lock_bh(&cmdq->cmdq_lock);
 
-	/* get the next write offset */
-	offset = NPS_PKT_IN_INSTR_BAOFF_DBELLX(cmdq->qno);
-	pkt_in_baoff_dbell.value = nitrox_read_csr(ndev, offset);
+	idx = cmdq->write_idx;
 	/* copy the instruction */
-	ent = cmdq->head + pkt_in_baoff_dbell.s.aoff;
+	ent = cmdq->head + (idx * cmdq->instr_size);
 	memcpy(ent, &sr->instr, cmdq->instr_size);
-	/* flush the command queue updates */
-	dma_wmb();
 
-	sr->tstamp = jiffies;
 	atomic_set(&sr->status, REQ_POSTED);
 	response_list_add(sr, cmdq);
+	sr->tstamp = jiffies;
+	/* flush the command queue updates */
+	dma_wmb();
 
 	/* Ring doorbell with count 1 */
 	writeq(1, cmdq->dbell_csr_addr);
 	/* orders the doorbell rings */
 	mmiowb();
 
+	cmdq->write_idx = incr_index(idx, 1, ndev->qlen);
+
 	spin_unlock_bh(&cmdq->cmdq_lock);
 }
 
@@ -460,6 +469,9 @@ static int post_backlog_cmds(struct nitrox_cmdq *cmdq)
 	struct nitrox_softreq *sr, *tmp;
 	int ret = 0;
 
+	if (!atomic_read(&cmdq->backlog_count))
+		return 0;
+
 	spin_lock_bh(&cmdq->backlog_lock);
 
 	list_for_each_entry_safe(sr, tmp, &cmdq->backlog_head, backlog) {
@@ -467,7 +479,7 @@ static int post_backlog_cmds(struct nitrox_cmdq *cmdq)
 
 		/* submit until space available */
 		if (unlikely(cmdq_full(cmdq, ndev->qlen))) {
-			ret = -EBUSY;
+			ret = -ENOSPC;
 			break;
 		}
 		/* delete from backlog list */
@@ -492,23 +504,20 @@ static int nitrox_enqueue_request(struct nitrox_softreq *sr)
 {
 	struct nitrox_cmdq *cmdq = sr->cmdq;
 	struct nitrox_device *ndev = sr->ndev;
-	int ret = -EBUSY;
+
+	/* try to post backlog requests */
+	post_backlog_cmds(cmdq);
 
 	if (unlikely(cmdq_full(cmdq, ndev->qlen))) {
 		if (!(sr->flags & CRYPTO_TFM_REQ_MAY_BACKLOG))
-			return -EAGAIN;
-
+			return -ENOSPC;
+		/* add to backlog list */
 		backlog_list_add(sr, cmdq);
-	} else {
-		ret = post_backlog_cmds(cmdq);
-		if (ret) {
-			backlog_list_add(sr, cmdq);
-			return ret;
-		}
-		post_se_instr(sr, cmdq);
-		ret = -EINPROGRESS;
+		return -EBUSY;
 	}
-	return ret;
+	post_se_instr(sr, cmdq);
+
+	return -EINPROGRESS;
 }
 
 /**
@@ -625,11 +634,9 @@ int nitrox_process_se_request(struct nitrox_device *ndev,
 	 */
 	sr->instr.fdata[0] = *((u64 *)&req->gph);
 	sr->instr.fdata[1] = 0;
-	/* flush the soft_req changes before posting the cmd */
-	wmb();
 
 	ret = nitrox_enqueue_request(sr);
-	if (ret == -EAGAIN)
+	if (ret == -ENOSPC)
 		goto send_fail;
 
 	return ret;
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 15/25] hwmon: (adt7475) Make adt7475_read_word() return errors
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (12 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 14/25] hwmon: (ina2xx) fix sysfs shunt resistor read access Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 16/25] Revert "ARM: dts: imx7d: Invert legacy PCI irq mapping" Sasha Levin
                   ` (9 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Dan Carpenter, Guenter Roeck, Sasha Levin

From: Dan Carpenter <dan.carpenter@oracle.com>

[ Upstream commit f196dec6d50abb2e65fb54a0621b2f1b4d922995 ]

The adt7475_read_word() function was meant to return negative error
codes on failure.

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Tokunori Ikegami <ikegami@allied-telesis.co.jp>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/hwmon/adt7475.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)

diff --git a/drivers/hwmon/adt7475.c b/drivers/hwmon/adt7475.c
index 9ef84998c7f3..37db2eb66ed7 100644
--- a/drivers/hwmon/adt7475.c
+++ b/drivers/hwmon/adt7475.c
@@ -303,14 +303,18 @@ static inline u16 volt2reg(int channel, long volt, u8 bypass_attn)
 	return clamp_val(reg, 0, 1023) & (0xff << 2);
 }
 
-static u16 adt7475_read_word(struct i2c_client *client, int reg)
+static int adt7475_read_word(struct i2c_client *client, int reg)
 {
-	u16 val;
+	int val1, val2;
 
-	val = i2c_smbus_read_byte_data(client, reg);
-	val |= (i2c_smbus_read_byte_data(client, reg + 1) << 8);
+	val1 = i2c_smbus_read_byte_data(client, reg);
+	if (val1 < 0)
+		return val1;
+	val2 = i2c_smbus_read_byte_data(client, reg + 1);
+	if (val2 < 0)
+		return val2;
 
-	return val;
+	return val1 | (val2 << 8);
 }
 
 static void adt7475_write_word(struct i2c_client *client, int reg, u16 val)
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 14/25] hwmon: (ina2xx) fix sysfs shunt resistor read access
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (11 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 13/25] crypto: cavium/nitrox - fix for command corruption in queue full case with backlog submissions Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 15/25] hwmon: (adt7475) Make adt7475_read_word() return errors Sasha Levin
                   ` (10 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Lothar Felten, Guenter Roeck, Sasha Levin

From: Lothar Felten <lothar.felten@gmail.com>

[ Upstream commit 3ad867001c91657c46dcf6656d52eb6080286fd5 ]

fix the sysfs shunt resistor read access: return the shunt resistor
value, not the calibration register contents.

update email address

Signed-off-by: Lothar Felten <lothar.felten@gmail.com>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 Documentation/hwmon/ina2xx           |  2 +-
 drivers/hwmon/ina2xx.c               | 13 +++++++++++--
 include/linux/platform_data/ina2xx.h |  2 +-
 3 files changed, 13 insertions(+), 4 deletions(-)

diff --git a/Documentation/hwmon/ina2xx b/Documentation/hwmon/ina2xx
index cfd31d94c872..f8bf14055c2f 100644
--- a/Documentation/hwmon/ina2xx
+++ b/Documentation/hwmon/ina2xx
@@ -32,7 +32,7 @@ Supported chips:
     Datasheet: Publicly available at the Texas Instruments website
                http://www.ti.com/
 
-Author: Lothar Felten <l-felten@ti.com>
+Author: Lothar Felten <lothar.felten@gmail.com>
 
 Description
 -----------
diff --git a/drivers/hwmon/ina2xx.c b/drivers/hwmon/ina2xx.c
index e9e6aeabbf84..71d3445ba869 100644
--- a/drivers/hwmon/ina2xx.c
+++ b/drivers/hwmon/ina2xx.c
@@ -17,7 +17,7 @@
  * Bi-directional Current/Power Monitor with I2C Interface
  * Datasheet: http://www.ti.com/product/ina230
  *
- * Copyright (C) 2012 Lothar Felten <l-felten@ti.com>
+ * Copyright (C) 2012 Lothar Felten <lothar.felten@gmail.com>
  * Thanks to Jan Volkering
  *
  * This program is free software; you can redistribute it and/or modify
@@ -329,6 +329,15 @@ static int ina2xx_set_shunt(struct ina2xx_data *data, long val)
 	return 0;
 }
 
+static ssize_t ina2xx_show_shunt(struct device *dev,
+			      struct device_attribute *da,
+			      char *buf)
+{
+	struct ina2xx_data *data = dev_get_drvdata(dev);
+
+	return snprintf(buf, PAGE_SIZE, "%li\n", data->rshunt);
+}
+
 static ssize_t ina2xx_store_shunt(struct device *dev,
 				  struct device_attribute *da,
 				  const char *buf, size_t count)
@@ -403,7 +412,7 @@ static SENSOR_DEVICE_ATTR(power1_input, S_IRUGO, ina2xx_show_value, NULL,
 
 /* shunt resistance */
 static SENSOR_DEVICE_ATTR(shunt_resistor, S_IRUGO | S_IWUSR,
-			  ina2xx_show_value, ina2xx_store_shunt,
+			  ina2xx_show_shunt, ina2xx_store_shunt,
 			  INA2XX_CALIBRATION);
 
 /* update interval (ina226 only) */
diff --git a/include/linux/platform_data/ina2xx.h b/include/linux/platform_data/ina2xx.h
index 9abc0ca7259b..9f0aa1b48c78 100644
--- a/include/linux/platform_data/ina2xx.h
+++ b/include/linux/platform_data/ina2xx.h
@@ -1,7 +1,7 @@
 /*
  * Driver for Texas Instruments INA219, INA226 power monitor chips
  *
- * Copyright (C) 2012 Lothar Felten <l-felten@ti.com>
+ * Copyright (C) 2012 Lothar Felten <lothar.felten@gmail.com>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 16/25] Revert "ARM: dts: imx7d: Invert legacy PCI irq mapping"
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (13 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 15/25] hwmon: (adt7475) Make adt7475_read_word() return errors Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 17/25] drm/amdgpu: Enable/disable gfx PG feature in rlc safe mode Sasha Levin
                   ` (8 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Leonard Crestez, Shawn Guo, Sasha Levin

From: Leonard Crestez <leonard.crestez@nxp.com>

[ Upstream commit 538d6e9d597584e80514698e24321645debde78f ]

This reverts commit 1c86c9dd82f859b474474a7fee0d5195da2c9c1d.

That commit followed the reference manual but unfortunately the imx7d
manual is incorrect.

Tested with ath9k pcie card and confirmed internally.

Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
Acked-by: Lucas Stach <l.stach@pengutronix.de>
Fixes: 1c86c9dd82f8 ("ARM: dts: imx7d: Invert legacy PCI irq mapping")
Signed-off-by: Shawn Guo <shawnguo@kernel.org>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 arch/arm/boot/dts/imx7d.dtsi | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/arch/arm/boot/dts/imx7d.dtsi b/arch/arm/boot/dts/imx7d.dtsi
index 4d308d17f040..119b63ffb0fe 100644
--- a/arch/arm/boot/dts/imx7d.dtsi
+++ b/arch/arm/boot/dts/imx7d.dtsi
@@ -144,10 +144,14 @@
 		interrupt-names = "msi";
 		#interrupt-cells = <1>;
 		interrupt-map-mask = <0 0 0 0x7>;
-		interrupt-map = <0 0 0 1 &intc GIC_SPI 122 IRQ_TYPE_LEVEL_HIGH>,
-				<0 0 0 2 &intc GIC_SPI 123 IRQ_TYPE_LEVEL_HIGH>,
-				<0 0 0 3 &intc GIC_SPI 124 IRQ_TYPE_LEVEL_HIGH>,
-				<0 0 0 4 &intc GIC_SPI 125 IRQ_TYPE_LEVEL_HIGH>;
+		/*
+		 * Reference manual lists pci irqs incorrectly
+		 * Real hardware ordering is same as imx6: D+MSI, C, B, A
+		 */
+		interrupt-map = <0 0 0 1 &intc GIC_SPI 125 IRQ_TYPE_LEVEL_HIGH>,
+				<0 0 0 2 &intc GIC_SPI 124 IRQ_TYPE_LEVEL_HIGH>,
+				<0 0 0 3 &intc GIC_SPI 123 IRQ_TYPE_LEVEL_HIGH>,
+				<0 0 0 4 &intc GIC_SPI 122 IRQ_TYPE_LEVEL_HIGH>;
 		clocks = <&clks IMX7D_PCIE_CTRL_ROOT_CLK>,
 			 <&clks IMX7D_PLL_ENET_MAIN_100M_CLK>,
 			 <&clks IMX7D_PCIE_PHY_ROOT_CLK>;
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 17/25] drm/amdgpu: Enable/disable gfx PG feature in rlc safe mode
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (14 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 16/25] Revert "ARM: dts: imx7d: Invert legacy PCI irq mapping" Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 18/25] drm/amdgpu: Update power state at the end of smu hw_init Sasha Levin
                   ` (7 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Rex Zhu, Hang Zhou, Alex Deucher, Sasha Levin

From: Rex Zhu <Rex.Zhu@amd.com>

[ Upstream commit 8ef23364b654d44244400d79988e677e504b21ba ]

This is required by gfx hw and can fix the rlc hang when
do s3 stree test on Cz/St.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Hang Zhou <hang.zhou@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index a7e54820a330..85bcd236890e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -5479,6 +5479,11 @@ static int gfx_v8_0_set_powergating_state(void *handle,
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
+	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_SMG |
+				AMD_PG_SUPPORT_RLC_SMU_HS |
+				AMD_PG_SUPPORT_CP |
+				AMD_PG_SUPPORT_GFX_DMG))
+		adev->gfx.rlc.funcs->enter_safe_mode(adev);
 	switch (adev->asic_type) {
 	case CHIP_CARRIZO:
 	case CHIP_STONEY:
@@ -5527,7 +5532,11 @@ static int gfx_v8_0_set_powergating_state(void *handle,
 	default:
 		break;
 	}
-
+	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_SMG |
+				AMD_PG_SUPPORT_RLC_SMU_HS |
+				AMD_PG_SUPPORT_CP |
+				AMD_PG_SUPPORT_GFX_DMG))
+		adev->gfx.rlc.funcs->exit_safe_mode(adev);
 	return 0;
 }
 
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 19/25] ata: ftide010: Add a quirk for SQ201
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (16 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 18/25] drm/amdgpu: Update power state at the end of smu hw_init Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 20/25] nvme-fcloop: Fix dropped LS's to removed target port Sasha Levin
                   ` (5 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Linus Walleij, Jens Axboe, Sasha Levin

From: Linus Walleij <linus.walleij@linaro.org>

[ Upstream commit 46cb52ad414ac829680d0bb8cc7090ac2b577ca7 ]

The DMA is broken on this specific device for some unknown
reason (probably badly designed or plain broken interface
electronics) and will only work with PIO. Other users of
the same hardware does not have this problem.

Add a specific quirk so that this Gemini device gets
DMA turned off. Also fix up some code around passing the
port information around in probe while we're at it.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/ata/pata_ftide010.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

diff --git a/drivers/ata/pata_ftide010.c b/drivers/ata/pata_ftide010.c
index 5d4b72e21161..569a4a662dcd 100644
--- a/drivers/ata/pata_ftide010.c
+++ b/drivers/ata/pata_ftide010.c
@@ -256,14 +256,12 @@ static struct ata_port_operations pata_ftide010_port_ops = {
 	.qc_issue	= ftide010_qc_issue,
 };
 
-static struct ata_port_info ftide010_port_info[] = {
-	{
-		.flags		= ATA_FLAG_SLAVE_POSS,
-		.mwdma_mask	= ATA_MWDMA2,
-		.udma_mask	= ATA_UDMA6,
-		.pio_mask	= ATA_PIO4,
-		.port_ops	= &pata_ftide010_port_ops,
-	},
+static struct ata_port_info ftide010_port_info = {
+	.flags		= ATA_FLAG_SLAVE_POSS,
+	.mwdma_mask	= ATA_MWDMA2,
+	.udma_mask	= ATA_UDMA6,
+	.pio_mask	= ATA_PIO4,
+	.port_ops	= &pata_ftide010_port_ops,
 };
 
 #if IS_ENABLED(CONFIG_SATA_GEMINI)
@@ -349,6 +347,7 @@ static int pata_ftide010_gemini_cable_detect(struct ata_port *ap)
 }
 
 static int pata_ftide010_gemini_init(struct ftide010 *ftide,
+				     struct ata_port_info *pi,
 				     bool is_ata1)
 {
 	struct device *dev = ftide->dev;
@@ -373,7 +372,13 @@ static int pata_ftide010_gemini_init(struct ftide010 *ftide,
 
 	/* Flag port as SATA-capable */
 	if (gemini_sata_bridge_enabled(sg, is_ata1))
-		ftide010_port_info[0].flags |= ATA_FLAG_SATA;
+		pi->flags |= ATA_FLAG_SATA;
+
+	/* This device has broken DMA, only PIO works */
+	if (of_machine_is_compatible("itian,sq201")) {
+		pi->mwdma_mask = 0;
+		pi->udma_mask = 0;
+	}
 
 	/*
 	 * We assume that a simple 40-wire cable is used in the PATA mode.
@@ -435,6 +440,7 @@ static int pata_ftide010_gemini_init(struct ftide010 *ftide,
 }
 #else
 static int pata_ftide010_gemini_init(struct ftide010 *ftide,
+				     struct ata_port_info *pi,
 				     bool is_ata1)
 {
 	return -ENOTSUPP;
@@ -446,7 +452,7 @@ static int pata_ftide010_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct device_node *np = dev->of_node;
-	const struct ata_port_info pi = ftide010_port_info[0];
+	struct ata_port_info pi = ftide010_port_info;
 	const struct ata_port_info *ppi[] = { &pi, NULL };
 	struct ftide010 *ftide;
 	struct resource *res;
@@ -490,6 +496,7 @@ static int pata_ftide010_probe(struct platform_device *pdev)
 		 * are ATA0. This will also set up the cable types.
 		 */
 		ret = pata_ftide010_gemini_init(ftide,
+				&pi,
 				(res->start == 0x63400000));
 		if (ret)
 			goto err_dis_clk;
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 18/25] drm/amdgpu: Update power state at the end of smu hw_init.
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (15 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 17/25] drm/amdgpu: Enable/disable gfx PG feature in rlc safe mode Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 19/25] ata: ftide010: Add a quirk for SQ201 Sasha Levin
                   ` (6 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Rex Zhu, Alex Deucher, Sasha Levin

From: Rex Zhu <Rex.Zhu@amd.com>

[ Upstream commit 2ab4d0e74256fc49b7b270f63c1d1e47c2455abc ]

For SI/Kv, the power state is managed by function
amdgpu_pm_compute_clocks.

when dpm enabled, we should call amdgpu_pm_compute_clocks
to update current power state instand of set boot state.

this change can fix the oops when kfd driver was enabled on Kv.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Tested-by: Michel Dänzer <michel.daenzer@amd.com>
Signed-off-by: Rex Zhu <Rex.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/gpu/drm/amd/amdgpu/kv_dpm.c | 4 +---
 drivers/gpu/drm/amd/amdgpu/si_dpm.c | 3 +--
 2 files changed, 2 insertions(+), 5 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
index 3bbf2ccfca89..c76073b422d6 100644
--- a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
@@ -1352,8 +1352,6 @@ static int kv_dpm_enable(struct amdgpu_device *adev)
 		return ret;
 	}
 
-	kv_update_current_ps(adev, adev->pm.dpm.boot_ps);
-
 	if (adev->irq.installed &&
 	    amdgpu_is_internal_thermal_sensor(adev->pm.int_thermal_type)) {
 		ret = kv_set_thermal_temperature_range(adev, KV_TEMP_RANGE_MIN, KV_TEMP_RANGE_MAX);
@@ -3054,7 +3052,7 @@ static int kv_dpm_hw_init(void *handle)
 	else
 		adev->pm.dpm_enabled = true;
 	mutex_unlock(&adev->pm.mutex);
-
+	amdgpu_pm_compute_clocks(adev);
 	return ret;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dpm.c b/drivers/gpu/drm/amd/amdgpu/si_dpm.c
index 6f1dea157a77..55613f425931 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dpm.c
@@ -6884,7 +6884,6 @@ static int si_dpm_enable(struct amdgpu_device *adev)
 
 	si_enable_auto_throttle_source(adev, AMDGPU_DPM_AUTO_THROTTLE_SRC_THERMAL, true);
 	si_thermal_start_thermal_controller(adev);
-	ni_update_current_ps(adev, boot_ps);
 
 	return 0;
 }
@@ -7758,7 +7757,7 @@ static int si_dpm_hw_init(void *handle)
 	else
 		adev->pm.dpm_enabled = true;
 	mutex_unlock(&adev->pm.mutex);
-
+	amdgpu_pm_compute_clocks(adev);
 	return ret;
 }
 
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 20/25] nvme-fcloop: Fix dropped LS's to removed target port
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (17 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 19/25] ata: ftide010: Add a quirk for SQ201 Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 21/25] ARM: dts: omap4-droid4: Fix emmc errors seen on some devices Sasha Levin
                   ` (4 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: James Smart, James Smart, Christoph Hellwig, Sasha Levin

From: James Smart <jsmart2021@gmail.com>

[ Upstream commit afd299ca996929f4f98ac20da0044c0cdc124879 ]

When a targetport is removed from the config, fcloop will avoid calling
the LS done() routine thinking the targetport is gone. This leaves the
initiator reset/reconnect hanging as it waits for a status on the
Create_Association LS for the reconnect.

Change the filter in the LS callback path. If tport null (set when
failed validation before "sending to remote port"), be sure to call
done. This was the main bug. But, continue the logic that only calls
done if tport was set but there is no remoteport (e.g. case where
remoteport has been removed, thus host doesn't expect a completion).

Signed-off-by: James Smart <james.smart@broadcom.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/nvme/target/fcloop.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/nvme/target/fcloop.c b/drivers/nvme/target/fcloop.c
index c0080f6ab2f5..0b0a4825b3eb 100644
--- a/drivers/nvme/target/fcloop.c
+++ b/drivers/nvme/target/fcloop.c
@@ -300,7 +300,7 @@ fcloop_tgt_lsrqst_done_work(struct work_struct *work)
 	struct fcloop_tport *tport = tls_req->tport;
 	struct nvmefc_ls_req *lsreq = tls_req->lsreq;
 
-	if (tport->remoteport)
+	if (!tport || tport->remoteport)
 		lsreq->done(lsreq, tls_req->status);
 }
 
@@ -318,6 +318,7 @@ fcloop_ls_req(struct nvme_fc_local_port *localport,
 
 	if (!rport->targetport) {
 		tls_req->status = -ECONNREFUSED;
+		tls_req->tport = NULL;
 		schedule_work(&tls_req->work);
 		return ret;
 	}
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 21/25] ARM: dts: omap4-droid4: Fix emmc errors seen on some devices
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (18 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 20/25] nvme-fcloop: Fix dropped LS's to removed target port Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 22/25] arm/arm64: smccc-1.1: Make return values unsigned long Sasha Levin
                   ` (3 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: Tony Lindgren, Marcel Partap, Merlijn Wajer, Michael Scott,
	NeKit, Pavel Machek, Sebastian Reichel, Sasha Levin

From: Tony Lindgren <tony@atomide.com>

[ Upstream commit 2d59bb602314a4b2593fde267734266b5e872dd0 ]

Otherwise we can get the following errors occasionally on some devices:

mmc1: tried to HW reset card, got error -110
mmcblk1: error -110 requesting status
mmcblk1: recovery failed!
print_req_error: I/O error, dev mmcblk1, sector 14329
...

I have one device that hits this error almost on every boot, and another
one that hits it only rarely with the other ones I've used behave without
problems. I'm not sure if the issue is related to a particular eMMC card
model, but in case it is, both of the machines with issues have:

# cat /sys/class/mmc_host/mmc1/mmc1:0001/manfid \
/sys/class/mmc_host/mmc1/mmc1:0001/oemid \
/sys/class/mmc_host/mmc1/mmc1:0001/name
0x000045
0x0100
SEM16G

and the working ones have:

0x000011
0x0100
016G92

Note that "ti,non-removable" is different as omap_hsmmc_reg_get() does not
call omap_hsmmc_disable_boot_regulators() if no_regulator_off_init is set.
And currently we set no_regulator_off_init only for "ti,non-removable" and
not for "non-removable". It seems that we should have "non-removable" with
some other mmc generic property behave in the same way instead of having to
use a non-generic property. But let's fix the issue first.

Fixes: 7e2f8c0ae670 ("ARM: dts: Add minimal support for motorola droid 4
xt894")
Cc: Marcel Partap <mpartap@gmx.net>
Cc: Merlijn Wajer <merlijn@wizzup.org>
Cc: Michael Scott <hashcode0f@gmail.com>
Cc: NeKit <nekit1000@gmail.com>
Cc: Pavel Machek <pavel@ucw.cz>
Cc: Sebastian Reichel <sre@kernel.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 arch/arm/boot/dts/omap4-droid4-xt894.dts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm/boot/dts/omap4-droid4-xt894.dts b/arch/arm/boot/dts/omap4-droid4-xt894.dts
index 8b93d37310f2..bad690b23081 100644
--- a/arch/arm/boot/dts/omap4-droid4-xt894.dts
+++ b/arch/arm/boot/dts/omap4-droid4-xt894.dts
@@ -314,7 +314,7 @@
 &mmc2 {
 	vmmc-supply = <&vsdio>;
 	bus-width = <8>;
-	non-removable;
+	ti,non-removable;
 };
 
 &mmc3 {
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 22/25] arm/arm64: smccc-1.1: Make return values unsigned long
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (19 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 21/25] ARM: dts: omap4-droid4: Fix emmc errors seen on some devices Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 24/25] i2c: i801: Allow ACPI AML access I/O ports not reserved for SMBus Sasha Levin
                   ` (2 subsequent siblings)
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Marc Zyngier, Will Deacon, Sasha Levin

From: Marc Zyngier <marc.zyngier@arm.com>

[ Upstream commit 1d8f574708a3fb6f18c85486d0c5217df893c0cf ]

An unfortunate consequence of having a strong typing for the input
values to the SMC call is that it also affects the type of the
return values, limiting r0 to 32 bits and r{1,2,3} to whatever
was passed as an input.

Let's turn everything into "unsigned long", which satisfies the
requirements of both architectures, and allows for the full
range of return values.

Reported-by: Julien Grall <julien.grall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 include/linux/arm-smccc.h | 20 ++++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/include/linux/arm-smccc.h b/include/linux/arm-smccc.h
index ca1d2cc2cdfa..5a91ff33720b 100644
--- a/include/linux/arm-smccc.h
+++ b/include/linux/arm-smccc.h
@@ -199,31 +199,31 @@ asmlinkage void __arm_smccc_hvc(unsigned long a0, unsigned long a1,
 
 #define __declare_arg_0(a0, res)					\
 	struct arm_smccc_res   *___res = res;				\
-	register u32           r0 asm("r0") = a0;			\
+	register unsigned long r0 asm("r0") = (u32)a0;			\
 	register unsigned long r1 asm("r1");				\
 	register unsigned long r2 asm("r2");				\
 	register unsigned long r3 asm("r3")
 
 #define __declare_arg_1(a0, a1, res)					\
 	struct arm_smccc_res   *___res = res;				\
-	register u32           r0 asm("r0") = a0;			\
-	register typeof(a1)    r1 asm("r1") = a1;			\
+	register unsigned long r0 asm("r0") = (u32)a0;			\
+	register unsigned long r1 asm("r1") = a1;			\
 	register unsigned long r2 asm("r2");				\
 	register unsigned long r3 asm("r3")
 
 #define __declare_arg_2(a0, a1, a2, res)				\
 	struct arm_smccc_res   *___res = res;				\
-	register u32           r0 asm("r0") = a0;			\
-	register typeof(a1)    r1 asm("r1") = a1;			\
-	register typeof(a2)    r2 asm("r2") = a2;			\
+	register unsigned long r0 asm("r0") = (u32)a0;			\
+	register unsigned long r1 asm("r1") = a1;			\
+	register unsigned long r2 asm("r2") = a2;			\
 	register unsigned long r3 asm("r3")
 
 #define __declare_arg_3(a0, a1, a2, a3, res)				\
 	struct arm_smccc_res   *___res = res;				\
-	register u32           r0 asm("r0") = a0;			\
-	register typeof(a1)    r1 asm("r1") = a1;			\
-	register typeof(a2)    r2 asm("r2") = a2;			\
-	register typeof(a3)    r3 asm("r3") = a3
+	register unsigned long r0 asm("r0") = (u32)a0;			\
+	register unsigned long r1 asm("r1") = a1;			\
+	register unsigned long r2 asm("r2") = a2;			\
+	register unsigned long r3 asm("r3") = a3
 
 #define __declare_arg_4(a0, a1, a2, a3, a4, res)			\
 	__declare_arg_3(a0, a1, a2, a3, res);				\
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 23/25] arm/arm64: smccc-1.1: Handle function result as parameters
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (21 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 24/25] i2c: i801: Allow ACPI AML access I/O ports not reserved for SMBus Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 25/25] x86/pti: Fix section mismatch warning/error Sasha Levin
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Marc Zyngier, Will Deacon, Sasha Levin

From: Marc Zyngier <marc.zyngier@arm.com>

[ Upstream commit 755a8bf5579d22eb5636685c516d8dede799e27b ]

If someone has the silly idea to write something along those lines:

	extern u64 foo(void);

	void bar(struct arm_smccc_res *res)
	{
		arm_smccc_1_1_smc(0xbad, foo(), res);
	}

they are in for a surprise, as this gets compiled as:

	0000000000000588 <bar>:
	 588:   a9be7bfd        stp     x29, x30, [sp, #-32]!
	 58c:   910003fd        mov     x29, sp
	 590:   f9000bf3        str     x19, [sp, #16]
	 594:   aa0003f3        mov     x19, x0
	 598:   aa1e03e0        mov     x0, x30
	 59c:   94000000        bl      0 <_mcount>
	 5a0:   94000000        bl      0 <foo>
	 5a4:   aa0003e1        mov     x1, x0
	 5a8:   d4000003        smc     #0x0
	 5ac:   b4000073        cbz     x19, 5b8 <bar+0x30>
	 5b0:   a9000660        stp     x0, x1, [x19]
	 5b4:   a9010e62        stp     x2, x3, [x19, #16]
	 5b8:   f9400bf3        ldr     x19, [sp, #16]
	 5bc:   a8c27bfd        ldp     x29, x30, [sp], #32
	 5c0:   d65f03c0        ret
	 5c4:   d503201f        nop

The call to foo "overwrites" the x0 register for the return value,
and we end up calling the wrong secure service.

A solution is to evaluate all the parameters before assigning
anything to specific registers, leading to the expected result:

	0000000000000588 <bar>:
	 588:   a9be7bfd        stp     x29, x30, [sp, #-32]!
	 58c:   910003fd        mov     x29, sp
	 590:   f9000bf3        str     x19, [sp, #16]
	 594:   aa0003f3        mov     x19, x0
	 598:   aa1e03e0        mov     x0, x30
	 59c:   94000000        bl      0 <_mcount>
	 5a0:   94000000        bl      0 <foo>
	 5a4:   aa0003e1        mov     x1, x0
	 5a8:   d28175a0        mov     x0, #0xbad
	 5ac:   d4000003        smc     #0x0
	 5b0:   b4000073        cbz     x19, 5bc <bar+0x34>
	 5b4:   a9000660        stp     x0, x1, [x19]
	 5b8:   a9010e62        stp     x2, x3, [x19, #16]
	 5bc:   f9400bf3        ldr     x19, [sp, #16]
	 5c0:   a8c27bfd        ldp     x29, x30, [sp], #32
	 5c4:   d65f03c0        ret

Reported-by: Julien Grall <julien.grall@arm.com>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 include/linux/arm-smccc.h | 30 ++++++++++++++++++++----------
 1 file changed, 20 insertions(+), 10 deletions(-)

diff --git a/include/linux/arm-smccc.h b/include/linux/arm-smccc.h
index 5a91ff33720b..18863d56273c 100644
--- a/include/linux/arm-smccc.h
+++ b/include/linux/arm-smccc.h
@@ -205,41 +205,51 @@ asmlinkage void __arm_smccc_hvc(unsigned long a0, unsigned long a1,
 	register unsigned long r3 asm("r3")
 
 #define __declare_arg_1(a0, a1, res)					\
+	typeof(a1) __a1 = a1;						\
 	struct arm_smccc_res   *___res = res;				\
 	register unsigned long r0 asm("r0") = (u32)a0;			\
-	register unsigned long r1 asm("r1") = a1;			\
+	register unsigned long r1 asm("r1") = __a1;			\
 	register unsigned long r2 asm("r2");				\
 	register unsigned long r3 asm("r3")
 
 #define __declare_arg_2(a0, a1, a2, res)				\
+	typeof(a1) __a1 = a1;						\
+	typeof(a2) __a2 = a2;						\
 	struct arm_smccc_res   *___res = res;				\
 	register unsigned long r0 asm("r0") = (u32)a0;			\
-	register unsigned long r1 asm("r1") = a1;			\
-	register unsigned long r2 asm("r2") = a2;			\
+	register unsigned long r1 asm("r1") = __a1;			\
+	register unsigned long r2 asm("r2") = __a2;			\
 	register unsigned long r3 asm("r3")
 
 #define __declare_arg_3(a0, a1, a2, a3, res)				\
+	typeof(a1) __a1 = a1;						\
+	typeof(a2) __a2 = a2;						\
+	typeof(a3) __a3 = a3;						\
 	struct arm_smccc_res   *___res = res;				\
 	register unsigned long r0 asm("r0") = (u32)a0;			\
-	register unsigned long r1 asm("r1") = a1;			\
-	register unsigned long r2 asm("r2") = a2;			\
-	register unsigned long r3 asm("r3") = a3
+	register unsigned long r1 asm("r1") = __a1;			\
+	register unsigned long r2 asm("r2") = __a2;			\
+	register unsigned long r3 asm("r3") = __a3
 
 #define __declare_arg_4(a0, a1, a2, a3, a4, res)			\
+	typeof(a4) __a4 = a4;						\
 	__declare_arg_3(a0, a1, a2, a3, res);				\
-	register typeof(a4) r4 asm("r4") = a4
+	register unsigned long r4 asm("r4") = __a4
 
 #define __declare_arg_5(a0, a1, a2, a3, a4, a5, res)			\
+	typeof(a5) __a5 = a5;						\
 	__declare_arg_4(a0, a1, a2, a3, a4, res);			\
-	register typeof(a5) r5 asm("r5") = a5
+	register unsigned long r5 asm("r5") = __a5
 
 #define __declare_arg_6(a0, a1, a2, a3, a4, a5, a6, res)		\
+	typeof(a6) __a6 = a6;						\
 	__declare_arg_5(a0, a1, a2, a3, a4, a5, res);			\
-	register typeof(a6) r6 asm("r6") = a6
+	register unsigned long r6 asm("r6") = __a6
 
 #define __declare_arg_7(a0, a1, a2, a3, a4, a5, a6, a7, res)		\
+	typeof(a7) __a7 = a7;						\
 	__declare_arg_6(a0, a1, a2, a3, a4, a5, a6, res);		\
-	register typeof(a7) r7 asm("r7") = a7
+	register unsigned long r7 asm("r7") = __a7
 
 #define ___declare_args(count, ...) __declare_arg_ ## count(__VA_ARGS__)
 #define __declare_args(count, ...)  ___declare_args(count, __VA_ARGS__)
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 24/25] i2c: i801: Allow ACPI AML access I/O ports not reserved for SMBus
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (20 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 22/25] arm/arm64: smccc-1.1: Make return values unsigned long Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 23/25] arm/arm64: smccc-1.1: Handle function result as parameters Sasha Levin
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 25/25] x86/pti: Fix section mismatch warning/error Sasha Levin
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel; +Cc: Mika Westerberg, Wolfram Sang, Sasha Levin

From: Mika Westerberg <mika.westerberg@linux.intel.com>

[ Upstream commit 7fd6d98b89f382d414e1db528e29a67bbd749457 ]

Commit 7ae81952cda ("i2c: i801: Allow ACPI SystemIO OpRegion to conflict
with PCI BAR") made it possible for AML code to access SMBus I/O ports
by installing custom SystemIO OpRegion handler and blocking i80i driver
access upon first AML read/write to this OpRegion.

However, while ThinkPad T560 does have SystemIO OpRegion declared under
the SMBus device, it does not access any of the SMBus registers:

    Device (SMBU)
    {
        ...

        OperationRegion (SMBP, PCI_Config, 0x50, 0x04)
        Field (SMBP, DWordAcc, NoLock, Preserve)
        {
            ,   5,
            TCOB,   11,
            Offset (0x04)
        }

        Name (TCBV, 0x00)
        Method (TCBS, 0, NotSerialized)
        {
            If ((TCBV == 0x00))
            {
            TCBV = (\_SB.PCI0.SMBU.TCOB << 0x05)
            }

            Return (TCBV) /* \_SB_.PCI0.SMBU.TCBV */
        }

        OperationRegion (TCBA, SystemIO, TCBS (), 0x10)
        Field (TCBA, ByteAcc, NoLock, Preserve)
        {
            Offset (0x04),
            ,   9,
            CPSC,   1
        }
    }

Problem with the current approach is that it blocks all I/O port access
and because this system has touchpad connected to the SMBus controller
after first AML access (happens during suspend/resume cycle) the
touchpad fails to work anymore.

Fix this so that we allow ACPI AML I/O port access if it does not touch
the region reserved for the SMBus.

Fixes: 7ae81952cda ("i2c: i801: Allow ACPI SystemIO OpRegion to conflict with PCI BAR")
Link: https://bugzilla.kernel.org/show_bug.cgi?id=200737
Reported-by: Yussuf Khalil <dev@pp3345.net>
Signed-off-by: Mika Westerberg <mika.westerberg@linux.intel.com>
Reviewed-by: Jean Delvare <jdelvare@suse.de>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 drivers/i2c/busses/i2c-i801.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c
index ba8df2fde1b2..ec10aa1655f7 100644
--- a/drivers/i2c/busses/i2c-i801.c
+++ b/drivers/i2c/busses/i2c-i801.c
@@ -1411,6 +1411,13 @@ static void i801_add_tco(struct i801_priv *priv)
 }
 
 #ifdef CONFIG_ACPI
+static bool i801_acpi_is_smbus_ioport(const struct i801_priv *priv,
+				      acpi_physical_address address)
+{
+	return address >= priv->smba &&
+	       address <= pci_resource_end(priv->pci_dev, SMBBAR);
+}
+
 static acpi_status
 i801_acpi_io_handler(u32 function, acpi_physical_address address, u32 bits,
 		     u64 *value, void *handler_context, void *region_context)
@@ -1426,7 +1433,7 @@ i801_acpi_io_handler(u32 function, acpi_physical_address address, u32 bits,
 	 */
 	mutex_lock(&priv->acpi_lock);
 
-	if (!priv->acpi_reserved) {
+	if (!priv->acpi_reserved && i801_acpi_is_smbus_ioport(priv, address)) {
 		priv->acpi_reserved = true;
 
 		dev_warn(&pdev->dev, "BIOS is accessing SMBus registers\n");
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

* [PATCH AUTOSEL 4.14 25/25] x86/pti: Fix section mismatch warning/error
  2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
                   ` (22 preceding siblings ...)
  2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 23/25] arm/arm64: smccc-1.1: Handle function result as parameters Sasha Levin
@ 2018-09-20  2:48 ` Sasha Levin
  23 siblings, 0 replies; 25+ messages in thread
From: Sasha Levin @ 2018-09-20  2:48 UTC (permalink / raw)
  To: stable, linux-kernel
  Cc: Randy Dunlap, Thomas Gleixner, Andy Lutomirski, Sasha Levin

From: Randy Dunlap <rdunlap@infradead.org>

[ Upstream commit ff924c5a1ec7548825cc2d07980b03be4224ffac ]

Fix the section mismatch warning in arch/x86/mm/pti.c:

WARNING: vmlinux.o(.text+0x6972a): Section mismatch in reference from the function pti_clone_pgtable() to the function .init.text:pti_user_pagetable_walk_pte()
The function pti_clone_pgtable() references
the function __init pti_user_pagetable_walk_pte().
This is often because pti_clone_pgtable lacks a __init
annotation or the annotation of pti_user_pagetable_walk_pte is wrong.
FATAL: modpost: Section mismatches detected.

Fixes: 85900ea51577 ("x86/pti: Map the vsyscall page if needed")
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Andy Lutomirski <luto@kernel.org>
Link: https://lkml.kernel.org/r/43a6d6a3-d69d-5eda-da09-0b1c88215a2a@infradead.org
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 arch/x86/mm/pti.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/x86/mm/pti.c b/arch/x86/mm/pti.c
index d6f11accd37a..3fa048a4eb6f 100644
--- a/arch/x86/mm/pti.c
+++ b/arch/x86/mm/pti.c
@@ -220,7 +220,7 @@ static __init pmd_t *pti_user_pagetable_walk_pmd(unsigned long address)
  *
  * Returns a pointer to a PTE on success, or NULL on failure.
  */
-static __init pte_t *pti_user_pagetable_walk_pte(unsigned long address)
+static pte_t *pti_user_pagetable_walk_pte(unsigned long address)
 {
 	gfp_t gfp = (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO);
 	pmd_t *pmd = pti_user_pagetable_walk_pmd(address);
-- 
2.17.1

^ permalink raw reply related	[flat|nested] 25+ messages in thread

end of thread, other threads:[~2018-09-20  2:55 UTC | newest]

Thread overview: 25+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-09-20  2:48 [PATCH AUTOSEL 4.14 01/25] qed: Wait for ready indication before rereading the shmem Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 02/25] qed: Wait for MCP halt and resume commands to take place Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 03/25] qed: Prevent a possible deadlock during driver load and unload Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 05/25] thermal: of-thermal: disable passive polling when thermal zone is disabled Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 04/25] qed: Avoid sending mailbox commands when MFW is not responsive Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 06/25] isofs: reject hardware sector size > 2048 bytes Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 07/25] tls: possible hang when do_tcp_sendpages hits sndbuf is full case Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 08/25] bpf: sockmap: write_space events need to be passed to TCP handler Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 09/25] net: hns: fix length and page_offset overflow when CONFIG_ARM64_64K_PAGES Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 10/25] net: hns: fix skb->truesize underestimation Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 11/25] e1000: check on netif_running() before calling e1000_up() Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 12/25] e1000: ensure to free old tx/rx rings in set_ringparam() Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 13/25] crypto: cavium/nitrox - fix for command corruption in queue full case with backlog submissions Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 14/25] hwmon: (ina2xx) fix sysfs shunt resistor read access Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 15/25] hwmon: (adt7475) Make adt7475_read_word() return errors Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 16/25] Revert "ARM: dts: imx7d: Invert legacy PCI irq mapping" Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 17/25] drm/amdgpu: Enable/disable gfx PG feature in rlc safe mode Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 18/25] drm/amdgpu: Update power state at the end of smu hw_init Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 19/25] ata: ftide010: Add a quirk for SQ201 Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 20/25] nvme-fcloop: Fix dropped LS's to removed target port Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 21/25] ARM: dts: omap4-droid4: Fix emmc errors seen on some devices Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 22/25] arm/arm64: smccc-1.1: Make return values unsigned long Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 24/25] i2c: i801: Allow ACPI AML access I/O ports not reserved for SMBus Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 23/25] arm/arm64: smccc-1.1: Handle function result as parameters Sasha Levin
2018-09-20  2:48 ` [PATCH AUTOSEL 4.14 25/25] x86/pti: Fix section mismatch warning/error Sasha Levin

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).