Skip to contentSkip to conversation

scsi: lpfc: Fix spinlock_irq issues in lpfc_els_flush_cmd()

PATCH In review Sasha Levin 2 participants 138 messages Dec 11, 2019 – Dec 14, 2019

Subsystems (26)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 1/134 scsi: lpfc: Fix spinlock_irq issues in lpfc_els_flush_cmd()

From: James Smart <[email protected]>

[ Upstream commit d38b4a527fe898f859f74a3a43d4308f48ac7855 ]

While reviewing the CT behavior, issues with spinlock_irq were seen. The driver should be using spinlock_irqsave/irqrestore in the els flush routine.

Changed to spinlock_irqsave/irqrestore.

---
 drivers/scsi/lpfc/lpfc_els.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)
drivers/scsi/lpfc/lpfc_els.c +9 -7
index d5303994bfd62..0052b341587d9 100644
@@ -7986,20 +7986,22 @@ lpfc_els_flush_cmd(struct lpfc_vport *vport)
 	struct lpfc_sli_ring *pring;
 	struct lpfc_iocbq *tmp_iocb, *piocb;
 	IOCB_t *cmd = NULL;
+	unsigned long iflags = 0;
 
 	lpfc_fabric_abort_vport(vport);
+
 	/*
 	 * For SLI3, only the hbalock is required.  But SLI4 needs to coordinate
 	 * with the ring insert operation.  Because lpfc_sli_issue_abort_iotag
 	 * ultimately grabs the ring_lock, the driver must splice the list into
 	 * a working list and release the locks before calling the abort.
 	 */
-	spin_lock_irq(&phba->hbalock);
+	spin_lock_irqsave(&phba->hbalock, iflags);
 	pring = lpfc_phba_elsring(phba);
 
 	/* Bail out if we've no ELS wq, like in PCI error recovery case. */
 	if (unlikely(!pring)) {
-		spin_unlock_irq(&phba->hbalock);
+		spin_unlock_irqrestore(&phba->hbalock, iflags);
 		return;
 	}
 
@@ -8037,21 +8039,21 @@ lpfc_els_flush_cmd(struct lpfc_vport *vport)
 
 	if (phba->sli_rev == LPFC_SLI_REV4)
 		spin_unlock(&pring->ring_lock);
-	spin_unlock_irq(&phba->hbalock);
+	spin_unlock_irqrestore(&phba->hbalock, iflags);
 
 	/* Abort each txcmpl iocb on aborted list and remove the dlist links. */
 	list_for_each_entry_safe(piocb, tmp_iocb, &abort_list, dlist) {
-		spin_lock_irq(&phba->hbalock);
+		spin_lock_irqsave(&phba->hbalock, iflags);
 		list_del_init(&piocb->dlist);
 		lpfc_sli_issue_abort_iotag(phba, pring, piocb);
-		spin_unlock_irq(&phba->hbalock);
+		spin_unlock_irqrestore(&phba->hbalock, iflags);
 	}
 	if (!list_empty(&abort_list))
 		lpfc_printf_vlog(vport, KERN_ERR, LOG_ELS,
 				 "3387 abort list for txq not empty\n");
 	INIT_LIST_HEAD(&abort_list);
 
-	spin_lock_irq(&phba->hbalock);
+	spin_lock_irqsave(&phba->hbalock, iflags);
 	if (phba->sli_rev == LPFC_SLI_REV4)
 		spin_lock(&pring->ring_lock);
 
@@ -8091,7 +8093,7 @@ lpfc_els_flush_cmd(struct lpfc_vport *vport)
 
 	if (phba->sli_rev == LPFC_SLI_REV4)
 		spin_unlock(&pring->ring_lock);
-	spin_unlock_irq(&phba->hbalock);
+	spin_unlock_irqrestore(&phba->hbalock, iflags);
 
 	/* Cancel all the IOCBs from the completions list */
 	lpfc_sli_cancel_iocbs(phba, &abort_list,
Signature · 1 line
2.20.1
author maintainer
#1 lore raw

PATCH 2/134 scsi: lpfc: Fix discovery failures when target device connectivity bounces

From: James Smart <[email protected]>

[ Upstream commit 3f97aed6117c7677eb16756c4ec8b86000fd5822 ]

An issue was seen discovering all SCSI Luns when a target device undergoes link bounce.

The driver currently does not qualify the FC4 support on the target. Therefore it will send a SCSI PRLI and an NVMe PRLI. The expectation is that the target will reject the PRLI if it is not supported. If a PRLI times out, the driver will retry. The driver will not proceed with the device until both SCSI and NVMe PRLIs are resolved. In the failure case, the device is FCP only and does not respond to the NVMe PRLI, thus initiating the wait/retry loop in the driver. During that time, a RSCN is received (device bounced) causing the driver to issue a GID_FT. The GID_FT response comes back before the PRLI mess is resolved and it prematurely cancels the PRLI retry logic and leaves the device in a STE_PRLI_ISSUE state. Discovery with the target never completes or resets.

Fix by resetting the node state back to STE_NPR_NODE when GID_FT completes, thereby restarting the discovery process for the node.

---
 drivers/scsi/lpfc/lpfc_hbadisc.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/scsi/lpfc/lpfc_hbadisc.c +6 -1
index 749286acdc173..f7c205e1da485 100644
@@ -5405,9 +5405,14 @@ lpfc_setup_disc_node(struct lpfc_vport *vport, uint32_t did)
 			/* If we've already received a PLOGI from this NPort
 			 * we don't need to try to discover it again.
 			 */
-			if (ndlp->nlp_flag & NLP_RCV_PLOGI)
+			if (ndlp->nlp_flag & NLP_RCV_PLOGI &&
+			    !(ndlp->nlp_type &
+			     (NLP_FCP_TARGET | NLP_NVME_TARGET)))
 				return NULL;
 
+			ndlp->nlp_prev_state = ndlp->nlp_state;
+			lpfc_nlp_set_state(vport, ndlp, NLP_STE_NPR_NODE);
+
 			spin_lock_irq(shost->host_lock);
 			ndlp->nlp_flag |= NLP_NPR_2B_DISC;
 			spin_unlock_irq(shost->host_lock);
Signature · 1 line
2.20.1
author maintainer
#2 lore raw

PATCH 3/134 scsi: mpt3sas: Fix clear pending bit in ioctl status

In reply to Sasha Levin

From: Sreekanth Reddy <[email protected]>

[ Upstream commit 782b281883caf70289ba6a186af29441a117d23e ]

When user issues diag register command from application with required size, and if driver unable to allocate the memory, then it will fail the register command. While failing the register command, driver is not currently clearing MPT3_CMD_PENDING bit in ctl_cmds.status variable which was set before trying to allocate the memory. As this bit is set, subsequent register command will be failed with BUSY status even when user wants to register the trace buffer will less memory.

Clear MPT3_CMD_PENDING bit in ctl_cmds.status before returning the diag register command with no memory status.

---
 drivers/scsi/mpt3sas/mpt3sas_ctl.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/mpt3sas/mpt3sas_ctl.c +2 -1
index 7d696952b3763..3c463e8f60740 100644
@@ -1584,7 +1584,8 @@ _ctl_diag_register_2(struct MPT3SAS_ADAPTER *ioc,
 			ioc_err(ioc, "%s: failed allocating memory for diag buffers, requested size(%d)\n",
 				__func__, request_data_sz);
 			mpt3sas_base_free_smid(ioc, smid);
-			return -ENOMEM;
+			rc = -ENOMEM;
+			goto out;
 		}
 		ioc->diag_buffer[buffer_type] = request_data;
 		ioc->diag_buffer_sz[buffer_type] = request_data_sz;
Signature · 1 line
2.20.1
author maintainer
#3 lore raw

PATCH 4/134 scsi: lpfc: Fix locking on mailbox command completion

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit 07b8582430370097238b589f4e24da7613ca6dd3 ]

Symptoms were seen of the driver not having valid data for mailbox commands. After debugging, the following sequence was found:

The driver maintains a port-wide pointer of the mailbox command that is currently in execution. Once finished, the port-wide pointer is cleared (done in lpfc_sli4_mq_release()). The next mailbox command issued will set the next pointer and so on.

The mailbox response data is only copied if there is a valid port-wide pointer.

In the failing case, it was seen that a new mailbox command was being attempted in parallel with the completion. The parallel path was seeing the mailbox no long in use (flag check under lock) and thus set the port pointer. The completion path had cleared the active flag under lock, but had not touched the port pointer. The port pointer is cleared after the lock is released. In this case, the completion path cleared the just-set value by the parallel path.

Fix by making the calls that clear mbox state/port pointer while under lock. Also slightly cleaned up the error path.

---
 drivers/scsi/lpfc/lpfc_sli.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/scsi/lpfc/lpfc_sli.c +7 -1
index 614f78dddafe0..e5413d52e49a2 100644
@@ -13158,13 +13158,19 @@ send_current_mbox:
 	phba->sli.sli_flag &= ~LPFC_SLI_MBOX_ACTIVE;
 	/* Setting active mailbox pointer need to be in sync to flag clear */
 	phba->sli.mbox_active = NULL;
+	if (bf_get(lpfc_trailer_consumed, mcqe))
+		lpfc_sli4_mq_release(phba->sli4_hba.mbx_wq);
 	spin_unlock_irqrestore(&phba->hbalock, iflags);
 	/* Wake up worker thread to post the next pending mailbox command */
 	lpfc_worker_wake_up(phba);
+	return workposted;
+
 out_no_mqe_complete:
+	spin_lock_irqsave(&phba->hbalock, iflags);
 	if (bf_get(lpfc_trailer_consumed, mcqe))
 		lpfc_sli4_mq_release(phba->sli4_hba.mbx_wq);
-	return workposted;
+	spin_unlock_irqrestore(&phba->hbalock, iflags);
+	return false;
 }
 
 /**
Signature · 1 line
2.20.1
author maintainer
#4 lore raw

PATCH 5/134 scsi: mpt3sas: Reject NVMe Encap cmnds to unsupported HBA

In reply to Sasha Levin

From: Sreekanth Reddy <[email protected]>

[ Upstream commit 77fd4f2c88bf83205a21f9ca49fdcc0c7868dba9 ]

If any faulty application issues an NVMe Encapsulated commands to HBA which doesn't support NVMe protocol then driver should return the command as invalid with the following message.

"HBA doesn't support NVMe. Rejecting NVMe Encapsulated request."

Otherwise below page fault kernel panic will be observed while building the PRPs as there is no PRP pools allocated for the HBA which doesn't support NVMe drives.

RIP: 0010:_base_build_nvme_prp+0x3b/0xf0 [mpt3sas]
Call Trace:
 _ctl_do_mpt_command+0x931/0x1120 [mpt3sas]
 _ctl_ioctl_main.isra.11+0xa28/0x11e0 [mpt3sas]
 ? prepare_to_wait+0xb0/0xb0
 ? tty_ldisc_deref+0x16/0x20
 _ctl_ioctl+0x1a/0x20 [mpt3sas]
 do_vfs_ioctl+0xaa/0x620
 ? vfs_read+0x117/0x140
 ksys_ioctl+0x67/0x90
 __x64_sys_ioctl+0x1a/0x20
 do_syscall_64+0x60/0x190
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

[mkp: tweaked error string]

---
 drivers/scsi/mpt3sas/mpt3sas_ctl.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)
drivers/scsi/mpt3sas/mpt3sas_ctl.c +12 -0
index 3c463e8f60740..b95f7d062ea44 100644
@@ -778,6 +778,18 @@ _ctl_do_mpt_command(struct MPT3SAS_ADAPTER *ioc, struct mpt3_ioctl_command karg,
 	case MPI2_FUNCTION_NVME_ENCAPSULATED:
 	{
 		nvme_encap_request = (Mpi26NVMeEncapsulatedRequest_t *)request;
+		if (!ioc->pcie_sg_lookup) {
+			dtmprintk(ioc, ioc_info(ioc,
+			    "HBA doesn't support NVMe. Rejecting NVMe Encapsulated request.\n"
+			    ));
+
+			if (ioc->logging_level & MPT_DEBUG_TM)
+				_debug_dump_mf(nvme_encap_request,
+				    ioc->request_sz/4);
+			mpt3sas_base_free_smid(ioc, smid);
+			ret = -EINVAL;
+			goto out;
+		}
 		/*
 		 * Get the Physical Address of the sense buffer.
 		 * Use Error Response buffer address field to hold the sense
Signature · 1 line
2.20.1
author maintainer
#5 lore raw

PATCH 6/134 gpio: mxc: Only get the second IRQ when there is more than one IRQ

In reply to Sasha Levin

From: Anson Huang <[email protected]>

[ Upstream commit c8f3d144004dd3f471ffd414690d15a005e4acd6 ]

On some of i.MX SoCs like i.MX8QXP, there is ONLY one IRQ for each GPIO bank, so it is better to check the IRQ count before getting second IRQ to avoid below error message during probe:

[    1.070908] gpio-mxc 5d080000.gpio: IRQ index 1 not found
[    1.077420] gpio-mxc 5d090000.gpio: IRQ index 1 not found
[    1.083766] gpio-mxc 5d0a0000.gpio: IRQ index 1 not found
[    1.090122] gpio-mxc 5d0b0000.gpio: IRQ index 1 not found
[    1.096470] gpio-mxc 5d0c0000.gpio: IRQ index 1 not found
[    1.102804] gpio-mxc 5d0d0000.gpio: IRQ index 1 not found
[    1.109144] gpio-mxc 5d0e0000.gpio: IRQ index 1 not found
[    1.115475] gpio-mxc 5d0f0000.gpio: IRQ index 1 not found
---
 drivers/gpio/gpio-mxc.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)
drivers/gpio/gpio-mxc.c +10 -3
index 7907a87558662..c77d474185f31 100644
@@ -411,6 +411,7 @@ static int mxc_gpio_probe(struct platform_device *pdev)
 {
 	struct device_node *np = pdev->dev.of_node;
 	struct mxc_gpio_port *port;
+	int irq_count;
 	int irq_base;
 	int err;
 
@@ -426,9 +427,15 @@ static int mxc_gpio_probe(struct platform_device *pdev)
 	if (IS_ERR(port->base))
 		return PTR_ERR(port->base);
 
-	port->irq_high = platform_get_irq(pdev, 1);
-	if (port->irq_high < 0)
-		port->irq_high = 0;
+	irq_count = platform_irq_count(pdev);
+	if (irq_count < 0)
+		return irq_count;
+
+	if (irq_count > 1) {
+		port->irq_high = platform_get_irq(pdev, 1);
+		if (port->irq_high < 0)
+			port->irq_high = 0;
+	}
 
 	port->irq = platform_get_irq(pdev, 0);
 	if (port->irq < 0)
Signature · 1 line
2.20.1
author maintainer
#6 lore raw

PATCH 7/134 scsi: lpfc: Fix list corruption in lpfc_sli_get_iocbq

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit 15498dc1a55b7aaea4b51ff03e3ff0f662e73f44 ]

After study, it was determined there was a double free of a CT iocb during execution of lpfc_offline_prep and lpfc_offline. The prep routine issued an abort for some CT iocbs, but the aborts did not complete fast enough for a subsequent routine that waits for completion. Thus the driver proceeded to lpfc_offline, which releases any pending iocbs. Unfortunately, the completions for the aborts were then received which re-released the ct iocbs.

Turns out the issue for why the aborts didn't complete fast enough was not their time on the wire/in the adapter. It was the lpfc_work_done routine, which requires the adapter state to be UP before it calls lpfc_sli_handle_slow_ring_event() to process the completions. The issue is the prep routine takes the link down as part of it's processing.

To fix, the following was performed:

- Prevent the offline routine from releasing iocbs that have had aborts issued on them. Defer to the abort completions. Also means the driver fully waits for the completions. Given this change, the recognition of "driver-generated" status which then releases the iocb is no longer valid. As such, the change made in the commit 296012285c90 is reverted. As recognition of "driver-generated" status is no longer valid, this patch reverts the changes made in
commit 296012285c90 ("scsi: lpfc: Fix leak of ELS completions on adapter reset")

- Modify lpfc_work_done to allow slow path completions so that the abort completions aren't ignored.

- Updated the fdmi path to recognize a CT request that fails due to the port being unusable. This stops FDMI retries. FDMI will be restarted on next link up.

---
 drivers/scsi/lpfc/lpfc_ct.c      | 6 ++++++
 drivers/scsi/lpfc/lpfc_els.c     | 3 +++
 drivers/scsi/lpfc/lpfc_hbadisc.c | 5 ++++-
 drivers/scsi/lpfc/lpfc_sli.c     | 3 ---
 4 files changed, 13 insertions(+), 4 deletions(-)
drivers/scsi/lpfc/lpfc_ct.c +6 -0
index 25e86706e2072..f883fac2d2b1d 100644
@@ -1868,6 +1868,12 @@ lpfc_cmpl_ct_disc_fdmi(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 		if (irsp->ulpStatus == IOSTAT_LOCAL_REJECT) {
 			switch ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK)) {
 			case IOERR_SLI_ABORTED:
+			case IOERR_SLI_DOWN:
+				/* Driver aborted this IO.  No retry as error
+				 * is likely Offline->Online or some adapter
+				 * error.  Recovery will try again.
+				 */
+				break;
 			case IOERR_ABORT_IN_PROGRESS:
 			case IOERR_SEQUENCE_TIMEOUT:
 			case IOERR_ILLEGAL_FRAME:
drivers/scsi/lpfc/lpfc_els.c +3 -0
index 0052b341587d9..f293b48616ae9 100644
@@ -8016,6 +8016,9 @@ lpfc_els_flush_cmd(struct lpfc_vport *vport)
 		if (piocb->vport != vport)
 			continue;
 
+		if (piocb->iocb_flag & LPFC_DRIVER_ABORTED)
+			continue;
+
 		/* On the ELS ring we can have ELS_REQUESTs or
 		 * GEN_REQUESTs waiting for a response.
 		 */
drivers/scsi/lpfc/lpfc_hbadisc.c +4 -1
index f7c205e1da485..1286c658ba34f 100644
@@ -700,7 +700,10 @@ lpfc_work_done(struct lpfc_hba *phba)
 			if (!(phba->hba_flag & HBA_SP_QUEUE_EVT))
 				set_bit(LPFC_DATA_READY, &phba->data_flags);
 		} else {
-			if (phba->link_state >= LPFC_LINK_UP ||
+			/* Driver could have abort request completed in queue
+			 * when link goes down.  Allow for this transition.
+			 */
+			if (phba->link_state >= LPFC_LINK_DOWN ||
 			    phba->link_flag & LS_MDS_LOOPBACK) {
 				pring->flag &= ~LPFC_DEFERRED_RING_EVENT;
 				lpfc_sli_handle_slow_ring_event(phba, pring,
drivers/scsi/lpfc/lpfc_sli.c +0 -3
index e5413d52e49a2..995a2b56a35ee 100644
@@ -11050,9 +11050,6 @@ lpfc_sli_abort_els_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 				irsp->ulpStatus, irsp->un.ulpWord[4]);
 
 		spin_unlock_irq(&phba->hbalock);
-		if (irsp->ulpStatus == IOSTAT_LOCAL_REJECT &&
-		    irsp->un.ulpWord[4] == IOERR_SLI_ABORTED)
-			lpfc_sli_release_iocbq(phba, abort_iocb);
 	}
 release_iocb:
 	lpfc_sli_release_iocbq(phba, cmdiocb);
Signature · 1 line
2.20.1
author maintainer
#7 lore raw

PATCH 8/134 Input: atmel_mxt_ts - disable IRQ across suspend

In reply to Sasha Levin

From: Evan Green <[email protected]>

[ Upstream commit 463fa44eec2fef50d111ed0199cf593235065c04 ]

Across suspend and resume, we are seeing error messages like the following:

atmel_mxt_ts i2c-PRP0001:00: __mxt_read_reg: i2c transfer failed (-121) atmel_mxt_ts i2c-PRP0001:00: Failed to read T44 and T5 (-121)

This occurs because the driver leaves its IRQ enabled. Upon resume, there is an IRQ pending, but the interrupt is serviced before both the driver and the underlying I2C bus have been resumed. This causes EREMOTEIO errors.

Disable the IRQ in suspend, and re-enable it on resume. If there are cases where the driver enters suspend with interrupts disabled, that's a bug we should fix separately.

---
 drivers/input/touchscreen/atmel_mxt_ts.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/input/touchscreen/atmel_mxt_ts.c +4 -0
index 24c4b691b1c99..ae60442efda0d 100644
@@ -3156,6 +3156,8 @@ static int __maybe_unused mxt_suspend(struct device *dev)
 
 	mutex_unlock(&input_dev->mutex);
 
+	disable_irq(data->irq);
+
 	return 0;
 }
 
@@ -3168,6 +3170,8 @@ static int __maybe_unused mxt_resume(struct device *dev)
 	if (!input_dev)
 		return 0;
 
+	enable_irq(data->irq);
+
 	mutex_lock(&input_dev->mutex);
 
 	if (input_dev->users)
Signature · 1 line
2.20.1
author maintainer
#8 lore raw

PATCH 9/134 f2fs: fix to update time in lazytime mode

In reply to Sasha Levin

From: Chao Yu <[email protected]>

[ Upstream commit fe1897eaa6646f5a64a4cee0e6473ed9887d324b ]

generic/018 reports an inconsistent status of atime, the testcase is as below:
- open file with O_SYNC
- write file to construct fraged space
- calc md5 of file
- record {a,c,m}time
- defrag file --- do nothing
- umount & mount
- check {a,c,m}time

The root cause is, as f2fs enables lazytime by default, atime update will dirty vfs inode, rather than dirtying f2fs inode (by set with FI_DIRTY_INODE), so later f2fs_write_inode() called from VFS will fail to update inode page due to our skip:

f2fs_write_inode()

	if (is_inode_flag_set(inode, FI_DIRTY_INODE))
		return 0;

So eventually, after evict(), we lose last atime for ever.

To fix this issue, we need to check whether {a,c,m,cr}time is consistent in between inode cache and inode page, and only skip f2fs_update_inode() if f2fs inode is not dirty and time is consistent as well.

---
 fs/f2fs/f2fs.h  | 23 +++++++++++++++--------
 fs/f2fs/inode.c |  6 +++++-
 2 files changed, 20 insertions(+), 9 deletions(-)
fs/f2fs/f2fs.h +15 -8
index 4024790028aab..f078cd20dab88 100644
@@ -2704,6 +2704,20 @@ static inline void clear_file(struct inode *inode, int type)
 	f2fs_mark_inode_dirty_sync(inode, true);
 }
 
+static inline bool f2fs_is_time_consistent(struct inode *inode)
+{
+	if (!timespec64_equal(F2FS_I(inode)->i_disk_time, &inode->i_atime))
+		return false;
+	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &inode->i_ctime))
+		return false;
+	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 2, &inode->i_mtime))
+		return false;
+	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 3,
+						&F2FS_I(inode)->i_crtime))
+		return false;
+	return true;
+}
+
 static inline bool f2fs_skip_inode_update(struct inode *inode, int dsync)
 {
 	bool ret;
@@ -2721,14 +2735,7 @@ static inline bool f2fs_skip_inode_update(struct inode *inode, int dsync)
 			i_size_read(inode) & ~PAGE_MASK)
 		return false;
 
-	if (!timespec64_equal(F2FS_I(inode)->i_disk_time, &inode->i_atime))
-		return false;
-	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 1, &inode->i_ctime))
-		return false;
-	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 2, &inode->i_mtime))
-		return false;
-	if (!timespec64_equal(F2FS_I(inode)->i_disk_time + 3,
-						&F2FS_I(inode)->i_crtime))
+	if (!f2fs_is_time_consistent(inode))
 		return false;
 
 	down_read(&F2FS_I(inode)->i_sem);
fs/f2fs/inode.c +5 -1
index db4fec30c30df..386ad54c13c3a 100644
@@ -615,7 +615,11 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
 			inode->i_ino == F2FS_META_INO(sbi))
 		return 0;
 
-	if (!is_inode_flag_set(inode, FI_DIRTY_INODE))
+	/*
+	 * atime could be updated without dirtying f2fs inode in lazytime mode
+	 */
+	if (f2fs_is_time_consistent(inode) &&
+		!is_inode_flag_set(inode, FI_DIRTY_INODE))
 		return 0;
 
 	if (!f2fs_is_checkpoint_ready(sbi))
Signature · 8 lines
2.20.1



_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
author maintainer
#9 lore raw

PATCH 10/134 powerpc/papr_scm: Fix an off-by-one check in papr_scm_meta_{get, set}

In reply to Sasha Levin

From: Vaibhav Jain <[email protected]>

[ Upstream commit 612ee81b9461475b5a5612c2e8d71559dd3c7920 ]

A validation check to prevent out of bounds read/write inside functions papr_scm_meta_{get,set}() is off-by-one that prevent reads and writes to the last byte of the label area.

This bug manifests as a failure to probe a dimm when libnvdimm is unable to read the entire config-area as advertised by ND_CMD_GET_CONFIG_SIZE. This usually happens when there are large number of namespaces created in the region backed by the dimm and the label-index spans max possible config-area. An error of the form below usually reported in the kernel logs:

[  255.293912] nvdimm: probe of nmem0 failed with error -22

The patch fixes these validation checks there by letting libnvdimm access the entire config-area.

Fixes 53e80bd042773('powerpc/nvdimm: Add support for multibyte read/write for metadata')
---
 arch/powerpc/platforms/pseries/papr_scm.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/platforms/pseries/papr_scm.c +2 -2
index 61883291defc3..ee07d0718bf1a 100644
@@ -152,7 +152,7 @@ static int papr_scm_meta_get(struct papr_scm_priv *p,
 	int len, read;
 	int64_t ret;
 
-	if ((hdr->in_offset + hdr->in_length) >= p->metadata_size)
+	if ((hdr->in_offset + hdr->in_length) > p->metadata_size)
 		return -EINVAL;
 
 	for (len = hdr->in_length; len; len -= read) {
@@ -206,7 +206,7 @@ static int papr_scm_meta_set(struct papr_scm_priv *p,
 	__be64 data_be;
 	int64_t ret;
 
-	if ((hdr->in_offset + hdr->in_length) >= p->metadata_size)
+	if ((hdr->in_offset + hdr->in_length) > p->metadata_size)
 		return -EINVAL;
 
 	for (len = hdr->in_length; len; len -= wrote) {
Signature · 1 line
2.20.1
author maintainer
#10 lore raw

PATCH 11/134 tools/power/x86/intel-speed-select: Remove warning for unused result

In reply to Sasha Levin

From: Srinivas Pandruvada <[email protected]>

[ Upstream commit abd120e3bdf3dd72ba1ed9ac077a861e0e3dc43a ]

Fix warning for:
isst-config.c: In function ‘set_cpu_online_offline’:
isst-config.c:221:3: warning: ignoring return value of ‘write’, declared with attribute warn_unused_result [-Wunused-result]
write(fd, "1\n", 2);

---
 tools/power/x86/intel-speed-select/isst-config.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
tools/power/x86/intel-speed-select/isst-config.c +6 -3
index 2a9890c8395a4..21fcfe621d3aa 100644
@@ -169,7 +169,7 @@ int get_topo_max_cpus(void)
 static void set_cpu_online_offline(int cpu, int state)
 {
 	char buffer[128];
-	int fd;
+	int fd, ret;
 
 	snprintf(buffer, sizeof(buffer),
 		 "/sys/devices/system/cpu/cpu%d/online", cpu);
@@ -179,9 +179,12 @@ static void set_cpu_online_offline(int cpu, int state)
 		err(-1, "%s open failed", buffer);
 
 	if (state)
-		write(fd, "1\n", 2);
+		ret = write(fd, "1\n", 2);
 	else
-		write(fd, "0\n", 2);
+		ret = write(fd, "0\n", 2);
+
+	if (ret == -1)
+		perror("Online/Offline: Operation failed\n");
 
 	close(fd);
 }
Signature · 1 line
2.20.1
author maintainer
#11 lore raw

PATCH 12/134 platform/x86: peaq-wmi: switch to using polled mode of input devices

In reply to Sasha Levin

From: Dmitry Torokhov <[email protected]>

[ Upstream commit 60d15095336cfb56dce5c7767ed3b8c6c1cf79a3 ]

We have added polled mode to the normal input devices with the intent of retiring input_polled_dev. This converts peaq-wmi driver to use the polling mode of standard input devices and removes dependency on INPUT_POLLDEV.

Because the new polling coded does not allow peeking inside the poller structure to get the poll interval, we change the "debounce" process to operate on the time basis, instead of counting events.

We also fix error handling during initialization, as previously we leaked input device structure when we failed to register it.

---
 drivers/platform/x86/Kconfig    |  1 -
 drivers/platform/x86/peaq-wmi.c | 66 +++++++++++++++++++++------------
 2 files changed, 42 insertions(+), 25 deletions(-)
drivers/platform/x86/Kconfig +0 -1
index ae21d08c65e8d..1cab993205142 100644
@@ -806,7 +806,6 @@ config PEAQ_WMI
 	tristate "PEAQ 2-in-1 WMI hotkey driver"
 	depends on ACPI_WMI
 	depends on INPUT
-	select INPUT_POLLDEV
 	help
 	 Say Y here if you want to support WMI-based hotkeys on PEAQ 2-in-1s.
 
drivers/platform/x86/peaq-wmi.c +42 -24

117 lines not on this page · raw message

Signature · 1 line
2.20.1
author maintainer
#12 lore raw

PATCH 13/134 splice: only read in as much information as there is pipe buffer space

In reply to Sasha Levin

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit 3253d9d093376d62b4a56e609f15d2ec5085ac73 ]

Andreas Grünbacher reports that on the two filesystems that support iomap directio, it's possible for splice() to return -EAGAIN (instead of a short splice) if the pipe being written to has less space available in its pipe buffers than the length supplied by the calling process.

Months ago we fixed splice_direct_to_actor to clamp the length of the read request to the size of the splice pipe. Do the same to do_splice.

Fixes 17614445576b6 ("splice: don't read more than available pipe space")
Reported-by [email protected]
---
 fs/splice.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)
fs/splice.c +11 -3
index 98412721f0562..e509239d7e06a 100644
@@ -945,12 +945,13 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd,
 	WARN_ON_ONCE(pipe->nrbufs != 0);
 
 	while (len) {
+		unsigned int pipe_pages;
 		size_t read_len;
 		loff_t pos = sd->pos, prev_pos = pos;
 
 		/* Don't try to read more the pipe has space for. */
-		read_len = min_t(size_t, len,
-				 (pipe->buffers - pipe->nrbufs) << PAGE_SHIFT);
+		pipe_pages = pipe->buffers - pipe->nrbufs;
+		read_len = min(len, (size_t)pipe_pages << PAGE_SHIFT);
 		ret = do_splice_to(in, &pos, pipe, read_len, flags);
 		if (unlikely(ret <= 0))
 			goto out_release;
@@ -1180,8 +1181,15 @@ static long do_splice(struct file *in, loff_t __user *off_in,
 
 		pipe_lock(opipe);
 		ret = wait_for_space(opipe, flags);
-		if (!ret)
+		if (!ret) {
+			unsigned int pipe_pages;
+
+			/* Don't try to read more the pipe has space for. */
+			pipe_pages = opipe->buffers - opipe->nrbufs;
+			len = min(len, (size_t)pipe_pages << PAGE_SHIFT);
+
 			ret = do_splice_to(in, &offset, opipe, len, flags);
+		}
 		pipe_unlock(opipe);
 		if (ret > 0)
 			wakeup_pipe_readers(opipe);
Signature · 1 line
2.20.1
author maintainer
#13 lore raw

PATCH 14/134 iommu: rockchip: Free domain on .domain_free

In reply to Sasha Levin

From: Ezequiel Garcia <[email protected]>

[ Upstream commit 42bb97b80f2e3bf592e3e99d109b67309aa1b30e ]

IOMMU domain resource life is well-defined, managed by .domain_alloc and .domain_free.

Therefore, domain-specific resources shouldn't be tied to the device life, but instead to its domain.

---
 drivers/iommu/rockchip-iommu.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/iommu/rockchip-iommu.c +5 -2
index 4dcbf68dfda43..0df091934361b 100644
@@ -980,13 +980,13 @@ static struct iommu_domain *rk_iommu_domain_alloc(unsigned type)
 	if (!dma_dev)
 		return NULL;
 
-	rk_domain = devm_kzalloc(dma_dev, sizeof(*rk_domain), GFP_KERNEL);
+	rk_domain = kzalloc(sizeof(*rk_domain), GFP_KERNEL);
 	if (!rk_domain)
 		return NULL;
 
 	if (type == IOMMU_DOMAIN_DMA &&
 	    iommu_get_dma_cookie(&rk_domain->domain))
-		return NULL;
+		goto err_free_domain;
 
 	/*
 	 * rk32xx iommus use a 2 level pagetable.
@@ -1021,6 +1021,8 @@ err_free_dt:
 err_put_cookie:
 	if (type == IOMMU_DOMAIN_DMA)
 		iommu_put_dma_cookie(&rk_domain->domain);
+err_free_domain:
+	kfree(rk_domain);
 
 	return NULL;
 }
@@ -1049,6 +1051,7 @@ static void rk_iommu_domain_free(struct iommu_domain *domain)
 
 	if (domain->type == IOMMU_DOMAIN_DMA)
 		iommu_put_dma_cookie(&rk_domain->domain);
+	kfree(rk_domain);
 }
 
 static int rk_iommu_add_device(struct device *dev)
Signature · 1 line
2.20.1
author maintainer
#14 lore raw

PATCH 15/134 iommu/tegra-smmu: Fix page tables in > 4 GiB memory

In reply to Sasha Levin

From: Thierry Reding <[email protected]>

[ Upstream commit 96d3ab802e4930a29a33934373157d6dff1b2c7e ]

Page tables that reside in physical memory beyond the 4 GiB boundary are currently not working properly. The reason is that when the physical address for page directory entries is read, it gets truncated at 32 bits and can cause crashes when passing that address to the DMA API.

Fix this by first casting the PDE value to a dma_addr_t and then using the page frame number mask for the SMMU instance to mask out the invalid bits, which are typically used for mapping attributes, etc.

---
 drivers/iommu/tegra-smmu.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/iommu/tegra-smmu.c +6 -5
index 7293fc3f796d6..dd486233e2828 100644
@@ -159,9 +159,9 @@ static bool smmu_dma_addr_valid(struct tegra_smmu *smmu, dma_addr_t addr)
 	return (addr & smmu->pfn_mask) == addr;
 }
 
-static dma_addr_t smmu_pde_to_dma(u32 pde)
+static dma_addr_t smmu_pde_to_dma(struct tegra_smmu *smmu, u32 pde)
 {
-	return pde << 12;
+	return (dma_addr_t)(pde & smmu->pfn_mask) << 12;
 }
 
 static void smmu_flush_ptc_all(struct tegra_smmu *smmu)
@@ -549,6 +549,7 @@ static u32 *tegra_smmu_pte_lookup(struct tegra_smmu_as *as, unsigned long iova,
 				  dma_addr_t *dmap)
 {
 	unsigned int pd_index = iova_pd_index(iova);
+	struct tegra_smmu *smmu = as->smmu;
 	struct page *pt_page;
 	u32 *pd;
 
@@ -557,7 +558,7 @@ static u32 *tegra_smmu_pte_lookup(struct tegra_smmu_as *as, unsigned long iova,
 		return NULL;
 
 	pd = page_address(as->pd);
-	*dmap = smmu_pde_to_dma(pd[pd_index]);
+	*dmap = smmu_pde_to_dma(smmu, pd[pd_index]);
 
 	return tegra_smmu_pte_offset(pt_page, iova);
 }
@@ -599,7 +600,7 @@ static u32 *as_get_pte(struct tegra_smmu_as *as, dma_addr_t iova,
 	} else {
 		u32 *pd = page_address(as->pd);
 
-		*dmap = smmu_pde_to_dma(pd[pde]);
+		*dmap = smmu_pde_to_dma(smmu, pd[pde]);
 	}
 
 	return tegra_smmu_pte_offset(as->pts[pde], iova);
@@ -624,7 +625,7 @@ static void tegra_smmu_pte_put_use(struct tegra_smmu_as *as, unsigned long iova)
 	if (--as->count[pde] == 0) {
 		struct tegra_smmu *smmu = as->smmu;
 		u32 *pd = page_address(as->pd);
-		dma_addr_t pte_dma = smmu_pde_to_dma(pd[pde]);
+		dma_addr_t pte_dma = smmu_pde_to_dma(smmu, pd[pde]);
 
 		tegra_smmu_set_pde(as, iova, 0);
 
Signature · 1 line
2.20.1
author maintainer
#15 lore raw

PATCH 16/134 dmaengine: xilinx_dma: Clear desc_pendingcount in xilinx_dma_reset

In reply to Sasha Levin

From: Nicholas Graumann <[email protected]>

[ Upstream commit 8a631a5a0f7d4a4a24dba8587d5d9152be0871cc ]

Whenever we reset the channel, we need to clear desc_pendingcount along with desc_submitcount. Otherwise when a new transaction is submitted, the irq coalesce level could be programmed to an incorrect value in the axidma case.

This behavior can be observed when terminating pending transactions with xilinx_dma_terminate_all() and then submitting new transactions without releasing and requesting the channel.

---
 drivers/dma/xilinx/xilinx_dma.c | 1 +
 1 file changed, 1 insertion(+)
drivers/dma/xilinx/xilinx_dma.c +1 -0
index 5d56f1e4d332c..43acba2a1c0ee 100644
@@ -1433,6 +1433,7 @@ static int xilinx_dma_reset(struct xilinx_dma_chan *chan)
 
 	chan->err = false;
 	chan->idle = true;
+	chan->desc_pendingcount = 0;
 	chan->desc_submitcount = 0;
 
 	return err;
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#16 lore raw

PATCH 17/134 scsi: target: compare full CHAP_A Algorithm strings

In reply to Sasha Levin

From: David Disseldorp <[email protected]>

[ Upstream commit 9cef2a7955f2754257a7cddedec16edae7b587d0 ]

RFC 2307 states:

For CHAP [RFC1994], in the first step, the initiator MUST send:

      CHAP_A=<A1,A2...>

Where A1,A2... are proposed algorithms, in order of preference.
...
For the Algorithm, as stated in [RFC1994], one value is required to be implemented:

5 (CHAP with MD5)

LIO currently checks for this value by only comparing a single byte in the tokenized Algorithm string, which means that any value starting with a '5' (e.g. "55") is interpreted as "CHAP with MD5". Fix this by comparing the entire tokenized string.

---
 drivers/target/iscsi/iscsi_target_auth.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/target/iscsi/iscsi_target_auth.c +1 -1
index 51ddca2033e0a..8fe9b12a07a4d 100644
@@ -70,7 +70,7 @@ static int chap_check_algorithm(const char *a_str)
 		if (!token)
 			goto out;
 
-		if (!strncmp(token, "5", 1)) {
+		if (!strcmp(token, "5")) {
 			pr_debug("Selected MD5 Algorithm\n");
 			kfree(orig);
 			return CHAP_DIGEST_MD5;
Signature · 1 line
2.20.1
author maintainer
#17 lore raw

PATCH 18/134 scsi: lpfc: Fix hardlockup in lpfc_abort_handler

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit 91a52b617cdb8bf6d298892101c061d438b84a19 ]

In lpfc_abort_handler, the lock acquire order is hbalock (irqsave), buf_lock (irq) and ring_lock (irq). The issue is that in two places the locks are released out of order - the buf_lock and the hbalock - resulting in the cpu preemption/lock flags getting restored out of order and deadlocking the cpu.

Fix the unlock order by fully releasing the hbalocks as well.

---
 drivers/scsi/lpfc/lpfc_scsi.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/scsi/lpfc/lpfc_scsi.c +3 -2
index 6822cd9ff8f1e..68b31b27fa057 100644
@@ -4843,20 +4843,21 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd)
 		ret_val = __lpfc_sli_issue_iocb(phba, LPFC_FCP_RING,
 						abtsiocb, 0);
 	}
-	/* no longer need the lock after this point */
-	spin_unlock_irqrestore(&phba->hbalock, flags);
 
 	if (ret_val == IOCB_ERROR) {
 		/* Indicate the IO is not being aborted by the driver. */
 		iocb->iocb_flag &= ~LPFC_DRIVER_ABORTED;
 		lpfc_cmd->waitq = NULL;
 		spin_unlock(&lpfc_cmd->buf_lock);
+		spin_unlock_irqrestore(&phba->hbalock, flags);
 		lpfc_sli_release_iocbq(phba, abtsiocb);
 		ret = FAILED;
 		goto out;
 	}
 
+	/* no longer need the lock after this point */
 	spin_unlock(&lpfc_cmd->buf_lock);
+	spin_unlock_irqrestore(&phba->hbalock, flags);
 
 	if (phba->cfg_poll & DISABLE_FCP_RING_INT)
 		lpfc_sli_handle_fast_ring_event(phba,
Signature · 1 line
2.20.1
author maintainer
#18 lore raw

PATCH 19/134 scsi: lpfc: Fix SLI3 hba in loop mode not discovering devices

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit feff8b3d84d3d9570f893b4d83e5eab6693d6a52 ]

When operating in private loop mode, PLOGI exchanges are racing and the driver tries to abort it's PLOGI. But the PLOGI abort ends up terminating the login with the other end causing the other end to abort its PLOGI as well. Discovery never fully completes.

Fix by disabling the PLOGI abort when private loop and letting the state machine play out.

---
 drivers/scsi/lpfc/lpfc_nportdisc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/scsi/lpfc/lpfc_nportdisc.c +3 -1
index fc6e4546d738a..6961713825585 100644
@@ -484,8 +484,10 @@ lpfc_rcv_plogi(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp,
 	 * single discovery thread, this will cause a huge delay in
 	 * discovery. Also this will cause multiple state machines
 	 * running in parallel for this node.
+	 * This only applies to a fabric environment.
 	 */
-	if (ndlp->nlp_state == NLP_STE_PLOGI_ISSUE) {
+	if ((ndlp->nlp_state == NLP_STE_PLOGI_ISSUE) &&
+	    (vport->fc_flag & FC_FABRIC)) {
 		/* software abort outstanding PLOGI */
 		lpfc_els_abort(phba, ndlp);
 	}
Signature · 1 line
2.20.1
author maintainer
#19 lore raw

PATCH 20/134 scsi: csiostor: Don't enable IRQs too early

In reply to Sasha Levin

From: Dan Carpenter <[email protected]>

[ Upstream commit d6c9b31ac3064fbedf8961f120a4c117daa59932 ]

These are called with IRQs disabled from csio_mgmt_tmo_handler() so we can't call spin_unlock_irq() or it will enable IRQs prematurely.

Fixes a3667aaed569 ("[SCSI] csiostor: Chelsio FCoE offload driver")
---
 drivers/scsi/csiostor/csio_lnode.c | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)
drivers/scsi/csiostor/csio_lnode.c +9 -6
index 66e58f0a75dc1..23cbe4cda760e 100644
@@ -301,6 +301,7 @@ csio_ln_fdmi_rhba_cbfn(struct csio_hw *hw, struct csio_ioreq *fdmi_req)
 	struct fc_fdmi_port_name *port_name;
 	uint8_t buf[64];
 	uint8_t *fc4_type;
+	unsigned long flags;
 
 	if (fdmi_req->wr_status != FW_SUCCESS) {
 		csio_ln_dbg(ln, "WR error:%x in processing fdmi rhba cmd\n",
@@ -385,13 +386,13 @@ csio_ln_fdmi_rhba_cbfn(struct csio_hw *hw, struct csio_ioreq *fdmi_req)
 	len = (uint32_t)(pld - (uint8_t *)cmd);
 
 	/* Submit FDMI RPA request */
-	spin_lock_irq(&hw->lock);
+	spin_lock_irqsave(&hw->lock, flags);
 	if (csio_ln_mgmt_submit_req(fdmi_req, csio_ln_fdmi_done,
 				FCOE_CT, &fdmi_req->dma_buf, len)) {
 		CSIO_INC_STATS(ln, n_fdmi_err);
 		csio_ln_dbg(ln, "Failed to issue fdmi rpa req\n");
 	}
-	spin_unlock_irq(&hw->lock);
+	spin_unlock_irqrestore(&hw->lock, flags);
 }
 
 /*
@@ -412,6 +413,7 @@ csio_ln_fdmi_dprt_cbfn(struct csio_hw *hw, struct csio_ioreq *fdmi_req)
 	struct fc_fdmi_rpl *reg_pl;
 	struct fs_fdmi_attrs *attrib_blk;
 	uint8_t buf[64];
+	unsigned long flags;
 
 	if (fdmi_req->wr_status != FW_SUCCESS) {
 		csio_ln_dbg(ln, "WR error:%x in processing fdmi dprt cmd\n",
@@ -491,13 +493,13 @@ csio_ln_fdmi_dprt_cbfn(struct csio_hw *hw, struct csio_ioreq *fdmi_req)
 	attrib_blk->numattrs = htonl(numattrs);
 
 	/* Submit FDMI RHBA request */
-	spin_lock_irq(&hw->lock);
+	spin_lock_irqsave(&hw->lock, flags);
 	if (csio_ln_mgmt_submit_req(fdmi_req, csio_ln_fdmi_rhba_cbfn,
 				FCOE_CT, &fdmi_req->dma_buf, len)) {
 		CSIO_INC_STATS(ln, n_fdmi_err);
 		csio_ln_dbg(ln, "Failed to issue fdmi rhba req\n");
 	}
-	spin_unlock_irq(&hw->lock);
+	spin_unlock_irqrestore(&hw->lock, flags);
 }
 
 /*
@@ -512,6 +514,7 @@ csio_ln_fdmi_dhba_cbfn(struct csio_hw *hw, struct csio_ioreq *fdmi_req)
 	void *cmd;
 	struct fc_fdmi_port_name *port_name;
 	uint32_t len;
+	unsigned long flags;
 
 	if (fdmi_req->wr_status != FW_SUCCESS) {
 		csio_ln_dbg(ln, "WR error:%x in processing fdmi dhba cmd\n",
@@ -542,13 +545,13 @@ csio_ln_fdmi_dhba_cbfn(struct csio_hw *hw, struct csio_ioreq *fdmi_req)
 	len += sizeof(*port_name);
 
 	/* Submit FDMI request */
-	spin_lock_irq(&hw->lock);
+	spin_lock_irqsave(&hw->lock, flags);
 	if (csio_ln_mgmt_submit_req(fdmi_req, csio_ln_fdmi_dprt_cbfn,
 				FCOE_CT, &fdmi_req->dma_buf, len)) {
 		CSIO_INC_STATS(ln, n_fdmi_err);
 		csio_ln_dbg(ln, "Failed to issue fdmi dprt req\n");
 	}
-	spin_unlock_irq(&hw->lock);
+	spin_unlock_irqrestore(&hw->lock, flags);
 }
 
 /**
Signature · 1 line
2.20.1
author maintainer
#20 lore raw

PATCH 21/134 scsi: hisi_sas: Replace in_softirq() check in hisi_sas_task_exec()

In reply to Sasha Levin

From: Xiang Chen <[email protected]>

[ Upstream commit 550c0d89d52d3bec5c299f69b4ed5d2ee6b8a9a6 ]

For IOs from upper layer, preemption may be disabled as it may be called by function __blk_mq_delay_run_hw_queue which will call get_cpu() (it disables preemption). So if flags HISI_SAS_REJECT_CMD_BIT is set in function hisi_sas_task_exec(), it may disable preempt twice after down() and up() which will cause following call trace:

BUG: scheduling while atomic: fio/60373/0x00000002
Call trace:
dump_backtrace+0x0/0x150
show_stack+0x24/0x30
dump_stack+0xa0/0xc4
__schedule_bug+0x68/0x88
__schedule+0x4b8/0x548
schedule+0x40/0xd0
schedule_timeout+0x200/0x378
__down+0x78/0xc8
down+0x54/0x70
hisi_sas_task_exec.isra.10+0x598/0x8d8 [hisi_sas_main]
hisi_sas_queue_command+0x28/0x38 [hisi_sas_main]
sas_queuecommand+0x168/0x1b0 [libsas]
scsi_queue_rq+0x2ac/0x980
blk_mq_dispatch_rq_list+0xb0/0x550
blk_mq_do_dispatch_sched+0x6c/0x110
blk_mq_sched_dispatch_requests+0x114/0x1d8
__blk_mq_run_hw_queue+0xb8/0x130
__blk_mq_delay_run_hw_queue+0x1c0/0x220
blk_mq_run_hw_queue+0xb0/0x128
blk_mq_sched_insert_requests+0xdc/0x208
blk_mq_flush_plug_list+0x1b4/0x3a0
blk_flush_plug_list+0xdc/0x110
blk_finish_plug+0x3c/0x50
blkdev_direct_IO+0x404/0x550
generic_file_read_iter+0x9c/0x848
blkdev_read_iter+0x50/0x78
aio_read+0xc8/0x170
io_submit_one+0x1fc/0x8d8
__arm64_sys_io_submit+0xdc/0x280
el0_svc_common.constprop.0+0xe0/0x1e0
el0_svc_handler+0x34/0x90
el0_svc+0x10/0x14

...

To solve the issue, check preemptible() to avoid disabling preempt multiple when flag HISI_SAS_REJECT_CMD_BIT is set.

---
 drivers/scsi/hisi_sas/hisi_sas_main.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/scsi/hisi_sas/hisi_sas_main.c +7 -1
index 0847e682797be..20f0cb4698b7f 100644
@@ -587,7 +587,13 @@ static int hisi_sas_task_exec(struct sas_task *task, gfp_t gfp_flags,
 	dev = hisi_hba->dev;
 
 	if (unlikely(test_bit(HISI_SAS_REJECT_CMD_BIT, &hisi_hba->flags))) {
-		if (in_softirq())
+		/*
+		 * For IOs from upper layer, it may already disable preempt
+		 * in the IO path, if disable preempt again in down(),
+		 * function schedule() will report schedule_bug(), so check
+		 * preemptible() before goto down().
+		 */
+		if (!preemptible())
 			return -EINVAL;
 
 		down(&hisi_hba->sem);
Signature · 1 line
2.20.1
author maintainer
#21 lore raw

PATCH 22/134 scsi: hisi_sas: Delete the debugfs folder of hisi_sas when the probe fails

In reply to Sasha Levin

From: Luo Jiaxing <[email protected]>

[ Upstream commit cabe7c10c97a0857a9fb14b6c772ab784947995d ]

Although if the debugfs initialization fails, we will delete the debugfs folder of hisi_sas, but we did not consider the scenario where debugfs was successfully initialized, but the probe failed for other reasons. We found out that hisi_sas folder is still remain after the probe failed.

When probe fail, we should delete debugfs folder to avoid the above issue.

---
 drivers/scsi/hisi_sas/hisi_sas_main.c  | 1 +
 drivers/scsi/hisi_sas/hisi_sas_v3_hw.c | 1 +
 2 files changed, 2 insertions(+)
drivers/scsi/hisi_sas/hisi_sas_main.c +1 -0
index 20f0cb4698b7f..633effb09c9cc 100644
@@ -2682,6 +2682,7 @@ int hisi_sas_probe(struct platform_device *pdev,
 err_out_register_ha:
 	scsi_remove_host(shost);
 err_out_ha:
+	hisi_sas_debugfs_exit(hisi_hba);
 	hisi_sas_free(hisi_hba);
 	scsi_host_put(shost);
 	return rc;
drivers/scsi/hisi_sas/hisi_sas_v3_hw.c +1 -0
index cb8d087762dbd..ef32ee12f6065 100644
@@ -3259,6 +3259,7 @@ hisi_sas_v3_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 err_out_register_ha:
 	scsi_remove_host(shost);
 err_out_ha:
+	hisi_sas_debugfs_exit(hisi_hba);
 	scsi_host_put(shost);
 err_out_regions:
 	pci_release_regions(pdev);
Signature · 1 line
2.20.1
author maintainer
#22 lore raw

PATCH 23/134 powerpc/pseries: Mark accumulate_stolen_time() as notrace

In reply to Sasha Levin

From: Michael Ellerman <[email protected]>

[ Upstream commit eb8e20f89093b64f48975c74ccb114e6775cee22 ]

accumulate_stolen_time() is called prior to interrupt state being reconciled, which can trip the warning in arch_local_irq_restore():

  WARNING: CPU: 5 PID: 1017 at arch/powerpc/kernel/irq.c:258 .arch_local_irq_restore+0x9c/0x130
  ...
  NIP .arch_local_irq_restore+0x9c/0x130
  LR  .rb_start_commit+0x38/0x80
  Call Trace:
    .ring_buffer_lock_reserve+0xe4/0x620
    .trace_function+0x44/0x210
    .function_trace_call+0x148/0x170
    .ftrace_ops_no_ops+0x180/0x1d0
    ftrace_call+0x4/0x8
    .accumulate_stolen_time+0x1c/0xb0
    decrementer_common+0x124/0x160

For now just mark it as notrace. We may change the ordering to call it after interrupt state has been reconciled, but that is a larger change.

---
 arch/powerpc/kernel/time.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/kernel/time.c +1 -1
index 694522308cd51..968ae97382b4e 100644
@@ -232,7 +232,7 @@ static u64 scan_dispatch_log(u64 stop_tb)
  * Accumulate stolen time by scanning the dispatch trace log.
  * Called on entry from user mode.
  */
-void accumulate_stolen_time(void)
+void notrace accumulate_stolen_time(void)
 {
 	u64 sst, ust;
 	unsigned long save_irq_soft_mask = irq_soft_mask_return();
Signature · 1 line
2.20.1
author maintainer
#23 lore raw

PATCH 24/134 powerpc/pseries: Don't fail hash page table insert for bolted mapping

In reply to Sasha Levin

From: "Aneesh Kumar K.V" <[email protected]>

[ Upstream commit 75838a3290cd4ebbd1f567f310ba04b6ef017ce4 ]

If the hypervisor returned H_PTEG_FULL for H_ENTER hcall, retry a hash page table insert by removing a random entry from the group.

After some runtime, it is very well possible to find all the 8 hash page table entry slot in the hpte group used for mapping. Don't fail a bolted entry insert in that case. With Storage class memory a user can find this error easily since a namespace enable/disable is equivalent to memory add/remove.

This results in failures as reported below:

$ ndctl create-namespace -r region1 -t pmem -m devdax -a 65536 -s 100M
libndctl: ndctl_dax_enable: dax1.3: failed to enable
  Error: namespace1.2: failed to enable

failed to create namespace: No such device or address

In kernel log we find the details as below:

Unable to create mapping for hot added memory 0xc000042006000000..0xc00004200d000000: -1
dax_pmem: probe of dax1.3 failed with error -14

This indicates that we failed to create a bolted hash table entry for direct-map address backing the namespace.

We also observe failures such that not all namespaces will be enabled with ndctl enable-namespace all command.

---
 arch/powerpc/mm/book3s64/hash_utils.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
arch/powerpc/mm/book3s64/hash_utils.c +8 -1
index 6c123760164e8..6e5a769ebcb80 100644
@@ -294,7 +294,14 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
 		ret = mmu_hash_ops.hpte_insert(hpteg, vpn, paddr, tprot,
 					       HPTE_V_BOLTED, psize, psize,
 					       ssize);
-
+		if (ret == -1) {
+			/* Try to remove a non bolted entry */
+			ret = mmu_hash_ops.hpte_remove(hpteg);
+			if (ret != -1)
+				ret = mmu_hash_ops.hpte_insert(hpteg, vpn, paddr, tprot,
+							       HPTE_V_BOLTED, psize, psize,
+							       ssize);
+		}
 		if (ret < 0)
 			break;
 
Signature · 1 line
2.20.1
author maintainer
#24 lore raw

PATCH 25/134 Input: st1232 - do not reset the chip too early

In reply to Sasha Levin

From: Dmitry Torokhov <[email protected]>

[ Upstream commit efd7bb08a762d4f6322054c6824bd942971ac563 ]

We should not be putting the chip into reset while interrupts are enabled and ISR may be running. Fix this by installing a custom devm action and powering off the device/resetting GPIO line from there. This ensures proper ordering.

---
 drivers/input/touchscreen/st1232.c | 22 ++++++++++++----------
 1 file changed, 12 insertions(+), 10 deletions(-)
drivers/input/touchscreen/st1232.c +12 -10
index 1139714e72e26..1c5f8875cb795 100644
@@ -149,6 +149,11 @@ static void st1232_ts_power(struct st1232_ts_data *ts, bool poweron)
 		gpiod_set_value_cansleep(ts->reset_gpio, !poweron);
 }
 
+static void st1232_ts_power_off(void *data)
+{
+	st1232_ts_power(data, false);
+}
+
 static const struct st_chip_info st1232_chip_info = {
 	.have_z		= true,
 	.max_x		= 0x31f, /* 800 - 1 */
@@ -229,6 +234,13 @@ static int st1232_ts_probe(struct i2c_client *client,
 
 	st1232_ts_power(ts, true);
 
+	error = devm_add_action_or_reset(&client->dev, st1232_ts_power_off, ts);
+	if (error) {
+		dev_err(&client->dev,
+			"Failed to install power off action: %d\n", error);
+		return error;
+	}
+
 	input_dev->name = "st1232-touchscreen";
 	input_dev->id.bustype = BUS_I2C;
 	input_dev->dev.parent = &client->dev;
@@ -271,15 +283,6 @@ static int st1232_ts_probe(struct i2c_client *client,
 	return 0;
 }
 
-static int st1232_ts_remove(struct i2c_client *client)
-{
-	struct st1232_ts_data *ts = i2c_get_clientdata(client);
-
-	st1232_ts_power(ts, false);
-
-	return 0;
-}
-
 static int __maybe_unused st1232_ts_suspend(struct device *dev)
 {
 	struct i2c_client *client = to_i2c_client(dev);
@@ -329,7 +332,6 @@ MODULE_DEVICE_TABLE(of, st1232_ts_dt_ids);
 
 static struct i2c_driver st1232_ts_driver = {
 	.probe		= st1232_ts_probe,
-	.remove		= st1232_ts_remove,
 	.id_table	= st1232_ts_id,
 	.driver = {
 		.name	= ST1232_TS_NAME,
Signature · 1 line
2.20.1
author maintainer
#25 lore raw

PATCH 26/134 selftests/powerpc: Fixup clobbers for TM tests

In reply to Sasha Levin

From: Michael Ellerman <[email protected]>

[ Upstream commit a02cbc7ffe529ed58b6bbe54652104fc2c88bd77 ]

Some of our TM (Transactional Memory) tests, list "r1" (the stack pointer) as a clobbered register.

GCC >= 9 doesn't accept this, and the build breaks:

ptrace-tm-spd-tar.c: In function 'tm_spd_tar':
ptrace-tm-spd-tar.c:31:2: error: listing the stack pointer register 'r1' in a clobber list is deprecated [-Werror=deprecated]

     31 |  asm __volatile__(
        |  ^~~

ptrace-tm-spd-tar.c:31:2: note: the value of the stack pointer after an 'asm' statement must be the same as it was before the statement

We do have some fairly large inline asm blocks in these tests, and some of them do change the value of r1. However they should all return to C with the value in r1 restored, so I think it's legitimate to say r1 is not clobbered.

As Segher points out, the r1 clobbers may have been added because of the use of or 1,1,1, however that doesn't actually clobber r1.

Segher also points out that some of these tests do clobber LR, because they call functions, and that is not listed in the clobbers, so add that where appropriate.

---
 tools/testing/selftests/powerpc/ptrace/ptrace-tm-spd-tar.c | 2 +-
 tools/testing/selftests/powerpc/ptrace/ptrace-tm-spd-vsx.c | 4 ++--
 tools/testing/selftests/powerpc/ptrace/ptrace-tm-tar.c     | 2 +-
 tools/testing/selftests/powerpc/ptrace/ptrace-tm-vsx.c     | 4 ++--
 4 files changed, 6 insertions(+), 6 deletions(-)
tools/testing/selftests/powerpc/ptrace/ptrace-tm-spd-tar.c +1 -1
index 25e23e73c72e6..2ecfa1158e2bc 100644
@@ -73,7 +73,7 @@ trans:
 		[sprn_texasr]"i"(SPRN_TEXASR), [tar_1]"i"(TAR_1),
 		[dscr_1]"i"(DSCR_1), [tar_2]"i"(TAR_2), [dscr_2]"i"(DSCR_2),
 		[tar_3]"i"(TAR_3), [dscr_3]"i"(DSCR_3)
-		: "memory", "r0", "r1", "r3", "r4", "r5", "r6"
+		: "memory", "r0", "r3", "r4", "r5", "r6", "lr"
 		);
 
 	/* TM failed, analyse */
tools/testing/selftests/powerpc/ptrace/ptrace-tm-spd-vsx.c +2 -2
index f603fe5a445b3..6f7fb51f08099 100644
@@ -74,8 +74,8 @@ trans:
 		"3: ;"
 		: [res] "=r" (result), [texasr] "=r" (texasr)
 		: [sprn_texasr] "i"  (SPRN_TEXASR)
-		: "memory", "r0", "r1", "r3", "r4",
-		"r7", "r8", "r9", "r10", "r11"
+		: "memory", "r0", "r3", "r4",
+		  "r7", "r8", "r9", "r10", "r11", "lr"
 		);
 
 	if (result) {
tools/testing/selftests/powerpc/ptrace/ptrace-tm-tar.c +1 -1
index e0d37f07bdeba..46ef378a15ecc 100644
@@ -62,7 +62,7 @@ trans:
 		[sprn_ppr]"i"(SPRN_PPR), [sprn_texasr]"i"(SPRN_TEXASR),
 		[tar_1]"i"(TAR_1), [dscr_1]"i"(DSCR_1), [tar_2]"i"(TAR_2),
 		[dscr_2]"i"(DSCR_2), [cptr1] "b" (&cptr[1])
-		: "memory", "r0", "r1", "r3", "r4", "r5", "r6"
+		: "memory", "r0", "r3", "r4", "r5", "r6"
 		);
 
 	/* TM failed, analyse */
tools/testing/selftests/powerpc/ptrace/ptrace-tm-vsx.c +2 -2
index 8027457b97b7c..70ca01234f79c 100644
@@ -62,8 +62,8 @@ trans:
 		"3: ;"
 		: [res] "=r" (result), [texasr] "=r" (texasr)
 		: [sprn_texasr] "i"  (SPRN_TEXASR), [cptr1] "b" (&cptr[1])
-		: "memory", "r0", "r1", "r3", "r4",
-		"r7", "r8", "r9", "r10", "r11"
+		: "memory", "r0", "r3", "r4",
+		  "r7", "r8", "r9", "r10", "r11", "lr"
 		);
 
 	if (result) {
Signature · 1 line
2.20.1
author maintainer
#26 lore raw

PATCH 27/134 powerpc/tools: Don't quote $objdump in scripts

In reply to Sasha Levin

From: Michael Ellerman <[email protected]>

[ Upstream commit e44ff9ea8f4c8a90c82f7b85bd4f5e497c841960 ]

Some of our scripts are passed $objdump and then call it as "$objdump". This doesn't work if it contains spaces because we're using ccache, for example you get errors such as:

./arch/powerpc/tools/relocs_check.sh: line 48: ccache ppc64le-objdump: No such file or directory ./arch/powerpc/tools/unrel_branch_check.sh: line 26: ccache ppc64le-objdump: No such file or directory

Fix it by not quoting the string when we expand it, allowing the shell to do the right thing for us.

Fixes a71aa05e1416 ("powerpc: Convert relocs_check to a shell script using grep")
Fixes 4ea80652dc75 ("powerpc/64s: Tool to flag direct branches from unrelocated interrupt vectors")
---
 arch/powerpc/tools/relocs_check.sh       | 2 +-
 arch/powerpc/tools/unrel_branch_check.sh | 4 ++--
 2 files changed, 3 insertions(+), 3 deletions(-)
arch/powerpc/tools/relocs_check.sh +1 -1
index 2b4e959caa365..7b9fe0a567cf3 100755
@@ -20,7 +20,7 @@ objdump="$1"
 vmlinux="$2"
 
 bad_relocs=$(
-"$objdump" -R "$vmlinux" |
+$objdump -R "$vmlinux" |
 	# Only look at relocation lines.
 	grep -E '\<R_' |
 	# These relocations are okay
arch/powerpc/tools/unrel_branch_check.sh +2 -2
index 1e972df3107ee..77114755dc6f2 100755
@@ -18,14 +18,14 @@ vmlinux="$2"
 #__end_interrupts should be located within the first 64K
 
 end_intr=0x$(
-"$objdump" -R "$vmlinux" -d --start-address=0xc000000000000000		\
+$objdump -R "$vmlinux" -d --start-address=0xc000000000000000           \
 		 --stop-address=0xc000000000010000 |
 grep '\<__end_interrupts>:' |
 awk '{print $1}'
 )
 
 BRANCHES=$(
-"$objdump" -R "$vmlinux" -D --start-address=0xc000000000000000		\
+$objdump -R "$vmlinux" -D --start-address=0xc000000000000000           \
 		--stop-address=${end_intr} |
 grep -e "^c[0-9a-f]*:[[:space:]]*\([0-9a-f][0-9a-f][[:space:]]\)\{4\}[[:space:]]*b" |
 grep -v '\<__start_initialization_multiplatform>' |
Signature · 1 line
2.20.1
author maintainer
#27 lore raw

PATCH 28/134 dma-debug: add a schedule point in debug_dma_dump_mappings()

In reply to Sasha Levin

From: Eric Dumazet <[email protected]>

[ Upstream commit 9ff6aa027dbb98755f0265695354f2dd07c0d1ce ]

debug_dma_dump_mappings() can take a lot of cpu cycles :

lpk43:/# time wc -l /sys/kernel/debug/dma-api/dump
163435 /sys/kernel/debug/dma-api/dump

real 0m0.463s
user 0m0.003s
sys 0m0.459s

Let's add a cond_resched() to avoid holding cpu for too long.

Cc Corentin Labbe <[email protected]>
Cc Christoph Hellwig <[email protected]>
Cc Marek Szyprowski <[email protected]>
---
 kernel/dma/debug.c | 1 +
 1 file changed, 1 insertion(+)
kernel/dma/debug.c +1 -0
index 099002d84f466..4ad74f5987ea9 100644
@@ -420,6 +420,7 @@ void debug_dma_dump_mappings(struct device *dev)
 		}
 
 		spin_unlock_irqrestore(&bucket->lock, flags);
+		cond_resched();
 	}
 }
 
Signature · 1 line
2.20.1
author maintainer
#28 lore raw

PATCH 29/134 dma-mapping: Add vmap checks to dma_map_single()

In reply to Sasha Levin

From: Kees Cook <[email protected]>

[ Upstream commit 4544b9f25e70eae9f70a243de0cc802aa5c8cb69 ]

As we've seen from USB and other areas[1], we need to always do runtime checks for DMA operating on memory regions that might be remapped. This adds vmap checks (similar to those already in USB but missing in other places) into dma_map_single() so all callers benefit from the checking.

[1] https://git.kernel.org/linus/3840c5b78803b2b6cc1ff820100a74a092c40cbb

---
 include/linux/dma-mapping.h | 4 ++++
 1 file changed, 4 insertions(+)
include/linux/dma-mapping.h +4 -0
index 4a1c4fca475ad..0aad641d662c3 100644
@@ -583,6 +583,10 @@ static inline unsigned long dma_get_merge_boundary(struct device *dev)
 static inline dma_addr_t dma_map_single_attrs(struct device *dev, void *ptr,
 		size_t size, enum dma_data_direction dir, unsigned long attrs)
 {
+	/* DMA must never operate on areas that might be remapped. */
+	if (dev_WARN_ONCE(dev, is_vmalloc_addr(ptr),
+			  "rejecting DMA map of vmalloc memory\n"))
+		return DMA_MAPPING_ERROR;
 	debug_dma_map_single(dev, ptr, size);
 	return dma_map_page_attrs(dev, virt_to_page(ptr), offset_in_page(ptr),
 			size, dir, attrs);
Signature · 1 line
2.20.1
author maintainer
#29 lore raw

PATCH 30/134 dma-mapping: fix handling of dma-ranges for reserved memory (again)

In reply to Sasha Levin

From: Vladimir Murzin <[email protected]>

[ Upstream commit a445e940ea686fc60475564009821010eb213be3 ]

Daniele reported that issue previously fixed in c41f9ea998f3 ("drivers: dma-coherent: Account dma_pfn_offset when used with device tree") reappear shortly after 43fc509c3efb ("dma-coherent: introduce interface for default DMA pool") where fix was accidentally dropped.

Lets put fix back in place and respect dma-ranges for reserved memory.

Fixes 43fc509c3efb ("dma-coherent: introduce interface for default DMA pool")
---
 arch/arm/mm/dma-mapping-nommu.c |  2 +-
 include/linux/dma-mapping.h     |  4 ++--
 kernel/dma/coherent.c           | 16 +++++++++-------
 3 files changed, 12 insertions(+), 10 deletions(-)
arch/arm/mm/dma-mapping-nommu.c +1 -1
index db92478983005..287ef898a55e1 100644
@@ -35,7 +35,7 @@ static void *arm_nommu_dma_alloc(struct device *dev, size_t size,
 				 unsigned long attrs)
 
 {
-	void *ret = dma_alloc_from_global_coherent(size, dma_handle);
+	void *ret = dma_alloc_from_global_coherent(dev, size, dma_handle);
 
 	/*
 	 * dma_alloc_from_global_coherent() may fail because:
include/linux/dma-mapping.h +2 -2
index 0aad641d662c3..4d450672b7d66 100644
@@ -162,7 +162,7 @@ int dma_release_from_dev_coherent(struct device *dev, int order, void *vaddr);
 int dma_mmap_from_dev_coherent(struct device *dev, struct vm_area_struct *vma,
 			    void *cpu_addr, size_t size, int *ret);
 
-void *dma_alloc_from_global_coherent(ssize_t size, dma_addr_t *dma_handle);
+void *dma_alloc_from_global_coherent(struct device *dev, ssize_t size, dma_addr_t *dma_handle);
 int dma_release_from_global_coherent(int order, void *vaddr);
 int dma_mmap_from_global_coherent(struct vm_area_struct *vma, void *cpu_addr,
 				  size_t size, int *ret);
@@ -172,7 +172,7 @@ int dma_mmap_from_global_coherent(struct vm_area_struct *vma, void *cpu_addr,
 #define dma_release_from_dev_coherent(dev, order, vaddr) (0)
 #define dma_mmap_from_dev_coherent(dev, vma, vaddr, order, ret) (0)
 
-static inline void *dma_alloc_from_global_coherent(ssize_t size,
+static inline void *dma_alloc_from_global_coherent(struct device *dev, ssize_t size,
 						   dma_addr_t *dma_handle)
 {
 	return NULL;
kernel/dma/coherent.c +9 -7
index 545e3869b0e32..551b0eb7028a3 100644
@@ -123,8 +123,9 @@ int dma_declare_coherent_memory(struct device *dev, phys_addr_t phys_addr,
 	return ret;
 }
 
-static void *__dma_alloc_from_coherent(struct dma_coherent_mem *mem,
-		ssize_t size, dma_addr_t *dma_handle)
+static void *__dma_alloc_from_coherent(struct device *dev,
+				       struct dma_coherent_mem *mem,
+				       ssize_t size, dma_addr_t *dma_handle)
 {
 	int order = get_order(size);
 	unsigned long flags;
@@ -143,7 +144,7 @@ static void *__dma_alloc_from_coherent(struct dma_coherent_mem *mem,
 	/*
 	 * Memory was found in the coherent area.
 	 */
-	*dma_handle = mem->device_base + (pageno << PAGE_SHIFT);
+	*dma_handle = dma_get_device_base(dev, mem) + (pageno << PAGE_SHIFT);
 	ret = mem->virt_base + (pageno << PAGE_SHIFT);
 	spin_unlock_irqrestore(&mem->spinlock, flags);
 	memset(ret, 0, size);
@@ -175,17 +176,18 @@ int dma_alloc_from_dev_coherent(struct device *dev, ssize_t size,
 	if (!mem)
 		return 0;
 
-	*ret = __dma_alloc_from_coherent(mem, size, dma_handle);
+	*ret = __dma_alloc_from_coherent(dev, mem, size, dma_handle);
 	return 1;
 }
 
-void *dma_alloc_from_global_coherent(ssize_t size, dma_addr_t *dma_handle)
+void *dma_alloc_from_global_coherent(struct device *dev, ssize_t size,
+				     dma_addr_t *dma_handle)
 {
 	if (!dma_coherent_default_memory)
 		return NULL;
 
-	return __dma_alloc_from_coherent(dma_coherent_default_memory, size,
-			dma_handle);
+	return __dma_alloc_from_coherent(dev, dma_coherent_default_memory, size,
+					 dma_handle);
 }
 
 static int __dma_release_from_coherent(struct dma_coherent_mem *mem,
Signature · 1 line
2.20.1
author maintainer
#30 lore raw

PATCH 31/134 dmaengine: fsl-qdma: Handle invalid qdma-queue0 IRQ

In reply to Sasha Levin

From: Krzysztof Kozlowski <[email protected]>

[ Upstream commit 41814c4eadf8a791b6d07114f96e7e120e59555c ]

platform_get_irq_byname() might return -errno which later would be cast to an unsigned int and used in IRQ handling code leading to usage of wrong ID and errors about wrong irq_base.

---
 drivers/dma/fsl-qdma.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/dma/fsl-qdma.c +3 -0
index 06664fbd2d911..89792083d62c5 100644
@@ -1155,6 +1155,9 @@ static int fsl_qdma_probe(struct platform_device *pdev)
 		return ret;
 
 	fsl_qdma->irq_base = platform_get_irq_byname(pdev, "qdma-queue0");
+	if (fsl_qdma->irq_base < 0)
+		return fsl_qdma->irq_base;
+
 	fsl_qdma->feature = of_property_read_bool(np, "big-endian");
 	INIT_LIST_HEAD(&fsl_qdma->dma_dev.channels);
 
Signature · 1 line
2.20.1
author maintainer
#31 lore raw

PATCH 32/134 leds: lm3692x: Handle failure to probe the regulator

In reply to Sasha Levin

From: Guido Günther <[email protected]>

[ Upstream commit 396128d2ffcba6e1954cfdc9a89293ff79cbfd7c ]

Instead use devm_regulator_get_optional since the regulator is optional and check for errors.

---
 drivers/leds/leds-lm3692x.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)
drivers/leds/leds-lm3692x.c +11 -2
index 3d381f2f73d04..1ac9a44570eed 100644
@@ -334,9 +334,18 @@ static int lm3692x_probe_dt(struct lm3692x_led *led)
 		return ret;
 	}
 
-	led->regulator = devm_regulator_get(&led->client->dev, "vled");
-	if (IS_ERR(led->regulator))
+	led->regulator = devm_regulator_get_optional(&led->client->dev, "vled");
+	if (IS_ERR(led->regulator)) {
+		ret = PTR_ERR(led->regulator);
+		if (ret != -ENODEV) {
+			if (ret != -EPROBE_DEFER)
+				dev_err(&led->client->dev,
+					"Failed to get vled regulator: %d\n",
+					ret);
+			return ret;
+		}
 		led->regulator = NULL;
+	}
 
 	child = device_get_next_child_node(&led->client->dev, child);
 	if (!child) {
Signature · 1 line
2.20.1
#32 lore raw

Re: [PATCH AUTOSEL 5.4 032/134] leds: lm3692x: Handle failure to probe the regulator

Quoted 6 lines · Sasha Levin

From: Guido Günther <[email protected]>

[ Upstream commit 396128d2ffcba6e1954cfdc9a89293ff79cbfd7c ]

Instead use devm_regulator_get_optional since the regulator is optional and check for errors.

Support for new hardware, more than serious bug.

Please drop. Also please drop leds from your monitoring scripts, it just adds unneccessary workload for everyone. If we see something _really_ critical, we'll cc: stable.

Actually, I believe your monitoring scripts are doing more harm than good, and it would be better to get positive confirmation from contributors before pushing random changes into stable.

Thanks,
Pavel

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
author maintainer
#33 lore raw

PATCH 33/134 leds: an30259a: add a check for devm_regmap_init_i2c

In reply to Sasha Levin

From: Chuhong Yuan <[email protected]>

[ Upstream commit fc7b5028f2627133c7c18734715a08829eab4d1f ]

an30259a_probe misses a check for devm_regmap_init_i2c and may cause problems.
Add a check and print errors like other leds drivers.

---
 drivers/leds/leds-an30259a.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/leds/leds-an30259a.c +7 -0
index 250dc9d6f6350..82350a28a5644 100644
@@ -305,6 +305,13 @@ static int an30259a_probe(struct i2c_client *client)
 
 	chip->regmap = devm_regmap_init_i2c(client, &an30259a_regmap_config);
 
+	if (IS_ERR(chip->regmap)) {
+		err = PTR_ERR(chip->regmap);
+		dev_err(&client->dev, "Failed to allocate register map: %d\n",
+			err);
+		goto exit;
+	}
+
 	for (i = 0; i < chip->num_leds; i++) {
 		struct led_init_data init_data = {};
 
Signature · 1 line
2.20.1
#34 lore raw

Re: [PATCH AUTOSEL 5.4 033/134] leds: an30259a: add a check for devm_regmap_init_i2c

Quoted 7 lines · Sasha Levin

From: Chuhong Yuan <[email protected]>

[ Upstream commit fc7b5028f2627133c7c18734715a08829eab4d1f ]

an30259a_probe misses a check for devm_regmap_init_i2c and may cause problems.
Add a check and print errors like other leds drivers.

Please drop.
Pavel

Quoted 27 lines
---
 drivers/leds/leds-an30259a.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/leds/leds-an30259a.c +7 -0
index 250dc9d6f6350..82350a28a5644 100644
@@ -305,6 +305,13 @@ static int an30259a_probe(struct i2c_client *client)
 
 	chip->regmap = devm_regmap_init_i2c(client, &an30259a_regmap_config);
 
+	if (IS_ERR(chip->regmap)) {
+		err = PTR_ERR(chip->regmap);
+		dev_err(&client->dev, "Failed to allocate register map: %d\n",
+			err);
+		goto exit;
+	}
+
 	for (i = 0; i < chip->num_leds; i++) {
 		struct led_init_data init_data = {};
 
Signature · 1 line
2.20.1
Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
author maintainer
#35 lore raw

PATCH 34/134 leds: trigger: netdev: fix handling on interface rename

In reply to Sasha Levin

From: Martin Schiller <[email protected]>

[ Upstream commit 5f820ed52371b4f5d8c43c93f03408d0dbc01e5b ]

The NETDEV_CHANGENAME code is not "unneeded" like it is stated in commit 4cb6560514fa ("leds: trigger: netdev: fix refcnt leak on interface rename").

The event was accidentally misinterpreted equivalent to NETDEV_UNREGISTER, but should be equivalent to NETDEV_REGISTER.

This was the case in the original code from the openwrt project.

Otherwise, you are unable to set netdev led triggers for (non-existent) netdevices, which has to be renamed. This is the case, for example, for ppp interfaces in openwrt.

Fixes 06f502f57d0d ("leds: trigger: Introduce a NETDEV trigger")
Fixes 4cb6560514fa ("leds: trigger: netdev: fix refcnt leak on interface rename")
---
 drivers/leds/trigger/ledtrig-netdev.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/leds/trigger/ledtrig-netdev.c +4 -1
index 136f86a1627d1..d5e774d830215 100644
@@ -302,10 +302,12 @@ static int netdev_trig_notify(struct notifier_block *nb,
 		container_of(nb, struct led_netdev_data, notifier);
 
 	if (evt != NETDEV_UP && evt != NETDEV_DOWN && evt != NETDEV_CHANGE
-	    && evt != NETDEV_REGISTER && evt != NETDEV_UNREGISTER)
+	    && evt != NETDEV_REGISTER && evt != NETDEV_UNREGISTER
+	    && evt != NETDEV_CHANGENAME)
 		return NOTIFY_DONE;
 
 	if (!(dev == trigger_data->net_dev ||
+	      (evt == NETDEV_CHANGENAME && !strcmp(dev->name, trigger_data->device_name)) ||
 	      (evt == NETDEV_REGISTER && !strcmp(dev->name, trigger_data->device_name))))
 		return NOTIFY_DONE;
 
@@ -315,6 +317,7 @@ static int netdev_trig_notify(struct notifier_block *nb,
 
 	clear_bit(NETDEV_LED_MODE_LINKUP, &trigger_data->mode);
 	switch (evt) {
+	case NETDEV_CHANGENAME:
 	case NETDEV_REGISTER:
 		if (trigger_data->net_dev)
 			dev_put(trigger_data->net_dev);
Signature · 1 line
2.20.1
#36 lore raw

Re: [PATCH AUTOSEL 5.4 034/134] leds: trigger: netdev: fix handling on interface rename

Quoted 8 lines · Sasha Levin

From: Martin Schiller <[email protected]>

[ Upstream commit 5f820ed52371b4f5d8c43c93f03408d0dbc01e5b ]

The NETDEV_CHANGENAME code is not "unneeded" like it is stated in commit 4cb6560514fa ("leds: trigger: netdev: fix refcnt leak on interface rename").

Quoted 8 lines

The event was accidentally misinterpreted equivalent to NETDEV_UNREGISTER, but should be equivalent to NETDEV_REGISTER.

This was the case in the original code from the openwrt project.

Otherwise, you are unable to set netdev led triggers for (non-existent) netdevices, which has to be renamed. This is the case, for example, for ppp interfaces in openwrt.

Please drop.
Pavel

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
author maintainer
#37 lore raw

Re: [PATCH AUTOSEL 5.4 034/134] leds: trigger: netdev: fix handling on interface rename

Quoted 12 lines · Pavel Machek
Quoted 11 lines · Sasha Levin

From: Martin Schiller <[email protected]>

[ Upstream commit 5f820ed52371b4f5d8c43c93f03408d0dbc01e5b ]

The NETDEV_CHANGENAME code is not "unneeded" like it is stated in commit 4cb6560514fa ("leds: trigger: netdev: fix refcnt leak on interface rename").

The event was accidentally misinterpreted equivalent to NETDEV_UNREGISTER, but should be equivalent to NETDEV_REGISTER.

Quoted 7 lines
Quoted 5 lines

This was the case in the original code from the openwrt project.

Otherwise, you are unable to set netdev led triggers for (non-existent) netdevices, which has to be renamed. This is the case, for example, for ppp interfaces in openwrt.

Please drop.

Here's a bug report from a user (fixed by this patch):

       https://forum.openwrt.org/t/18-06-4-serious-led-problems-td-w8970-v1/40417

He has titled the report "18.06.4:Serious ‘LED’ Problems!", and in the bug report itself he has mentioned:

The LED's are really important to me. please let me know how to fix this by my self! I cannot report it and wait for next few months for patch, again!

There are two other similar bug reports:

       https://bugs.openwrt.org/index.php?do=details&task_id=2193
       https://bugs.openwrt.org/index.php?do=details&task_id=2239

So this is obviously an issue that affects users and needs to be fixed.

Beyond the above, the patch is upstream, it fixes a single issue, and is shorter than 100 lines.

I'm going to go ahead and ignore your input for this and the rest of the led patches in the series for similar reasons.

Signature · 2 lines
Thanks,
Sasha
author maintainer
#38 lore raw

PATCH 35/134 clocksource/drivers/asm9260: Add a check for of_clk_get

In reply to Sasha Levin

From: Chuhong Yuan <[email protected]>

[ Upstream commit 6e001f6a4cc73cd06fc7b8c633bc4906c33dd8ad ]

asm9260_timer_init misses a check for of_clk_get.
Add a check for it and print errors like other clocksource drivers.

---
 drivers/clocksource/asm9260_timer.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/clocksource/asm9260_timer.c +4 -0
index 9f09a59161e70..5b39d3701fa3c 100644
@@ -194,6 +194,10 @@ static int __init asm9260_timer_init(struct device_node *np)
 	}
 
 	clk = of_clk_get(np, 0);
+	if (IS_ERR(clk)) {
+		pr_err("Failed to get clk!\n");
+		return PTR_ERR(clk);
+	}
 
 	ret = clk_prepare_enable(clk);
 	if (ret) {
Signature · 1 line
2.20.1
author maintainer
#39 lore raw

PATCH 36/134 clocksource/drivers/timer-of: Use unique device name instead of timer

In reply to Sasha Levin

From: Geert Uytterhoeven <[email protected]>

[ Upstream commit 4411464d6f8b5e5759637235a6f2b2a85c2be0f1 ]

If a hardware-specific driver does not provide a name, the timer-of core falls back to device_node.name. Due to generic DT node naming policies, that name is almost always "timer", and thus doesn't identify the actual timer used.

Fix this by using device_node.full_name instead, which includes the unit addrees.

Example impact on /proc/timer_list:

-Clock Event Device: timer
+Clock Event Device: timer@fcfec400

---
 drivers/clocksource/timer-of.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clocksource/timer-of.c +1 -1
index 11ff701ff4bb9..a3c73e972fce1 100644
@@ -192,7 +192,7 @@ int __init timer_of_init(struct device_node *np, struct timer_of *to)
 	}
 
 	if (!to->clkevt.name)
-		to->clkevt.name = np->name;
+		to->clkevt.name = np->full_name;
 
 	to->np = np;
 
Signature · 1 line
2.20.1
author maintainer
#40 lore raw

PATCH 37/134 dtc: Use pkg-config to locate libyaml

In reply to Sasha Levin

From: Pavel Modilaynen <[email protected]>

[ Upstream commit 067c650c456e758f933aaf87a202f841d34be269 ]

Using Makefile's wildcard with absolute path to detect the presence of libyaml results in false-positive detection when cross-compiling e.g. in yocto environment. The latter results in build error:

| scripts/dtc/yamltree.o: In function `yaml_propval_int':
| yamltree.c: undefined reference to `yaml_sequence_start_event_initialize'
| yamltree.c: undefined reference to `yaml_emitter_emit'
| yamltree.c: undefined reference to `yaml_scalar_event_initialize'

...
Use pkg-config to locate libyaml to address this scenario.

---
 scripts/dtc/Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
scripts/dtc/Makefile +2 -2
index 82160808765c3..b5a5b1c548c9b 100644
@@ -11,7 +11,7 @@ dtc-objs	+= dtc-lexer.lex.o dtc-parser.tab.o
 # Source files need to get at the userspace version of libfdt_env.h to compile
 HOST_EXTRACFLAGS := -I $(srctree)/$(src)/libfdt
 
-ifeq ($(wildcard /usr/include/yaml.h),)
+ifeq ($(shell pkg-config --exists yaml-0.1 2>/dev/null && echo yes),)
 ifneq ($(CHECK_DTBS),)
 $(error dtc needs libyaml for DT schema validation support. \
 	Install the necessary libyaml development package.)
@@ -19,7 +19,7 @@ endif
 HOST_EXTRACFLAGS += -DNO_YAML
 else
 dtc-objs	+= yamltree.o
-HOSTLDLIBS_dtc	:= -lyaml
+HOSTLDLIBS_dtc	:= $(shell pkg-config yaml-0.1 --libs)
 endif
 
 # Generated files need one more search path to include headers in source tree
Signature · 1 line
2.20.1
author maintainer
#41 lore raw

PATCH 38/134 selftests/powerpc: Skip tm-signal-sigreturn-nt if TM not available

In reply to Sasha Levin

From: Michael Ellerman <[email protected]>

[ Upstream commit 505127068d9b705a6cf335143239db91bfe7bbe2 ]

On systems where TM (Transactional Memory) is disabled the tm-signal-sigreturn-nt test causes a SIGILL:

test: tm_signal_sigreturn_nt
tags: git_version:7c202575ef63
!! child died by signal 4
failure: tm_signal_sigreturn_nt

We should skip the test if TM is not available.

Fixes 34642d70ac7e ("selftests/powerpc: Add checks for transactional sigreturn")
---
 tools/testing/selftests/powerpc/tm/tm-signal-sigreturn-nt.c | 4 ++++
 1 file changed, 4 insertions(+)
tools/testing/selftests/powerpc/tm/tm-signal-sigreturn-nt.c +4 -0
index 56fbf9f6bbf30..07c388147b75e 100644
@@ -10,10 +10,12 @@
  */
 
 #define _GNU_SOURCE
+#include <stdio.h>
 #include <stdlib.h>
 #include <signal.h>
 
 #include "utils.h"
+#include "tm.h"
 
 void trap_signal_handler(int signo, siginfo_t *si, void *uc)
 {
@@ -29,6 +31,8 @@ int tm_signal_sigreturn_nt(void)
 {
 	struct sigaction trap_sa;
 
+	SKIP_IF(!have_htm());
+
 	trap_sa.sa_flags = SA_SIGINFO;
 	trap_sa.sa_sigaction = trap_signal_handler;
 
Signature · 1 line
2.20.1
author maintainer
#42 lore raw

PATCH 39/134 powerpc/security/book3s64: Report L1TF status in sysfs

In reply to Sasha Levin

From: Anthony Steinhauser <[email protected]>

[ Upstream commit 8e6b6da91ac9b9ec5a925b6cb13f287a54bd547d ]

Some PowerPC CPUs are vulnerable to L1TF to the same extent as to Meltdown. It is also mitigated by flushing the L1D on privilege transition.

Currently the sysfs gives a false negative on L1TF on CPUs that I verified to be vulnerable, a Power9 Talos II Boston 004e 1202, PowerNV T2P9D01.

[mpe: Just have cpu_show_l1tf() call cpu_show_meltdown() directly]
---
 arch/powerpc/kernel/security.c | 5 +++++
 1 file changed, 5 insertions(+)
arch/powerpc/kernel/security.c +5 -0
index bd91dceb70105..298a2e3ad6f4c 100644
@@ -168,6 +168,11 @@ ssize_t cpu_show_meltdown(struct device *dev, struct device_attribute *attr, cha
 
 	return sprintf(buf, "Vulnerable\n");
 }
+
+ssize_t cpu_show_l1tf(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	return cpu_show_meltdown(dev, attr, buf);
+}
 #endif
 
 ssize_t cpu_show_spectre_v1(struct device *dev, struct device_attribute *attr, char *buf)
Signature · 1 line
2.20.1
author maintainer
#43 lore raw

PATCH 40/134 powerpc/book3s64/hash: Add cond_resched to avoid soft lockup warning

In reply to Sasha Levin

From: "Aneesh Kumar K.V" <[email protected]>

[ Upstream commit 16f6b67cf03cb43db7104acb2ca877bdc2606c92 ]

With large memory (8TB and more) hotplug, we can get soft lockup warnings as below. These were caused by a long loop without any explicit cond_resched which is a problem for !PREEMPT kernels.

Avoid this using cond_resched() while inserting hash page table entries. We already do similar cond_resched() in __add_pages(), see commit f64ac5e6e306 ("mm, memory_hotplug: add scheduling point to __add_pages").

rcu: 3-....: (24002 ticks this GP) idle=13e/1/0x4000000000000002 softirq=722/722 fqs=12001
(t=24003 jiffies g=4285 q=2002)
NMI backtrace for cpu 3

  CPU: 3 PID: 3870 Comm: ndctl Not tainted 5.3.0-197.18-default+ #2
  Call Trace:
    dump_stack+0xb0/0xf4 (unreliable)
    nmi_cpu_backtrace+0x124/0x130
    nmi_trigger_cpumask_backtrace+0x1ac/0x1f0
    arch_trigger_cpumask_backtrace+0x28/0x3c
    rcu_dump_cpu_stacks+0xf8/0x154
    rcu_sched_clock_irq+0x878/0xb40
    update_process_times+0x48/0x90
    tick_sched_handle.isra.16+0x4c/0x80
    tick_sched_timer+0x68/0xe0
    __hrtimer_run_queues+0x180/0x430
    hrtimer_interrupt+0x110/0x300
    timer_interrupt+0x108/0x2f0
    decrementer_common+0x114/0x120
  --- interrupt: 901 at arch_add_memory+0xc0/0x130
      LR = arch_add_memory+0x74/0x130
    memremap_pages+0x494/0x650
    devm_memremap_pages+0x3c/0xa0
    pmem_attach_disk+0x188/0x750
    nvdimm_bus_probe+0xac/0x2c0
    really_probe+0x148/0x570
    driver_probe_device+0x19c/0x1d0
    device_driver_attach+0xcc/0x100
    bind_store+0x134/0x1c0
    drv_attr_store+0x44/0x60
    sysfs_kf_write+0x64/0x90
    kernfs_fop_write+0x1a0/0x270
    __vfs_write+0x3c/0x70
    vfs_write+0xd0/0x260
    ksys_write+0xdc/0x130
    system_call+0x5c/0x68
---
 arch/powerpc/mm/book3s64/hash_utils.c | 1 +
 1 file changed, 1 insertion(+)
arch/powerpc/mm/book3s64/hash_utils.c +1 -0
index 6e5a769ebcb80..83c51a7d7eee6 100644
@@ -305,6 +305,7 @@ int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
 		if (ret < 0)
 			break;
 
+		cond_resched();
 #ifdef CONFIG_DEBUG_PAGEALLOC
 		if (debug_pagealloc_enabled() &&
 			(paddr >> PAGE_SHIFT) < linear_map_hash_count)
Signature · 1 line
2.20.1
author maintainer
#44 lore raw

PATCH 41/134 ext4: update direct I/O read lock pattern for IOCB_NOWAIT

In reply to Sasha Levin

From: Matthew Bobrowski <[email protected]>

[ Upstream commit 548feebec7e93e58b647dba70b3303dcb569c914 ]

This patch updates the lock pattern in ext4_direct_IO_read() to not block on inode lock in cases of IOCB_NOWAIT direct I/O reads. The locking condition implemented here is similar to that of 942491c9e6d6 ("xfs: fix AIM7 regression").

---
 fs/ext4/inode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
fs/ext4/inode.c +7 -1
index d691d1783ed62..207b8c23e53d9 100644
@@ -3827,7 +3827,13 @@ static ssize_t ext4_direct_IO_read(struct kiocb *iocb, struct iov_iter *iter)
 	 * writes & truncates and since we take care of writing back page cache,
 	 * we are protected against page writeback as well.
 	 */
-	inode_lock_shared(inode);
+	if (iocb->ki_flags & IOCB_NOWAIT) {
+		if (!inode_trylock_shared(inode))
+			return -EAGAIN;
+	} else {
+		inode_lock_shared(inode);
+	}
+
 	ret = filemap_write_and_wait_range(mapping, iocb->ki_pos,
 					   iocb->ki_pos + count - 1);
 	if (ret)
Signature · 1 line
2.20.1
author maintainer
#45 lore raw

PATCH 42/134 ext4: iomap that extends beyond EOF should be marked dirty

In reply to Sasha Levin

From: Matthew Bobrowski <[email protected]>

[ Upstream commit 2e9b51d78229d5145725a481bb5464ebc0a3f9b2 ]

This patch addresses what Dave Chinner had discovered and fixed within
commit: 7684e2c4384d. This changes does not have any user visible impact for ext4 as none of the current users of ext4_iomap_begin() that extend files depend on IOMAP_F_DIRTY.

When doing a direct IO that spans the current EOF, and there are written blocks beyond EOF that extend beyond the current write, the only metadata update that needs to be done is a file size extension.

However, we don't mark such iomaps as IOMAP_F_DIRTY to indicate that there is IO completion metadata updates required, and hence we may fail to correctly sync file size extensions made in IO completion when O_DSYNC writes are being used and the hardware supports FUA.

Hence when setting IOMAP_F_DIRTY, we need to also take into account whether the iomap spans the current EOF. If it does, then we need to mark it dirty so that IO completion will call generic_write_sync() to flush the inode size update to stable storage correctly.

---
 fs/ext4/inode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
fs/ext4/inode.c +7 -1
index 207b8c23e53d9..02313f10e8897 100644
@@ -3523,8 +3523,14 @@ retry:
 			return ret;
 	}
 
+	/*
+	 * Writes that span EOF might trigger an I/O size update on completion,
+	 * so consider them to be dirty for the purposes of O_DSYNC, even if
+	 * there is no other metadata changes being made or are pending here.
+	 */
 	iomap->flags = 0;
-	if (ext4_inode_datasync_dirty(inode))
+	if (ext4_inode_datasync_dirty(inode) ||
+	    offset + length > i_size_read(inode))
 		iomap->flags |= IOMAP_F_DIRTY;
 	iomap->bdev = inode->i_sb->s_bdev;
 	iomap->dax_dev = sbi->s_daxdev;
Signature · 1 line
2.20.1
author maintainer
#46 lore raw

PATCH 43/134 jbd2: Fix statistics for the number of logged blocks

In reply to Sasha Levin

From: Jan Kara <[email protected]>

[ Upstream commit 015c6033068208d6227612c878877919f3fcf6b6 ]

jbd2 statistics counting number of blocks logged in a transaction was wrong. It didn't count the commit block and more importantly it didn't count revoke descriptor blocks. Make sure these get properly counted.

---
 fs/jbd2/commit.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
fs/jbd2/commit.c +3 -1
index 132fb92098c71..c43591cd70f1c 100644
@@ -727,7 +727,6 @@ start_journal_io:
 				submit_bh(REQ_OP_WRITE, REQ_SYNC, bh);
 			}
 			cond_resched();
-			stats.run.rs_blocks_logged += bufs;
 
 			/* Force a new descriptor to be generated next
                            time round the loop. */
@@ -814,6 +813,7 @@ start_journal_io:
 		if (unlikely(!buffer_uptodate(bh)))
 			err = -EIO;
 		jbd2_unfile_log_bh(bh);
+		stats.run.rs_blocks_logged++;
 
 		/*
 		 * The list contains temporary buffer heads created by
@@ -859,6 +859,7 @@ start_journal_io:
 		BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile");
 		clear_buffer_jwrite(bh);
 		jbd2_unfile_log_bh(bh);
+		stats.run.rs_blocks_logged++;
 		__brelse(bh);		/* One for getblk */
 		/* AKPM: bforget here */
 	}
@@ -880,6 +881,7 @@ start_journal_io:
 	}
 	if (cbh)
 		err = journal_wait_on_commit_record(journal, cbh);
+	stats.run.rs_blocks_logged++;
 	if (jbd2_has_feature_async_commit(journal) &&
 	    journal->j_flags & JBD2_BARRIER) {
 		blkdev_issue_flush(journal->j_dev, GFP_NOFS, NULL);
Signature · 1 line
2.20.1
author maintainer
#47 lore raw

PATCH 44/134 scsi: tracing: Fix handling of TRANSFER LENGTH == 0 for READ(6) and WRITE(6)

In reply to Sasha Levin

From: Bart Van Assche <[email protected]>

[ Upstream commit f6b8540f40201bff91062dd64db8e29e4ddaaa9d ]

According to SBC-2 a TRANSFER LENGTH field of zero means that 256 logical blocks must be transferred. Make the SCSI tracing code follow SBC-2.

Fixes bf8162354233 ("[SCSI] add scsi trace core functions and put trace points")
Cc Christoph Hellwig <[email protected]>
Cc Hannes Reinecke <[email protected]>
Cc Douglas Gilbert <[email protected]>
---
 drivers/scsi/scsi_trace.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
drivers/scsi/scsi_trace.c +7 -4
index 0f17e7dac1b08..07a2425ffa2c2 100644
@@ -18,15 +18,18 @@ static const char *
 scsi_trace_rw6(struct trace_seq *p, unsigned char *cdb, int len)
 {
 	const char *ret = trace_seq_buffer_ptr(p);
-	sector_t lba = 0, txlen = 0;
+	u32 lba = 0, txlen;
 
 	lba |= ((cdb[1] & 0x1F) << 16);
 	lba |=  (cdb[2] << 8);
 	lba |=   cdb[3];
-	txlen = cdb[4];
+	/*
+	 * From SBC-2: a TRANSFER LENGTH field set to zero specifies that 256
+	 * logical blocks shall be read (READ(6)) or written (WRITE(6)).
+	 */
+	txlen = cdb[4] ? cdb[4] : 256;
 
-	trace_seq_printf(p, "lba=%llu txlen=%llu",
-			 (unsigned long long)lba, (unsigned long long)txlen);
+	trace_seq_printf(p, "lba=%u txlen=%u", lba, txlen);
 	trace_seq_putc(p, 0);
 
 	return ret;
Signature · 1 line
2.20.1
author maintainer
#48 lore raw

PATCH 45/134 scsi: lpfc: Fix unexpected error messages during RSCN handling

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit 2332e6e475b016e2026763f51333f84e2e6c57a3 ]

During heavy RCN activity and log_verbose = 0 we see these messages:

2754 PRLI failure DID:521245 Status:x9/xb2c00, data: x0 0231 RSCN timeout Data: x0 x3
0230 Unexpected timeout, hba link state x5

This is due to delayed RSCN activity.

Correct by avoiding the timeout thus the messages by restarting the discovery timeout whenever an rscn is received.

Filter PRLI responses such that severity depends on whether expected for the configuration or not. For example, PRLI errors on a fabric will be informational (they are expected), but Point-to-Point errors are not necessarily expected so they are raised to an error level.

---
 drivers/scsi/lpfc/lpfc_els.c | 21 +++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)
drivers/scsi/lpfc/lpfc_els.c +19 -2
index f293b48616ae9..4794a58deaf3c 100644
@@ -2236,6 +2236,7 @@ lpfc_cmpl_els_prli(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 	struct Scsi_Host  *shost = lpfc_shost_from_vport(vport);
 	IOCB_t *irsp;
 	struct lpfc_nodelist *ndlp;
+	char *mode;
 
 	/* we pass cmdiocb to state machine which needs rspiocb as well */
 	cmdiocb->context_un.rsp_iocb = rspiocb;
@@ -2273,8 +2274,17 @@ lpfc_cmpl_els_prli(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 			goto out;
 		}
 
+		/* If we don't send GFT_ID to Fabric, a PRLI error
+		 * could be expected.
+		 */
+		if ((vport->fc_flag & FC_FABRIC) ||
+		    (vport->cfg_enable_fc4_type != LPFC_ENABLE_BOTH))
+			mode = KERN_ERR;
+		else
+			mode = KERN_INFO;
+
 		/* PRLI failed */
-		lpfc_printf_vlog(vport, KERN_ERR, LOG_ELS,
+		lpfc_printf_vlog(vport, mode, LOG_ELS,
 				 "2754 PRLI failure DID:%06X Status:x%x/x%x, "
 				 "data: x%x\n",
 				 ndlp->nlp_DID, irsp->ulpStatus,
@@ -6455,7 +6465,7 @@ lpfc_els_rcv_rscn(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb,
 	uint32_t payload_len, length, nportid, *cmd;
 	int rscn_cnt;
 	int rscn_id = 0, hba_id = 0;
-	int i;
+	int i, tmo;
 
 	pcmd = (struct lpfc_dmabuf *) cmdiocb->context2;
 	lp = (uint32_t *) pcmd->virt;
@@ -6561,6 +6571,13 @@ lpfc_els_rcv_rscn(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb,
 
 		spin_lock_irq(shost->host_lock);
 		vport->fc_flag |= FC_RSCN_DEFERRED;
+
+		/* Restart disctmo if its already running */
+		if (vport->fc_flag & FC_DISC_TMO) {
+			tmo = ((phba->fc_ratov * 3) + 3);
+			mod_timer(&vport->fc_disctmo,
+				  jiffies + msecs_to_jiffies(1000 * tmo));
+		}
 		if ((rscn_cnt < FC_MAX_HOLD_RSCN) &&
 		    !(vport->fc_flag & FC_RSCN_DISCOVERY)) {
 			vport->fc_flag |= FC_RSCN_MODE;
Signature · 1 line
2.20.1
author maintainer
#49 lore raw

PATCH 46/134 scsi: lpfc: Fix duplicate unreg_rpi error in port offline flow

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit 7cfd5639d99bec0d27af089d0c8c114330e43a72 ]

If the driver receives a login that is later then LOGO'd by the remote port (aka ndlp), the driver, upon the completion of the LOGO ACC transmission, will logout the node and unregister the rpi that is being used for the node. As part of the unreg, the node's rpi value is replaced by the LPFC_RPI_ALLOC_ERROR value. If the port is subsequently offlined, the offline walks the nodes and ensures they are logged out, which possibly entails unreg'ing their rpi values. This path does not validate the node's rpi value, thus doesn't detect that it has been unreg'd already. The replaced rpi value is then used when accessing the rpi bitmask array which tracks active rpi values. As the LPFC_RPI_ALLOC_ERROR value is not a valid index for the bitmask, it may fault the system.

Revise the rpi release code to detect when the rpi value is the replaced RPI_ALLOC_ERROR value and ignore further release steps.

---
 drivers/scsi/lpfc/lpfc_sli.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/scsi/lpfc/lpfc_sli.c +7 -0
index 995a2b56a35ee..cf95249985bc4 100644
@@ -18184,6 +18184,13 @@ lpfc_sli4_alloc_rpi(struct lpfc_hba *phba)
 static void
 __lpfc_sli4_free_rpi(struct lpfc_hba *phba, int rpi)
 {
+	/*
+	 * if the rpi value indicates a prior unreg has already
+	 * been done, skip the unreg.
+	 */
+	if (rpi == LPFC_RPI_ALLOC_ERROR)
+		return;
+
 	if (test_and_clear_bit(rpi, phba->sli4_hba.rpi_bmask)) {
 		phba->sli4_hba.rpi_count--;
 		phba->sli4_hba.max_cfg_param.rpi_used--;
Signature · 1 line
2.20.1
author maintainer
#50 lore raw

PATCH 47/134 f2fs: fix to update dir's i_pino during cross_rename

In reply to Sasha Levin

From: Chao Yu <[email protected]>

[ Upstream commit 2a60637f06ac94869b2e630eaf837110d39bf291 ]

As Eric reported:

RENAME_EXCHANGE support was just added to fsstress in xfstests:

	commit 65dfd40a97b6bbbd2a22538977bab355c5bc0f06
	Author: kaixuxia <[email protected]>
	Date:   Thu Oct 31 14:41:48 2019 +0800

	    fsstress: add EXCHANGE renameat2 support

This is causing xfstest generic/579 to fail due to fsck.f2fs reporting errors. I'm not sure what the problem is, but it still happens even with all the fs-verity stuff in the test commented out, so that the test just runs fsstress.

generic/579 23s ... [10:02:25]

[    7.745370] run fstests generic/579 at 2019-11-04 10:02:25

_check_generic_filesystem: filesystem on /dev/vdc is inconsistent (see /results/f2fs/results-default/generic/579.full for details)
[10:02:47]
Ran: generic/579
Failures: generic/579
Failed 1 of 1 tests
Xunit report: /results/f2fs/results-default/result.xml

Here's the contents of 579.full:

_check_generic_filesystem: filesystem on /dev/vdc is inconsistent
*** fsck.f2fs output ***
[ASSERT] (__chk_dots_dentries:1378) --> Bad inode number[0x24] for '..', parent parent ino is [0xd10]

The root cause is that we forgot to update directory's i_pino during cross_rename, fix it.

Fixes 32f9bc25cbda0 ("f2fs: support ->rename2()")
---
 fs/f2fs/namei.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)
fs/f2fs/namei.c +12 -3
index 4faf06e8bf899..a1c507b0b4ac4 100644
@@ -981,7 +981,8 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (!old_dir_entry || whiteout)
 		file_lost_pino(old_inode);
 	else
-		F2FS_I(old_inode)->i_pino = new_dir->i_ino;
+		/* adjust dir's i_pino to pass fsck check */
+		f2fs_i_pino_write(old_inode, new_dir->i_ino);
 	up_write(&F2FS_I(old_inode)->i_sem);
 
 	old_inode->i_ctime = current_time(old_inode);
@@ -1141,7 +1142,11 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	f2fs_set_link(old_dir, old_entry, old_page, new_inode);
 
 	down_write(&F2FS_I(old_inode)->i_sem);
-	file_lost_pino(old_inode);
+	if (!old_dir_entry)
+		file_lost_pino(old_inode);
+	else
+		/* adjust dir's i_pino to pass fsck check */
+		f2fs_i_pino_write(old_inode, new_dir->i_ino);
 	up_write(&F2FS_I(old_inode)->i_sem);
 
 	old_dir->i_ctime = current_time(old_dir);
@@ -1156,7 +1161,11 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	f2fs_set_link(new_dir, new_entry, new_page, old_inode);
 
 	down_write(&F2FS_I(new_inode)->i_sem);
-	file_lost_pino(new_inode);
+	if (!new_dir_entry)
+		file_lost_pino(new_inode);
+	else
+		/* adjust dir's i_pino to pass fsck check */
+		f2fs_i_pino_write(new_inode, old_dir->i_ino);
 	up_write(&F2FS_I(new_inode)->i_sem);
 
 	new_dir->i_ctime = current_time(new_dir);
Signature · 8 lines
2.20.1



_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
author maintainer
#51 lore raw

PATCH 48/134 clk: qcom: smd: Add missing pnoc clock

In reply to Sasha Levin

From: Jeffrey Hugo <[email protected]>

[ Upstream commit ba1d366de261981c0dd04fac44d2ce3a5eba2eaa ]

When MSM8998 support was added, and analysis was done to determine what clocks would be consumed. That analysis had a flaw, which caused the pnoc to be skipped. The pnoc clock needs to be on to access the uart for the console. The clock is on from boot, but has no consumer votes in the RPM. When we attempt to boot the modem, it causes the RPM to turn off pnoc, which kills our access to the console and causes CPU hangs.

We need pnoc to be defined, so that clk_smd_rpm_handoff() will put in an implicit vote for linux and prevent issues when booting modem. Hopefully pnoc can be consumed by the interconnect framework in future so that Linux can rely on explicit votes.

Fixes 6131dc81211c ("clk: qcom: smd: Add support for MSM8998 rpm clocks")
---
 drivers/clk/qcom/clk-smd-rpm.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/clk/qcom/clk-smd-rpm.c +3 -0
index fef5e81570614..930fa4a4c52a8 100644
@@ -648,6 +648,7 @@ static const struct rpm_smd_clk_desc rpm_clk_qcs404 = {
 };
 
 /* msm8998 */
+DEFINE_CLK_SMD_RPM(msm8998, pcnoc_clk, pcnoc_a_clk, QCOM_SMD_RPM_BUS_CLK, 0);
 DEFINE_CLK_SMD_RPM(msm8998, snoc_clk, snoc_a_clk, QCOM_SMD_RPM_BUS_CLK, 1);
 DEFINE_CLK_SMD_RPM(msm8998, cnoc_clk, cnoc_a_clk, QCOM_SMD_RPM_BUS_CLK, 2);
 DEFINE_CLK_SMD_RPM(msm8998, ce1_clk, ce1_a_clk, QCOM_SMD_RPM_CE_CLK, 0);
@@ -670,6 +671,8 @@ DEFINE_CLK_SMD_RPM_XO_BUFFER_PINCTRL(msm8998, rf_clk2_pin, rf_clk2_a_pin, 5);
 DEFINE_CLK_SMD_RPM_XO_BUFFER(msm8998, rf_clk3, rf_clk3_a, 6);
 DEFINE_CLK_SMD_RPM_XO_BUFFER_PINCTRL(msm8998, rf_clk3_pin, rf_clk3_a_pin, 6);
 static struct clk_smd_rpm *msm8998_clks[] = {
+	[RPM_SMD_PCNOC_CLK] = &msm8998_pcnoc_clk,
+	[RPM_SMD_PCNOC_A_CLK] = &msm8998_pcnoc_a_clk,
 	[RPM_SMD_SNOC_CLK] = &msm8998_snoc_clk,
 	[RPM_SMD_SNOC_A_CLK] = &msm8998_snoc_a_clk,
 	[RPM_SMD_CNOC_CLK] = &msm8998_cnoc_clk,
Signature · 1 line
2.20.1
author maintainer
#52 lore raw

PATCH 49/134 clk: qcom: Allow constant ratio freq tables for rcg

In reply to Sasha Levin

From: Jeffrey Hugo <[email protected]>

[ Upstream commit efd164b5520afd6fb2883b68e0d408a7de29c491 ]

Some RCGs (the gfx_3d_src_clk in msm8998 for example) are basically just some constant ratio from the input across the entire frequency range. It would be great if we could specify the frequency table as a single entry constant ratio instead of a long list, ie:

	{ .src = P_GPUPLL0_OUT_EVEN, .pre_div = 3 },
        { }

So, lets support that.

We need to fix a corner case in qcom_find_freq() where if the freq table is non-null, but has no frequencies, we end up returning an "entry" before the table array, which is bad. Then, we need ignore the freq from the table, and instead base everything on the requested freq.

---
 drivers/clk/qcom/clk-rcg2.c | 2 ++
 drivers/clk/qcom/common.c   | 3 +++
 2 files changed, 5 insertions(+)
drivers/clk/qcom/clk-rcg2.c +2 -0
index b98b81ef43a14..5a89ed88cc27a 100644
@@ -220,6 +220,8 @@ static int _freq_tbl_determine_rate(struct clk_hw *hw, const struct freq_tbl *f,
 	if (clk_flags & CLK_SET_RATE_PARENT) {
 		rate = f->freq;
 		if (f->pre_div) {
+			if (!rate)
+				rate = req->rate;
 			rate /= 2;
 			rate *= f->pre_div + 1;
 		}
drivers/clk/qcom/common.c +3 -0
index 28ddc747d7035..bdeacebbf0e47 100644
@@ -29,6 +29,9 @@ struct freq_tbl *qcom_find_freq(const struct freq_tbl *f, unsigned long rate)
 	if (!f)
 		return NULL;
 
+	if (!f->freq)
+		return f;
+
 	for (; f->freq; f++)
 		if (rate <= f->freq)
 			return f;
Signature · 1 line
2.20.1
author maintainer
#53 lore raw

PATCH 50/134 clk: clk-gpio: propagate rate change to parent

In reply to Sasha Levin

From: Michael Hennerich <[email protected]>

[ Upstream commit fc59462c5ce60da119568fac325c92fc6b7c6175 ]

For an external clock source, which is gated via a GPIO, the rate change should typically be propagated to the parent clock.

The situation where we are requiring this propagation, is when an external clock is connected to override an internal clock (which typically has a fixed rate). The external clock can have a different rate than the internal one, and may also be variable, thus requiring the rate propagation.

This rate change wasn't propagated until now, and it's unclear about cases where this shouldn't be propagated. Thus, it's unclear whether this is fixing a bug, or extending the current driver behavior. Also, it's unsure about whether this may break any existing setups; in the case that it does, a device-tree property may be added to disable this flag.

---
 drivers/clk/clk-gpio.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clk/clk-gpio.c +1 -1
index 9d930edd65162..13304cf5f2a8f 100644
@@ -280,7 +280,7 @@ static int gpio_clk_driver_probe(struct platform_device *pdev)
 	else
 		clk = clk_register_gpio_gate(&pdev->dev, node->name,
 				parent_names ?  parent_names[0] : NULL, gpiod,
-				0);
+				CLK_SET_RATE_PARENT);
 	if (IS_ERR(clk))
 		return PTR_ERR(clk);
 
Signature · 1 line
2.20.1
author maintainer
#54 lore raw

PATCH 51/134 scsi: qla2xxx: Fix a dma_pool_free() call

In reply to Sasha Levin

From: Bart Van Assche <[email protected]>

[ Upstream commit 162b805e38327135168cb0938bd37b131b481cb0 ]

This patch fixes the following kernel warning:

DMA-API: qla2xxx 0000:00:0a.0: device driver frees DMA memory with different size [device address=0x00000000c7b60000] [map size=4088 bytes] [unmap size=512 bytes]

WARNING: CPU: 3 PID: 1122 at kernel/dma/debug.c:1021 check_unmap+0x4d0/0xbd0
CPU: 3 PID: 1122 Comm: rmmod Tainted: G           O      5.4.0-rc1-dbg+ #1
RIP: 0010:check_unmap+0x4d0/0xbd0
Call Trace:
 debug_dma_free_coherent+0x123/0x173
 dma_free_attrs+0x76/0xe0
 qla2x00_mem_free+0x329/0xc40 [qla2xxx_scst]
 qla2x00_free_device+0x170/0x1c0 [qla2xxx_scst]
 qla2x00_remove_one+0x4f0/0x6d0 [qla2xxx_scst]
 pci_device_remove+0xd5/0x1f0
 device_release_driver_internal+0x159/0x280
 driver_detach+0x8b/0xf2
 bus_remove_driver+0x9a/0x15a
 driver_unregister+0x51/0x70
 pci_unregister_driver+0x2d/0x130
 qla2x00_module_exit+0x1c/0xbc [qla2xxx_scst]
 __x64_sys_delete_module+0x22a/0x300
 do_syscall_64+0x6f/0x2e0
 entry_SYSCALL_64_after_hwframe+0x49/0xbe
Fixes 3f006ac342c0 ("scsi: qla2xxx: Secure flash update support for ISP28XX") # v5.2-rc1~130^2~270.
Cc Michael Hernandez <[email protected]>
Cc Himanshu Madhani <[email protected]>
---
 drivers/scsi/qla2xxx/qla_os.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/qla2xxx/qla_os.c +2 -1
index 726ad4cbf4a64..9fc0789c9d3c3 100644
@@ -4666,7 +4666,8 @@ qla2x00_mem_free(struct qla_hw_data *ha)
 	ha->sfp_data = NULL;
 
 	if (ha->flt)
-		dma_free_coherent(&ha->pdev->dev, SFP_DEV_SIZE,
+		dma_free_coherent(&ha->pdev->dev,
+		    sizeof(struct qla_flt_header) + FLT_REGIONS_SIZE,
 		    ha->flt, ha->flt_dma);
 	ha->flt = NULL;
 	ha->flt_dma = 0;
Signature · 1 line
2.20.1
author maintainer
#55 lore raw

PATCH 52/134 irqchip/irq-bcm7038-l1: Enable parent IRQ if necessary

In reply to Sasha Levin

From: Florian Fainelli <[email protected]>

[ Upstream commit 27eebb60357ed5aa6659442f92907c0f7368d6ae ]

If the 'brcm,irq-can-wake' property is specified, make sure we also enable the corresponding parent interrupt we are attached to.

---
 drivers/irqchip/irq-bcm7038-l1.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/irqchip/irq-bcm7038-l1.c +4 -0
index fc75c61233aaf..58bec21269661 100644
@@ -281,6 +281,10 @@ static int __init bcm7038_l1_init_one(struct device_node *dn,
 		pr_err("failed to map parent interrupt %d\n", parent_irq);
 		return -EINVAL;
 	}
+
+	if (of_property_read_bool(dn, "brcm,irq-can-wake"))
+		enable_irq_wake(parent_irq);
+
 	irq_set_chained_handler_and_data(parent_irq, bcm7038_l1_irq_handle,
 					 intc);
 
Signature · 1 line
2.20.1
author maintainer
#56 lore raw

PATCH 53/134 irqchip: ingenic: Error out if IRQ domain creation failed

In reply to Sasha Levin

From: Paul Cercueil <[email protected]>

[ Upstream commit 52ecc87642f273a599c9913b29fd179c13de457b ]

If we cannot create the IRQ domain, the driver should fail to probe instead of succeeding with just a warning message.

---
 drivers/irqchip/irq-ingenic.c | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)
drivers/irqchip/irq-ingenic.c +10 -5
index f126255b3260c..dda512dfe2c17 100644
@@ -108,6 +108,14 @@ static int __init ingenic_intc_of_init(struct device_node *node,
 		goto out_unmap_irq;
 	}
 
+	domain = irq_domain_add_legacy(node, num_chips * 32,
+				       JZ4740_IRQ_BASE, 0,
+				       &irq_domain_simple_ops, NULL);
+	if (!domain) {
+		err = -ENOMEM;
+		goto out_unmap_base;
+	}
+
 	for (i = 0; i < num_chips; i++) {
 		/* Mask all irqs */
 		writel(0xffffffff, intc->base + (i * CHIP_SIZE) +
@@ -134,14 +142,11 @@ static int __init ingenic_intc_of_init(struct device_node *node,
 				       IRQ_NOPROBE | IRQ_LEVEL);
 	}
 
-	domain = irq_domain_add_legacy(node, num_chips * 32, JZ4740_IRQ_BASE, 0,
-				       &irq_domain_simple_ops, NULL);
-	if (!domain)
-		pr_warn("unable to register IRQ domain\n");
-
 	setup_irq(parent_irq, &intc_cascade_action);
 	return 0;
 
+out_unmap_base:
+	iounmap(intc->base);
 out_unmap_irq:
 	irq_dispose_mapping(parent_irq);
 out_free:
Signature · 1 line
2.20.1
author maintainer
#57 lore raw

PATCH 54/134 dma-direct: check for overflows on 32 bit DMA addresses

In reply to Sasha Levin

From: Nicolas Saenz Julienne <[email protected]>

[ Upstream commit b12d66278dd627cbe1ea7c000aa4715aaf8830c8 ]

As seen on the new Raspberry Pi 4 and sta2x11's DMA implementation it is possible for a device configured with 32 bit DMA addresses and a partial DMA mapping located at the end of the address space to overflow. It happens when a higher physical address, not DMAable, is translated to it's DMA counterpart.

For example the Raspberry Pi 4, configurable up to 4 GB of memory, has an interconnect capable of addressing the lower 1 GB of physical memory with a DMA offset of 0xc0000000. It transpires that, any attempt to translate physical addresses higher than the first GB will result in an overflow which dma_capable() can't detect as it only checks for addresses bigger then the maximum allowed DMA address.

Fix this by verifying in dma_capable() if the DMA address range provided is at any point lower than the minimum possible DMA address on the bus.

---
 include/linux/dma-direct.h | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)
include/linux/dma-direct.h +10 -2
index adf993a3bd580..6a18a97b76a87 100644
@@ -3,8 +3,11 @@
 #define _LINUX_DMA_DIRECT_H 1
 
 #include <linux/dma-mapping.h>
+#include <linux/memblock.h> /* for min_low_pfn */
 #include <linux/mem_encrypt.h>
 
+static inline dma_addr_t phys_to_dma(struct device *dev, phys_addr_t paddr);
+
 #ifdef CONFIG_ARCH_HAS_PHYS_TO_DMA
 #include <asm/dma-direct.h>
 #else
@@ -24,11 +27,16 @@ static inline phys_addr_t __dma_to_phys(struct device *dev, dma_addr_t dev_addr)
 
 static inline bool dma_capable(struct device *dev, dma_addr_t addr, size_t size)
 {
+	dma_addr_t end = addr + size - 1;
+
 	if (!dev->dma_mask)
 		return false;
 
-	return addr + size - 1 <=
-		min_not_zero(*dev->dma_mask, dev->bus_dma_mask);
+	if (!IS_ENABLED(CONFIG_ARCH_DMA_ADDR_T_64BIT) &&
+	    min(addr, end) < phys_to_dma(dev, PFN_PHYS(min_low_pfn)))
+		return false;
+
+	return end <= min_not_zero(*dev->dma_mask, dev->bus_dma_mask);
 }
 #endif /* !CONFIG_ARCH_HAS_PHYS_TO_DMA */
 
Signature · 1 line
2.20.1
author maintainer
#58 lore raw

PATCH 55/134 mfd: mfd-core: Honour Device Tree's request to disable a child-device

In reply to Sasha Levin

From: Lee Jones <[email protected]>

[ Upstream commit 6b5c350648b857047b47acf74a57087ad27d6183 ]

Until now, MFD has assumed all child devices passed to it (via mfd_cells) are to be registered. It does not take into account requests from Device Tree and the like to disable child devices on a per-platform basis.

Well now it does.

---
 drivers/mfd/mfd-core.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/mfd/mfd-core.c +5 -0
index 23276a80e3b48..b0afefdc9eac8 100644
@@ -174,6 +174,11 @@ static int mfd_add_device(struct device *parent, int id,
 	if (parent->of_node && cell->of_compatible) {
 		for_each_child_of_node(parent->of_node, np) {
 			if (of_device_is_compatible(np, cell->of_compatible)) {
+				if (!of_device_is_available(np)) {
+					/* Ignore disabled devices error free */
+					ret = 0;
+					goto fail_alias;
+				}
 				pdev->dev.of_node = np;
 				pdev->dev.fwnode = &np->fwnode;
 				break;
Signature · 1 line
2.20.1
author maintainer
#59 lore raw

PATCH 56/134 fs/quota: handle overflows of sysctl fs.quota.* and report as unsigned long

In reply to Sasha Levin

From: Konstantin Khlebnikov <[email protected]>

[ Upstream commit 6fcbcec9cfc7b3c6a2c1f1a23ebacedff7073e0a ]

Quota statistics counted as 64-bit per-cpu counter. Reading sums per-cpu fractions as signed 64-bit int, filters negative values and then reports lower half as signed 32-bit int.

Result may looks like:

fs.quota.allocated_dquots = 22327
fs.quota.cache_hits = -489852115
fs.quota.drops = -487288718
fs.quota.free_dquots = 22083
fs.quota.lookups = -486883485
fs.quota.reads = 22327
fs.quota.syncs = 335064
fs.quota.writes = 3088689

Values bigger than 2^31-1 reported as negative.

All counters except "allocated_dquots" and "free_dquots" are monotonic, thus they should be reported as is without filtering negative values.

Kernel doesn't have generic helper for 64-bit sysctl yet, let's use at least unsigned long.

---
 fs/quota/dquot.c      | 29 +++++++++++++++++------------
 include/linux/quota.h |  2 +-
 2 files changed, 18 insertions(+), 13 deletions(-)
fs/quota/dquot.c +17 -12
index 6e826b454082c..fa6ec4f96791d 100644
@@ -2860,68 +2860,73 @@ EXPORT_SYMBOL(dquot_quotactl_sysfile_ops);
 static int do_proc_dqstats(struct ctl_table *table, int write,
 		     void __user *buffer, size_t *lenp, loff_t *ppos)
 {
-	unsigned int type = (int *)table->data - dqstats.stat;
+	unsigned int type = (unsigned long *)table->data - dqstats.stat;
+	s64 value = percpu_counter_sum(&dqstats.counter[type]);
+
+	/* Filter negative values for non-monotonic counters */
+	if (value < 0 && (type == DQST_ALLOC_DQUOTS ||
+			  type == DQST_FREE_DQUOTS))
+		value = 0;
 
 	/* Update global table */
-	dqstats.stat[type] =
-			percpu_counter_sum_positive(&dqstats.counter[type]);
-	return proc_dointvec(table, write, buffer, lenp, ppos);
+	dqstats.stat[type] = value;
+	return proc_doulongvec_minmax(table, write, buffer, lenp, ppos);
 }
 
 static struct ctl_table fs_dqstats_table[] = {
 	{
 		.procname	= "lookups",
 		.data		= &dqstats.stat[DQST_LOOKUPS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "drops",
 		.data		= &dqstats.stat[DQST_DROPS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "reads",
 		.data		= &dqstats.stat[DQST_READS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "writes",
 		.data		= &dqstats.stat[DQST_WRITES],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "cache_hits",
 		.data		= &dqstats.stat[DQST_CACHE_HITS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "allocated_dquots",
 		.data		= &dqstats.stat[DQST_ALLOC_DQUOTS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "free_dquots",
 		.data		= &dqstats.stat[DQST_FREE_DQUOTS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
 	{
 		.procname	= "syncs",
 		.data		= &dqstats.stat[DQST_SYNCS],
-		.maxlen		= sizeof(int),
+		.maxlen		= sizeof(unsigned long),
 		.mode		= 0444,
 		.proc_handler	= do_proc_dqstats,
 	},
include/linux/quota.h +1 -1
index f32dd270b8e3f..27aab84fcbaac 100644
@@ -263,7 +263,7 @@ enum {
 };
 
 struct dqstats {
-	int stat[_DQST_DQSTAT_LAST];
+	unsigned long stat[_DQST_DQSTAT_LAST];
 	struct percpu_counter counter[_DQST_DQSTAT_LAST];
 };
 
Signature · 1 line
2.20.1
author maintainer
#60 lore raw

PATCH 57/134 iommu/arm-smmu-v3: Don't display an error when IRQ lines are missing

In reply to Sasha Levin

From: Jean-Philippe Brucker <[email protected]>

[ Upstream commit f7aff1a93f52047739af31072de0ad8d149641f3 ]

Since commit 7723f4c5ecdb ("driver core: platform: Add an error message to platform_get_irq*()"), platform_get_irq_byname() displays an error when the IRQ isn't found. Since the SMMUv3 driver uses that function to query which interrupt method is available, the message is now displayed during boot for any SMMUv3 that doesn't implement the combined interrupt, or that implements MSIs.

[   20.700337] arm-smmu-v3 arm-smmu-v3.7.auto: IRQ combined not found
[   20.706508] arm-smmu-v3 arm-smmu-v3.7.auto: IRQ eventq not found
[   20.712503] arm-smmu-v3 arm-smmu-v3.7.auto: IRQ priq not found
[   20.718325] arm-smmu-v3 arm-smmu-v3.7.auto: IRQ gerror not found

Use platform_get_irq_byname_optional() to avoid displaying a spurious error.

Fixes 7723f4c5ecdb ("driver core: platform: Add an error message to platform_get_irq*()")
---
 drivers/iommu/arm-smmu-v3.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/iommu/arm-smmu-v3.c +4 -4
index 8da93e730d6fd..ed90361b84dc7 100644
@@ -3611,19 +3611,19 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
 
 	/* Interrupt lines */
 
-	irq = platform_get_irq_byname(pdev, "combined");
+	irq = platform_get_irq_byname_optional(pdev, "combined");
 	if (irq > 0)
 		smmu->combined_irq = irq;
 	else {
-		irq = platform_get_irq_byname(pdev, "eventq");
+		irq = platform_get_irq_byname_optional(pdev, "eventq");
 		if (irq > 0)
 			smmu->evtq.q.irq = irq;
 
-		irq = platform_get_irq_byname(pdev, "priq");
+		irq = platform_get_irq_byname_optional(pdev, "priq");
 		if (irq > 0)
 			smmu->priq.q.irq = irq;
 
-		irq = platform_get_irq_byname(pdev, "gerror");
+		irq = platform_get_irq_byname_optional(pdev, "gerror");
 		if (irq > 0)
 			smmu->gerr_irq = irq;
 	}
Signature · 1 line
2.20.1
author maintainer
#61 lore raw

PATCH 58/134 i2c: stm32f7: fix & reorder remove & probe error handling

In reply to Sasha Levin

From: Alain Volmat <[email protected]>

[ Upstream commit 53aaaa5d9b1e95eb40e877fbffa6f964a8394bb7 ]

Add missing dma channels free calls in case of error during probe and reorder the remove function so that dma channels are freed after the i2c adapter is deleted.
Overall, reorder the remove function so that probe error handling order and remove function order are same.

Fixes 7ecc8cfde553 ("i2c: i2c-stm32f7: Add DMA support")
---
 drivers/i2c/busses/i2c-stm32f7.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
drivers/i2c/busses/i2c-stm32f7.c +12 -7
index b24e7b937f210..84cfed17ff4f5 100644
@@ -1985,6 +1985,11 @@ pm_disable:
 	pm_runtime_set_suspended(i2c_dev->dev);
 	pm_runtime_dont_use_autosuspend(i2c_dev->dev);
 
+	if (i2c_dev->dma) {
+		stm32_i2c_dma_free(i2c_dev->dma);
+		i2c_dev->dma = NULL;
+	}
+
 clk_free:
 	clk_disable_unprepare(i2c_dev->clk);
 
@@ -1995,21 +2000,21 @@ static int stm32f7_i2c_remove(struct platform_device *pdev)
 {
 	struct stm32f7_i2c_dev *i2c_dev = platform_get_drvdata(pdev);
 
-	if (i2c_dev->dma) {
-		stm32_i2c_dma_free(i2c_dev->dma);
-		i2c_dev->dma = NULL;
-	}
-
 	i2c_del_adapter(&i2c_dev->adap);
 	pm_runtime_get_sync(i2c_dev->dev);
 
-	clk_disable_unprepare(i2c_dev->clk);
-
 	pm_runtime_put_noidle(i2c_dev->dev);
 	pm_runtime_disable(i2c_dev->dev);
 	pm_runtime_set_suspended(i2c_dev->dev);
 	pm_runtime_dont_use_autosuspend(i2c_dev->dev);
 
+	if (i2c_dev->dma) {
+		stm32_i2c_dma_free(i2c_dev->dma);
+		i2c_dev->dma = NULL;
+	}
+
+	clk_disable_unprepare(i2c_dev->clk);
+
 	return 0;
 }
 
Signature · 1 line
2.20.1
author maintainer
#62 lore raw

PATCH 59/134 iomap: fix return value of iomap_dio_bio_actor on 32bit systems

In reply to Sasha Levin

From: Jan Stancek <[email protected]>

[ Upstream commit e9f930ac88a8936ccc2d021110c98810cf5aa810 ]

Naresh reported LTP diotest4 failing for 32bit x86 and arm -next kernels on ext4. Same problem exists in 5.4-rc7 on xfs.

The failure comes down to:
openat(AT_FDCWD, "testdata-4.5918", O_RDWR|O_DIRECT) = 4 mmap2(NULL, 4096, PROT_READ, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0xb7f7b000
read(4, 0xb7f7b000, 4096) = 0 // expects -EFAULT

Problem is conversion at iomap_dio_bio_actor() return. Ternary operator has a return type and an attempt is made to convert each of operands to the type of the other. In this case "ret" (int) is converted to type of "copied" (unsigned long). Both have size of 4 bytes:

    size_t copied = 0;
    int ret = -14;
    long long actor_ret = copied ? copied : ret;

    On x86_64: actor_ret == -14;
    On x86   : actor_ret == 4294967282

Replace ternary operator with 2 return statements to avoid this unwanted conversion.

Fixes 4721a6010990 ("iomap: dio data corruption and spurious errors when pipes fill")
---
 fs/iomap/direct-io.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
fs/iomap/direct-io.c +3 -1
index 1fc28c2da2790..7c58f51d7da74 100644
@@ -318,7 +318,9 @@ zero_tail:
 		if (pad)
 			iomap_dio_zero(dio, iomap, pos, fs_block_size - pad);
 	}
-	return copied ? copied : ret;
+	if (copied)
+		return copied;
+	return ret;
 }
 
 static loff_t
Signature · 1 line
2.20.1
author maintainer
#63 lore raw

PATCH 60/134 Input: ili210x - handle errors from input_mt_init_slots()

In reply to Sasha Levin

From: Dmitry Torokhov <[email protected]>

[ Upstream commit 43f06a4c639de8ee89fc348a9a3ecd70320a04dd ]

input_mt_init_slots() may fail and we need to handle such failures.

---
 drivers/input/touchscreen/ili210x.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/input/touchscreen/ili210x.c +6 -1
index e9006407c9bc0..f4ebdab062806 100644
@@ -334,7 +334,12 @@ static int ili210x_i2c_probe(struct i2c_client *client,
 	input_set_abs_params(input, ABS_MT_POSITION_X, 0, 0xffff, 0, 0);
 	input_set_abs_params(input, ABS_MT_POSITION_Y, 0, 0xffff, 0, 0);
 	touchscreen_parse_properties(input, true, &priv->prop);
-	input_mt_init_slots(input, priv->max_touches, INPUT_MT_DIRECT);
+
+	error = input_mt_init_slots(input, priv->max_touches, INPUT_MT_DIRECT);
+	if (error) {
+		dev_err(dev, "Unable to set up slots, err: %d\n", error);
+		return error;
+	}
 
 	error = devm_add_action(dev, ili210x_cancel_work, priv);
 	if (error)
Signature · 1 line
2.20.1
author maintainer
#64 lore raw

PATCH 61/134 scsi: lpfc: fix: Coverity: lpfc_cmpl_els_rsp(): Null pointer dereferences

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit 6c6d59e0fe5b86cf273d6d744a6a9768c4ecc756 ]

Coverity reported the following:

*** CID 101747: Null pointer dereferences (FORWARD_NULL) /drivers/scsi/lpfc/lpfc_els.c: 4439 in lpfc_cmpl_els_rsp()

4433     			kfree(mp);
4434     		}
4435     		mempool_free(mbox, phba->mbox_mem_pool);
4436     	}
4437     out:
4438     	if (ndlp && NLP_CHK_NODE_ACT(ndlp)) {
vvv     CID 101747:  Null pointer dereferences  (FORWARD_NULL)
vvv     Dereferencing null pointer "shost".
4439     		spin_lock_irq(shost->host_lock);
4440     		ndlp->nlp_flag &= ~(NLP_ACC_REGLOGIN | NLP_RM_DFLT_RPI);
4441     		spin_unlock_irq(shost->host_lock);

4442
4443 /* If the node is not being used by another discovery thread,
4444 * and we are sending a reject, we are done with it.

Fix by adding a check for non-null shost in line 4438. The scenario when shost is set to null is when ndlp is null. As such, the ndlp check present was sufficient. But better safe than sorry so add the shost check.

Addresses-Coverity-ID: 101747 ("Null pointer dereferences")

Fixes 2e0fef85e098 ("[SCSI] lpfc: NPIV: split ports")
---
 drivers/scsi/lpfc/lpfc_els.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/lpfc/lpfc_els.c +1 -1
index 4794a58deaf3c..66f8867dd8377 100644
@@ -4440,7 +4440,7 @@ lpfc_cmpl_els_rsp(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb,
 		mempool_free(mbox, phba->mbox_mem_pool);
 	}
 out:
-	if (ndlp && NLP_CHK_NODE_ACT(ndlp)) {
+	if (ndlp && NLP_CHK_NODE_ACT(ndlp) && shost) {
 		spin_lock_irq(shost->host_lock);
 		ndlp->nlp_flag &= ~(NLP_ACC_REGLOGIN | NLP_RM_DFLT_RPI);
 		spin_unlock_irq(shost->host_lock);
Signature · 1 line
2.20.1
author maintainer
#65 lore raw

PATCH 62/134 scsi: zorro_esp: Limit DMA transfers to 65536 bytes (except on Fastlane)

In reply to Sasha Levin

From: Kars de Jong <[email protected]>

[ Upstream commit 02f7e9f351a9de95577eafdc3bd413ed1c3b589f ]

When using this driver on a Blizzard 1260, there were failures whenever DMA transfers from the SCSI bus to memory of 65535 bytes were followed by a DMA transfer of 1 byte. This caused the byte at offset 65535 to be overwritten with 0xff. The Blizzard hardware can't handle single byte DMA transfers.

Besides this issue, limiting the DMA length to something that is not a multiple of the page size is very inefficient on most file systems.

It seems this limit was chosen because the DMA transfer counter of the ESP by default is 16 bits wide, thus limiting the length to 65535 bytes. However, the value 0 means 65536 bytes, which is handled by the ESP and the Blizzard just fine. It is also the default maximum used by esp_scsi when drivers don't provide their own dma_length_limit() function.

The limit of 65536 bytes can be used by all boards except the Fastlane. The old driver used a limit of 65532 bytes (0xfffc), which is reintroduced in this patch.

Fixes b7ded0e8b0d1 ("scsi: zorro_esp: Limit DMA transfers to 65535 bytes")
---
 drivers/scsi/zorro_esp.c | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)
drivers/scsi/zorro_esp.c +9 -2
index ca8e3abeb2c7a..a23a8e5794f5c 100644
@@ -218,7 +218,14 @@ static int fastlane_esp_irq_pending(struct esp *esp)
 static u32 zorro_esp_dma_length_limit(struct esp *esp, u32 dma_addr,
 					u32 dma_len)
 {
-	return dma_len > 0xFFFF ? 0xFFFF : dma_len;
+	return dma_len > (1U << 16) ? (1U << 16) : dma_len;
+}
+
+static u32 fastlane_esp_dma_length_limit(struct esp *esp, u32 dma_addr,
+					u32 dma_len)
+{
+	/* The old driver used 0xfffc as limit, so do that here too */
+	return dma_len > 0xfffc ? 0xfffc : dma_len;
 }
 
 static void zorro_esp_reset_dma(struct esp *esp)
@@ -604,7 +611,7 @@ static const struct esp_driver_ops fastlane_esp_ops = {
 	.esp_write8		= zorro_esp_write8,
 	.esp_read8		= zorro_esp_read8,
 	.irq_pending		= fastlane_esp_irq_pending,
-	.dma_length_limit	= zorro_esp_dma_length_limit,
+	.dma_length_limit	= fastlane_esp_dma_length_limit,
 	.reset_dma		= zorro_esp_reset_dma,
 	.dma_drain		= zorro_esp_dma_drain,
 	.dma_invalidate		= fastlane_esp_dma_invalidate,
Signature · 1 line
2.20.1
author maintainer
#66 lore raw

PATCH 63/134 PCI: rpaphp: Fix up pointer to first drc-info entry

In reply to Sasha Levin

From: Tyrel Datwyler <[email protected]>

[ Upstream commit 9723c25f99aff0451cfe6392e1b9fdd99d0bf9f0 ]

The first entry of the ibm,drc-info property is an int encoded count of the number of drc-info entries that follow. The "value" pointer returned by of_prop_next_u32() is still pointing at the this value when we call of_read_drc_info_cell(), but the helper function expects that value to be pointing at the first element of an entry.

Fix up by incrementing the "value" pointer to point at the first element of the first drc-info entry prior.

---
 drivers/pci/hotplug/rpaphp_core.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/pci/hotplug/rpaphp_core.c +2 -0
index 18627bb21e9ec..e3502644a45c3 100644
@@ -239,6 +239,8 @@ static int rpaphp_check_drc_props_v2(struct device_node *dn, char *drc_name,
 	value = of_prop_next_u32(info, NULL, &entries);
 	if (!value)
 		return -EINVAL;
+	else
+		value++;
 
 	for (j = 0; j < entries; j++) {
 		of_read_drc_info_cell(&info, &value, &drc);
Signature · 1 line
2.20.1
author maintainer
#67 lore raw

PATCH 64/134 scsi: ufs: fix potential bug which ends in system hang

In reply to Sasha Levin

From: Bean Huo <[email protected]>

[ Upstream commit cfcbae3895b86c390ede57b2a8f601dd5972b47b ]

In function __ufshcd_query_descriptor(), in the event of an error happening, we directly goto out_unlock and forget to invaliate hba->dev_cmd.query.descriptor pointer. This results in this pointer still valid in ufshcd_copy_query_response() for other query requests which go through ufshcd_exec_raw_upiu_cmd(). This will cause __memcpy() crash and system hangs. Log as shown below:

Unable to handle kernel paging request at virtual address

ffff000012233c40
Mem abort info:
ESR = 0x96000047
Exception class = DABT (current EL), IL = 32 bits SET = 0, FnV = 0
EA = 0, S1PTW = 0
Data abort info:
ISV = 0, ISS = 0x00000047
CM = 0, WnR = 1
swapper pgtable: 4k pages, 48-bit VAs, pgdp = 0000000028cc735c [ffff000012233c40] pgd=00000000bffff003, pud=00000000bfffe003, pmd=00000000ba8b8003, pte=0000000000000000

 Internal error: Oops: 96000047 [#2] PREEMPT SMP
 ...
 Call trace:
  __memcpy+0x74/0x180
  ufshcd_issue_devman_upiu_cmd+0x250/0x3c0
  ufshcd_exec_raw_upiu_cmd+0xfc/0x1a8
  ufs_bsg_request+0x178/0x3b0
  bsg_queue_rq+0xc0/0x118
  blk_mq_dispatch_rq_list+0xb0/0x538
  blk_mq_sched_dispatch_requests+0x18c/0x1d8
  __blk_mq_run_hw_queue+0xb4/0x118
  blk_mq_run_work_fn+0x28/0x38
  process_one_work+0x1ec/0x470
  worker_thread+0x48/0x458
  kthread+0x130/0x138
  ret_from_fork+0x10/0x1c
 Code: 540000ab a8c12027 a88120c7 a8c12027 (a88120c7)
 ---[ end trace 793e1eb5dff69f2d ]---
 note: kworker/0:2H[2054] exited with preempt_count 1

This patch is to move "descriptor = NULL" down to below the label "out_unlock".

---
 drivers/scsi/ufs/ufshcd.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/ufs/ufshcd.c +1 -1
index 11a87f51c442a..358ff7b015680 100644
@@ -2986,10 +2986,10 @@ static int __ufshcd_query_descriptor(struct ufs_hba *hba,
 		goto out_unlock;
 	}
 
-	hba->dev_cmd.query.descriptor = NULL;
 	*buf_len = be16_to_cpu(response->upiu_res.length);
 
 out_unlock:
+	hba->dev_cmd.query.descriptor = NULL;
 	mutex_unlock(&hba->dev_cmd.lock);
 out:
 	ufshcd_release(hba);
Signature · 1 line
2.20.1
author maintainer
#68 lore raw

PATCH 65/134 powerpc/pseries/cmm: Implement release() function for sysfs device

In reply to Sasha Levin

From: David Hildenbrand <[email protected]>

[ Upstream commit 7d8212747435c534c8d564fbef4541a463c976ff ]

When unloading the module, one gets

  ------------[ cut here ]------------
  Device 'cmm0' does not have a release() function, it is broken and must be fixed. See Documentation/kobject.txt.
  WARNING: CPU: 0 PID: 19308 at drivers/base/core.c:1244 .device_release+0xcc/0xf0
  ...

We only have one static fake device. There is nothing to do when releasing the device (via cmm_exit()).

---
 arch/powerpc/platforms/pseries/cmm.c | 5 +++++
 1 file changed, 5 insertions(+)
arch/powerpc/platforms/pseries/cmm.c +5 -0
index b33251d75927b..572651a5c87bb 100644
@@ -411,6 +411,10 @@ static struct bus_type cmm_subsys = {
 	.dev_name = "cmm",
 };
 
+static void cmm_release_device(struct device *dev)
+{
+}
+
 /**
  * cmm_sysfs_register - Register with sysfs
  *
@@ -426,6 +430,7 @@ static int cmm_sysfs_register(struct device *dev)
 
 	dev->id = 0;
 	dev->bus = &cmm_subsys;
+	dev->release = cmm_release_device;
 
 	if ((rc = device_register(dev)))
 		goto subsys_unregister;
Signature · 1 line
2.20.1
author maintainer
#69 lore raw

PATCH 66/134 PCI: rpaphp: Don't rely on firmware feature to imply drc-info support

In reply to Sasha Levin

From: Tyrel Datwyler <[email protected]>

[ Upstream commit 52e2b0f16574afd082cff0f0e8567b2d9f68c033 ]

In the event that the partition is migrated to a platform with older firmware that doesn't support the ibm,drc-info property the device tree is modified to remove the ibm,drc-info property and replace it with the older style ibm,drc-* properties for types, names, indexes, and power-domains. One of the requirements of the drc-info firmware feature is that the client is able to handle both the new property, and old style properties at runtime. Therefore we can't rely on the firmware feature alone to dictate which property is currently present in the device tree.

Fix this short coming by checking explicitly for the ibm,drc-info property, and falling back to the older ibm,drc-* properties if it doesn't exist.

---
 drivers/pci/hotplug/rpaphp_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/pci/hotplug/rpaphp_core.c +1 -1
index e3502644a45c3..e18e9a0e959c8 100644
@@ -275,7 +275,7 @@ int rpaphp_check_drc_props(struct device_node *dn, char *drc_name,
 		return -EINVAL;
 	}
 
-	if (firmware_has_feature(FW_FEATURE_DRC_INFO))
+	if (of_find_property(dn->parent, "ibm,drc-info", NULL))
 		return rpaphp_check_drc_props_v2(dn, drc_name, drc_type,
 						*my_index);
 	else
Signature · 1 line
2.20.1
author maintainer
#70 lore raw

PATCH 67/134 PCI: rpaphp: Annotate and correctly byte swap DRC properties

In reply to Sasha Levin

From: Tyrel Datwyler <[email protected]>

[ Upstream commit 0737686778c6dbe0908d684dd5b9c05b127526ba ]

The device tree is in big endian format and any properties directly retrieved using OF helpers that don't explicitly byte swap should be annotated. In particular there are several places where we grab the opaque property value for the old ibm,drc-* properties and the ibm,my-drc-index property.

Fix this for better static checking by annotating values we know to explicitly big endian, and byte swap where appropriate.

---
 drivers/pci/hotplug/rpaphp_core.c | 29 +++++++++++++++--------------
 1 file changed, 15 insertions(+), 14 deletions(-)
drivers/pci/hotplug/rpaphp_core.c +15 -14
index e18e9a0e959c8..abb10b3c0b70f 100644
@@ -154,11 +154,11 @@ static enum pci_bus_speed get_max_bus_speed(struct slot *slot)
 	return speed;
 }
 
-static int get_children_props(struct device_node *dn, const int **drc_indexes,
-		const int **drc_names, const int **drc_types,
-		const int **drc_power_domains)
+static int get_children_props(struct device_node *dn, const __be32 **drc_indexes,
+			      const __be32 **drc_names, const __be32 **drc_types,
+			      const __be32 **drc_power_domains)
 {
-	const int *indexes, *names, *types, *domains;
+	const __be32 *indexes, *names, *types, *domains;
 
 	indexes = of_get_property(dn, "ibm,drc-indexes", NULL);
 	names = of_get_property(dn, "ibm,drc-names", NULL);
@@ -194,8 +194,8 @@ static int rpaphp_check_drc_props_v1(struct device_node *dn, char *drc_name,
 				char *drc_type, unsigned int my_index)
 {
 	char *name_tmp, *type_tmp;
-	const int *indexes, *names;
-	const int *types, *domains;
+	const __be32 *indexes, *names;
+	const __be32 *types, *domains;
 	int i, rc;
 
 	rc = get_children_props(dn->parent, &indexes, &names, &types, &domains);
@@ -208,7 +208,7 @@ static int rpaphp_check_drc_props_v1(struct device_node *dn, char *drc_name,
 
 	/* Iterate through parent properties, looking for my-drc-index */
 	for (i = 0; i < be32_to_cpu(indexes[0]); i++) {
-		if ((unsigned int) indexes[i + 1] == my_index)
+		if (be32_to_cpu(indexes[i + 1]) == my_index)
 			break;
 
 		name_tmp += (strlen(name_tmp) + 1);
@@ -267,7 +267,7 @@ static int rpaphp_check_drc_props_v2(struct device_node *dn, char *drc_name,
 int rpaphp_check_drc_props(struct device_node *dn, char *drc_name,
 			char *drc_type)
 {
-	const unsigned int *my_index;
+	const __be32 *my_index;
 
 	my_index = of_get_property(dn, "ibm,my-drc-index", NULL);
 	if (!my_index) {
@@ -277,10 +277,10 @@ int rpaphp_check_drc_props(struct device_node *dn, char *drc_name,
 
 	if (of_find_property(dn->parent, "ibm,drc-info", NULL))
 		return rpaphp_check_drc_props_v2(dn, drc_name, drc_type,
-						*my_index);
+						be32_to_cpu(*my_index));
 	else
 		return rpaphp_check_drc_props_v1(dn, drc_name, drc_type,
-						*my_index);
+						be32_to_cpu(*my_index));
 }
 EXPORT_SYMBOL_GPL(rpaphp_check_drc_props);
 
@@ -311,10 +311,11 @@ static int is_php_type(char *drc_type)
  * for built-in pci slots (even when the built-in slots are
  * dlparable.)
  */
-static int is_php_dn(struct device_node *dn, const int **indexes,
-		const int **names, const int **types, const int **power_domains)
+static int is_php_dn(struct device_node *dn, const __be32 **indexes,
+		     const __be32 **names, const __be32 **types,
+		     const __be32 **power_domains)
 {
-	const int *drc_types;
+	const __be32 *drc_types;
 	int rc;
 
 	rc = get_children_props(dn, indexes, names, &drc_types, power_domains);
@@ -349,7 +350,7 @@ int rpaphp_add_slot(struct device_node *dn)
 	struct slot *slot;
 	int retval = 0;
 	int i;
-	const int *indexes, *names, *types, *power_domains;
+	const __be32 *indexes, *names, *types, *power_domains;
 	char *name, *type;
 
 	if (!dn->name || strcmp(dn->name, "pci"))
Signature · 1 line
2.20.1
author maintainer
#71 lore raw

PATCH 68/134 PCI: rpaphp: Correctly match ibm, my-drc-index to drc-name when using drc-info

In reply to Sasha Levin

From: Tyrel Datwyler <[email protected]>

[ Upstream commit 4f9f2d3d7a434b7f882b72550194c9278f4a3925 ]

The newer ibm,drc-info property is a condensed description of the old ibm,drc-* properties (ie. names, types, indexes, and power-domains). When matching a drc-index to a drc-name we need to verify that the index is within the start and last drc-index range and map it to a drc-name using the drc-name-prefix and logical index.

Fix the mapping by checking that the index is within the range of the current drc-info entry, and build the name from the drc-name-prefix concatenated with the starting drc-name-suffix value and the sequential index obtained by subtracting ibm,my-drc-index from this entries drc-start-index.

---
 drivers/pci/hotplug/rpaphp_core.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/pci/hotplug/rpaphp_core.c +3 -2
index abb10b3c0b70f..32eab1776cfea 100644
@@ -248,9 +248,10 @@ static int rpaphp_check_drc_props_v2(struct device_node *dn, char *drc_name,
 		/* Should now know end of current entry */
 
 		/* Found it */
-		if (my_index <= drc.last_drc_index) {
+		if (my_index >= drc.drc_index_start && my_index <= drc.last_drc_index) {
+			int index = my_index - drc.drc_index_start;
 			sprintf(cell_drc_name, "%s%d", drc.drc_name_prefix,
-				my_index);
+				drc.drc_name_suffix_start + index);
 			break;
 		}
 	}
Signature · 1 line
2.20.1
author maintainer
#72 lore raw

PATCH 69/134 powerpc/security: Fix wrong message when RFI Flush is disable

In reply to Sasha Levin

From: "Gustavo L. F. Walbon" <[email protected]>

[ Upstream commit 4e706af3cd8e1d0503c25332b30cad33c97ed442 ]

The issue was showing "Mitigation" message via sysfs whatever the state of "RFI Flush", but it should show "Vulnerable" when it is disabled.

If you have "L1D private" feature enabled and not "RFI Flush" you are vulnerable to meltdown attacks.

"RFI Flush" is the key feature to mitigate the meltdown whatever the "L1D private" state.

SEC_FTR_L1D_THREAD_PRIV is a feature for Power9 only.

So the message should be as the truth table shows:

  CPU | L1D private | RFI Flush |                sysfs
  ----|-------------|-----------|-------------------------------------
   P9 |    False    |   False   | Vulnerable
   P9 |    False    |   True    | Mitigation: RFI Flush
   P9 |    True     |   False   | Vulnerable: L1D private per thread
   P9 |    True     |   True    | Mitigation: RFI Flush, L1D private per thread
   P8 |    False    |   False   | Vulnerable
   P8 |    False    |   True    | Mitigation: RFI Flush

Output before this fix:

  # cat /sys/devices/system/cpu/vulnerabilities/meltdown
  Mitigation: RFI Flush, L1D private per thread
  # echo 0 > /sys/kernel/debug/powerpc/rfi_flush
  # cat /sys/devices/system/cpu/vulnerabilities/meltdown
  Mitigation: L1D private per thread

Output after fix:

  # cat /sys/devices/system/cpu/vulnerabilities/meltdown
  Mitigation: RFI Flush, L1D private per thread
  # echo 0 > /sys/kernel/debug/powerpc/rfi_flush
  # cat /sys/devices/system/cpu/vulnerabilities/meltdown
  Vulnerable: L1D private per thread
---
 arch/powerpc/kernel/security.c | 16 ++++++----------
 1 file changed, 6 insertions(+), 10 deletions(-)
arch/powerpc/kernel/security.c +6 -10
index 298a2e3ad6f4c..d341b464f23c6 100644
@@ -142,26 +142,22 @@ ssize_t cpu_show_meltdown(struct device *dev, struct device_attribute *attr, cha
 
 	thread_priv = security_ftr_enabled(SEC_FTR_L1D_THREAD_PRIV);
 
-	if (rfi_flush || thread_priv) {
+	if (rfi_flush) {
 		struct seq_buf s;
 		seq_buf_init(&s, buf, PAGE_SIZE - 1);
 
-		seq_buf_printf(&s, "Mitigation: ");
-
-		if (rfi_flush)
-			seq_buf_printf(&s, "RFI Flush");
-
-		if (rfi_flush && thread_priv)
-			seq_buf_printf(&s, ", ");
-
+		seq_buf_printf(&s, "Mitigation: RFI Flush");
 		if (thread_priv)
-			seq_buf_printf(&s, "L1D private per thread");
+			seq_buf_printf(&s, ", L1D private per thread");
 
 		seq_buf_printf(&s, "\n");
 
 		return s.len;
 	}
 
+	if (thread_priv)
+		return sprintf(buf, "Vulnerable: L1D private per thread\n");
+
 	if (!security_ftr_enabled(SEC_FTR_L1D_FLUSH_HV) &&
 	    !security_ftr_enabled(SEC_FTR_L1D_FLUSH_PR))
 		return sprintf(buf, "Not affected\n");
Signature · 1 line
2.20.1
author maintainer
#73 lore raw

PATCH 70/134 powerpc/eeh: differentiate duplicate detection message

In reply to Sasha Levin

From: Sam Bobroff <[email protected]>

[ Upstream commit de84ffc3ccbeec3678f95a3d898fc188efa0d9c5 ]

Currently when an EEH error is detected, the system log receives the same (or almost the same) message twice:

EEH: PHB#0 failure detected, location: N/A
EEH: PHB#0 failure detected, location: N/A
or
EEH: eeh_dev_check_failure: Frozen PHB#0-PE#0 detected
EEH: Frozen PHB#0-PE#0 detected

This looks like a bug, but in fact the messages are from different functions and mean slightly different things. So keep both but change one of the messages slightly, so that it's clear they are different:

EEH: PHB#0 failure detected, location: N/A
EEH: Recovering PHB#0, location: N/A
or
EEH: eeh_dev_check_failure: Frozen PHB#0-PE#0 detected
EEH: Recovering PHB#0-PE#0

---
 arch/powerpc/kernel/eeh_driver.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/kernel/eeh_driver.c +2 -2
index d9279d0ee9f54..c031be8d41ffd 100644
@@ -897,12 +897,12 @@ void eeh_handle_normal_event(struct eeh_pe *pe)
 
 	/* Log the event */
 	if (pe->type & EEH_PE_PHB) {
-		pr_err("EEH: PHB#%x failure detected, location: %s\n",
+		pr_err("EEH: Recovering PHB#%x, location: %s\n",
 			pe->phb->global_number, eeh_pe_loc_get(pe));
 	} else {
 		struct eeh_pe *phb_pe = eeh_phb_pe_get(pe->phb);
 
-		pr_err("EEH: Frozen PHB#%x-PE#%x detected\n",
+		pr_err("EEH: Recovering PHB#%x-PE#%x\n",
 		       pe->phb->global_number, pe->addr);
 		pr_err("EEH: PE location: %s, PHB location: %s\n",
 		       eeh_pe_loc_get(pe), eeh_pe_loc_get(phb_pe));
Signature · 1 line
2.20.1
author maintainer
#74 lore raw

PATCH 71/134 powerpc/book3s/mm: Update Oops message to print the correct translation in use

In reply to Sasha Levin

From: "Aneesh Kumar K.V" <[email protected]>

[ Upstream commit d7e02f7b7991dbe14a2acfb0e53d675cd149001c ]

Avoids confusion when printing Oops message like below

Faulting instruction address: 0xc00000000008bdb4

 Oops: Kernel access of bad area, sig: 11 [#1]
 LE PAGE_SIZE=64K MMU=Radix MMU=Hash SMP NR_CPUS=2048 NUMA PowerNV

This was because we never clear the MMU_FTR_HPTE_TABLE feature flag even if we run with radix translation. It was discussed that we should look at this feature flag as an indication of the capability to run hash translation and we should not clear the flag even if we run in radix translation. All the code paths check for radix_enabled() check and if found true consider we are running with radix translation. Follow the same sequence for finding the MMU translation string to be used in Oops message.

---
 arch/powerpc/kernel/traps.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)
arch/powerpc/kernel/traps.c +11 -4
index 82f43535e6867..014ff0701f245 100644
@@ -250,15 +250,22 @@ static void oops_end(unsigned long flags, struct pt_regs *regs,
 }
 NOKPROBE_SYMBOL(oops_end);
 
+static char *get_mmu_str(void)
+{
+	if (early_radix_enabled())
+		return " MMU=Radix";
+	if (early_mmu_has_feature(MMU_FTR_HPTE_TABLE))
+		return " MMU=Hash";
+	return "";
+}
+
 static int __die(const char *str, struct pt_regs *regs, long err)
 {
 	printk("Oops: %s, sig: %ld [#%d]\n", str, err, ++die_counter);
 
-	printk("%s PAGE_SIZE=%luK%s%s%s%s%s%s%s %s\n",
+	printk("%s PAGE_SIZE=%luK%s%s%s%s%s%s %s\n",
 	       IS_ENABLED(CONFIG_CPU_LITTLE_ENDIAN) ? "LE" : "BE",
-	       PAGE_SIZE / 1024,
-	       early_radix_enabled() ? " MMU=Radix" : "",
-	       early_mmu_has_feature(MMU_FTR_HPTE_TABLE) ? " MMU=Hash" : "",
+	       PAGE_SIZE / 1024, get_mmu_str(),
 	       IS_ENABLED(CONFIG_PREEMPT) ? " PREEMPT" : "",
 	       IS_ENABLED(CONFIG_SMP) ? " SMP" : "",
 	       IS_ENABLED(CONFIG_SMP) ? (" NR_CPUS=" __stringify(NR_CPUS)) : "",
Signature · 1 line
2.20.1
author maintainer
#75 lore raw

PATCH 72/134 scsi: atari_scsi: sun3_scsi: Set sg_tablesize to 1 instead of SG_NONE

In reply to Sasha Levin

From: Finn Thain <[email protected]>

[ Upstream commit 79172ab20bfd8437b277254028efdb68484e2c21 ]

Since the scsi subsystem adopted the blk-mq API, a host with zero sg_tablesize crashes with a NULL pointer dereference.

blk_queue_max_segments: set to minimum 1
scsi 0:0:0:0: Direct-Access QEMU QEMU HARDDISK 2.5+ PQ: 0 ANSI: 5
scsi target0:0:0: Beginning Domain Validation
scsi target0:0:0: Domain Validation skipping write tests
scsi target0:0:0: Ending Domain Validation
blk_queue_max_segments: set to minimum 1
scsi 0:0:1:0: Direct-Access QEMU QEMU HARDDISK 2.5+ PQ: 0 ANSI: 5
scsi target0:0:1: Beginning Domain Validation
scsi target0:0:1: Domain Validation skipping write tests
scsi target0:0:1: Ending Domain Validation
blk_queue_max_segments: set to minimum 1
scsi 0:0:2:0: CD-ROM QEMU QEMU CD-ROM 2.5+ PQ: 0 ANSI: 5
scsi target0:0:2: Beginning Domain Validation
scsi target0:0:2: Domain Validation skipping write tests
scsi target0:0:2: Ending Domain Validation
blk_queue_max_segments: set to minimum 1
blk_queue_max_segments: set to minimum 1
blk_queue_max_segments: set to minimum 1
blk_queue_max_segments: set to minimum 1
sr 0:0:2:0: Power-on or device reset occurred
sd 0:0:0:0: Power-on or device reset occurred
sd 0:0:1:0: Power-on or device reset occurred
sd 0:0:0:0: [sda] 10485762 512-byte logical blocks: (5.37 GB/5.00 GiB)
sd 0:0:0:0: [sda] Write Protect is off
sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA

Unable to handle kernel NULL pointer dereference at virtual address (ptrval)
Oops: 00000000
Modules linked in:
PC: [<001cd874>] blk_mq_free_request+0x66/0xe2
SR: 2004  SP: (ptrval)  a2: 00874520
d0: 00000000    d1: 00000000    d2: 009ba800    d3: 00000000
d4: 00000000    d5: 08000002    a0: 0087be68    a1: 009a81e0

Process kworker/u2:2 (pid: 15, task=(ptrval)) Frame format=7 eff addr=0000007a ssw=0505 faddr=0000007a wb 1 stat/addr/data: 0000 00000000 00000000
wb 2 stat/addr/data: 0000 00000000 00000000
wb 3 stat/addr/data: 0000 0000007a 00000000 push data: 00000000 00000000 00000000 00000000 Stack from 0087bd98:

        00000002 00000000 0087be72 009a7820 0087bdb4 001c4f6c 009a7820 0087bdd4
        0024d200 009a7820 0024d0dc 0087be72 009baa00 0087be68 009a5000 0087be7c
        00265d10 009a5000 0087be72 00000003 00000000 00000000 00000000 0087be68
        00000bb8 00000005 00000000 00000000 00000000 00000000 00265c56 00000000
        009ba60c 0036ddf4 00000002 ffffffff 009baa00 009ba600 009a50d6 0087be74
        00227ba0 009baa08 00000001 009baa08 009ba60c 0036ddf4 00000000 00000000
Call Trace: [<001c4f6c>] blk_put_request+0xe/0x14
 [<0024d200>] __scsi_execute+0x124/0x174
 [<0024d0dc>] __scsi_execute+0x0/0x174
 [<00265d10>] sd_revalidate_disk+0xba/0x1f02
 [<00265c56>] sd_revalidate_disk+0x0/0x1f02
 [<0036ddf4>] strlen+0x0/0x22
 [<00227ba0>] device_add+0x3da/0x604
 [<0036ddf4>] strlen+0x0/0x22
 [<00267e64>] sd_probe+0x30c/0x4b4
 [<0002da44>] process_one_work+0x0/0x402
 [<0022b978>] really_probe+0x226/0x354
 [<0022bc34>] driver_probe_device+0xa4/0xf0
 [<0002da44>] process_one_work+0x0/0x402
 [<0022bcd0>] __driver_attach_async_helper+0x50/0x70
 [<00035dae>] async_run_entry_fn+0x36/0x130
 [<0002db88>] process_one_work+0x144/0x402
 [<0002e1aa>] worker_thread+0x0/0x570
 [<0002e29a>] worker_thread+0xf0/0x570
 [<0002e1aa>] worker_thread+0x0/0x570
 [<003768d8>] schedule+0x0/0xb8
 [<0003f58c>] __init_waitqueue_head+0x0/0x12
 [<00033e92>] kthread+0xc2/0xf6
 [<000331e8>] kthread_parkme+0x0/0x4e
 [<003768d8>] schedule+0x0/0xb8
 [<00033dd0>] kthread+0x0/0xf6
 [<00002c10>] ret_from_kernel_thread+0xc/0x14
Code: 0280 0006 0800 56c0 4400 0280 0000 00ff <52b4> 0c3a 082b 0006 0013 6706 2042 53a8 00c4 4ab9 0047 3374 6640 202d 000c 670c

Disabling lock debugging due to kernel taint

Avoid this by setting sg_tablesize = 1.

References: commit 68ab2d76e4be ("scsi: cxlflash: Set sg_tablesize to 1 instead of SG_NONE")

---
 drivers/scsi/atari_scsi.c | 6 +++---
 drivers/scsi/mac_scsi.c   | 2 +-
 drivers/scsi/sun3_scsi.c  | 4 ++--
 3 files changed, 6 insertions(+), 6 deletions(-)
drivers/scsi/atari_scsi.c +3 -3
index e809493d0d063..a82b63a666356 100644
@@ -742,7 +742,7 @@ static int __init atari_scsi_probe(struct platform_device *pdev)
 		atari_scsi_template.sg_tablesize = SG_ALL;
 	} else {
 		atari_scsi_template.can_queue    = 1;
-		atari_scsi_template.sg_tablesize = SG_NONE;
+		atari_scsi_template.sg_tablesize = 1;
 	}
 
 	if (setup_can_queue > 0)
@@ -751,8 +751,8 @@ static int __init atari_scsi_probe(struct platform_device *pdev)
 	if (setup_cmd_per_lun > 0)
 		atari_scsi_template.cmd_per_lun = setup_cmd_per_lun;
 
-	/* Leave sg_tablesize at 0 on a Falcon! */
-	if (ATARIHW_PRESENT(TT_SCSI) && setup_sg_tablesize >= 0)
+	/* Don't increase sg_tablesize on Falcon! */
+	if (ATARIHW_PRESENT(TT_SCSI) && setup_sg_tablesize > 0)
 		atari_scsi_template.sg_tablesize = setup_sg_tablesize;
 
 	if (setup_hostid >= 0) {
drivers/scsi/mac_scsi.c +1 -1
index 9c5566217ef6a..b5dde9d0d0545 100644
@@ -464,7 +464,7 @@ static int __init mac_scsi_probe(struct platform_device *pdev)
 		mac_scsi_template.can_queue = setup_can_queue;
 	if (setup_cmd_per_lun > 0)
 		mac_scsi_template.cmd_per_lun = setup_cmd_per_lun;
-	if (setup_sg_tablesize >= 0)
+	if (setup_sg_tablesize > 0)
 		mac_scsi_template.sg_tablesize = setup_sg_tablesize;
 	if (setup_hostid >= 0)
 		mac_scsi_template.this_id = setup_hostid & 7;
drivers/scsi/sun3_scsi.c +2 -2
index 955e4c938d49e..701b842296f05 100644
@@ -501,7 +501,7 @@ static struct scsi_host_template sun3_scsi_template = {
 	.eh_host_reset_handler	= sun3scsi_host_reset,
 	.can_queue		= 16,
 	.this_id		= 7,
-	.sg_tablesize		= SG_NONE,
+	.sg_tablesize		= 1,
 	.cmd_per_lun		= 2,
 	.dma_boundary		= PAGE_SIZE - 1,
 	.cmd_size		= NCR5380_CMD_SIZE,
@@ -523,7 +523,7 @@ static int __init sun3_scsi_probe(struct platform_device *pdev)
 		sun3_scsi_template.can_queue = setup_can_queue;
 	if (setup_cmd_per_lun > 0)
 		sun3_scsi_template.cmd_per_lun = setup_cmd_per_lun;
-	if (setup_sg_tablesize >= 0)
+	if (setup_sg_tablesize > 0)
 		sun3_scsi_template.sg_tablesize = setup_sg_tablesize;
 	if (setup_hostid >= 0)
 		sun3_scsi_template.this_id = setup_hostid & 7;
Signature · 1 line
2.20.1
author maintainer
#76 lore raw

PATCH 73/134 clk: pxa: fix one of the pxa RTC clocks

In reply to Sasha Levin

From: Robert Jarzmik <[email protected]>

[ Upstream commit 46acbcb4849b2ca2e6e975e7c8130c1d61c8fd0c ]

The pxa27x platforms have a single IP with 2 drivers, sa1100-rtc and rtc-pxa drivers.

A previous patch fixed the sa1100-rtc case, but the pxa-rtc wasn't fixed. This patch completes the previous one.

Fixes 8b6d10345e16 ("clk: pxa: add missing pxa27x clocks for Irda and sa1100-rtc")
---
 drivers/clk/pxa/clk-pxa27x.c | 1 +
 1 file changed, 1 insertion(+)
drivers/clk/pxa/clk-pxa27x.c +1 -0
index 287fdeae7c7c2..7b123105b5de4 100644
@@ -459,6 +459,7 @@ struct dummy_clk {
 };
 static struct dummy_clk dummy_clks[] __initdata = {
 	DUMMY_CLK(NULL, "pxa27x-gpio", "osc_32_768khz"),
+	DUMMY_CLK(NULL, "pxa-rtc", "osc_32_768khz"),
 	DUMMY_CLK(NULL, "sa1100-rtc", "osc_32_768khz"),
 	DUMMY_CLK("UARTCLK", "pxa2xx-ir", "STUART"),
 };
Signature · 1 line
2.20.1
author maintainer
#77 lore raw

PATCH 74/134 bcache: at least try to shrink 1 node in bch_mca_scan()

In reply to Sasha Levin

From: Coly Li <[email protected]>

[ Upstream commit 9fcc34b1a6dd4b8e5337e2b6ef45e428897eca6b ]

In bch_mca_scan(), the number of shrinking btree node is calculated by code like this,

	unsigned long nr = sc->nr_to_scan;

        nr /= c->btree_pages;
        nr = min_t(unsigned long, nr, mca_can_free(c));

variable sc->nr_to_scan is number of objects (here is bcache B+tree nodes' number) to shrink, and pointer variable sc is sent from memory management code as parametr of a callback.

If sc->nr_to_scan is smaller than c->btree_pages, after the above calculation, variable 'nr' will be 0 and nothing will be shrunk. It is frequeently observed that only 1 or 2 is set to sc->nr_to_scan and make nr to be zero. Then bch_mca_scan() will do nothing more then acquiring and releasing mutex c->bucket_lock.

This patch checkes whether nr is 0 after the above calculation, if 0 is the result then set 1 to variable 'n'. Then at least bch_mca_scan() will try to shrink a single B+tree node.

---
 drivers/md/bcache/btree.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/md/bcache/btree.c +2 -0
index ba434d9ac7204..46a8b5a91c386 100644
@@ -723,6 +723,8 @@ static unsigned long bch_mca_scan(struct shrinker *shrink,
 	 * IO can always make forward progress:
 	 */
 	nr /= c->btree_pages;
+	if (nr == 0)
+		nr = 1;
 	nr = min_t(unsigned long, nr, mca_can_free(c));
 
 	i = 0;
Signature · 1 line
2.20.1
author maintainer
#78 lore raw

PATCH 75/134 HID: quirks: Add quirk for HP MSU1465 PIXART OEM mouse

In reply to Sasha Levin

From: Jinke Fan <[email protected]>

[ Upstream commit f1a0094cbbe97a5f8aca7bdc64bfe43ac9dc6879 ]

The PixArt OEM mouse disconnets/reconnects every minute on Linux. All contents of dmesg are repetitive:

[ 1465.810014] usb 1-2.2: USB disconnect, device number 20
[ 1467.431509] usb 1-2.2: new low-speed USB device number 21 using xhci_hcd
[ 1467.654982] usb 1-2.2: New USB device found, idVendor=03f0,idProduct=1f4a, bcdDevice= 1.00
[ 1467.654985] usb 1-2.2: New USB device strings: Mfr=1, Product=2,SerialNumber=0
[ 1467.654987] usb 1-2.2: Product: HP USB Optical Mouse
[ 1467.654988] usb 1-2.2: Manufacturer: PixArt
[ 1467.699722] input: PixArt HP USB Optical Mouse as /devices/pci0000:00/0000:00:07.1/0000:05:00.3/usb1/1-2/1-2.2/1-2.2:1.0/0003:03F0:1F4A.0012/input/input19
[ 1467.700124] hid-generic 0003:03F0:1F4A.0012: input,hidraw0: USB HID v1.11 Mouse [PixArt HP USB Optical Mouse] on usb-0000:05:00.3-2.2/input0

So add HID_QUIRK_ALWAYS_POLL for this one as well.
Test the patch, the mouse is no longer disconnected and there are no duplicate logs in dmesg.

Reference:
https://github.com/sriemer/fix-linux-mouse

---
 drivers/hid/hid-ids.h    | 1 +
 drivers/hid/hid-quirks.c | 1 +
 2 files changed, 2 insertions(+)
drivers/hid/hid-ids.h +1 -0
index 447e8db21174a..00904537e17c4 100644
@@ -573,6 +573,7 @@
 #define USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_094A	0x094a
 #define USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_0941	0x0941
 #define USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_0641	0x0641
+#define USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_1f4a	0x1f4a
 
 #define USB_VENDOR_ID_HUION		0x256c
 #define USB_DEVICE_ID_HUION_TABLET	0x006e
drivers/hid/hid-quirks.c +1 -0
index c50bcd967d994..9a35af1e26623 100644
@@ -94,6 +94,7 @@ static const struct hid_device_id hid_quirks[] = {
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_094A), HID_QUIRK_ALWAYS_POLL },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_0941), HID_QUIRK_ALWAYS_POLL },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_0641), HID_QUIRK_ALWAYS_POLL },
+	{ HID_USB_DEVICE(USB_VENDOR_ID_HP, USB_PRODUCT_ID_HP_PIXART_OEM_USB_OPTICAL_MOUSE_1f4a), HID_QUIRK_ALWAYS_POLL },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_IDEACOM, USB_DEVICE_ID_IDEACOM_IDC6680), HID_QUIRK_MULTI_INPUT },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_INNOMEDIA, USB_DEVICE_ID_INNEX_GENESIS_ATARI), HID_QUIRK_MULTI_INPUT },
 	{ HID_USB_DEVICE(USB_VENDOR_ID_KYE, USB_DEVICE_ID_KYE_EASYPEN_M610X), HID_QUIRK_MULTI_INPUT },
Signature · 1 line
2.20.1
author maintainer
#79 lore raw

PATCH 76/134 dt-bindings: Improve validation build error handling

In reply to Sasha Levin

From: Rob Herring <[email protected]>

[ Upstream commit 93512dad334deb444619505f1fbb761156f7471b ]

Schema errors can cause make to exit before useful information is printed. This leaves developers wondering what's wrong. It can be overcome passing '-k' to make, but that's not an obvious solution. There's 2 scenarios where this happens.

When using DT_SCHEMA_FILES to validate with a single schema, any error in the schema results in processed-schema.yaml being empty causing a make error. The result is the specific errors in the schema are never shown because processed-schema.yaml is the first target built. Simply making processed-schema.yaml last in extra-y ensures the full schema validation with detailed error messages happen first.

The 2nd problem is while schema errors are ignored for processed-schema.yaml, full validation of the schema still runs in parallel and any schema validation errors will still stop the build when running validation of dts files. The fix is to not add the schema examples to extra-y in this case. This means 'dtbs_check' is no longer a superset of 'dt_binding_check'. Update the documentation to make this clear.

---
 Documentation/devicetree/bindings/Makefile  | 5 ++++-
 Documentation/devicetree/writing-schema.rst | 6 ++++--
 2 files changed, 8 insertions(+), 3 deletions(-)
Documentation/devicetree/bindings/Makefile +4 -1
index 5138a2f6232aa..646cb35253733 100644
@@ -12,7 +12,6 @@ $(obj)/%.example.dts: $(src)/%.yaml FORCE
 	$(call if_changed,chk_binding)
 
 DT_TMP_SCHEMA := processed-schema.yaml
-extra-y += $(DT_TMP_SCHEMA)
 
 quiet_cmd_mk_schema = SCHEMA  $@
       cmd_mk_schema = $(DT_MK_SCHEMA) $(DT_MK_SCHEMA_FLAGS) -o $@ $(real-prereqs)
@@ -26,8 +25,12 @@ DT_DOCS = $(shell \
 
 DT_SCHEMA_FILES ?= $(addprefix $(src)/,$(DT_DOCS))
 
+ifeq ($(CHECK_DTBS),)
 extra-y += $(patsubst $(src)/%.yaml,%.example.dts, $(DT_SCHEMA_FILES))
 extra-y += $(patsubst $(src)/%.yaml,%.example.dt.yaml, $(DT_SCHEMA_FILES))
+endif
 
 $(obj)/$(DT_TMP_SCHEMA): $(DT_SCHEMA_FILES) FORCE
 	$(call if_changed,mk_schema)
+
+extra-y += $(DT_TMP_SCHEMA)
Documentation/devicetree/writing-schema.rst +4 -2
index f4a638072262f..83e04e5c342da 100644
@@ -130,11 +130,13 @@ binding schema. All of the DT binding documents can be validated using the
 
     make dt_binding_check
 
-In order to perform validation of DT source files, use the `dtbs_check` target::
+In order to perform validation of DT source files, use the ``dtbs_check`` target::
 
     make dtbs_check
 
-This will first run the `dt_binding_check` which generates the processed schema.
+Note that ``dtbs_check`` will skip any binding schema files with errors. It is
+necessary to use ``dt_binding_check`` to get all the validation errors in the
+binding schema files.
 
 It is also possible to run checks with a single schema file by setting the
 ``DT_SCHEMA_FILES`` variable to a specific schema file.
Signature · 1 line
2.20.1
author maintainer
#80 lore raw

PATCH 77/134 ext4: fix a bug in ext4_wait_for_tail_page_commit

In reply to Sasha Levin

From: yangerkun <[email protected]>

[ Upstream commit 565333a1554d704789e74205989305c811fd9c7a ]

No need to wait for any commit once the page is fully truncated. Besides, it may confuse e.g. concurrent ext4_writepage() with the page still be dirty (will be cleared by truncate_pagecache() in ext4_setattr()) but buffers has been freed; and then trigger a bug show as below:

[   26.057508] ------------[ cut here ]------------
[   26.058531] kernel BUG at fs/ext4/inode.c:2134!
...
[   26.088130] Call trace:
[   26.088695]  ext4_writepage+0x914/0xb28
[   26.089541]  writeout.isra.4+0x1b4/0x2b8
[   26.090409]  move_to_new_page+0x3b0/0x568
[   26.091338]  __unmap_and_move+0x648/0x988
[   26.092241]  unmap_and_move+0x48c/0xbb8
[   26.093096]  migrate_pages+0x220/0xb28
[   26.093945]  kernel_mbind+0x828/0xa18
[   26.094791]  __arm64_sys_mbind+0xc8/0x138
[   26.095716]  el0_svc_common+0x190/0x490
[   26.096571]  el0_svc_handler+0x60/0xd0
[   26.097423]  el0_svc+0x8/0xc

Run the procedure (generate by syzkaller) parallel with ext3.

void main()

{
	int fd, fd1, ret;
	void *addr;
	size_t length = 4096;
	int flags;
	off_t offset = 0;
	char *str = "12345";

	fd = open("a", O_RDWR | O_CREAT);
	assert(fd >= 0);

	/* Truncate to 4k */
	ret = ftruncate(fd, length);
	assert(ret == 0);

	/* Journal data mode */
	flags = 0xc00f;
	ret = ioctl(fd, _IOW('f', 2, long), &flags);
	assert(ret == 0);

	/* Truncate to 0 */
	fd1 = open("a", O_TRUNC | O_NOATIME);
	assert(fd1 >= 0);

	addr = mmap(NULL, length, PROT_WRITE | PROT_READ,
					MAP_SHARED, fd, offset);
	assert(addr != (void *)-1);

	memcpy(addr, str, 5);
	mbind(addr, length, 0, 0, 0, MPOL_MF_MOVE);
}

And the bug will be triggered once we seen the below order.

reproduce1 reproduce2

...                            |   ...
truncate to 4k                 |
change to journal data mode    |
                               |   memcpy(set page dirty)
truncate to 0:                 |
ext4_setattr:                  |
...                            |
ext4_wait_for_tail_page_commit |
                               |   mbind(trigger bug)
truncate_pagecache(clean dirty)|   ...
...                            |

mbind will call ext4_writepage() since the page still be dirty, and then report the bug since the buffers has been free. Fix it by return directly once offset equals to 0 which means the page has been fully truncated.

---
 fs/ext4/inode.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
fs/ext4/inode.c +8 -4
index 02313f10e8897..8b7c704b0de5e 100644
@@ -5462,11 +5462,15 @@ static void ext4_wait_for_tail_page_commit(struct inode *inode)
 
 	offset = inode->i_size & (PAGE_SIZE - 1);
 	/*
-	 * All buffers in the last page remain valid? Then there's nothing to
-	 * do. We do the check mainly to optimize the common PAGE_SIZE ==
-	 * blocksize case
+	 * If the page is fully truncated, we don't need to wait for any commit
+	 * (and we even should not as __ext4_journalled_invalidatepage() may
+	 * strip all buffers from the page but keep the page dirty which can then
+	 * confuse e.g. concurrent ext4_writepage() seeing dirty page without
+	 * buffers). Also we don't need to wait for any commit if all buffers in
+	 * the page remain valid. This is most beneficial for the common case of
+	 * blocksize == PAGESIZE.
 	 */
-	if (offset > PAGE_SIZE - i_blocksize(inode))
+	if (!offset || offset > (PAGE_SIZE - i_blocksize(inode)))
 		return;
 	while (1) {
 		page = find_lock_page(inode->i_mapping,
Signature · 1 line
2.20.1
author maintainer
#81 lore raw

PATCH 78/134 HID: logitech-hidpp: Silence intermittent get_battery_capacity errors

In reply to Sasha Levin

From: Hans de Goede <[email protected]>

[ Upstream commit 61005d65b6c7dcf61c19516e6ebe5acc02d2cdda ]

My Logitech M185 (PID:4038) 2.4 GHz wireless HID++ mouse is causing intermittent errors like these in the log:

[11091.034857] logitech-hidpp-device 0003:046D:4038.0006: hidpp20_batterylevel_get_battery_capacity: received protocol error 0x09
[12388.031260] logitech-hidpp-device 0003:046D:4038.0006: hidpp20_batterylevel_get_battery_capacity: received protocol error 0x09
[16613.718543] logitech-hidpp-device 0003:046D:4038.0006: hidpp20_batterylevel_get_battery_capacity: received protocol error 0x09
[23529.938728] logitech-hidpp-device 0003:046D:4038.0006: hidpp20_batterylevel_get_battery_capacity: received protocol error 0x09

We are already silencing error-code 0x09 (HIDPP_ERROR_RESOURCE_ERROR) errors in other places, lets do the same in hidpp20_batterylevel_get_battery_capacity to remove these harmless, but scary looking errors from the dmesg output.

---
 drivers/hid/hid-logitech-hidpp.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/hid/hid-logitech-hidpp.c +3 -0
index 8e91e2f06cb4f..cd9193078525b 100644
@@ -1102,6 +1102,9 @@ static int hidpp20_batterylevel_get_battery_capacity(struct hidpp_device *hidpp,
 	ret = hidpp_send_fap_command_sync(hidpp, feature_index,
 					  CMD_BATTERY_LEVEL_STATUS_GET_BATTERY_LEVEL_STATUS,
 					  NULL, 0, &response);
+	/* Ignore these intermittent errors */
+	if (ret == HIDPP_ERROR_RESOURCE_ERROR)
+		return -EIO;
 	if (ret > 0) {
 		hid_err(hidpp->hid_dev, "%s: received protocol error 0x%02x\n",
 			__func__, ret);
Signature · 1 line
2.20.1
author maintainer
#82 lore raw

PATCH 79/134 HID: i2c-hid: fix no irq after reset on raydium 3118

In reply to Sasha Levin

From: Aaron Ma <[email protected]>

[ Upstream commit 0c8432236dea20a95f68fa17989ea3f8af0186a5 ]

On some ThinkPad L390 some raydium 3118 touchscreen devices doesn't response any data after reset, but some does.

Add this ID to no irq quirk,
then don't wait for any response alike on these touchscreens. All kinds of raydium 3118 devices work fine.

---
 drivers/hid/hid-ids.h              | 1 +
 drivers/hid/i2c-hid/i2c-hid-core.c | 2 ++
 2 files changed, 3 insertions(+)
drivers/hid/hid-ids.h +1 -0
index 00904537e17c4..6273e7178e785 100644
@@ -960,6 +960,7 @@
 
 #define I2C_VENDOR_ID_RAYDIUM		0x2386
 #define I2C_PRODUCT_ID_RAYDIUM_4B33	0x4b33
+#define I2C_PRODUCT_ID_RAYDIUM_3118	0x3118
 
 #define USB_VENDOR_ID_RAZER            0x1532
 #define USB_DEVICE_ID_RAZER_BLADE_14   0x011D
drivers/hid/i2c-hid/i2c-hid-core.c +2 -0
index 04c088131e044..7608ee053114c 100644
@@ -170,6 +170,8 @@ static const struct i2c_hid_quirks {
 		I2C_HID_QUIRK_SET_PWR_WAKEUP_DEV },
 	{ I2C_VENDOR_ID_HANTICK, I2C_PRODUCT_ID_HANTICK_5288,
 		I2C_HID_QUIRK_NO_IRQ_AFTER_RESET },
+	{ I2C_VENDOR_ID_RAYDIUM, I2C_PRODUCT_ID_RAYDIUM_3118,
+		I2C_HID_QUIRK_NO_IRQ_AFTER_RESET },
 	{ USB_VENDOR_ID_ELAN, HID_ANY_ID,
 		 I2C_HID_QUIRK_BOGUS_IRQ },
 	{ 0, 0 }
Signature · 1 line
2.20.1
author maintainer
#83 lore raw

PATCH 80/134 ARM: 8937/1: spectre-v2: remove Brahma-B53 from hardening

In reply to Sasha Levin

From: Doug Berger <[email protected]>

[ Upstream commit 4ae5061a19b550dfe25397843427ed2ebab16b16 ]

When the default processor handling was added to the function cpu_v7_spectre_init() it only excluded other ARM implemented processor cores. The Broadcom Brahma B53 core is not implemented by ARM so it ended up falling through into the set of processors that attempt to use the ARM_SMCCC_ARCH_WORKAROUND_1 service to harden the branch predictor.

Since this workaround is not necessary for the Brahma-B53 this commit explicitly checks for it and prevents it from applying a branch predictor hardening workaround.

Fixes 10115105cb3a ("ARM: spectre-v2: add firmware based hardening")
---
 arch/arm/mm/proc-v7-bugs.c | 3 +++
 1 file changed, 3 insertions(+)
arch/arm/mm/proc-v7-bugs.c +3 -0
index 9a07916af8dd2..a6554fdb56c54 100644
@@ -65,6 +65,9 @@ static void cpu_v7_spectre_init(void)
 		break;
 
 #ifdef CONFIG_ARM_PSCI
+	case ARM_CPU_PART_BRAHMA_B53:
+		/* Requires no workaround */
+		break;
 	default:
 		/* Other ARM CPUs require no workaround */
 		if (read_cpuid_implementor() == ARM_CPU_IMP_ARM)
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#84 lore raw

PATCH 81/134 libnvdimm/btt: fix variable 'rc' set but not used

In reply to Sasha Levin

From: Qian Cai <[email protected]>

[ Upstream commit 4e24e37d5313edca8b4ab86f240c046c731e28d6 ]

drivers/nvdimm/btt.c: In function 'btt_read_pg':
drivers/nvdimm/btt.c:1264:8: warning: variable 'rc' set but not used [-Wunused-but-set-variable]

    int rc;
        ^~

Add a ratelimited message in case a storm of errors is encountered.

Fixes d9b83c756953 ("libnvdimm, btt: rework error clearing")
---
 drivers/nvdimm/btt.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/nvdimm/btt.c +4 -4
index 3e9f45aec8d18..5129543a04739 100644
@@ -1261,11 +1261,11 @@ static int btt_read_pg(struct btt *btt, struct bio_integrity_payload *bip,
 
 		ret = btt_data_read(arena, page, off, postmap, cur_len);
 		if (ret) {
-			int rc;
-
 			/* Media error - set the e_flag */
-			rc = btt_map_write(arena, premap, postmap, 0, 1,
-				NVDIMM_IO_ATOMIC);
+			if (btt_map_write(arena, premap, postmap, 0, 1, NVDIMM_IO_ATOMIC))
+				dev_warn_ratelimited(to_dev(arena),
+					"Error persistently tracking bad blocks at %#x\n",
+					premap);
 			goto out_rtt;
 		}
 
Signature · 4 lines
2.20.1
_______________________________________________
Linux-nvdimm mailing list -- [email protected]
To unsubscribe send an email to [email protected]
author maintainer
#85 lore raw

PATCH 82/134 HID: Improve Windows Precision Touchpad detection.

In reply to Sasha Levin

From: Blaž Hrastnik <[email protected]>

[ Upstream commit 2dbc6f113acd74c66b04bf49fb027efd830b1c5a ]

Per Microsoft spec, usage 0xC5 (page 0xFF) returns a blob containing data used to verify the touchpad as a Windows Precision Touchpad.

   0x85, REPORTID_PTPHQA,    //    REPORT_ID (PTPHQA)
    0x09, 0xC5,              //    USAGE (Vendor Usage 0xC5)
    0x15, 0x00,              //    LOGICAL_MINIMUM (0)
    0x26, 0xff, 0x00,        //    LOGICAL_MAXIMUM (0xff)
    0x75, 0x08,              //    REPORT_SIZE (8)
    0x96, 0x00, 0x01,        //    REPORT_COUNT (0x100 (256))
    0xb1, 0x02,              //    FEATURE (Data,Var,Abs)

However, some devices, namely Microsoft's Surface line of products instead implement a "segmented device certification report" (usage 0xC6) which returns the same report, but in smaller chunks.

    0x06, 0x00, 0xff,        //     USAGE_PAGE (Vendor Defined)
    0x85, REPORTID_PTPHQA,   //     REPORT_ID (PTPHQA)
    0x09, 0xC6,              //     USAGE (Vendor usage for segment #)
    0x25, 0x08,              //     LOGICAL_MAXIMUM (8)
    0x75, 0x08,              //     REPORT_SIZE (8)
    0x95, 0x01,              //     REPORT_COUNT (1)
    0xb1, 0x02,              //     FEATURE (Data,Var,Abs)
    0x09, 0xC7,              //     USAGE (Vendor Usage)
    0x26, 0xff, 0x00,        //     LOGICAL_MAXIMUM (0xff)
    0x95, 0x20,              //     REPORT_COUNT (32)
    0xb1, 0x02,              //     FEATURE (Data,Var,Abs)

By expanding Win8 touchpad detection to also look for the segmented report, all Surface touchpads are now properly recognized by hid-multitouch.

---
 drivers/hid/hid-core.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/hid/hid-core.c +4 -0
index 2fa3587d974f4..e0b241bd3070c 100644
@@ -781,6 +781,10 @@ static void hid_scan_feature_usage(struct hid_parser *parser, u32 usage)
 	if (usage == 0xff0000c5 && parser->global.report_count == 256 &&
 	    parser->global.report_size == 8)
 		parser->scan_flags |= HID_SCAN_FLAG_MT_WIN_8;
+
+	if (usage == 0xff0000c6 && parser->global.report_count == 1 &&
+	    parser->global.report_size == 8)
+		parser->scan_flags |= HID_SCAN_FLAG_MT_WIN_8;
 }
 
 static void hid_scan_collection(struct hid_parser *parser, unsigned type)
Signature · 1 line
2.20.1
author maintainer
#86 lore raw

PATCH 83/134 HID: rmi: Check that the RMI_STARTED bit is set before unregistering the RMI transport device

In reply to Sasha Levin

From: Andrew Duggan <[email protected]>

[ Upstream commit 8725aa4fa7ded30211ebd28bb1c9bae806eb3841 ]

In the event that the RMI device is unreachable, the calls to rmi_set_mode() or rmi_set_page() will fail before registering the RMI transport device. When the device is removed, rmi_remove() will call rmi_unregister_transport_device() which will attempt to access the rmi_dev pointer which was not set. This patch adds a check of the RMI_STARTED bit before calling rmi_unregister_transport_device(). The RMI_STARTED bit is only set after rmi_register_transport_device() completes successfully.

The kernel oops was reported in this message:
https://www.spinics.net/lists/linux-input/msg58433.html

---
 drivers/hid/hid-rmi.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/hid/hid-rmi.c +2 -1
index 7c6abd7e09797..9ce22acdfaca2 100644
@@ -744,7 +744,8 @@ static void rmi_remove(struct hid_device *hdev)
 {
 	struct rmi_data *hdata = hid_get_drvdata(hdev);
 
-	if (hdata->device_flags & RMI_DEVICE) {
+	if ((hdata->device_flags & RMI_DEVICE)
+	    && test_bit(RMI_STARTED, &hdata->flags)) {
 		clear_bit(RMI_STARTED, &hdata->flags);
 		cancel_work_sync(&hdata->reset_work);
 		rmi_unregister_transport_device(&hdata->xport);
Signature · 1 line
2.20.1
author maintainer
#87 lore raw

PATCH 84/134 watchdog: imx7ulp: Fix reboot hang

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 6083ab7b2f3f25022e2e8f4c42f14a8521f47873 ]

The following hang is observed when a 'reboot' command is issued:

# reboot
# Stopping network: OK
Stopping klogd: OK
Stopping syslogd: OK
umount: devtmpfs busy - remounted read-only
[    8.612079] EXT4-fs (mmcblk0p2): re-mounted. Opts: (null)
The system is going down NOW!
Sent SIGTERM to all processes
Sent SIGKILL to all processes
Requesting system reboot
[   10.694753] reboot: Restarting system
[   11.699008] Reboot failed -- System halted

Fix this problem by adding a .restart ops member.

---
 drivers/watchdog/imx7ulp_wdt.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/watchdog/imx7ulp_wdt.c +16 -0
index 5ce51026989a4..ba5d535a6db2e 100644
@@ -106,12 +106,28 @@ static int imx7ulp_wdt_set_timeout(struct watchdog_device *wdog,
 	return 0;
 }
 
+static int imx7ulp_wdt_restart(struct watchdog_device *wdog,
+			       unsigned long action, void *data)
+{
+	struct imx7ulp_wdt_device *wdt = watchdog_get_drvdata(wdog);
+
+	imx7ulp_wdt_enable(wdt->base, true);
+	imx7ulp_wdt_set_timeout(&wdt->wdd, 1);
+
+	/* wait for wdog to fire */
+	while (true)
+		;
+
+	return NOTIFY_DONE;
+}
+
 static const struct watchdog_ops imx7ulp_wdt_ops = {
 	.owner = THIS_MODULE,
 	.start = imx7ulp_wdt_start,
 	.stop  = imx7ulp_wdt_stop,
 	.ping  = imx7ulp_wdt_ping,
 	.set_timeout = imx7ulp_wdt_set_timeout,
+	.restart = imx7ulp_wdt_restart,
 };
 
 static const struct watchdog_info imx7ulp_wdt_info = {
Signature · 1 line
2.20.1
author maintainer
#88 lore raw

PATCH 85/134 watchdog: aspeed: Fix clock behaviour for ast2600

In reply to Sasha Levin

From: Joel Stanley <[email protected]>

[ Upstream commit c04571251b3d842096f1597f5d4badb508be016d ]

The ast2600 no longer uses bit 4 in the control register to indicate a 1MHz clock (It now controls whether this watchdog is reset by a SOC reset). This means we do not want to set it. It also does not need to be set for the ast2500, as it is read-only on that SoC.

The comment next to the clock rate selection wandered away from where it was set, so put it back next to the register setting it's describing.

---
 drivers/watchdog/aspeed_wdt.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)
drivers/watchdog/aspeed_wdt.c +10 -6
index 4ec0906bf12c0..7e00960651fa2 100644
@@ -258,11 +258,6 @@ static int aspeed_wdt_probe(struct platform_device *pdev)
 	if (IS_ERR(wdt->base))
 		return PTR_ERR(wdt->base);
 
-	/*
-	 * The ast2400 wdt can run at PCLK, or 1MHz. The ast2500 only
-	 * runs at 1MHz. We chose to always run at 1MHz, as there's no
-	 * good reason to have a faster watchdog counter.
-	 */
 	wdt->wdd.info = &aspeed_wdt_info;
 	wdt->wdd.ops = &aspeed_wdt_ops;
 	wdt->wdd.max_hw_heartbeat_ms = WDT_MAX_TIMEOUT_MS;
@@ -278,7 +273,16 @@ static int aspeed_wdt_probe(struct platform_device *pdev)
 		return -EINVAL;
 	config = ofdid->data;
 
-	wdt->ctrl = WDT_CTRL_1MHZ_CLK;
+	/*
+	 * On clock rates:
+	 *  - ast2400 wdt can run at PCLK, or 1MHz
+	 *  - ast2500 only runs at 1MHz, hard coding bit 4 to 1
+	 *  - ast2600 always runs at 1MHz
+	 *
+	 * Set the ast2400 to run at 1MHz as it simplifies the driver.
+	 */
+	if (of_device_is_compatible(np, "aspeed,ast2400-wdt"))
+		wdt->ctrl = WDT_CTRL_1MHZ_CLK;
 
 	/*
 	 * Control reset on a per-device basis to ensure the
Signature · 1 line
2.20.1
author maintainer
#89 lore raw

PATCH 86/134 watchdog: prevent deferral of watchdogd wakeup on RT

In reply to Sasha Levin

From: Julia Cartwright <[email protected]>

[ Upstream commit a19f89335f4bda3d77d991c96583e3e51856acbb ]

When PREEMPT_RT is enabled, all hrtimer expiry functions are deferred for execution into the context of ksoftirqd unless otherwise annotated.

Deferring the expiry of the hrtimer used by the watchdog core, however, is a waste, as the callback does nothing but queue a kthread work item and wakeup watchdogd.

It's worst then that, too: the deferral through ksoftirqd also means that for correct behavior a user must adjust the scheduling parameters of both watchdogd _and_ ksoftirqd, which is unnecessary and has other side effects (like causing unrelated expiry functions to execute at potentially elevated priority).

Instead, mark the hrtimer used by the watchdog core as being _HARD to allow it's execution directly from hardirq context. The work done in this expiry function is well-bounded and minimal.

A user still must adjust the scheduling parameters of the watchdogd to be correct w.r.t. their application needs.

---
 drivers/watchdog/watchdog_dev.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/watchdog/watchdog_dev.c +6 -4
index dbd2ad4c92948..d3acc0a7256ca 100644
@@ -158,7 +158,8 @@ static inline void watchdog_update_worker(struct watchdog_device *wdd)
 		ktime_t t = watchdog_next_keepalive(wdd);
 
 		if (t > 0)
-			hrtimer_start(&wd_data->timer, t, HRTIMER_MODE_REL);
+			hrtimer_start(&wd_data->timer, t,
+				      HRTIMER_MODE_REL_HARD);
 	} else {
 		hrtimer_cancel(&wd_data->timer);
 	}
@@ -177,7 +178,7 @@ static int __watchdog_ping(struct watchdog_device *wdd)
 	if (ktime_after(earliest_keepalive, now)) {
 		hrtimer_start(&wd_data->timer,
 			      ktime_sub(earliest_keepalive, now),
-			      HRTIMER_MODE_REL);
+			      HRTIMER_MODE_REL_HARD);
 		return 0;
 	}
 
@@ -971,7 +972,7 @@ static int watchdog_cdev_register(struct watchdog_device *wdd, dev_t devno)
 		return -ENODEV;
 
 	kthread_init_work(&wd_data->work, watchdog_ping_work);
-	hrtimer_init(&wd_data->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	hrtimer_init(&wd_data->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD);
 	wd_data->timer.function = watchdog_timer_expired;
 
 	if (wdd->id == 0) {
@@ -1019,7 +1020,8 @@ static int watchdog_cdev_register(struct watchdog_device *wdd, dev_t devno)
 		__module_get(wdd->ops->owner);
 		kref_get(&wd_data->kref);
 		if (handle_boot_enabled)
-			hrtimer_start(&wd_data->timer, 0, HRTIMER_MODE_REL);
+			hrtimer_start(&wd_data->timer, 0,
+				      HRTIMER_MODE_REL_HARD);
 		else
 			pr_info("watchdog%d running and kernel based pre-userspace handler disabled\n",
 				wdd->id);
Signature · 1 line
2.20.1
author maintainer
#90 lore raw

PATCH 87/134 watchdog: Fix the race between the release of watchdog_core_data and cdev

In reply to Sasha Levin

From: Kevin Hao <[email protected]>

[ Upstream commit 72139dfa2464e43957d330266994740bb7be2535 ]

The struct cdev is embedded in the struct watchdog_core_data. In the
current code, we manage the watchdog_core_data with a kref, but the
cdev is manged by a kobject. There is no any relationship between
this kref and kobject. So it is possible that the watchdog_core_data is
freed before the cdev is entirely released. We can easily get the
following call trace with CONFIG_DEBUG_KOBJECT_RELEASE and CONFIG_DEBUG_OBJECTS_TIMERS enabled.
ODEBUG: free active (active state 0) object type: timer_list hint: delayed_work_timer_fn+0x0/0x38

  WARNING: CPU: 23 PID: 1028 at lib/debugobjects.c:481 debug_print_object+0xb0/0xf0
  Modules linked in: softdog(-) deflate ctr twofish_generic twofish_common camellia_generic serpent_generic blowfish_generic blowfish_common cast5_generic cast_common cmac xcbc af_key sch_fq_codel openvswitch nsh nf_conncount nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4
  CPU: 23 PID: 1028 Comm: modprobe Not tainted 5.3.0-next-20190924-yoctodev-standard+ #180
  Hardware name: Marvell OcteonTX CN96XX board (DT)
  pstate: 00400009 (nzcv daif +PAN -UAO)
  pc : debug_print_object+0xb0/0xf0
  lr : debug_print_object+0xb0/0xf0
  sp : ffff80001cbcfc70
  x29: ffff80001cbcfc70 x28: ffff800010ea2128
  x27: ffff800010bad000 x26: 0000000000000000
  x25: ffff80001103c640 x24: ffff80001107b268
  x23: ffff800010bad9e8 x22: ffff800010ea2128
  x21: ffff000bc2c62af8 x20: ffff80001103c600
  x19: ffff800010e867d8 x18: 0000000000000060
  x17: 0000000000000000 x16: 0000000000000000
  x15: ffff000bd7240470 x14: 6e6968207473696c
  x13: 5f72656d6974203a x12: 6570797420746365
  x11: 6a626f2029302065 x10: 7461747320657669
  x9 : 7463612820657669 x8 : 3378302f3078302b
  x7 : 0000000000001d7a x6 : ffff800010fd5889
  x5 : 0000000000000000 x4 : 0000000000000000
  x3 : 0000000000000000 x2 : ffff000bff948548
  x1 : 276a1c9e1edc2300 x0 : 0000000000000000
  Call trace:
   debug_print_object+0xb0/0xf0
   debug_check_no_obj_freed+0x1e8/0x210
   kfree+0x1b8/0x368
   watchdog_cdev_unregister+0x88/0xc8
   watchdog_dev_unregister+0x38/0x48
   watchdog_unregister_device+0xa8/0x100
   softdog_exit+0x18/0xfec4 [softdog]
   __arm64_sys_delete_module+0x174/0x200
   el0_svc_handler+0xd0/0x1c8
   el0_svc+0x8/0xc

This is a common issue when using cdev embedded in a struct. Fortunately, we already have a mechanism to solve this kind of issue. Please see commit 233ed09d7fda ("chardev: add helper function to register char devs with a struct device") for more detail.

In this patch, we choose to embed the struct device into the watchdog_core_data, and use the API provided by the commit 233ed09d7fda to make sure that the release of watchdog_core_data and cdev are in sequence.

---
 drivers/watchdog/watchdog_dev.c | 70 +++++++++++++++------------------
 1 file changed, 32 insertions(+), 38 deletions(-)
drivers/watchdog/watchdog_dev.c +32 -38
index d3acc0a7256ca..62483a99105c0 100644
@@ -34,7 +34,6 @@
 #include <linux/init.h>		/* For __init/__exit/... */
 #include <linux/hrtimer.h>	/* For hrtimers */
 #include <linux/kernel.h>	/* For printk/panic/... */
-#include <linux/kref.h>		/* For data references */
 #include <linux/kthread.h>	/* For kthread_work */
 #include <linux/miscdevice.h>	/* For handling misc devices */
 #include <linux/module.h>	/* For module stuff/... */
@@ -52,14 +51,14 @@
 
 /*
  * struct watchdog_core_data - watchdog core internal data
- * @kref:	Reference count.
+ * @dev:	The watchdog's internal device
  * @cdev:	The watchdog's Character device.
  * @wdd:	Pointer to watchdog device.
  * @lock:	Lock for watchdog core.
  * @status:	Watchdog core internal status bits.
  */
 struct watchdog_core_data {
-	struct kref kref;
+	struct device dev;
 	struct cdev cdev;
 	struct watchdog_device *wdd;
 	struct mutex lock;
@@ -840,7 +839,7 @@ static int watchdog_open(struct inode *inode, struct file *file)
 	file->private_data = wd_data;
 
 	if (!hw_running)
-		kref_get(&wd_data->kref);
+		get_device(&wd_data->dev);
 
 	/*
 	 * open_timeout only applies for the first open from
@@ -861,11 +860,11 @@ out_clear:
 	return err;
 }
 
-static void watchdog_core_data_release(struct kref *kref)
+static void watchdog_core_data_release(struct device *dev)
 {
 	struct watchdog_core_data *wd_data;
 
-	wd_data = container_of(kref, struct watchdog_core_data, kref);
+	wd_data = container_of(dev, struct watchdog_core_data, dev);
 
 	kfree(wd_data);
 }
@@ -925,7 +924,7 @@ done:
 	 */
 	if (!running) {
 		module_put(wd_data->cdev.owner);
-		kref_put(&wd_data->kref, watchdog_core_data_release);
+		put_device(&wd_data->dev);
 	}
 	return 0;
 }
@@ -944,17 +943,22 @@ static struct miscdevice watchdog_miscdev = {
 	.fops		= &watchdog_fops,
 };
 
+static struct class watchdog_class = {
+	.name =		"watchdog",
+	.owner =	THIS_MODULE,
+	.dev_groups =	wdt_groups,
+};
+
 /*
  *	watchdog_cdev_register: register watchdog character device
  *	@wdd: watchdog device
- *	@devno: character device number
  *
  *	Register a watchdog character device including handling the legacy
  *	/dev/watchdog node. /dev/watchdog is actually a miscdevice and
  *	thus we set it up like that.
  */
 
-static int watchdog_cdev_register(struct watchdog_device *wdd, dev_t devno)
+static int watchdog_cdev_register(struct watchdog_device *wdd)
 {
 	struct watchdog_core_data *wd_data;
 	int err;
@@ -962,7 +966,6 @@ static int watchdog_cdev_register(struct watchdog_device *wdd, dev_t devno)
 	wd_data = kzalloc(sizeof(struct watchdog_core_data), GFP_KERNEL);
 	if (!wd_data)
 		return -ENOMEM;
-	kref_init(&wd_data->kref);
 	mutex_init(&wd_data->lock);
 
 	wd_data->wdd = wdd;
@@ -991,23 +994,33 @@ static int watchdog_cdev_register(struct watchdog_device *wdd, dev_t devno)
 		}
 	}
 
+	device_initialize(&wd_data->dev);
+	wd_data->dev.devt = MKDEV(MAJOR(watchdog_devt), wdd->id);
+	wd_data->dev.class = &watchdog_class;
+	wd_data->dev.parent = wdd->parent;
+	wd_data->dev.groups = wdd->groups;
+	wd_data->dev.release = watchdog_core_data_release;
+	dev_set_drvdata(&wd_data->dev, wdd);
+	dev_set_name(&wd_data->dev, "watchdog%d", wdd->id);
+
 	/* Fill in the data structures */
 	cdev_init(&wd_data->cdev, &watchdog_fops);
-	wd_data->cdev.owner = wdd->ops->owner;
 
 	/* Add the device */
-	err = cdev_add(&wd_data->cdev, devno, 1);
+	err = cdev_device_add(&wd_data->cdev, &wd_data->dev);
 	if (err) {
 		pr_err("watchdog%d unable to add device %d:%d\n",
 			wdd->id,  MAJOR(watchdog_devt), wdd->id);
 		if (wdd->id == 0) {
 			misc_deregister(&watchdog_miscdev);
 			old_wd_data = NULL;
-			kref_put(&wd_data->kref, watchdog_core_data_release);
+			put_device(&wd_data->dev);
 		}
 		return err;
 	}
 
+	wd_data->cdev.owner = wdd->ops->owner;
+
 	/* Record time of most recent heartbeat as 'just before now'. */
 	wd_data->last_hw_keepalive = ktime_sub(ktime_get(), 1);
 	watchdog_set_open_deadline(wd_data);
@@ -1018,7 +1031,7 @@ static int watchdog_cdev_register(struct watchdog_device *wdd, dev_t devno)
 	 */
 	if (watchdog_hw_running(wdd)) {
 		__module_get(wdd->ops->owner);
-		kref_get(&wd_data->kref);
+		get_device(&wd_data->dev);
 		if (handle_boot_enabled)
 			hrtimer_start(&wd_data->timer, 0,
 				      HRTIMER_MODE_REL_HARD);
@@ -1042,7 +1055,7 @@ static void watchdog_cdev_unregister(struct watchdog_device *wdd)
 {
 	struct watchdog_core_data *wd_data = wdd->wd_data;
 
-	cdev_del(&wd_data->cdev);
+	cdev_device_del(&wd_data->cdev, &wd_data->dev);
 	if (wdd->id == 0) {
 		misc_deregister(&watchdog_miscdev);
 		old_wd_data = NULL;
@@ -1061,15 +1074,9 @@ static void watchdog_cdev_unregister(struct watchdog_device *wdd)
 	hrtimer_cancel(&wd_data->timer);
 	kthread_cancel_work_sync(&wd_data->work);
 
-	kref_put(&wd_data->kref, watchdog_core_data_release);
+	put_device(&wd_data->dev);
 }
 
-static struct class watchdog_class = {
-	.name =		"watchdog",
-	.owner =	THIS_MODULE,
-	.dev_groups =	wdt_groups,
-};
-
 static int watchdog_reboot_notifier(struct notifier_block *nb,
 				    unsigned long code, void *data)
 {
@@ -1100,27 +1107,14 @@ static int watchdog_reboot_notifier(struct notifier_block *nb,
 
 int watchdog_dev_register(struct watchdog_device *wdd)
 {
-	struct device *dev;
-	dev_t devno;
 	int ret;
 
-	devno = MKDEV(MAJOR(watchdog_devt), wdd->id);
-
-	ret = watchdog_cdev_register(wdd, devno);
+	ret = watchdog_cdev_register(wdd);
 	if (ret)
 		return ret;
 
-	dev = device_create_with_groups(&watchdog_class, wdd->parent,
-					devno, wdd, wdd->groups,
-					"watchdog%d", wdd->id);
-	if (IS_ERR(dev)) {
-		watchdog_cdev_unregister(wdd);
-		return PTR_ERR(dev);
-	}
-
 	ret = watchdog_register_pretimeout(wdd);
 	if (ret) {
-		device_destroy(&watchdog_class, devno);
 		watchdog_cdev_unregister(wdd);
 		return ret;
 	}
@@ -1128,7 +1122,8 @@ int watchdog_dev_register(struct watchdog_device *wdd)
 	if (test_bit(WDOG_STOP_ON_REBOOT, &wdd->status)) {
 		wdd->reboot_nb.notifier_call = watchdog_reboot_notifier;
 
-		ret = devm_register_reboot_notifier(dev, &wdd->reboot_nb);
+		ret = devm_register_reboot_notifier(&wdd->wd_data->dev,
+						    &wdd->reboot_nb);
 		if (ret) {
 			pr_err("watchdog%d: Cannot register reboot notifier (%d)\n",
 			       wdd->id, ret);
@@ -1150,7 +1145,6 @@ int watchdog_dev_register(struct watchdog_device *wdd)
 void watchdog_dev_unregister(struct watchdog_device *wdd)
 {
 	watchdog_unregister_pretimeout(wdd);
-	device_destroy(&watchdog_class, wdd->wd_data->cdev.dev);
 	watchdog_cdev_unregister(wdd);
 }
 
Signature · 1 line
2.20.1
author maintainer
#91 lore raw

PATCH 88/134 powerpc/fixmap: Use __fix_to_virt() instead of fix_to_virt()

In reply to Sasha Levin

From: Christophe Leroy <[email protected]>

[ Upstream commit 77693a5fb57be4606a6024ec8e3076f9499b906b ]

Modify back __set_fixmap() to using __fix_to_virt() instead of fix_to_virt() otherwise the following happens because it seems GCC doesn't see idx as a builtin const.

CC mm/early_ioremap.o
In file included from ./include/linux/kernel.h:11:0,

                 from mm/early_ioremap.c:11:

In function ‘fix_to_virt’,
inlined from ‘__set_fixmap’ at ./arch/powerpc/include/asm/fixmap.h:87:2,
inlined from ‘__early_ioremap’ at mm/early_ioremap.c:156:4:
./include/linux/compiler.h:350:38: error: call to ‘__compiletime_assert_32’ declared with attribute error: BUILD_BUG_ON failed: idx >= __end_of_fixed_addresses
_compiletime_assert(condition, msg, __compiletime_assert_, __LINE__)

                                      ^
./include/linux/compiler.h:331:4: note: in definition of macro ‘__compiletime_assert’
    prefix ## suffix();    \
    ^

./include/linux/compiler.h:350:2: note: in expansion of macro ‘_compiletime_assert’
_compiletime_assert(condition, msg, __compiletime_assert_, __LINE__) ^
./include/linux/build_bug.h:39:37: note: in expansion of macro ‘compiletime_assert’

 #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
                                     ^

./include/linux/build_bug.h:50:2: note: in expansion of macro ‘BUILD_BUG_ON_MSG’
BUILD_BUG_ON_MSG(condition, "BUILD_BUG_ON failed: " #condition)
^
./include/asm-generic/fixmap.h:32:2: note: in expansion of macro ‘BUILD_BUG_ON’
BUILD_BUG_ON(idx >= __end_of_fixed_addresses);
^

---
 arch/powerpc/include/asm/fixmap.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
arch/powerpc/include/asm/fixmap.h +6 -1
index 0cfc365d814ba..722289a1d000e 100644
@@ -77,7 +77,12 @@ enum fixed_addresses {
 static inline void __set_fixmap(enum fixed_addresses idx,
 				phys_addr_t phys, pgprot_t flags)
 {
-	map_kernel_page(fix_to_virt(idx), phys, flags);
+	if (__builtin_constant_p(idx))
+		BUILD_BUG_ON(idx >= __end_of_fixed_addresses);
+	else if (WARN_ON(idx >= __end_of_fixed_addresses))
+		return;
+
+	map_kernel_page(__fix_to_virt(idx), phys, flags);
 }
 
 #endif /* !__ASSEMBLY__ */
Signature · 1 line
2.20.1
author maintainer
#92 lore raw

PATCH 89/134 ext4: work around deleting a file with i_nlink == 0 safely

In reply to Sasha Levin

From: Theodore Ts'o <[email protected]>

[ Upstream commit c7df4a1ecb8579838ec8c56b2bb6a6716e974f37 ]

If the file system is corrupted such that a file's i_links_count is too small, then it's possible that when unlinking that file, i_nlink will already be zero. Previously we were working around this kind of corruption by forcing i_nlink to one; but we were doing this before trying to delete the directory entry --- and if the file system is corrupted enough that ext4_delete_entry() fails, then we exit with i_nlink elevated, and this causes the orphan inode list handling to be FUBAR'ed, such that when we unmount the file system, the orphan inode list can get corrupted.

A better way to fix this is to simply skip trying to call drop_nlink() if i_nlink is already zero, thus moving the check to the place where it makes the most sense.

https://bugzilla.kernel.org/show_bug.cgi?id=205433

---
 fs/ext4/namei.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
fs/ext4/namei.c +5 -6
index a427d2031a8da..923476e3aefbc 100644
@@ -3182,18 +3182,17 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry)
 	if (IS_DIRSYNC(dir))
 		ext4_handle_sync(handle);
 
-	if (inode->i_nlink == 0) {
-		ext4_warning_inode(inode, "Deleting file '%.*s' with no links",
-				   dentry->d_name.len, dentry->d_name.name);
-		set_nlink(inode, 1);
-	}
 	retval = ext4_delete_entry(handle, dir, de, bh);
 	if (retval)
 		goto end_unlink;
 	dir->i_ctime = dir->i_mtime = current_time(dir);
 	ext4_update_dx_flag(dir);
 	ext4_mark_inode_dirty(handle, dir);
-	drop_nlink(inode);
+	if (inode->i_nlink == 0)
+		ext4_warning_inode(inode, "Deleting file '%.*s' with no links",
+				   dentry->d_name.len, dentry->d_name.name);
+	else
+		drop_nlink(inode);
 	if (!inode->i_nlink)
 		ext4_orphan_add(handle, inode);
 	inode->i_ctime = current_time(inode);
Signature · 1 line
2.20.1
author maintainer
#93 lore raw

PATCH 90/134 clk: Fix memory leak in clk_unregister()

In reply to Sasha Levin

From: Kishon Vijay Abraham I <[email protected]>

[ Upstream commit 8247470772beb38822f226c99a2ed8c195f6b438 ]

Memory allocated in alloc_clk() for 'struct clk' and 'const char *con_id' while invoking clk_register() is never freed in clk_unregister(), resulting in kmemleak showing the following backtrace.

backtrace:

    [<00000000546f5dd0>] kmem_cache_alloc+0x18c/0x270
    [<0000000073a32862>] alloc_clk+0x30/0x70
    [<0000000082942480>] __clk_register+0xc8/0x760
    [<000000005c859fca>] devm_clk_register+0x54/0xb0
    [<00000000868834a8>] 0xffff800008c60950
    [<00000000d5a80534>] platform_drv_probe+0x50/0xa0
    [<000000001b3889fc>] really_probe+0x108/0x348
    [<00000000953fa60a>] driver_probe_device+0x58/0x100
    [<0000000008acc17c>] device_driver_attach+0x6c/0x90
    [<0000000022813df3>] __driver_attach+0x84/0xc8
    [<00000000448d5443>] bus_for_each_dev+0x74/0xc8
    [<00000000294aa93f>] driver_attach+0x20/0x28
    [<00000000e5e52626>] bus_add_driver+0x148/0x1f0
    [<000000001de21efc>] driver_register+0x60/0x110
    [<00000000af07c068>] __platform_driver_register+0x40/0x48
    [<0000000060fa80ee>] 0xffff800008c66020

Fix it here.

Cc Tomi Valkeinen <[email protected]>
Cc Tero Kristo <[email protected]>
Fixes 1df4046a93e0 ("clk: Combine __clk_get() and __clk_create_clk()")
---
 drivers/clk/clk.c | 1 +
 1 file changed, 1 insertion(+)
drivers/clk/clk.c +1 -0
index 1c677d7f7f530..2f2eea26c3750 100644
@@ -3879,6 +3879,7 @@ void clk_unregister(struct clk *clk)
 					__func__, clk->core->name);
 
 	kref_put(&clk->core->ref, __clk_release);
+	free_clk(clk);
 unlock:
 	clk_prepare_unlock();
 }
Signature · 1 line
2.20.1
author maintainer
#94 lore raw

PATCH 91/134 scsi: pm80xx: Fix for SATA device discovery

In reply to Sasha Levin

From: peter chang <[email protected]>

[ Upstream commit ce21c63ee995b7a8b7b81245f2cee521f8c3c220 ]

Driver was missing complete() call in mpi_sata_completion which result in SATA abort error handling timing out. That causes the device to be left in the in_recovery state so subsequent commands sent to the device fail and the OS removes access to it.

---
 drivers/scsi/pm8001/pm80xx_hwi.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/scsi/pm8001/pm80xx_hwi.c +2 -0
index 73261902d75d5..161bf4760eac7 100644
@@ -2382,6 +2382,8 @@ mpi_sata_completion(struct pm8001_hba_info *pm8001_ha, void *piomb)
 			pm8001_printk("task 0x%p done with io_status 0x%x"
 			" resp 0x%x stat 0x%x but aborted by upper layer!\n",
 			t, status, ts->resp, ts->stat));
+		if (t->slow_task)
+			complete(&t->slow_task->completion);
 		pm8001_ccb_task_free(pm8001_ha, t, ccb, tag);
 	} else {
 		spin_unlock_irqrestore(&t->task_state_lock, flags);
Signature · 1 line
2.20.1
author maintainer
#95 lore raw

PATCH 92/134 scsi: ufs: Fix error handing during hibern8 enter

In reply to Sasha Levin

From: Subhash Jadavani <[email protected]>

[ Upstream commit 6d303e4b19d694cdbebf76bcdb51ada664ee953d ]

During clock gating (ufshcd_gate_work()), we first put the link hibern8 by calling ufshcd_uic_hibern8_enter() and if ufshcd_uic_hibern8_enter() returns success (0) then we gate all the clocks. Now let’s zoom in to what ufshcd_uic_hibern8_enter() does internally: It calls __ufshcd_uic_hibern8_enter() and if failure is encountered, link recovery shall put the link back to the highest HS gear and returns success (0) to ufshcd_uic_hibern8_enter() which is the issue as link is still in active state due to recovery! Now ufshcd_uic_hibern8_enter() returns success to ufshcd_gate_work() and hence it goes ahead with gating the UFS clock while link is still in active state hence I believe controller would raise UIC error interrupts. But when we service the interrupt, clocks might have already been disabled!

This change fixes for this by returning failure from __ufshcd_uic_hibern8_enter() if recovery succeeds as link is still not in hibern8, upon receiving the error ufshcd_hibern8_enter() would initiate retry to put the link state back into hibern8.

---
 drivers/scsi/ufs/ufshcd.c | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)
drivers/scsi/ufs/ufshcd.c +14 -5
index 358ff7b015680..0036dcffc4a90 100644
@@ -3885,15 +3885,24 @@ static int __ufshcd_uic_hibern8_enter(struct ufs_hba *hba)
 			     ktime_to_us(ktime_sub(ktime_get(), start)), ret);
 
 	if (ret) {
+		int err;
+
 		dev_err(hba->dev, "%s: hibern8 enter failed. ret = %d\n",
 			__func__, ret);
 
 		/*
-		 * If link recovery fails then return error so that caller
-		 * don't retry the hibern8 enter again.
+		 * If link recovery fails then return error code returned from
+		 * ufshcd_link_recovery().
+		 * If link recovery succeeds then return -EAGAIN to attempt
+		 * hibern8 enter retry again.
 		 */
-		if (ufshcd_link_recovery(hba))
-			ret = -ENOLINK;
+		err = ufshcd_link_recovery(hba);
+		if (err) {
+			dev_err(hba->dev, "%s: link recovery failed", __func__);
+			ret = err;
+		} else {
+			ret = -EAGAIN;
+		}
 	} else
 		ufshcd_vops_hibern8_notify(hba, UIC_CMD_DME_HIBER_ENTER,
 								POST_CHANGE);
@@ -3907,7 +3916,7 @@ static int ufshcd_uic_hibern8_enter(struct ufs_hba *hba)
 
 	for (retries = UIC_HIBERN8_ENTER_RETRIES; retries > 0; retries--) {
 		ret = __ufshcd_uic_hibern8_enter(hba);
-		if (!ret || ret == -ENOLINK)
+		if (!ret)
 			goto out;
 	}
 out:
Signature · 1 line
2.20.1
author maintainer
#96 lore raw

PATCH 93/134 scsi: scsi_debug: num_tgts must be >= 0

In reply to Sasha Levin

From: Maurizio Lombardi <[email protected]>

[ Upstream commit aa5334c4f3014940f11bf876e919c956abef4089 ]

Passing the parameter "num_tgts=-1" will start an infinite loop that exhausts the system memory

---
 drivers/scsi/scsi_debug.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/scsi/scsi_debug.c +5 -0
index d323523f5f9de..32965ec76965a 100644
@@ -5263,6 +5263,11 @@ static int __init scsi_debug_init(void)
 		return -EINVAL;
 	}
 
+	if (sdebug_num_tgts < 0) {
+		pr_err("num_tgts must be >= 0\n");
+		return -EINVAL;
+	}
+
 	if (sdebug_guard > 1) {
 		pr_err("guard must be 0 or 1\n");
 		return -EINVAL;
Signature · 1 line
2.20.1
author maintainer
#97 lore raw

PATCH 94/134 scsi: NCR5380: Add disconnect_mask module parameter

In reply to Sasha Levin

From: Finn Thain <[email protected]>

[ Upstream commit 0b7a223552d455bcfba6fb9cfc5eef2b5fce1491 ]

Add a module parameter to inhibit disconnect/reselect for individual targets. This gains compatibility with Aztec PowerMonster SCSI/SATA adapters with buggy firmware. (No fix is available from the vendor.)

Apparently these adapters pass-through the product/vendor of the attached SATA device. Since they can't be identified from the response to an INQUIRY command, a device blacklist flag won't work.

---
 drivers/scsi/NCR5380.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/scsi/NCR5380.c +5 -1
index 536426f25e866..d4401c768a0c7 100644
@@ -129,6 +129,9 @@
 #define NCR5380_release_dma_irq(x)
 #endif
 
+static unsigned int disconnect_mask = ~0;
+module_param(disconnect_mask, int, 0444);
+
 static int do_abort(struct Scsi_Host *);
 static void do_reset(struct Scsi_Host *);
 static void bus_reset_cleanup(struct Scsi_Host *);
@@ -954,7 +957,8 @@ static bool NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd)
 	int err;
 	bool ret = true;
 	bool can_disconnect = instance->irq != NO_IRQ &&
-			      cmd->cmnd[0] != REQUEST_SENSE;
+			      cmd->cmnd[0] != REQUEST_SENSE &&
+			      (disconnect_mask & BIT(scmd_id(cmd)));
 
 	NCR5380_dprint(NDEBUG_ARBITRATION, instance);
 	dsprintk(NDEBUG_ARBITRATION, instance, "starting arbitration, id = %d\n",
Signature · 1 line
2.20.1
author maintainer
#98 lore raw

PATCH 95/134 scsi: target: core: Release SPC-2 reservations when closing a session

In reply to Sasha Levin

From: Bart Van Assche <[email protected]>

[ Upstream commit 80647a89eaf3f2549741648f3230cd6ff68c23b4 ]

The SCSI specs require releasing SPC-2 reservations when a session is closed. Make sure that the target core does this.

Running the libiscsi tests triggers the KASAN complaint shown below. This patch fixes that use-after-free.

BUG: KASAN: use-after-free in target_check_reservation+0x171/0x980 [target_core_mod]
Read of size 8 at addr ffff88802ecd1878 by task iscsi_trx/17200

CPU: 0 PID: 17200 Comm: iscsi_trx Not tainted 5.4.0-rc1-dbg+ #1
Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011
Call Trace:
 dump_stack+0x8a/0xd6
 print_address_description.constprop.0+0x40/0x60
 __kasan_report.cold+0x1b/0x34
 kasan_report+0x16/0x20
 __asan_load8+0x58/0x90
 target_check_reservation+0x171/0x980 [target_core_mod]
 __target_execute_cmd+0xb1/0xf0 [target_core_mod]
 target_execute_cmd+0x22d/0x4d0 [target_core_mod]
 transport_generic_new_cmd+0x31f/0x5b0 [target_core_mod]
 transport_handle_cdb_direct+0x6f/0x90 [target_core_mod]
 iscsit_execute_cmd+0x381/0x3f0 [iscsi_target_mod]
 iscsit_sequence_cmd+0x13b/0x1f0 [iscsi_target_mod]
 iscsit_process_scsi_cmd+0x4c/0x130 [iscsi_target_mod]
 iscsit_get_rx_pdu+0x8e8/0x15f0 [iscsi_target_mod]
 iscsi_target_rx_thread+0x105/0x1b0 [iscsi_target_mod]
 kthread+0x1bc/0x210
 ret_from_fork+0x24/0x30

Allocated by task 1079:
 save_stack+0x23/0x90
 __kasan_kmalloc.constprop.0+0xcf/0xe0
 kasan_slab_alloc+0x12/0x20
 kmem_cache_alloc+0xfe/0x3a0
 transport_alloc_session+0x29/0x80 [target_core_mod]
 iscsi_target_login_thread+0xceb/0x1920 [iscsi_target_mod]
 kthread+0x1bc/0x210
 ret_from_fork+0x24/0x30

Freed by task 17193:
 save_stack+0x23/0x90
 __kasan_slab_free+0x13a/0x190
 kasan_slab_free+0x12/0x20
 kmem_cache_free+0xc8/0x3e0
 transport_free_session+0x179/0x2f0 [target_core_mod]
 transport_deregister_session+0x121/0x170 [target_core_mod]
 iscsit_close_session+0x12c/0x350 [iscsi_target_mod]
 iscsit_logout_post_handler+0x136/0x380 [iscsi_target_mod]
 iscsit_response_queue+0x8fa/0xc00 [iscsi_target_mod]
 iscsi_target_tx_thread+0x28e/0x390 [iscsi_target_mod]
 kthread+0x1bc/0x210
 ret_from_fork+0x24/0x30

The buggy address belongs to the object at ffff88802ecd1860
 which belongs to the cache se_sess_cache of size 352
The buggy address is located 24 bytes inside of
 352-byte region [ffff88802ecd1860, ffff88802ecd19c0)
The buggy address belongs to the page:
page:ffffea0000bb3400 refcount:1 mapcount:0 mapping:ffff8880bef2ed00 index:0x0 compound_mapcount: 0
flags: 0x1000000000010200(slab|head)
raw: 1000000000010200 dead000000000100 dead000000000122 ffff8880bef2ed00
raw: 0000000000000000 0000000080270027 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff88802ecd1700: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88802ecd1780: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
>ffff88802ecd1800: fb fb fb fb fc fc fc fc fc fc fc fc fb fb fb fb
                                                                ^
 ffff88802ecd1880: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff88802ecd1900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
---
 drivers/target/target_core_transport.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
drivers/target/target_core_transport.c +15 -0
index 7f06a62f8661d..eda8b4736c158 100644
@@ -584,6 +584,15 @@ void transport_free_session(struct se_session *se_sess)
 }
 EXPORT_SYMBOL(transport_free_session);
 
+static int target_release_res(struct se_device *dev, void *data)
+{
+	struct se_session *sess = data;
+
+	if (dev->reservation_holder = sess)
+		target_release_reservation(dev);
+	return 0;
+}
+
 void transport_deregister_session(struct se_session *se_sess)
 {
 	struct se_portal_group *se_tpg = se_sess->se_tpg;
@@ -600,6 +609,12 @@ void transport_deregister_session(struct se_session *se_sess)
 	se_sess->fabric_sess_ptr = NULL;
 	spin_unlock_irqrestore(&se_tpg->session_lock, flags);
 
+	/*
+	 * Since the session is being removed, release SPC-2
+	 * reservations held by the session that is disappearing.
+	 */
+	target_for_each_device(target_release_res, se_sess);
+
 	pr_debug("TARGET_CORE[%s]: Deregistered fabric_sess\n",
 		se_tpg->se_tpg_tfo->fabric_name);
 	/*
Signature · 1 line
2.20.1
author maintainer
#99 lore raw

PATCH 96/134 scsi: ufs: Fix up auto hibern8 enablement

In reply to Sasha Levin

From: Can Guo <[email protected]>

[ Upstream commit 71d848b8d97ec0f8e993d63cf9de6ac8b3f7c43d ]

Fix up possible unclocked register access to auto hibern8 register in resume path and through sysfs entry. Meanwhile, enable auto hibern8 only after device is fully initialized in probe path.

---
 drivers/scsi/ufs/ufs-sysfs.c | 15 +++++++++------
 drivers/scsi/ufs/ufshcd.c    | 14 +++++++-------
 drivers/scsi/ufs/ufshcd.h    |  2 ++
 3 files changed, 18 insertions(+), 13 deletions(-)
drivers/scsi/ufs/ufs-sysfs.c +9 -6
index 969a36b15897b..ad2abc96c0f19 100644
@@ -126,13 +126,16 @@ static void ufshcd_auto_hibern8_update(struct ufs_hba *hba, u32 ahit)
 		return;
 
 	spin_lock_irqsave(hba->host->host_lock, flags);
-	if (hba->ahit == ahit)
-		goto out_unlock;
-	hba->ahit = ahit;
-	if (!pm_runtime_suspended(hba->dev))
-		ufshcd_writel(hba, hba->ahit, REG_AUTO_HIBERNATE_IDLE_TIMER);
-out_unlock:
+	if (hba->ahit != ahit)
+		hba->ahit = ahit;
 	spin_unlock_irqrestore(hba->host->host_lock, flags);
+	if (!pm_runtime_suspended(hba->dev)) {
+		pm_runtime_get_sync(hba->dev);
+		ufshcd_hold(hba, false);
+		ufshcd_auto_hibern8_enable(hba);
+		ufshcd_release(hba);
+		pm_runtime_put(hba->dev);
+	}
 }
 
 /* Convert Auto-Hibernate Idle Timer register value to microseconds */
drivers/scsi/ufs/ufshcd.c +7 -7
index 0036dcffc4a90..25a6a25b17a28 100644
@@ -3950,7 +3950,7 @@ static int ufshcd_uic_hibern8_exit(struct ufs_hba *hba)
 	return ret;
 }
 
-static void ufshcd_auto_hibern8_enable(struct ufs_hba *hba)
+void ufshcd_auto_hibern8_enable(struct ufs_hba *hba)
 {
 	unsigned long flags;
 
@@ -6890,9 +6890,6 @@ static int ufshcd_probe_hba(struct ufs_hba *hba)
 	/* UniPro link is active now */
 	ufshcd_set_link_active(hba);
 
-	/* Enable Auto-Hibernate if configured */
-	ufshcd_auto_hibern8_enable(hba);
-
 	ret = ufshcd_verify_dev_init(hba);
 	if (ret)
 		goto out;
@@ -6943,6 +6940,9 @@ static int ufshcd_probe_hba(struct ufs_hba *hba)
 	/* set the state as operational after switching to desired gear */
 	hba->ufshcd_state = UFSHCD_STATE_OPERATIONAL;
 
+	/* Enable Auto-Hibernate if configured */
+	ufshcd_auto_hibern8_enable(hba);
+
 	/*
 	 * If we are in error handling context or in power management callbacks
 	 * context, no need to scan the host
@@ -7959,12 +7959,12 @@ static int ufshcd_resume(struct ufs_hba *hba, enum ufs_pm_op pm_op)
 	if (hba->clk_scaling.is_allowed)
 		ufshcd_resume_clkscaling(hba);
 
-	/* Schedule clock gating in case of no access to UFS device yet */
-	ufshcd_release(hba);
-
 	/* Enable Auto-Hibernate if configured */
 	ufshcd_auto_hibern8_enable(hba);
 
+	/* Schedule clock gating in case of no access to UFS device yet */
+	ufshcd_release(hba);
+
 	goto out;
 
 set_old_link_state:
drivers/scsi/ufs/ufshcd.h +2 -0
index c94cfda528290..52c9676a12425 100644
@@ -916,6 +916,8 @@ int ufshcd_query_attr(struct ufs_hba *hba, enum query_opcode opcode,
 int ufshcd_query_flag(struct ufs_hba *hba, enum query_opcode opcode,
 	enum flag_idn idn, bool *flag_res);
 
+void ufshcd_auto_hibern8_enable(struct ufs_hba *hba);
+
 #define SD_ASCII_STD true
 #define SD_RAW false
 int ufshcd_read_string_desc(struct ufs_hba *hba, u8 desc_index,
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#100 lore raw

PATCH 97/134 scsi: iscsi: Don't send data to unbound connection

In reply to Sasha Levin

From: Anatol Pomazau <[email protected]>

[ Upstream commit 238191d65d7217982d69e21c1d623616da34b281 ]

If a faulty initiator fails to bind the socket to the iSCSI connection before emitting a command, for instance, a subsequent send_pdu, it will crash the kernel due to a null pointer dereference in sock_sendmsg(), as shown in the log below. This patch makes sure the bind succeeded before trying to use the socket.

BUG: kernel NULL pointer dereference, address: 0000000000000018
 #PF: supervisor read access in kernel mode
 #PF: error_code(0x0000) - not-present page
PGD 0 P4D 0
Oops: 0000 [#1] SMP PTI
CPU: 3 PID: 7 Comm: kworker/u8:0 Not tainted 5.4.0-rc2.iscsi+ #13
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014
[   24.158246] Workqueue: iscsi_q_0 iscsi_xmitworker
[   24.158883] RIP: 0010:apparmor_socket_sendmsg+0x5/0x20
[...]
[   24.161739] RSP: 0018:ffffab6440043ca0 EFLAGS: 00010282
[   24.162400] RAX: ffffffff891c1c00 RBX: ffffffff89d53968 RCX: 0000000000000001
[   24.163253] RDX: 0000000000000030 RSI: ffffab6440043d00 RDI: 0000000000000000
[   24.164104] RBP: 0000000000000030 R08: 0000000000000030 R09: 0000000000000030
[   24.165166] R10: ffffffff893e66a0 R11: 0000000000000018 R12: ffffab6440043d00
[   24.166038] R13: 0000000000000000 R14: 0000000000000000 R15: ffff9d5575a62e90
[   24.166919] FS:  0000000000000000(0000) GS:ffff9d557db80000(0000) knlGS:0000000000000000
[   24.167890] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   24.168587] CR2: 0000000000000018 CR3: 000000007a838000 CR4: 00000000000006e0
[   24.169451] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   24.170320] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[   24.171214] Call Trace:
[   24.171537]  security_socket_sendmsg+0x3a/0x50
[   24.172079]  sock_sendmsg+0x16/0x60
[   24.172506]  iscsi_sw_tcp_xmit_segment+0x77/0x120
[   24.173076]  iscsi_sw_tcp_pdu_xmit+0x58/0x170
[   24.173604]  ? iscsi_dbg_trace+0x63/0x80
[   24.174087]  iscsi_tcp_task_xmit+0x101/0x280
[   24.174666]  iscsi_xmit_task+0x83/0x110
[   24.175206]  iscsi_xmitworker+0x57/0x380
[   24.175757]  ? __schedule+0x2a2/0x700
[   24.176273]  process_one_work+0x1b5/0x360
[   24.176837]  worker_thread+0x50/0x3c0
[   24.177353]  kthread+0xf9/0x130
[   24.177799]  ? process_one_work+0x360/0x360
[   24.178401]  ? kthread_park+0x90/0x90
[   24.178915]  ret_from_fork+0x35/0x40
[   24.179421] Modules linked in:
[   24.179856] CR2: 0000000000000018
[   24.180327] ---[ end trace b4b7674b6df5f480 ]---
---
 drivers/scsi/iscsi_tcp.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/scsi/iscsi_tcp.c +8 -0
index 7bedbe8777049..0bc63a7ab41c8 100644
@@ -369,8 +369,16 @@ static int iscsi_sw_tcp_pdu_xmit(struct iscsi_task *task)
 {
 	struct iscsi_conn *conn = task->conn;
 	unsigned int noreclaim_flag;
+	struct iscsi_tcp_conn *tcp_conn = conn->dd_data;
+	struct iscsi_sw_tcp_conn *tcp_sw_conn = tcp_conn->dd_data;
 	int rc = 0;
 
+	if (!tcp_sw_conn->sock) {
+		iscsi_conn_printk(KERN_ERR, conn,
+				  "Transport not bound to socket!\n");
+		return -EINVAL;
+	}
+
 	noreclaim_flag = memalloc_noreclaim_save();
 
 	while (iscsi_sw_tcp_xmit_qlen(conn)) {
Signature · 1 line
2.20.1
author maintainer
#101 lore raw

PATCH 98/134 scsi: target: iscsi: Wait for all commands to finish before freeing a session

In reply to Sasha Levin

From: Bart Van Assche <[email protected]>

[ Upstream commit e9d3009cb936bd0faf0719f68d98ad8afb1e613b ]

The iSCSI target driver is the only target driver that does not wait for ongoing commands to finish before freeing a session. Make the iSCSI target driver wait for ongoing commands to finish before freeing a session. This patch fixes the following KASAN complaint:

BUG: KASAN: use-after-free in __lock_acquire+0xb1a/0x2710
Read of size 8 at addr ffff8881154eca70 by task kworker/0:2/247

CPU: 0 PID: 247 Comm: kworker/0:2 Not tainted 5.4.0-rc1-dbg+ #6
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014
Workqueue: target_completion target_complete_ok_work [target_core_mod]
Call Trace:
 dump_stack+0x8a/0xd6
 print_address_description.constprop.0+0x40/0x60
 __kasan_report.cold+0x1b/0x33
 kasan_report+0x16/0x20
 __asan_load8+0x58/0x90
 __lock_acquire+0xb1a/0x2710
 lock_acquire+0xd3/0x200
 _raw_spin_lock_irqsave+0x43/0x60
 target_release_cmd_kref+0x162/0x7f0 [target_core_mod]
 target_put_sess_cmd+0x2e/0x40 [target_core_mod]
 lio_check_stop_free+0x12/0x20 [iscsi_target_mod]
 transport_cmd_check_stop_to_fabric+0xd8/0xe0 [target_core_mod]
 target_complete_ok_work+0x1b0/0x790 [target_core_mod]
 process_one_work+0x549/0xa40
 worker_thread+0x7a/0x5d0
 kthread+0x1bc/0x210
 ret_from_fork+0x24/0x30

Allocated by task 889:
 save_stack+0x23/0x90
 __kasan_kmalloc.constprop.0+0xcf/0xe0
 kasan_slab_alloc+0x12/0x20
 kmem_cache_alloc+0xf6/0x360
 transport_alloc_session+0x29/0x80 [target_core_mod]
 iscsi_target_login_thread+0xcd6/0x18f0 [iscsi_target_mod]
 kthread+0x1bc/0x210
 ret_from_fork+0x24/0x30

Freed by task 1025:
 save_stack+0x23/0x90
 __kasan_slab_free+0x13a/0x190
 kasan_slab_free+0x12/0x20
 kmem_cache_free+0x146/0x400
 transport_free_session+0x179/0x2f0 [target_core_mod]
 transport_deregister_session+0x130/0x180 [target_core_mod]
 iscsit_close_session+0x12c/0x350 [iscsi_target_mod]
 iscsit_logout_post_handler+0x136/0x380 [iscsi_target_mod]
 iscsit_response_queue+0x8de/0xbe0 [iscsi_target_mod]
 iscsi_target_tx_thread+0x27f/0x370 [iscsi_target_mod]
 kthread+0x1bc/0x210
 ret_from_fork+0x24/0x30

The buggy address belongs to the object at ffff8881154ec9c0
 which belongs to the cache se_sess_cache of size 352
The buggy address is located 176 bytes inside of
 352-byte region [ffff8881154ec9c0, ffff8881154ecb20)
The buggy address belongs to the page:
page:ffffea0004553b00 refcount:1 mapcount:0 mapping:ffff888101755400 index:0x0 compound_mapcount: 0
flags: 0x2fff000000010200(slab|head)
raw: 2fff000000010200 dead000000000100 dead000000000122 ffff888101755400
raw: 0000000000000000 0000000080130013 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff8881154ec900: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
 ffff8881154ec980: fc fc fc fc fc fc fc fc fb fb fb fb fb fb fb fb
>ffff8881154eca00: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                                                             ^
 ffff8881154eca80: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
 ffff8881154ecb00: fb fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc
---
 drivers/target/iscsi/iscsi_target.c | 10 ++++++++--
 include/scsi/iscsi_proto.h          |  1 +
 2 files changed, 9 insertions(+), 2 deletions(-)
drivers/target/iscsi/iscsi_target.c +8 -2
index d19e051f2bc23..f194ffc4699e0 100644
@@ -1165,7 +1165,9 @@ int iscsit_setup_scsi_cmd(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 		hdr->cmdsn, be32_to_cpu(hdr->data_length), payload_length,
 		conn->cid);
 
-	target_get_sess_cmd(&cmd->se_cmd, true);
+	if (target_get_sess_cmd(&cmd->se_cmd, true) < 0)
+		return iscsit_add_reject_cmd(cmd,
+				ISCSI_REASON_WAITING_FOR_LOGOUT, buf);
 
 	cmd->sense_reason = transport_lookup_cmd_lun(&cmd->se_cmd,
 						     scsilun_to_int(&hdr->lun));
@@ -2002,7 +2004,9 @@ iscsit_handle_task_mgt_cmd(struct iscsi_conn *conn, struct iscsi_cmd *cmd,
 			      conn->sess->se_sess, 0, DMA_NONE,
 			      TCM_SIMPLE_TAG, cmd->sense_buffer + 2);
 
-	target_get_sess_cmd(&cmd->se_cmd, true);
+	if (target_get_sess_cmd(&cmd->se_cmd, true) < 0)
+		return iscsit_add_reject_cmd(cmd,
+				ISCSI_REASON_WAITING_FOR_LOGOUT, buf);
 
 	/*
 	 * TASK_REASSIGN for ERL=2 / connection stays inside of
@@ -4232,6 +4236,8 @@ int iscsit_close_connection(
 	 * must wait until they have completed.
 	 */
 	iscsit_check_conn_usage_count(conn);
+	target_sess_cmd_list_set_waiting(sess->se_sess);
+	target_wait_for_sess_cmds(sess->se_sess);
 
 	ahash_request_free(conn->conn_tx_hash);
 	if (conn->conn_rx_hash) {
include/scsi/iscsi_proto.h +1 -0
index b71b5c4f418c5..533f56733ba84 100644
@@ -627,6 +627,7 @@ struct iscsi_reject {
 #define ISCSI_REASON_BOOKMARK_INVALID	9
 #define ISCSI_REASON_BOOKMARK_NO_RESOURCES	10
 #define ISCSI_REASON_NEGOTIATION_RESET	11
+#define ISCSI_REASON_WAITING_FOR_LOGOUT	12
 
 /* Max. number of Key=Value pairs in a text message */
 #define MAX_KEY_VALUE_PAIRS	8192
Signature · 1 line
2.20.1
author maintainer
#102 lore raw

PATCH 99/134 f2fs: Fix deadlock in f2fs_gc() context during atomic files handling

In reply to Sasha Levin

From: Sahitya Tummala <[email protected]>

[ Upstream commit 677017d196ba2a4cfff13626b951cc9a206b8c7c ]

The FS got stuck in the below stack when the storage is almost full/dirty condition (when FG_GC is being done).

schedule_timeout
io_schedule_timeout
congestion_wait
f2fs_drop_inmem_pages_all
f2fs_gc
f2fs_balance_fs
__write_node_page
f2fs_fsync_node_pages
f2fs_do_sync_file
f2fs_ioctl

The root cause for this issue is there is a potential infinite loop in f2fs_drop_inmem_pages_all() for the case where gc_failure is true and when there an inode whose i_gc_failures[GC_FAILURE_ATOMIC] is not set. Fix this by keeping track of the total atomic files currently opened and using that to exit from this condition.

---
 fs/f2fs/f2fs.h    |  1 +
 fs/f2fs/file.c    |  1 +
 fs/f2fs/segment.c | 21 +++++++++++++++------
 3 files changed, 17 insertions(+), 6 deletions(-)
fs/f2fs/f2fs.h +1 -0
index f078cd20dab88..9046432b87c2d 100644
@@ -1289,6 +1289,7 @@ struct f2fs_sb_info {
 	unsigned int gc_mode;			/* current GC state */
 	unsigned int next_victim_seg[2];	/* next segment in victim section */
 	/* for skip statistic */
+	unsigned int atomic_files;              /* # of opened atomic file */
 	unsigned long long skipped_atomic_files[2];	/* FG_GC and BG_GC */
 	unsigned long long skipped_gc_rwsem;		/* FG_GC only */
 
fs/f2fs/file.c +1 -0
index 29bc0a542759a..8ed8e4328bd1a 100644
@@ -1890,6 +1890,7 @@ static int f2fs_ioc_start_atomic_write(struct file *filp)
 	spin_lock(&sbi->inode_lock[ATOMIC_FILE]);
 	if (list_empty(&fi->inmem_ilist))
 		list_add_tail(&fi->inmem_ilist, &sbi->inode_list[ATOMIC_FILE]);
+	sbi->atomic_files++;
 	spin_unlock(&sbi->inode_lock[ATOMIC_FILE]);
 
 	/* add inode in inmem_list first and set atomic_file */
fs/f2fs/segment.c +15 -6
index 8087095814819..7d85784012678 100644
@@ -288,6 +288,8 @@ void f2fs_drop_inmem_pages_all(struct f2fs_sb_info *sbi, bool gc_failure)
 	struct list_head *head = &sbi->inode_list[ATOMIC_FILE];
 	struct inode *inode;
 	struct f2fs_inode_info *fi;
+	unsigned int count = sbi->atomic_files;
+	unsigned int looped = 0;
 next:
 	spin_lock(&sbi->inode_lock[ATOMIC_FILE]);
 	if (list_empty(head)) {
@@ -296,22 +298,26 @@ next:
 	}
 	fi = list_first_entry(head, struct f2fs_inode_info, inmem_ilist);
 	inode = igrab(&fi->vfs_inode);
+	if (inode)
+		list_move_tail(&fi->inmem_ilist, head);
 	spin_unlock(&sbi->inode_lock[ATOMIC_FILE]);
 
 	if (inode) {
 		if (gc_failure) {
-			if (fi->i_gc_failures[GC_FAILURE_ATOMIC])
-				goto drop;
-			goto skip;
+			if (!fi->i_gc_failures[GC_FAILURE_ATOMIC])
+				goto skip;
 		}
-drop:
 		set_inode_flag(inode, FI_ATOMIC_REVOKE_REQUEST);
 		f2fs_drop_inmem_pages(inode);
+skip:
 		iput(inode);
 	}
-skip:
 	congestion_wait(BLK_RW_ASYNC, HZ/50);
 	cond_resched();
+	if (gc_failure) {
+		if (++looped >= count)
+			return;
+	}
 	goto next;
 }
 
@@ -327,13 +333,16 @@ void f2fs_drop_inmem_pages(struct inode *inode)
 		mutex_unlock(&fi->inmem_lock);
 	}
 
-	clear_inode_flag(inode, FI_ATOMIC_FILE);
 	fi->i_gc_failures[GC_FAILURE_ATOMIC] = 0;
 	stat_dec_atomic_write(inode);
 
 	spin_lock(&sbi->inode_lock[ATOMIC_FILE]);
 	if (!list_empty(&fi->inmem_ilist))
 		list_del_init(&fi->inmem_ilist);
+	if (f2fs_is_atomic_file(inode)) {
+		clear_inode_flag(inode, FI_ATOMIC_FILE);
+		sbi->atomic_files--;
+	}
 	spin_unlock(&sbi->inode_lock[ATOMIC_FILE]);
 }
 
Signature · 8 lines
2.20.1



_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
author maintainer
#103 lore raw

PATCH 100/134 habanalabs: skip VA block list update in reset flow

In reply to Sasha Levin

From: Omer Shpigelman <[email protected]>

[ Upstream commit 71c5e55e7c077fa17c42fbda91a8d14322825c44 ]

Reduce context close time by skipping the VA block free list update in order to avoid hard reset with open contexts.
Reset with open contexts can potentially lead to a kernel crash as the generic pool of the MMU hops is destroyed while it is not empty because some unmap operations are not done.
The commit affect mainly when running on simulator.

---
 drivers/misc/habanalabs/memory.c | 30 ++++++++++++++++++++----------
 1 file changed, 20 insertions(+), 10 deletions(-)
drivers/misc/habanalabs/memory.c +20 -10
index 365fb0cb8dfff..22566b75ca50c 100644
@@ -965,17 +965,19 @@ init_page_pack_err:
  *
  * @ctx                 : current context
  * @vaddr               : device virtual address to unmap
+ * @ctx_free            : true if in context free flow, false otherwise.
  *
  * This function does the following:
  * - Unmap the physical pages related to the given virtual address
  * - return the device virtual block to the virtual block list
  */
-static int unmap_device_va(struct hl_ctx *ctx, u64 vaddr)
+static int unmap_device_va(struct hl_ctx *ctx, u64 vaddr, bool ctx_free)
 {
 	struct hl_device *hdev = ctx->hdev;
 	struct hl_vm_phys_pg_pack *phys_pg_pack = NULL;
 	struct hl_vm_hash_node *hnode = NULL;
 	struct hl_userptr *userptr = NULL;
+	struct hl_va_range *va_range;
 	enum vm_type_t *vm_type;
 	u64 next_vaddr, i;
 	u32 page_size;
@@ -1003,6 +1005,7 @@ static int unmap_device_va(struct hl_ctx *ctx, u64 vaddr)
 
 	if (*vm_type == VM_TYPE_USERPTR) {
 		is_userptr = true;
+		va_range = &ctx->host_va_range;
 		userptr = hnode->ptr;
 		rc = init_phys_pg_pack_from_userptr(ctx, userptr,
 				&phys_pg_pack);
@@ -1014,6 +1017,7 @@ static int unmap_device_va(struct hl_ctx *ctx, u64 vaddr)
 		}
 	} else if (*vm_type == VM_TYPE_PHYS_PACK) {
 		is_userptr = false;
+		va_range = &ctx->dram_va_range;
 		phys_pg_pack = hnode->ptr;
 	} else {
 		dev_warn(hdev->dev,
@@ -1052,12 +1056,18 @@ static int unmap_device_va(struct hl_ctx *ctx, u64 vaddr)
 
 	mutex_unlock(&ctx->mmu_lock);
 
-	if (add_va_block(hdev,
-			is_userptr ? &ctx->host_va_range : &ctx->dram_va_range,
-			vaddr,
-			vaddr + phys_pg_pack->total_size - 1))
-		dev_warn(hdev->dev, "add va block failed for vaddr: 0x%llx\n",
-				vaddr);
+	/*
+	 * No point in maintaining the free VA block list if the context is
+	 * closing as the list will be freed anyway
+	 */
+	if (!ctx_free) {
+		rc = add_va_block(hdev, va_range, vaddr,
+					vaddr + phys_pg_pack->total_size - 1);
+		if (rc)
+			dev_warn(hdev->dev,
+					"add va block failed for vaddr: 0x%llx\n",
+					vaddr);
+	}
 
 	atomic_dec(&phys_pg_pack->mapping_cnt);
 	kfree(hnode);
@@ -1189,8 +1199,8 @@ int hl_mem_ioctl(struct hl_fpriv *hpriv, void *data)
 		break;
 
 	case HL_MEM_OP_UNMAP:
-		rc = unmap_device_va(ctx,
-				args->in.unmap.device_virt_addr);
+		rc = unmap_device_va(ctx, args->in.unmap.device_virt_addr,
+					false);
 		break;
 
 	default:
@@ -1620,7 +1630,7 @@ void hl_vm_ctx_fini(struct hl_ctx *ctx)
 		dev_dbg(hdev->dev,
 			"hl_mem_hash_node of vaddr 0x%llx of asid %d is still alive\n",
 			hnode->vaddr, ctx->asid);
-		unmap_device_va(ctx, hnode->vaddr);
+		unmap_device_va(ctx, hnode->vaddr, true);
 	}
 
 	spin_lock(&vm->idr_lock);
Signature · 1 line
2.20.1
author maintainer
#104 lore raw

PATCH 101/134 gpio/mpc8xxx: fix qoriq GPIO reading

In reply to Sasha Levin

From: Russell King <[email protected]>

[ Upstream commit 787b64a43f7acacf8099329ea08872e663f1e74f ]

Qoriq requires the IBE register to be set to enable GPIO inputs to be read. Set it.

---
 drivers/gpio/gpio-mpc8xxx.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/gpio/gpio-mpc8xxx.c +3 -0
index 16a47de29c94c..b863421ae7309 100644
@@ -386,6 +386,9 @@ static int mpc8xxx_probe(struct platform_device *pdev)
 
 	gc->to_irq = mpc8xxx_gpio_to_irq;
 
+	if (of_device_is_compatible(np, "fsl,qoriq-gpio"))
+		gc->write_reg(mpc8xxx_gc->regs + GPIO_IBE, 0xffffffff);
+
 	ret = gpiochip_add_data(gc, mpc8xxx_gc);
 	if (ret) {
 		pr_err("%pOF: GPIO chip registration failed with status %d\n",
Signature · 1 line
2.20.1
author maintainer
#105 lore raw

PATCH 102/134 platform/x86: intel_pmc_core: Fix the SoC naming inconsistency

In reply to Sasha Levin

From: Gayatri Kammela <[email protected]>

[ Upstream commit 43e82d8aa92503d264309fb648b251b2d85caf1a ]

Intel's SoCs follow a naming convention which spells out the SoC name as two words instead of one word (E.g: Cannon Lake vs Cannonlake). Thus fix the naming inconsistency across the intel_pmc_core driver, so future SoCs can follow the naming consistency as below.

Cometlake -> Comet Lake
Tigerlake -> Tiger Lake
Elkhartlake -> Elkhart Lake

Cc Mario Limonciello <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Srinivas Pandruvada <[email protected]>
Cc Andy Shevchenko <[email protected]>
Cc Kan Liang <[email protected]>
Cc David E. Box <[email protected]>
Cc Rajneesh Bhardwaj <[email protected]>
Cc Tony Luck <[email protected]>
---
 drivers/platform/x86/intel_pmc_core.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
drivers/platform/x86/intel_pmc_core.c +7 -7
index 94a008efb09b8..6b6edc30f8359 100644
@@ -158,7 +158,7 @@ static const struct pmc_reg_map spt_reg_map = {
 	.pm_vric1_offset = SPT_PMC_VRIC1_OFFSET,
 };
 
-/* Cannonlake: PGD PFET Enable Ack Status Register(s) bitmap */
+/* Cannon Lake: PGD PFET Enable Ack Status Register(s) bitmap */
 static const struct pmc_bit_map cnp_pfear_map[] = {
 	{"PMC",                 BIT(0)},
 	{"OPI-DMI",             BIT(1)},
@@ -185,7 +185,7 @@ static const struct pmc_bit_map cnp_pfear_map[] = {
 	{"SDX",                 BIT(4)},
 	{"SPE",                 BIT(5)},
 	{"Fuse",                BIT(6)},
-	/* Reserved for Cannonlake but valid for Icelake */
+	/* Reserved for Cannon Lake but valid for Ice Lake */
 	{"SBR8",		BIT(7)},
 
 	{"CSME_FSC",            BIT(0)},
@@ -229,12 +229,12 @@ static const struct pmc_bit_map cnp_pfear_map[] = {
 	{"HDA_PGD4",            BIT(2)},
 	{"HDA_PGD5",            BIT(3)},
 	{"HDA_PGD6",            BIT(4)},
-	/* Reserved for Cannonlake but valid for Icelake */
+	/* Reserved for Cannon Lake but valid for Ice Lake */
 	{"PSF6",		BIT(5)},
 	{"PSF7",		BIT(6)},
 	{"PSF8",		BIT(7)},
 
-	/* Icelake generation onwards only */
+	/* Ice Lake generation onwards only */
 	{"RES_65",		BIT(0)},
 	{"RES_66",		BIT(1)},
 	{"RES_67",		BIT(2)},
@@ -324,7 +324,7 @@ static const struct pmc_bit_map cnp_ltr_show_map[] = {
 	{"ISH",			CNP_PMC_LTR_ISH},
 	{"UFSX2",		CNP_PMC_LTR_UFSX2},
 	{"EMMC",		CNP_PMC_LTR_EMMC},
-	/* Reserved for Cannonlake but valid for Icelake */
+	/* Reserved for Cannon Lake but valid for Ice Lake */
 	{"WIGIG",		ICL_PMC_LTR_WIGIG},
 	/* Below two cannot be used for LTR_IGNORE */
 	{"CURRENT_PLATFORM",	CNP_PMC_LTR_CUR_PLT},
@@ -871,8 +871,8 @@ static int pmc_core_probe(struct platform_device *pdev)
 	pmcdev->map = (struct pmc_reg_map *)cpu_id->driver_data;
 
 	/*
-	 * Coffeelake has CPU ID of Kabylake and Cannonlake PCH. So here
-	 * Sunrisepoint PCH regmap can't be used. Use Cannonlake PCH regmap
+	 * Coffee Lake has CPU ID of Kaby Lake and Cannon Lake PCH. So here
+	 * Sunrisepoint PCH regmap can't be used. Use Cannon Lake PCH regmap
 	 * in this case.
 	 */
 	if (pmcdev->map == &spt_reg_map && !pci_dev_present(pmc_pci_ids))
Signature · 1 line
2.20.1
author maintainer
#106 lore raw

PATCH 103/134 platform/x86: intel_pmc_core: Add Comet Lake (CML) platform support to intel_pmc_core driver

In reply to Sasha Levin

From: Gayatri Kammela <[email protected]>

[ Upstream commit 5406327d43edd9a171bd260f49c752d148727eaf ]

Add Comet Lake to the list of the platforms that intel_pmc_core driver supports for pmc_core device.

Just like Ice Lake, Comet Lake can also reuse all the Cannon Lake PCH IPs. No additional effort is needed to enable but to simply reuse them.

Cc Mario Limonciello <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Srinivas Pandruvada <[email protected]>
Cc Andy Shevchenko <[email protected]>
Cc Kan Liang <[email protected]>
Cc David E. Box <[email protected]>
Cc Rajneesh Bhardwaj <[email protected]>
Cc Tony Luck <[email protected]>
---
 drivers/platform/x86/intel_pmc_core.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/platform/x86/intel_pmc_core.c +5 -2
index 6b6edc30f8359..571b4754477c5 100644
@@ -160,6 +160,7 @@ static const struct pmc_reg_map spt_reg_map = {
 
 /* Cannon Lake: PGD PFET Enable Ack Status Register(s) bitmap */
 static const struct pmc_bit_map cnp_pfear_map[] = {
+	/* Reserved for Cannon Lake but valid for Comet Lake */
 	{"PMC",                 BIT(0)},
 	{"OPI-DMI",             BIT(1)},
 	{"SPI/eSPI",            BIT(2)},
@@ -185,7 +186,7 @@ static const struct pmc_bit_map cnp_pfear_map[] = {
 	{"SDX",                 BIT(4)},
 	{"SPE",                 BIT(5)},
 	{"Fuse",                BIT(6)},
-	/* Reserved for Cannon Lake but valid for Ice Lake */
+	/* Reserved for Cannon Lake but valid for Ice Lake and Comet Lake */
 	{"SBR8",		BIT(7)},
 
 	{"CSME_FSC",            BIT(0)},
@@ -229,7 +230,7 @@ static const struct pmc_bit_map cnp_pfear_map[] = {
 	{"HDA_PGD4",            BIT(2)},
 	{"HDA_PGD5",            BIT(3)},
 	{"HDA_PGD6",            BIT(4)},
-	/* Reserved for Cannon Lake but valid for Ice Lake */
+	/* Reserved for Cannon Lake but valid for Ice Lake and Comet Lake */
 	{"PSF6",		BIT(5)},
 	{"PSF7",		BIT(6)},
 	{"PSF8",		BIT(7)},
@@ -813,6 +814,8 @@ static const struct x86_cpu_id intel_pmc_core_ids[] = {
 	INTEL_CPU_FAM6(CANNONLAKE_L, cnp_reg_map),
 	INTEL_CPU_FAM6(ICELAKE_L, icl_reg_map),
 	INTEL_CPU_FAM6(ICELAKE_NNPI, icl_reg_map),
+	INTEL_CPU_FAM6(COMETLAKE, cnp_reg_map),
+	INTEL_CPU_FAM6(COMETLAKE_L, cnp_reg_map),
 	{}
 };
 
Signature · 1 line
2.20.1
author maintainer
#107 lore raw

PATCH 104/134 gpio: mpc8xxx: Don't overwrite default irq_set_type callback

In reply to Sasha Levin

From: Vladimir Oltean <[email protected]>

[ Upstream commit 4e50573f39229d5e9c985fa3b4923a8b29619ade ]

The per-SoC devtype structures can contain their own callbacks that overwrite mpc8xxx_gpio_devtype_default.

The clear intention is that mpc8xxx_irq_set_type is used in case the SoC does not specify a more specific callback. But what happens is that if the SoC doesn't specify one, its .irq_set_type is de-facto NULL, and this overwrites mpc8xxx_irq_set_type to a no-op. This means that the following SoCs are affected:

- fsl,mpc8572-gpio
- fsl,ls1028a-gpio
- fsl,ls1088a-gpio

On these boards, the irq_set_type does exactly nothing, and the GPIO controller keeps its GPICR register in the hardware-default state. On the LS1028A, that is ACTIVE_BOTH, which means 2 interrupts are raised even if the IRQ client requests LEVEL_HIGH. Another implication is that the IRQs are not checked (e.g. level-triggered interrupts are not rejected, although they are not supported).

Fixes 82e39b0d8566 ("gpio: mpc8xxx: handle differences between incarnations at a single place")
---
 drivers/gpio/gpio-mpc8xxx.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpio/gpio-mpc8xxx.c +2 -1
index b863421ae7309..a031cbcdf6ef9 100644
@@ -377,7 +377,8 @@ static int mpc8xxx_probe(struct platform_device *pdev)
 	 * It's assumed that only a single type of gpio controller is available
 	 * on the current machine, so overwriting global data is fine.
 	 */
-	mpc8xxx_irq_chip.irq_set_type = devtype->irq_set_type;
+	if (devtype->irq_set_type)
+		mpc8xxx_irq_chip.irq_set_type = devtype->irq_set_type;
 
 	if (devtype->gpio_dir_out)
 		gc->direction_output = devtype->gpio_dir_out;
Signature · 1 line
2.20.1
author maintainer
#108 lore raw

PATCH 105/134 gpio: lynxpoint: Setup correct IRQ handlers

In reply to Sasha Levin

From: Andy Shevchenko <[email protected]>

[ Upstream commit e272f7ec070d212b9301d5a465bc8952f8dcf908 ]

When commit 75e99bf5ed8f ("gpio: lynxpoint: set default handler to be handle_bad_irq()") switched default handler to be handle_bad_irq() the lp_irq_type() function remained untouched. It means that even request_irq() can't change the handler and we are not able to handle IRQs properly anymore. Fix it by setting correct handlers in the lp_irq_type() callback.

Fixes 75e99bf5ed8f ("gpio: lynxpoint: set default handler to be handle_bad_irq()")
---
 drivers/gpio/gpio-lynxpoint.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/gpio/gpio-lynxpoint.c +6 -0
index e9e47c0d5be75..490ce7bae25ec 100644
@@ -164,6 +164,12 @@ static int lp_irq_type(struct irq_data *d, unsigned type)
 		value |= TRIG_SEL_BIT | INT_INV_BIT;
 
 	outl(value, reg);
+
+	if (type & IRQ_TYPE_EDGE_BOTH)
+		irq_set_handler_locked(d, handle_edge_irq);
+	else if (type & IRQ_TYPE_LEVEL_MASK)
+		irq_set_handler_locked(d, handle_level_irq);
+
 	spin_unlock_irqrestore(&lg->lock, flags);
 
 	return 0;
Signature · 1 line
2.20.1
author maintainer
#109 lore raw

PATCH 106/134 tools/power/x86/intel-speed-select: Ignore missing config level

In reply to Sasha Levin

From: Srinivas Pandruvada <[email protected]>

[ Upstream commit 20183ccd3e4d01d23b0a01fe9f3ee73fbae312fa ]

It is possible that certain config levels are not available, even if the max level includes the level. There can be missing levels in some platforms. So ignore the level when called for information dump for all levels and fail if specifically ask for the missing level.

Here the changes is to continue reading information about other levels even if we fail to get information for the current level. But use the "processed" flag to indicate the failure. When the "processed" flag is not set, don't dump information about that level.

---
 tools/power/x86/intel-speed-select/isst-core.c    | 8 ++++----
 tools/power/x86/intel-speed-select/isst-display.c | 3 ++-
 2 files changed, 6 insertions(+), 5 deletions(-)
tools/power/x86/intel-speed-select/isst-core.c +4 -4
index 6dee5332c9d37..fde3f9cefc6db 100644
@@ -553,7 +553,6 @@ int isst_get_process_ctdp(int cpu, int tdp_level, struct isst_pkg_ctdp *pkg_dev)
 			     i);
 		ctdp_level = &pkg_dev->ctdp_level[i];
 
-		ctdp_level->processed = 1;
 		ctdp_level->level = i;
 		ctdp_level->control_cpu = cpu;
 		ctdp_level->pkg_id = get_physical_package_id(cpu);
@@ -561,7 +560,10 @@ int isst_get_process_ctdp(int cpu, int tdp_level, struct isst_pkg_ctdp *pkg_dev)
 
 		ret = isst_get_ctdp_control(cpu, i, ctdp_level);
 		if (ret)
-			return ret;
+			continue;
+
+		pkg_dev->processed = 1;
+		ctdp_level->processed = 1;
 
 		ret = isst_get_tdp_info(cpu, i, ctdp_level);
 		if (ret)
@@ -614,8 +616,6 @@ int isst_get_process_ctdp(int cpu, int tdp_level, struct isst_pkg_ctdp *pkg_dev)
 		}
 	}
 
-	pkg_dev->processed = 1;
-
 	return 0;
 }
 
tools/power/x86/intel-speed-select/isst-display.c +2 -1
index 40346d534f789..b11575c3e8864 100644
@@ -314,7 +314,8 @@ void isst_ctdp_display_information(int cpu, FILE *outf, int tdp_level,
 	char value[256];
 	int i, base_level = 1;
 
-	print_package_info(cpu, outf);
+	if (pkg_dev->processed)
+		print_package_info(cpu, outf);
 
 	for (i = 0; i <= pkg_dev->levels; ++i) {
 		struct isst_pkg_ctdp_level_info *ctdp_level;
Signature · 1 line
2.20.1
author maintainer
#110 lore raw

PATCH 107/134 Drivers: hv: vmbus: Fix crash handler reset of Hyper-V synic

In reply to Sasha Levin

From: Michael Kelley <[email protected]>

[ Upstream commit 7a1323b5dfe44a9013a2cc56ef2973034a00bf88 ]

The crash handler calls hv_synic_cleanup() to shutdown the Hyper-V synthetic interrupt controller. But if the CPU that calls hv_synic_cleanup() has a VMbus channel interrupt assigned to it (which is likely the case in smaller VM sizes), hv_synic_cleanup() returns an error and the synthetic interrupt controller isn't shutdown. While the lack of being shutdown hasn't caused a known problem, it still should be fixed for highest reliability.

So directly call hv_synic_disable_regs() instead of hv_synic_cleanup(), which ensures that the synic is always shutdown.

---
 drivers/hv/vmbus_drv.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hv/vmbus_drv.c +1 -1
index 53a60c81e220d..05ead1735c6e3 100644
@@ -2308,7 +2308,7 @@ static void hv_crash_handler(struct pt_regs *regs)
 	vmbus_connection.conn_state = DISCONNECTED;
 	cpu = smp_processor_id();
 	hv_stimer_cleanup(cpu);
-	hv_synic_cleanup(cpu);
+	hv_synic_disable_regs(cpu);
 	hyperv_cleanup();
 };
 
Signature · 1 line
2.20.1
author maintainer
#111 lore raw

PATCH 108/134 apparmor: fix unsigned len comparison with less than zero

In reply to Sasha Levin

From: Colin Ian King <[email protected]>

[ Upstream commit 00e0590dbaec6f1bcaa36a85467d7e3497ced522 ]

The sanity check in macro update_for_len checks to see if len is less than zero, however, len is a size_t so it can never be less than zero, so this sanity check is a no-op. Fix this by making len a ssize_t so the comparison will work and add ulen that is a size_t copy of len so that the min() macro won't throw warnings about comparing different types.

Addresses-Coverity: ("Macro compares unsigned to 0")

Fixes f1bd904175e8 ("apparmor: add the base fns() for domain labels")
---
 security/apparmor/label.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
security/apparmor/label.c +7 -5
index 59f1cc2557a73..470693239e64f 100644
@@ -1458,11 +1458,13 @@ static inline bool use_label_hname(struct aa_ns *ns, struct aa_label *label,
 /* helper macro for snprint routines */
 #define update_for_len(total, len, size, str)	\
 do {					\
+	size_t ulen = len;		\
+					\
 	AA_BUG(len < 0);		\
-	total += len;			\
-	len = min(len, size);		\
-	size -= len;			\
-	str += len;			\
+	total += ulen;			\
+	ulen = min(ulen, size);		\
+	size -= ulen;			\
+	str += ulen;			\
 } while (0)
 
 /**
@@ -1597,7 +1599,7 @@ int aa_label_snxprint(char *str, size_t size, struct aa_ns *ns,
 	struct aa_ns *prev_ns = NULL;
 	struct label_it i;
 	int count = 0, total = 0;
-	size_t len;
+	ssize_t len;
 
 	AA_BUG(!str && size != 0);
 	AA_BUG(!label);
Signature · 1 line
2.20.1
author maintainer
#112 lore raw

PATCH 109/134 drm/amdgpu: Call find_vma under mmap_sem

In reply to Sasha Levin

From: Jason Gunthorpe <[email protected]>

[ Upstream commit a9ae8731e6e52829a935d81a65d7f925cb95dbac ]

find_vma() must be called under the mmap_sem, reorganize this code to do the vma check after entering the lock.

Further, fix the unlocked use of struct task_struct's mm, instead use the mm from hmm_mirror which has an active mm_grab. Also the mm_grab must be converted to a mm_get before acquiring mmap_sem or calling find_vma().

Fixes 66c45500bfdc ("drm/amdgpu: use new HMM APIs and helpers")
Fixes 0919195f2b0d ("drm/amdgpu: Enable amdgpu_ttm_tt_get_user_pages in worker threads")
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 37 ++++++++++++++-----------
 1 file changed, 21 insertions(+), 16 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +21 -16
index dff41d0a85fe9..c0e41f1f0c236 100644
@@ -35,6 +35,7 @@
 #include <linux/hmm.h>
 #include <linux/pagemap.h>
 #include <linux/sched/task.h>
+#include <linux/sched/mm.h>
 #include <linux/seq_file.h>
 #include <linux/slab.h>
 #include <linux/swap.h>
@@ -788,7 +789,7 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages)
 	struct hmm_mirror *mirror = bo->mn ? &bo->mn->mirror : NULL;
 	struct ttm_tt *ttm = bo->tbo.ttm;
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
-	struct mm_struct *mm = gtt->usertask->mm;
+	struct mm_struct *mm;
 	unsigned long start = gtt->userptr;
 	struct vm_area_struct *vma;
 	struct hmm_range *range;
@@ -796,25 +797,14 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages)
 	uint64_t *pfns;
 	int r = 0;
 
-	if (!mm) /* Happens during process shutdown */
-		return -ESRCH;
-
 	if (unlikely(!mirror)) {
 		DRM_DEBUG_DRIVER("Failed to get hmm_mirror\n");
-		r = -EFAULT;
-		goto out;
+		return -EFAULT;
 	}
 
-	vma = find_vma(mm, start);
-	if (unlikely(!vma || start < vma->vm_start)) {
-		r = -EFAULT;
-		goto out;
-	}
-	if (unlikely((gtt->userflags & AMDGPU_GEM_USERPTR_ANONONLY) &&
-		vma->vm_file)) {
-		r = -EPERM;
-		goto out;
-	}
+	mm = mirror->hmm->mmu_notifier.mm;
+	if (!mmget_not_zero(mm)) /* Happens during process shutdown */
+		return -ESRCH;
 
 	range = kzalloc(sizeof(*range), GFP_KERNEL);
 	if (unlikely(!range)) {
@@ -847,6 +837,17 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages)
 	hmm_range_wait_until_valid(range, HMM_RANGE_DEFAULT_TIMEOUT);
 
 	down_read(&mm->mmap_sem);
+	vma = find_vma(mm, start);
+	if (unlikely(!vma || start < vma->vm_start)) {
+		r = -EFAULT;
+		goto out_unlock;
+	}
+	if (unlikely((gtt->userflags & AMDGPU_GEM_USERPTR_ANONONLY) &&
+		vma->vm_file)) {
+		r = -EPERM;
+		goto out_unlock;
+	}
+
 	r = hmm_range_fault(range, 0);
 	up_read(&mm->mmap_sem);
 
@@ -865,15 +866,19 @@ int amdgpu_ttm_tt_get_user_pages(struct amdgpu_bo *bo, struct page **pages)
 	}
 
 	gtt->range = range;
+	mmput(mm);
 
 	return 0;
 
+out_unlock:
+	up_read(&mm->mmap_sem);
 out_free_pfns:
 	hmm_range_unregister(range);
 	kvfree(pfns);
 out_free_ranges:
 	kfree(range);
 out:
+	mmput(mm);
 	return r;
 }
 
Signature · 6 lines
2.20.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
author maintainer
#113 lore raw

PATCH 110/134 scripts/kallsyms: fix definitely-lost memory leak

In reply to Sasha Levin

From: Masahiro Yamada <[email protected]>

[ Upstream commit 21915eca088dc271c970e8351290e83d938114ac ]

build_initial_tok_table() overwrites unused sym_entry to shrink the table size. Before the entry is overwritten, table[i].sym must be freed since it is malloc'ed data.

This fixes the 'definitely lost' report from valgrind. I ran valgrind against x86_64_defconfig of v5.4-rc8 kernel, and here is the summary:

[Before the fix]

LEAK SUMMARY:

     definitely lost: 53,184 bytes in 2,874 blocks

[After the fix]

LEAK SUMMARY:
definitely lost: 0 bytes in 0 blocks

---
 scripts/kallsyms.c | 2 ++
 1 file changed, 2 insertions(+)
scripts/kallsyms.c +2 -0
index ae6504d07fd61..fb15f09e0e38c 100644
@@ -489,6 +489,8 @@ static void build_initial_tok_table(void)
 				table[pos] = table[i];
 			learn_symbol(table[pos].sym, table[pos].len);
 			pos++;
+		} else {
+			free(table[i].sym);
 		}
 	}
 	table_cnt = pos;
Signature · 1 line
2.20.1
author maintainer
#114 lore raw

PATCH 111/134 powerpc: Don't add -mabi= flags when building with Clang

In reply to Sasha Levin

From: Nathan Chancellor <[email protected]>

[ Upstream commit 465bfd9c44dea6b55962b5788a23ac87a467c923 ]

When building pseries_defconfig, building vdso32 errors out:

error: unknown target ABI 'elfv1'

This happens because -m32 in clang changes the target to 32-bit, which does not allow the ABI to be changed.

Commit 4dc831aa8813 ("powerpc: Fix compiling a BE kernel with a powerpc64le toolchain") added these flags to fix building big endian kernels with a little endian GCC.

Clang doesn't need -mabi because the target triple controls the default value. -mlittle-endian and -mbig-endian manipulate the triple into either powerpc64-* or powerpc64le-*, which properly sets the default ABI.

Adding a debug print out in the PPC64TargetInfo constructor after line 383 above shows this:

  $ echo | ./clang -E --target=powerpc64-linux -mbig-endian -o /dev/null -
  Default ABI: elfv1
  $ echo | ./clang -E --target=powerpc64-linux -mlittle-endian -o /dev/null -
  Default ABI: elfv2
  $ echo | ./clang -E --target=powerpc64le-linux -mbig-endian -o /dev/null -
  Default ABI: elfv1
  $ echo | ./clang -E --target=powerpc64le-linux -mlittle-endian -o /dev/null -
  Default ABI: elfv2

Don't specify -mabi when building with clang to avoid the build error with -m32 and not change any code generation.

-mcall-aixdesc is not an implemented flag in clang so it can be safely excluded as well, see commit 238abecde8ad ("powerpc: Don't use gcc specific options on clang").

pseries_defconfig successfully builds after this patch and powernv_defconfig and ppc44x_defconfig don't regress.

---
 arch/powerpc/Makefile | 4 ++++
 1 file changed, 4 insertions(+)
arch/powerpc/Makefile +4 -0
index 83522c9fc7b66..37ac731a556b8 100644
@@ -91,11 +91,13 @@ MULTIPLEWORD	:= -mmultiple
 endif
 
 ifdef CONFIG_PPC64
+ifndef CONFIG_CC_IS_CLANG
 cflags-$(CONFIG_CPU_BIG_ENDIAN)		+= $(call cc-option,-mabi=elfv1)
 cflags-$(CONFIG_CPU_BIG_ENDIAN)		+= $(call cc-option,-mcall-aixdesc)
 aflags-$(CONFIG_CPU_BIG_ENDIAN)		+= $(call cc-option,-mabi=elfv1)
 aflags-$(CONFIG_CPU_LITTLE_ENDIAN)	+= -mabi=elfv2
 endif
+endif
 
 ifndef CONFIG_CC_IS_CLANG
   cflags-$(CONFIG_CPU_LITTLE_ENDIAN)	+= -mno-strict-align
@@ -141,6 +143,7 @@ endif
 endif
 
 CFLAGS-$(CONFIG_PPC64)	:= $(call cc-option,-mtraceback=no)
+ifndef CONFIG_CC_IS_CLANG
 ifdef CONFIG_CPU_LITTLE_ENDIAN
 CFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mabi=elfv2,$(call cc-option,-mcall-aixdesc))
 AFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mabi=elfv2)
@@ -149,6 +152,7 @@ CFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mabi=elfv1)
 CFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mcall-aixdesc)
 AFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mabi=elfv1)
 endif
+endif
 CFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mcmodel=medium,$(call cc-option,-mminimal-toc))
 CFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mno-pointers-to-nested-functions)
 
Signature · 1 line
2.20.1
author maintainer
#115 lore raw

PATCH 112/134 f2fs: choose hardlimit when softlimit is larger than hardlimit in f2fs_statfs_project()

In reply to Sasha Levin

From: Chengguang Xu <[email protected]>

[ Upstream commit 909110c060f22e65756659ec6fa957ae75777e00 ]

Setting softlimit larger than hardlimit seems meaningless for disk quota but currently it is allowed. In this case, there may be a bit of comfusion for users when they run df comamnd to directory which has project quota.

For example, we set 20M softlimit and 10M hardlimit of block usage limit for project quota of test_dir(project id 123).

[root@hades f2fs]# repquota -P -a
*** Report for project quotas on device /dev/nvme0n1p8 Block grace time: 7days; Inode grace time: 7days Block limits File limits
Project used soft hard grace used soft hard grace
----------------------------------------------------------------------
0 -- 4 0 0 1 0 0
123 +- 10248 20480 10240 2 0 0

The result of df command as below:

[root@hades f2fs]# df -h /mnt/f2fs/test
Filesystem Size Used Avail Use% Mounted on
/dev/nvme0n1p8 20M 11M 10M 51% /mnt/f2fs

Even though it looks like there is another 10M free space to use, if we write new data to diretory test(inherit project id), the write will fail with errno(-EDQUOT).

After this patch, the df result looks like below.

[root@hades f2fs]# df -h /mnt/f2fs/test
Filesystem Size Used Avail Use% Mounted on
/dev/nvme0n1p8 10M 10M 0 100% /mnt/f2fs

---
 fs/f2fs/super.c | 20 ++++++++++++++------
 1 file changed, 14 insertions(+), 6 deletions(-)
fs/f2fs/super.c +14 -6
index 1443cee158633..a2af155567b80 100644
@@ -1213,9 +1213,13 @@ static int f2fs_statfs_project(struct super_block *sb,
 		return PTR_ERR(dquot);
 	spin_lock(&dquot->dq_dqb_lock);
 
-	limit = (dquot->dq_dqb.dqb_bsoftlimit ?
-		 dquot->dq_dqb.dqb_bsoftlimit :
-		 dquot->dq_dqb.dqb_bhardlimit) >> sb->s_blocksize_bits;
+	limit = 0;
+	if (dquot->dq_dqb.dqb_bsoftlimit)
+		limit = dquot->dq_dqb.dqb_bsoftlimit;
+	if (dquot->dq_dqb.dqb_bhardlimit &&
+			(!limit || dquot->dq_dqb.dqb_bhardlimit < limit))
+		limit = dquot->dq_dqb.dqb_bhardlimit;
+
 	if (limit && buf->f_blocks > limit) {
 		curblock = dquot->dq_dqb.dqb_curspace >> sb->s_blocksize_bits;
 		buf->f_blocks = limit;
@@ -1224,9 +1228,13 @@ static int f2fs_statfs_project(struct super_block *sb,
 			 (buf->f_blocks - curblock) : 0;
 	}
 
-	limit = dquot->dq_dqb.dqb_isoftlimit ?
-		dquot->dq_dqb.dqb_isoftlimit :
-		dquot->dq_dqb.dqb_ihardlimit;
+	limit = 0;
+	if (dquot->dq_dqb.dqb_isoftlimit)
+		limit = dquot->dq_dqb.dqb_isoftlimit;
+	if (dquot->dq_dqb.dqb_ihardlimit &&
+			(!limit || dquot->dq_dqb.dqb_ihardlimit < limit))
+		limit = dquot->dq_dqb.dqb_ihardlimit;
+
 	if (limit && buf->f_files > limit) {
 		buf->f_files = limit;
 		buf->f_ffree =
Signature · 8 lines
2.20.1



_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
author maintainer
#116 lore raw

PATCH 113/134 cifs: Fix use-after-free bug in cifs_reconnect()

In reply to Sasha Levin

From: "Paulo Alcantara (SUSE)" <[email protected]>

[ Upstream commit 8354d88efdab72b4da32fc4f032448fcef22dab4 ]

Ensure we grab an active reference in cifs superblock while doing failover to prevent automounts (DFS links) of expiring and then destroying the superblock pointer.

This patch fixes the following KASAN report:

[  464.301462] BUG: KASAN: use-after-free in
cifs_reconnect+0x6ab/0x1350
[  464.303052] Read of size 8 at addr ffff888155e580d0 by task

cifsd/1107

[  464.304682] CPU: 3 PID: 1107 Comm: cifsd Not tainted 5.4.0-rc4+ #13
[  464.305552] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009),
BIOS rel-1.12.1-0-ga5cab58-rebuilt.opensuse.org 04/01/2014
[  464.307146] Call Trace:
[  464.307875]  dump_stack+0x5b/0x90
[  464.308631]  print_address_description.constprop.0+0x16/0x200
[  464.309478]  ? cifs_reconnect+0x6ab/0x1350
[  464.310253]  ? cifs_reconnect+0x6ab/0x1350
[  464.311040]  __kasan_report.cold+0x1a/0x41
[  464.311811]  ? cifs_reconnect+0x6ab/0x1350
[  464.312563]  kasan_report+0xe/0x20
[  464.313300]  cifs_reconnect+0x6ab/0x1350
[  464.314062]  ? extract_hostname.part.0+0x90/0x90
[  464.314829]  ? printk+0xad/0xde
[  464.315525]  ? _raw_spin_lock+0x7c/0xd0
[  464.316252]  ? _raw_read_lock_irq+0x40/0x40
[  464.316961]  ? ___ratelimit+0xed/0x182
[  464.317655]  cifs_readv_from_socket+0x289/0x3b0
[  464.318386]  cifs_read_from_socket+0x98/0xd0
[  464.319078]  ? cifs_readv_from_socket+0x3b0/0x3b0
[  464.319782]  ? try_to_wake_up+0x43c/0xa90
[  464.320463]  ? cifs_small_buf_get+0x4b/0x60
[  464.321173]  ? allocate_buffers+0x98/0x1a0
[  464.321856]  cifs_demultiplex_thread+0x218/0x14a0
[  464.322558]  ? cifs_handle_standard+0x270/0x270
[  464.323237]  ? __switch_to_asm+0x40/0x70
[  464.323893]  ? __switch_to_asm+0x34/0x70
[  464.324554]  ? __switch_to_asm+0x40/0x70
[  464.325226]  ? __switch_to_asm+0x40/0x70
[  464.325863]  ? __switch_to_asm+0x34/0x70
[  464.326505]  ? __switch_to_asm+0x40/0x70
[  464.327161]  ? __switch_to_asm+0x34/0x70
[  464.327784]  ? finish_task_switch+0xa1/0x330
[  464.328414]  ? __switch_to+0x363/0x640
[  464.329044]  ? __schedule+0x575/0xaf0
[  464.329655]  ? _raw_spin_lock_irqsave+0x82/0xe0
[  464.330301]  kthread+0x1a3/0x1f0
[  464.330884]  ? cifs_handle_standard+0x270/0x270
[  464.331624]  ? kthread_create_on_node+0xd0/0xd0
[  464.332347]  ret_from_fork+0x35/0x40
[  464.333577] Allocated by task 1110:
[  464.334381]  save_stack+0x1b/0x80
[  464.335123]  __kasan_kmalloc.constprop.0+0xc2/0xd0
[  464.335848]  cifs_smb3_do_mount+0xd4/0xb00
[  464.336619]  legacy_get_tree+0x6b/0xa0
[  464.337235]  vfs_get_tree+0x41/0x110
[  464.337975]  fc_mount+0xa/0x40
[  464.338557]  vfs_kern_mount.part.0+0x6c/0x80
[  464.339227]  cifs_dfs_d_automount+0x336/0xd29
[  464.339846]  follow_managed+0x1b1/0x450
[  464.340449]  lookup_fast+0x231/0x4a0
[  464.341039]  path_openat+0x240/0x1fd0
[  464.341634]  do_filp_open+0x126/0x1c0
[  464.342277]  do_sys_open+0x1eb/0x2c0
[  464.342957]  do_syscall_64+0x5e/0x190
[  464.343555]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[  464.344772] Freed by task 0:
[  464.345347]  save_stack+0x1b/0x80
[  464.345966]  __kasan_slab_free+0x12c/0x170
[  464.346576]  kfree+0xa6/0x270
[  464.347211]  rcu_core+0x39c/0xc80
[  464.347800]  __do_softirq+0x10d/0x3da
[  464.348919] The buggy address belongs to the object at
ffff888155e58000
                which belongs to the cache kmalloc-256 of size 256
[  464.350222] The buggy address is located 208 bytes inside of
                256-byte region [ffff888155e58000, ffff888155e58100)
[  464.351575] The buggy address belongs to the page:
[  464.352333] page:ffffea0005579600 refcount:1 mapcount:0
mapping:ffff88815a803400 index:0x0 compound_mapcount: 0
[  464.353583] flags: 0x200000000010200(slab|head)
[  464.354209] raw: 0200000000010200 ffffea0005576200 0000000400000004
ffff88815a803400
[  464.355353] raw: 0000000000000000 0000000080100010 00000001ffffffff
0000000000000000
[  464.356458] page dumped because: kasan: bad access detected
[  464.367005] Memory state around the buggy address:
[  464.367787]  ffff888155e57f80: fc fc fc fc fc fc fc fc fc fc fc fc
fc fc fc fc
[  464.368877]  ffff888155e58000: fb fb fb fb fb fb fb fb fb fb fb fb
fb fb fb fb
[  464.369967] >ffff888155e58080: fb fb fb fb fb fb fb fb fb fb fb fb
fb fb fb fb
[  464.371111]                                                  ^
[  464.371775]  ffff888155e58100: fc fc fc fc fc fc fc fc fc fc fc fc
fc fc fc fc
[  464.372893]  ffff888155e58180: fc fc fc fc fc fc fc fc fc fc fc fc
fc fc fc fc
[  464.373983] ==================================================================
---
 fs/cifs/connect.c | 46 +++++++++++++++++++++++++++++++++++-----------
 1 file changed, 35 insertions(+), 11 deletions(-)
fs/cifs/connect.c +35 -11
index ccaa8bad336f9..a4ae4d944a3ab 100644
@@ -387,7 +387,7 @@ static inline int reconn_set_ipaddr(struct TCP_Server_Info *server)
 #ifdef CONFIG_CIFS_DFS_UPCALL
 struct super_cb_data {
 	struct TCP_Server_Info *server;
-	struct cifs_sb_info *cifs_sb;
+	struct super_block *sb;
 };
 
 /* These functions must be called with server->srv_mutex held */
@@ -398,25 +398,39 @@ static void super_cb(struct super_block *sb, void *arg)
 	struct cifs_sb_info *cifs_sb;
 	struct cifs_tcon *tcon;
 
-	if (d->cifs_sb)
+	if (d->sb)
 		return;
 
 	cifs_sb = CIFS_SB(sb);
 	tcon = cifs_sb_master_tcon(cifs_sb);
 	if (tcon->ses->server == d->server)
-		d->cifs_sb = cifs_sb;
+		d->sb = sb;
 }
 
-static inline struct cifs_sb_info *
-find_super_by_tcp(struct TCP_Server_Info *server)
+static struct super_block *get_tcp_super(struct TCP_Server_Info *server)
 {
 	struct super_cb_data d = {
 		.server = server,
-		.cifs_sb = NULL,
+		.sb = NULL,
 	};
 
 	iterate_supers_type(&cifs_fs_type, super_cb, &d);
-	return d.cifs_sb ? d.cifs_sb : ERR_PTR(-ENOENT);
+
+	if (unlikely(!d.sb))
+		return ERR_PTR(-ENOENT);
+	/*
+	 * Grab an active reference in order to prevent automounts (DFS links)
+	 * of expiring and then freeing up our cifs superblock pointer while
+	 * we're doing failover.
+	 */
+	cifs_sb_active(d.sb);
+	return d.sb;
+}
+
+static inline void put_tcp_super(struct super_block *sb)
+{
+	if (!IS_ERR_OR_NULL(sb))
+		cifs_sb_deactive(sb);
 }
 
 static void reconn_inval_dfs_target(struct TCP_Server_Info *server,
@@ -480,6 +494,7 @@ cifs_reconnect(struct TCP_Server_Info *server)
 	struct mid_q_entry *mid_entry;
 	struct list_head retry_list;
 #ifdef CONFIG_CIFS_DFS_UPCALL
+	struct super_block *sb = NULL;
 	struct cifs_sb_info *cifs_sb = NULL;
 	struct dfs_cache_tgt_list tgt_list = {0};
 	struct dfs_cache_tgt_iterator *tgt_it = NULL;
@@ -489,13 +504,15 @@ cifs_reconnect(struct TCP_Server_Info *server)
 	server->nr_targets = 1;
 #ifdef CONFIG_CIFS_DFS_UPCALL
 	spin_unlock(&GlobalMid_Lock);
-	cifs_sb = find_super_by_tcp(server);
-	if (IS_ERR(cifs_sb)) {
-		rc = PTR_ERR(cifs_sb);
+	sb = get_tcp_super(server);
+	if (IS_ERR(sb)) {
+		rc = PTR_ERR(sb);
 		cifs_dbg(FYI, "%s: will not do DFS failover: rc = %d\n",
 			 __func__, rc);
-		cifs_sb = NULL;
+		sb = NULL;
 	} else {
+		cifs_sb = CIFS_SB(sb);
+
 		rc = reconn_setup_dfs_targets(cifs_sb, &tgt_list, &tgt_it);
 		if (rc && (rc != -EOPNOTSUPP)) {
 			cifs_server_dbg(VFS, "%s: no target servers for DFS failover\n",
@@ -512,6 +529,10 @@ cifs_reconnect(struct TCP_Server_Info *server)
 		/* the demux thread will exit normally
 		next time through the loop */
 		spin_unlock(&GlobalMid_Lock);
+#ifdef CONFIG_CIFS_DFS_UPCALL
+		dfs_cache_free_tgts(&tgt_list);
+		put_tcp_super(sb);
+#endif
 		return rc;
 	} else
 		server->tcpStatus = CifsNeedReconnect;
@@ -638,7 +659,10 @@ cifs_reconnect(struct TCP_Server_Info *server)
 				 __func__, rc);
 		}
 		dfs_cache_free_tgts(&tgt_list);
+
 	}
+
+	put_tcp_super(sb);
 #endif
 	if (server->tcpStatus == CifsNeedNegotiate)
 		mod_delayed_work(cifsiod_wq, &server->echo, 0);
Signature · 1 line
2.20.1
author maintainer
#117 lore raw

PATCH 114/134 um: virtio: Keep reading on -EAGAIN

In reply to Sasha Levin

From: Johannes Berg <[email protected]>

[ Upstream commit 7e60746005573a06149cdee7acedf428906f3a59 ]

When we get an interrupt from the socket getting readable, and start reading, there's a possibility for a race. This depends on the implementation of the device, but e.g. with qemu's libvhost-user, we can see:

device virtio_uml
---------------------------------------
write header

                         get interrupt
                         read header
                         read body -> returns -EAGAIN

write body

The -EAGAIN return is because the socket is non-blocking, and then this leads us to abandon this message.

In fact, we've already read the header, so when the get another signal/interrupt for the body, we again read it as though it's a new message header, and also abandon it for the same reason (wrong size etc.)

This essentially breaks things, and if that message was one that required a response, it leads to a deadlock as the device is waiting for the response but we'll never reply.

Fix this by spinning on -EAGAIN as well when we read the message body. We need to handle -EAGAIN as "no message" while reading the header, since we share an interrupt.

Note that this situation is highly unlikely to occur in normal usage, since there will be very few messages and only in the startup phase. With the inband call feature this does tend to happen (eventually) though.

---
 arch/um/drivers/virtio_uml.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
arch/um/drivers/virtio_uml.c +4 -4
index fc8c52cff5aa8..c5643a59a8c76 100644
@@ -83,7 +83,7 @@ static int full_sendmsg_fds(int fd, const void *buf, unsigned int len,
 	return 0;
 }
 
-static int full_read(int fd, void *buf, int len)
+static int full_read(int fd, void *buf, int len, bool abortable)
 {
 	int rc;
 
@@ -93,7 +93,7 @@ static int full_read(int fd, void *buf, int len)
 			buf += rc;
 			len -= rc;
 		}
-	} while (len && (rc > 0 || rc == -EINTR));
+	} while (len && (rc > 0 || rc == -EINTR || (!abortable && rc == -EAGAIN)));
 
 	if (rc < 0)
 		return rc;
@@ -104,7 +104,7 @@ static int full_read(int fd, void *buf, int len)
 
 static int vhost_user_recv_header(int fd, struct vhost_user_msg *msg)
 {
-	return full_read(fd, msg, sizeof(msg->header));
+	return full_read(fd, msg, sizeof(msg->header), true);
 }
 
 static int vhost_user_recv(int fd, struct vhost_user_msg *msg,
@@ -118,7 +118,7 @@ static int vhost_user_recv(int fd, struct vhost_user_msg *msg,
 	size = msg->header.size;
 	if (size > max_payload_size)
 		return -EPROTO;
-	return full_read(fd, &msg->payload, size);
+	return full_read(fd, &msg->payload, size, false);
 }
 
 static int vhost_user_recv_resp(struct virtio_uml_device *vu_dev,
Signature · 1 line
2.20.1
author maintainer
#118 lore raw

PATCH 115/134 io_uring: io_allocate_scq_urings() should return a sane state

In reply to Sasha Levin

From: Jens Axboe <[email protected]>

[ Upstream commit eb065d301e8c83643367bdb0898becc364046bda ]

We currently rely on the ring destroy on cleaning things up in case of failure, but io_allocate_scq_urings() can leave things half initialized if only parts of it fails.

Be nice and return with either everything setup in success, or return an error with things nicely cleaned up.

---
 fs/io_uring.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
fs/io_uring.c +8 -2
index cbe8dabb6479c..7e900bfd24718 100644
@@ -3756,12 +3756,18 @@ static int io_allocate_scq_urings(struct io_ring_ctx *ctx,
 	ctx->cq_entries = rings->cq_ring_entries;
 
 	size = array_size(sizeof(struct io_uring_sqe), p->sq_entries);
-	if (size == SIZE_MAX)
+	if (size == SIZE_MAX) {
+		io_mem_free(ctx->rings);
+		ctx->rings = NULL;
 		return -EOVERFLOW;
+	}
 
 	ctx->sq_sqes = io_mem_alloc(size);
-	if (!ctx->sq_sqes)
+	if (!ctx->sq_sqes) {
+		io_mem_free(ctx->rings);
+		ctx->rings = NULL;
 		return -ENOMEM;
+	}
 
 	return 0;
 }
Signature · 1 line
2.20.1
author maintainer
#119 lore raw

PATCH 116/134 of: unittest: fix memory leak in attach_node_and_children

In reply to Sasha Levin

From: Erhard Furtner <[email protected]>

[ Upstream commit 2aacace6dbbb6b6ce4e177e6c7ea901f389c0472 ]

In attach_node_and_children memory is allocated for full_name via kasprintf. If the condition of the 1st if is not met the function returns early without freeing the memory. Add a kfree() to fix that.

This has been detected with kmemleak:

It looks like the leak was introduced by this commit:

Fixes 5babefb7f7ab ("of: unittest: allow base devicetree to have symbol metadata")
---
 drivers/of/unittest.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/of/unittest.c +3 -1
index 92e895d864584..ca7823eef2b40 100644
@@ -1146,8 +1146,10 @@ static void attach_node_and_children(struct device_node *np)
 	full_name = kasprintf(GFP_KERNEL, "%pOF", np);
 
 	if (!strcmp(full_name, "/__local_fixups__") ||
-	    !strcmp(full_name, "/__fixups__"))
+	    !strcmp(full_name, "/__fixups__")) {
+		kfree(full_name);
 		return;
+	}
 
 	dup = of_find_node_by_path(full_name);
 	kfree(full_name);
Signature · 1 line
2.20.1
author maintainer
#120 lore raw

PATCH 117/134 cdrom: respect device capabilities during opening action

In reply to Sasha Levin

From: Diego Elio Pettenò <[email protected]>

[ Upstream commit 366ba7c71ef77c08d06b18ad61b26e2df7352338 ]

Reading the TOC only works if the device can play audio, otherwise these commands fail (and possibly bring the device to an unhealthy state.)

Similarly, cdrom_mmc3_profile() should only be called if the device supports generic packet commands.

To: Jens Axboe <[email protected]>

---
 drivers/cdrom/cdrom.c | 12 +++++++++++-
 1 file changed, 11 insertions(+), 1 deletion(-)
drivers/cdrom/cdrom.c +11 -1
index ac42ae4651ce7..eebdcbef0578f 100644
@@ -996,6 +996,12 @@ static void cdrom_count_tracks(struct cdrom_device_info *cdi, tracktype *tracks)
 	tracks->xa = 0;
 	tracks->error = 0;
 	cd_dbg(CD_COUNT_TRACKS, "entering cdrom_count_tracks\n");
+
+	if (!CDROM_CAN(CDC_PLAY_AUDIO)) {
+		tracks->error = CDS_NO_INFO;
+		return;
+	}
+
 	/* Grab the TOC header so we can see how many tracks there are */
 	ret = cdi->ops->audio_ioctl(cdi, CDROMREADTOCHDR, &header);
 	if (ret) {
@@ -1162,7 +1168,8 @@ int cdrom_open(struct cdrom_device_info *cdi, struct block_device *bdev,
 		ret = open_for_data(cdi);
 		if (ret)
 			goto err;
-		cdrom_mmc3_profile(cdi);
+		if (CDROM_CAN(CDC_GENERIC_PACKET))
+			cdrom_mmc3_profile(cdi);
 		if (mode & FMODE_WRITE) {
 			ret = -EROFS;
 			if (cdrom_open_write(cdi))
@@ -2882,6 +2889,9 @@ int cdrom_get_last_written(struct cdrom_device_info *cdi, long *last_written)
 	   it doesn't give enough information or fails. then we return
 	   the toc contents. */
 use_toc:
+	if (!CDROM_CAN(CDC_PLAY_AUDIO))
+		return -ENOSYS;
+
 	toc.cdte_format = CDROM_MSF;
 	toc.cdte_track = CDROM_LEADOUT;
 	if ((ret = cdi->ops->audio_ioctl(cdi, CDROMREADTOCENTRY, &toc)))
Signature · 1 line
2.20.1
author maintainer
#121 lore raw

PATCH 118/134 cifs: move cifsFileInfo_put logic into a work-queue

In reply to Sasha Levin

From: Ronnie Sahlberg <[email protected]>

[ Upstream commit 32546a9586aa4565035bb557e191648e022b29e8 ]

This patch moves the final part of the cifsFileInfo_put() logic where we need a write lock on lock_sem to be processed in a separate thread that holds no other locks.
This is to prevent deadlocks like the one below:

Quoted 232 lines

there are 6 processes looping to while trying to down_write cinode->lock_sem, 5 of them from _cifsFileInfo_put, and one from cifs_new_fileinfo

and there are 5 other processes which are blocked, several of them waiting on either PG_writeback or PG_locked (which are both set), all for the same page of the file

2 inode_lock() (inode->i_rwsem) for the file
1 wait_on_page_writeback() for the page
1 down_read(inode->i_rwsem) for the inode of the directory 1 inode_lock()(inode->i_rwsem) for the inode of the directory 1 __lock_page

so processes are blocked waiting on:
page flags PG_locked and PG_writeback for one specific page inode->i_rwsem for the directory
inode->i_rwsem for the file
cifsInodeInflock_sem

here are the more gory details (let me know if I need to provide anything more/better):

[0 00:48:22.765] [UN] PID: 8863 TASK: ffff8c691547c5c0 CPU: 3
COMMAND: "reopen_file"
#0 [ffff9965007e3ba8] __schedule at ffffffff9b6e6095
#1 [ffff9965007e3c38] schedule at ffffffff9b6e64df
#2 [ffff9965007e3c48] rwsem_down_write_slowpath at ffffffff9af283d7 #3 [ffff9965007e3cb8] legitimize_path at ffffffff9b0f975d #4 [ffff9965007e3d08] path_openat at ffffffff9b0fe55d
#5 [ffff9965007e3dd8] do_filp_open at ffffffff9b100a33 #6 [ffff9965007e3ee0] do_sys_open at ffffffff9b0eb2d6
#7 [ffff9965007e3f38] do_syscall_64 at ffffffff9ae04315
* (I think legitimize_path is bogus)

in path_openat

        } else {
                const char *s = path_init(nd, flags);
                while (!(error = link_path_walk(s, nd)) &&
                        (error = do_last(nd, file, op)) > 0) {  <<<<

do_last:
if (open_flag & O_CREAT)
inode_lock(dir->d_inode); <<<<
else
so it's trying to take inode->i_rwsem for the directory

DENTRY INODE SUPERBLK TYPE PATH
ffff8c68bb8e79c0 ffff8c691158ef20 ffff8c6915bf9000 DIR /mnt/vm1_smb/ inode.i_rwsem is ffff8c691158efc0

<struct rw_semaphore 0xffff8c691158efc0>:

        owner: <struct task_struct 0xffff8c6914275d00> (UN -   8856 -
reopen_file), counter: 0x0000000000000003
        waitlist: 2
        0xffff9965007e3c90     8863   reopen_file      UN 0  1:29:22.926
  RWSEM_WAITING_FOR_WRITE
        0xffff996500393e00     9802   ls               UN 0  1:17:26.700

RWSEM_WAITING_FOR_READ

the owner of the inode.i_rwsem of the directory is:

[0 00:00:00.109] [UN] PID: 8856 TASK: ffff8c6914275d00 CPU: 3
COMMAND: "reopen_file"
#0 [ffff99650065b828] __schedule at ffffffff9b6e6095
#1 [ffff99650065b8b8] schedule at ffffffff9b6e64df
#2 [ffff99650065b8c8] schedule_timeout at ffffffff9b6e9f89
#3 [ffff99650065b940] msleep at ffffffff9af573a9
#4 [ffff99650065b948] _cifsFileInfo_put.cold.63 at ffffffffc0a42dd6 [cifs] #5 [ffff99650065ba38] cifs_writepage_locked at ffffffffc0a0b8f3 [cifs] #6 [ffff99650065bab0] cifs_launder_page at ffffffffc0a0bb72 [cifs] #7 [ffff99650065bb30] invalidate_inode_pages2_range at ffffffff9b04d4bd #8 [ffff99650065bcb8] cifs_invalidate_mapping at ffffffffc0a11339 [cifs] #9 [ffff99650065bcd0] cifs_revalidate_mapping at ffffffffc0a1139a [cifs]
#10 [ffff99650065bcf0] cifs_d_revalidate at ffffffffc0a014f6 [cifs] #11 [ffff99650065bd08] path_openat at ffffffff9b0fe7f7
#12 [ffff99650065bdd8] do_filp_open at ffffffff9b100a33
#13 [ffff99650065bee0] do_sys_open at ffffffff9b0eb2d6
#14 [ffff99650065bf38] do_syscall_64 at ffffffff9ae04315

cifs_launder_page is for page 0xffffd1e2c07d2480

crash> page.index,mapping,flags 0xffffd1e2c07d2480

      index = 0x8
      mapping = 0xffff8c68f3cd0db0

flags = 0xfffffc0008095

  PAGE-FLAG       BIT  VALUE
  PG_locked         0  0000001
  PG_uptodate       2  0000004
  PG_lru            4  0000010
  PG_waiters        7  0000080
  PG_writeback     15  0008000

inode is ffff8c68f3cd0c40
inode.i_rwsem is ffff8c68f3cd0ce0
DENTRY INODE SUPERBLK TYPE PATH
ffff8c68a1f1b480 ffff8c68f3cd0c40 ffff8c6915bf9000 REG /mnt/vm1_smb/testfile.8853

this process holds the inode->i_rwsem for the parent directory, is laundering a page attached to the inode of the file it's opening, and in _cifsFileInfo_put is trying to down_write the cifsInodeInflock_sem for the file itself.

<struct rw_semaphore 0xffff8c68f3cd0ce0>:

        owner: <struct task_struct 0xffff8c6914272e80> (UN -   8854 -
reopen_file), counter: 0x0000000000000003
        waitlist: 1
        0xffff9965005dfd80     8855   reopen_file      UN 0  1:29:22.912

RWSEM_WAITING_FOR_WRITE

this is the inode.i_rwsem for the file

the owner:

[0 00:48:22.739] [UN] PID: 8854 TASK: ffff8c6914272e80 CPU: 2
COMMAND: "reopen_file"
#0 [ffff99650054fb38] __schedule at ffffffff9b6e6095
#1 [ffff99650054fbc8] schedule at ffffffff9b6e64df
#2 [ffff99650054fbd8] io_schedule at ffffffff9b6e68e2
#3 [ffff99650054fbe8] __lock_page at ffffffff9b03c56f
#4 [ffff99650054fc80] pagecache_get_page at ffffffff9b03dcdf
#5 [ffff99650054fcc0] grab_cache_page_write_begin at ffffffff9b03ef4c #6 [ffff99650054fcd0] cifs_write_begin at ffffffffc0a064ec [cifs] #7 [ffff99650054fd30] generic_perform_write at ffffffff9b03bba4 #8 [ffff99650054fda8] __generic_file_write_iter at ffffffff9b04060a #9 [ffff99650054fdf0] cifs_strict_writev.cold.70 at ffffffffc0a4469b [cifs]
#10 [ffff99650054fe48] new_sync_write at ffffffff9b0ec1dd
#11 [ffff99650054fed0] vfs_write at ffffffff9b0eed35
#12 [ffff99650054ff00] ksys_write at ffffffff9b0eefd9
#13 [ffff99650054ff38] do_syscall_64 at ffffffff9ae04315

the process holds the inode->i_rwsem for the file to which it's writing, and is trying to __lock_page for the same page as in the other processes

the other tasks:
[0 00:00:00.028] [UN] PID: 8859 TASK: ffff8c6915479740 CPU: 2
COMMAND: "reopen_file"
#0 [ffff9965007b39d8] __schedule at ffffffff9b6e6095
#1 [ffff9965007b3a68] schedule at ffffffff9b6e64df
#2 [ffff9965007b3a78] schedule_timeout at ffffffff9b6e9f89
#3 [ffff9965007b3af0] msleep at ffffffff9af573a9
#4 [ffff9965007b3af8] cifs_new_fileinfo.cold.61 at ffffffffc0a42a07 [cifs] #5 [ffff9965007b3b78] cifs_open at ffffffffc0a0709d [cifs]
#6 [ffff9965007b3cd8] do_dentry_open at ffffffff9b0e9b7a
#7 [ffff9965007b3d08] path_openat at ffffffff9b0fe34f
#8 [ffff9965007b3dd8] do_filp_open at ffffffff9b100a33
#9 [ffff9965007b3ee0] do_sys_open at ffffffff9b0eb2d6
#10 [ffff9965007b3f38] do_syscall_64 at ffffffff9ae04315

this is opening the file, and is trying to down_write cinode->lock_sem

[0 00:00:00.041] [UN] PID: 8860 TASK: ffff8c691547ae80 CPU: 2
COMMAND: "reopen_file"
[0 00:00:00.057] [UN] PID: 8861 TASK: ffff8c6915478000 CPU: 3
COMMAND: "reopen_file"
[0 00:00:00.059] [UN] PID: 8858 TASK: ffff8c6914271740 CPU: 2
COMMAND: "reopen_file"
[0 00:00:00.109] [UN] PID: 8862 TASK: ffff8c691547dd00 CPU: 6
COMMAND: "reopen_file"
#0 [ffff9965007c3c78] __schedule at ffffffff9b6e6095
#1 [ffff9965007c3d08] schedule at ffffffff9b6e64df
#2 [ffff9965007c3d18] schedule_timeout at ffffffff9b6e9f89
#3 [ffff9965007c3d90] msleep at ffffffff9af573a9
#4 [ffff9965007c3d98] _cifsFileInfo_put.cold.63 at ffffffffc0a42dd6 [cifs] #5 [ffff9965007c3e88] cifs_close at ffffffffc0a07aaf [cifs]
#6 [ffff9965007c3ea0] __fput at ffffffff9b0efa6e
#7 [ffff9965007c3ee8] task_work_run at ffffffff9aef1614
#8 [ffff9965007c3f20] exit_to_usermode_loop at ffffffff9ae03d6f #9 [ffff9965007c3f38] do_syscall_64 at ffffffff9ae0444c

closing the file, and trying to down_write cifsi->lock_sem

[0 00:48:22.839] [UN] PID: 8857 TASK: ffff8c6914270000 CPU: 7
COMMAND: "reopen_file"
#0 [ffff9965006a7cc8] __schedule at ffffffff9b6e6095
#1 [ffff9965006a7d58] schedule at ffffffff9b6e64df
#2 [ffff9965006a7d68] io_schedule at ffffffff9b6e68e2
#3 [ffff9965006a7d78] wait_on_page_bit at ffffffff9b03cac6 #4 [ffff9965006a7e10] __filemap_fdatawait_range at ffffffff9b03b028 #5 [ffff9965006a7ed8] filemap_write_and_wait at ffffffff9b040165 #6 [ffff9965006a7ef0] cifs_flush at ffffffffc0a0c2fa [cifs] #7 [ffff9965006a7f10] filp_close at ffffffff9b0e93f1
#8 [ffff9965006a7f30] __x64_sys_close at ffffffff9b0e9a0e #9 [ffff9965006a7f38] do_syscall_64 at ffffffff9ae04315

in __filemap_fdatawait_range

                        wait_on_page_writeback(page);

for the same page of the file

[0 00:48:22.718] [UN] PID: 8855 TASK: ffff8c69142745c0 CPU: 7
COMMAND: "reopen_file"
#0 [ffff9965005dfc98] __schedule at ffffffff9b6e6095
#1 [ffff9965005dfd28] schedule at ffffffff9b6e64df
#2 [ffff9965005dfd38] rwsem_down_write_slowpath at ffffffff9af283d7 #3 [ffff9965005dfdf0] cifs_strict_writev at ffffffffc0a0c40a [cifs] #4 [ffff9965005dfe48] new_sync_write at ffffffff9b0ec1dd #5 [ffff9965005dfed0] vfs_write at ffffffff9b0eed35
#6 [ffff9965005dff00] ksys_write at ffffffff9b0eefd9
#7 [ffff9965005dff38] do_syscall_64 at ffffffff9ae04315

        inode_lock(inode);

and one 'ls' later on, to see whether the rest of the mount is available (the test file is in the root, so we get blocked up on the directory ->i_rwsem), so the entire mount is unavailable

[0 00:36:26.473] [UN] PID: 9802 TASK: ffff8c691436ae80 CPU: 4
COMMAND: "ls"
#0 [ffff996500393d28] __schedule at ffffffff9b6e6095
#1 [ffff996500393db8] schedule at ffffffff9b6e64df
#2 [ffff996500393dc8] rwsem_down_read_slowpath at ffffffff9b6e9421 #3 [ffff996500393e78] down_read_killable at ffffffff9b6e95e2 #4 [ffff996500393e88] iterate_dir at ffffffff9b103c56 #5 [ffff996500393ec8] ksys_getdents64 at ffffffff9b104b0c #6 [ffff996500393f30] __x64_sys_getdents64 at ffffffff9b104bb6 #7 [ffff996500393f38] do_syscall_64 at ffffffff9ae04315

Quoted 6 lines

in iterate_dir:

        if (shared)
                res = down_read_killable(&inode->i_rwsem);  <<<<
        else
                res = down_write_killable(&inode->i_rwsem);
---
 fs/cifs/cifsfs.c   | 13 +++++++-
 fs/cifs/cifsglob.h |  5 +++-
 fs/cifs/file.c     | 74 ++++++++++++++++++++++++++++++----------------
 3 files changed, 65 insertions(+), 27 deletions(-)
fs/cifs/cifsfs.c +12 -1
index 1a135d1b85bd8..07d8ace61f777 100644
@@ -119,6 +119,7 @@ extern mempool_t *cifs_mid_poolp;
 
 struct workqueue_struct	*cifsiod_wq;
 struct workqueue_struct	*decrypt_wq;
+struct workqueue_struct	*fileinfo_put_wq;
 struct workqueue_struct	*cifsoplockd_wq;
 __u32 cifs_lock_secret;
 
@@ -1554,11 +1555,18 @@ init_cifs(void)
 		goto out_destroy_cifsiod_wq;
 	}
 
+	fileinfo_put_wq = alloc_workqueue("cifsfileinfoput",
+				     WQ_UNBOUND|WQ_FREEZABLE|WQ_MEM_RECLAIM, 0);
+	if (!fileinfo_put_wq) {
+		rc = -ENOMEM;
+		goto out_destroy_decrypt_wq;
+	}
+
 	cifsoplockd_wq = alloc_workqueue("cifsoplockd",
 					 WQ_FREEZABLE|WQ_MEM_RECLAIM, 0);
 	if (!cifsoplockd_wq) {
 		rc = -ENOMEM;
-		goto out_destroy_decrypt_wq;
+		goto out_destroy_fileinfo_put_wq;
 	}
 
 	rc = cifs_fscache_register();
@@ -1624,6 +1632,8 @@ out_unreg_fscache:
 	cifs_fscache_unregister();
 out_destroy_cifsoplockd_wq:
 	destroy_workqueue(cifsoplockd_wq);
+out_destroy_fileinfo_put_wq:
+	destroy_workqueue(fileinfo_put_wq);
 out_destroy_decrypt_wq:
 	destroy_workqueue(decrypt_wq);
 out_destroy_cifsiod_wq:
@@ -1653,6 +1663,7 @@ exit_cifs(void)
 	cifs_fscache_unregister();
 	destroy_workqueue(cifsoplockd_wq);
 	destroy_workqueue(decrypt_wq);
+	destroy_workqueue(fileinfo_put_wq);
 	destroy_workqueue(cifsiod_wq);
 	cifs_proc_clean();
 }
fs/cifs/cifsglob.h +4 -1
index d78bfcc191560..e0a7dc3f62b29 100644
@@ -1265,6 +1265,7 @@ struct cifsFileInfo {
 	struct mutex fh_mutex; /* prevents reopen race after dead ses*/
 	struct cifs_search_info srch_inf;
 	struct work_struct oplock_break; /* work for oplock breaks */
+	struct work_struct put; /* work for the final part of _put */
 };
 
 struct cifs_io_parms {
@@ -1370,7 +1371,8 @@ cifsFileInfo_get_locked(struct cifsFileInfo *cifs_file)
 }
 
 struct cifsFileInfo *cifsFileInfo_get(struct cifsFileInfo *cifs_file);
-void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_hdlr);
+void _cifsFileInfo_put(struct cifsFileInfo *cifs_file, bool wait_oplock_hdlr,
+		       bool offload);
 void cifsFileInfo_put(struct cifsFileInfo *cifs_file);
 
 #define CIFS_CACHE_READ_FLG	1
@@ -1907,6 +1909,7 @@ void cifs_queue_oplock_break(struct cifsFileInfo *cfile);
 extern const struct slow_work_ops cifs_oplock_break_ops;
 extern struct workqueue_struct *cifsiod_wq;
 extern struct workqueue_struct *decrypt_wq;
+extern struct workqueue_struct *fileinfo_put_wq;
 extern struct workqueue_struct *cifsoplockd_wq;
 extern __u32 cifs_lock_secret;
 
fs/cifs/file.c +49 -25

134 lines not on this page · raw message

Signature · 1 line
2.20.1
author maintainer
#122 lore raw

PATCH 119/134 perf diff: Use llabs() with 64-bit values

In reply to Sasha Levin

From: Arnaldo Carvalho de Melo <[email protected]>

[ Upstream commit 98e93245113d0f5c279ef77f4a9e7d097323ad71 ]

To fix these build errors on a debian mipsel cross build environment:

builtin-diff.c: In function 'block_cycles_diff_cmp':
builtin-diff.c:550:6: error: absolute value function 'labs' given an argument of type 's64' {aka 'long long int'} but has parameter of type 'long int' which may cause truncation of value [-Werror=absolute-value]

    550 |  l = labs(left->diff.cycles);
        |      ^~~~
  builtin-diff.c:551:6: error: absolute value function 'labs' given an argument of type 's64' {aka 'long long int'} but has parameter of type 'long int' which may cause truncation of value [-Werror=absolute-value]
    551 |  r = labs(right->diff.cycles);
        |      ^~~~
Fixes 99150a1faab2 ("perf diff: Use hists to manage basic blocks per symbol")
Cc Jin Yao <[email protected]>
Cc Adrian Hunter <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Namhyung Kim <[email protected]>
---
 tools/perf/builtin-diff.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-diff.c +2 -2
index c37a786779555..265682296836d 100644
@@ -575,8 +575,8 @@ static int64_t block_cycles_diff_cmp(struct hist_entry *left,
 	if (!pairs_left && !pairs_right)
 		return 0;
 
-	l = labs(left->diff.cycles);
-	r = labs(right->diff.cycles);
+	l = llabs(left->diff.cycles);
+	r = llabs(right->diff.cycles);
 	return r - l;
 }
 
Signature · 1 line
2.20.1
author maintainer
#123 lore raw

PATCH 120/134 perf script: Fix invalid LBR/binary mismatch error

In reply to Sasha Levin

From: Adrian Hunter <[email protected]>

[ Upstream commit 5172672da02e483d9b3c4d814c3482d0c8ffb1a6 ]

The 'len' returned by grab_bb() includes an extra MAXINSN bytes to allow for the last instruction, so the the final 'offs' will not be 'len'. Fix the error condition logic accordingly.

Before:

  $ perf record -e '{intel_pt//,cpu/mem_inst_retired.all_loads,aux-sample-size=8192/pp}:u' grep -rqs jhgjhg /boot
  [ perf record: Woken up 19 times to write data ]
  [ perf record: Captured and wrote 2.274 MB perf.data ]
  $ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
            grep 13759 [002]  8091.310257:       1862                                        instructions:uH:      5641d58069eb bmexec+0x86b (/bin/grep)
        bmexec+2485:
        00005641d5806b35                        jnz 0x5641d5806bd0              # MISPRED
        00005641d5806bd0                        movzxb  (%r13,%rdx,1), %eax
        00005641d5806bd6                        add %rdi, %rax
        00005641d5806bd9                        movzxb  -0x1(%rax), %edx
        00005641d5806bdd                        cmp %rax, %r14
        00005641d5806be0                        jnb 0x5641d58069c0              # MISPRED
        mismatch of LBR data and executable
        00005641d58069c0                        movzxb  (%r13,%rdx,1), %edi

After:

  $ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
            grep 13759 [002]  8091.310257:       1862                                        instructions:uH:      5641d58069eb bmexec+0x86b (/bin/grep)
        bmexec+2485:
        00005641d5806b35                        jnz 0x5641d5806bd0              # MISPRED
        00005641d5806bd0                        movzxb  (%r13,%rdx,1), %eax
        00005641d5806bd6                        add %rdi, %rax
        00005641d5806bd9                        movzxb  -0x1(%rax), %edx
        00005641d5806bdd                        cmp %rax, %r14
        00005641d5806be0                        jnb 0x5641d58069c0              # MISPRED
        00005641d58069c0                        movzxb  (%r13,%rdx,1), %edi
        00005641d58069c6                        add %rax, %rdi
Fixes e98df280bc2a ("perf script brstackinsn: Fix recovery from LBR/binary mismatch")
Cc Jiri Olsa <[email protected]>
---
 tools/perf/builtin-script.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/builtin-script.c +1 -1
index 67be8d31afab6..6dba8b728d232 100644
@@ -1084,7 +1084,7 @@ static int perf_sample__fprintf_brstackinsn(struct perf_sample *sample,
 				insn++;
 			}
 		}
-		if (off != (unsigned)len)
+		if (off != end - start)
 			printed += fprintf(fp, "\tmismatch of LBR data and executable\n");
 	}
 
Signature · 1 line
2.20.1
author maintainer
#124 lore raw

PATCH 121/134 perf script: Fix brstackinsn for AUXTRACE

In reply to Sasha Levin

From: Adrian Hunter <[email protected]>

[ Upstream commit 0cd032d3b5fcebf5454315400ab310746a81ca53 ]

brstackinsn must be allowed to be set by the user when AUX area data has been captured because, in that case, the branch stack might be synthesized on the fly. This fixes the following error:

Before:

  $ perf record -e '{intel_pt//,cpu/mem_inst_retired.all_loads,aux-sample-size=8192/pp}:u' grep -rqs jhgjhg /boot
  [ perf record: Woken up 19 times to write data ]
  [ perf record: Captured and wrote 2.274 MB perf.data ]
  $ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
  Display of branch stack assembler requested, but non all-branch filter set
  Hint: run 'perf record -b ...'

After:

  $ perf record -e '{intel_pt//,cpu/mem_inst_retired.all_loads,aux-sample-size=8192/pp}:u' grep -rqs jhgjhg /boot
  [ perf record: Woken up 19 times to write data ]
  [ perf record: Captured and wrote 2.274 MB perf.data ]
  $ perf script -F +brstackinsn --xed --itrace=i1usl100 | head
            grep 13759 [002]  8091.310257:       1862                                        instructions:uH:      5641d58069eb bmexec+0x86b (/bin/grep)
        bmexec+2485:
        00005641d5806b35                        jnz 0x5641d5806bd0              # MISPRED
        00005641d5806bd0                        movzxb  (%r13,%rdx,1), %eax
        00005641d5806bd6                        add %rdi, %rax
        00005641d5806bd9                        movzxb  -0x1(%rax), %edx
        00005641d5806bdd                        cmp %rax, %r14
        00005641d5806be0                        jnb 0x5641d58069c0              # MISPRED
        mismatch of LBR data and executable
        00005641d58069c0                        movzxb  (%r13,%rdx,1), %edi
Fixes 48d02a1d5c13 ("perf script: Add 'brstackinsn' for branch stacks")
Cc Jiri Olsa <[email protected]>
---
 tools/perf/builtin-script.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/builtin-script.c +1 -1
index 6dba8b728d232..3983d6ccd14df 100644
@@ -448,7 +448,7 @@ static int perf_evsel__check_attr(struct evsel *evsel,
 		       "selected. Hence, no address to lookup the source line number.\n");
 		return -EINVAL;
 	}
-	if (PRINT_FIELD(BRSTACKINSN) &&
+	if (PRINT_FIELD(BRSTACKINSN) && !allow_user_set &&
 	    !(perf_evlist__combined_branch_type(session->evlist) &
 	      PERF_SAMPLE_BRANCH_ANY)) {
 		pr_err("Display of branch stack assembler requested, but non all-branch filter set\n"
Signature · 1 line
2.20.1
author maintainer
#125 lore raw

PATCH 122/134 perf regs: Make perf_reg_name() return "unknown" instead of NULL

In reply to Sasha Levin

From: Arnaldo Carvalho de Melo <[email protected]>

[ Upstream commit 5b596e0ff0e1852197d4c82d3314db5e43126bf7 ]

To avoid breaking the build on arches where this is not wired up, at least all the other features should be made available and when using this specific routine, the "unknown" should point the user/developer to the need to wire this up on this particular hardware architecture.

Detected in a container mipsel debian cross build environment, where it shows up as:

In file included from /usr/mipsel-linux-gnu/include/stdio.h:867,

                   from /git/linux/tools/perf/lib/include/perf/cpumap.h:6,
                   from util/session.c:13:
  In function 'printf',
      inlined from 'regs_dump__printf' at util/session.c:1103:3,
      inlined from 'regs__printf' at util/session.c:1131:2:
  /usr/mipsel-linux-gnu/include/bits/stdio2.h:107:10: error: '%-5s' directive argument is null [-Werror=format-overflow=]
    107 |   return __printf_chk (__USE_FORTIFY_LEVEL - 1, __fmt, __va_arg_pack ());
        |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

cross compiler details:

mipsel-linux-gnu-gcc (Debian 9.2.1-8) 9.2.1 20190909

Also on mips64:

In file included from /usr/mips64-linux-gnuabi64/include/stdio.h:867,

                   from /git/linux/tools/perf/lib/include/perf/cpumap.h:6,
                   from util/session.c:13:
  In function 'printf',
      inlined from 'regs_dump__printf' at util/session.c:1103:3,
      inlined from 'regs__printf' at util/session.c:1131:2,
      inlined from 'regs_user__printf' at util/session.c:1139:3,
      inlined from 'dump_sample' at util/session.c:1246:3,
      inlined from 'machines__deliver_event' at util/session.c:1421:3:
  /usr/mips64-linux-gnuabi64/include/bits/stdio2.h:107:10: error: '%-5s' directive argument is null [-Werror=format-overflow=]
    107 |   return __printf_chk (__USE_FORTIFY_LEVEL - 1, __fmt, __va_arg_pack ());
        |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  In function 'printf',
      inlined from 'regs_dump__printf' at util/session.c:1103:3,
      inlined from 'regs__printf' at util/session.c:1131:2,
      inlined from 'regs_intr__printf' at util/session.c:1147:3,
      inlined from 'dump_sample' at util/session.c:1249:3,
      inlined from 'machines__deliver_event' at util/session.c:1421:3:
  /usr/mips64-linux-gnuabi64/include/bits/stdio2.h:107:10: error: '%-5s' directive argument is null [-Werror=format-overflow=]
    107 |   return __printf_chk (__USE_FORTIFY_LEVEL - 1, __fmt, __va_arg_pack ());
        |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

cross compiler details:

mips64-linux-gnuabi64-gcc (Debian 9.2.1-8) 9.2.1 20190909

Fixes 2bcd355b71da ("perf tools: Add interface to arch registers sets")
Cc Adrian Hunter <[email protected]>
Cc Jiri Olsa <[email protected]>
Cc Namhyung Kim <[email protected]>
---
 tools/perf/util/perf_regs.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/util/perf_regs.h +1 -1
index 47fe34e5f7d5b..ec7640cc4c911 100644
@@ -41,7 +41,7 @@ int perf_reg_value(u64 *valp, struct regs_dump *regs, int id);
 
 static inline const char *perf_reg_name(int id __maybe_unused)
 {
-	return NULL;
+	return "unknown";
 }
 
 static inline int perf_reg_value(u64 *valp __maybe_unused,
Signature · 1 line
2.20.1
author maintainer
#126 lore raw

PATCH 123/134 s390/zcrypt: handle new reply code FILTERED_BY_HYPERVISOR

In reply to Sasha Levin

From: Harald Freudenberger <[email protected]>

[ Upstream commit 6733775a92eacd612ac88afa0fd922e4ffeb2bc7 ]

This patch introduces support for a new architectured reply code 0x8B indicating that a hypervisor layer (if any) has rejected an ap message.

Linux may run as a guest on top of a hypervisor like zVM or KVM. So the crypto hardware seen by the ap bus may be restricted by the hypervisor for example only a subset like only clear key crypto requests may be supported. Other requests will be filtered out - rejected by the hypervisor. The new reply code 0x8B will appear in such cases and needs to get recognized by the ap bus and zcrypt device driver zoo.

---
 drivers/s390/crypto/zcrypt_error.h | 2 ++
 1 file changed, 2 insertions(+)
drivers/s390/crypto/zcrypt_error.h +2 -0
index f34ee41cbed88..4f4dd9d727c9e 100644
@@ -61,6 +61,7 @@ struct error_hdr {
 #define REP82_ERROR_EVEN_MOD_IN_OPND	    0x85
 #define REP82_ERROR_RESERVED_FIELD	    0x88
 #define REP82_ERROR_INVALID_DOMAIN_PENDING  0x8A
+#define REP82_ERROR_FILTERED_BY_HYPERVISOR  0x8B
 #define REP82_ERROR_TRANSPORT_FAIL	    0x90
 #define REP82_ERROR_PACKET_TRUNCATED	    0xA0
 #define REP82_ERROR_ZERO_BUFFER_LEN	    0xB0
@@ -91,6 +92,7 @@ static inline int convert_error(struct zcrypt_queue *zq,
 	case REP82_ERROR_INVALID_DOMAIN_PRECHECK:
 	case REP82_ERROR_INVALID_DOMAIN_PENDING:
 	case REP82_ERROR_INVALID_SPECIAL_CMD:
+	case REP82_ERROR_FILTERED_BY_HYPERVISOR:
 	//   REP88_ERROR_INVALID_KEY		// '82' CEX2A
 	//   REP88_ERROR_OPERAND		// '84' CEX2A
 	//   REP88_ERROR_OPERAND_EVEN_MOD	// '85' CEX2A
Signature · 1 line
2.20.1
author maintainer
#127 lore raw

PATCH 124/134 mailbox: imx: Clear the right interrupts at shutdown

In reply to Sasha Levin

From: Daniel Baluta <[email protected]>

[ Upstream commit 5f0af07e89199ac51cdd4f25bc303bdc703f4e9c ]

Make sure to only clear enabled interrupts keeping count of the connection type.

---
 drivers/mailbox/imx-mailbox.c | 15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)
drivers/mailbox/imx-mailbox.c +13 -2
index 9f74dee1a58c7..d28bbd47ff882 100644
@@ -217,8 +217,19 @@ static void imx_mu_shutdown(struct mbox_chan *chan)
 	if (cp->type == IMX_MU_TYPE_TXDB)
 		tasklet_kill(&cp->txdb_tasklet);
 
-	imx_mu_xcr_rmw(priv, 0, IMX_MU_xCR_TIEn(cp->idx) |
-		       IMX_MU_xCR_RIEn(cp->idx) | IMX_MU_xCR_GIEn(cp->idx));
+	switch (cp->type) {
+	case IMX_MU_TYPE_TX:
+		imx_mu_xcr_rmw(priv, 0, IMX_MU_xCR_TIEn(cp->idx));
+		break;
+	case IMX_MU_TYPE_RX:
+		imx_mu_xcr_rmw(priv, 0, IMX_MU_xCR_RIEn(cp->idx));
+		break;
+	case IMX_MU_TYPE_RXDB:
+		imx_mu_xcr_rmw(priv, 0, IMX_MU_xCR_GIEn(cp->idx));
+		break;
+	default:
+		break;
+	}
 
 	free_irq(priv->irq, chan);
 }
Signature · 1 line
2.20.1
author maintainer
#128 lore raw

PATCH 125/134 libfdt: define INT32_MAX and UINT32_MAX in libfdt_env.h

In reply to Sasha Levin

From: Masahiro Yamada <[email protected]>

[ Upstream commit a8de1304b7df30e3a14f2a8b9709bb4ff31a0385 ]

The DTC v1.5.1 added references to (U)INT32_MAX.

This is no problem for user-space programs since <stdint.h> defines (U)INT32_MAX along with (u)int32_t.

For the kernel space, libfdt_env.h needs to be adjusted before we pull in the changes.

In the kernel, we usually use s/u32 instead of (u)int32_t for the fixed-width types.

Accordingly, we already have S/U32_MAX for their max values. So, we should not add (U)INT32_MAX to <linux/limits.h> any more.

Instead, add them to the in-kernel libfdt_env.h to compile the latest libfdt.

---
 arch/arm/boot/compressed/libfdt_env.h | 4 +++-
 arch/powerpc/boot/libfdt_env.h        | 2 ++
 include/linux/libfdt_env.h            | 3 +++
 3 files changed, 8 insertions(+), 1 deletion(-)
arch/arm/boot/compressed/libfdt_env.h +3 -1
index b36c0289a308e..6a0f1f524466e 100644
@@ -2,11 +2,13 @@
 #ifndef _ARM_LIBFDT_ENV_H
 #define _ARM_LIBFDT_ENV_H
 
+#include <linux/limits.h>
 #include <linux/types.h>
 #include <linux/string.h>
 #include <asm/byteorder.h>
 
-#define INT_MAX			((int)(~0U>>1))
+#define INT32_MAX	S32_MAX
+#define UINT32_MAX	U32_MAX
 
 typedef __be16 fdt16_t;
 typedef __be32 fdt32_t;
arch/powerpc/boot/libfdt_env.h +2 -0
index 2abc8e83b95e9..9757d4f6331e7 100644
@@ -6,6 +6,8 @@
 #include <string.h>
 
 #define INT_MAX			((int)(~0U>>1))
+#define UINT32_MAX		((u32)~0U)
+#define INT32_MAX		((s32)(UINT32_MAX >> 1))
 
 #include "of.h"
 
include/linux/libfdt_env.h +3 -0
index edb0f0c309044..1adf54aad2df1 100644
@@ -7,6 +7,9 @@
 
 #include <asm/byteorder.h>
 
+#define INT32_MAX	S32_MAX
+#define UINT32_MAX	U32_MAX
+
 typedef __be16 fdt16_t;
 typedef __be32 fdt32_t;
 typedef __be64 fdt64_t;
Signature · 1 line
2.20.1
author maintainer
#129 lore raw

PATCH 126/134 s390/unwind: filter out unreliable bogus %r14

In reply to Sasha Levin

From: Vasily Gorbik <[email protected]>

[ Upstream commit bf018ee644897d7982e1b8dd8b15e97db6e1a4da ]

Currently unwinder unconditionally returns %r14 from the first frame pointed by %r15 from pt_regs. A task could be interrupted when a function already allocated this frame (if it needs it) for its callees or to store local variables. In that case this frame would contain random values from stack or values stored there by a callee. As we are only interested in %r14 to get potential return address, skip bogus return addresses which doesn't belong to kernel text.

This helps to avoid duplicating filtering logic in unwider users, most of which use unwind_get_return_address() and would choke on bogus 0 address returned by it otherwise.

---
 arch/s390/kernel/unwind_bc.c | 5 +++++
 1 file changed, 5 insertions(+)
arch/s390/kernel/unwind_bc.c +5 -0
index a8204f952315d..6e609b13c0cec 100644
@@ -60,6 +60,11 @@ bool unwind_next_frame(struct unwind_state *state)
 		ip = READ_ONCE_NOCHECK(sf->gprs[8]);
 		reliable = false;
 		regs = NULL;
+		if (!__kernel_text_address(ip)) {
+			/* skip bogus %r14 */
+			state->regs = NULL;
+			return unwind_next_frame(state);
+		}
 	} else {
 		sf = (struct stack_frame *) state->sp;
 		sp = READ_ONCE_NOCHECK(sf->back_chain);
Signature · 1 line
2.20.1
author maintainer
#130 lore raw

PATCH 127/134 s390/cpum_sf: Check for SDBT and SDB consistency

In reply to Sasha Levin

From: Thomas Richter <[email protected]>

[ Upstream commit 247f265fa502e7b17a0cb0cc330e055a36aafce4 ]

Each SBDT is located at a 4KB page and contains 512 entries. Each entry of a SDBT points to a SDB, a 4KB page containing sampled data. The last entry is a link to another SDBT page.

When an event is created the function sequence executed is:

__hw_perf_event_init()

  +--> allocate_buffers()
       +--> realloc_sampling_buffers()
	    +---> alloc_sample_data_block()

Both functions realloc_sampling_buffers() and alloc_sample_data_block() allocate pages and the allocation can fail. This is handled correctly and all allocated pages are freed and error -ENOMEM is returned to the top calling function. Finally the event is not created.

Once the event has been created, the amount of initially allocated SDBT and SDB can be too low. This is detected during measurement interrupt handling, where the amount of lost samples is calculated. If the number of lost samples is too high considering sampling frequency and already allocated SBDs, the number of SDBs is enlarged during the next execution of cpumsf_pmu_enable().

If more SBDs need to be allocated, functions

       realloc_sampling_buffers()
       +---> alloc-sample_data_block()

are called to allocate more pages. Page allocation may fail and the returned error is ignored. A SDBT and SDB setup already exists.

However the modified SDBTs and SDBs might end up in a situation where the first entry of an SDBT does not point to an SDB, but another SDBT, basicly an SBDT without payload. This can not be handled by the interrupt handler, where an SDBT must have at least one entry pointing to an SBD.

Add a check to avoid SDBTs with out payload (SDBs) when enlarging the buffer setup.

---
 arch/s390/kernel/perf_cpum_sf.c | 17 +++++++++++++++--
 1 file changed, 15 insertions(+), 2 deletions(-)
arch/s390/kernel/perf_cpum_sf.c +15 -2
index 3d8b12a9a6ff4..7511b71d29313 100644
@@ -193,7 +193,7 @@ static int realloc_sampling_buffer(struct sf_buffer *sfb,
 				   unsigned long num_sdb, gfp_t gfp_flags)
 {
 	int i, rc;
-	unsigned long *new, *tail;
+	unsigned long *new, *tail, *tail_prev = NULL;
 
 	if (!sfb->sdbt || !sfb->tail)
 		return -EINVAL;
@@ -232,6 +232,7 @@ static int realloc_sampling_buffer(struct sf_buffer *sfb,
 			sfb->num_sdbt++;
 			/* Link current page to tail of chain */
 			*tail = (unsigned long)(void *) new + 1;
+			tail_prev = tail;
 			tail = new;
 		}
 
@@ -241,10 +242,22 @@ static int realloc_sampling_buffer(struct sf_buffer *sfb,
 		 * issue, a new realloc call (if required) might succeed.
 		 */
 		rc = alloc_sample_data_block(tail, gfp_flags);
-		if (rc)
+		if (rc) {
+			/* Undo last SDBT. An SDBT with no SDB at its first
+			 * entry but with an SDBT entry instead can not be
+			 * handled by the interrupt handler code.
+			 * Avoid this situation.
+			 */
+			if (tail_prev) {
+				sfb->num_sdbt--;
+				free_page((unsigned long) new);
+				tail = tail_prev;
+			}
 			break;
+		}
 		sfb->num_sdb++;
 		tail++;
+		tail_prev = new = NULL;	/* Allocated at least one SBD */
 	}
 
 	/* Link sampling buffer to its origin */
Signature · 1 line
2.20.1
author maintainer
#131 lore raw

PATCH 128/134 ocfs2: fix passing zero to 'PTR_ERR' warning

In reply to Sasha Levin

From: Ding Xiang <[email protected]>

[ Upstream commit 188c523e1c271d537f3c9f55b6b65bf4476de32f ]

Fix a static code checker warning:
fs/ocfs2/acl.c:331
ocfs2_acl_chmod() warn: passing zero to 'PTR_ERR'

Fixes 5ee0fbd50fd ("ocfs2: revert using ocfs2_acl_chmod to avoid inode cluster lock hang")
Cc Mark Fasheh <[email protected]>
Cc Joel Becker <[email protected]>
Cc Junxiao Bi <[email protected]>
Cc Changwei Ge <[email protected]>
Cc Gang He <[email protected]>
Cc Jun Piao <[email protected]>
---
 fs/ocfs2/acl.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/ocfs2/acl.c +2 -2
index 3e7da392aa6f8..bb981ec76456b 100644
@@ -327,8 +327,8 @@ int ocfs2_acl_chmod(struct inode *inode, struct buffer_head *bh)
 	down_read(&OCFS2_I(inode)->ip_xattr_sem);
 	acl = ocfs2_get_acl_nolock(inode, ACL_TYPE_ACCESS, bh);
 	up_read(&OCFS2_I(inode)->ip_xattr_sem);
-	if (IS_ERR(acl) || !acl)
-		return PTR_ERR(acl);
+	if (IS_ERR_OR_NULL(acl))
+		return PTR_ERR_OR_ZERO(acl);
 	ret = __posix_acl_chmod(&acl, GFP_KERNEL, inode->i_mode);
 	if (ret)
 		return ret;
Signature · 1 line
2.20.1
author maintainer
#132 lore raw

PATCH 129/134 mailbox: imx: Fix Tx doorbell shutdown path

In reply to Sasha Levin

From: Daniel Baluta <[email protected]>

[ Upstream commit bf159d151a0b844be28882f39e316b5800acaa2b ]

Tx doorbell is handled by txdb_tasklet and doesn't have an associated IRQ.

Anyhow, imx_mu_shutdown ignores this and tries to free an IRQ that wasn't requested for Tx DB resulting in the following warning:

[    1.967644] Trying to free already-free IRQ 26
[    1.972108] WARNING: CPU: 2 PID: 157 at kernel/irq/manage.c:1708 __free_irq+0xc0/0x358
[    1.980024] Modules linked in:
[    1.983088] CPU: 2 PID: 157 Comm: kworker/2:1 Tainted: G
[    1.993524] Hardware name: Freescale i.MX8QXP MEK (DT)
[    1.998668] Workqueue: events deferred_probe_work_func
[    2.003812] pstate: 60000085 (nZCv daIf -PAN -UAO)
[    2.008607] pc : __free_irq+0xc0/0x358
[    2.012364] lr : __free_irq+0xc0/0x358
[    2.016111] sp : ffff00001179b7e0
[    2.019422] x29: ffff00001179b7e0 x28: 0000000000000018
[    2.024736] x27: ffff000011233000 x26: 0000000000000004
[    2.030053] x25: 000000000000001a x24: ffff80083bec74d4
[    2.035369] x23: 0000000000000000 x22: ffff80083bec7588
[    2.040686] x21: ffff80083b1fe8d8 x20: ffff80083bec7400
[    2.046003] x19: 0000000000000000 x18: ffffffffffffffff
[    2.051320] x17: 0000000000000000 x16: 0000000000000000
[    2.056637] x15: ffff0000111296c8 x14: ffff00009179b517
[    2.061953] x13: ffff00001179b525 x12: ffff000011142000
[    2.067270] x11: ffff000011129f20 x10: ffff0000105da970
[    2.072587] x9 : 00000000ffffffd0 x8 : 0000000000000194
[    2.077903] x7 : 612065657266206f x6 : ffff0000111e7b09
[    2.083220] x5 : 0000000000000003 x4 : 0000000000000000
[    2.088537] x3 : 0000000000000000 x2 : 00000000ffffffff
[    2.093854] x1 : 28b70f0a2b60a500 x0 : 0000000000000000
[    2.099173] Call trace:
[    2.101618]  __free_irq+0xc0/0x358
[    2.105021]  free_irq+0x38/0x98
[    2.108170]  imx_mu_shutdown+0x90/0xb0
[    2.111921]  mbox_free_channel.part.2+0x24/0xb8
[    2.116453]  mbox_free_channel+0x18/0x28

This bug is present from the beginning of times.

---
 drivers/mailbox/imx-mailbox.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/mailbox/imx-mailbox.c +3 -1
index d28bbd47ff882..afe625e88a5c5 100644
@@ -214,8 +214,10 @@ static void imx_mu_shutdown(struct mbox_chan *chan)
 	struct imx_mu_priv *priv = to_imx_mu_priv(chan->mbox);
 	struct imx_mu_con_priv *cp = chan->con_priv;
 
-	if (cp->type == IMX_MU_TYPE_TXDB)
+	if (cp->type == IMX_MU_TYPE_TXDB) {
 		tasklet_kill(&cp->txdb_tasklet);
+		return;
+	}
 
 	switch (cp->type) {
 	case IMX_MU_TYPE_TX:
Signature · 1 line
2.20.1
author maintainer
#133 lore raw

PATCH 130/134 s390: disable preemption when switching to nodat stack with CALL_ON_STACK

In reply to Sasha Levin

From: Vasily Gorbik <[email protected]>

[ Upstream commit 7f28dad395243c5026d649136823bbc40029a828 ]

Make sure preemption is disabled when temporary switching to nodat stack with CALL_ON_STACK helper, because nodat stack is per cpu.

---
 arch/s390/kernel/machine_kexec.c |  2 ++
 arch/s390/mm/maccess.c           | 12 +++++++++---
 2 files changed, 11 insertions(+), 3 deletions(-)
arch/s390/kernel/machine_kexec.c +2 -0
index 444a19125a815..dcaadceaf6efc 100644
@@ -164,7 +164,9 @@ static bool kdump_csum_valid(struct kimage *image)
 #ifdef CONFIG_CRASH_DUMP
 	int rc;
 
+	preempt_disable();
 	rc = CALL_ON_STACK(do_start_kdump, S390_lowcore.nodat_stack, 1, image);
+	preempt_enable();
 	return rc == 0;
 #else
 	return false;
arch/s390/mm/maccess.c +9 -3
index 1864a8bb9622f..88d5fcf67a2f1 100644
@@ -115,9 +115,15 @@ static unsigned long _memcpy_real(unsigned long dest, unsigned long src,
  */
 int memcpy_real(void *dest, void *src, size_t count)
 {
-	if (S390_lowcore.nodat_stack != 0)
-		return CALL_ON_STACK(_memcpy_real, S390_lowcore.nodat_stack,
-				     3, dest, src, count);
+	int rc;
+
+	if (S390_lowcore.nodat_stack != 0) {
+		preempt_disable();
+		rc = CALL_ON_STACK(_memcpy_real, S390_lowcore.nodat_stack, 3,
+				   dest, src, count);
+		preempt_enable();
+		return rc;
+	}
 	/*
 	 * This is a really early memcpy_real call, the stacks are
 	 * not set up yet. Just call _memcpy_real on the early boot
Signature · 1 line
2.20.1
author maintainer
#134 lore raw

PATCH 131/134 selftests: vm: add fragment CONFIG_TEST_VMALLOC

In reply to Sasha Levin

From: Anders Roxell <[email protected]>

[ Upstream commit 746dd4012d215b53152f0001a48856e41ea31730 ]

When running test_vmalloc.sh smoke the following print out states that the fragment is missing.

# ./test_vmalloc.sh: You must have the following enabled in your kernel:
# CONFIG_TEST_VMALLOC=m

Rework to add the fragment 'CONFIG_TEST_VMALLOC=m' to the config file.

Fixes a05ef00c9790 ("selftests/vm: add script helper for CONFIG_TEST_VMALLOC_MODULE")
Cc Shuah Khan <[email protected]>
Cc "Uladzislau Rezki (Sony)" <[email protected]>
---
 tools/testing/selftests/vm/config | 1 +
 1 file changed, 1 insertion(+)
tools/testing/selftests/vm/config +1 -0
index 1c0d76cb5adfd..93b90a9b1eebf 100644
@@ -1,2 +1,3 @@
 CONFIG_SYSVIPC=y
 CONFIG_USERFAULTFD=y
+CONFIG_TEST_VMALLOC=m
Signature · 1 line
2.20.1
author maintainer
#135 lore raw

PATCH 132/134 mm/hugetlbfs: fix error handling when setting up mounts

In reply to Sasha Levin

From: Mike Kravetz <[email protected]>

[ Upstream commit 8fc312b32b25c6b0a8b46fab4df8c68df5af1223 ]

It is assumed that the hugetlbfs_vfsmount[] array will contain either a valid vfsmount pointer or NULL for each hstate after initialization. Changes made while converting to use fs_context broke this assumption.

While fixing the hugetlbfs_vfsmount issue, it was discovered that init_hugetlbfs_fs never did correctly clean up when encountering a vfs mount error.

It was found during code inspection. A small memory allocation failure would be the most likely cause of taking a error path with the bug. This is unlikely to happen as this is early init code.

---
 fs/hugetlbfs/inode.c | 31 ++++++++++++++++++++++---------
 1 file changed, 22 insertions(+), 9 deletions(-)
fs/hugetlbfs/inode.c +22 -9
index a478df0356517..26e3906c18fe9 100644
@@ -1461,28 +1461,41 @@ static int __init init_hugetlbfs_fs(void)
 					sizeof(struct hugetlbfs_inode_info),
 					0, SLAB_ACCOUNT, init_once);
 	if (hugetlbfs_inode_cachep == NULL)
-		goto out2;
+		goto out;
 
 	error = register_filesystem(&hugetlbfs_fs_type);
 	if (error)
-		goto out;
+		goto out_free;
 
+	/* default hstate mount is required */
+	mnt = mount_one_hugetlbfs(&hstates[default_hstate_idx]);
+	if (IS_ERR(mnt)) {
+		error = PTR_ERR(mnt);
+		goto out_unreg;
+	}
+	hugetlbfs_vfsmount[default_hstate_idx] = mnt;
+
+	/* other hstates are optional */
 	i = 0;
 	for_each_hstate(h) {
+		if (i == default_hstate_idx)
+			continue;
+
 		mnt = mount_one_hugetlbfs(h);
-		if (IS_ERR(mnt) && i == 0) {
-			error = PTR_ERR(mnt);
-			goto out;
-		}
-		hugetlbfs_vfsmount[i] = mnt;
+		if (IS_ERR(mnt))
+			hugetlbfs_vfsmount[i] = NULL;
+		else
+			hugetlbfs_vfsmount[i] = mnt;
 		i++;
 	}
 
 	return 0;
 
- out:
+ out_unreg:
+	(void)unregister_filesystem(&hugetlbfs_fs_type);
+ out_free:
 	kmem_cache_destroy(hugetlbfs_inode_cachep);
- out2:
+ out:
 	return error;
 }
 fs_initcall(init_hugetlbfs_fs)
Signature · 1 line
2.20.1
author maintainer
#136 lore raw

PATCH 133/134 kernel: sysctl: make drop_caches write-only

In reply to Sasha Levin

From: Johannes Weiner <[email protected]>

[ Upstream commit 204cb79ad42f015312a5bbd7012d09c93d9b46fb ]

Currently, the drop_caches proc file and sysctl read back the last value written, suggesting this is somehow a stateful setting instead of a one-time command. Make it write-only, like e.g. compact_memory.

While mitigating a VM problem at scale in our fleet, there was confusion about whether writing to this file will permanently switch the kernel into a non-caching mode. This influences the decision making in a tense situation, where tens of people are trying to fix tens of thousands of affected machines: Do we need a rollback strategy? What are the performance implications of operating in a non-caching state for several days? It also caused confusion when the kernel team said we may need to write the file several times to make sure it's effective ("But it already reads back 3?").

---
 kernel/sysctl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/sysctl.c +1 -1
index b6f2f35d0bcf5..70665934d53e2 100644
@@ -1466,7 +1466,7 @@ static struct ctl_table vm_table[] = {
 		.procname	= "drop_caches",
 		.data		= &sysctl_drop_caches,
 		.maxlen		= sizeof(int),
-		.mode		= 0644,
+		.mode		= 0200,
 		.proc_handler	= drop_caches_sysctl_handler,
 		.extra1		= SYSCTL_ONE,
 		.extra2		= &four,
Signature · 1 line
2.20.1
author maintainer
#137 lore raw

PATCH 134/134 userfaultfd: require CAP_SYS_PTRACE for UFFD_FEATURE_EVENT_FORK

In reply to Sasha Levin

From: Mike Rapoport <[email protected]>

[ Upstream commit 3c1c24d91ffd536de0a64688a9df7f49e58fadbc ]

A while ago Andy noticed
(http://lkml.kernel.org/r/CALCETrWY+5ynDct7eU_nDUqx=okQvjm=Y5wJvA4ahBja=CQXGw@mail.gmail.com)
that UFFD_FEATURE_EVENT_FORK used by an unprivileged user may have security implications.

As the first step of the solution the following patch limits the availably of UFFD_FEATURE_EVENT_FORK only for those having CAP_SYS_PTRACE.

The usage of CAP_SYS_PTRACE ensures compatibility with CRIU.

Yet, if there are other users of non-cooperative userfaultfd that run without CAP_SYS_PTRACE, they would be broken :(

Current implementation of UFFD_FEATURE_EVENT_FORK modifies the file descriptor table from the read() implementation of uffd, which may have security implications for unprivileged use of the userfaultfd.

Limit availability of UFFD_FEATURE_EVENT_FORK only for callers that have CAP_SYS_PTRACE.

---
 fs/userfaultfd.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)
fs/userfaultfd.c +11 -7
index f9fd18670e22d..d99d166fd8926 100644
@@ -1834,13 +1834,12 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
 	if (copy_from_user(&uffdio_api, buf, sizeof(uffdio_api)))
 		goto out;
 	features = uffdio_api.features;
-	if (uffdio_api.api != UFFD_API || (features & ~UFFD_API_FEATURES)) {
-		memset(&uffdio_api, 0, sizeof(uffdio_api));
-		if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api)))
-			goto out;
-		ret = -EINVAL;
-		goto out;
-	}
+	ret = -EINVAL;
+	if (uffdio_api.api != UFFD_API || (features & ~UFFD_API_FEATURES))
+		goto err_out;
+	ret = -EPERM;
+	if ((features & UFFD_FEATURE_EVENT_FORK) && !capable(CAP_SYS_PTRACE))
+		goto err_out;
 	/* report all available features and ioctls to userland */
 	uffdio_api.features = UFFD_API_FEATURES;
 	uffdio_api.ioctls = UFFD_API_IOCTLS;
@@ -1853,6 +1852,11 @@ static int userfaultfd_api(struct userfaultfd_ctx *ctx,
 	ret = 0;
 out:
 	return ret;
+err_out:
+	memset(&uffdio_api, 0, sizeof(uffdio_api));
+	if (copy_to_user(buf, &uffdio_api, sizeof(uffdio_api)))
+		ret = -EFAULT;
+	goto out;
 }
 
 static long userfaultfd_ioctl(struct file *file, unsigned cmd,
Signature · 1 line
2.20.1