aboutsummaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2018-06-28serdev: fix memleak on module unloadJohan Hovold1-0/+1
Make sure to free all resources associated with the ida on module exit. Fixes: cd6484e1830b ("serdev: Introduce new bus for serial attached devices") Cc: stable <[email protected]> # 4.11 Signed-off-by: Johan Hovold <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28serial: 8250_pci: Remove stalled entries in blacklistAndy Shevchenko1-2/+0
After the commit 7d8905d06405 ("serial: 8250_pci: Enable device after we check black list") pure serial multi-port cards, such as CH355, got blacklisted and thus not being enumerated anymore. Previously, it seems, blacklisting them was on purpose to shut up pciserial_init_one() about record duplication. So, remove the entries from blacklist in order to get cards enumerated. Fixes: 7d8905d06405 ("serial: 8250_pci: Enable device after we check black list") Reported-by: Matt Turner <[email protected]> Cc: Sergej Pupykin <[email protected]> Cc: Alexandr Petrenko <[email protected]> Signed-off-by: Andy Shevchenko <[email protected]> Reviewed-and-Tested-by: Matt Turner <[email protected]> Cc: stable <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28n_tty: Access echo_* variables carefully.Tetsuo Handa1-18/+24
syzbot is reporting stalls at __process_echoes() [1]. This is because since ldata->echo_commit < ldata->echo_tail becomes true for some reason, the discard loop is serving as almost infinite loop. This patch tries to avoid falling into ldata->echo_commit < ldata->echo_tail situation by making access to echo_* variables more carefully. Since reset_buffer_flags() is called without output_lock held, it should not touch echo_* variables. And omit a call to reset_buffer_flags() from n_tty_open() by using vzalloc(). Since add_echo_byte() is called without output_lock held, it needs memory barrier between storing into echo_buf[] and incrementing echo_head counter. echo_buf() needs corresponding memory barrier before reading echo_buf[]. Lack of handling the possibility of not-yet-stored multi-byte operation might be the reason of falling into ldata->echo_commit < ldata->echo_tail situation, for if I do WARN_ON(ldata->echo_commit == tail + 1) prior to echo_buf(ldata, tail + 1), the WARN_ON() fires. Also, explicitly masking with buffer for the former "while" loop, and use ldata->echo_commit > tail for the latter "while" loop. [1] https://syzkaller.appspot.com/bug?id=17f23b094cd80df750e5b0f8982c521ee6bcbf40 Signed-off-by: Tetsuo Handa <[email protected]> Reported-by: syzbot <[email protected]> Cc: Peter Hurley <[email protected]> Cc: stable <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28n_tty: Fix stall at n_tty_receive_char_special().Tetsuo Handa1-5/+8
syzbot is reporting stalls at n_tty_receive_char_special() [1]. This is because comparison is not working as expected since ldata->read_head can change at any moment. Mitigate this by explicitly masking with buffer size when checking condition for "while" loops. [1] https://syzkaller.appspot.com/bug?id=3d7481a346958d9469bebbeb0537d5f056bdd6e8 Signed-off-by: Tetsuo Handa <[email protected]> Reported-by: syzbot <[email protected]> Fixes: bc5a5e3f45d04784 ("n_tty: Don't wrap input buffer indices at buffer size") Cc: stable <[email protected]> Cc: Peter Hurley <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28swiotlb: export swiotlb_dma_opsChristoph Hellwig1-0/+1
For architectures that do not use per-device dma ops we need to export the dma_map_ops structure returned from get_arch_dma_ops(). Fixes: 10314e09 ("riscv: add swiotlb support") Signed-off-by: Christoph Hellwig <[email protected]> Reported-by: Andreas Schwab <[email protected]>
2018-06-28usb: typec: function for checking cable plug orientationHeikki Krogerus2-0/+13
This adds function typec_get_orientation() that can be used for checking the current cable plug orientation. Signed-off-by: Heikki Krogerus <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: pd: include kernel.hHeikki Krogerus1-0/+1
This makes life a bit easier for the drivers that include pd.h. All pd_header_*_le() inline functions defined in pd.h call le16_to_cpu(), and all *_LE() macros in pd.h call cpu_to_le16(). Signed-off-by: Heikki Krogerus <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28staging: typec: tcpci: move tcpci drivers out of stagingLi Jun10-32/+17
Move TCPCI(Typec port controller interface) driver and rt1711h driver out of staging. Signed-off-by: Li Jun <[email protected]> Reviewed-by: Guenter Roeck <[email protected]> Acked-by: Heikki Krogerus <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28staging: typec: tcpci: Only touch target bit when enable vconnLi Jun1-6/+3
We need regmap_update_bits to avoid touch any other bits when enable or disable vconn. Reviewed-by: Guenter Roeck <[email protected]> Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28staging: typec: tcpci: keep the disconnected cc line openLi Jun1-4/+14
While set polarity, we should keep the disconnected cc line to be open. Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Reviewed-by: Guenter Roeck <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28typec: tcpm: add starting value for drp togglingLi Jun1-5/+5
As DRP port autonomously toggles the Rp/Rd need a start value to begin with, so add one parameter for it in tcpm_start_drp_toggling. Reviewed-by: Guenter Roeck <[email protected]> Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28staging: typec: tcpci: enable vbus detectionLi Jun1-0/+6
TCPCI implementation may need SW to enable VBUS detection to generate power status events. Reviewed-by: Guenter Roeck <[email protected]> Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28staging: typec: tcpci: register port before request irqPeter Chen1-6/+9
With that we can clear any pending events and the port is registered so driver can be ready to handle typec events once we request irq. Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Peter Chen <[email protected]> Signed-off-by: Li Jun <[email protected]> Reviewed-by: Guenter Roeck <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28staging: typec: tcpci: use IS_ERR() instead of PTR_ERR_OR_ZERO()Li Jun1-2/+2
As tcpm_register_port() and tcpci_register_port() never return NULL and NULL is not a success in this case, use IS_ERR() to check the return value of both. Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Reviewed-by: Guenter Roeck <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28staging: typec: tcpci: remove unused tcpci_tcpc_configLi Jun1-7/+0
Since we will use config settings via device properties, so remove the hard code tcpci_tcpc_config. Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Reviewed-by: Guenter Roeck <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: typec: tcpm: support get typec and pd config from device propertiesLi Jun1-22/+109
This patch adds support of get typec and power delivery config from firmware description. Reviewed-by: Heikki Krogerus <[email protected]> Reviewed-by: Guenter Roeck <[email protected]> Signed-off-by: Li Jun <[email protected]> Acked-by: Adam Thomson <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: typec: add API to get typec basic port power and data configLi Jun2-4/+57
This patch adds 3 APIs to get the typec port power and data type, and preferred power role by its name string. Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: typec: add fwnode to tcpcLi Jun3-0/+10
Add fwnode handle to get the fwnode so we can get typec configs it contains. Suggested-by: Heikki Krogerus <[email protected]> Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Reviewed-by: Guenter Roeck <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28staging: typec: tcpci: add compatible string for nxp ptn5110Li Jun1-1/+1
Add nxp ptn5110 typec controller compatible string: "nxp,ptn5110", which is a standard tcpci chip with power delivery support. Meanwhile remove "usb,tcpci" because it doesn't follow the binding format rule and has not been used yet. Reviewed-by: Heikki Krogerus <[email protected]> Signed-off-by: Li Jun <[email protected]> Reviewed-by: Guenter Roeck <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28dt-bindings: usb: add documentation for typec port controller(TCPCI)Li Jun1-0/+49
TCPCI stands for typec port controller interface, its implementation has full typec port control with power delivery support, it's a standard i2c slave with GPIO input as irq interface, detail see spec "Universal Serial Bus Type-C Port Controller Interface Specification Revision 1.0, Version 1.1" Reviewed-by: Rob Herring <[email protected]> Signed-off-by: Li Jun <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28dt-bindings: connector: add properties for typecLi Jun2-0/+106
Add bindings supported by current typec driver, so user can pass all those properties via dt. Reviewed-by: Rob Herring <[email protected]> Signed-off-by: Li Jun <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: typec: tps6598x: Remove VLA usageKees Cook1-1/+10
In the quest to remove all stack VLA usage from the kernel[1], this uses the maximum buffer size and adds a sanity check. While 25 bytes is the size of the largest current things coming through, Heikki Krogerus pointed out that the actual max in 64 bytes, as per ch 1.3.2 http://www.ti.com/lit/ug/slvuan1a/slvuan1a.pdf [1] https://lkml.kernel.org/r/CA+55aFzCG-zNmZwX4A2FQpadafLfEzK6CC=qPXydAacU1RqZWA@mail.gmail.com Signed-off-by: Kees Cook <[email protected]> Acked-by: Heikki Krogerus <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: cdc-wdm: use irqsave() in USB's complete callbackSebastian Andrzej Siewior1-6/+10
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Cc: Oliver Neukum <[email protected]> Cc: "Bjørn Mork" <[email protected]> Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28NFC: nfcmrvl_usb: use irqsave() in USB's complete callbackSebastian Andrzej Siewior1-2/+3
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Cc: Samuel Ortiz <[email protected]> Cc: [email protected] Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28USB: cdc-wdm: don't enable interrupts in USB-givebackSebastian Andrzej Siewior1-7/+24
In the code path __usb_hcd_giveback_urb() -> wdm_in_callback() -> service_outstanding_interrupt() The function service_outstanding_interrupt() will unconditionally enable interrupts during unlock and invoke usb_submit_urb() with GFP_KERNEL. If the HCD completes in BH (like ehci does) then the context remains atomic due local_bh_disable() and enabling interrupts does not change this. Defer the error case handling to a workqueue as suggested by Oliver Neukum. In case of an error the worker performs the read out and wakes the user. Fixes: c1da59dad0eb ("cdc-wdm: Clear read pipeline in case of error") Cc: Robert Foss <[email protected]> Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Acked-by: Oliver Neukum <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: legousbtower: use irqsave() in USB's complete callbackSebastian Andrzej Siewior1-2/+3
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Cc: Juergen Stuber <[email protected]> Cc: [email protected] Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: ldusb: use irqsave() in USB's complete callbackSebastian Andrzej Siewior1-3/+4
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: iowarrior: remove intr_idx_lockSebastian Andrzej Siewior1-5/+0
The intr_idx_lock lock is acquired only in the completion callback of the ->int_in_urb (iowarrior_callback()). There is only one URB that is scheduled / completed so there can't be more than one user of the lock. The comment says that it protects ->intr_idx and the callback is the only place in driver that writes to it. Remove the intr_idx_lock lock because it is superfluous. Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: adutux: use irqsave() in USB's complete callbackSebastian Andrzej Siewior1-4/+6
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: wusbcore: remove excessive irqsaveJohn Ogness1-4/+4
wa_urb_dequeue() locks multiple spinlocks while disabling interrupts: spin_lock_irqsave(&lock1, flags); spin_lock_irqsave(&lock2, flags2); Obviously there is no need for the second irqsave and "flags2" variable since interrupts are disabled at that point. Remove the second irqsave: spin_lock_irqsave(&lock1, flags); spin_lock(&lock2); and eliminate the flags2 variable. Signed-off-by: John Ogness <[email protected]> Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: usb-skeleton: use irqsave() in USB's complete callbackJohn Ogness1-4/+6
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Signed-off-by: John Ogness <[email protected]> Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: usblp: use irqsave() in USB's complete callbackSebastian Andrzej Siewior1-4/+6
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Acked-by: Pete Zaitcev <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: usbfs: use irqsave() in USB's complete callbackSebastian Andrzej Siewior1-2/+3
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Acked-by: Alan Stern <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: core: use irqsave() in sg_complete() complete callbackSebastian Andrzej Siewior1-4/+5
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Acked-by: Alan Stern <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28usb: cdc-acm: use irqsave() in USB's complete callbackSebastian Andrzej Siewior1-2/+3
The USB completion callback does not disable interrupts while acquiring the lock. We want to remove the local_irq_disable() invocation from __usb_hcd_giveback_urb() and therefore it is required for the callback handler to disable the interrupts while acquiring the lock. The callback may be invoked either in IRQ or BH context depending on the USB host controller. Use the _irqsave() variant of the locking primitives. Cc: Oliver Neukum <[email protected]> Signed-off-by: Sebastian Andrzej Siewior <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28USB: host: ehci-npcm7xx: Fix some error codes in probeDan Carpenter1-2/+2
We accidentally return 1 instead of negative error codes. Fixes: df44831ee2dd ("USB host: Add USB ehci support for nuvoton npcm7xx platform") Signed-off-by: Dan Carpenter <[email protected]> Signed-off-by: Avi Fishman <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2018-06-28Merge tag 'usb-serial-4.18-rc3' of ↵Greg Kroah-Hartman1-0/+14
https://git.kernel.org/pub/scm/linux/kernel/git/johan/usb-serial into usb-linus Johan writes: USB-serial fixes for v4.18-rc3 Here are bunch of new device ids for cp210x. All have been in linux-next with no reported issues. Signed-off-by: Johan Hovold <[email protected]>
2018-06-28Btrfs: fix mount failure when qgroup rescan is in progressFilipe Manana1-3/+10
If a power failure happens while the qgroup rescan kthread is running, the next mount operation will always fail. This is because of a recent regression that makes qgroup_rescan_init() incorrectly return -EINVAL when we are mounting the filesystem (through btrfs_read_qgroup_config()). This causes the -EINVAL error to be returned regardless of any qgroup flags being set instead of returning the error only when neither of the flags BTRFS_QGROUP_STATUS_FLAG_RESCAN nor BTRFS_QGROUP_STATUS_FLAG_ON are set. A test case for fstests follows up soon. Fixes: 9593bf49675e ("btrfs: qgroup: show more meaningful qgroup_rescan_init error message") Signed-off-by: Filipe Manana <[email protected]> Reviewed-by: Qu Wenruo <[email protected]> Signed-off-by: David Sterba <[email protected]>
2018-06-28Btrfs: fix regression in btrfs_page_mkwrite() from vm_fault_t conversionChris Mason1-1/+2
The vm_fault_t conversion commit introduced a ret2 variable for tracking the integer return values from internal btrfs functions. It was sometimes returning VM_FAULT_LOCKED for pages that were actually invalid and had been removed from the radix. Something like this: ret2 = btrfs_delalloc_reserve_space() // returns zero on success lock_page(page) if (page->mapping != inode->i_mapping) goto out_unlock; ... out_unlock: if (!ret2) { ... return VM_FAULT_LOCKED; } This ends up triggering this WARNING in btrfs_destroy_inode() WARN_ON(BTRFS_I(inode)->block_rsv.size); xfstests generic/095 was able to reliably reproduce the errors. Since out_unlock: is only used for errors, this fix moves it below the if (!ret2) check we use to return VM_FAULT_LOCKED for success. Fixes: a528a2415087 (btrfs: change return type of btrfs_page_mkwrite to vm_fault_t) Signed-off-by: Chris Mason <[email protected]> Reviewed-by: David Sterba <[email protected]> Signed-off-by: David Sterba <[email protected]>
2018-06-28btrfs: quota: Set rescan progress to (u64)-1 if we hit last leafQu Wenruo1-1/+3
Commit ff3d27a048d9 ("btrfs: qgroup: Finish rescan when hit the last leaf of extent tree") added a new exit for rescan finish. However after finishing quota rescan, we set fs_info->qgroup_rescan_progress to (u64)-1 before we exit through the original exit path. While we missed that assignment of (u64)-1 in the new exit path. The end result is, the quota status item doesn't have the same value. (-1 vs the last bytenr + 1) Although it doesn't affect quota accounting, it's still better to keep the original behavior. Reported-by: Misono Tomohiro <[email protected]> Fixes: ff3d27a048d9 ("btrfs: qgroup: Finish rescan when hit the last leaf of extent tree") Signed-off-by: Qu Wenruo <[email protected]> Reviewed-by: Misono Tomohiro <[email protected]> Signed-off-by: David Sterba <[email protected]>
2018-06-28Merge branch 'drm-fixes-4.18' of git://people.freedesktop.org/~agd5f/linux ↵Dave Airlie1-4/+6
into drm-fixes Single amdgpu regression fix for stable. Signed-off-by: Dave Airlie <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2018-06-28Merge branch 'for-upstream/malidp-fixes' of git://linux-arm.org/linux-ld ↵Dave Airlie3-5/+10
into drm-fixes Misc set of malidp fixes. Signed-off-by: Dave Airlie <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2018-06-27proc: add proc_seq_releaseChunyu Hu1-1/+10
kmemleak reported some memory leak on reading proc files. After adding some debug lines, find that proc_seq_fops is using seq_release as release handler, which won't handle the free of 'private' field of seq_file, while in fact the open handler proc_seq_open could create the private data with __seq_open_private when state_size is greater than zero. So after reading files created with proc_create_seq_private, such as /proc/timer_list and /proc/vmallocinfo, the private mem of a seq_file is not freed. Fix it by adding the paired proc_seq_release as the default release handler of proc_seq_ops instead of seq_release. Fixes: 44414d82cfe0 ("proc: introduce proc_create_seq_private") Reviewed-by: Christoph Hellwig <[email protected]> CC: Christoph Hellwig <[email protected]> Signed-off-by: Chunyu Hu <[email protected]> Signed-off-by: Al Viro <[email protected]>
2018-06-27ARM64: dts: meson-gxl: fix Mali GPU compatible stringMartin Blumenstingl1-1/+1
meson-gxl-mali.dtsi is only used on GXL SoCs. Thus it should use the GXL specific compatible string instead of the GXBB one. For now this is purely cosmetic since the (out-of-tree) lima driver for this GPU currently uses the "arm,mali-450" match instead of the SoC specific one. However, update the .dts to match the documentation since this driver behavior might change in the future. Signed-off-by: Martin Blumenstingl <[email protected]> Acked-by: Neil Armstrong <[email protected]> Signed-off-by: Kevin Hilman <[email protected]>
2018-06-27ARM64: dts: meson-axg: fix ethernet stability issueJerome Brunet1-1/+14
Like the odroid-c2 and wetek, the s400 uses the RTL8211F and seems to suffer from the kind of stability issue. Doing an iperf3 download test, we can see a significant number of LPI interrupts on the tx path. After a short while (5 to 15 seconds), the network connection dies. If using rootfs over NFS, the connection may also break during the boot sequence. We still don't have a real explanation for this problem so let's disable EEE once again. Fixes: f6f6ac914b82 ("ARM64: dts: meson-axg: enable ethernet for A113D S400 board") Signed-off-by: Jerome Brunet <[email protected]> Reviewed-by: Neil Armstrong <[email protected]> Signed-off-by: Kevin Hilman <[email protected]>
2018-06-27ARM64: dts: meson-gx: fix ATF reserved memory regionKevin Hilman2-8/+6
Vendor firmware/uboot has different reserved regions depending on firmware version, but current codebase reserves the same regions on GXL and GXBB, so move the additional reserved memory region to common .dtsi. Found when putting a recent vendor u-boot on meson-gxbb-p200. Suggested-by: Neil Armstrong <[email protected]> Cc: [email protected] Signed-off-by: Kevin Hilman <[email protected]>
2018-06-27ARM64: dts: meson-gxl-s905x-p212: Add phy-supply for usb0Neil Armstrong1-0/+7
Like LibreTech-CC, the USB0 needs the 5V regulator to be enabled to power the devices on the P212 Reference Design based boards. Fixes: b9f07cb4f41f ("ARM64: dts: meson-gxl-s905x-p212: enable the USB controller") Signed-off-by: Neil Armstrong <[email protected]> Acked-by: Martin Blumenstingl <[email protected]> Signed-off-by: Kevin Hilman <[email protected]>
2018-06-27ARM64: dts: meson: fix register ranges for SD/eMMCKevin Hilman2-5/+5
Based on updated information from Amlogic, correct the register range for the SD/eMMC blocks to the right size. Reported-by: Yixun Lan <[email protected]> Tested-by: Yixun Lan <[email protected]> Signed-off-by: Kevin Hilman <[email protected]>
2018-06-27ARM64: dts: meson: disable sd-uhs modes on the libretech-ccJerome Brunet1-3/+0
There is a problem with the sd-uhs mode when doing a soft reboot. Switching back from 1.8v to 3.3v messes with the card, which no longer respond (timeout errors). According to the specification, we should perform a card reset (power cycling the card) but this is something we cannot control on this design. Then the only solution to restore the communication with the card is an "unplug-plug" which is not acceptable Until we find a solution, if any, disable the sd-uhs modes on this design. For the people using uhs at the moment, there will a performance drop as a result. Fixes: 3cde63ebc85c ("ARM64: dts: meson-gxl: libretech-cc: enable high speed modes") Signed-off-by: Jerome Brunet <[email protected]> Cc: [email protected] Signed-off-by: Kevin Hilman <[email protected]>
2018-06-27drm/amd/display: release spinlock before committing updates to streamShirish S1-4/+2
Currently, amdgpu_do_flip() spinlocks crtc->dev->event_lock and releases it only after committing updates to the stream. dc_commit_updates_for_stream() should be moved out of spinlock for the below reasons: 1. event_lock is supposed to protect access to acrct->pflip_status _only_ 2. dc_commit_updates_for_stream() has potential sleep's and also its not appropriate to be in an atomic state for such long sequences of code. Signed-off-by: Shirish S <[email protected]> Suggested-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
Powered by cgit, Gruvboxed by Blaster4385.