linux/drivers/tty/serial, branch v4.4

Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc

2015-12-25T21:15:23Z

Pull sparc fixes from David Miller: 1) Finally make perf stack backtraces stable on sparc, several problems (mostly due to the context in which the user copies from the stack are done) contributed to this. From Rob Gardner. 2) Export ADI capability if the cpu supports it. 3) Hook up userfaultfd system call. 4) When faults happen during user copies we really have to clean up and restore the FPU state fully. Also from Rob Gardner * git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc: tty/serial: Skip 'NULL' char after console break when sysrq enabled sparc64: fix FP corruption in user copy functions sparc64: Perf should save/restore fault info sparc64: Ensure perf can access user stacks sparc64: Don't set %pil in rtrap_nmi too early sparc64: Add ADI capability to cpu capabilities tty: serial: constify sunhv_ops structs sparc: Hook up userfaultfd system call

tty/serial: Skip 'NULL' char after console break when sysrq enabled

2015-12-24T17:13:37Z

When sysrq is triggered from console, serial driver for SUN hypervisor console receives a console break and enables the sysrq. It expects a valid sysrq char following with break. Meanwhile if driver receives 'NULL' ASCII char then it disables sysrq and sysrq handler will never be invoked. This fix skips calling uart sysrq handler when 'NULL' is received while sysrq is enabled. Signed-off-by: Vijay Kumar Acked-by: Karl Volz Signed-off-by: David S. Miller

tty: serial: constify sunhv_ops structs

2015-12-24T17:04:18Z

Constifies sunhv_ops structures in tty's serial driver since they are not modified after their initialization. Detected and found using Coccinelle. Suggested-by: Julia Lawall Signed-off-by: Aya Mahfouz Signed-off-by: David S. Miller

serial: earlycon: Add missing spinlock initialization

2015-12-13T07:05:28Z

If an earlycon console driver needs to acquire the uart_port.lock spinlock for serial console output, and CONFIG_DEBUG_SPINLOCK=y: BUG: spinlock bad magic on CPU#0, swapper/0 lock: sci_ports+0x0/0x3480, .magic: 00000000, .owner: /-1, .owner_cpu: 0 CPU: 0 PID: 0 Comm: swapper Not tainted 4.4.0-rc2-koelsch-g62ea5edf143bb1d0-dirty #2083 Hardware name: Generic R8A7791 (Flattened Device Tree) [] (unwind_backtrace) from [] (show_stack+0x10/0x14) [] (show_stack) from [] (dump_stack+0x70/0x8c) [] (dump_stack) from [] (do_raw_spin_lock+0x20/0x190) [] (do_raw_spin_lock) from [] (serial_console_write+0x4c/0x130) [] (serial_console_write) from [] (call_console_drivers.constprop.13+0xc8/0xec) [] (call_console_drivers.constprop.13) from [] (console_unlock+0x354/0x440) [] (console_unlock) from [] (register_console+0x2a0/0x394) [] (register_console) from [] (of_setup_earlycon+0x90/0xa4) [] (of_setup_earlycon) from [] (setup_of_earlycon+0x118/0x13c) [] (setup_of_earlycon) from [] (do_early_param+0x64/0xb4) [] (do_early_param) from [] (parse_args+0x254/0x350) [] (parse_args) from [] (parse_early_options+0x2c/0x3c) [] (parse_early_options) from [] (parse_early_param+0x2c/0x40) [] (parse_early_param) from [] (setup_arch+0x520/0xaf0) [] (setup_arch) from [] (start_kernel+0x94/0x370) [] (start_kernel) from [<40008090>] (0x40008090) Initialize the spinlock in of_setup_earlycon() and register_earlycon(), to fix this for both DT-based and legacy earlycon. If the driver would reinitialize the spinlock again, this is harmless, as it's allowed to reinitialize an unlocked spinlock. Alternatives are: - Drivers having an early_serial_console_write() that only performs the core functionality of serial_console_write(), without acquiring the lock (which may be unsafe, depending on the hardware), - Drivers initializing the spinlock in their private earlycon setup functions. As uart_port is owned by generic serial_core, and uart_port.lock is initialized by uart_add_one_port() for the normal case, this can better be handled in the earlycon core. Signed-off-by: Geert Uytterhoeven Reviewed-by: Peter Hurley Reported-by: Bjorn Andersson Signed-off-by: Greg Kroah-Hartman

serial: sh-sci: Fix length of scatterlist

2015-12-13T05:42:31Z

This patch fixes an issue that the "length" of scatterlist should be set using sg_dma_len(). Otherwise, a dmaengine driver cannot work correctly if CONFIG_NEED_SG_DMA_LENGTH=y. Fixes: 7b39d90184 (serial: sh-sci: Fix NULL pointer dereference if HIGHMEM is enabled) Signed-off-by: Yoshihiro Shimoda Signed-off-by: Geert Uytterhoeven Acked-by: Simon Horman Signed-off-by: Greg Kroah-Hartman

serial: 8250_uniphier: fix dl_read and dl_write functions

2015-12-13T05:42:31Z

The register offset must be shifted by regshift, otherwise the baudrate is not set. I missed the issue probably because the divisor register was already set by the boot loader. Fixes: 1a8d2903cb6a ("serial: 8250_uniphier: add UniPhier serial driver") Signed-off-by: Masahiro Yamada Signed-off-by: Greg Kroah-Hartman

serial: export fsl8250_handle_irq

2015-11-21T00:19:54Z

fsl8250_handle_irq is now used by the of_serial driver, and that fails if it is a loadable module: ERROR: "fsl8250_handle_irq" [drivers/tty/serial/of_serial.ko] undefined! This exports the symbol to avoid randconfig errors. Signed-off-by: Arnd Bergmann Fixes: d43b54d269d2 ("serial: Enable Freescale 16550 workaround on arm") Cc: Scott Wood Signed-off-by: Jeff Mahoney Signed-off-by: Greg Kroah-Hartman

serial: 8250_mid: Add missing dependency

2015-11-21T00:19:54Z

8250_mid uses rational_best_approximation() function, so the driver needs to select CONFIG_RATIONAL option. This fixes build error when CONFIG_RATIONAL is not enabled: drivers/built-in.o: In function `mid8250_set_termios': 8250_mid.c:(.text+0x10169a): undefined reference to `rational_best_approximation' Reported-by: Randy Dunlap Signed-off-by: Heikki Krogerus Acked-by: Andy Shevchenko Signed-off-by: Arnd Bergmann Signed-off-by: Greg Kroah-Hartman

serial: etraxfs-uart: Fix crash

2015-11-21T00:19:54Z

Since commit 7d8c70d8048c ("serial: mctrl-gpio: rename init function"), crisv32 either do not build or crash as follows. Unable to handle kernel NULL pointer dereference Linux 4.3.0-rc7-next-20151101 #1 Sun Nov 1 11:41:28 PST 2015 ... Call Trace: [] show_stack+0x0/0x9e [] printk+0x0/0x2c [] show_registers+0x14a/0x1c2 [] printk+0x0/0x2c [] die_if_kernel+0x7c/0x9e [] do_page_fault+0x32e/0x3e6 [] of_get_property+0x0/0x2c [] of_irq_parse_raw+0x12a/0x376 [] of_get_property+0x0/0x2c [] get_page_from_freelist+0x73e/0x856 [] of_get_property+0x0/0x2c [] d_mmu_refill+0x10a/0x112 [] devm_kmalloc+0x40/0x56 [] add_dr+0xc/0x1c [] devm_add_action+0x2/0x4e [] mctrl_gpio_init_noauto+0x1c/0x76 [] mctrl_gpio_init+0x22/0x110 The function call in the etraxfs-uart driver was not renamed, possibly due to interference with commit 7b9c5162c182 ("serial: etraxfs-uart: use mctrl_gpio helpers for handling modem signals"). Fixes: 7d8c70d8048c ("serial: mctrl-gpio: rename init function") Signed-off-by: Guenter Roeck Acked-by: Uwe Kleine-König Acked-by: Niklas Cassel Signed-off-by: Greg Kroah-Hartman

serial: fsl_lpuart: Fix earlycon support

2015-11-21T00:19:54Z

Earlycon support for Freescale lpuart should only be enabled when console support is enabled. Fixes: 1d59b382f1c4 ("serial: fsl_lpuart: add earlycon support") Acked-by: Stefan Agner Signed-off-by: Peter Hurley Acked-by: Arnd Bergmann Signed-off-by: Greg Kroah-Hartman