kernel.git/arch/xtensa/kernel/entry.S, branch linux-5.7.y

xtensa: add missing exclusive access state management

2020-08-21T11:07:23Z

commit a0fc1436f1f4f84e93144480bf30e0c958d135b6 upstream. The result of the s32ex opcode is recorded in the ATOMCTL special register and must be retrieved with the getex opcode. Context switch between s32ex and getex may trash the ATOMCTL register and result in duplicate update or missing update of the atomic variable. Add atomctl8 field to the struct thread_info and use getex to swap ATOMCTL bit 8 as a part of context switch. Clear exclusive access monitor on kernel entry. Cc: stable@vger.kernel.org Fixes: f7c34874f04a ("xtensa: add exclusive atomics support") Signed-off-by: Max Filippov Signed-off-by: Greg Kroah-Hartman

Merge tag 'xtensa-20200206' of git://github.com/jcmvbkbc/linux-xtensa

2020-02-07T20:22:29Z

Pull xtensa updates from Max Filippov: - reorganize exception vectors placement - small cleanups (drop unused functions/headers/defconfig entries, spelling fixes) * tag 'xtensa-20200206' of git://github.com/jcmvbkbc/linux-xtensa: xtensa: ISS: improve simcall assembly xtensa: reorganize vectors placement xtensa: separate SMP and XIP support xtensa: move fast exception handlers close to vectors arch/xtensa: fix Kconfig typos for HAVE_SMP xtensa: clean up optional XCHAL_* definitions xtensa: drop unused function fast_coprocessor_double xtensa: drop empty platform_* functions from platforms xtensa: clean up platform headers xtensa: drop set_except_vector declaration xtensa: configs: Cleanup old Kconfig IO scheduler options

xtensa: move fast exception handlers close to vectors

2020-02-05T05:53:39Z

On XIP kernels it makes sense to have exception vectors and fast exception handlers together (in a fast memory). In addition, with MTD XIP support both vectors and fast exception handlers must be outside of the FLASH. Add section .exception.text and move fast exception handlers to it. Put it together with vectors when vectors are outside of the .text. Signed-off-by: Max Filippov

sched/rt, xtensa: Use CONFIG_PREEMPTION

2019-12-08T13:37:35Z

CONFIG_PREEMPTION is selected by CONFIG_PREEMPT and by CONFIG_PREEMPT_RT. Both PREEMPT and PREEMPT_RT require the same functionality which today depends on CONFIG_PREEMPT. Switch the entry code over to use CONFIG_PREEMPTION. Add PREEMPT_RT output to die(). [bigeasy: +traps.c] Signed-off-by: Thomas Gleixner Signed-off-by: Sebastian Andrzej Siewior Signed-off-by: Thomas Gleixner Acked-by: Max Filippov Cc: Chris Zankel Cc: Linus Torvalds Cc: Peter Zijlstra Cc: linux-xtensa@linux-xtensa.org Link: https://lore.kernel.org/r/20191015191821.11479-21-bigeasy@linutronix.de Signed-off-by: Ingo Molnar

xtensa: clean up system_call/xtensa_rt_sigreturn interaction

2019-11-30T03:37:12Z

system_call assembly code always pushes pointer to struct pt_regs as the last additional parameter for all system calls. The only user of this feature is xtensa_rt_sigreturn. Avoid this special case. Define xtensa_rt_sigreturn as accepting no argiments. Use current_pt_regs to get pointer to struct pt_regs in xtensa_rt_sigreturn. Don't pass additional parameter from system_call code. Signed-off-by: Max Filippov

xtensa: fix system_call interaction with ptrace

2019-11-29T23:47:54Z

Don't overwrite return value if system call was cancelled at entry by ptrace. Return status code from do_syscall_trace_enter so that pt_regs::syscall doesn't need to be changed to skip syscall. Signed-off-by: Max Filippov

xtensa: rearrange syscall tracing

2019-11-29T22:39:12Z

system_call saves and restores syscall number across system call to make clone and execv entry and exit tracing match. This complicates things when syscall code may be changed by ptrace. Preserve syscall code in copy_thread and start_thread directly instead of doing tricks in system_call. Signed-off-by: Max Filippov

xtensa: entry: Remove unneeded need_resched() loop

2019-11-26T19:33:39Z

Since the enabling and disabling of IRQs within preempt_schedule_irq() is contained in a need_resched() loop, we don't need the outer arch code loop. Acked-by: Max Filippov Signed-off-by: Valentin Schneider Message-Id: <20190923143620.29334-10-valentin.schneider@arm.com> Signed-off-by: Max Filippov

xtensa: add support for call0 ABI in userspace

2019-09-01T20:11:57Z

Provide a Kconfig choice to select whether only the default ABI, only call0 ABI or both are supported. The default for XEA2 is windowed, but it may change for XEA3. Call0 only runs userspace with PS.WOE disabled. Supporting both windowed and call0 ABIs is tricky, as there's no indication in the ELF binaries which ABI they use. So it is done by probing: each process is started with PS.WOE disabled, but the handler of an illegal instruction exception taken with PS.WOE retries faulting instruction after enabling PS.WOE. It must happen before any signal is delivered to the process, otherwise it may be delivered incorrectly. Signed-off-by: Max Filippov

xtensa: clean up PS_WOE_BIT usage

2019-09-01T20:11:57Z

PS_WOE_BIT is mainly used to generate PS.WOE mask in the code. Introduce PS_WOE_MASK macro and use it instead. Signed-off-by: Max Filippov