linux-dev/arch/csky/kernel/entry.S, branch linus/master

irq: csky: perform irqentry in entry code

2021-10-26T09:13:27Z

In preparation for removing HANDLE_DOMAIN_IRQ_IRQENTRY, have arch/csky perform all the irqentry accounting in its entry code. As arch/csky uses GENERIC_IRQ_MULTI_HANDLER, we can use generic_handle_arch_irq() to do so. There should be no functional change as a result of this patch. Signed-off-by: Mark Rutland Reviewed-by: Guo Ren Reviewed-by: Marc Zyngier Cc: Thomas Gleixner

csky: Remove duplicate include in arch/csky/kernel/entry.S

2021-04-20T01:58:29Z

'asm/setup.h' included in 'arch/csky/kernel/entry.S' is duplicated. Signed-off-by: Zhang Yunkai Signed-off-by: Guo Ren

csky: Fixup do_page_fault parent irq status

2021-01-12T01:52:41Z

We must succeed parent's context irq status in page fault handler. Signed-off-by: Guo Ren

csky: Remove prologue of page fault handler in entry.S

2021-01-12T01:52:40Z

There is a prologue on page fault handler which marking pages dirty and/or accessed in page attributes, but all of these have been handled in handle_pte_fault. - Add flush_tlb_one in vmalloc page fault instead of prologue. - Using cmxchg_fixup C codes in do_page_fault instead of ASM one. Signed-off-by: Guo Ren

csky: Add memory layout 2.5G(user):1.5G(kernel)

2021-01-12T01:52:40Z

There are two ways for translating va to pa for csky: - Use TLB(Translate Lookup Buffer) and PTW (Page Table Walk) - Use SSEG0/1 (Simple Segment Mapping) We use tlb mapping 0-2G and 3G-4G virtual address area and SSEG0/1 are for 2G-2.5G and 2.5G-3G translation. We could disable SSEG0 to use 2G-2.5G as TLB user mapping. Signed-off-by: Guo Ren

csky: Add context tracking support

2020-08-01T08:17:51Z

This patch support context tracking with no hz full. Here is the test result with dynticks-testing (see tick_stop): cat /sys/kernel/debug/tracing/per_cpu/cpu0/trace tracer: nop entries-in-buffer/entries-written: 356/356 #P:1 _-----=> irqs-off / _----=> need-resched | / _---=> hardirq/softirq || / _--=> preempt-depth ||| / delay TASK-PID CPU# |||| TIMESTAMP FUNCTION | | | |||| | | ... sleep-192 [000] d.h. 167.088270: hrtimer_expire_entry: hrtimer=(ptrval) function=tick_sched_timer now=166436355700 sleep-192 [000] d.h. 167.092279: hrtimer_expire_entry: hrtimer=(ptrval) function=tick_sched_timer now=166440365700 -0 [000] d.h2 167.096492: hrtimer_expire_entry: hrtimer=(ptrval) function=tick_sched_timer now=166444578400 -0 [000] d..1 167.097876: tick_stop: success=1 dependency=NONE ^^^^^^^^^ -0 [000] d.h1 168.818206: hrtimer_expire_entry: hrtimer=(ptrval) function=tick_sched_timer now=168166280900 kworker/u2:0-7 [000] .... 168.821760: workqueue_execute_start: work struct (ptrval): function wb_workfn kworker/u2:0-7 [000] d.h1 168.824464: hrtimer_expire_entry: hrtimer=(ptrval) function=tick_sched_timer now=168172547100 kworker/0:1-18 [000] .... 168.825053: workqueue_execute_start: work struct (ptrval): function vmstat_update kworker/0:1-18 [000] .... 168.825238: workqueue_execute_start: work struct (ptrval): function vmstat_shepherd kworker/0:1-18 [000] .... 168.825516: workqueue_execute_start: work struct (ptrval): function neigh_periodic_work -0 [000] d..1 168.826121: tick_stop: success=1 dependency=NONE kworker/u2:0-7 [000] .... 169.377327: workqueue_execute_start: work struct (ptrval): function flush_to_ldisc -0 [000] d..1 169.379832: tick_stop: success=1 dependency=NONE kworker/u2:0-7 [000] .... 169.607935: workqueue_execute_start: work struct (ptrval): function flush_to_ldisc kworker/u2:0-7 [000] d.h1 169.608148: hrtimer_expire_entry: hrtimer=(ptrval) function=tick_sched_timer now=168956235500 -0 [000] d..1 169.608654: tick_stop: success=1 dependency=NONE Signed-off-by: Guo Ren Cc: Greentime Hu Cc: Arnd Bergmann

csky: Add SECCOMP_FILTER supported

2020-07-31T01:51:05Z

secure_computing() is called first in syscall_trace_enter() so that a system call will be aborted quickly without doing succeeding syscall tracing if seccomp rules want to deny that system call. TODO: - Update https://github.com/seccomp/libseccomp csky support Signed-off-by: Guo Ren Cc: Arnd Bergmann

csky: Fixup CONFIG_DEBUG_RSEQ

2020-05-28T00:18:36Z

Put the rseq_syscall check point at the prologue of the syscall will break the a0 ... a7. This will casue system call bug when DEBUG_RSEQ is enabled. So move it to the epilogue of syscall, but before syscall_trace. Signed-off-by: Guo Ren

csky: Coding convention in entry.S

2020-05-28T00:18:36Z

There is no fixup or feature in the patch, we only cleanup with: - Remove unnecessary reg used (r11, r12), just use r9 & r10 & syscallid regs as temp useage. - Add _TIF_SYSCALL_WORK and _TIF_WORK_MASK to gather macros. Signed-off-by: Guo Ren

csky: Fixup abiv2 syscall_trace break a4 & a5

2020-05-28T00:18:36Z

Current implementation could destory a4 & a5 when strace, so we need to get them from pt_regs by SAVE_ALL. Signed-off-by: Guo Ren