diff options
author | Anton Blanchard <anton@samba.org> | 2011-11-30 00:23:10 +0000 |
---|---|---|
committer | Benjamin Herrenschmidt <benh@kernel.crashing.org> | 2011-12-08 14:02:22 +1100 |
commit | 9b00ac06978c54788f13eefd34a07b77db48d567 (patch) | |
tree | 4727d71f5d4071b5595d4268c0fe05c8bf2ae02a /arch/powerpc/kernel | |
parent | 58154c8ce71a7854d969d73468fd00e5eeeab708 (diff) | |
download | op-kernel-dev-9b00ac06978c54788f13eefd34a07b77db48d567.zip op-kernel-dev-9b00ac06978c54788f13eefd34a07b77db48d567.tar.gz |
powerpc: Remove broken and complicated kdump system reset code
We have a lot of complicated logic that handles possible recursion between
kdump and a system reset exception. We can solve this in a much simpler
way using the same setjmp/longjmp tricks xmon does.
As a first step, this patch removes the old system reset code.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Diffstat (limited to 'arch/powerpc/kernel')
-rw-r--r-- | arch/powerpc/kernel/crash.c | 87 | ||||
-rw-r--r-- | arch/powerpc/kernel/traps.c | 33 |
2 files changed, 25 insertions, 95 deletions
diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index d879809..3d87b20 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -47,7 +47,6 @@ /* This keeps a track of which one is crashing cpu. */ int crashing_cpu = -1; static cpumask_t cpus_in_crash = CPU_MASK_NONE; -cpumask_t cpus_in_sr = CPU_MASK_NONE; #define CRASH_HANDLER_MAX 3 /* NULL terminated list of shutdown handles */ @@ -55,7 +54,6 @@ static crash_shutdown_t crash_shutdown_handles[CRASH_HANDLER_MAX+1]; static DEFINE_SPINLOCK(crash_handlers_lock); #ifdef CONFIG_SMP -static atomic_t enter_on_soft_reset = ATOMIC_INIT(0); void crash_ipi_callback(struct pt_regs *regs) { @@ -70,23 +68,8 @@ void crash_ipi_callback(struct pt_regs *regs) cpumask_set_cpu(cpu, &cpus_in_crash); /* - * Entered via soft-reset - could be the kdump - * process is invoked using soft-reset or user activated - * it if some CPU did not respond to an IPI. - * For soft-reset, the secondary CPU can enter this func - * twice. 1 - using IPI, and 2. soft-reset. - * Tell the kexec CPU that entered via soft-reset and ready - * to go down. - */ - if (cpumask_test_cpu(cpu, &cpus_in_sr)) { - cpumask_clear_cpu(cpu, &cpus_in_sr); - atomic_inc(&enter_on_soft_reset); - } - - /* * Starting the kdump boot. * This barrier is needed to make sure that all CPUs are stopped. - * If not, soft-reset will be invoked to bring other CPUs. */ while (!cpumask_test_cpu(crashing_cpu, &cpus_in_crash)) cpu_relax(); @@ -103,25 +86,14 @@ void crash_ipi_callback(struct pt_regs *regs) /* NOTREACHED */ } -/* - * Wait until all CPUs are entered via soft-reset. - */ -static void crash_soft_reset_check(int cpu) -{ - unsigned int ncpus = num_online_cpus() - 1;/* Excluding the panic cpu */ - - cpumask_clear_cpu(cpu, &cpus_in_sr); - while (atomic_read(&enter_on_soft_reset) != ncpus) - cpu_relax(); -} - - static void crash_kexec_prepare_cpus(int cpu) { unsigned int msecs; unsigned int ncpus = num_online_cpus() - 1;/* Excluding the panic cpu */ + printk(KERN_EMERG "Sending IPI to other CPUs\n"); + crash_send_ipi(crash_ipi_callback); smp_wmb(); @@ -131,7 +103,6 @@ static void crash_kexec_prepare_cpus(int cpu) * respond. * Delay of at least 10 seconds. */ - printk(KERN_EMERG "Sending IPI to other cpus...\n"); msecs = 10000; while ((cpumask_weight(&cpus_in_crash) < ncpus) && (--msecs > 0)) { cpu_relax(); @@ -140,69 +111,36 @@ static void crash_kexec_prepare_cpus(int cpu) /* Would it be better to replace the trap vector here? */ - /* - * FIXME: In case if we do not get all CPUs, one possibility: ask the - * user to do soft reset such that we get all. - * Soft-reset will be used until better mechanism is implemented. - */ if (cpumask_weight(&cpus_in_crash) < ncpus) { - printk(KERN_EMERG "done waiting: %d cpu(s) not responding\n", + printk(KERN_EMERG "ERROR: %d CPU(s) not responding\n", ncpus - cpumask_weight(&cpus_in_crash)); - printk(KERN_EMERG "Activate soft-reset to stop other cpu(s)\n"); - cpumask_clear(&cpus_in_sr); - atomic_set(&enter_on_soft_reset, 0); - while (cpumask_weight(&cpus_in_crash) < ncpus) - cpu_relax(); } - /* - * Make sure all CPUs are entered via soft-reset if the kdump is - * invoked using soft-reset. - */ - if (cpumask_test_cpu(cpu, &cpus_in_sr)) - crash_soft_reset_check(cpu); - /* Leave the IPI callback set */ + + printk(KERN_EMERG "IPI complete\n"); } /* - * This function will be called by secondary cpus or by kexec cpu - * if soft-reset is activated to stop some CPUs. + * This function will be called by secondary cpus. */ void crash_kexec_secondary(struct pt_regs *regs) { - int cpu = smp_processor_id(); unsigned long flags; - int msecs = 5; + int msecs = 500; local_irq_save(flags); - /* Wait 5ms if the kexec CPU is not entered yet. */ + + /* Wait 500ms for the primary crash CPU to signal its progress */ while (crashing_cpu < 0) { if (--msecs < 0) { - /* - * Either kdump image is not loaded or - * kdump process is not started - Probably xmon - * exited using 'x'(exit and recover) or - * kexec_should_crash() failed for all running tasks. - */ - cpumask_clear_cpu(cpu, &cpus_in_sr); + /* No response, kdump image may not have been loaded */ local_irq_restore(flags); return; } + mdelay(1); cpu_relax(); } - if (cpu == crashing_cpu) { - /* - * Panic CPU will enter this func only via soft-reset. - * Wait until all secondary CPUs entered and - * then start kexec boot. - */ - crash_soft_reset_check(cpu); - cpumask_set_cpu(crashing_cpu, &cpus_in_crash); - if (ppc_md.kexec_cpu_down) - ppc_md.kexec_cpu_down(1, 0); - machine_kexec(kexec_crash_image); - /* NOTREACHED */ - } + crash_ipi_callback(regs); } @@ -225,7 +163,6 @@ static void crash_kexec_prepare_cpus(int cpu) void crash_kexec_secondary(struct pt_regs *regs) { - cpumask_clear(&cpus_in_sr); } #endif /* CONFIG_SMP */ diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c index 9137787..014f88f 100644 --- a/arch/powerpc/kernel/traps.c +++ b/arch/powerpc/kernel/traps.c @@ -162,10 +162,20 @@ int die(const char *str, struct pt_regs *regs, long err) printk("\n"); raw_spin_unlock_irqrestore(&die.lock, flags); - if (kexec_should_crash(current) || - kexec_sr_activated(smp_processor_id())) + /* + * A system reset (0x100) is a request to dump, so we always send + * it through the crashdump code. + */ + if (kexec_should_crash(current) || (TRAP(regs) == 0x100)) { crash_kexec(regs); - crash_kexec_secondary(regs); + + /* + * We aren't the primary crash CPU. We need to send it + * to a holding pattern to avoid it ending up in the panic + * code. + */ + crash_kexec_secondary(regs); + } /* * While our oops output is serialised by a spinlock, output @@ -232,25 +242,8 @@ void system_reset_exception(struct pt_regs *regs) return; } -#ifdef CONFIG_KEXEC - cpumask_set_cpu(smp_processor_id(), &cpus_in_sr); -#endif - die("System Reset", regs, SIGABRT); - /* - * Some CPUs when released from the debugger will execute this path. - * These CPUs entered the debugger via a soft-reset. If the CPU was - * hung before entering the debugger it will return to the hung - * state when exiting this function. This causes a problem in - * kdump since the hung CPU(s) will not respond to the IPI sent - * from kdump. To prevent the problem we call crash_kexec_secondary() - * here. If a kdump had not been initiated or we exit the debugger - * with the "exit and recover" command (x) crash_kexec_secondary() - * will return after 5ms and the CPU returns to its previous state. - */ - crash_kexec_secondary(regs); - /* Must die if the interrupt is not recoverable */ if (!(regs->msr & MSR_RI)) panic("Unrecoverable System Reset"); |