From patchwork Mon Jul 22 15:37:31 2013 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mark Rutland X-Patchwork-Id: 2831434 Return-Path: X-Original-To: patchwork-linux-arm@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.19.201]) by patchwork1.web.kernel.org (Postfix) with ESMTP id D5AEF9F243 for ; Mon, 22 Jul 2013 15:41:21 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id D12B4201F7 for ; Mon, 22 Jul 2013 15:41:17 +0000 (UTC) Received: from casper.infradead.org (casper.infradead.org [85.118.1.10]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id E3E72201EE for ; Mon, 22 Jul 2013 15:41:11 +0000 (UTC) Received: from merlin.infradead.org ([2001:4978:20e::2]) by casper.infradead.org with esmtps (Exim 4.80.1 #2 (Red Hat Linux)) id 1V1ICt-0006mh-9M; Mon, 22 Jul 2013 15:39:28 +0000 Received: from localhost ([::1] helo=merlin.infradead.org) by merlin.infradead.org with esmtp (Exim 4.80.1 #2 (Red Hat Linux)) id 1V1ICN-00074W-4p; Mon, 22 Jul 2013 15:38:55 +0000 Received: from cam-admin0.cambridge.arm.com ([217.140.96.50]) by merlin.infradead.org with esmtp (Exim 4.80.1 #2 (Red Hat Linux)) id 1V1ICG-00072B-Qi for linux-arm-kernel@lists.infradead.org; Mon, 22 Jul 2013 15:38:52 +0000 Received: from e106331-lin.cambridge.arm.com (e106331-lin.cambridge.arm.com [10.1.205.41]) by cam-admin0.cambridge.arm.com (8.12.6/8.12.6) with ESMTP id r6MFbZkk007531; Mon, 22 Jul 2013 16:38:24 +0100 (BST) From: Mark Rutland To: linux-arm-kernel@lists.infradead.org Subject: [PATCHv2 2/5] arm64: factor out spin-table boot method Date: Mon, 22 Jul 2013 16:37:31 +0100 Message-Id: <1374507454-4573-3-git-send-email-mark.rutland@arm.com> X-Mailer: git-send-email 1.8.1.1 In-Reply-To: <1374507454-4573-1-git-send-email-mark.rutland@arm.com> References: <1374507454-4573-1-git-send-email-mark.rutland@arm.com> X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20130722_113849_187067_3DF52C6A X-CRM114-Status: GOOD ( 25.89 ) X-Spam-Score: -8.4 (--------) Cc: Mark Rutland , Lorenzo.Pieralisi@arm.com, graeme.gregory@linaro.org, nico@linaro.org, Marc.Zyngier@arm.com, Catalin.Marinas@arm.com, sboyd@codeaurora.org, santosh.shilimkar@ti.com, hanjun.guo@linaro.org X-BeenThere: linux-arm-kernel@lists.infradead.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+patchwork-linux-arm=patchwork.kernel.org@lists.infradead.org X-Spam-Status: No, score=-5.6 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP The arm64 kernel has an internal holding pen, which is necessary for some systems where we can't bring CPUs online individually and must hold multiple CPUs in a safe area until the kernel is able to handle them. The current SMP infrastructure for arm64 is closely coupled to this holding pen, and alternative boot methods must launch CPUs into the pen, from whence they are launched into the kernel proper. With PSCI (and possibly other future boot methods), we can bring CPUs online individually, and need not perform the secondary_holding_pen dance. Instead, this patch factors the holding pen management code out to the spin-table boot method code, as it is the only boot method requiring the pen. A new entry point for secondaries, secondary_entry is added for other boot methods to use, which bypasses the holding pen and its associated overhead when bringing CPUs online. The smp.pen.text section is also removed, as the pen can live in head.text without problem. The smp_operations structure is extended with two new functions, cpu_boot and cpu_postboot, for bringing a cpu into the kernel and performing any post-boot cleanup required by a bootmethod (e.g. resetting the secondary_holding_pen_release to INVALID_HWID). Signed-off-by: Mark Rutland --- arch/arm64/include/asm/smp.h | 17 ++++++++- arch/arm64/kernel/head.S | 12 +++++- arch/arm64/kernel/smp.c | 67 +++------------------------------- arch/arm64/kernel/smp_psci.c | 16 ++++---- arch/arm64/kernel/smp_spin_table.c | 75 ++++++++++++++++++++++++++++++++++++++ arch/arm64/kernel/vmlinux.lds.S | 1 - 6 files changed, 115 insertions(+), 73 deletions(-) diff --git a/arch/arm64/include/asm/smp.h b/arch/arm64/include/asm/smp.h index 90626b6..af39644 100644 --- a/arch/arm64/include/asm/smp.h +++ b/arch/arm64/include/asm/smp.h @@ -60,8 +60,7 @@ struct secondary_data { void *stack; }; extern struct secondary_data secondary_data; -extern void secondary_holding_pen(void); -extern volatile unsigned long secondary_holding_pen_release; +extern void secondary_entry(void); extern void arch_send_call_function_single_ipi(int cpu); extern void arch_send_call_function_ipi_mask(const struct cpumask *mask); @@ -70,8 +69,22 @@ struct device_node; struct smp_operations { const char *name; + /* + * Check devicetree data for cpu + */ int (*cpu_init)(struct device_node *, unsigned int); + /* + * Test if cpu is present and bootable + */ int (*cpu_prepare)(unsigned int); + /* + * Boot cpu into the kernel + */ + int (*cpu_boot)(unsigned int); + /* + * Performs post-boot cleanup + */ + void (*cpu_postboot)(void); }; extern const struct smp_operations smp_spin_table_ops; diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S index 53dcae4..3532ca6 100644 --- a/arch/arm64/kernel/head.S +++ b/arch/arm64/kernel/head.S @@ -217,7 +217,6 @@ ENTRY(__boot_cpu_mode) .quad PAGE_OFFSET #ifdef CONFIG_SMP - .pushsection .smp.pen.text, "ax" .align 3 1: .quad . .quad secondary_holding_pen_release @@ -242,7 +241,16 @@ pen: ldr x4, [x3] wfe b pen ENDPROC(secondary_holding_pen) - .popsection + + /* + * Secondary entry point that jumps straight into the kernel. Only to + * be used where CPUs are brought online dynamically by the kernel. + */ +ENTRY(secondary_entry) + bl __calc_phys_offset // x2=phys offset + bl el2_setup // Drop to EL1 + b secondary_startup +ENDPROC(secondary_entry) ENTRY(secondary_startup) /* diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c index 533f405..72c2823 100644 --- a/arch/arm64/kernel/smp.c +++ b/arch/arm64/kernel/smp.c @@ -54,7 +54,6 @@ * where to place its SVC stack */ struct secondary_data secondary_data; -volatile unsigned long secondary_holding_pen_release = INVALID_HWID; enum ipi_msg_type { IPI_RESCHEDULE, @@ -63,22 +62,7 @@ enum ipi_msg_type { IPI_CPU_STOP, }; -static DEFINE_RAW_SPINLOCK(boot_lock); - -/* - * Write secondary_holding_pen_release in a way that is guaranteed to be - * visible to all observers, irrespective of whether they're taking part - * in coherency or not. This is necessary for the hotplug code to work - * reliably. - */ -static void write_pen_release(u64 val) -{ - void *start = (void *)&secondary_holding_pen_release; - unsigned long size = sizeof(secondary_holding_pen_release); - - secondary_holding_pen_release = val; - __flush_dcache_area(start, size); -} +static const struct smp_operations *smp_ops[NR_CPUS]; /* * Boot a secondary CPU, and assign it the specified idle task. @@ -86,38 +70,10 @@ static void write_pen_release(u64 val) */ static int boot_secondary(unsigned int cpu, struct task_struct *idle) { - unsigned long timeout; - - /* - * Set synchronisation state between this boot processor - * and the secondary one - */ - raw_spin_lock(&boot_lock); - - /* - * Update the pen release flag. - */ - write_pen_release(cpu_logical_map(cpu)); - - /* - * Send an event, causing the secondaries to read pen_release. - */ - sev(); - - timeout = jiffies + (1 * HZ); - while (time_before(jiffies, timeout)) { - if (secondary_holding_pen_release == INVALID_HWID) - break; - udelay(10); - } - - /* - * Now the secondary core is starting up let it run its - * calibrations, then wait for it to finish - */ - raw_spin_unlock(&boot_lock); + if (smp_ops[cpu]->cpu_boot) + return smp_ops[cpu]->cpu_boot(cpu); - return secondary_holding_pen_release != INVALID_HWID ? -ENOSYS : 0; + return -EOPNOTSUPP; } static DECLARE_COMPLETION(cpu_running); @@ -187,17 +143,8 @@ asmlinkage void secondary_start_kernel(void) preempt_disable(); trace_hardirqs_off(); - /* - * Let the primary processor know we're out of the - * pen, then head off into the C entry point - */ - write_pen_release(INVALID_HWID); - - /* - * Synchronise with the boot thread. - */ - raw_spin_lock(&boot_lock); - raw_spin_unlock(&boot_lock); + if (smp_ops[cpu]->cpu_postboot) + smp_ops[cpu]->cpu_postboot(); /* * OK, now it's safe to let the boot CPU continue. Wait for @@ -242,8 +189,6 @@ static const struct smp_operations *supported_smp_ops[] __initconst = { NULL, }; -static const struct smp_operations *smp_ops[NR_CPUS]; - static const struct smp_operations * __init smp_get_ops(const char *name) { const struct smp_operations **ops = supported_smp_ops; diff --git a/arch/arm64/kernel/smp_psci.c b/arch/arm64/kernel/smp_psci.c index 2f0d3dd..20499bc 100644 --- a/arch/arm64/kernel/smp_psci.c +++ b/arch/arm64/kernel/smp_psci.c @@ -30,24 +30,26 @@ static int smp_psci_cpu_init(struct device_node *dn, unsigned int cpu) static int smp_psci_cpu_prepare(unsigned int cpu) { - int err; - if (!psci_ops.cpu_on) { pr_err("psci: no cpu_on method, not booting CPU%d\n", cpu); return -ENODEV; } - err = psci_ops.cpu_on(cpu_logical_map(cpu), __pa(secondary_holding_pen)); - if (err) { + return 0; +} + +static int smp_psci_cpu_boot(unsigned int cpu) +{ + int err = psci_ops.cpu_on(cpu_logical_map(cpu), __pa(secondary_entry)); + if (err) pr_err("psci: failed to boot CPU%d (%d)\n", cpu, err); - return err; - } - return 0; + return err; } const struct smp_operations smp_psci_ops = { .name = "psci", .cpu_init = smp_psci_cpu_init, .cpu_prepare = smp_psci_cpu_prepare, + .cpu_boot = smp_psci_cpu_boot, }; diff --git a/arch/arm64/kernel/smp_spin_table.c b/arch/arm64/kernel/smp_spin_table.c index 5fecffc..87af6bb 100644 --- a/arch/arm64/kernel/smp_spin_table.c +++ b/arch/arm64/kernel/smp_spin_table.c @@ -16,13 +16,36 @@ * along with this program. If not, see . */ +#include #include #include #include #include +#include +#include + +extern void secondary_holding_pen(void); +volatile unsigned long secondary_holding_pen_release = INVALID_HWID; static phys_addr_t cpu_release_addr[NR_CPUS]; +static DEFINE_RAW_SPINLOCK(boot_lock); + +/* + * Write secondary_holding_pen_release in a way that is guaranteed to be + * visible to all observers, irrespective of whether they're taking part + * in coherency or not. This is necessary for the hotplug code to work + * reliably. + */ +static void write_pen_release(u64 val) +{ + void *start = (void *)&secondary_holding_pen_release; + unsigned long size = sizeof(secondary_holding_pen_release); + + secondary_holding_pen_release = val; + __flush_dcache_area(start, size); +} + static int smp_spin_table_cpu_init(struct device_node *dn, unsigned int cpu) { @@ -59,8 +82,60 @@ static int smp_spin_table_cpu_prepare(unsigned int cpu) return 0; } +static int smp_spin_table_cpu_boot(unsigned int cpu) +{ + unsigned long timeout; + + /* + * Set synchronisation state between this boot processor + * and the secondary one + */ + raw_spin_lock(&boot_lock); + + /* + * Update the pen release flag. + */ + write_pen_release(cpu_logical_map(cpu)); + + /* + * Send an event, causing the secondaries to read pen_release. + */ + sev(); + + timeout = jiffies + (1 * HZ); + while (time_before(jiffies, timeout)) { + if (secondary_holding_pen_release == INVALID_HWID) + break; + udelay(10); + } + + /* + * Now the secondary core is starting up let it run its + * calibrations, then wait for it to finish + */ + raw_spin_unlock(&boot_lock); + + return secondary_holding_pen_release != INVALID_HWID ? -ENOSYS : 0; +} + +void smp_spin_table_cpu_postboot(void) +{ + /* + * Let the primary processor know we're out of the pen. + */ + write_pen_release(INVALID_HWID); + + /* + * Synchronise with the boot thread. + */ + raw_spin_lock(&boot_lock); + raw_spin_unlock(&boot_lock); +} + const struct smp_operations smp_spin_table_ops = { .name = "spin-table", .cpu_init = smp_spin_table_cpu_init, .cpu_prepare = smp_spin_table_cpu_prepare, + .cpu_boot = smp_spin_table_cpu_boot, + .cpu_postboot = smp_spin_table_cpu_postboot, }; diff --git a/arch/arm64/kernel/vmlinux.lds.S b/arch/arm64/kernel/vmlinux.lds.S index f5e5574..d8ca8d9 100644 --- a/arch/arm64/kernel/vmlinux.lds.S +++ b/arch/arm64/kernel/vmlinux.lds.S @@ -54,7 +54,6 @@ SECTIONS } .text : { /* Real text segment */ _stext = .; /* Text and read-only data */ - *(.smp.pen.text) __exception_text_start = .; *(.exception.text) __exception_text_end = .;