Re: [PATCH] oprofile: falling back on timer interrupt mode
From: Akinobu Mita <hidden>
Date: 2005-01-27 14:53:30
Also in:
lkml
On Thursday 27 January 2005 04:05, Olaf Hering wrote:
On Wed, Jan 26, Linux Kernel Mailing List wrote:quoted
ChangeSet 1.2038, 2005/01/25 20:31:01-08:00, amgta@yacht.ocn.ne.jp [PATCH] oprofile: falling back on timer interrupt mode
This misses arch/ppc
Thanks for pointing that out. This is opofile timer-mode fallback fix for ppc. Signed-off-by: Akinobu Mita <redacted>
--- 2.6-bk/arch/ppc/oprofile/common.c.orig 2005-01-27 23:28:44.000000000 +0900
+++ 2.6-bk/arch/ppc/oprofile/common.c 2005-01-27 23:32:27.000000000 +0900@@ -124,7 +124,7 @@ static struct oprofile_operations oprof_ .cpu_type = NULL /* To be filled in below. */ }; -void __init oprofile_arch_init(struct oprofile_operations *ops) +int __init oprofile_arch_init(struct oprofile_operations *ops) { char *name; int cpu_id = smp_processor_id();
@@ -132,14 +132,13 @@ void __init oprofile_arch_init(struct op #ifdef CONFIG_FSL_BOOKE model = &op_model_fsl_booke; #else - printk(KERN_ERR "oprofile enabled on unsupported processor!\n"); - return; + return -ENODEV; #endif name = kmalloc(32, GFP_KERNEL); if (NULL == name) - return; + return -ENOMEM; sprintf(name, "ppc/%s", cur_cpu_spec[cpu_id]->cpu_name);
@@ -151,6 +150,8 @@ void __init oprofile_arch_init(struct op printk(KERN_INFO "oprofile: using %s performance monitoring.\n", oprof_ppc32_ops.cpu_type); + + return 0; } void oprofile_arch_exit(void)