Re: git pull on ia64 linux tree

2007-05-08 Thread Zou Nan hai
On Tue, 2007-05-08 at 15:26, Simon Horman wrote: > On Tue, May 08, 2007 at 09:13:18AM +0800, Zou Nan hai wrote: > > > > Hi Tony, > > If the per-cpu TR entry is removed, > > I think the relocate_kernel.S also need a fix to remove the > per-cpu TR > > purge code. > > Just to clarify, i

Re: [PATCH Resend] - SN: validate smp_affinity mask on intr redirect

2007-05-08 Thread Peter Chubb
Jack> } Jack> + Jack> +bool is_affinity_mask_valid(cpumask_t cpumask) Jack> +{ Jack> + if (ia64_platform_is("sn2")) { Jack> + /* Only allow one CPU to be specified in the smp_affinity mask */ Jack> + if (cpus_weight(cpumask) != 1) Jack> + return false; Why not ju

Re: [PATCH Resend] - SN: validate smp_affinity mask on intr

2007-05-08 Thread John Keller
> > On Tue, 8 May 2007 13:14:26 -0700 > "Luck, Tony" <[EMAIL PROTECTED]> wrote: > > > > It had a dopey little bug: > > > > > > -#define is_affinity_mask_valid() 1 > > > +#define is_affinity_mask_valid(val) 1 > > > > That would fix warnings on non-ia64 systems (which is > > a step in the right di

Re: [PATCH Resend] - SN: validate smp_affinity mask on intr redirect

2007-05-08 Thread Andrew Morton
On Tue, 8 May 2007 13:14:26 -0700 "Luck, Tony" <[EMAIL PROTECTED]> wrote: > > It had a dopey little bug: > > > > -#define is_affinity_mask_valid() 1 > > +#define is_affinity_mask_valid(val) 1 > > That would fix warnings on non-ia64 systems (which is > a step in the right direction). But on ia64

RE: [PATCH Resend] - SN: validate smp_affinity mask on intr redirect

2007-05-08 Thread Luck, Tony
> It had a dopey little bug: > > -#define is_affinity_mask_valid() 1 > +#define is_affinity_mask_valid(val) 1 That would fix warnings on non-ia64 systems (which is a step in the right direction). But on ia64 I have the #define is_affinity_mask_valid is_affinity_mask_valid in play at that point,

Re: [PATCH Resend] - SN: validate smp_affinity mask on intr redirect

2007-05-08 Thread Andrew Morton
On Tue, 8 May 2007 11:03:20 -0700 "Luck, Tony" <[EMAIL PROTECTED]> wrote: > +#ifndef is_affinity_mask_valid > +#define is_affinity_mask_valid() 1 > +#endif > + > int no_irq_affinity; > static int irq_affinity_write_proc(struct file *file, const char __user > *buffer, >

RE: [PATCH Resend] - SN: validate smp_affinity mask on intr redirect

2007-05-08 Thread Luck, Tony
+#ifndef is_affinity_mask_valid +#define is_affinity_mask_valid() 1 +#endif + int no_irq_affinity; static int irq_affinity_write_proc(struct file *file, const char __user *buffer, unsigned long count, void *data) @@ -42,6 +46,9 @@ static int irq_affinity_writ

[patch] save and restore cpus_allowed in cpu_idle_wait

2007-05-08 Thread Siddha, Suresh B
Save and Restore the task's cpus_allowed mask, across the set_cpus_allowed() in cpu_idle_wait(). Without this, we will endup corrupting task's cpu affinity. Signed-off-by: Suresh Siddha <[EMAIL PROTECTED]> --- diff --git a/arch/ia64/kernel/process.c b/arch/ia64/kernel/process.c index ae96d41..cba

[PATCH Resend] - SN: validate smp_affinity mask on intr redirect

2007-05-08 Thread John Keller
On SN, only allow one bit to be set in the smp_affinty mask when redirecting an interrupt. Currently setting multiple bits is allowed, but only the first bit is used in determining the CPU to redirect to. This has caused confusion among some customers. Signed-off-by: John Keller <[EMAIL PROTECTED]

[patch 2/3] Use per-cpu elf_prstatus

2007-05-08 Thread Simon Horman
Use a pre-allocated per-cpu variable for saving the cpu registers, as per the IA64 specific code. This is in order to reduce possible stack contention at crash-time as per the inline comment. A following patch will update ia64 to use crash_save_cpu() rather than its own re-implementation of the ro

[patch 3/3] Use generic elf code on ia64

2007-05-08 Thread Simon Horman
Make use of the generic implementation of crash_save_cpu(). The code appears to work, however as the place where the registers are captured has changed natrually some of their values have also changed. This makes looking for problems by comparing the new and old output a little tricky. Signed-off

[patch 1/3] Add kdump_elf_core_copy_regs()

2007-05-08 Thread Simon Horman
Facilitate using crash_save_cpu() on ia64 by adding kdump_elf_core_copy_regs(). By default kdump_elf_core_copy_regs() is just defined to be elf_core_copy_regs(), which is what crash_save_cpu() previously used. ia64 (and other architectures) are able to define their own implementations as needed. S

[patch 0/3] [KEXEC] Use generic elf code on ia64

2007-05-08 Thread Simon Horman
Hi, the following series of three patches changes ia64 to use the generic elf note creation code. I previously posted it as a single patch, and have broken it up on Nanhai Zou's suggestion in order to make it easier to review. -- -- Horms H: http://www.vergenet.net/~horms/ W: http://www.va

Re: git pull on ia64 linux tree

2007-05-08 Thread Simon Horman
On Tue, May 08, 2007 at 09:13:18AM +0800, Zou Nan hai wrote: > > Hi Tony, > If the per-cpu TR entry is removed, > I think the relocate_kernel.S also need a fix to remove the per-cpu > TR > purge code. Just to clarify, is this what you are thinking about? If so, it seems sensible

Re: kexec/kdump: Use generic elf code on ia64

2007-05-08 Thread Simon Horman
On Tue, May 08, 2007 at 02:55:44PM +0800, Zou Nan hai wrote: > Hi, > The patch looks ok to me. > However split the patch in 2 parts, > 1. Put pr_status to percpu data to save stack usage. > 2. IA64 specific part. >may help others to review the patches. Sure, will do. -- Horms H: