[PATCH]gx-suspmod.c use boot_cpu_data instead of current_cpu_data

[PATCH]gx-suspmod.c use boot_cpu_data instead of current_cpu_data

[Dave Young]

Hi,
in preemptible kernel will report BUG: using smp_processor_id() in preemptible, so use boot_cpu_data instead of current_cpu_data.

Signed-off-by: Dave Young <hidave.darkstar@gmail.com> 

---
arch/i386/kernel/cpu/cpufreq/gx-suspmod.c |    4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff -pur linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c
--- linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c	2007-07-25 14:11:06.000000000 +0000
+++ linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c	2007-07-25 13:57:29.000000000 +0000
@@ -181,8 +181,8 @@ static __init struct pci_dev *gx_detect_
 	struct pci_dev *gx_pci = NULL;
 
 	/* check if CPU is a MediaGX or a Geode. */
-	if ((current_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
-	    (current_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
+	if ((boot_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
+	    (boot_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
 		dprintk("error: no MediaGX/Geode processor found!\n");
 		return NULL;
 	}

Re: [PATCH]gx-suspmod.c use boot_cpu_data instead of current_cpu_data

[Andrew Morton]

On Wed, 25 Jul 2007 14:19:05 +0000 Dave Young <hidave.darkstar@gmail.com> wrote:

> Hi,
> in preemptible kernel will report BUG: using smp_processor_id() in preemptible, so use boot_cpu_data instead of current_cpu_data.
> 
> Signed-off-by: Dave Young <hidave.darkstar@gmail.com> 
> 
> ---
> arch/i386/kernel/cpu/cpufreq/gx-suspmod.c |    4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff -pur linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c
> --- linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c	2007-07-25 14:11:06.000000000 +0000
> +++ linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c	2007-07-25 13:57:29.000000000 +0000
> @@ -181,8 +181,8 @@ static __init struct pci_dev *gx_detect_
>  	struct pci_dev *gx_pci = NULL;
>  
>  	/* check if CPU is a MediaGX or a Geode. */
> -	if ((current_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
> -	    (current_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
> +	if ((boot_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
> +	    (boot_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
>  		dprintk("error: no MediaGX/Geode processor found!\n");
>  		return NULL;
>  	}

um, I suspect it really wants to get at the current CPU.  But putting a
preempt_disable() around just that code is meaningless: the current CPU
could change immediately before or after the code block. It needs deeper
fixing, methinks.

Re: [PATCH]gx-suspmod.c use boot_cpu_data instead of current_cpu_data

[Dave Young]

>On 7/25/07, Andrew Morton <akpm@linux-foundation.org> wrote:
> On Wed, 25 Jul 2007 14:19:05 +0000 Dave Young <hidave.darkstar@gmail.com> wrote:
>
> > Hi,
> > in preemptible kernel will report BUG: using smp_processor_id() in preemptible, so use boot_cpu_data instead of current_cpu_data.
> >
> > Signed-off-by: Dave Young <hidave.darkstar@gmail.com>
> >
> > ---
> > arch/i386/kernel/cpu/cpufreq/gx-suspmod.c |    4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff -pur linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c
> > --- linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c   2007-07-25 14:11:06.000000000 +0000
> > +++ linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c       2007-07-25 13:57:29.000000000 +0000
> > @@ -181,8 +181,8 @@ static __init struct pci_dev *gx_detect_
> >       struct pci_dev *gx_pci = NULL;
> >
> >       /* check if CPU is a MediaGX or a Geode. */
> > -     if ((current_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
> > -         (current_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
> > +     if ((boot_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
> > +         (boot_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
> >               dprintk("error: no MediaGX/Geode processor found!\n");
> >               return NULL;
> >       }
>
> um, I suspect it really wants to get at the current CPU.  But putting a
> preempt_disable() around just that code is meaningless: the current CPU
> could change immediately before or after the code block. It needs deeper
> fixing, methinks.
The only target is to get the cpu vendor, so boot_cpu_data is enough,
the drivers/mtd/nand/cs553x_nand.c has the same usage.

Re: [PATCH]gx-suspmod.c use boot_cpu_data instead of current_cpu_data

[Andrew Morton]

On Thu, 26 Jul 2007 04:20:10 +0000 "Dave Young" <hidave.darkstar@gmail.com> wrote:

> >On 7/25/07, Andrew Morton <akpm@linux-foundation.org> wrote:
> > On Wed, 25 Jul 2007 14:19:05 +0000 Dave Young <hidave.darkstar@gmail.com> wrote:
> >
> > > Hi,
> > > in preemptible kernel will report BUG: using smp_processor_id() in preemptible, so use boot_cpu_data instead of current_cpu_data.
> > >
> > > Signed-off-by: Dave Young <hidave.darkstar@gmail.com>
> > >
> > > ---
> > > arch/i386/kernel/cpu/cpufreq/gx-suspmod.c |    4 ++--
> > > 1 file changed, 2 insertions(+), 2 deletions(-)
> > >
> > > diff -pur linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c
> > > --- linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c   2007-07-25 14:11:06.000000000 +0000
> > > +++ linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c       2007-07-25 13:57:29.000000000 +0000
> > > @@ -181,8 +181,8 @@ static __init struct pci_dev *gx_detect_
> > >       struct pci_dev *gx_pci = NULL;
> > >
> > >       /* check if CPU is a MediaGX or a Geode. */
> > > -     if ((current_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
> > > -         (current_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
> > > +     if ((boot_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
> > > +         (boot_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
> > >               dprintk("error: no MediaGX/Geode processor found!\n");
> > >               return NULL;
> > >       }
> >
> > um, I suspect it really wants to get at the current CPU.  But putting a
> > preempt_disable() around just that code is meaningless: the current CPU
> > could change immediately before or after the code block. It needs deeper
> > fixing, methinks.
> The only target is to get the cpu vendor, so boot_cpu_data is enough,
> the drivers/mtd/nand/cs553x_nand.c has the same usage.

I think there's some vague ambition in there to support non-identical CPUs.
In which case reading from the local CPU would make more sense.  (waves
frantically at cpufreq developers).

otoh, it'll take some work I suspect.  It'll need to sort out the overall
scope of "local cpu".  At what point and for how long should this code pin
itself on a cpu?

Re: [PATCH]gx-suspmod.c use boot_cpu_data instead of current_cpu_data

[Dave Jones]

On Wed, Jul 25, 2007 at 09:26:38PM -0700, Andrew Morton wrote:
 > On Thu, 26 Jul 2007 04:20:10 +0000 "Dave Young" <hidave.darkstar@gmail.com> wrote:
 > 
 > > >On 7/25/07, Andrew Morton <akpm@linux-foundation.org> wrote:
 > > > On Wed, 25 Jul 2007 14:19:05 +0000 Dave Young <hidave.darkstar@gmail.com> wrote:
 > > >
 > > > > Hi,
 > > > > in preemptible kernel will report BUG: using smp_processor_id() in preemptible, so use boot_cpu_data instead of current_cpu_data.
 > > > >
 > > > > Signed-off-by: Dave Young <hidave.darkstar@gmail.com>
 > > > >
 > > > > ---
 > > > > arch/i386/kernel/cpu/cpufreq/gx-suspmod.c |    4 ++--
 > > > > 1 file changed, 2 insertions(+), 2 deletions(-)
 > > > >
 > > > > diff -pur linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c
 > > > > --- linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c   2007-07-25 14:11:06.000000000 +0000
 > > > > +++ linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c       2007-07-25 13:57:29.000000000 +0000
 > > > > @@ -181,8 +181,8 @@ static __init struct pci_dev *gx_detect_
 > > > >       struct pci_dev *gx_pci = NULL;
 > > > >
 > > > >       /* check if CPU is a MediaGX or a Geode. */
 > > > > -     if ((current_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
 > > > > -         (current_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
 > > > > +     if ((boot_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
 > > > > +         (boot_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
 > > > >               dprintk("error: no MediaGX/Geode processor found!\n");
 > > > >               return NULL;
 > > > >       }
 > > >
 > > > um, I suspect it really wants to get at the current CPU.  But putting a
 > > > preempt_disable() around just that code is meaningless: the current CPU
 > > > could change immediately before or after the code block. It needs deeper
 > > > fixing, methinks.
 > > The only target is to get the cpu vendor, so boot_cpu_data is enough,
 > > the drivers/mtd/nand/cs553x_nand.c has the same usage.
 > 
 > I think there's some vague ambition in there to support non-identical CPUs.
 > In which case reading from the local CPU would make more sense.  (waves
 > frantically at cpufreq developers).

a non-identical CPU system with a geode + something else doesn't exist
and is very unlikely to happen.

	Dave

-- 
http://www.codemonkey.org.uk

Re: [PATCH]gx-suspmod.c use boot_cpu_data instead of current_cpu_data

[Alan Cox]

O> a non-identical CPU system with a geode + something else doesn't exist
> and is very unlikely to happen.

As the Geode isn't used SMP (and afaik isnt SMP capable) that would be a
reasonable assumption

Re: [PATCH]gx-suspmod.c use boot_cpu_data instead of current_cpu_data

[Hiroshi Miura]

2007/7/25, Andrew Morton <akpm@linux-foundation.org>:
> On Wed, 25 Jul 2007 14:19:05 +0000 Dave Young <hidave.darkstar@gmail.com> wrote:
>
> > Hi,
> > in preemptible kernel will report BUG: using smp_processor_id() in preemptible, so use boot_cpu_data instead of current_cpu_data.
> >
> > Signed-off-by: Dave Young <hidave.darkstar@gmail.com>
> >
> > ---
> > arch/i386/kernel/cpu/cpufreq/gx-suspmod.c |    4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff -pur linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c
> > --- linux/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c   2007-07-25 14:11:06.000000000 +0000
> > +++ linux.new/arch/i386/kernel/cpu/cpufreq/gx-suspmod.c       2007-07-25 13:57:29.000000000 +0000
> > @@ -181,8 +181,8 @@ static __init struct pci_dev *gx_detect_
> >       struct pci_dev *gx_pci = NULL;
> >
> >       /* check if CPU is a MediaGX or a Geode. */
> > -     if ((current_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
> > -         (current_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
> > +     if ((boot_cpu_data.x86_vendor != X86_VENDOR_NSC) &&
> > +         (boot_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) {
> >               dprintk("error: no MediaGX/Geode processor found!\n");
> >               return NULL;
> >       }
>
> um, I suspect it really wants to get at the current CPU.  But putting a
> preempt_disable() around just that code is meaningless: the current CPU
> could change immediately before or after the code block. It needs deeper
> fixing, methinks.

I think we can  remove these part. It checks chipset 's pci id
(vendor id/device id) after this part as follows;

        /* detect which companion chip is used */
        while ((gx_pci = pci_get_device(PCI_ANY_ID, PCI_ANY_ID,
gx_pci)) != NULL) {
                if ((pci_match_id(gx_chipset_tbl, gx_pci)) != NULL)
                        return gx_pci;
        }

It drives cpu frequency  through chipset register, not CPU regsiter.
It may be enough to check chipset pci id.

Hiroshi

-- 
HIroshi Miura
NTT DATA Corp. and IPA OSS center
miura@da-cha.org
（株）NTTデータ /（独）情報処理推進機構
三浦広志