Removing dead CONFIG_BLK_DEV_IDE

Prev: ipmi: Make sure drivers were registered before unregistering them
Next: Removing dead code

From: Shane McDonald on 10 Jun 2010 14:30

---

On Wed, Jun 9, 2010 at 5:21 AM, Christoph Egger <siccegge(a)cs.fau.de> wrote:
> CONFIG_BLK_DEV_IDE doesn't exist in Kconfig, therefore removing all
> references for it from the source code.
>
> Signed-off-by: Christoph Egger <siccegge(a)cs.fau.de>
> ---
> �arch/mips/mti-malta/malta-setup.c | � 25 -------------------------
> �1 files changed, 0 insertions(+), 25 deletions(-)
>
> diff --git a/arch/mips/mti-malta/malta-setup.c b/arch/mips/mti-malta/malta-setup.c
> index b7f37d4..d85143c 100644
> --- a/arch/mips/mti-malta/malta-setup.c
> +++ b/arch/mips/mti-malta/malta-setup.c
> @@ -105,28 +105,6 @@ static void __init fd_activate(void)
> �}
> �#endif
>
> -#ifdef CONFIG_BLK_DEV_IDE
> -static void __init pci_clock_check(void)
> -{
> - � � � unsigned int __iomem *jmpr_p =
> - � � � � � � � (unsigned int *) ioremap(MALTA_JMPRS_REG, sizeof(unsigned int));
> - � � � int jmpr = (__raw_readl(jmpr_p) >> 2) & 0x07;
> - � � � static const int pciclocks[] __initdata = {
> - � � � � � � � 33, 20, 25, 30, 12, 16, 37, 10
> - � � � };
> - � � � int pciclock = pciclocks[jmpr];
> - � � � char *argptr = prom_getcmdline();
> -
> - � � � if (pciclock != 33 && !strstr(argptr, "idebus=")) {
> - � � � � � � � printk(KERN_WARNING "WARNING: PCI clock is %dMHz, "
> - � � � � � � � � � � � � � � � "setting idebus\n", pciclock);
> - � � � � � � � argptr += strlen(argptr);
> - � � � � � � � sprintf(argptr, " idebus=%d", pciclock);
> - � � � � � � � if (pciclock < 20 || pciclock > 66)
> - � � � � � � � � � � � printk(KERN_WARNING "WARNING: IDE timing "
> - � � � � � � � � � � � � � � � � � � � "calculations will be incorrect\n");
> - � � � }
> -}
> �#endif
>
> �#if defined(CONFIG_VT) && defined(CONFIG_VGA_CONSOLE)
> @@ -207,9 +185,6 @@ void __init plat_mem_setup(void)
> � � � �if (mips_revision_sconid == MIPS_REVISION_SCON_BONITO)
> � � � � � � � �bonito_quirks_setup();
>
> -#ifdef CONFIG_BLK_DEV_IDE
> - � � � pci_clock_check();
> -#endif
>
> �#ifdef CONFIG_BLK_DEV_FD
> � � � �fd_activate();
> --
> 1.6.3.3

I wonder if, instead of deleting the code, the constant should be
changed from CONFIG_BLK_DEV_IDE to CONFIG_IDE. The original
patch that removed CONFIG_BLK_DEV_IDE seemed to make this change:
http://kerneltrap.org/mailarchive/linux-kernel/2008/8/13/2929444

Shane
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

From: Christoph Egger on 16 Jun 2010 07:40

---

On Thu, Jun 10, 2010 at 12:23:06PM -0600, Shane McDonald wrote:
> I wonder if, instead of deleting the code, the constant should be
> changed from CONFIG_BLK_DEV_IDE to CONFIG_IDE. The original
> patch that removed CONFIG_BLK_DEV_IDE seemed to make this change:
> http://kerneltrap.org/mailarchive/linux-kernel/2008/8/13/2929444
>
> Shane

You're probably right, updated patch below

-------
From: Christoph Egger <siccegge(a)cs.fau.de>
Date: Mon, 7 Jun 2010 17:29:48 +0200
Subject: [PATCH 5/9] Removing dead CONFIG_BLK_DEV_IDE

CONFIG_BLK_DEV_IDE doesn't exist in Kconfig, therefore removing all
references for it from the source code.

Signed-off-by: Christoph Egger <siccegge(a)cs.fau.de>
---
diff --git a/arch/mips/mti-malta/malta-setup.c
b/arch/mips/mti-malta/malta-setup.c
index b7f37d4..f6a5ea8 100644
--- a/arch/mips/mti-malta/malta-setup.c
+++ b/arch/mips/mti-malta/malta-setup.c
@@ -105,7 +105,7 @@ static void __init fd_activate(void)
}
#endif

-#ifdef CONFIG_BLK_DEV_IDE
+#ifdef CONFIG_IDE
static void __init pci_clock_check(void)
{
unsigned int __iomem *jmpr_p =
@@ -207,7 +207,7 @@ void __init plat_mem_setup(void)
if (mips_revision_sconid == MIPS_REVISION_SCON_BONITO)
bonito_quirks_setup();

-#ifdef CONFIG_BLK_DEV_IDE
+#ifdef CONFIG_IDE
pci_clock_check();
#endif

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

From: Ralf Baechle on 5 Jul 2010 04:20

---

On Wed, Jun 16, 2010 at 01:35:48PM +0200, Christoph Egger wrote:

> On Thu, Jun 10, 2010 at 12:23:06PM -0600, Shane McDonald wrote:
> > I wonder if, instead of deleting the code, the constant should be
> > changed from CONFIG_BLK_DEV_IDE to CONFIG_IDE. The original
> > patch that removed CONFIG_BLK_DEV_IDE seemed to make this change:
> > http://kerneltrap.org/mailarchive/linux-kernel/2008/8/13/2929444
> >
> > Shane
>
> You're probably right, updated patch below
>
> -------
> From: Christoph Egger <siccegge(a)cs.fau.de>
> Date: Mon, 7 Jun 2010 17:29:48 +0200
> Subject: [PATCH 5/9] Removing dead CONFIG_BLK_DEV_IDE
>
> CONFIG_BLK_DEV_IDE doesn't exist in Kconfig, therefore removing all
> references for it from the source code.
>
> Signed-off-by: Christoph Egger <siccegge(a)cs.fau.de>
> ---
> diff --git a/arch/mips/mti-malta/malta-setup.c
> b/arch/mips/mti-malta/malta-setup.c
> index b7f37d4..f6a5ea8 100644
> --- a/arch/mips/mti-malta/malta-setup.c
> +++ b/arch/mips/mti-malta/malta-setup.c
> @@ -105,7 +105,7 @@ static void __init fd_activate(void)
> }
> #endif
>
> -#ifdef CONFIG_BLK_DEV_IDE
> +#ifdef CONFIG_IDE

This doesn't fly too well either. CONFIG_BLK_DEV_IDE was a bool but
CONFIG_IDE is tristate. But changing the ifdef to

#if defined(CONFIG_IDE) || defined(CONFIG_IDE_MODULE)

wouldn't really work either. I think this needs some generic infrastructure
to get the PCI clock.

Or maybe this is just another reason to scrap IDE support.

Ralf
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

 | 
Pages: 1
Prev: ipmi: Make sure drivers were registered before unregistering them
Next: Removing dead code