From: tony@atomide.com (Tony Lindgren)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 08/14] omap: Use ioremap for omap4 L4 code
Date: Sat, 17 Oct 2009 09:19:47 -0700 [thread overview]
Message-ID: <20091017161947.GK12576@atomide.com> (raw)
In-Reply-To: <EAF47CD23C76F840A9E7FCE10091EFAB02BB37D7AA@dbde02.ent.ti.com>
* Shilimkar, Santosh <santosh.shilimkar@ti.com> [091017 03:49]:
> Thanks Tony for this ioremap series!!
>
> > -----Original Message-----
> > From: Tony Lindgren [mailto:tony at atomide.com]
> > Sent: Saturday, October 17, 2009 5:16 AM
> > To: linux-arm-kernel at lists.infradead.org
> > Cc: linux-omap at vger.kernel.org; Shilimkar, Santosh
> > Subject: Re: [PATCH 08/14] omap: Use ioremap for omap4 L4 code
> >
> > * Tony Lindgren <tony@atomide.com> [091016 16:42]:
> > > Use ioremap for omap4 L4 code
> >
> > Santosh, can you please check and ack this patch?
> >
> > Especially some of the ioremap sizes may be wrong. Also, this one
> > I cannot boot test, so it's just compile tested.
>
> I have checked the patch and it looks all right except few typo. Also the scu_base is needed bit earlier.
OK
> >
> > > Cc: Santosh Shilimkar <santosh.shilimkar@ti.com>
> > > Signed-off-by: Tony Lindgren <tony@atomide.com>
> > > ---
> > > arch/arm/mach-omap2/board-4430sdp.c | 13 ++++++++++--
> > > arch/arm/mach-omap2/omap-smp.c | 28
> > ++++++++++++++++++++-----
> > > arch/arm/mach-omap2/timer-gp.c | 3 ++-
> > > arch/arm/plat-omap/include/mach/entry-macro.S | 2 ++
> > > arch/arm/plat-omap/include/mach/omap44xx.h | 5 ----
> > > 5 files changed, 37 insertions(+), 14 deletions(-)
> > >
> > > diff --git a/arch/arm/mach-omap2/board-4430sdp.c b/arch/arm/mach-
> > omap2/board-4430sdp.c
> > > index eb37c40..2a7b901 100644
> > > --- a/arch/arm/mach-omap2/board-4430sdp.c
> > > +++ b/arch/arm/mach-omap2/board-4430sdp.c
> > > @@ -52,8 +52,17 @@ static struct omap_board_config_kernel
> > sdp4430_config[] __initdata = {
> > >
> > > static void __init gic_init_irq(void)
> > > {
> > > - gic_dist_init(0, OMAP2_IO_ADDRESS(OMAP44XX_GIC_DIST_BASE), 29);
> > > - gic_cpu_init(0, OMAP2_IO_ADDRESS(OMAP44XX_GIC_CPU_BASE));
> > > + void __iomem *base;
> > > +
> > > + /* Static mapping, never released */
> > > + base = ioremap(OMAP44XX_GIC_DIST_BASE, SZ_4K);
> > > + BUG_ON(!base);
> > > + gic_dist_init(0, base, 29);
> > > +
> > > + /* Static mapping, never released */
> > > + base = ioremap(OMAP44XX_GIC_CPU_BASE, SZ_256);
> Should be "SZ_512"
OK
> > > + BUG_ON(!base);
> > > + gic_cpu_init(0, OMAP44XX_GIC_CPU_BASE);
> This should be gic_cpu_init(0, base);
Oops :)
> > > }
> > >
> > > static void __init omap_4430sdp_init_irq(void)
> > > diff --git a/arch/arm/mach-omap2/omap-smp.c b/arch/arm/mach-omap2/omap-
> > smp.c
> > > index 48ee295..a1b618c 100644
> > > --- a/arch/arm/mach-omap2/omap-smp.c
> > > +++ b/arch/arm/mach-omap2/omap-smp.c
> > > @@ -26,11 +26,11 @@
> > > #include <mach/hardware.h>
> > >
> > > /* Registers used for communicating startup information */
> > > -#define OMAP4_AUXCOREBOOT_REG0 (OMAP44XX_VA_WKUPGEN_BASE +
> > 0x800)
> > > -#define OMAP4_AUXCOREBOOT_REG1 (OMAP44XX_VA_WKUPGEN_BASE +
> > 0x804)
> > > +static void __iomem *omap4_auxcoreboot_reg0;
> > > +static void __iomem *omap4_auxcoreboot_reg1;
> > >
> > > /* SCU base address */
> > > -static void __iomem *scu_base = OMAP44XX_VA_SCU_BASE;
> > > +static void __iomem *scu_base;
> > >
> > > /*
> > > * Use SCU config register to count number of cores
> > > @@ -46,6 +46,8 @@ static DEFINE_SPINLOCK(boot_lock);
> > >
> > > void __cpuinit platform_secondary_init(unsigned int cpu)
> > > {
> > > + void __iomem *gic_cpu_base;
> > > +
> > > trace_hardirqs_off();
> > >
> > > /*
> > > @@ -54,7 +56,10 @@ void __cpuinit platform_secondary_init(unsigned int
> > cpu)
> > > * for us: do so
> > > */
> > >
> > > - gic_cpu_init(0, OMAP2_IO_ADDRESS(OMAP44XX_GIC_CPU_BASE));
> > > + /* Static mapping, never released */
> > > + gic_cpu_base = ioremap(OMAP44XX_GIC_CPU_BASE, SZ_256);
> Although underneath static mapping will take care of not doing dual mapping, we can map this ones in gic_init_irq() and use the same variable here.
Yeah, good.
> > > + BUG_ON(!gic_cpu_base);
> > > + gic_cpu_init(0, gic_cpu_base);
> > >
> > > /*
> > > * Synchronise with the boot thread.
> > > @@ -79,7 +84,7 @@ int __cpuinit boot_secondary(unsigned int cpu, struct
> > task_struct *idle)
> > > * the AuxCoreBoot1 register is updated with cpu state
> > > * A barrier is added to ensure that write buffer is drained
> > > */
> > > - __raw_writel(cpu, OMAP4_AUXCOREBOOT_REG1);
> > > + __raw_writel(cpu, omap4_auxcoreboot_reg1);
> > > smp_wmb();
> > >
> > > timeout = jiffies + (1 * HZ);
> > > @@ -104,7 +109,7 @@ static void __init wakeup_secondary(void)
> > > * A barrier is added to ensure that write buffer is drained
> > > */
> > > __raw_writel(virt_to_phys(omap_secondary_startup), \
> > > - OMAP4_AUXCOREBOOT_REG0);
> > > + omap4_auxcoreboot_reg0);
> > > smp_wmb();
> > >
> > > /*
> > > @@ -130,6 +135,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
> > > {
> > > unsigned int ncores = get_core_count();
> This call will not give correct count. Also there is one more call earlier and hence scu_base ioremaps needs to be done earlier.
OK
> > > unsigned int cpu = smp_processor_id();
> > > + void __iomem *omap4_wkupgen_base;
> > > int i;
> > >
> > > /* sanity check */
> > > @@ -161,6 +167,16 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
> > > for (i = 0; i < max_cpus; i++)
> > > set_cpu_present(i, true);
> > >
> > > + /* Never released */
> > > + omap4_wkupgen_base = ioremap(OMAP44XX_WKUPGEN_BASE, SZ_4K);
> > > + BUG_ON(!omap4_wkupgen_base);
> > > + omap4_auxcoreboot_reg0 = omap4_wkupgen_base + 0x800;
> > > + omap4_auxcoreboot_reg0 = omap4_wkupgen_base + 0x804;
> Should be "omap4_auxcoreboot_reg1 = omap4_wkupgen_base + 0x804;"
Oops!
> > > +
> > > + /* Never released */
> > > + scu_base = ioremap(OMAP44XX_GIC_CPU_BASE, SZ_256);
> > > + BUG_ON(!scu_base);
> > > +
> > > if (max_cpus > 1) {
> > > /*
> > > * Enable the local timer or broadcast device for the
> > > diff --git a/arch/arm/mach-omap2/timer-gp.c b/arch/arm/mach-omap2/timer-
> > gp.c
> > > index e2338c0..cd72970 100644
> > > --- a/arch/arm/mach-omap2/timer-gp.c
> > > +++ b/arch/arm/mach-omap2/timer-gp.c
> > > @@ -231,7 +231,8 @@ static void __init omap2_gp_clocksource_init(void)
> > > static void __init omap2_gp_timer_init(void)
> > > {
> > > #ifdef CONFIG_LOCAL_TIMERS
> > > - twd_base = OMAP2_IO_ADDRESS(OMAP44XX_LOCAL_TWD_BASE);
> > > + twd_base = ioremap(OMAP44XX_LOCAL_TWD_BASE, SZ_256);
> > > + BUG_ON(!twd_base);
> > > #endif
> > > omap_dm_timer_init();
> > >
> > > diff --git a/arch/arm/plat-omap/include/mach/entry-macro.S
> > b/arch/arm/plat-omap/include/mach/entry-macro.S
> > > index abe0864..3bad928 100644
> > > --- a/arch/arm/plat-omap/include/mach/entry-macro.S
> > > +++ b/arch/arm/plat-omap/include/mach/entry-macro.S
> > > @@ -104,6 +104,8 @@
> > >
> > > .endm
> > > #else
> > > +#define OMAP44XX_VA_GIC_CPU_BASE
> > OMAP2_IO_ADDRESS(OMAP44XX_GIC_CPU_BASE)
> > > +
> > > /*
> > > * The interrupt numbering scheme is defined in the
> > > * interrupt controller spec. To wit:
> > > diff --git a/arch/arm/plat-omap/include/mach/omap44xx.h b/arch/arm/plat-
> > omap/include/mach/omap44xx.h
> > > index b3ba5ac..8b4a578 100644
> > > --- a/arch/arm/plat-omap/include/mach/omap44xx.h
> > > +++ b/arch/arm/plat-omap/include/mach/omap44xx.h
> > > @@ -33,14 +33,9 @@
> > > #define IRQ_SIR_IRQ 0x0040
> > > #define OMAP44XX_GIC_DIST_BASE 0x48241000
> > > #define OMAP44XX_GIC_CPU_BASE 0x48240100
> > > -#define OMAP44XX_VA_GIC_CPU_BASE
> > OMAP2_IO_ADDRESS(OMAP44XX_GIC_CPU_BASE)
> > > #define OMAP44XX_SCU_BASE 0x48240000
> > > -#define OMAP44XX_VA_SCU_BASE
> > OMAP2_IO_ADDRESS(OMAP44XX_SCU_BASE)
> > > #define OMAP44XX_LOCAL_TWD_BASE 0x48240600
> > > -#define OMAP44XX_VA_LOCAL_TWD_BASE
> > OMAP2_IO_ADDRESS(OMAP44XX_LOCAL_TWD_BASE)
> > > -#define OMAP44XX_LOCAL_TWD_SIZE 0x00000100
> > > #define OMAP44XX_WKUPGEN_BASE 0x48281000
> > > -#define OMAP44XX_VA_WKUPGEN_BASE
> > OMAP2_IO_ADDRESS(OMAP44XX_WKUPGEN_BASE)
> > >
> > > #endif /* __ASM_ARCH_OMAP44XX_H */
>
> Here is the patch on top of ioremap series which should fix above mentioned things. With below patch + two serial related patches the kernel boots on OMAP4430.
Great, will update. Thanks for taking the time to fix it!
Regards,
Tony
> From c7db626e320b73c02ce7578e81b8b8c65c41c955 Mon Sep 17 00:00:00 2001
> From: Santosh Shilimkar <santosh.shilimkar@ti.com>
> Date: Sat, 17 Oct 2009 15:25:00 +0530
> Subject: [PATCH] omap: ioremap fixes for omap4
>
> This patch fixes the below ioremap for omap4 patch.
>
> http://patchwork.kernel.org/patch/54457/
>
> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
> Signed-off-by: Tony Lindgren <tony@atomide.com>
> ---
> arch/arm/mach-omap2/board-4430sdp.c | 4 ++--
> arch/arm/mach-omap2/omap-smp.c | 23 ++++++++++-------------
> arch/arm/plat-omap/common.c | 3 +++
> arch/arm/plat-omap/include/mach/common.h | 3 +++
> 4 files changed, 18 insertions(+), 15 deletions(-)
>
> diff --git a/arch/arm/mach-omap2/board-4430sdp.c b/arch/arm/mach-omap2/board-4430sdp.c
> index 50a62f2..4b644a9 100644
> --- a/arch/arm/mach-omap2/board-4430sdp.c
> +++ b/arch/arm/mach-omap2/board-4430sdp.c
> @@ -60,9 +60,9 @@ static void __init gic_init_irq(void)
> gic_dist_init(0, base, 29);
>
> /* Static mapping, never released */
> - base = ioremap(OMAP44XX_GIC_CPU_BASE, SZ_256);
> + gic_cpu_base_addr = ioremap(OMAP44XX_GIC_CPU_BASE, SZ_512);
> BUG_ON(!base);
> - gic_cpu_init(0, OMAP44XX_GIC_CPU_BASE);
> + gic_cpu_init(0, gic_cpu_base_addr);
> }
>
> static void __init omap_4430sdp_init_irq(void)
> diff --git a/arch/arm/mach-omap2/omap-smp.c b/arch/arm/mach-omap2/omap-smp.c
> index a1b618c..8813ac2 100644
> --- a/arch/arm/mach-omap2/omap-smp.c
> +++ b/arch/arm/mach-omap2/omap-smp.c
> @@ -24,6 +24,7 @@
> #include <asm/localtimer.h>
> #include <asm/smp_scu.h>
> #include <mach/hardware.h>
> +#include <mach/common.h>
>
> /* Registers used for communicating startup information */
> static void __iomem *omap4_auxcoreboot_reg0;
> @@ -46,8 +47,6 @@ static DEFINE_SPINLOCK(boot_lock);
>
> void __cpuinit platform_secondary_init(unsigned int cpu)
> {
> - void __iomem *gic_cpu_base;
> -
> trace_hardirqs_off();
>
> /*
> @@ -55,11 +54,7 @@ void __cpuinit platform_secondary_init(unsigned int cpu)
> * core (e.g. timer irq), then they will not have been enabled
> * for us: do so
> */
> -
> - /* Static mapping, never released */
> - gic_cpu_base = ioremap(OMAP44XX_GIC_CPU_BASE, SZ_256);
> - BUG_ON(!gic_cpu_base);
> - gic_cpu_init(0, gic_cpu_base);
> + gic_cpu_init(0, gic_cpu_base_addr);
>
> /*
> * Synchronise with the boot thread.
> @@ -125,7 +120,13 @@ static void __init wakeup_secondary(void)
> */
> void __init smp_init_cpus(void)
> {
> - unsigned int i, ncores = get_core_count();
> + unsigned int i, ncores;
> +
> + /* Never released */
> + scu_base = ioremap(OMAP44XX_SCU_BASE, SZ_256);
> + BUG_ON(!scu_base);
> +
> + ncores = get_core_count();
>
> for (i = 0; i < ncores; i++)
> set_cpu_possible(i, true);
> @@ -171,11 +172,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
> omap4_wkupgen_base = ioremap(OMAP44XX_WKUPGEN_BASE, SZ_4K);
> BUG_ON(!omap4_wkupgen_base);
> omap4_auxcoreboot_reg0 = omap4_wkupgen_base + 0x800;
> - omap4_auxcoreboot_reg0 = omap4_wkupgen_base + 0x804;
> -
> - /* Never released */
> - scu_base = ioremap(OMAP44XX_GIC_CPU_BASE, SZ_256);
> - BUG_ON(!scu_base);
> + omap4_auxcoreboot_reg1 = omap4_wkupgen_base + 0x804;
>
> if (max_cpus > 1) {
> /*
> diff --git a/arch/arm/plat-omap/common.c b/arch/arm/plat-omap/common.c
> index 1aa58d2..8b3ef17 100644
> --- a/arch/arm/plat-omap/common.c
> +++ b/arch/arm/plat-omap/common.c
> @@ -49,6 +49,9 @@ int omap_bootloader_tag_len;
> struct omap_board_config_kernel *omap_board_config;
> int omap_board_config_size;
>
> +/* used by omap-smp.c and board-4430sdp.c */
> +void __iomem *gic_cpu_base_addr;
> +
> static const void *get_config(u16 tag, size_t len, int skip, size_t *len_out)
> {
> struct omap_board_config_kernel *kinfo = NULL;
> diff --git a/arch/arm/plat-omap/include/mach/common.h b/arch/arm/plat-omap/include/mach/common.h
> index fdeab42..064f173 100644
> --- a/arch/arm/plat-omap/include/mach/common.h
> +++ b/arch/arm/plat-omap/include/mach/common.h
> @@ -31,6 +31,9 @@
>
> struct sys_timer;
>
> +/* used by omap-smp.c and board-4430sdp.c */
> +extern void __iomem *gic_cpu_base_addr;
> +
> extern void omap_map_common_io(void);
> extern struct sys_timer omap_timer;
> #if defined(CONFIG_I2C_OMAP) || defined(CONFIG_I2C_OMAP_MODULE)
> --
> 1.5.4.7
>
>
>
> Regards,
> Santosh
next prev parent reply other threads:[~2009-10-17 16:19 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-16 23:40 [PATCH 00/14] Use ioremap for omap, split OMAP2_IO_ADDRESS for more space Tony Lindgren
2009-10-16 23:40 ` [PATCH 01/14] omap: Change low-level serial init to use ioremap Tony Lindgren
2009-10-16 23:40 ` [PATCH 02/14] omap: Use ioremap in irq.c Tony Lindgren
2009-10-16 23:40 ` [PATCH 03/14] omap: Use ioremap in dma.c Tony Lindgren
2009-10-16 23:40 ` [PATCH 04/14] omap: Use ioremap in dmtimer.c Tony Lindgren
2009-10-16 23:41 ` [PATCH 05/14] omap: Use ioremap in gpio.c Tony Lindgren
2009-10-17 8:10 ` Russell King - ARM Linux
2009-10-17 16:13 ` Tony Lindgren
2009-10-19 0:19 ` Tony Lindgren
2009-10-16 23:41 ` [PATCH 06/14] omap: Use ioremap in omap_hwmod.c Tony Lindgren
2009-10-16 23:41 ` [PATCH 07/14] omap: Use getnstimeofday for omap_device Tony Lindgren
2009-10-16 23:41 ` [PATCH 08/14] omap: Use ioremap for omap4 L4 code Tony Lindgren
2009-10-16 23:46 ` Tony Lindgren
2009-10-17 10:48 ` Shilimkar, Santosh
2009-10-17 16:19 ` Tony Lindgren [this message]
2009-10-19 0:24 ` Tony Lindgren
2009-10-19 7:27 ` Shilimkar, Santosh
2009-10-19 17:12 ` Tony Lindgren
2009-10-19 17:21 ` Aguirre Rodriguez, Sergio Alberto
2009-10-19 17:38 ` Tony Lindgren
2009-10-19 18:07 ` Aguirre Rodriguez, Sergio Alberto
2009-10-20 6:34 ` Shilimkar, Santosh
2009-10-16 23:41 ` [PATCH 09/14] omap: Use ioremap in dispc.c Tony Lindgren
2009-10-16 23:42 ` [PATCH 10/14] omap: Split OMAP2_IO_ADDRESS to L3 and L4 Tony Lindgren
2009-10-16 23:42 ` [PATCH 11/14] omap: Remap L3, L4 to get more kernel io address space Tony Lindgren
2009-10-20 0:20 ` Tony Lindgren
2009-10-20 6:32 ` Shilimkar, Santosh
2009-10-16 23:42 ` [PATCH 12/14] omap: Move SRAM map to claim more io space Tony Lindgren
2009-10-16 23:42 ` [PATCH 13/14] omap: Fix DEBUG_LL UART io address Tony Lindgren
2009-10-16 23:42 ` [PATCH 14/14] omap: Add OMAP4 L3 and L4 peripherals Tony Lindgren
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20091017161947.GK12576@atomide.com \
--to=tony@atomide.com \
--cc=linux-arm-kernel@lists.infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).