diff mbox

Commit 384a290283fde63ba8dc671fca5420111cdac19a seems to break 11MPCore boot

Message ID 20130130171918.GO23505@n2100.arm.linux.org.uk (mailing list archive)
State New, archived
Headers show

Commit Message

Russell King - ARM Linux Jan. 30, 2013, 5:19 p.m. UTC 
On Wed, Jan 30, 2013 at 04:45:35PM +0000, Russell King - ARM Linux wrote:
> What we could do is scan interrupts 0-31 for a non-zero value.  If they're
> all zero, we should complain.  Otherwise, we use the first non-zero value
> we find and validate it for a single bit set.

And here's a patch to do this - I've not run this but it's just built
successfully here.  Anyone want to give it a go?

I've decided that if we do hit the mask==0 case, we should just wail
loudly - panic'ing will bring the kernel to a halt right there and then,
which may be before any console drivers have been initialized (and the
kernel message buffer is no longer easy to read).  Moreover, panic()ing,
along with the possibility of rebooting won't really fix this kind of
error - it's rather fatal as far as that goes.  So, I think just wailing
at CRIT level is fine for this condition that should not occur.

 arch/arm/common/gic.c |   23 +++++++++++++++++++++--
 1 files changed, 21 insertions(+), 2 deletions(-)

Comments

Nicolas Pitre Jan. 30, 2013, 5:30 p.m. UTC | #1 
On Wed, 30 Jan 2013, Russell King - ARM Linux wrote:

> On Wed, Jan 30, 2013 at 04:45:35PM +0000, Russell King - ARM Linux wrote:
> > What we could do is scan interrupts 0-31 for a non-zero value.  If they're
> > all zero, we should complain.  Otherwise, we use the first non-zero value
> > we find and validate it for a single bit set.
> 
> And here's a patch to do this - I've not run this but it's just built
> successfully here.  Anyone want to give it a go?
> 
> I've decided that if we do hit the mask==0 case, we should just wail
> loudly - panic'ing will bring the kernel to a halt right there and then,
> which may be before any console drivers have been initialized (and the
> kernel message buffer is no longer easy to read).  Moreover, panic()ing,
> along with the possibility of rebooting won't really fix this kind of
> error - it's rather fatal as far as that goes.  So, I think just wailing
> at CRIT level is fine for this condition that should not occur.
> 
>  arch/arm/common/gic.c |   23 +++++++++++++++++++++--
>  1 files changed, 21 insertions(+), 2 deletions(-)
> 
> diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
> index 36ae03a..3bcef49 100644
> --- a/arch/arm/common/gic.c
> +++ b/arch/arm/common/gic.c
> @@ -351,6 +351,23 @@ void __init gic_cascade_irq(unsigned int gic_nr, unsigned int irq)
>  	irq_set_chained_handler(irq, gic_handle_cascade_irq);
>  }
>  
> +static u8 gic_get_cpumask(struct gic_chip_data *gic)
> +{
> +	void __iomem *base = gic_data_dist_base(gic);
> +	u8 mask, i;
> +
> +	for (i = mask = 0; i < 32; i++) {
> +		mask = readl_relaxed(base + GIC_DIST_TARGET + i);
> +		if (mask)
> +			break;
> +	}

That should probably be:

	u32 mask;

	for (i = 0; i < 32; i += 4) {
		mask = readl_relaxed(base + GIC_DIST_TARGET + i);
		mask |= (mask >> 16);
		mask |= (mask >> 8);
		if (mask)
			return mask;
	}

I know that the spec says that the GIC should accept byte sized 
accesses, but that too is known not to work on all implementations.


Nicolas
Will Deacon Jan. 30, 2013, 5:57 p.m. UTC | #2 
On Wed, Jan 30, 2013 at 05:30:06PM +0000, Nicolas Pitre wrote:
> On Wed, 30 Jan 2013, Russell King - ARM Linux wrote:
> 
> > On Wed, Jan 30, 2013 at 04:45:35PM +0000, Russell King - ARM Linux wrote:
> > > What we could do is scan interrupts 0-31 for a non-zero value.  If they're
> > > all zero, we should complain.  Otherwise, we use the first non-zero value
> > > we find and validate it for a single bit set.
> > 
> > And here's a patch to do this - I've not run this but it's just built
> > successfully here.  Anyone want to give it a go?
> > 
> > I've decided that if we do hit the mask==0 case, we should just wail
> > loudly - panic'ing will bring the kernel to a halt right there and then,
> > which may be before any console drivers have been initialized (and the
> > kernel message buffer is no longer easy to read).  Moreover, panic()ing,
> > along with the possibility of rebooting won't really fix this kind of
> > error - it's rather fatal as far as that goes.  So, I think just wailing
> > at CRIT level is fine for this condition that should not occur.
> > 
> >  arch/arm/common/gic.c |   23 +++++++++++++++++++++--
> >  1 files changed, 21 insertions(+), 2 deletions(-)
> > 
> > diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
> > index 36ae03a..3bcef49 100644
> > --- a/arch/arm/common/gic.c
> > +++ b/arch/arm/common/gic.c
> > @@ -351,6 +351,23 @@ void __init gic_cascade_irq(unsigned int gic_nr, unsigned int irq)
> >  	irq_set_chained_handler(irq, gic_handle_cascade_irq);
> >  }
> >  
> > +static u8 gic_get_cpumask(struct gic_chip_data *gic)
> > +{
> > +	void __iomem *base = gic_data_dist_base(gic);
> > +	u8 mask, i;
> > +
> > +	for (i = mask = 0; i < 32; i++) {
> > +		mask = readl_relaxed(base + GIC_DIST_TARGET + i);
> > +		if (mask)
> > +			break;
> > +	}
> 
> That should probably be:
> 
> 	u32 mask;
> 
> 	for (i = 0; i < 32; i += 4) {
> 		mask = readl_relaxed(base + GIC_DIST_TARGET + i);
> 		mask |= (mask >> 16);
> 		mask |= (mask >> 8);
> 		if (mask)
> 			return mask;
> 	}
> 
> I know that the spec says that the GIC should accept byte sized 
> accesses, but that too is known not to work on all implementations.

Oh, for a rotation operator!

Tested-by: Will Deacon <will.deacon@arm.com>

Will
Russell King - ARM Linux Jan. 30, 2013, 6:03 p.m. UTC | #3 
On Wed, Jan 30, 2013 at 12:30:06PM -0500, Nicolas Pitre wrote:
> That should probably be:
> 
> 	u32 mask;
> 
> 	for (i = 0; i < 32; i += 4) {
> 		mask = readl_relaxed(base + GIC_DIST_TARGET + i);
> 		mask |= (mask >> 16);
> 		mask |= (mask >> 8);
> 		if (mask)
> 			return mask;
> 	}
> 
> I know that the spec says that the GIC should accept byte sized 
> accesses, but that too is known not to work on all implementations.

Well, it really doesn't make that much difference... because from what
I can see the kernel is rather sodded up as it currently stands.

There was a time when the kernel ran on the ARM Realview EB.  But it
doesn't today...

Why is it when I walk away from anything kernel related and then come
back to it, it's always sodded up and rather terminally broken?
Russell King - ARM Linux Jan. 30, 2013, 6:16 p.m. UTC | #4 
On Wed, Jan 30, 2013 at 06:03:31PM +0000, Russell King - ARM Linux wrote:
> On Wed, Jan 30, 2013 at 12:30:06PM -0500, Nicolas Pitre wrote:
> > That should probably be:
> > 
> > 	u32 mask;
> > 
> > 	for (i = 0; i < 32; i += 4) {
> > 		mask = readl_relaxed(base + GIC_DIST_TARGET + i);
> > 		mask |= (mask >> 16);
> > 		mask |= (mask >> 8);
> > 		if (mask)
> > 			return mask;
> > 	}
> > 
> > I know that the spec says that the GIC should accept byte sized 
> > accesses, but that too is known not to work on all implementations.
> 
> Well, it really doesn't make that much difference... because from what
> I can see the kernel is rather sodded up as it currently stands.
> 
> There was a time when the kernel ran on the ARM Realview EB.  But it
> doesn't today...
> 
> Why is it when I walk away from anything kernel related and then come
> back to it, it's always sodded up and rather terminally broken?

Finally, it now tries to mount a rootfs, and dies with the following:

BUG: spinlock lockup suspected on CPU#0, init/1                               
 lock: 0xcf8bde10, .magic: dead4ead, .owner: init/1, .owner_cpu: 0            
Backtrace:                                                                      
[<c00185d8>] (dump_backtrace+0x0/0x10c) from [<c03294e8>] (dump_stack+0x18/0x1c)
 r6:cf8bde10 r5:cf83d1c0 r4:cf8bde10 r3:cf83d1c0                                
[<c03294d0>] (dump_stack+0x0/0x1c) from [<c018926c>] (spin_dump+0x84/0x98)      
[<c01891e8>] (spin_dump+0x0/0x98) from [<c0189460>] (do_raw_spin_lock+0x100/0x19
8)                                                                              
 r5:00000000 r4:025e4000                                                        
[<c0189360>] (do_raw_spin_lock+0x0/0x198) from [<c032cbac>] (_raw_spin_lock+0x3c
/0x44)                                                                          
[<c032cb70>] (_raw_spin_lock+0x0/0x44) from [<c01c9224>] (pl011_console_write+0x
e8/0x11c)                                                                       
 r5:c0481780 r4:cf8bde10                                                        
[<c01c913c>] (pl011_console_write+0x0/0x11c) from [<c002aea8>] (call_console_dri
vers.clone.7+0xdc/0x104)                                                        
[<c002adcc>] (call_console_drivers.clone.7+0x0/0x104) from [<c002b320>] (console
_unlock+0x2e8/0x454)                                                            
[<c002b038>] (console_unlock+0x0/0x454) from [<c002b8b4>] (vprintk_emit+0x2d8/0x
594)                                                                            
[<c002b5dc>] (vprintk_emit+0x0/0x594) from [<c0329718>] (printk+0x3c/0x44)      
[<c03296dc>] (printk+0x0/0x44) from [<c002929c>] (warn_slowpath_common+0x28/0x6c
)                                                                               
 r3:00000009 r2:c0070ab0 r1:00000ab3 r0:c0334b60                                
[<c0029274>] (warn_slowpath_common+0x0/0x6c) from [<c0029304>] (warn_slowpath_nu
ll+0x24/0x2c)                                                                   
 r8:c00beb00 r7:000000d0 r6:cf8001c0 r5:60000093 r4:cf83e000                    
r3:00000009                                                                     
[<c00292e0>] (warn_slowpath_null+0x0/0x2c) from [<c0070ab0>] (lockdep_trace_allo
c+0xd8/0xf0)                                                                    
[<c00709d8>] (lockdep_trace_alloc+0x0/0xf0) from [<c00c0850>] (kmem_cache_alloc+
0x24/0x11c)                                                                     
 r5:20000008 r4:000080d0                                                        
[<c00c082c>] (kmem_cache_alloc+0x0/0x11c) from [<c00bb044>] (__get_vm_area_node.
clone.24+0x7c/0x16c)                                                            
 r7:000000d0 r6:00000001 r5:20000008 r4:cf83fbc0                                
[<c00bafc8>] (__get_vm_area_node.clone.24+0x0/0x16c) from [<c00bb7b8>] (get_vm_a
rea_caller+0x48/0x54)                                                           
[<c00bb770>] (get_vm_area_caller+0x0/0x54) from [<c0020064>] (__alloc_remap_buff
er.clone.15+0x38/0xb8)                                                          
[<c002002c>] (__alloc_remap_buffer.clone.15+0x0/0xb8) from [<c0020244>] (__dma_a
lloc+0x160/0x2c8)                                                               
 r8:00000000 r7:00002000 r6:00002000 r5:ffffffff r4:ffffffff                    
r3:cf83fc54                                                                     
[<c00200e4>] (__dma_alloc+0x0/0x2c8) from [<c00204d8>] (arm_dma_alloc+0x88/0xa0)
[<c0020450>] (arm_dma_alloc+0x0/0xa0) from [<c00beb00>] (dma_pool_alloc+0xcc/0x1
a8)                                                                             
[<c00bea34>] (dma_pool_alloc+0x0/0x1a8) from [<c01a9d14>] (pl08x_fill_llis_for_d
esc+0x28/0x568)                                                                 
[<c01a9cec>] (pl08x_fill_llis_for_desc+0x0/0x568) from [<c01aab8c>] (pl08x_prep_
slave_sg+0x258/0x3b0)                                                           
[<c01aa934>] (pl08x_prep_slave_sg+0x0/0x3b0) from [<c01c9f74>] (pl011_dma_tx_ref
ill+0x140/0x288)                                                                
[<c01c9e34>] (pl011_dma_tx_refill+0x0/0x288) from [<c01ca748>] (pl011_start_tx+0
xe4/0x120)                                                                      
[<c01ca664>] (pl011_start_tx+0x0/0x120) from [<c01c54a4>] (__uart_start+0x48/0x4
c)                                                                              
 r5:cfa7c400 r4:cf8bde10                                                        
[<c01c545c>] (__uart_start+0x0/0x4c) from [<c01c632c>] (uart_start+0x2c/0x3c)   
[<c01c6300>] (uart_start+0x0/0x3c) from [<c01c795c>] (uart_write+0xcc/0xf4)     
 r6:cfa8b037 r5:cf906000 r4:00000000 r3:cf83d1c0                                
[<c01c7890>] (uart_write+0x0/0xf4) from [<c01b0384>] (n_tty_write+0x1c0/0x3e4)  
[<c01b01c4>] (n_tty_write+0x0/0x3e4) from [<c01acfe8>] (tty_write+0x144/0x240)  
[<c01acea4>] (tty_write+0x0/0x240) from [<c01ad17c>] (redirected_tty_write+0x98/
0xac)                                                                           
[<c01ad0e4>] (redirected_tty_write+0x0/0xac) from [<c00c371c>] (vfs_write+0xbc/0
x150)                                                                           
 r8:00000000 r7:cf83ff70 r6:be83cc04 r5:00000038 r4:cf8b1d80                    
r3:cf83ff70                                                                     
[<c00c3660>] (vfs_write+0x0/0x150) from [<c00c39c0>] (sys_write+0x4c/0x78)      
 r8:00000000 r7:be83cc04 r6:cf83ff70 r5:00000000 r4:cf8b1d80                    
[<c00c3974>] (sys_write+0x0/0x78) from [<c0014460>] (ret_fast_syscall+0x0/0x3c) 

Now, this _used_ to work fine before the DMA code was changed...  What
joy running ARM kernels bring.  Thankfully I hardly ever do that nowadays.
I wonder why that is...
Russell King - ARM Linux Jan. 30, 2013, 6:19 p.m. UTC | #5 
On Wed, Jan 30, 2013 at 06:16:51PM +0000, Russell King - ARM Linux wrote:
> On Wed, Jan 30, 2013 at 06:03:31PM +0000, Russell King - ARM Linux wrote:
> > On Wed, Jan 30, 2013 at 12:30:06PM -0500, Nicolas Pitre wrote:
> > > That should probably be:
> > > 
> > > 	u32 mask;
> > > 
> > > 	for (i = 0; i < 32; i += 4) {
> > > 		mask = readl_relaxed(base + GIC_DIST_TARGET + i);
> > > 		mask |= (mask >> 16);
> > > 		mask |= (mask >> 8);
> > > 		if (mask)
> > > 			return mask;
> > > 	}
> > > 
> > > I know that the spec says that the GIC should accept byte sized 
> > > accesses, but that too is known not to work on all implementations.
> > 
> > Well, it really doesn't make that much difference... because from what
> > I can see the kernel is rather sodded up as it currently stands.
> > 
> > There was a time when the kernel ran on the ARM Realview EB.  But it
> > doesn't today...
> > 
> > Why is it when I walk away from anything kernel related and then come
> > back to it, it's always sodded up and rather terminally broken?
> 
> Finally, it now tries to mount a rootfs, and dies with the following:
> 
> BUG: spinlock lockup suspected on CPU#0, init/1                               
>  lock: 0xcf8bde10, .magic: dead4ead, .owner: init/1, .owner_cpu: 0            
> Backtrace:                                                                      
> [<c00185d8>] (dump_backtrace+0x0/0x10c) from [<c03294e8>] (dump_stack+0x18/0x1c)
>  r6:cf8bde10 r5:cf83d1c0 r4:cf8bde10 r3:cf83d1c0                                
> [<c03294d0>] (dump_stack+0x0/0x1c) from [<c018926c>] (spin_dump+0x84/0x98)      
> [<c01891e8>] (spin_dump+0x0/0x98) from [<c0189460>] (do_raw_spin_lock+0x100/0x19
> 8)                                                                              
>  r5:00000000 r4:025e4000                                                        
> [<c0189360>] (do_raw_spin_lock+0x0/0x198) from [<c032cbac>] (_raw_spin_lock+0x3c
> /0x44)                                                                          
> [<c032cb70>] (_raw_spin_lock+0x0/0x44) from [<c01c9224>] (pl011_console_write+0x
> e8/0x11c)                                                                       
>  r5:c0481780 r4:cf8bde10                                                        
> [<c01c913c>] (pl011_console_write+0x0/0x11c) from [<c002aea8>] (call_console_dri
> vers.clone.7+0xdc/0x104)                                                        
> [<c002adcc>] (call_console_drivers.clone.7+0x0/0x104) from [<c002b320>] (console
> _unlock+0x2e8/0x454)                                                            
> [<c002b038>] (console_unlock+0x0/0x454) from [<c002b8b4>] (vprintk_emit+0x2d8/0x
> 594)                                                                            
> [<c002b5dc>] (vprintk_emit+0x0/0x594) from [<c0329718>] (printk+0x3c/0x44)      
> [<c03296dc>] (printk+0x0/0x44) from [<c002929c>] (warn_slowpath_common+0x28/0x6c
> )                                                                               
>  r3:00000009 r2:c0070ab0 r1:00000ab3 r0:c0334b60                                
> [<c0029274>] (warn_slowpath_common+0x0/0x6c) from [<c0029304>] (warn_slowpath_nu
> ll+0x24/0x2c)                                                                   
>  r8:c00beb00 r7:000000d0 r6:cf8001c0 r5:60000093 r4:cf83e000                    
> r3:00000009                                                                     
> [<c00292e0>] (warn_slowpath_null+0x0/0x2c) from [<c0070ab0>] (lockdep_trace_allo
> c+0xd8/0xf0)                                                                    
> [<c00709d8>] (lockdep_trace_alloc+0x0/0xf0) from [<c00c0850>] (kmem_cache_alloc+
> 0x24/0x11c)                                                                     
>  r5:20000008 r4:000080d0                                                        
> [<c00c082c>] (kmem_cache_alloc+0x0/0x11c) from [<c00bb044>] (__get_vm_area_node.
> clone.24+0x7c/0x16c)                                                            
>  r7:000000d0 r6:00000001 r5:20000008 r4:cf83fbc0                                
> [<c00bafc8>] (__get_vm_area_node.clone.24+0x0/0x16c) from [<c00bb7b8>] (get_vm_a
> rea_caller+0x48/0x54)                                                           
> [<c00bb770>] (get_vm_area_caller+0x0/0x54) from [<c0020064>] (__alloc_remap_buff
> er.clone.15+0x38/0xb8)                                                          
> [<c002002c>] (__alloc_remap_buffer.clone.15+0x0/0xb8) from [<c0020244>] (__dma_a
> lloc+0x160/0x2c8)                                                               
>  r8:00000000 r7:00002000 r6:00002000 r5:ffffffff r4:ffffffff                    
> r3:cf83fc54                                                                     
> [<c00200e4>] (__dma_alloc+0x0/0x2c8) from [<c00204d8>] (arm_dma_alloc+0x88/0xa0)
> [<c0020450>] (arm_dma_alloc+0x0/0xa0) from [<c00beb00>] (dma_pool_alloc+0xcc/0x1
> a8)                                                                             
> [<c00bea34>] (dma_pool_alloc+0x0/0x1a8) from [<c01a9d14>] (pl08x_fill_llis_for_d
> esc+0x28/0x568)                                                                 
> [<c01a9cec>] (pl08x_fill_llis_for_desc+0x0/0x568) from [<c01aab8c>] (pl08x_prep_
> slave_sg+0x258/0x3b0)                                                           
> [<c01aa934>] (pl08x_prep_slave_sg+0x0/0x3b0) from [<c01c9f74>] (pl011_dma_tx_ref
> ill+0x140/0x288)                                                                
> [<c01c9e34>] (pl011_dma_tx_refill+0x0/0x288) from [<c01ca748>] (pl011_start_tx+0
> xe4/0x120)                                                                      
> [<c01ca664>] (pl011_start_tx+0x0/0x120) from [<c01c54a4>] (__uart_start+0x48/0x4
> c)                                                                              
>  r5:cfa7c400 r4:cf8bde10                                                        
> [<c01c545c>] (__uart_start+0x0/0x4c) from [<c01c632c>] (uart_start+0x2c/0x3c)   
> [<c01c6300>] (uart_start+0x0/0x3c) from [<c01c795c>] (uart_write+0xcc/0xf4)     
>  r6:cfa8b037 r5:cf906000 r4:00000000 r3:cf83d1c0                                
> [<c01c7890>] (uart_write+0x0/0xf4) from [<c01b0384>] (n_tty_write+0x1c0/0x3e4)  
> [<c01b01c4>] (n_tty_write+0x0/0x3e4) from [<c01acfe8>] (tty_write+0x144/0x240)  
> [<c01acea4>] (tty_write+0x0/0x240) from [<c01ad17c>] (redirected_tty_write+0x98/
> 0xac)                                                                           
> [<c01ad0e4>] (redirected_tty_write+0x0/0xac) from [<c00c371c>] (vfs_write+0xbc/0
> x150)                                                                           
>  r8:00000000 r7:cf83ff70 r6:be83cc04 r5:00000038 r4:cf8b1d80                    
> r3:cf83ff70                                                                     
> [<c00c3660>] (vfs_write+0x0/0x150) from [<c00c39c0>] (sys_write+0x4c/0x78)      
>  r8:00000000 r7:be83cc04 r6:cf83ff70 r5:00000000 r4:cf8b1d80                    
> [<c00c3974>] (sys_write+0x0/0x78) from [<c0014460>] (ret_fast_syscall+0x0/0x3c) 
> 
> Now, this _used_ to work fine before the DMA code was changed...  What
> joy running ARM kernels bring.  Thankfully I hardly ever do that nowadays.
> I wonder why that is...

Oh, and there's also this:

WARNING: at /home/rmk/git/linux-rmk/arch/arm/common/gic.c:757 gic_init_bases+0x1
2c/0x2ec()                                                                      
Cannot allocate irq_descs @ IRQ96, assuming pre-allocated                       
Modules linked in:                                                              
Backtrace:                                                                      
[<c00185d8>] (dump_backtrace+0x0/0x10c) from [<c03294e8>] (dump_stack+0x18/0x1c)
 r6:000002f5 r5:c042c62c r4:c044ff40 r3:c045f240                                
[<c03294d0>] (dump_stack+0x0/0x1c) from [<c00292c8>] (warn_slowpath_common+0x54/
0x6c)                                                                           
[<c0029274>] (warn_slowpath_common+0x0/0x6c) from [<c0029384>] (warn_slowpath_fm
t+0x38/0x40)                                                                    
 r8:00000060 r7:00000020 r6:00000060 r5:00000040 r4:00000001                    
r3:00000009                                                                     
[<c002934c>] (warn_slowpath_fmt+0x0/0x40) from [<c042c62c>] (gic_init_bases+0x12
c/0x2ec)                                                                        
 r3:00000060 r2:c03340b4                                                        
[<c042c500>] (gic_init_bases+0x0/0x2ec) from [<c042cdc8>] (gic_init_irq+0x8c/0xd
8)                                                                              
[<c042cd3c>] (gic_init_irq+0x0/0xd8) from [<c042827c>] (init_IRQ+0x1c/0x24)     
 r4:ffffffff                                                                    
[<c0428260>] (init_IRQ+0x0/0x24) from [<c04256c8>] (start_kernel+0x1a4/0x300)   
[<c0425524>] (start_kernel+0x0/0x300) from [<70008070>] (0x70008070)            
 r7:c04605dc r6:c044585c r5:c045c88c r4:00c5387d                                
---[ end trace 1b75b31a2719ed1c ]---                                            
------------[ cut here ]------------                                            
WARNING: at /home/rmk/git/linux-rmk/kernel/irq/irqdomain.c:234 irq_domain_add_le
gacy+0x80/0x140()                                                               
Modules linked in:                                                              
Backtrace:                                                                      
[<c00185d8>] (dump_backtrace+0x0/0x10c) from [<c03294e8>] (dump_stack+0x18/0x1c)
 r6:000000ea r5:c0081a38 r4:00000000 r3:c045f240                                
[<c03294d0>] (dump_stack+0x0/0x1c) from [<c00292c8>] (warn_slowpath_common+0x54/
0x6c)                                                                           
[<c0029274>] (warn_slowpath_common+0x0/0x6c) from [<c0029304>] (warn_slowpath_nu
ll+0x24/0x2c)                                                                   
 r8:00000020 r7:00000020 r6:cf8019c0 r5:00000040 r4:00000060                    
r3:00000009                                                                     
[<c00292e0>] (warn_slowpath_null+0x0/0x2c) from [<c0081a38>] (irq_domain_add_leg
acy+0x80/0x140)                                                                 
[<c00819b8>] (irq_domain_add_legacy+0x0/0x140) from [<c042c64c>] (gic_init_bases
+0x14c/0x2ec)                                                                   
[<c042c500>] (gic_init_bases+0x0/0x2ec) from [<c042cdc8>] (gic_init_irq+0x8c/0xd
8)                                                                              
[<c042cd3c>] (gic_init_irq+0x0/0xd8) from [<c042827c>] (init_IRQ+0x1c/0x24)     
 r4:ffffffff                                                                    
[<c0428260>] (init_IRQ+0x0/0x24) from [<c04256c8>] (start_kernel+0x1a4/0x300)   
[<c0425524>] (start_kernel+0x0/0x300) from [<70008070>] (0x70008070)            
 r7:c04605dc r6:c044585c r5:c045c88c r4:00c5387d                                
---[ end trace 1b75b31a2719ed1d ]---                                            
------------[ cut here ]------------                                            
WARNING: at /home/rmk/git/linux-rmk/arch/arm/common/gic.c:762 gic_init_bases+0x1
70/0x2ec()                                                                      
Modules linked in:                                                              
Backtrace:                                                                      
[<c00185d8>] (dump_backtrace+0x0/0x10c) from [<c03294e8>] (dump_stack+0x18/0x1c)
 r6:000002fa r5:c042c670 r4:00000000 r3:c045f240                                
[<c03294d0>] (dump_stack+0x0/0x1c) from [<c00292c8>] (warn_slowpath_common+0x54/
0x6c)                                                                           
[<c0029274>] (warn_slowpath_common+0x0/0x6c) from [<c0029304>] (warn_slowpath_nu
ll+0x24/0x2c)                                                                   
 r8:00000060 r7:00000020 r6:00000060 r5:00000040 r4:00000001                    
r3:00000009                                                                     
[<c00292e0>] (warn_slowpath_null+0x0/0x2c) from [<c042c670>] (gic_init_bases+0x1
70/0x2ec)                                                                       
[<c042c500>] (gic_init_bases+0x0/0x2ec) from [<c042cdc8>] (gic_init_irq+0x8c/0xd
8)                                                                              
[<c042cd3c>] (gic_init_irq+0x0/0xd8) from [<c042827c>] (init_IRQ+0x1c/0x24)     
 r4:ffffffff                                                                    
[<c0428260>] (init_IRQ+0x0/0x24) from [<c04256c8>] (start_kernel+0x1a4/0x300)   
[<c0425524>] (start_kernel+0x0/0x300) from [<70008070>] (0x70008070)            
 r7:c04605dc r6:c044585c r5:c045c88c r4:00c5387d                                
---[ end trace 1b75b31a2719ed1e ]---                                            

Way to go, people who DT-ized the GIC stuff, that's really great, three
warnings for one thing.  Really excellent.

Giving up trying to test this.  Someone else with more patience can do
it, and while they're at it they can fix the sodding kernel too.
diff mbox

Patch

diff --git a/arch/arm/common/gic.c b/arch/arm/common/gic.c
index 36ae03a..3bcef49 100644
--- a/arch/arm/common/gic.c
+++ b/arch/arm/common/gic.c
@@ -351,6 +351,23 @@  void __init gic_cascade_irq(unsigned int gic_nr, unsigned int irq)
 	irq_set_chained_handler(irq, gic_handle_cascade_irq);
 }
 
+static u8 gic_get_cpumask(struct gic_chip_data *gic)
+{
+	void __iomem *base = gic_data_dist_base(gic);
+	u8 mask, i;
+
+	for (i = mask = 0; i < 32; i++) {
+		mask = readl_relaxed(base + GIC_DIST_TARGET + i);
+		if (mask)
+			break;
+	}
+
+	if (!mask)
+		pr_crit("GIC CPU mask not found - kernel will fail to boot.\n");
+
+	return mask;
+}
+
 static void __init gic_dist_init(struct gic_chip_data *gic)
 {
 	unsigned int i;
@@ -369,7 +386,9 @@  static void __init gic_dist_init(struct gic_chip_data *gic)
 	/*
 	 * Set all global interrupts to this CPU only.
 	 */
-	cpumask = readl_relaxed(base + GIC_DIST_TARGET + 0);
+	cpumask = gic_get_cpumask(gic);
+	cpumask |= cpumask << 8;
+	cpumask |= cpumask << 16;
 	for (i = 32; i < gic_irqs; i += 4)
 		writel_relaxed(cpumask, base + GIC_DIST_TARGET + i * 4 / 4);
 
@@ -400,7 +419,7 @@  static void __cpuinit gic_cpu_init(struct gic_chip_data *gic)
 	 * Get what the GIC says our CPU mask is.
 	 */
 	BUG_ON(cpu >= NR_GIC_CPU_IF);
-	cpu_mask = readl_relaxed(dist_base + GIC_DIST_TARGET + 0);
+	cpu_mask = gic_get_cpumask(gic);
 	gic_cpu_map[cpu] = cpu_mask;
 
 	/*