: sparc32: Use OF device probing for sun4m irq
David Miller
maintainerPATCH 2/2 : sparc32: Use OF device probing for sun4m irq
sparc32: Use OF device probing for sun4m irq registers.
--- arch/sparc/kernel/entry.S | 38 ++++---- arch/sparc/kernel/sun4m_irq.c | 192 +++++++++++++++++------------------------ 2 files changed, 99 insertions(+), 131 deletions(-)
arch/sparc/kernel/entry.S +19 -19
@@ -272,17 +272,18 @@ smp4m_ticker: */ maybe_smp4m_msg: GET_PROCESSOR4M_ID(o3) - set sun4m_interrupts, %l5 - ld [%l5], %o5 + sethi %hi(sun4m_irq_percpu), %l5 + ld [%l5 + %lo(sun4m_irq_percpu)], %o5 sethi %hi(0x40000000), %o2 - sll %o3, 12, %o3 + sll %o3, 2, %o3 ld [%o5 + %o3], %o1 + ld [%o1 + 0x00], %o1 ! sun4m_irq_percpu[cpu]->pending andcc %o1, %o2, %g0 be,a smp4m_ticker cmp %l7, 14 - st %o2, [%o5 + 0x4] + st %o2, [%o1 + 0x04] ! sun4m_irq_percpu[cpu]->clear=0x40000000 WRITE_PAUSE - ld [%o5], %g0 + ld [%o1 + 0x00], %g0 ! sun4m_irq_percpu[cpu]->pending WRITE_PAUSE or %l0, PSR_PIL, %l4 wr %l4, 0x0, %psr
@@ -300,16 +301,16 @@ linux_trap_ipi15_sun4m: SAVE_ALL sethi %hi(0x80000000), %o2 GET_PROCESSOR4M_ID(o0) - set sun4m_interrupts, %l5 - ld [%l5], %o5 - sll %o0, 12, %o0 - add %o5, %o0, %o5 - ld [%o5], %o3 + sethi %hi(sun4m_irq_percpu), %l5 + ld [%l5 + %lo(sun4m_irq_percpu)], %o5 + sll %o0, 2, %o0 + ld [%o5 + %o0], %o5 + ld [%o5 + 0x00], %o3 ! sun4m_irq_percpu[cpu]->pending andcc %o3, %o2, %g0 be 1f ! Must be an NMI async memory error - st %o2, [%o5 + 4] + st %o2, [%o5 + 0x04] ! sun4m_irq_percpu[cpu]->clear=0x80000000 WRITE_PAUSE - ld [%o5], %g0 + ld [%o5 + 0x00], %g0 ! sun4m_irq_percpu[cpu]->pending WRITE_PAUSE or %l0, PSR_PIL, %l4 wr %l4, 0x0, %psr
@@ -323,12 +324,11 @@ linux_trap_ipi15_sun4m: 1: /* NMI async memory error handling. */ sethi %hi(0x80000000), %l4 - sethi %hi(0x4000), %o3 - sub %o5, %o0, %o5 - add %o5, %o3, %l5 - st %l4, [%l5 + 0xc] + sethi %hi(sun4m_irq_global), %o5 + ld [%o5 + %lo(sun4m_irq_global)], %l5 + st %l4, [%l5 + 0x0c] ! sun4m_irq_global->mask_set=0x80000000 WRITE_PAUSE - ld [%l5], %g0 + ld [%l5 + 0x00], %g0 ! sun4m_irq_global->pending WRITE_PAUSE or %l0, PSR_PIL, %l4 wr %l4, 0x0, %psr
@@ -337,9 +337,9 @@ linux_trap_ipi15_sun4m: WRITE_PAUSE call sun4m_nmi nop - st %l4, [%l5 + 0x8] + st %l4, [%l5 + 0x08] ! sun4m_irq_global->mask_clear=0x80000000 WRITE_PAUSE - ld [%l5], %g0 + ld [%l5 + 0x00], %g0 ! sun4m_irq_global->pending WRITE_PAUSE RESTORE_ALL
arch/sparc/kernel/sun4m_irq.c +80 -112
265 lines not on this page · raw message
Signature · 1 line
1.5.6.5.GIT
Re: [PATCH 2/2]: sparc32: Use OF device probing for sun4m irq registers.
With both patches applied, I get:
<#0> ok boot
Boot device: /iommu/sbus/SUNW,fas/sd@0,0 File and args:
SILO Version 1.4.13
boot:
Uncompressing image...
Loading kernel version 2.6.27
Loading initial ramdisk (2738820 bytes at 0x3000000 phys, 0x60000000 virt)...
PROMLIB: obio_ranges 5
OF stdout device is: /iommu@f,e0000000/sbus@f,e0001000/cgsix@3,0 Booting Linux...
Level 15 Interrupt
<#2 ok>
David Miller
maintainerFrom: Robert Reif <[email protected]>
Date: Tue, 09 Sep 2008 22:25:47 -0400
Quoted 1 line
With both patches applied, I get:
Thanks for testing, can you revert the second one, ie. have only the first one applied?
Thanks.
Re: [PATCH 2/2]: sparc32: Use OF device probing for sun4m irq registers.
In reply to David MillerQuoted 6 lines · David Miller
Thanks for testing, can you revert the second one, ie. have only the first one applied?
Thanks.
No joy. Rather than a level 15 interrupt, I now just get a hang after:
Booting Linux...
David Miller
maintainerFrom: Robert Reif <[email protected]>
Date: Wed, 10 Sep 2008 19:41:09 -0400
Quoted 9 lines
Quoted 6 lines · David Miller
Thanks for testing, can you revert the second one, ie. have only the first one applied?
Thanks.
No joy. Rather than a level 15 interrupt, I now just get a hang after:
Booting Linux...
Thanks a lot for testing, I'll try to figure out what is wrong and post new patches.
Re: [PATCH 2/2]: sparc32: Use OF device probing for sun4m irq registers.
In reply to David MillerQuoted 13 lines · David Miller
From: Robert Reif <[email protected]>
Date: Wed, 10 Sep 2008 19:41:09 -0400Quoted 9 lines
Quoted 7 lines · David Miller
Thanks for testing, can you revert the second one, ie. have only the first one applied?
Thanks.
Quoted 8 lines
Quoted 3 lines
No joy. Rather than a level 15 interrupt, I now just get a hang after:
Booting Linux...Thanks a lot for testing, I'll try to figure out what is wrong and post new patches.
Hang on. I looked at my command line history and applied the second patch rather than the first.
Give me a few minutes to try the right patch.
David Miller
maintainerFrom: Robert Reif <[email protected]>
Date: Wed, 10 Sep 2008 19:45:04 -0400
Quoted 4 lines
Hang on. I looked at my command line history and applied the second patch rather than the first.
Give me a few minutes to try the right patch.
OK :)