[PATCH 2/2] : Add local_irq_restore in error handling code

[PATCH 2/2] : Add local_irq_restore in error handling code

[Julia Lawall]

From: Julia Lawall <julia@diku.dk>

There is a call to local_irq_restore in the normal exit case, so it would
seem that there should be one on an error return as well.

The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@@
expression l;
expression E,E1,E2;
@@

local_irq_save(l);
... when != local_irq_restore(l)
    when != spin_unlock_irqrestore(E,l)
    when any
    when strict
(
if (...) { ... when != local_irq_restore(l)
               when != spin_unlock_irqrestore(E1,l)
+   local_irq_restore(l);
    return ...;
}
|
if (...)
+   {local_irq_restore(l);
    return ...;
+   }
|
spin_unlock_irqrestore(E2,l);
|
local_irq_restore(l);
)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>

---
 arch/mips/mm/tlb-r3k.c        |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/arch/mips/mm/tlb-r3k.c b/arch/mips/mm/tlb-r3k.c
index a782549..7d7e822 100644
--- a/arch/mips/mm/tlb-r3k.c
+++ b/arch/mips/mm/tlb-r3k.c
@@ -247,6 +247,7 @@ void __init add_wired_entry(unsigned long entrylo0, unsigned long entrylo1,
 		w = read_c0_wired();
 		write_c0_wired(w + 1);
 		if (read_c0_wired() != w + 1) {
+			local_irq_restore(flags);
 			printk("[tlbwired] No WIRED reg?\n");
 			return;
 		}

[PATCH 2/2] : Add local_irq_restore in error handling code

[Julia Lawall]

From: Julia Lawall <julia@diku.dk>

There is a call to local_irq_restore in the normal exit case, so it would
seem that there should be one on an error return as well.

The semantic patch that makes this change is as follows:
(http://www.emn.fr/x-info/coccinelle/)

// <smpl>
@@
expression l;
expression E,E1,E2;
@@

local_irq_save(l);
... when != local_irq_restore(l)
    when != spin_unlock_irqrestore(E,l)
    when any
    when strict
(
if (...) { ... when != local_irq_restore(l)
               when != spin_unlock_irqrestore(E1,l)
+   local_irq_restore(l);
    return ...;
}
|
if (...)
+   {local_irq_restore(l);
    return ...;
+   }
|
spin_unlock_irqrestore(E2,l);
|
local_irq_restore(l);
)
// </smpl>

Signed-off-by: Julia Lawall <julia@diku.dk>

---
 arch/mips/mm/tlb-r3k.c        |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/arch/mips/mm/tlb-r3k.c b/arch/mips/mm/tlb-r3k.c
index a782549..7d7e822 100644
--- a/arch/mips/mm/tlb-r3k.c
+++ b/arch/mips/mm/tlb-r3k.c
@@ -247,6 +247,7 @@ void __init add_wired_entry(unsigned long entrylo0, unsigned long entrylo1,
 		w = read_c0_wired();
 		write_c0_wired(w + 1);
 		if (read_c0_wired() != w + 1) {
+			local_irq_restore(flags);
 			printk("[tlbwired] No WIRED reg?\n");
 			return;
 		}

Re: [PATCH 2/2] : Add local_irq_restore in error handling code

[Ralf Baechle]

Hi Julia,

On Tue, Jul 22, 2008 at 06:44:48PM +0200, Julia Lawall wrote:

> From: Julia Lawall <julia@diku.dk>
> 
> There is a call to local_irq_restore in the normal exit case, so it would
> seem that there should be one on an error return as well.
> 
> The semantic patch that makes this change is as follows:
> (http://www.emn.fr/x-info/coccinelle/)

Correctly spotted - but I decieded to go for below patch instead.

Thanks,

  Ralf

From: Ralf Baechle <ralf@linux-mips.org>

[MIPS] tlb-r4k: Nuke broken paranoia error test.

Bug originally found and reported by Julia Lawall <julia@diku.dk>.  I
decieded that the whole error check was mostly useless paranoia and should
be discarded.  It would only ever trigger if r3k_have_wired_reg has a wrong
value.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

diff --git a/arch/mips/mm/tlb-r3k.c b/arch/mips/mm/tlb-r3k.c
index a782549..f0cf46a 100644
--- a/arch/mips/mm/tlb-r3k.c
+++ b/arch/mips/mm/tlb-r3k.c
@@ -246,10 +246,6 @@ void __init add_wired_entry(unsigned long entrylo0, unsigned long entrylo1,
 		old_pagemask = read_c0_pagemask();
 		w = read_c0_wired();
 		write_c0_wired(w + 1);
-		if (read_c0_wired() != w + 1) {
-			printk("[tlbwired] No WIRED reg?\n");
-			return;
-		}
 		write_c0_index(w << 8);
 		write_c0_pagemask(pagemask);
 		write_c0_entryhi(entryhi);

Re: [PATCH 2/2] : Add local_irq_restore in error handling code

[Ralf Baechle]

Hi Julia,

On Tue, Jul 22, 2008 at 06:44:48PM +0200, Julia Lawall wrote:

> From: Julia Lawall <julia@diku.dk>
> 
> There is a call to local_irq_restore in the normal exit case, so it would
> seem that there should be one on an error return as well.
> 
> The semantic patch that makes this change is as follows:
> (http://www.emn.fr/x-info/coccinelle/)

Correctly spotted - but I decieded to go for below patch instead.

Thanks,

  Ralf

From: Ralf Baechle <ralf@linux-mips.org>

[MIPS] tlb-r4k: Nuke broken paranoia error test.

Bug originally found and reported by Julia Lawall <julia@diku.dk>.  I
decieded that the whole error check was mostly useless paranoia and should
be discarded.  It would only ever trigger if r3k_have_wired_reg has a wrong
value.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>

diff --git a/arch/mips/mm/tlb-r3k.c b/arch/mips/mm/tlb-r3k.c
index a782549..f0cf46a 100644
--- a/arch/mips/mm/tlb-r3k.c
+++ b/arch/mips/mm/tlb-r3k.c
@@ -246,10 +246,6 @@ void __init add_wired_entry(unsigned long entrylo0, unsigned long entrylo1,
 		old_pagemask = read_c0_pagemask();
 		w = read_c0_wired();
 		write_c0_wired(w + 1);
-		if (read_c0_wired() != w + 1) {
-			printk("[tlbwired] No WIRED reg?\n");
-			return;
-		}
 		write_c0_index(w << 8);
 		write_c0_pagemask(pagemask);
 		write_c0_entryhi(entryhi);