While tracking down the reason for an ioremap() failure I was distracted
by the WARN_ONCE() in __ioremap_caller(). Can we move the sanity check
later in the function until _after_ the mapping has been performed?
rtg
--
Tim Gardner [email protected]
On Thu, 2011-04-28 at 10:00 -0700, Tim Gardner wrote:
> While tracking down the reason for an ioremap() failure I was
> distracted
> by the WARN_ONCE() in __ioremap_caller(). Can we move the sanity
> check
> later in the function until _after_ the mapping has been performed?
>
> rtg
> --
> Tim Gardner [email protected]
>
>
>
>
>
>
>
> differences
> between files
> attachment
> (0001-ioremap-Delay-sanity-check-until-after-a-successful-.patch)
>
> From 31dec327a254888fcd0b6aa963414b09626d3168 Mon Sep 17 00:00:00 2001
> From: Tim Gardner <[email protected]>
> Date: Thu, 28 Apr 2011 10:07:51 -0600
> Subject: [PATCH] ioremap: Delay sanity check until after a successful
> mapping
>
> Performing a WARN_ONCE() sanity check before the mapping
> is successful seems pointless if the caller sends bad values.
> A case in point is when the BIOS provides erroneous screen_info values
> causing vesafb_probe() to request an outrageuous size. The
> WARN_ONCE is then wasted on bogosity. Move the warning to a
> point where the mapping has been successfully allocated.
>
> http://bugs.launchpad.net/bugs/772042
Sounds good to me.
Reviewed-by: Suresh Siddha <[email protected]>
thanks.
Commit-ID: c7a7b814c9dca9ee01b38e63b4a46de87156d3b6
Gitweb: http://git.kernel.org/tip/c7a7b814c9dca9ee01b38e63b4a46de87156d3b6
Author: Tim Gardner <[email protected]>
AuthorDate: Thu, 28 Apr 2011 11:00:30 -0600
Committer: Ingo Molnar <[email protected]>
CommitDate: Fri, 29 Apr 2011 08:02:47 +0200
ioremap: Delay sanity check until after a successful mapping
While tracking down the reason for an ioremap() failure I was
distracted by the WARN_ONCE() in __ioremap_caller().
Performing a WARN_ONCE() sanity check before the mapping
is successful seems pointless if the caller sends bad values.
A case in point is when the BIOS provides erroneous screen_info
values causing vesafb_probe() to request an outrageuous size.
The WARN_ONCE is then wasted on bogosity. Move the warning to a
point where the mapping has been successfully allocated.
Addresses:
http://bugs.launchpad.net/bugs/772042
Reviewed-by: Suresh Siddha <[email protected]>
Signed-off-by: Tim Gardner <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
---
arch/x86/mm/ioremap.c | 14 +++++++-------
1 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c
index 0369843..be1ef57 100644
--- a/arch/x86/mm/ioremap.c
+++ b/arch/x86/mm/ioremap.c
@@ -91,13 +91,6 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr,
return (__force void __iomem *)phys_to_virt(phys_addr);
/*
- * Check if the request spans more than any BAR in the iomem resource
- * tree.
- */
- WARN_ONCE(iomem_map_sanity_check(phys_addr, size),
- KERN_INFO "Info: mapping multiple BARs. Your kernel is fine.");
-
- /*
* Don't allow anybody to remap normal RAM that we're using..
*/
last_pfn = last_addr >> PAGE_SHIFT;
@@ -170,6 +163,13 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr,
ret_addr = (void __iomem *) (vaddr + offset);
mmiotrace_ioremap(unaligned_phys_addr, unaligned_size, ret_addr);
+ /*
+ * Check if the request spans more than any BAR in the iomem resource
+ * tree.
+ */
+ WARN_ONCE(iomem_map_sanity_check(unaligned_phys_addr, unaligned_size),
+ KERN_INFO "Info: mapping multiple BARs. Your kernel is fine.");
+
return ret_addr;
err_free_area:
free_vm_area(area);