Commit 7a1ab6a2 authored by John David Anglin's avatar John David Anglin Committed by Greg Kroah-Hartman

parisc: Fix race in pci-dma.c

commit c0452fb9 upstream.

We are still troubled by occasional random segmentation faults and
memory memory corruption on SMP machines.  The causes quite a few
package builds to fail on the Debian buildd machines for parisc.  When
gcc-6 failed to build three times in a row, I looked again at the TLB
related code.  I found a couple of issues.  This is the first.

In general, we need to ensure page table updates and corresponding TLB
purges are atomic.  The attached patch fixes an instance in pci-dma.c
where the page table update was not guarded by the TLB lock.

Tested on rp3440 and c8000.  So far, no further random segmentation
faults have been observed.
Signed-off-by: default avatarJohn David Anglin  <dave.anglin@bell.net>
Signed-off-by: default avatarHelge Deller <deller@gmx.de>
Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
parent e541fd81
...@@ -95,8 +95,8 @@ static inline int map_pte_uncached(pte_t * pte, ...@@ -95,8 +95,8 @@ static inline int map_pte_uncached(pte_t * pte,
if (!pte_none(*pte)) if (!pte_none(*pte))
printk(KERN_ERR "map_pte_uncached: page already exists\n"); printk(KERN_ERR "map_pte_uncached: page already exists\n");
set_pte(pte, __mk_pte(*paddr_ptr, PAGE_KERNEL_UNC));
purge_tlb_start(flags); purge_tlb_start(flags);
set_pte(pte, __mk_pte(*paddr_ptr, PAGE_KERNEL_UNC));
pdtlb_kernel(orig_vaddr); pdtlb_kernel(orig_vaddr);
purge_tlb_end(flags); purge_tlb_end(flags);
vaddr += PAGE_SIZE; vaddr += PAGE_SIZE;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment