Skip to content

Commit

Permalink
spufs: use new vm_ops->access to allow local state access from gdb
Browse files Browse the repository at this point in the history
This uses the new vm_ops->access to allow gdb to access the SPU local
store.  We currently prevent access to problem state registers, this can
be done later if really needed but it's safer not to.

[akpm@linux-foundation.org: fix typo]
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: Dave Airlie <airlied@linux.ie>
Cc: Hugh Dickins <hugh@veritas.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
  • Loading branch information
Benjamin Herrenschmidt authored and Linus Torvalds committed Jul 24, 2008
1 parent a1f242f commit a352894
Showing 1 changed file with 23 additions and 0 deletions.
23 changes: 23 additions & 0 deletions arch/powerpc/platforms/cell/spufs/file.c
Original file line number Diff line number Diff line change
Expand Up @@ -288,9 +288,32 @@ spufs_mem_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
return VM_FAULT_NOPAGE;
}

static int spufs_mem_mmap_access(struct vm_area_struct *vma,
unsigned long address,
void *buf, int len, int write)
{
struct spu_context *ctx = vma->vm_file->private_data;
unsigned long offset = address - vma->vm_start;
char *local_store;

if (write && !(vma->vm_flags & VM_WRITE))
return -EACCES;
if (spu_acquire(ctx))
return -EINTR;
if ((offset + len) > vma->vm_end)
len = vma->vm_end - offset;
local_store = ctx->ops->get_ls(ctx);
if (write)
memcpy_toio(local_store + offset, buf, len);
else
memcpy_fromio(buf, local_store + offset, len);
spu_release(ctx);
return len;
}

static struct vm_operations_struct spufs_mem_mmap_vmops = {
.fault = spufs_mem_mmap_fault,
.access = spufs_mem_mmap_access,
};

static int spufs_mem_mmap(struct file *file, struct vm_area_struct *vma)
Expand Down

0 comments on commit a352894

Please sign in to comment.