The problem is introduced by commit 2332616 (exec: Support 64-bit operations in address_space_rw, 2013-07-08). Before that commit, memory_access_size would only return 1/2/4.
Since alignment is already handled above, reduce l to the largest power of two that is smaller than l. Cc: [email protected] Reported-by: Oleksii Shevchuk <[email protected]> Tested-by: Oleksii Shevchuk <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> --- exec.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/exec.c b/exec.c index 87b0b39..b52ec80 100644 --- a/exec.c +++ b/exec.c @@ -1913,6 +1913,9 @@ static int memory_access_size(MemoryRegion *mr, unsigned l, hwaddr addr) if (l > access_size_max) { l = access_size_max; } + if (l & (l - 1)) { + l = 1 << (qemu_fls(l) - 1); + } return l; } -- 1.8.3.1
