mem_alignment.rst (2704B)
1================ 2Memory alignment 3================ 4 5Too many problems popped up because of unnoticed misaligned memory access in 6kernel code lately. Therefore the alignment fixup is now unconditionally 7configured in for SA11x0 based targets. According to Alan Cox, this is a 8bad idea to configure it out, but Russell King has some good reasons for 9doing so on some f***ed up ARM architectures like the EBSA110. However 10this is not the case on many design I'm aware of, like all SA11x0 based 11ones. 12 13Of course this is a bad idea to rely on the alignment trap to perform 14unaligned memory access in general. If those access are predictable, you 15are better to use the macros provided by include/asm/unaligned.h. The 16alignment trap can fixup misaligned access for the exception cases, but at 17a high performance cost. It better be rare. 18 19Now for user space applications, it is possible to configure the alignment 20trap to SIGBUS any code performing unaligned access (good for debugging bad 21code), or even fixup the access by software like for kernel code. The later 22mode isn't recommended for performance reasons (just think about the 23floating point emulation that works about the same way). Fix your code 24instead! 25 26Please note that randomly changing the behaviour without good thought is 27real bad - it changes the behaviour of all unaligned instructions in user 28space, and might cause programs to fail unexpectedly. 29 30To change the alignment trap behavior, simply echo a number into 31/proc/cpu/alignment. The number is made up from various bits: 32 33=== ======================================================== 34bit behavior when set 35=== ======================================================== 360 A user process performing an unaligned memory access 37 will cause the kernel to print a message indicating 38 process name, pid, pc, instruction, address, and the 39 fault code. 40 411 The kernel will attempt to fix up the user process 42 performing the unaligned access. This is of course 43 slow (think about the floating point emulator) and 44 not recommended for production use. 45 462 The kernel will send a SIGBUS signal to the user process 47 performing the unaligned access. 48=== ======================================================== 49 50Note that not all combinations are supported - only values 0 through 5. 51(6 and 7 don't make sense). 52 53For example, the following will turn on the warnings, but without 54fixing up or sending SIGBUS signals:: 55 56 echo 1 > /proc/cpu/alignment 57 58You can also read the content of the same file to get statistical 59information on unaligned access occurrences plus the current mode of 60operation for user space code. 61 62 63Nicolas Pitre, Mar 13, 2001. Modified Russell King, Nov 30, 2001.