On 23 December 2013 20:15, Richard Henderson <r...@twiddle.net> wrote: > Ping.
I made it through as far as patch 38, but really I think this series is just way too big to review as a single set. There doesn't seem to be any particularly significant reason for it to be a single set either. I'll try to work through the remaining 23 patches later this week, but for future stuff I'd definitely appreciate more but shorter patch series.. I noticed a few checkpatch warnings (mostly missing braces, spaces at end of line, overlong lines), generally where you're touching code that's already noncompliant. You might like to go through and fix those. A note on these MO_ op values: it's kinda confusing that sometimes we use 8/16/32/64 and sometimes BWLQ (particularly obvious in patches that eg switch from MO_8 &c to MO_SB &c). Personally I'd rather we dropped the BWLQ versions entirely in favour of _8/_16/_32/_64. "word" in particular is horrifically ambiguous -- my default assumption is that "word" means "32 bits", but for QEMU it's 16. This is just a general observation, not a request for a change in any of these patches, though. The patch removing gen_op_movl_T1_im says "T0" in the subject. Otherwise, patches 1 to 38 Reviewed-by: Peter Maydell <peter.mayd...@linaro.org> thanks -- PMM
