On 20 October 2015 at 17:31, Kyrill Tkachov <kyrylo.tkac...@arm.com> wrote: > > On 20/10/15 17:28, Ramana Radhakrishnan wrote: >> >> On Tue, Oct 20, 2015 at 4:26 PM, Marcus Shawcroft >> <marcus.shawcr...@gmail.com> wrote: >>> >>> On 19 October 2015 at 14:57, Kyrill Tkachov <kyrylo.tkac...@arm.com> >>> wrote: >>> >>>> 2015-10-19 Kyrylo Tkachov <kyrylo.tkac...@arm.com> >>>> >>>> * config/aarch64/aarch64.md >>>> (*aarch64_fcvt<su_optab><GPF:mode><GPI:mode>2_mult): New pattern. >>>> * config/aarch64/aarch64-simd.md >>>> (*aarch64_fcvt<su_optab><VDQF:mode><fcvt_target>2_mult): Likewise. >>>> * config/aarch64/aarch64.c (aarch64_rtx_costs): Handle above >>>> patterns. >>>> (aarch64_fpconst_pow_of_2): New function. >>>> (aarch64_vec_fpconst_pow_of_2): Likewise. >>>> * config/aarch64/aarch64-protos.h (aarch64_fpconst_pow_of_2): >>>> Declare >>>> prototype. >>>> (aarch64_vec_fpconst_pow_of_2): Likewise. >>>> * config/aarch64/predicates.md (aarch64_fp_pow2): New predicate. >>>> (aarch64_fp_vec_pow2): Likewise. >>>> >>>> 2015-10-19 Kyrylo Tkachov <kyrylo.tkac...@arm.com> >>>> >>>> * gcc.target/aarch64/fmul_fcvt_1.c: New test. >>>> * gcc.target/aarch64/fmul_fcvt_2.c: Likewise. >>> >>> + char buf[64]; >>> + sprintf (buf, "fcvtz<su>\\t%%0.<Vtype>, %%1.<Vtype>, #%d", fbits); >>> >>> Prefer snprintf please >> >> Should we also update our coding standards for this ? i.e. use the non >> `n' versions of the string functions only if you have a very good >> reason. Even more so in the run time support libraries ! > > > My understanding is one should *really* use snprintf if any part of the > string comes from the user. So in this case it probably doesn't make a > difference. However, it's not hard to use the 'n' version and thinking > about the maximum length is probably a good practice to get into.
Using snprintf here means that if/when someone fettles around with the pattern in the future they won't silently over run the buffer that they forgot to resize for their new longer pattern. Although in the case the likely hood of that happening seems rather small it is good practice to simply avoid the none n version. Cheers /Marcus
