On Tue, Jun 24, 2025 at 02:12:53PM +0300, Edvard Fagerholm wrote: > RTE_FORCE_INTRINSICS with a C++ compiler causes rte_atomic.h and > rte_byteorder.h to emit on x86 an extern "C" without a closing > brace. > ---
Unfortunately, patch is missing a signoff, so isn't eligible for merge. Can you please submit a v2 with the appropriate signoff included. See [1] for more details. For the patch itself, it looks ok to me, but really the various #endif and #else tags need comments on them to indicate which #ifdef they belong to (unless it's a very short ifdef). If you do a v2 would you consider adding those comments to help make the code clearer? With or without that suggestion, you can include my ack on v2. Acked-by: Bruce Richardson <bruce.richard...@intel.com> Regards, /Bruce [1] https://doc.dpdk.org/guides/contributing/patches.html > lib/eal/x86/include/rte_atomic.h | 4 ++-- > lib/eal/x86/include/rte_byteorder.h | 9 +++++---- > 2 files changed, 7 insertions(+), 6 deletions(-) > > diff --git a/lib/eal/x86/include/rte_atomic.h > b/lib/eal/x86/include/rte_atomic.h > index c72c47c83e..68629016b4 100644 > --- a/lib/eal/x86/include/rte_atomic.h > +++ b/lib/eal/x86/include/rte_atomic.h > @@ -282,10 +282,10 @@ static inline int > rte_atomic32_dec_and_test(rte_atomic32_t *v) > return ret != 0; > } > > -#ifdef __cplusplus > -} > #endif > > +#ifdef __cplusplus > +} > #endif > > #ifdef RTE_ARCH_I686 > diff --git a/lib/eal/x86/include/rte_byteorder.h > b/lib/eal/x86/include/rte_byteorder.h > index 5a49ffcd50..62a87c1aa6 100644 > --- a/lib/eal/x86/include/rte_byteorder.h > +++ b/lib/eal/x86/include/rte_byteorder.h > @@ -48,10 +48,6 @@ static inline uint32_t rte_arch_bswap32(uint32_t _x) > return x; > } > > -#ifdef __cplusplus > -} > -#endif > - > #define rte_bswap16(x) ((uint16_t)(__builtin_constant_p(x) ? \ > rte_constant_bswap16(x) : \ > rte_arch_bswap16(x))) > @@ -69,6 +65,11 @@ static inline uint32_t rte_arch_bswap32(uint32_t _x) > #else > #include "rte_byteorder_64.h" > #endif > + > +#endif > + > +#ifdef __cplusplus > +} > #endif > > #define rte_cpu_to_le_16(x) (x) > -- > 2.34.1 >
