Serious bugs in socket.c

Question

github.com/Wiznet/W7500P/issues/3 This code : github.com/Wiznet/W7500P/blob/m … t/socket.c ctlsocket() and getsockopt() were seriously broken without the ‘break;’ statements. Hope you patch this soon and also check other versions of this code… I can not understand how these kind of errors can go unnoticed for so long, no peer review? Good luck, Edwin van den Oetelaar diff -r e377caf76636 W7500P/W7500P_FW/ioLibrary/Ethernet/socket.c --- a/W7500P/W7500P_FW/ioLibrary/Ethernet/socket.c Tue Apr 05 21:28:05 2016 +0200 +++ b/W7500P/W7500P_FW/ioLibrary/Ethernet/socket.c Sun Apr 10 17:20:23 2016 +0200 @@ -52,12 +52,12 @@ #include "socket.h" #include "W7500x_wztoe.h" -#define SOCK_ANY_PORT_NUM 0xC000; +#define SOCK_ANY_PORT_NUM (0xC000) static uint16_t sock_any_port = SOCK_ANY_PORT_NUM; static uint16_t sock_io_mode = 0; static uint16_t sock_is_sending = 0; -static uint16_t sock_remained_size[_WIZCHIP_SOCK_NUM_] = {0,0,}; +static uint16_t sock_remained_size[_WIZCHIP_SOCK_NUM_] = {0,}; static uint8_t sock_pack_info[_WIZCHIP_SOCK_NUM_] = {0,}; #if _WIZCHIP_ == 5200 @@ -589,6 +589,7 @@ int8_t ctlsocket(uint8_t sn, ctlsock_ty break; case CS_GET_INTMASK: *((uint8_t*)arg) = getSn_IMR(sn); + break; default: return SOCKERR_ARG; } @@ -664,6 +665,7 @@ int8_t getsockopt(uint8_t sn, sockopt_t break; case SO_MSS: *(uint8_t*) arg = getSn_MSSR(sn); + break; case SO_DESTIP: getSn_DIPR(sn, (uint8_t*)arg); break; @@ -678,8 +680,10 @@ int8_t getsockopt(uint8_t sn, sockopt_t #endif case SO_SENDBUF: *(uint16_t*) arg = getSn_TX_FSR(sn); + break; case SO_RECVBUF: *(uint16_t*) arg = getSn_RX_RSR(sn); + break; case SO_STATUS: *(uint8_t*) arg = getSn_SR(sn); break;

lawrence · Answer

Hello, oetelaar Thank you for your interest in W7500P. And So sorry not to notice your issue. We will update soon. I checked ioLibrary in Gethub right after it. And fortunately, it is correct. So I guess someone had mistakes during socket.c in W7500P porting from ioLibrary. Thank you again! lawrence