Code improvements #228
Code improvements #228
Conversation
| uint8 *cp; | ||
| uint32 lowpc, highpc, text_start; | ||
| uint32 lowpc, highpc; | ||
| uint32 text_start = 0; |
There was a problem hiding this comment.
The text_start was not initialised and when if (low_pc == 0 && high_pc == 0) { was false this had no initial value at p->text_start = text_start; on line 175
| # ifdef D | ||
| # undef D | ||
| # endif // D | ||
| #define D(x) ; |
There was a problem hiding this comment.
The compiler was complaining the the D is redefined
| if (res) { | ||
| struct Clib4Node c2n; | ||
| char varbuf[8] = {0}; | ||
| char uuid[UUID4_LEN + 1] = {0}; |
There was a problem hiding this comment.
Those were not used anywhere
| libBase->libNode.lib_Version = VERSION; | ||
| libBase->libNode.lib_Revision = REVISION; | ||
| libBase->libNode.lib_IdString = VSTRING; | ||
| libBase->libNode.lib_IdString = (char *)VSTRING; |
There was a problem hiding this comment.
The compiler was complaining for losing the const character
| int clib4ProcessCompare(const void *a, const void *b, void *udata); | ||
|
|
||
| static void _start_ctors(void (*__CTOR_LIST__[])(void)); | ||
| static void _start_dtors(void (*__DTOR_LIST__[])(void)); |
There was a problem hiding this comment.
These were set here but were not defined. These are defined and used in main.c, so I thought that they are not needed here.
There was a problem hiding this comment.
Revert this please. Don't touch them
There was a problem hiding this comment.
@afxgroup With my last commit I reverted those changes.
| int saw_digit, octets, ch; | ||
| unsigned char tmp[NS_INADDRSZ], *tp; | ||
| unsigned char tmp[NS_INADDRSZ] = {0}; | ||
| unsigned char *tp; |
There was a problem hiding this comment.
The tmp was not initialised and later, based on conditions, an initial value is required
| } | ||
| ret2 = vsnprintf(ptr, ret, format, ap); | ||
| if (ret2 < 0) { | ||
| free(ptr); |
| bp = (char *) malloc(malloc_size); | ||
| if (bp == NULL) { | ||
| __set_errno(ENOMEM); | ||
| free(buf.beg); |
library/unistd/common_pathconf.c
Outdated
| break; | ||
| case _PC_MAX_INPUT: { | ||
| uint32_t Bufsize; | ||
| uint32_t Bufsize = 0L; |
There was a problem hiding this comment.
The Bufsize was not initialised and later, based on conditions, an initial value is required, but I am not sure about the value I put here. Is that right?
There was a problem hiding this comment.
Should be UL, but it depends on the architecture if L is actually needed. int is normally a target's natural word size, but also can be as low as 16bits. Perhaps U would work?
There was a problem hiding this comment.
@3246251196 With my last commit, I changed this to 2048, because a few lines later (line 127) there is this comment ret = Bufsize; /* Default is 2048 bytes. */
| host_entity = gethostbyname(addr_host); | ||
| if (host_entity == NULL) { | ||
| // No ip found for hostname | ||
| free(ip); |
| printf( "Length in bytes of multibyte character %x: %u\n", pmb, len ); | ||
|
|
||
|
|
||
| free(pwcs); |
There was a problem hiding this comment.
memory leaks fixes
| free(pwcs); | ||
| } | ||
|
|
||
| free(pmb); |
| return; | ||
| } | ||
|
|
||
| ar[n] = tmp; |
There was a problem hiding this comment.
With these changes the following critical error is fixed, referring to this line.
Address of local auto-variable assigned to a function parameter.
|
so many leaks plugged...here's hoping |
|
Unfortuntely all those memory leaks are in tests file.. :( |
|
ah, ok shows how much I understand of coding |
|
Half of the memory leak fixes are in the clib4 library, and to be more specific in the following files |
4 participants
I did a few code improvements based on the compiler warnings produced and the cppcheck report after running it on the library folder.