Open XVilka opened 4 years ago
can you update?
or just share a link to the last log
Last build log: https://travis-ci.com/github/radareorg/radare2/jobs/351993035 Seems the amount of bugs increased
https://travis-ci.com/github/radareorg/radare2/jobs/358672729
p/../arch/8051/8051_disas.c:64:14: warning: ‘r_8051_disas’ defined but not used [-Wunused-function]
static char *r_8051_disas(ut64 pc, const ut8 *buf, int len, int *olen) {
^~~~~~~~~~~~
/home/travis/build/radareorg/radare2/libr/..//libr/anal/p/../..//asm/arch/tms320/c55x_plus/hashvector.c:97:25: warning: ‘ins_hash’ defined but not used [-Wunused-variable]
static HASHCODE_ENTRY_T ins_hash[] = {
^~~~~~~~
p/debug_native.c: In function ‘r_debug_native_bp’:
p/debug_native.c:1450:10: warning: unused variable ‘dbg’ [-Wunused-variable]
RDebug *dbg = bp->user;
^~~
p/debug_native.c: At top level:
p/debug_native.c:1649:2: warning: #warning Unsupported architecture [-Wcpp]
#warning Unsupported architecture
^~~~~~~
p/debug_native.c:1260:13: warning: ‘set_drx_regs’ defined but not used [-Wunused-function]
static void set_drx_regs (RDebug *dbg, drxt *regs, size_t num_regs) {
^~~~~~~~~~~~
p/debug_native.c:1234:13: warning: ‘sync_drx_regs’ defined but not used [-Wunused-function]
static void sync_drx_regs (RDebug *dbg, drxt *regs, size_t num_regs) {
^~~~~~~~~~~~~
p/native/linux/linux_debug.c: In function ‘print_fpu’:
p/native/linux/linux_debug.c:971:2: warning: #warning print_fpu not implemented for this platform [-Wcpp]
#warning print_fpu not implemented for this platform
^~~~~~~
p/native/linux/linux_debug.c: In function ‘linux_reg_read’:
p/native/linux/linux_debug.c:1065:3: warning: #warning getfpregs not implemented for this platform [-Wcpp]
#warning getfpregs not implemented for this platform
^~~~~~~
p/native/linux/linux_debug.c:1080:54: warning: passing argument 4 of ‘r_debug_ptrace’ makes pointer from integer without a cast [-Wint-conversion]
ret = r_debug_ptrace (dbg, PTRACE_GETREGSET, pid, 1, &io);
^
In file included from p/native/linux/linux_debug.c:6:0:
/home/travis/build/radareorg/radare2/libr/include/r_debug.h:618:20: note: expected ‘void *’ but argument is of type ‘int’
static inline long r_debug_ptrace(RDebug *dbg, r_ptrace_request_t request, pid_t pid, void *addr, r_ptrace_data_t data) {
^~~~~~~~~~~~~~
p/native/linux/linux_debug.c:976:7: warning: variable ‘showfpu’ set but not used [-Wunused-but-set-variable]
bool showfpu = false;
^~~~~~~
At top level:
p/native/linux/linux_debug.c:883:13: warning: ‘print_fpu’ defined but not used [-Wunused-function]
static void print_fpu(void *f){
^~~~~~~~~
cmd.c: In function ‘r_core_cmd_subst’:
cmd.c:2583:8: warning: variable ‘cmt’ set but not used [-Wunused-but-set-variable]
char *cmt, *colon = NULL, *icmd = NULL;
^~~
canal.c: In function ‘is_skippable_addr’:
canal.c:1943:45: warning: passing argument 3 of ‘r_list_find’ from incompatible pointer type [-Wincompatible-pointer-types]
return !(flags && r_list_find (flags, fcn, find_sym_flag));
^~~~~~~~~~~~~
In file included from canal.c:4:0:
/home/travis/build/radareorg/radare2/libr/include/r_list.h:115:18: note: expected ‘RListComparator {aka int (*)(const void *, const void *)}’ but argument is of type ‘int (*)(void *, void *)’
R_API RListIter *r_list_find(const RList *list, const void *p, RListComparator cmp);
^~~~~~~~~~~
It seems hugely broken now:
[**] db/anal/arm 11076 OK 927 BR 1808 XX 20 FX
Looks like problem might be somewhere in IO layer:
test_dwarf4_cpp_multiple_modules ERR
[XX] Fail at line 176: Wrong attribute data: expected 1, got 72057594037927936.
test_graph_to_agraph ERR
[XX] Fail at line 34: Wrong node out-nodes: expected 2, got 4.
test_r_buf_sparse2 ERR
[XX] Fail at line 251: right 10 bytes: expected ff ff ff 61 61 61 61 ff ff ff, got ff ff ff 61 61 61 61 ff ff 00
Warning: File is not PE
test_parse_dwarf_types_version2 ERR
[XX] Fail at line 24: pe/vista-glass.exe binary could not be opened
and so on for various DWARF and PDB failures, @HoundThe please take a look - here are the logs: https://travis-ci.com/github/radareorg/radare2/jobs/370233920
Seems that some tests fail at r_bin_open()
and from a first look it seems that some readings have endianess issues. I'll have a deeper look
I've just now noticed that DWARF has custom macros to read that looks like they care about endianness, but they assume little endian everywhere
#define READ32(x) \
(((x) + sizeof (ut32) < buf_end) ? r_read_ble32 (x, 0) : 0); \
(x) += sizeof (ut32)
thats bad
These are gone now, but the amount of broken tests is still roughly the same
The most common reason of failure it seems this one:
+WARNING: No calling convention defined for this file, analysis may be inaccurate.
+WARNING: No calling convention defined for this file, analysis may be inaccurate.
Cannot find file
Cannot find file
Cannot find file
https://travis-ci.com/github/radareorg/radare2/jobs/432758538#L43967
cc @eagleoflqj
See full log here: https://travis-ci.com/github/radareorg/radare2/jobs/351993035