search for: overflow_arg_area

http://llvm.org/bugs/show_bug.cgi?id=14792 Problem: In the i386 ABI Page 3-10, it said that the stack is aligned. However, the two example code show that does not handle the alignment correctly when using variadic function. For example, if the size of the first argument is 17, the overflow_arg_area in va_list will be set to "address of first argument + 16" instead of "address of first argument + 24" after calling va_start. In addition, #6636 showed the same problem because in AMD64, arguments is passed by register at first, then pass by memory when run out of register (A...

...end of this mail. A VAARG call requires a lot of computation. Typically the C code for va_arg(ap, int) is: int va_arg_gpr(ap_list ap) { int idx = ap->gpr; if (idx < 8) { ap->gpr = idx + 1; return ap->reg_save_area[idx]; } else { int res = ap->overflow_arg_area[0]; ap->gpr = idx + 1; ap->overflow_arg_area += 4; return res; } } Actually, all VAARG implementations for other backends use DAGs, but this may be too complicated to do for ELF32. What would you suggest to code the VAARG instruction efficiently? Should I code va_arg_...

...br i1 %cmp7, label %for.body.lr.ph, label %for.end for.body.lr.ph: ; preds = %entry %gp_offset_p = getelementptr inbounds [1 x %struct.__va_list_tag]* %ap, i64 0, i64 0, i32 0 %0 = getelementptr inbounds [1 x %struct.__va_list_tag]* %ap, i64 0, i64 0, i32 3 %overflow_arg_area_p = getelementptr inbounds [1 x %struct.__va_list_tag]* %ap, i64 0, i64 0, i32 2 %gp_offset.pre = load i32* %gp_offset_p, align 16 br label %for.body for.body: ; preds = %vaarg.end, % for.body.lr.ph %gp_offset = phi i32 [ %gp_offset.pre, %for.body.lr.p...

...In the i386 ABI Page 3-10, it said that the stack is aligned. However, >> > the >> > two example code show that does not handle the alignment correctly when >> > using variadic function. For example, if the size of the first argument >> > is >> > 17, the overflow_arg_area in va_list will be set to "address of first >> > argument + 16" instead of "address of first argument + 24" after calling >> > va_start. >> > In addition, #6636 showed the same problem because in AMD64, arguments >> > is >> > passed...

...n C, are you seeing llvm.vaarg in the output .ll file? -Chris > is: > > int va_arg_gpr(ap_list ap) { > int idx = ap->gpr; > if (idx < 8) { > ap->gpr = idx + 1; > return ap->reg_save_area[idx]; > } > else { > int res = ap->overflow_arg_area[0]; > ap->gpr = idx + 1; > ap->overflow_arg_area += 4; > return res; > } > } > > Actually, all VAARG implementations for other backends use DAGs, but > this may be > too complicated to do for ELF32. > > What would you suggest to code the VAAR...

...gt; -Chris > > >> is: >> >> int va_arg_gpr(ap_list ap) { >> int idx = ap->gpr; >> if (idx < 8) { >> ap->gpr = idx + 1; >> return ap->reg_save_area[idx]; >> } >> else { >> int res = ap->overflow_arg_area[0]; >> ap->gpr = idx + 1; >> ap->overflow_arg_area += 4; >> return res; >> } >> } >> >> Actually, all VAARG implementations for other backends use DAGs, but >> this may be >> too complicated to do for ELF32. >> >...

...c:657 status = 0 #4 0x00007f64dab30d4e in i_panic (format=format at entry=0x7f64dae79848 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:267 ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0} args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7fffb9d88df0, reg_save_area = 0x7fffb9d88d30}} #5 0x00007f64dae59fbc in mail_index_keywords_unref (_keywords=<optimized out>) at mail-index.c:380 keywords = <optimized out> __FUNCTION__ = "mail_index_keywords_unref" #6 0x00007f64dae22657 in mailbox_co...

...status = 0 #4 0x00007f8bb654ac39 in i_panic (format=0x7f8bb65ba898 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:275 ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0, timestamp_usecs = 0} args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7ffef81dab50, reg_save_area = 0x7ffef81daa90}} #5 0x00007f8bb656bcb0 in io_remove_full (_io=0x220a100, closed=false) at ioloop.c:126 io = 0x22694a0 __FUNCTION__ = "io_remove_full" #6 0x00007f8bb656be31 in io_remove (io=0x220a100) at ioloop.c:169 No locals. #7 0x...

...???????? status = 0 #4? 0x00007f8bb654ac39 in i_panic (format=0x7f8bb65ba898 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:275 ???????? ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0, timestamp_usecs = 0} ???????? args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7ffef81dab50, reg_save_area = 0x7ffef81daa90}} #5? 0x00007f8bb656bcb0 in io_remove_full (_io=0x220a100, closed=false) at ioloop.c:126 ???????? io = 0x22694a0 ???????? __FUNCTION__ = "io_remove_full" #6? 0x00007f8bb656be31 in io_remove (io=0x220a100) at ioloop.c:169 No locals. #7? 0x...

...a in i_internal_fatal_handler (type=LOG_TYPE_PANIC, status=0, fmt=<value optimized out>, args=<value optimized out>) at failures.c:593 No locals. #4 0x00002b5e8240d163 in i_panic (format=<value optimized out>) at failures.c:230 args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7fffbc0b8d40, reg_save_area = 0x7fffbc0b8c80}} #5 0x00002b5e83e2ee6d in squat_build_word (ctx=0x127776e0, uid=574, data=0x126ef79f "\a\b\021?\n", char_lengths=0x126ef4b7 "\001\001\001\004\001\001", size=5) at squat-trie.c:876 trie = (struct squat_trie *) 0x12780...

....so.6 No symbol table info available. #1 0x00007f9dd8cc6732 in vsnprintf () from /lib/libc.so.6 No symbol table info available. #2 0x00007f9dd904d0db in str_vprintfa (str=0x11aa4f8, fmt=0x409184 "%u %s", args=0x7ffffab4fff0) at str.c:155 args2 = {{gp_offset = 16, fp_offset = 48, overflow_arg_area = 0x7ffffab500e0, reg_save_area = 0x7ffffab50010}} init_size = 4231558 pos = 0 ret = <value optimized out> ret2 = <value optimized out> __FUNCTION__ = "str_vprintfa" #3 0x00000000004055ff in client_send_line (client=0x11d1e50, fmt=<v...

Hi, I've got report about issue when dovecot fails to start and there is no error logged (error goes only to stderr) situation: 1) dovecot is running 2) dovecot is automatically updated to new version (by yum update daemon), after update, dovecot is restarted (it's part of update script) 3) new dovecot fails to start (for whatever reason) result: dovecot not running and no error

...tus = 0 #4 0x00007f497f080fde in i_panic (format=format at entry=0x7f497f47af00 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:275 ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0, timestamp_usecs = 0} args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7ffc829fd450, reg_save_area = 0x7ffc829fd390}} #5 0x00007f497f3cb6dc in mailbox_notify_changes_stop (box=<optimized out>) at mail-storage.c:1950 __FUNCTION__ = "mailbox_notify_changes_stop" #6 0x00007f497dc5aa6f in virtual_notify_changes (box=0x5636400cdbd8) at virtual...

Hello, I'm using the Debian builds of dovecot-2.2 by S[r]us. Today I started getting the following panic: log: Panic: file priorityq.c: line 122 (priorityq_remove_idx): assertion failed: (idx < count) log: Error: Raw backtrace: /usr/lib/dovecot/libdovecot.so.0(+0x7f1ae) [0x7fd368dc31ae] -> /usr/lib/dovecot/libdovecot.so.0(+0x7f269) [0x7fd368dc3269] ->

...32\177\000\000\001\000\000\000\000\000\000\000\211\001\000\000\000\000\000\000\260\006}S\375\177\000\000\240\257\235\003\000\000\000\000\260\006}S\375\177\000\000\200"... workstart = 0x0 workend = 0x7ffd537d0748 "" ap_save = {{gp_offset = 8, fp_offset = 48, overflow_arg_area = 0x7ffd537d0b60, reg_save_area = 0x7ffd537d0aa0}} nspecs_done = 2 save_errno = 0 readonly_format = 0 args_malloced = 0x0 specs = 0xa00000001 specs_malloced = false jump_table = "\001\000\000\004\000\016\000\006\000\000\a\002\000...

...10} sigs = {__val = {32, 0 <repeats 15 times>}} #2 0x00007fdb930e2c00 in __libc_message (do_abort=do_abort at entry=2, fmt=fmt at entry=0x7fdb931d7d98 "*** Error in `%s': %s: 0x%s ***\n") at ../sysdeps/posix/libc_fatal.c:175 ap = {{gp_offset = 40, fp_offset = 0, overflow_arg_area = 0x7ffdc467c420, reg_save_area = 0x7ffdc467c3b0}} fd = 2 on_2 = <optimized out> list = <optimized out> nlist = <optimized out> cp = <optimized out> written = <optimized out> #3 0x00007fdb930e8fc6 in malloc_printerr (act...

...0x00007f497f080fde in i_panic (format=format at entry=0x7f497f47af00 "file %s: line %d (%s): assertion failed: (%s)") at failures.c:275 > ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0, timestamp_usecs = 0} > args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7ffc829fd450, reg_save_area = 0x7ffc829fd390}} > #5 0x00007f497f3cb6dc in mailbox_notify_changes_stop (box=<optimized out>) at mail-storage.c:1950 > __FUNCTION__ = "mailbox_notify_changes_stop" > #6 0x00007f497dc5aa6f in virtual_notify_changes (box=0x5636400cd...

Hello! I'm having crashes with LMTP delivery when user is over quota on the latest CentOS 7.4 with the latest Dovecot 2.3.0.1 from Dovecot repo. I see the issue has been fixed on January 17, but it doesn't seem to have made it into 2.3.0.1 (I compared with the source from https://dovecot.org/releases/2.3/dovecot-2.3.0.1.tar.gz).

...was_recursion = 0 i = <value optimized out> #3 0x00002af7ef2cf433 in IA__g_log ( log_domain=0x52a9 <Address 0x52a9 out of bounds>, log_level=21161, format=0x6 <Address 0x6 out of bounds>) at gmessages.c:517 args = {{gp_offset = 40, fp_offset = 48, overflow_arg_area = 0x7fffbf5d95e0, reg_save_area = 0x7fffbf5d9520}} #4 0x00002aaaafbea39a in swfdec_as_frame_return (frame=0x36a5d80, return_value=0x7fffbf5d9690) at swfdec_as_frame.c:472 __o = <value optimized out> context = (SwfdecAsContext *) 0x1bfa750 retval = {type = 5730240...

...s = 0 > #11 0x00007f73f1723e11 in i_panic (format=0x1310 <Address 0x1310 out of > bounds>) at failures.c:306 > ??????? ctx = {type = LOG_TYPE_PANIC, exit_status = 0, timestamp = 0x0, > timestamp_usecs = 0, log_prefix = 0x0} > ??????? args = {{gp_offset = 16, fp_offset = 48, overflow_arg_area = > 0x7ffd0ddd84a0, reg_save_area = 0x7ffd0ddd83e0}} > #12 0x00007f73f17ab83a in mem_block_alloc (min_size=512) at > data-stack.c:360 > ??????? block = <value optimized out> > ??????? prev_size = <value optimized out> > ??????? alloc_size = 134217728 > #13 0x00...