https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120513
--- Comment #31 from Greg Chandler <chandleg at wizardsworks dot org> ---
This hangs:
root@bigbang:/tmp# gcc -S 1.c
>From ps -faux:
root 391 0.0 0.1 4832 3912 ttyS0 Ss 03:28 0:00 -bash
root 432 0.4 0.1 6616 2512 ttyS0 S+ 03:57 0:00 \_ gcc -S 1.c
root 433 3.5 0.8 54176 17968 ttyS0 S+ 03:57 0:00 \_
/usr/libexec/gcc/alpha-linux-gnu/14.3.0/cc1
The ps list does not show the 1.c being passed to the cc1, and running that
alone will cause a wait for input like I have been seeing.
Like the prior compiles if run by hand with the correct flag:
root@bigbang:/tmp# /usr/libexec/gcc/alpha-linux-gnu/14.3.0/cc1 1.c
main
Analyzing compilation unit
Performing interprocedural optimizations
<*free_lang_data> {heap 824k} <visibility> {heap 824k} <build_ssa_passes>
{heap 824k} <opt_local_passes> {heap 1160k} <remove_symbols> {heap 1160k}
<targetclone> {heap 1160k} <free-fnsummary> {heap 1160k}Streaming LTO
<whole-program> {heap 1160k} <fnsummary> {heap 1160k} <inline> {heap 1160k}
<modref> {heap 1160k} <free-fnsummary> {heap 1160k} <single-use> {heap 1160k}
<comdats> {heap 1160k}Assembling functions:
main
Time variable usr sys wall
GGC
phase setup : 0.11 ( 59%) 0.14 ( 32%) 1.42 ( 46%)
1255k ( 88%)
phase parsing : 0.02 ( 9%) 0.07 ( 17%) 0.19 ( 6%)
112k ( 8%)
phase opt and generate : 0.06 ( 31%) 0.21 ( 50%) 1.45 ( 48%)
52k ( 4%)
callgraph construction : 0.00 ( 1%) 0.00 ( 0%) 0.11 ( 4%)
4528 ( 0%)
callgraph optimization : 0.00 ( 1%) 0.01 ( 3%) 0.16 ( 5%)
0 ( 0%)
callgraph ipa passes : 0.01 ( 4%) 0.08 ( 19%) 0.29 ( 9%)
3936 ( 0%)
ipa function summary : 0.00 ( 0%) 0.00 ( 0%) 0.05 ( 2%)
1952 ( 0%)
ipa inheritance graph : 0.00 ( 0%) 0.01 ( 3%) 0.02 ( 1%)
0 ( 0%)
rebuild jump labels : 0.00 ( 1%) 0.00 ( 0%) 0.06 ( 2%)
0 ( 0%)
preprocessing : 0.00 ( 2%) 0.06 ( 15%) 0.09 ( 3%)
9128 ( 1%)
parser (global) : 0.01 ( 8%) 0.00 ( 1%) 0.06 ( 2%)
102k ( 7%)
parser function body : 0.00 ( 1%) 0.00 ( 1%) 0.04 ( 1%)
1248 ( 0%)
inline parameters : 0.00 ( 1%) 0.06 ( 15%) 0.07 ( 2%)
544 ( 0%)
tree gimplify : 0.00 ( 1%) 0.00 ( 0%) 0.24 ( 8%)
1936 ( 0%)
tree CFG construction : 0.00 ( 1%) 0.01 ( 2%) 0.01 ( 0%)
1200 ( 0%)
tree CFG cleanup : 0.00 ( 1%) 0.00 ( 0%) 0.01 ( 0%)
0 ( 0%)
tree switch lowering : 0.00 ( 0%) 0.00 ( 0%) 0.03 ( 1%)
0 ( 0%)
out of ssa : 0.00 ( 1%) 0.04 ( 11%) 0.04 ( 1%)
184 ( 0%)
expand : 0.00 ( 0%) 0.00 ( 0%) 0.11 ( 4%)
3448 ( 0%)
integrated RA : 0.01 ( 6%) 0.00 ( 0%) 0.07 ( 2%)
12k ( 1%)
reload : 0.00 ( 0%) 0.00 ( 0%) 0.02 ( 1%)
56 ( 0%)
thread pro- & epilogue : 0.00 ( 1%) 0.00 ( 0%) 0.09 ( 3%)
2376 ( 0%)
final : 0.00 ( 2%) 0.06 ( 15%) 0.07 ( 2%)
2736 ( 0%)
initialize rtl : 0.03 ( 14%) 0.00 ( 0%) 0.28 ( 9%)
15k ( 1%)
TOTAL : 0.19 0.42 3.06
1420k
root@bigbang:/tmp# cat 1.s
.set noreorder
.set volatile
.set noat
.set nomacro
.arch ev4
.text
.align 2
.globl main
.ent main
$main..ng:
main:
.frame $15,16,$26,0
.mask 0x4008000,-16
$LFB0:
.cfi_startproc
lda $30,-16($30)
.cfi_def_cfa_offset 16
stq $26,0($30)
stq $15,8($30)
.cfi_offset 26, -16
.cfi_offset 15, -8
mov $30,$15
.cfi_def_cfa_register 15
.prologue 0
mov $31,$1
mov $1,$0
mov $15,$30
ldq $26,0($30)
ldq $15,8($30)
lda $30,16($30)
.cfi_restore 15
.cfi_restore 26
.cfi_def_cfa 30, 0
ret $31,($26),1
.cfi_endproc
$LFE0:
.end main
.ident "GCC: (GNU) 14.3.0"
.section .note.GNU-stack,"",@progbits
So I guess I at least have a place to start digging now. I'm hoping it's as
simple as the gcc driver not passing arguments to the compiler.
