Hey,
Steffen Nurpmeso wrote in
<09f36b9e8c5793662b0f788b3396d02e7f1cb9ed.1661902741.git.steffen@sdaoden\
.eu>:
|The former implementation was not correct regarding whiteouts in
|?: conditional branches. The new one also parses a bit better, in
|effect on equal level than bash with its recursive descendent parser.
I am quite sorry on how this was happening. Anyhow, i stumbled
upon a bug that somehow did not trigger for months
- su_ALIGNOF(sizeof(*sasp->sas_ops) * 2))
+ (su_ALIGNOF(*sasp->sas_ops) * 2))
a_SHEXP_ARITH_ERROR_TRACK(
+ sizeof(*sasp->sas_error_track) * 2 ))
)){
@@ -433,7 +432,7 @@ jenomem:
++i;
j = su_ALIGNOF(*sasp->sas_nums) * (i >> 1);
- o = su_ALIGNOF(sizeof(*sasp->sas_ops) * i);
+ o = su_ALIGNOF(*sasp->sas_ops) * i;
Two out of four were used falsely, and i did not see it.
Does not trigger with my default build of busybox, but did trigger
with my MUA upon syntax error (r=(xa ((512 + 511) & ~511) >> 9)).
I did then change to "normal alignment directives", saving some
alloc space.
Anyhow -- i was so shocked i immediately embedded the entire
header into math.c, using tabulator indentation and up to ~120
columns, ripping off practically all compat shims. 'Had to remove
the >>> unsigned right shift operator to get away with only
arith_t, and its support.
Not yet ready, 'want to have this running a few days.
But geee, i was _so_ thrilled to come back to tabulators after
seeing how nice it can be to work with a 120{-140} column screen,
and how many bytes that saves! Only did 80 (79) for all my life!!
'Only working with ~<30% of my screen width. I immediately
changed two old projects of mine accordingly, and think will turn
all to the new era of wide screens. So it is.
Ciao!
--steffen
|
|Der Kragenbaer, The moon bear,
|der holt sich munter he cheerfully and one by one
|einen nach dem anderen runter wa.ks himself off
|(By Robert Gernhardt)
_______________________________________________
busybox mailing list
[email protected]
http://lists.busybox.net/mailman/listinfo/busybox
