| 01:14:10 | bug1 | its cheaper to pass an int than a short or char as a function paramater |
| 01:14:27 | bug1 | but when your in the function its cheaper to declare a short or char than it is an int |
| 01:15:02 | bug1 | weird |
| 01:15:12 | mjn3 | hmm |
| 01:15:42 | bug1 | ive only just tried it in th unzip.c im working on |
| 01:16:18 | bug1 | i guess there are standard prototypes for function arguments, and its more common to use ints |
| 01:16:42 | bug1 | static unsigned int fill_bitbuffer(unsigned int bitbuffer, unsigned int *current, const unsigned int required) |
| 01:16:50 | bug1 | is smaller than |
| 01:17:03 | bug1 | static unsigned int fill_bitbuffer(unsigned int bitbuffer, unsigned char *current, const unsigned char required) |
| 01:17:14 | mjn3 | i've noticed that sometimes declaring a short or a char is cheaper than an int, and sometimes not. and it depends on the compiler to of course. as far as passing a parameter, that may not be that surprising, since the compiler needs to ensure the unused stuff is masked off |
| 01:17:32 | bug1 | smaller by 30 bytes |
| 01:20:32 | bug1 | unused stuff is masked off? so if you declare a char as an argument internally it might use an int and mark 3 bytes as being unuable ? |
| 01:21:22 | mjn3 | what i mean is that it needs to do a cast-to-char or cast-to-short on the val passed |
| 01:21:38 | bug1 | ahh |
| 01:21:58 | bug1 | yes in my example it would have had to do a cast |
| 01:22:07 | bug1 | (if i use a char) |
| 02:10:03 | andersee | is back (gone 05:04:11) |
| 05:54:47 | andersee | Evening Tim |
| 05:54:59 | andersee | BZFlag: did you end up making it to sllug? |
| 07:44:00 | BZFlag | yep. headed to bed now. |
| 07:44:27 | BZFlag | I'll likely show up Monday morning for setup. have to be in the office later though. |
| 07:44:41 | BZFlag | we'll see. next week could be busy in the office. |
| 07:44:43 | BZFlag | nite |
| 14:41:42 | mjn3 | BZFlag: i emailed the tmpfs/ramfs to you yesterday. just mentioning it in case i entered the wrong email address |
| 20:00:35 | Russ|werk | anyone notice any problems with todays cvs? |