Hi Wolfgang,
On Tue, Mar 6, 2012 at 11:51 AM, Wolfgang Denk wd@denx.de wrote:
Dear Simon Glass,
In message 1329285566-30386-1-git-send-email-sjg@chromium.org you wrote:
At present two parsers have similar code to execute commands. Also cmd_usage() is called all over the place. This series adds a single function which processes commands called cmd_process().
This new function understands return codes, and in particular CMD_RET_USAGE to indicate a usage error. So rather than calling cmd_usage() themselves, the command handlers can just return this error.
There appears to be a run_command2() which is used to run commands with the selected parser. This series changes this in two separate steps to just run_command(), and renames the old run_command() to builtin_run_command(). No one should call this outside main.c since if the hush parser is being used it is wrong to call it. The built-in parser code could move into a separate file perhaps in a future patch.
The overall series reduces code size on ARM by about 1KB on my ~160KB U-Boot text region when the hush parser is used, and around 60 bytes when it isn't.
I tested this on PPC (TQM5200 config):
add/remove: 2/0 grow/shrink: 3/40 up/down: 268/-740 (-472)
and ARM (m28evk config):
add/remove: 1/0 grow/shrink: 5/38 up/down: 212/-1404 (-1192)
so this looks good to me.
OK good. Hoping it sorts out the command success/failure problem you saw with the last version also.
Out of interest what tool are you using to display that information?
Regards, Simon
Best regards,
Wolfgang Denk
-- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de NOTE: The Most Fundamental Particles in This Product Are Held Together by a "Gluing" Force About Which Little is Currently Known and Whose Adhesive Power Can Therefore Not Be Permanently Guaranteed.