Skip to contentSkip to conversation

dspbridge: kill camelcase

ANNOUNCE Omar Ramirez Luna 3 participants 4 messages Feb 24, 2010

lore
Share & tools
lore
#0 lore raw

Hi All,

I'm currently working on removing camel case + hungarian notation from dspbridge code in the following branch:

dspbridge-next-camelcase on d.o-z

I'll appreciate if you can mail any unsent patch otherwise it will need to wait (and be rebased) after the camelcase changes.

If nobody disagrees I'll be marking dspbridge 0.2 (as per private discussion) in the upcoming days.

Thanks,

Omar

#1 lore raw
Quoted 6 lines · Omar Ramirez Luna

Hi All,

I'm currently working on removing camel case + hungarian notation from dspbridge code in the following branch:

Quoted 7 lines

dspbridge-next-camelcase on d.o-z

I'll appreciate if you can mail any unsent patch otherwise it will need to wait (and be rebased) after the camelcase changes.

If nobody disagrees I'll be marking dspbridge 0.2 (as per private discussion) in the upcoming days.

Maybe do a perl or sed script that allows you to convert one string at a time? That way you can regenerate the patches as needed.

We used something like that to convert the musb code a few years ago.

I recommend converting one variable at a time and then compile and boot test in between. Otherwise you can easily convert substrings accidentally and then things will not compile..

Regards,

Tony

#2 lore raw

Tony Lindgren said the following on 02/24/2010 05:13 AM:

Quoted 18 lines · Omar Ramirez Luna
Quoted 5 lines

Hi All,

I'm currently working on removing camel case + hungarian notation from dspbridge code in the following branch:

Quoted 7 lines

dspbridge-next-camelcase on d.o-z

I'll appreciate if you can mail any unsent patch otherwise it will need to wait (and be rebased) after the camelcase changes.

If nobody disagrees I'll be marking dspbridge 0.2 (as per private discussion) in the upcoming days.

Maybe do a perl or sed script that allows you to convert one string at a time? That way you can regenerate the patches as needed.

Quoted 6 lines

We used something like that to convert the musb code a few years ago.

I recommend converting one variable at a time and then compile and boot test in between. Otherwise you can easily convert substrings accidentally and then things will not compile..

I think we discussed this quiet some time back:
http://marc.info/?t=125319739500001&r=1&w=2 - yep there is script available.

Signature · 2 lines
Regards,
Nishanth Menon
#3 lore raw

Hi,

Quoted 22 lines · Nishanth

Tony Lindgren said the following on 02/24/2010 05:13 AM:

Quoted 21 lines · Omar Ramirez Luna
Quoted 5 lines

Hi All,

I'm currently working on removing camel case + hungarian notation from dspbridge code in the following branch:

Quoted 7 lines

dspbridge-next-camelcase on d.o-z

I'll appreciate if you can mail any unsent patch otherwise it will need to wait (and be rebased) after the camelcase changes.

If nobody disagrees I'll be marking dspbridge 0.2 (as per private discussion) in the upcoming days.

Maybe do a perl or sed script that allows you to convert one string at a time? That way you can regenerate the patches as needed.

We used something like that to convert the musb code a few years ago.

Quoted 8 lines
Quoted 3 lines

I recommend converting one variable at a time and then compile and boot test in between. Otherwise you can easily convert substrings accidentally and then things will not compile..

I think we discussed this quiet some time back:
http://marc.info/?t=125319739500001&r=1&w=2 - yep there is script available.

Tony,

Yes, it is not a bad idea to generate single patches, probably I can create another branch if needed for each change, in my laziness I decided to replace safe stuff in one shot.

Nishanth,

I have been playing with the script, which is great to parse the tags file and create the new tokens, but not so hot for replacing stuff.

If you have time, you can criticize my rationale ;)

http://dev.omapzoom.org/?p=tidspbridge/kernel-dspbridge.git;a=commit;h=7c0065d4219b7b4ba15709870ef9fb819f7e71c8

All,

As of know, with the safe symbols (this means symbols are not duplicated in bridge code), I had to fix:

dspbridge: conflicting types for 'mem_init'
dspbridge: variable or field 'nop' declared void
dspbridge: conflicting types for 'clk_init'
dspbridge: conflicting types for 'io_schedule'
dspbridge: conflicting types for 'find_symbol'

These are duplicated on kernel sources so renamed:
mem_init ==> services_mem_init
clk_init ==> services_clk_init
io_schedule ==> iosm_schedule
find_symbol ==> dbll_find_symbol

Right now count stands:

151 files changed, 12579 insertions(+), 12579 deletions(-)

I'll now proceed to rename the symbols classified as "dangerous", because might get duplicated with new replacements.

Thanks,

- omar