Re: [resend PATCH] coccinelle: Add devm_kmalloc_objs conversion script

[Joe Perches <joe@perches.com>]

On 2026-11-01 15:51, Julia Lawall wrote:
> On Sun, 1 Nov 2026, Joe Perches wrote:
> 
>> Like commit bdc5071d7f7b ("coccinelle: Add kmalloc_objs conversion 
>> script")
>> with commit 1c7bbaeed110 ("coccinelle: kmalloc_obj: Remove default 
>> GFP_KERNEL arg")
>> 
>> Finds and converts sized devm_kmalloc-family of allocations into the 
>> typed
>> devm_kmalloc_obj-family of allocations.
> 
> Applied
> 
>> 
>> Signed-off-by: Joe Perches <joe@perches.com>
>> ---
>> 
>> Maybe this might be useful to run and apply just prior to some -rc1
>> 
>> This script would change more than 8000 instances in more than
>> 5000 files in Linus' current master.
>> 
>> $ spatch -D patch --very-quiet --cocci-file 
>> ./scripts/coccinelle/api/devm_kmalloc_objs.cocci \
>>     --include-headers-for-types --all-includes --include-headers 
>> --keep-comments \
>>     -I ./arch/x86/include -I ./arch/x86/include/generated -I ./include 
>> \
>>     -I ./arch/x86/include/uapi -I ./arch/x86/include/generated/uapi -I 
>> ./include/uapi \
>>     -I ./include/generated/uapi --include 
>> ./include/linux/compiler-version.h \
>>     --include ./include/linux/kconfig.h .
> 
> I'm not sure that --include-headers-for-types is useful.  That is 
> useful
> when there is an expression metavariable that is required to have a
> particular type.  I don't see that here.  I also got around 5000 
> changes
> in Linus's current master without using that argument.  I didn't try 
> it,
> but I think --no-includes would thus be fine.  --include-headers will 
> get
> the changes in the header files.
> 
> If the --keep comments argument is useful, it could be added to the
> semantic patch, by putting #spatch --keep-comments at the top of the 
> file.

Ok. Whatever is appropriate.

I just copied the command line from the one that applied the commits 
above.