Messages in this thread | | | From | "Gustavo A. R. Silva" <> | Subject | Re: [GIT PULL] Wimplicit-fallthrough patches for 5.3-rc1 | Date | Thu, 18 Jul 2019 15:28:48 -0500 |
| |
On 7/10/19 4:36 PM, Stephen Rothwell wrote: > Hi Gustavo, > > On Wed, 10 Jul 2019 13:14:10 -0500 "Gustavo A. R. Silva" <gustavo@embeddedor.com> wrote: >> >> At some point during this development cycle, we reached the quota of zero >> fall-through warnings, but people continued introducing such warnings. So, >> it seems we are now pretty much ready for enabling -Wimplicit-fallthrough >> globally. Before it turns into a never ending story. :) > > Sounds good to me. My mail was, I guess, just a heads up to Linus that > he will see some new warnings in his test build if he merges your > tree. Thanks for addressing them. > Yep. Thanks, Stephen.
Linus:
After you have merged all the trees containing fall-through patches, you might see the following warnings:
arch/x86/events/intel/core.c:4957:8: warning: this statement may fall through [-Wimplicit-fallthrough=] arch/x86/events/intel/core.c:5006:8: warning: this statement may fall through [-Wimplicit-fallthrough=] drivers/mtd/nand/onenand/onenand_base.c:3261:6: warning: this statement may fall through [-Wimplicit-fallthrough=]
for which I already have patches ready to be applied, but I didn't include them in my pull-request because such patches don't apply to 5.2-rc2, on which I based my -next tree for v5.3.
We can coordinate and I can send you the patches that address those and any other warning that you might see after merging my pull-request and just before you release 5.3-rc1. So we can have the -Wimplicit-fallthrough option globally enabled in 5.3-rc1 and zero fall-through warnings.
What do you think?
Thanks -- Gustavo
| |