Re: [PATCH 00/12] Compile QEMU with -Wimplicit-fallthrough

[Thomas Huth]

On 11/12/2020 16.24, Thomas Huth wrote:
> Coverity is already reporting switch-case statements where code
> can fall through from one case to another without a proper comment
> (since this could indicate a missing "break" and thus a bug).
> However, it's cumbersome to fix these issues after they have been
> merged already, it would be better if the author of the code would
> already take care of this when writing the patch. Fortunately,
> GCC and Clang can already warn about those code spots, too.
> So let's fix our remaining statements that fall through without
> a proper comment, so we can finally turn on -Wimplicit-fallthrough
> for all compilation runs.
> 
> Chen Qun (6):
>   hw/timer/renesas_tmr: silence the compiler warnings
>   target/i386: silence the compiler warnings in gen_shiftd_rm_T1
>   hw/intc/arm_gicv3_kvm: silence the compiler warnings
>   accel/tcg/user-exec: silence the compiler warnings
>   target/sparc/translate: silence the compiler warnings
>   target/sparc/win_helper: silence the compiler warnings
> 
> Thomas Huth (6):
>   disas/libvixl: Fix fall-through annotation for GCC >= 7
>   target/unicore32/translate: Add missing fallthrough annotations
>   hw/rtc/twl92230: Silence warnings about missing fallthrough statements
>   tcg/optimize: Add fallthrough annotations
>   tests/fp: Do not emit implicit-fallthrough warnings in the softfloat
>     tests
>   configure: Compile with -Wimplicit-fallthrough=2

I forgot to mention: There are two more spots in the ppc code that currently
cause some warnings, but the patches to fix those are already in David
Gibson's ppc pull request, so they should be handled soon.

 Thomas