Re: [PATCH] [v2] ata: ahci: fix enum constants for gcc-13

[Arnd Bergmann]

On Wed, Dec 7, 2022, at 07:41, Jiri Slaby wrote:
> Arnd,
>
> I just noticed this on stable@. Do you have more of the gcc-13-enum 
> patches? I sent some (this one incl.), but didn't have time for v2 of 
> some of them. So should I respin the rest or have you fixed them all yet?

I only have this one, as this is the only one that Luis asked me
about. I haven't actually tried using gcc-13 myself yet.

>  > [PATCH] ath11k (gcc13): synchronize 
> ath11k_mac_he_gi_to_nl80211_he_gi()'s return type - "Jiri Slaby (SUSE)" 
> <jirislaby@xxxxxxxxxx> - 2022-10-31 1243.eml:Message-Id: 
> <20221031114341.10377-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] block_blk-iocost (gcc13): cast enum members to int in prints 
> - "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1245.eml:Message-Id: <20221031114520.10518-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] bonding (gcc13): synchronize bond_{a,t}lb_xmit() types - 
> "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1244.eml:Message-Id: <20221031114409.10417-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] drm_amd_display (gcc13): fix enum mismatch - "Jiri Slaby 
> (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 1242.eml:Message-Id: 
> <20221031114247.10309-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] drm_nouveau_kms_nv50- (gcc13): fix nv50_wndw_new_ prototype - 
> "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1242.eml:Message-Id: <20221031114229.10289-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] init: Kconfig (gcc13): disable -Warray-bounds on gcc-13 too - 
> "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1242.eml:Message-Id: <20221031114212.10266-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] i40e (gcc13): synchronize allocate_free functions return type 
> & values - "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1244.eml:Message-Id: <20221031114456.10482-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] qed (gcc13): use u16 for fid to be big enough - "Jiri Slaby 
> (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 1243.eml:Message-Id: 
> <20221031114354.10398-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] RDMA_srp (gcc13): force int types for max_send_sge and 
> can_queue - "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1245.eml:Message-Id: <20221031114506.10501-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] sfc (gcc13): synchronize ef100_enqueue_skb()'s return type - 
> "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1244.eml:Message-Id: <20221031114440.10461-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] thunderbolt (gcc13): synchronize tb_port_is_clx_enabled()'s 
> 2nd param - "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1243.eml:Message-Id: <20221031114323.10356-1-jirislaby@xxxxxxxxxx>
>  > [PATCH] wireguard (gcc13): cast enum limits members to int in prints 
> - "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1244.eml:Message-Id: <20221031114424.10438-1-jirislaby@xxxxxxxxxx>
>  > [PATCH 1_2] ata: ahci (gcc13): use BIT() for bit definitions in enum 
> - "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1243.eml:Message-Id: <20221031114310.10337-1-jirislaby@xxxxxxxxxx>
>  > [PATCH 2_2] ata: ahci (gcc13): use U suffix for enum definitions - 
> "Jiri Slaby (SUSE)" <jirislaby@xxxxxxxxxx> - 2022-10-31 
> 1243.eml:Message-Id: <20221031114310.10337-2-jirislaby@xxxxxxxxxx>

I had a look, these all look good to me, as expected. My guess is that
these all only cause a build failure with -Werror, so Luis didn't
get stuck on the warnings.

Some of them appear to overlap with work that Nathan Huckleberry
did a while ago to fix related clang-14 warnings, see e.g.
https://lore.kernel.org/lkml/20220912214523.929094-1-nhuck@xxxxxxxxxx/

I don't know what the state of those is.

    Arnd