Re: [PATCH net-next 2/3] net: hns3: add missing const qualifier to hclge_log_error() reg parameter

[Simon Horman <horms@kernel.org>]

On Thu, Aug 06, 2026 at 03:48:20PM +0800, Jijie Shao wrote:
> 
> on 2026/8/6 0:52, Simon Horman wrote:
> > This is an AI-generated review of your patch. The human sending this
> > email has considered the AI review valid, or at least plausible.
> > Full review at: https://netdev-ai.bots.linux.dev/sashiko/#/patchset/20260804130933.880552-2-shaojijie@huawei.com?part=2
> > ---
> > --- Patch [2]: [PATCH net-next 2/3] net: hns3: add missing const qualifier to hclge_log_error() reg parameter ---
> > commit c77e6cbb6d302d584ade8cd89a76f4750fe00694
> > Author: Jijie Shao <shaojijie@huawei.com>
> > 
> > net: hns3: add missing const qualifier to hclge_log_error() reg parameter
> > 
> > This adds a const qualifier to the reg parameter of hclge_log_error() in
> > the hns3 PF error handling code. The parameter is only read inside the
> > function, so the change itself is a straightforward const-correctness
> > cleanup.
> > 
> > > diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
> > > index 7e124e2c718da..6093a60d257b1 100644
> > > --- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
> > > +++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c
> > > @@ -1762,7 +1762,7 @@ static const struct hclge_hw_type_id hclge_hw_type_id_st[] = {
> > >   	},
> > >   };
> > > -static void hclge_log_error(struct device *dev, char *reg,
> > > +static void hclge_log_error(struct device *dev, const char *reg,
> > >   			    const struct hclge_hw_error *err,
> > >   			    u32 err_sts, unsigned long *reset_requests)
> > >   {
> > [Severity: Low]
> > [Finding: main-1-0]
> > [Sources: opus-5]
> > This isn't a bug in the code, but is the justification in the changelog
> > accurate?  The commit message says:
> > 
> >      Callers pass const strings, causing a compiler warning about
> >      discarding the 'const' qualifier.
> > 
> > hclge_log_error() is static, and every call site is in hclge_err.c.  They
> > all pass a bare string literal, for example in hclge_handle_mpf_ras_error():
> > 
> >      hclge_log_error(dev, "IMP_TCM_ECC_INT_STS",
> >                      &hclge_imp_tcm_ecc_int[0], status,
> >                      &ae_dev->hw_err_reset_req);
> > 
> > and similarly for "SSU_ECC_MULTI_BIT_INT_0" and the remaining register
> > names down to "PPU_PF_ABNORMAL_INT_ST" in the MSI-X handlers.
> > 
> > In C a string literal has type char[N] rather than const char[N], so
> > passing one to a char *reg parameter does not discard a qualifier, and the
> > kernel build does not enable -Wwrite-strings.  None of the call sites pass
> > a value held in a const char * lvalue (such as
> > struct hclge_mod_reg_info::reg_name), which is the case that would produce
> > the described diagnostic.
> > 
> > Could the changelog be reworded to describe this as a const-correctness
> > cleanup instead of a build warning fix?  As written it suggests a broken
> > build and reads like a stable backport candidate.
> 
> The warning is real in our internal build, which enables
> -Wwrite-strings. Would it be acceptable to keep the commit message
> as-is, or should I reword it to const-correctness cleanup and note
> the internal build context in the changelog?
> 
> This series is cleanup, not bugfix — no Fixes: tags, not intended for stable.

Thanks, I agree with your analysis.
Sorry for the noise on this one.

Reviewed-by: Simon Horman <horms@kernel.org>