Re: [PATCH net-next] net: freescale: use ethtool string helpers

[Michael Ellerman <mpe@ellerman.id.au>]

Rosen Penev <rosenp@gmail.com> writes:
> On Fri, Oct 25, 2024 at 5:57 AM Simon Horman <horms@kernel.org> wrote:
>>
>> On Thu, Oct 24, 2024 at 01:52:57PM -0700, Rosen Penev wrote:
>> > The latter is the preferred way to copy ethtool strings.
>> >
>> > Avoids manually incrementing the pointer. Cleans up the code quite well.
>> >
>> > Signed-off-by: Rosen Penev <rosenp@gmail.com>
>>
>> ...
>>
>> > diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
>> > index b0060cf96090..10c5fa4d23d2 100644
>> > --- a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
>> > +++ b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
>> > @@ -243,38 +243,24 @@ static void dpaa_get_ethtool_stats(struct net_device *net_dev,
>> >  static void dpaa_get_strings(struct net_device *net_dev, u32 stringset,
>> >                            u8 *data)
>> >  {
>> > -     unsigned int i, j, num_cpus, size;
>> > -     char string_cpu[ETH_GSTRING_LEN];
>> > -     u8 *strings;
>> > +     unsigned int i, j, num_cpus;
>> >
>> > -     memset(string_cpu, 0, sizeof(string_cpu));
>> > -     strings   = data;
>> > -     num_cpus  = num_online_cpus();
>> > -     size      = DPAA_STATS_GLOBAL_LEN * ETH_GSTRING_LEN;
>> > +     num_cpus = num_online_cpus();
>> >
>> >       for (i = 0; i < DPAA_STATS_PERCPU_LEN; i++) {
>> > -             for (j = 0; j < num_cpus; j++) {
>> > -                     snprintf(string_cpu, ETH_GSTRING_LEN, "%s [CPU %d]",
>> > -                              dpaa_stats_percpu[i], j);
>> > -                     memcpy(strings, string_cpu, ETH_GSTRING_LEN);
>> > -                     strings += ETH_GSTRING_LEN;
>> > -             }
>> > -             snprintf(string_cpu, ETH_GSTRING_LEN, "%s [TOTAL]",
>> > -                      dpaa_stats_percpu[i]);
>> > -             memcpy(strings, string_cpu, ETH_GSTRING_LEN);
>> > -             strings += ETH_GSTRING_LEN;
>> > -     }
>> > -     for (j = 0; j < num_cpus; j++) {
>> > -             snprintf(string_cpu, ETH_GSTRING_LEN,
>> > -                      "bpool [CPU %d]", j);
>> > -             memcpy(strings, string_cpu, ETH_GSTRING_LEN);
>> > -             strings += ETH_GSTRING_LEN;
>> > +             for (j = 0; j < num_cpus; j++)
>> > +                     ethtool_sprintf(&data, "%s [CPU %d]",
>> > +                                     dpaa_stats_percpu[i], j);
>> > +
>> > +             ethtool_sprintf(&data, "%s [TOTAL]", dpaa_stats_percpu[i]);
>> >       }
>> > -     snprintf(string_cpu, ETH_GSTRING_LEN, "bpool [TOTAL]");
>> > -     memcpy(strings, string_cpu, ETH_GSTRING_LEN);
>> > -     strings += ETH_GSTRING_LEN;
>> > +     for (i = 0; j < num_cpus; i++)
>>
>> Perhaps this should consistently use i, rather than i and j:
>>
>>         for (i = 0; i < num_cpus; i++)
>>
>> Flagged by W=1 builds with clang-18.

> I really need to compile test this on a PPC system.

Cross compiling should be sufficient.

There's some pointers here:
  https://github.com/linuxppc/wiki/wiki/Building-powerpc-kernels

Or there's also libc-less cross compilers on kernel.org, eg:
  https://mirrors.edge.kernel.org/pub/tools/crosstool/files/bin/x86_64/14.2.0/x86_64-gcc-14.2.0-nolibc-powerpc64-linux.tar.xz


cheers