Re: linux-next: build failure after merge of the net-next tree

Hi Geert,

Geert Uytterhoeven [off-list ref] writes:

On Fri, Oct 8, 2021 at 7:55 AM Kalle Valo [off-list ref] wrote:

quoted

Stephen Rothwell [off-list ref] writes:

quoted

After merging the net-next tree, today's linux-next build (xtensa,
m68k allmodconfig) failed like this:

In file included from <command-line>:0:0:
In function 'ath11k_peer_assoc_h_smps',
    inlined from 'ath11k_peer_assoc_prepare' at drivers/net/wireless/ath/ath11k/mac.c:2362:2:
include/linux/compiler_types.h:317:38: error: call to '__compiletime_assert_650' declared with attribute error: FIELD_GET: type of reg too small for mask
  _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
                                      ^
include/linux/compiler_types.h:298:4: note: in definition of macro '__compiletime_assert'
    prefix ## suffix();    \
    ^
include/linux/compiler_types.h:317:2: note: in expansion of macro '_compiletime_assert'
  _compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
  ^
include/linux/build_bug.h:39:37: note: in expansion of macro 'compiletime_assert'
 #define BUILD_BUG_ON_MSG(cond, msg) compiletime_assert(!(cond), msg)
                                     ^
include/linux/bitfield.h:52:3: note: in expansion of macro 'BUILD_BUG_ON_MSG'
   BUILD_BUG_ON_MSG((_mask) > (typeof(_reg))~0ull,  \
   ^
include/linux/bitfield.h:108:3: note: in expansion of macro '__BF_FIELD_CHECK'
   __BF_FIELD_CHECK(_mask, _reg, 0U, "FIELD_GET: "); \
   ^
drivers/net/wireless/ath/ath11k/mac.c:2079:10: note: in expansion of macro 'FIELD_GET'
   smps = FIELD_GET(IEEE80211_HE_6GHZ_CAP_SM_PS,
          ^

Caused by commit

  6f4d70308e5e ("ath11k: support SMPS configuration for 6 GHz")

Thanks for the report, weird that I don't see it on x86. I can't look at
this in detail now, maybe later today, but I wonder if the diff below
fixes the issue?

It seems to be related to passing "le16_to_cpu(sta->he_6ghz_capa.capa)".
Probably typeof(le16_to_cpu(sta->he_6ghz_capa.capa)) goes berserk.
le16_to_cpu() is a complex macro on big-endian. I had expected to see
a similar issue on powerpc, but I don't.
Using an intermediate "u16 capa = le16_to_cpu(sta->he_6ghz_capa.capa)"
fixes the problem.

quoted

At least it's cleaner than using FIELD_GET(), actually ath11k should be
cleaned up to use xx_get_bits() all over.

Kalle

diff --git a/drivers/net/wireless/ath/ath11k/mac.c b/drivers/net/wireless/ath/ath11k/mac.c
index d897020dd52d..3e7e569f284b 100644
--- a/drivers/net/wireless/ath/ath11k/mac.c
+++ b/drivers/net/wireless/ath/ath11k/mac.c

@@ -2076,8 +2076,8 @@ static void ath11k_peer_assoc_h_smps(struct ieee80211_sta *sta,
                smps = ht_cap->cap & IEEE80211_HT_CAP_SM_PS;
                smps >>= IEEE80211_HT_CAP_SM_PS_SHIFT;
        } else {
-               smps = FIELD_GET(IEEE80211_HE_6GHZ_CAP_SM_PS,
-                                le16_to_cpu(sta->he_6ghz_capa.capa));
+               smps = le16_get_bits(sta->he_6ghz_capa.capa,
+                                    IEEE80211_HE_6GHZ_CAP_SM_PS);
        }

Thanks, that works, too, so (compile)
Tested-by: Geert Uytterhoeven <geert@linux-m68k.org>

Thanks Geert, you helped a lot! I now submitted a patch to netdev:

https://patchwork.kernel.org/project/netdevbpf/patch/20211008143932.23884-1-kvalo@codeaurora.org/

Dave & Jakub, if you can please take the patch directly to net-next.

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches

From: Kalle Valo <redacted>
Date: Fri, 08 Oct 2021 17:43:39 +0300

Thanks Geert, you helped a lot! I now submitted a patch to netdev:

https://patchwork.kernel.org/project/netdevbpf/patch/20211008143932.23884-1-kvalo@codeaurora.org/

Dave & Jakub, if you can please take the patch directly to net-next.

Done.