On 4/23/25 09:17, Philippe Mathieu-Daudé wrote:
> On 23/4/25 18:09, Pierrick Bouvier wrote:
>> On 4/23/25 03:27, Philippe Mathieu-Daudé wrote:
>>> On 22/4/25 21:27, Richard Henderson wrote:
>>>> From: Pierrick Bouvier <pierrick.bouvier@linaro.org>
>>>>
>>>> Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
>>>> Signed-off-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
>>>> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
>>>> Message-ID: <20250325045915.994760-7-pierrick.bouvier@linaro.org>
>>>> ---
>>>> include/exec/cpu-all.h | 1 -
>>>> target/ppc/mmu-hash64.h | 2 ++
>>>> target/i386/tcg/system/excp_helper.c | 1 +
>>>> target/i386/xsave_helper.c | 1 +
>>>> target/riscv/vector_helper.c | 1 +
>>>> 5 files changed, 5 insertions(+), 1 deletion(-)
>>>>
>>>> diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h
>>>> index 957c86886e..bfa039ab76 100644
>>>> --- a/include/exec/cpu-all.h
>>>> +++ b/include/exec/cpu-all.h
>>>> @@ -21,7 +21,6 @@
>>>> #include "exec/cpu-common.h"
>>>> #include "exec/cpu-interrupt.h"
>>>> -#include "exec/tswap.h"
>>>> #include "hw/core/cpu.h"
>>>> #include "exec/cpu-defs.h"
>>>> #include "exec/target_page.h"
>>>> diff --git a/target/ppc/mmu-hash64.h b/target/ppc/mmu-hash64.h
>>>> index ae8d4b37ae..b8fb12a970 100644
>>>> --- a/target/ppc/mmu-hash64.h
>>>> +++ b/target/ppc/mmu-hash64.h
>>>> @@ -1,6 +1,8 @@
>>>> #ifndef MMU_HASH64_H
>>>> #define MMU_HASH64_H
>>>> +#include "exec/tswap.h"
>>>
>>> This header doesn't use "exec/tswap.h". Maybe you want this
>>> reviewed patch instead:
>>>
>>> https://lore.kernel.org/qemu-devel/20241211230357.97036-6-
>>> philmd@linaro.org/
>>>
>>>> +
>>>> #ifndef CONFIG_USER_ONLY
>>>> #ifdef TARGET_PPC64
>>>
>>
>> See: [PATCH 002/147] exec/tswap: implement {ld,st}.*_p as functions
>> instead of macros.
>> It removes macro definitions from cpu-all.h, and implement those as
>> proper functions in exec/tswap.h.
>>
>> ppc/mmu-hash64.h uses ld*_p, which are defined in exec/tswap.h, so it's
>> the correct place for this include.
>
> Ah I missed that, OK then.
>
>
No worries, it's a bit confusing with all the variants we have, and I
had to reproduce the compile error to find what was the rationale, so it
makes sense you missed it.