Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <bac3f11b-db5f-113f-9cc3-8abf0e8e6ed6@redhat.com>
Date:   Tue, 10 Jan 2023 11:05:38 +0100
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.6.0
To:     Mirsad Goran Todorovac <mirsad.todorovac@alu.unizg.hr>,
        linux-mm@kvack.org
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        Shuah Khan <shuah@kernel.org>, linux-kselftest@vger.kernel.org,
        linux-kernel@vger.kernel.org
References: <0f117203-3227-cd16-61c2-2dd3de75ecc7@alu.unizg.hr>
 <0ee389dc-5e47-5b7e-4db5-637eb2b3fbc9@redhat.com>
 <0e692e52-0a4f-3892-ed25-f3fa12892b6f@alu.unizg.hr>
Content-Language: en-US
From:   David Hildenbrand <david@redhat.com>
Organization: Red Hat
Subject: =?UTF-8?Q?Re=3a_PROBLEM=3a_selftest/vm/cow=2ec_failed_to_compile_?=
 =?UTF-8?B?KOKAmE1BRFZfUEFHRU9VVOKAmSB1bmRlY2xhcmVkKQ==?=
In-Reply-To: <0e692e52-0a4f-3892-ed25-f3fa12892b6f@alu.unizg.hr>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 09.01.23 22:41, Mirsad Goran Todorovac wrote:
> On 1/9/2023 5:42 PM, David Hildenbrand wrote:
> 
> Hi, thank you for your reply.
> 
>>> I hope this is enough information for you to debug the issue.
>>>
>>> I am standing by for any additional diagnostics needed.
>>
>> Won't userfaultfd.c fail in a similar way?
>>
>> Anyhow, khugepaged.c jas
>>
>> #ifndef MADV_PAGEOUT
>> #define MADV_PAGEOUT 21
>> #endif
>>
>> So most probably we should do the same.
> 
> Actually, David, it turned out that userfaultfd.c compiled
> out-of-the-box, and side-by-side comparison showed that it also included
> "/home/marvin/linux/kernel/linux_torvalds/usr/include/asm-generic/mman-common.h"
> 
> The only remaining difference was including <linux/mman.h>, which fixed
> the issue w/o #ifdef ... #endif
> 
> Hope this helps.
> 
> Please find the following diff.
> 
> Regards,
> Mirsad
> 
> ------------------------------------------------------------------------------
> diff --git a/tools/testing/selftests/vm/cow.c
> b/tools/testing/selftests/vm/cow.c
> index 26f6ea3079e2..dd8cf12c6776 100644
> --- a/tools/testing/selftests/vm/cow.c
> +++ b/tools/testing/selftests/vm/cow.c
> @@ -16,6 +16,7 @@
>    #include <fcntl.h>
>    #include <dirent.h>
>    #include <assert.h>
> +#include <linux/mman.h>
>    #include <sys/mman.h>
>    #include <sys/ioctl.h>
>    #include <sys/wait.h>
>

I already sent a different fix [1]. I suspect when including
linux/mman.h, it would still be problematic with older kernel
headers that lack MADV_PAGEOUT (< v5.4).


But yeah, I saw that userfaultfd.c was fixed that way:

commit b773827e361952b3f53ac6fa4c4e39ccd632102e
Author: Chengming Zhou <zhouchengming@bytedance.com>
Date:   Fri Mar 4 20:29:04 2022 -0800

     kselftest/vm: fix tests build with old libc
     
     The error message when I build vm tests on debian10 (GLIBC 2.28):
     
         userfaultfd.c: In function `userfaultfd_pagemap_test':
         userfaultfd.c:1393:37: error: `MADV_PAGEOUT' undeclared (first use
         in this function); did you mean `MADV_RANDOM'?
           if (madvise(area_dst, test_pgsize, MADV_PAGEOUT))
                                              ^~~~~~~~~~~~
                                              MADV_RANDOM
     
     This patch includes these newer definitions from UAPI linux/mman.h, is
     useful to fix tests build on systems without these definitions in glibc
     sys/mman.h.


[1] https://lkml.kernel.org/r/20230109171255.488749-1-david@redhat.com

-- 
Thanks,

David / dhildenb