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: <cf327d9f-eb29-9107-7fcb-6ae510c2abee@redhat.com>
Date:   Tue, 11 Oct 2022 11:36:06 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.3.1
Subject: Re: [PATCH v2 1/5] memblock tests: add range tests for
 memblock_alloc_exact_nid_raw
Content-Language: en-US
To:     Rebecca Mckeever <remckee0@gmail.com>
Cc:     Mike Rapoport <rppt@kernel.org>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
References: <cover.1664753534.git.remckee0@gmail.com>
 <c73a6fd7fc3fa4529a6e444b9602d9c63ded91fd.1664753534.git.remckee0@gmail.com>
 <b44eee1a-0d61-f5bb-7b0c-a01f13a0b297@redhat.com>
 <20221011041910.GA7782@sophie>
From:   David Hildenbrand <david@redhat.com>
Organization: Red Hat
In-Reply-To: <20221011041910.GA7782@sophie>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 11.10.22 06:19, Rebecca Mckeever wrote:
> On Mon, Oct 10, 2022 at 01:16:03PM +0200, David Hildenbrand wrote:
>> On 03.10.22 01:41, Rebecca Mckeever wrote:
>>> Add tests for memblock_alloc_exact_nid_raw() that are very similar to
>>> the range tests for memblock_alloc_try_nid_raw().
>>>
>>> Signed-off-by: Rebecca Mckeever <remckee0@gmail.com>
>>> ---
>>>    tools/testing/memblock/Makefile               |    2 +-
>>>    tools/testing/memblock/main.c                 |    2 +
>>>    .../memblock/tests/alloc_exact_nid_api.c      | 1208 +++++++++++++++++
>>>    .../memblock/tests/alloc_exact_nid_api.h      |    9 +
>>>    4 files changed, 1220 insertions(+), 1 deletion(-)
>>>    create mode 100644 tools/testing/memblock/tests/alloc_exact_nid_api.c
>>>    create mode 100644 tools/testing/memblock/tests/alloc_exact_nid_api.h
>>>
>>> diff --git a/tools/testing/memblock/Makefile b/tools/testing/memblock/Makefile
>>> index 246f7ac8489b..2310ac4d080e 100644
>>> --- a/tools/testing/memblock/Makefile
>>> +++ b/tools/testing/memblock/Makefile
>>> @@ -7,7 +7,7 @@ CFLAGS += -I. -I../../include -Wall -O2 -fsanitize=address \
>>>    LDFLAGS += -fsanitize=address -fsanitize=undefined
>>>    TARGETS = main
>>>    TEST_OFILES = tests/alloc_nid_api.o tests/alloc_helpers_api.o tests/alloc_api.o \
>>> -		  tests/basic_api.o tests/common.o
>>> +		  tests/basic_api.o tests/common.o tests/alloc_exact_nid_api.o
>>>    DEP_OFILES = memblock.o lib/slab.o mmzone.o slab.o
>>>    OFILES = main.o $(DEP_OFILES) $(TEST_OFILES)
>>>    EXTR_SRC = ../../../mm/memblock.c
>>> diff --git a/tools/testing/memblock/main.c b/tools/testing/memblock/main.c
>>> index 4ca1024342b1..278f9dec5008 100644
>>> --- a/tools/testing/memblock/main.c
>>> +++ b/tools/testing/memblock/main.c
>>> @@ -3,6 +3,7 @@
>>>    #include "tests/alloc_api.h"
>>>    #include "tests/alloc_helpers_api.h"
>>>    #include "tests/alloc_nid_api.h"
>>> +#include "tests/alloc_exact_nid_api.h"
>>>    #include "tests/common.h"
>>>    int main(int argc, char **argv)
>>> @@ -12,6 +13,7 @@ int main(int argc, char **argv)
>>>    	memblock_alloc_checks();
>>>    	memblock_alloc_helpers_checks();
>>>    	memblock_alloc_nid_checks();
>>> +	memblock_alloc_exact_nid_checks();
>>>    	return 0;
>>>    }
>>
>>
>> memblock_alloc_exact_nid_raw(NUMA_NO_NODE) behaves exactly the way
>> memblock_alloc_try_nid_raw(NUMA_NO_NODE) behaves -- which is essentially
>> memblock_alloc_raw().
>>
>> So do we really need a separate set of tests for these?
>>
> Instead of a separate set of tests, I could add a flag for
> memblock_alloc_exact_nid_raw() to test_flags so that the range tests in
> alloc_nid_api.c could be run with that flag. Do you think I should do
> that, or omit tests for memblock_alloc_exact_nid_raw(NUMA_NO_NODE)
> altogether?

Good question. From an API point of view, it makes sense to test that 
API somehow. With implementation in mind, it doesn't make too much sense 
to test it twice.

Whatever you prefer :) Just wanted to point out that separate tests 
don't make sense -- using a flag would be cleaner.

-- 
Thanks,

David / dhildenb