qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
To: Max Reitz <mreitz@redhat.com>, qemu-block@nongnu.org
Cc: qemu-devel@nongnu.org, fam@euphon.net, stefanha@redhat.com,
	kwolf@redhat.com, eesposit@redhat.com, pbonzini@redhat.com
Subject: Re: [PATCH v2 3/9] test-write-threshold: rewrite test_threshold_(not_)trigger tests
Date: Wed, 5 May 2021 18:20:20 +0300	[thread overview]
Message-ID: <880d2a4a-68ae-fdb8-05d6-357ff74b32a1@virtuozzo.com> (raw)
In-Reply-To: <f082362f-ec9e-9367-eede-9df7d14ef85b@redhat.com>

05.05.2021 17:28, Max Reitz wrote:
> On 04.05.21 10:25, Vladimir Sementsov-Ogievskiy wrote:
>> These tests use bdrv_write_threshold_exceeded() API, which is used only
>> for test.
> 
> Well, now.  That used to be different before patch 1.
> 
>> Better is testing real API, which is used in block.c as well.
>>
>> So, let's call bdrv_write_threshold_check_write(), and check is
>> bs->write_threshold_offset cleared or not (it's cleared iff threshold
>> triggered).
>>
>> Also we get rid of BdrvTrackedRequest use here. Tracked requests are
>> unrelated to write-threshold since we get rid of write notifiers.
> 
> The purpose behind the BdrvTrackedRequest was clearly because this is the object bdrv_write_threshold_exceeded() expected.  This reads like there was some other purpose (i.e. actually tracked requests), but there wasn’t.
> 
> The question that comes to my mind is why we had the bdrv_check_request() calls there, and why this patch removes them.  They seem unrelated to the write threshold, but someone must have thought something when adding them.

I wanted to add a note for it but forget. Something like:

   "Such small requests are obviously good, no reason to check them" :)

> 
> Looking into git blame, that someone is you :) (8b1170012b1)

Oops:) When I read your "but someone must have thought something", I really doubt in it :) But that was me, and I really thought something. Respect for your punctuality!

> Looks like you added those calls because BdrvTrackedRequest is a block layer structure, so getting rid of it means the reason for bdrv_check_request() disappears.  OK.

Yes, I was worried about the fact that BdrvTrackedRequest is a public structure and is a potential backdoor for not-checked offset/bytes things. At some moment we'd better close it (hide structure and add some interfaces).

> 
> Reviewed-by: Max Reitz <mreitz@redhat.com>
> 
>> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
>> ---
>>   tests/unit/test-write-threshold.c | 22 ++++------------------
>>   1 file changed, 4 insertions(+), 18 deletions(-)
>>
>> diff --git a/tests/unit/test-write-threshold.c b/tests/unit/test-write-threshold.c
>> index fc1c45a2eb..fd40a815b8 100644
>> --- a/tests/unit/test-write-threshold.c
>> +++ b/tests/unit/test-write-threshold.c
>> @@ -55,41 +55,27 @@ static void test_threshold_multi_set_get(void)
>>   static void test_threshold_not_trigger(void)
>>   {
>> -    uint64_t amount = 0;
>>       uint64_t threshold = 4 * 1024 * 1024;
>>       BlockDriverState bs;
>> -    BdrvTrackedRequest req;
>>       memset(&bs, 0, sizeof(bs));
>> -    memset(&req, 0, sizeof(req));
>> -    req.offset = 1024;
>> -    req.bytes = 1024;
>> -
>> -    bdrv_check_request(req.offset, req.bytes, &error_abort);
>>       bdrv_write_threshold_set(&bs, threshold);
>> -    amount = bdrv_write_threshold_exceeded(&bs, &req);
>> -    g_assert_cmpuint(amount, ==, 0);
>> +    bdrv_write_threshold_check_write(&bs, 1024, 1024);
>> +    g_assert_cmpuint(bdrv_write_threshold_get(&bs), ==, threshold);
>>   }
>>   static void test_threshold_trigger(void)
>>   {
>> -    uint64_t amount = 0;
>>       uint64_t threshold = 4 * 1024 * 1024;
>>       BlockDriverState bs;
>> -    BdrvTrackedRequest req;
>>       memset(&bs, 0, sizeof(bs));
>> -    memset(&req, 0, sizeof(req));
>> -    req.offset = (4 * 1024 * 1024) - 1024;
>> -    req.bytes = 2 * 1024;
>> -
>> -    bdrv_check_request(req.offset, req.bytes, &error_abort);
>>       bdrv_write_threshold_set(&bs, threshold);
>> -    amount = bdrv_write_threshold_exceeded(&bs, &req);
>> -    g_assert_cmpuint(amount, >=, 1024);
>> +    bdrv_write_threshold_check_write(&bs, threshold - 1024, 2 * 1024);
>> +    g_assert_cmpuint(bdrv_write_threshold_get(&bs), ==, 0);
>>   }
>>   typedef struct TestStruct {
>>
> 


-- 
Best regards,
Vladimir


  reply	other threads:[~2021-05-05 15:24 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-04  8:25 [PATCH v2 0/9] block: refactor write threshold Vladimir Sementsov-Ogievskiy
2021-05-04  8:25 ` [PATCH v2 1/9] block/write-threshold: don't use write notifiers Vladimir Sementsov-Ogievskiy
2021-05-05 12:37   ` Max Reitz
2021-05-05 13:27     ` Vladimir Sementsov-Ogievskiy
2021-05-05 13:35     ` Vladimir Sementsov-Ogievskiy
2021-05-05 14:29       ` Max Reitz
2021-05-04  8:25 ` [PATCH v2 2/9] block: drop " Vladimir Sementsov-Ogievskiy
2021-05-05 12:40   ` Max Reitz
2021-05-04  8:25 ` [PATCH v2 3/9] test-write-threshold: rewrite test_threshold_(not_)trigger tests Vladimir Sementsov-Ogievskiy
2021-05-05 14:28   ` Max Reitz
2021-05-05 15:20     ` Vladimir Sementsov-Ogievskiy [this message]
2021-05-04  8:25 ` [PATCH v2 4/9] block/write-threshold: drop extra APIs Vladimir Sementsov-Ogievskiy
2021-05-05 14:41   ` Max Reitz
2021-05-04  8:25 ` [PATCH v2 5/9] block/write-threshold: don't use aio context lock Vladimir Sementsov-Ogievskiy
2021-05-05 16:09   ` Max Reitz
2021-05-04  8:25 ` [PATCH v2 6/9] test-write-threshold: drop extra tests Vladimir Sementsov-Ogievskiy
2021-05-05 16:11   ` Max Reitz
2021-05-04  8:25 ` [PATCH v2 7/9] test-write-threshold: drop extra TestStruct structure Vladimir Sementsov-Ogievskiy
2021-05-05 16:13   ` Max Reitz
2021-05-04  8:25 ` [PATCH v2 8/9] test-write-threshold: drop extra includes Vladimir Sementsov-Ogievskiy
2021-05-05 16:14   ` Max Reitz
2021-05-04  8:25 ` [PATCH v2 9/9] block/write-threshold: " Vladimir Sementsov-Ogievskiy
2021-05-05 16:23   ` Max Reitz
2021-05-05 20:34     ` Vladimir Sementsov-Ogievskiy
2021-05-06  7:41       ` Max Reitz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=880d2a4a-68ae-fdb8-05d6-357ff74b32a1@virtuozzo.com \
    --to=vsementsov@virtuozzo.com \
    --cc=eesposit@redhat.com \
    --cc=fam@euphon.net \
    --cc=kwolf@redhat.com \
    --cc=mreitz@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-block@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).