qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: John Snow <jsnow@redhat.com>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: kwolf@redhat.com, qemu-devel@nongnu.org, stefanha@redhat.com,
	armbru@redhat.com
Subject: Re: [Qemu-devel] [PATCH 7/8] qtest/ahci: add qcow2 support to ahci-test
Date: Wed, 25 Feb 2015 17:40:00 -0500	[thread overview]
Message-ID: <54EE4F40.90108@redhat.com> (raw)
In-Reply-To: <20150225152757.GD2922@stefanha-thinkpad.redhat.com>



On 02/25/2015 10:27 AM, Stefan Hajnoczi wrote:
> On Thu, Feb 19, 2015 at 05:30:02PM -0500, John Snow wrote:
>> diff --git a/tests/libqos/libqos.c b/tests/libqos/libqos.c
>> index bc8beb2..3577401 100644
>> --- a/tests/libqos/libqos.c
>> +++ b/tests/libqos/libqos.c
>> @@ -61,3 +61,33 @@ void qtest_shutdown(QOSState *qs)
>>       qtest_quit(qs->qts);
>>       g_free(qs);
>>   }
>> +
>> +int mkqcow2(const char *file, unsigned size_mb)
>> +{
>> +    pid_t pid;
>> +    int rc;
>> +    char buff[32];
>> +
>> +    snprintf(buff, 32, "%uM", size_mb);
>> +
>> +    pid = fork();
>> +    switch (pid) {
>> +    case -1:
>> +        perror("fork failed");
>> +        return -1;
>> +    case 0:
>> +        close(fileno(stdout));
>> +        rc = open("/dev/null", O_WRONLY);
>> +        g_assert_cmpint(rc, ==, fileno(stdout));
>> +        execl("./qemu-img", "qemu-img", "create", "-f", "qcow2",
>> +              file, buff, NULL);
>
> Where is the current working directory set?
>

Nowhere in-program. We assume you are running this via `make check` 
which further assumes that qemu-img is in your build tree. We actually 
never set an explicit path for where the QEMU binary is either: it's set 
to e.g.:

QTEST_QEMU_BINARY=x86_64-softmmu/qemu-system-x86_64
and then executed via execlp("exec QTEST_QEMU_BINARY", ...)

This patch pulls a similar shortcut.

The Makefile blob doesn't actually appear to really ever explicitly set 
a path for where the binaries should be; qemu-img is not defined to be 
in any sort of path, so we can assume that all paths are relative to the 
Makefile.

What I will do is set QTEST_QEMU_IMG to at least match the direct name 
of the QEMU_IMG we generated, so it will have $(EXESUF) attached to it, 
for instance.

> Will this work for both in-tree and out-of-tree builds?
>

Yes. I usually run out of tree, but it works fine in-tree too. As long 
as relative to the Makefile ./qemu-img$(EXESUF) exists, you will be fine.

If you decide to run this outside of make check, you can always set 
QTEST_QEMU_IMG yourself.

> BTW, glib abstracts process spawning, maybe all this can be done in a
> single function call:
> https://developer.gnome.org/glib/stable/glib-Spawning-Processes.html#g-spawn-sync
>

OK, I've implemented it with this call instead. Will send a V2.

>> +        perror("execl failed");
>> +        exit(-1);
>
> EXIT_FAILURE is provided by the C standard for returning failure.  (-1
> is confusing because the shell exit code value cannot be negative.)
>
>> +    default:
>> +        wait(&rc);
>> +        g_assert_cmpint(rc, ==, 0);
>> +        return rc;
>> +    }
>> +
>> +    g_assert_not_reached();
>> +}
>> diff --git a/tests/libqos/libqos.h b/tests/libqos/libqos.h
>> index 612d41e..5abd2bd 100644
>> --- a/tests/libqos/libqos.h
>> +++ b/tests/libqos/libqos.h
>> @@ -19,6 +19,7 @@ typedef struct QOSState {
>>   QOSState *qtest_vboot(QOSOps *ops, const char *cmdline_fmt, va_list ap);
>>   QOSState *qtest_boot(QOSOps *ops, const char *cmdline_fmt, ...);
>>   void qtest_shutdown(QOSState *qs);
>> +int mkqcow2(const char *file, unsigned size_mb);
>>
>>   static inline uint64_t qmalloc(QOSState *q, size_t bytes)
>>   {
>> --
>> 1.9.3
>>
>>

  reply	other threads:[~2015-02-25 22:40 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-02-19 22:29 [Qemu-devel] [PATCH 0/8] ahci: add more IO tests John Snow
2015-02-19 22:29 ` [Qemu-devel] [PATCH 1/8] libqos/ahci: Zero-fill AHCI headers John Snow
2015-02-19 22:29 ` [Qemu-devel] [PATCH 2/8] qtest/ahci: Add a macro bootup routine John Snow
2015-02-19 22:29 ` [Qemu-devel] [PATCH 3/8] libqos/ahci: add ahci command helpers John Snow
2015-02-19 22:29 ` [Qemu-devel] [PATCH 4/8] qtest/ahci: Add DMA test variants John Snow
2015-02-19 22:30 ` [Qemu-devel] [PATCH 5/8] qtest/ahci: Add PIO and LBA48 tests John Snow
2015-02-19 22:30 ` [Qemu-devel] [PATCH 6/8] qtest/ahci: add fragmented dma test John Snow
2015-02-19 22:30 ` [Qemu-devel] [PATCH 7/8] qtest/ahci: add qcow2 support to ahci-test John Snow
2015-02-25 15:27   ` Stefan Hajnoczi
2015-02-25 22:40     ` John Snow [this message]
2015-02-26 11:01       ` Stefan Hajnoczi
2015-02-19 22:30 ` [Qemu-devel] [PATCH 8/8] qtest/ahci: test different disk sectors John Snow
2015-02-24 17:58 ` [Qemu-devel] [PATCH 0/8] ahci: add more IO tests John Snow
2015-02-25 15:29 ` Stefan Hajnoczi

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=54EE4F40.90108@redhat.com \
    --to=jsnow@redhat.com \
    --cc=armbru@redhat.com \
    --cc=kwolf@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@gmail.com \
    --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).