Re: [Qemu-devel] [PATCH] io: Fix double shift usages on QIOChannel features

[Felipe Franciosi]

Hi Daniel,

> On 27 Sep 2016, at 18:23, Daniel P. Berrange <address@hidden> wrote:
> 
> On Tue, Sep 27, 2016 at 09:49:18AM -0700, Felipe Franciosi wrote:
>> When QIOChannels were introduced in 666a3af9, the feature bits were
>> defined shifted. However, when using them, the code was shifting them
>> again. The incorrect use was consistent until 74b6ce43, where
>> QIO_CHANNEL_FEATURE_LISTEN was defined shifted but tested unshifted.
> 
> I'm more inclined to actually change the header file, so that they
> are defined unshifted, and fix the single place that tests
> unshifted.  They are defined in an enum, so it was kind of odd to
> use shifted values in the first place.

It's not uncommon to specify shifted features/flags on enums:
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/include/uapi/linux/nl80211.h#n2661
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/include/linux/nvme.h#n322
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/include/linux/blk-mq.h#n194

I actually prefer defining them shifted, as proposed in my patch. And perhaps 
adding a qio_channel_set_feature() to completely abstract their usage. But I 
don't have strong preferences towards this and can change it if you really want 
me to.

> 
>> This patch fixes all uses of QIOChannel features. They are defined
>> shifted and therefore set unshifted. It also makes all feature tests to
>> use the qio_channel_has_feature() function.
> 
> Switching to use of qio_channel_has_feature() is a useful, but
> independant fix, so should be a separate commit really.

Sure I can separate that in another patch. Should I also add a 
qio_channel_set_feature()? I think that's a good idea.

> 
>> 
>> Signed-off-by: Felipe Franciosi <address@hidden>
>> ---
>> io/channel-socket.c           |   11 ++++++-----
>> io/channel-tls.c              |    4 ++--
>> io/channel-websock.c          |    4 ++--
>> io/channel.c                  |   10 ++++------
>> migration/qemu-file-channel.c |    3 +--
>> qemu-char.c                   |    3 +--
>> 6 files changed, 16 insertions(+), 19 deletions(-)
> 
> The test/test-io-channel-*.c unit tests should be updated to
> validate the correct handling.

Oh, good point. Totally missed that.

> 
>> 
>> diff --git a/io/channel-socket.c b/io/channel-socket.c
>> index 196a4f1..bb0a0a7 100644
>> --- a/io/channel-socket.c
>> +++ b/io/channel-socket.c
>> @@ -55,7 +55,7 @@ qio_channel_socket_new(void)
>>     sioc->fd = -1;
>> 
>>     ioc = QIO_CHANNEL(sioc);
>> -    ioc->features |= (1 << QIO_CHANNEL_FEATURE_SHUTDOWN);
>> +    ioc->features |= QIO_CHANNEL_FEATURE_SHUTDOWN;
>> 
>> #ifdef WIN32
>>     ioc->event = CreateEvent(NULL, FALSE, FALSE, NULL);
>> @@ -107,12 +107,12 @@ qio_channel_socket_set_fd(QIOChannelSocket *sioc,
>> #ifndef WIN32
>>     if (sioc->localAddr.ss_family == AF_UNIX) {
>>         QIOChannel *ioc = QIO_CHANNEL(sioc);
>> -        ioc->features |= (1 << QIO_CHANNEL_FEATURE_FD_PASS);
>> +        ioc->features |= QIO_CHANNEL_FEATURE_FD_PASS;
>>     }
>> #endif /* WIN32 */
>>     if (getsockopt(fd, SOL_SOCKET, SO_ACCEPTCONN, &val, &len) == 0 && val) {
>>         QIOChannel *ioc = QIO_CHANNEL(sioc);
>> -        ioc->features |= (1 << QIO_CHANNEL_FEATURE_LISTEN);
>> +        ioc->features |= QIO_CHANNEL_FEATURE_LISTEN;
>>     }
>> 
>>     return 0;
>> @@ -380,7 +380,7 @@ qio_channel_socket_accept(QIOChannelSocket *ioc,
>> 
>> #ifndef WIN32
>>     if (cioc->localAddr.ss_family == AF_UNIX) {
>> -        QIO_CHANNEL(cioc)->features |= (1 << QIO_CHANNEL_FEATURE_FD_PASS);
>> +        QIO_CHANNEL(cioc)->features |= QIO_CHANNEL_FEATURE_FD_PASS;
>>     }
>> #endif /* WIN32 */
>> 
>> @@ -403,7 +403,8 @@ static void qio_channel_socket_finalize(Object *obj)
>>     QIOChannelSocket *ioc = QIO_CHANNEL_SOCKET(obj);
>> 
>>     if (ioc->fd != -1) {
>> -        if (QIO_CHANNEL(ioc)->features & QIO_CHANNEL_FEATURE_LISTEN) {
>> +        if (qio_channel_has_feature(QIO_CHANNEL(ioc),
>> +                                    QIO_CHANNEL_FEATURE_LISTEN)) {
>>             Error *err = NULL;
>> 
>>             socket_listen_cleanup(ioc->fd, &err);
>> diff --git a/io/channel-tls.c b/io/channel-tls.c
>> index 9a8525c..d22996b 100644
>> --- a/io/channel-tls.c
>> +++ b/io/channel-tls.c
>> @@ -111,8 +111,8 @@ qio_channel_tls_new_client(QIOChannel *master,
>>     ioc = QIO_CHANNEL(tioc);
>> 
>>     tioc->master = master;
>> -    if (master->features & (1 << QIO_CHANNEL_FEATURE_SHUTDOWN)) {
>> -        ioc->features |= (1 << QIO_CHANNEL_FEATURE_SHUTDOWN);
>> +    if (qio_channel_has_feature(master, QIO_CHANNEL_FEATURE_SHUTDOWN)) {
>> +        ioc->features |= QIO_CHANNEL_FEATURE_SHUTDOWN;
>>     }
>>     object_ref(OBJECT(master));
>> 
>> diff --git a/io/channel-websock.c b/io/channel-websock.c
>> index 533bd4b..9accd16 100644
>> --- a/io/channel-websock.c
>> +++ b/io/channel-websock.c
>> @@ -497,8 +497,8 @@ qio_channel_websock_new_server(QIOChannel *master)
>>     ioc = QIO_CHANNEL(wioc);
>> 
>>     wioc->master = master;
>> -    if (master->features & (1 << QIO_CHANNEL_FEATURE_SHUTDOWN)) {
>> -        ioc->features |= (1 << QIO_CHANNEL_FEATURE_SHUTDOWN);
>> +    if (qio_channel_has_feature(master, QIO_CHANNEL_FEATURE_SHUTDOWN)) {
>> +        ioc->features |= QIO_CHANNEL_FEATURE_SHUTDOWN;
>>     }
>>     object_ref(OBJECT(master));
>> 
>> diff --git a/io/channel.c b/io/channel.c
>> index 923c465..ebe9f86 100644
>> --- a/io/channel.c
>> +++ b/io/channel.c
>> @@ -23,13 +23,11 @@
>> #include "qapi/error.h"
>> #include "qemu/coroutine.h"
>> 
>> -bool qio_channel_has_feature(QIOChannel *ioc,
>> -                             QIOChannelFeature feature)
>> +bool qio_channel_has_feature(QIOChannel *ioc, QIOChannelFeature feature)
> 
> Don't mix in whitespace changes like this.

I think the line break was a leftover from further back. There's clearly no 
need for it here, but I can separate it in another commit, too.

> 
>> {
>> -    return ioc->features & (1 << feature);
>> +    return ioc->features & feature;
>> }
> 
> This is logically wrong - 'feature' can now contain multiple
> bits, but this is returning true if any single one of them
> is present, rather than if all are present. IMHO this is an
> example of why we should define them unshifted.

This looks correct to me. It's only wrong if we change the definition to be 
unshifted, which I believe is still on the table. :)

> 
>> 
>> -
>> ssize_t qio_channel_readv_full(QIOChannel *ioc,
>>                                const struct iovec *iov,
>>                                size_t niov,
>> @@ -40,7 +38,7 @@ ssize_t qio_channel_readv_full(QIOChannel *ioc,
>>     QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
>> 
>>     if ((fds || nfds) &&
>> -        !(ioc->features & (1 << QIO_CHANNEL_FEATURE_FD_PASS))) {
>> +        !qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_FD_PASS)) {
>>         error_setg_errno(errp, EINVAL,
>>                          "Channel does not support file descriptor passing");
>>         return -1;
>> @@ -60,7 +58,7 @@ ssize_t qio_channel_writev_full(QIOChannel *ioc,
>>     QIOChannelClass *klass = QIO_CHANNEL_GET_CLASS(ioc);
>> 
>>     if ((fds || nfds) &&
>> -        !(ioc->features & (1 << QIO_CHANNEL_FEATURE_FD_PASS))) {
>> +        !qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_FD_PASS)) {
>>         error_setg_errno(errp, EINVAL,
>>                          "Channel does not support file descriptor passing");
>>         return -1;
>> diff --git a/migration/qemu-file-channel.c b/migration/qemu-file-channel.c
>> index 45c13f1..a39abd5 100644
>> --- a/migration/qemu-file-channel.c
>> +++ b/migration/qemu-file-channel.c
>> @@ -105,8 +105,7 @@ static int channel_shutdown(void *opaque,
>> {
>>     QIOChannel *ioc = QIO_CHANNEL(opaque);
>> 
>> -    if (qio_channel_has_feature(ioc,
>> -                                QIO_CHANNEL_FEATURE_SHUTDOWN)) {
>> +    if (qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_SHUTDOWN)) {
>>         QIOChannelShutdown mode;
>>         if (rd && wr) {
>>             mode = QIO_CHANNEL_SHUTDOWN_BOTH;
> 
> That's an unrelated arbitrary whitespace change. Please don't mix that
> with other functional changes.

As above, I'll put those on a separate patch.

> 
>> diff --git a/qemu-char.c b/qemu-char.c
>> index 8826419..b825331 100644
>> --- a/qemu-char.c
>> +++ b/qemu-char.c
>> @@ -2787,8 +2787,7 @@ static int tcp_set_msgfds(CharDriverState *chr, int 
>> *fds, int num)
>>     s->write_msgfds_num = 0;
>> 
>>     if (!s->connected ||
>> -        !qio_channel_has_feature(s->ioc,
>> -                                 QIO_CHANNEL_FEATURE_FD_PASS)) {
>> +        !qio_channel_has_feature(s->ioc, QIO_CHANNEL_FEATURE_FD_PASS)) {
>>         return -1;
>>     }
> 
> Again, don't do whitespace changes like this.

As above, I'll put those on a separate patch.

Let me know what you think about the (un)shifted definition before I send a v2.

Thanks,
Felipe

> 
> Regards,
> Daniel
> -- 
> |: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org              -o-             http://virt-manager.org :|
> |: http://entangle-photo.org       -o-    http://search.cpan.org/~danberr/ :|