Message ID | 1321023744-16249-1-git-send-email-linkinjeon@gmail.com |
---|---|
State | Rejected, archived |
Headers | show |
Any feedback for this patch? We surely don't need 'ret2' variable in this code. Thanks & Regards, Amit Sahrawat On Fri, Nov 11, 2011 at 8:32 PM, Namjae Jeon <linkinjeon@gmail.com> wrote: > ret2 is not needed in ext4_flush_completed_IO(). > > Signed-off-by: Namjae Jeon <linkinjeon@gmail.com> > Signed-off-by: Amit Sahrawat <amit.sahrawat83@gmail.com> > --- > fs/ext4/fsync.c | 5 +---- > 1 files changed, 1 insertions(+), 4 deletions(-) > > diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c > index 00a2cb7..40397ac 100644 > --- a/fs/ext4/fsync.c > +++ b/fs/ext4/fsync.c > @@ -81,7 +81,6 @@ int ext4_flush_completed_IO(struct inode *inode) > struct ext4_inode_info *ei = EXT4_I(inode); > unsigned long flags; > int ret = 0; > - int ret2 = 0; > > dump_completed_IO(inode); > spin_lock_irqsave(&ei->i_completed_io_lock, flags); > @@ -105,12 +104,10 @@ int ext4_flush_completed_IO(struct inode *inode) > */ > spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); > ret = ext4_end_io_nolock(io); > - if (ret < 0) > - ret2 = ret; > spin_lock_irqsave(&ei->i_completed_io_lock, flags); > } > spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); > - return (ret2 < 0) ? ret2 : 0; > + return (ret < 0) ? ret : 0; > } > > /* > -- > 1.7.4.4 > > -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 11/11/2011 08:32 PM, Namjae Jeon wrote: > ret2 is not needed in ext4_flush_completed_IO(). > Not needed? I went through the code briefly, and I don't agree. > Signed-off-by: Namjae Jeon <linkinjeon@gmail.com> > Signed-off-by: Amit Sahrawat <amit.sahrawat83@gmail.com> > --- > fs/ext4/fsync.c | 5 +---- > 1 files changed, 1 insertions(+), 4 deletions(-) > > diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c > index 00a2cb7..40397ac 100644 > --- a/fs/ext4/fsync.c > +++ b/fs/ext4/fsync.c > @@ -81,7 +81,6 @@ int ext4_flush_completed_IO(struct inode *inode) > struct ext4_inode_info *ei = EXT4_I(inode); > unsigned long flags; > int ret = 0; > - int ret2 = 0; > > dump_completed_IO(inode); > spin_lock_irqsave(&ei->i_completed_io_lock, flags); > @@ -105,12 +104,10 @@ int ext4_flush_completed_IO(struct inode *inode) > */ > spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); > ret = ext4_end_io_nolock(io); > - if (ret < 0) > - ret2 = ret; > spin_lock_irqsave(&ei->i_completed_io_lock, flags); > } > spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); > - return (ret2 < 0) ? ret2 : 0; > + return (ret < 0) ? ret : 0; > } Please note that there is a while loop involved here. Which means, that ret2 is used to store the last negative value of ret. And due to the loop, ret can be over-written in the next loop iteration, which we can afford, because we have already stored what we need to save, in ret2. And this ret2 value is used to return appropriate value to the caller. Thanks, Srivatsa S. Bhat -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 11/20/2011 01:41 AM, Srivatsa S. Bhat wrote: > On 11/11/2011 08:32 PM, Namjae Jeon wrote: >> ret2 is not needed in ext4_flush_completed_IO(). >> > > Not needed? I went through the code briefly, and I don't agree. > >> Signed-off-by: Namjae Jeon <linkinjeon@gmail.com> >> Signed-off-by: Amit Sahrawat <amit.sahrawat83@gmail.com> >> --- >> fs/ext4/fsync.c | 5 +---- >> 1 files changed, 1 insertions(+), 4 deletions(-) >> >> diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c >> index 00a2cb7..40397ac 100644 >> --- a/fs/ext4/fsync.c >> +++ b/fs/ext4/fsync.c >> @@ -81,7 +81,6 @@ int ext4_flush_completed_IO(struct inode *inode) >> struct ext4_inode_info *ei = EXT4_I(inode); >> unsigned long flags; >> int ret = 0; >> - int ret2 = 0; >> >> dump_completed_IO(inode); >> spin_lock_irqsave(&ei->i_completed_io_lock, flags); >> @@ -105,12 +104,10 @@ int ext4_flush_completed_IO(struct inode *inode) >> */ >> spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); >> ret = ext4_end_io_nolock(io); >> - if (ret < 0) >> - ret2 = ret; >> spin_lock_irqsave(&ei->i_completed_io_lock, flags); >> } >> spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); >> - return (ret2 < 0) ? ret2 : 0; >> + return (ret < 0) ? ret : 0; >> } > > Please note that there is a while loop involved here. Which means, that ret2 > is used to store the last negative value of ret. And due to the loop, ret can > be over-written in the next loop iteration, which we can afford, because we > have already stored what we need to save, in ret2. And this ret2 value is used > to return appropriate value to the caller. > Actually, what I really meant was, removing ret2 as merely "unneeded" might not be the right thing to do because once you apply your patch, you end up altering the value returned by this function! If the return value is indeed wrong in the current code, you should rather be saying that this is a bug fix, with appropriate justification IMO. Thanks, Srivatsa S. Bhat -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
2011/11/20 Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>: > On 11/20/2011 01:41 AM, Srivatsa S. Bhat wrote: >> On 11/11/2011 08:32 PM, Namjae Jeon wrote: >>> ret2 is not needed in ext4_flush_completed_IO(). >>> >> >> Not needed? I went through the code briefly, and I don't agree. >> >>> Signed-off-by: Namjae Jeon <linkinjeon@gmail.com> >>> Signed-off-by: Amit Sahrawat <amit.sahrawat83@gmail.com> >>> --- >>> fs/ext4/fsync.c | 5 +---- >>> 1 files changed, 1 insertions(+), 4 deletions(-) >>> >>> diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c >>> index 00a2cb7..40397ac 100644 >>> --- a/fs/ext4/fsync.c >>> +++ b/fs/ext4/fsync.c >>> @@ -81,7 +81,6 @@ int ext4_flush_completed_IO(struct inode *inode) >>> struct ext4_inode_info *ei = EXT4_I(inode); >>> unsigned long flags; >>> int ret = 0; >>> - int ret2 = 0; >>> >>> dump_completed_IO(inode); >>> spin_lock_irqsave(&ei->i_completed_io_lock, flags); >>> @@ -105,12 +104,10 @@ int ext4_flush_completed_IO(struct inode *inode) >>> */ >>> spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); >>> ret = ext4_end_io_nolock(io); >>> - if (ret < 0) >>> - ret2 = ret; >>> spin_lock_irqsave(&ei->i_completed_io_lock, flags); >>> } >>> spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); >>> - return (ret2 < 0) ? ret2 : 0; >>> + return (ret < 0) ? ret : 0; >>> } >> >> Please note that there is a while loop involved here. Which means, that ret2 >> is used to store the last negative value of ret. And due to the loop, ret can >> be over-written in the next loop iteration, which we can afford, because we >> have already stored what we need to save, in ret2. And this ret2 value is used >> to return appropriate value to the caller. >> > > Actually, what I really meant was, removing ret2 as merely "unneeded" might not > be the right thing to do because once you apply your patch, you end up altering > the value returned by this function! > > If the return value is indeed wrong in the current code, you should rather > be saying that this is a bug fix, with appropriate justification IMO. > Thanks for your review. My thinking was reaching for to while loop(). I will remember your advice. Thanks Srivatsa. > Thanks, > Srivatsa S. Bhat > > -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 00a2cb7..40397ac 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -81,7 +81,6 @@ int ext4_flush_completed_IO(struct inode *inode) struct ext4_inode_info *ei = EXT4_I(inode); unsigned long flags; int ret = 0; - int ret2 = 0; dump_completed_IO(inode); spin_lock_irqsave(&ei->i_completed_io_lock, flags); @@ -105,12 +104,10 @@ int ext4_flush_completed_IO(struct inode *inode) */ spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); ret = ext4_end_io_nolock(io); - if (ret < 0) - ret2 = ret; spin_lock_irqsave(&ei->i_completed_io_lock, flags); } spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); - return (ret2 < 0) ? ret2 : 0; + return (ret < 0) ? ret : 0; } /*