On Fri 11-03-22 20:59:05, Charan Teja Kalla wrote: > The process_madvise() system call returns error even after processing > some VMA's passed in the 'struct iovec' vector list which leaves the > user confused to know where to restart the advise next. It is also > against this syscall man page[1] documentation where it mentions that > "return value may be less than the total number of requested bytes, if > an error occurred after some iovec elements were already processed.". > > Consider a user passed 10 VMA's in the 'struct iovec' vector list of > which 9 are processed but one. Then it just returns the error caused on > that failed VMA despite the first 9 VMA's processed, leaving the user > confused about on which VMA it is failed. Returning the number of bytes > processed here can help the user to know which VMA it is failed on and > thus can retry/skip the advise on that VMA. > > [1]https://man7.org/linux/man-pages/man2/process_madvise.2.html. > > Fixes: ecb8ac8b1f14("mm/madvise: introduce process_madvise() syscall: an external memory hinting API") > Cc: <stable@xxxxxxxxxxxxxxx> # 5.10+ > Signed-off-by: Charan Teja Kalla <quic_charante@xxxxxxxxxxx> Acked-by: Michal Hocko <mhocko@xxxxxxxx> > --- > Changes in V2: > -- Separated the ENOMEM handling and return bytes processed, as per Minchan comments. > -- This contains correcting return bytes processed with process_madvise(). > > Changes in V1: > -- Fixed the ENOMEM handling and return bytes processed by process_madvise. > -- https://patchwork.kernel.org/project/linux-mm/patch/1646803679-11433-1-git-send-email-quic_charante@xxxxxxxxxxx/ > > mm/madvise.c | 3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/mm/madvise.c b/mm/madvise.c > index 38d0f51..e97e6a9 100644 > --- a/mm/madvise.c > +++ b/mm/madvise.c > @@ -1433,8 +1433,7 @@ SYSCALL_DEFINE5(process_madvise, int, pidfd, const struct iovec __user *, vec, > iov_iter_advance(&iter, iovec.iov_len); > } > > - if (ret == 0) > - ret = total_len - iov_iter_count(&iter); > + ret = (total_len - iov_iter_count(&iter)) ? : ret; > > release_mm: > mmput(mm); > -- > 2.7.4 -- Michal Hocko SUSE Labs
