From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from userp1040.oracle.com ([156.151.31.81]:34118 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750775AbdI1Tba (ORCPT ); Thu, 28 Sep 2017 15:31:30 -0400 Date: Thu, 28 Sep 2017 12:31:20 -0700 From: "Darrick J. Wong" Subject: Re: [PATCH 3/3] Allow partial writes Message-ID: <20170928193120.GR5020@magnolia> References: <20170928185426.9070-1-rgoldwyn@suse.de> <20170928185426.9070-3-rgoldwyn@suse.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170928185426.9070-3-rgoldwyn@suse.de> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Goldwyn Rodrigues Cc: linux-xfs@vger.kernel.org, david@fromorbit.com, Goldwyn Rodrigues On Thu, Sep 28, 2017 at 01:54:26PM -0500, Goldwyn Rodrigues wrote: > From: Goldwyn Rodrigues > > Partial writes are performed when there is an error midway > while performing the I/O. Perform the write exactly once and > return the number of bytes written. > > Signed-off-by: Goldwyn Rodrigues > --- > io/io.h | 1 + > io/pwrite.c | 25 ++++++++++++++++++++++++- > man/man8/xfs_io.8 | 3 +++ > 3 files changed, 28 insertions(+), 1 deletion(-) > > diff --git a/io/io.h b/io/io.h > index 2bc7ac4a..5875e85a 100644 > --- a/io/io.h > +++ b/io/io.h > @@ -24,6 +24,7 @@ > #define IO_RANDOM ( 0) > #define IO_FORWARD ( 1) > #define IO_BACKWARD (-1) > +#define IO_PARTIAL (2) I would name this CALL_ONCE instead of PARTIAL since we don't know beforehand that the write will return partial results -- it could succeed at writing the entire buffer. --D > > /* > * File descriptor options > diff --git a/io/pwrite.c b/io/pwrite.c > index 8289c3ee..40f537b3 100644 > --- a/io/pwrite.c > +++ b/io/pwrite.c > @@ -47,6 +47,7 @@ pwrite_help(void) > " -W -- call fsync(2) at the end (included in timing results)\n" > " -B -- write backwards through the range from offset (backwards N bytes)\n" > " -F -- write forwards through the range of bytes from offset (default)\n" > +" -P -- return Partial writes, perform one call and return bytes written\n" > " -R -- write at random offsets in the specified range of bytes\n" > " -Z N -- zeed the random number generator (used when writing randomly)\n" > " (heh, zorry, the -s/-S arguments were already in use in pwrite)\n" > @@ -255,6 +256,22 @@ write_buffer( > return ops; > } > > +static int > +write_partial( > + off64_t offset, > + long long count, > + long long *total, > + int pwritev2_flags) > +{ > + size_t bytes; > + bytes = do_pwrite(file->fd, offset, count, count, pwritev2_flags); > + if (bytes < 0) > + return -1; > + *total = bytes; > + return 1; > +} > + > + > static int > pwrite_f( > int argc, > @@ -276,7 +293,7 @@ pwrite_f( > init_cvtnum(&fsblocksize, &fssectsize); > bsize = fsblocksize; > > - while ((c = getopt(argc, argv, "b:BCdf:Fi:NqRs:S:uV:wWZ:")) != EOF) { > + while ((c = getopt(argc, argv, "b:BCdf:Fi:NqRs:PS:uV:wWZ:")) != EOF) { > switch (c) { > case 'b': > tmp = cvtnum(fsblocksize, fssectsize, optarg); > @@ -298,6 +315,9 @@ pwrite_f( > case 'R': > direction = IO_RANDOM; > break; > + case 'P': > + direction = IO_PARTIAL; > + break; > case 'd': > dflag = 1; > break; > @@ -389,6 +409,9 @@ pwrite_f( > case IO_BACKWARD: > c = write_backward(offset, &count, &total, pwritev2_flags); > break; > + case IO_PARTIAL: > + c = write_partial(offset, count, &total, pwritev2_flags); > + break; > default: > total = 0; > ASSERT(0); > diff --git a/man/man8/xfs_io.8 b/man/man8/xfs_io.8 > index f410ce9a..b21a8145 100644 > --- a/man/man8/xfs_io.8 > +++ b/man/man8/xfs_io.8 > @@ -229,6 +229,9 @@ write the buffers in a reserve sequential direction. > .B \-R > write the buffers in the give range in a random order. > .TP > +.B \-P > +return Partial writes, perform one call and return the bytes written. > +.TP > .B \-Z seed > specify the random number seed used for random write > .TP > -- > 2.14.1 > > -- > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html