Bei Guan
2012-May-11 14:08 UTC
Errors of doing "make install-tools" with xen-4.2-unstable?
Hi, When I do the "make install-tools" with xen-4.2-unstable, there are some errors about "warn_unused_result". Is it the error in code or the error in the compiling environment? Thank you so much. gcc -O1 -fno-omit-frame-pointer -m64 -g -fno-strict-aliasing -std=gnu99 -Wall -Wstrict-prototypes -Wdeclaration-after-statement -D__XEN_TOOLS__ -MMD -MF .tapdisk-queue.o.d -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -fno-optimize-sibling-calls -Werror -g -Wno-unused -fno-strict-aliasing -I../include -I../drivers -I/root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libxc -I/root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/include -D_GNU_SOURCE -DUSE_NFS_LOCKS -fPIC -I /root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c -o tapdisk-queue.o tapdisk-queue.c cc1: warnings being treated as errors tapdisk-queue.c: In function ‘tapdisk_lio_ack_event’: tapdisk-queue.c:438: error: ignoring return value of ‘read’, declared with attribute warn_unused_result make[5]: *** [tapdisk-queue.o] Error 1 make[5]: Leaving directory `/root/Xen/xen-4.2-unstable/tools/blktap2/drivers'' make[4]: *** [subdir-install-drivers] Error 2 make[4]: Leaving directory `/root/Xen/xen-4.2-unstable/tools/blktap2'' make[3]: *** [subdirs-install] Error 2 make[3]: Leaving directory `/root/Xen/xen-4.2-unstable/tools/blktap2'' make[2]: *** [subdir-install-blktap2] Error 2 make[2]: Leaving directory `/root/Xen/xen-4.2-unstable/tools'' make[1]: *** [subdirs-install] Error 2 make[1]: Leaving directory `/root/Xen/xen-4.2-unstable/tools'' make: *** [install-tools] Error 2 ------ root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c -o tapdisk-log.o tapdisk-log.c cc1: warnings being treated as errors tapdisk-log.c: In function ‘tlog_flush’: tapdisk-log.c:250: error: ignoring return value of ‘write’, declared with attribute warn_unused_result ------ /root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c -o tapdisk2.o tapdisk2.c cc1: warnings being treated as errors tapdisk2.c: In function ‘main’: tapdisk2.c:82: error: ignoring return value of ‘chdir’, declared with attribute warn_unused_result ------ cc1: warnings being treated as errors tapdisk-stream.c: In function ‘tapdisk_stream_poll_clear’: tapdisk-stream.c:148: error: ignoring return value of ‘read’, declared with attribute warn_unused_result tapdisk-stream.c: In function ‘tapdisk_stream_poll_set’: tapdisk-stream.c:158: error: ignoring return value of ‘write’, declared with attribute warn_unused_result tapdisk-stream.c: In function ‘tapdisk_stream_print_request’: tapdisk-stream.c:206: error: ignoring return value of ‘write’, declared with attribute warn_unused_result Best Regards, Bei Guan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Ian Campbell
2012-May-11 14:16 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On Fri, 2012-05-11 at 15:08 +0100, Bei Guan wrote:> When I do the "make install-tools" with xen-4.2-unstable, there are > some errors about "warn_unused_result". > Is it the error in code or the error in the compiling environment? > Thank you so much.This commit changeset: 25275:27d63b9f111a user: Keir Fraser <keir@xen.org> date: Thu May 10 11:22:18 2012 +0100 summary: blktap2: Do not build with -O0 has caused more warnings to be generated by the compiler, which in turn leads to build failures. Since this is a compiler specific thing we are still tracking them all down. Can you confirm which versionof xen-unstable.hg you are using -- there have been several fixup patches since the above. If you are completely up to date and you still see errors we can dig into whatever is left. Ian. (starting to think that this change wasn't suitable during a feature freeze....)> > > gcc -O1 -fno-omit-frame-pointer -m64 -g -fno-strict-aliasing > -std=gnu99 -Wall -Wstrict-prototypes -Wdeclaration-after-statement > -D__XEN_TOOLS__ -MMD -MF .tapdisk-queue.o.d -D_LARGEFILE_SOURCE > -D_LARGEFILE64_SOURCE -fno-optimize-sibling-calls -Werror -g > -Wno-unused -fno-strict-aliasing -I../include -I../drivers > -I/root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libxc -I/root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/include -D_GNU_SOURCE -DUSE_NFS_LOCKS -fPIC -I /root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c -o tapdisk-queue.o tapdisk-queue.c > cc1: warnings being treated as errors > tapdisk-queue.c: In function ‘tapdisk_lio_ack_event’: > tapdisk-queue.c:438: error: ignoring return value of ‘read’, declared > with attribute warn_unused_result > make[5]: *** [tapdisk-queue.o] Error 1 > make[5]: Leaving directory > `/root/Xen/xen-4.2-unstable/tools/blktap2/drivers' > make[4]: *** [subdir-install-drivers] Error 2 > make[4]: Leaving directory `/root/Xen/xen-4.2-unstable/tools/blktap2' > make[3]: *** [subdirs-install] Error 2 > make[3]: Leaving directory `/root/Xen/xen-4.2-unstable/tools/blktap2' > make[2]: *** [subdir-install-blktap2] Error 2 > make[2]: Leaving directory `/root/Xen/xen-4.2-unstable/tools' > make[1]: *** [subdirs-install] Error 2 > make[1]: Leaving directory `/root/Xen/xen-4.2-unstable/tools' > make: *** [install-tools] Error 2 > ------ > root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c -o tapdisk-log.o tapdisk-log.c > cc1: warnings being treated as errors > tapdisk-log.c: In function ‘tlog_flush’: > tapdisk-log.c:250: error: ignoring return value of ‘write’, declared > with attribute warn_unused_result > ------ > /root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c -o tapdisk2.o tapdisk2.c > cc1: warnings being treated as errors > tapdisk2.c: In function ‘main’: > tapdisk2.c:82: error: ignoring return value of ‘chdir’, declared with > attribute warn_unused_result > ------ > cc1: warnings being treated as errors > tapdisk-stream.c: In function ‘tapdisk_stream_poll_clear’: > tapdisk-stream.c:148: error: ignoring return value of ‘read’, declared > with attribute warn_unused_result > tapdisk-stream.c: In function ‘tapdisk_stream_poll_set’: > tapdisk-stream.c:158: error: ignoring return value of ‘write’, > declared with attribute warn_unused_result > tapdisk-stream.c: In function ‘tapdisk_stream_print_request’: > tapdisk-stream.c:206: error: ignoring return value of ‘write’, > declared with attribute warn_unused_result > > > > Best Regards, > Bei Guan >_______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Olaf Hering
2012-May-11 14:19 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On Fri, May 11, Bei Guan wrote:> Hi, > > When I do the "make install-tools" with xen-4.2-unstable, there are some errors > about "warn_unused_result". > Is it the error in code or the error in the compiling environment? Thank you so > much.I suggest to remove all -Werror from the Makefiles and introduce a --disable-Werror/--enable-Werror configure option, similar to what binutils and other projects have (not sure what the exact configure option name is in those projects). Olaf
Bei Guan
2012-May-11 15:12 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
2012/5/11 Ian Campbell <Ian.Campbell@citrix.com>> On Fri, 2012-05-11 at 15:08 +0100, Bei Guan wrote: > > When I do the "make install-tools" with xen-4.2-unstable, there are > > some errors about "warn_unused_result". > > Is it the error in code or the error in the compiling environment? > > Thank you so much. > > This commit > changeset: 25275:27d63b9f111a > user: Keir Fraser <keir@xen.org> > date: Thu May 10 11:22:18 2012 +0100 > summary: blktap2: Do not build with -O0 > has caused more warnings to be generated by the compiler, which in turn > leads to build failures. Since this is a compiler specific thing we are > still tracking them all down. > > Can you confirm which versionof xen-unstable.hg you are using -- there > have been several fixup patches since the above. If you are completely > up to date and you still see errors we can dig into whatever is left. >Thank you for your reply. I have used "hg pull; hg update" to update to the latest Xen and it is the version info: root@gavin-desktop:~/Xen/xen-4.2-unstable# hg tip changeset: 25287:54c8c9eaee92 tag: tip user: Dietmar Hahn <dietmar.hahn@ts.fujitsu.com> date: Fri Apr 27 11:09:26 2012 +0200 summary: docs: add vpmu description to xen-command-line.markdown But, it still has the same errors. Thanks, Bei Guan> > Ian. (starting to think that this change wasn''t suitable during a > feature freeze....) > > > > > > > gcc -O1 -fno-omit-frame-pointer -m64 -g -fno-strict-aliasing > > -std=gnu99 -Wall -Wstrict-prototypes -Wdeclaration-after-statement > > -D__XEN_TOOLS__ -MMD -MF .tapdisk-queue.o.d -D_LARGEFILE_SOURCE > > -D_LARGEFILE64_SOURCE -fno-optimize-sibling-calls -Werror -g > > -Wno-unused -fno-strict-aliasing -I../include -I../drivers > > -I/root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libxc > -I/root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/include > -D_GNU_SOURCE -DUSE_NFS_LOCKS -fPIC -I > /root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src > -c -o tapdisk-queue.o tapdisk-queue.c > > cc1: warnings being treated as errors > > tapdisk-queue.c: In function ‘tapdisk_lio_ack_event’: > > tapdisk-queue.c:438: error: ignoring return value of ‘read’, declared > > with attribute warn_unused_result > > make[5]: *** [tapdisk-queue.o] Error 1 > > make[5]: Leaving directory > > `/root/Xen/xen-4.2-unstable/tools/blktap2/drivers'' > > make[4]: *** [subdir-install-drivers] Error 2 > > make[4]: Leaving directory `/root/Xen/xen-4.2-unstable/tools/blktap2'' > > make[3]: *** [subdirs-install] Error 2 > > make[3]: Leaving directory `/root/Xen/xen-4.2-unstable/tools/blktap2'' > > make[2]: *** [subdir-install-blktap2] Error 2 > > make[2]: Leaving directory `/root/Xen/xen-4.2-unstable/tools'' > > make[1]: *** [subdirs-install] Error 2 > > make[1]: Leaving directory `/root/Xen/xen-4.2-unstable/tools'' > > make: *** [install-tools] Error 2 > > ------ > > > root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src > -c -o tapdisk-log.o tapdisk-log.c > > cc1: warnings being treated as errors > > tapdisk-log.c: In function ‘tlog_flush’: > > tapdisk-log.c:250: error: ignoring return value of ‘write’, declared > > with attribute warn_unused_result > > ------ > > > /root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src > -c -o tapdisk2.o tapdisk2.c > > cc1: warnings being treated as errors > > tapdisk2.c: In function ‘main’: > > tapdisk2.c:82: error: ignoring return value of ‘chdir’, declared with > > attribute warn_unused_result > > ------ > > cc1: warnings being treated as errors > > tapdisk-stream.c: In function ‘tapdisk_stream_poll_clear’: > > tapdisk-stream.c:148: error: ignoring return value of ‘read’, declared > > with attribute warn_unused_result > > tapdisk-stream.c: In function ‘tapdisk_stream_poll_set’: > > tapdisk-stream.c:158: error: ignoring return value of ‘write’, > > declared with attribute warn_unused_result > > tapdisk-stream.c: In function ‘tapdisk_stream_print_request’: > > tapdisk-stream.c:206: error: ignoring return value of ‘write’, > > declared with attribute warn_unused_result > > > > > > > > Best Regards, > > Bei Guan > > > > >-- Best Regards, Bei Guan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Bei Guan
2012-May-11 15:26 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
2012/5/11 Olaf Hering <olaf@aepfle.de>> On Fri, May 11, Bei Guan wrote: > > > Hi, > > > > When I do the "make install-tools" with xen-4.2-unstable, there are some > errors > > about "warn_unused_result". > > Is it the error in code or the error in the compiling environment? Thank > you so > > much. > > I suggest to remove all -Werror from the Makefiles and introduce a > --disable-Werror/--enable-Werror configure option, similar to what > binutils and other projects have (not sure what the exact configure > option name is in those projects). >Thank you so much for your reply too. It is better to have such as a choice to enable or disable -Werror option. But, now it doesn''t have in the configure file. Thanks, Bei Guan> > Olaf > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel >-- Best Regards, Bei Guan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Ian Campbell
2012-May-11 15:47 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On Fri, 2012-05-11 at 16:12 +0100, Bei Guan wrote:> > 2012/5/11 Ian Campbell <Ian.Campbell@citrix.com> > On Fri, 2012-05-11 at 15:08 +0100, Bei Guan wrote: > > When I do the "make install-tools" with xen-4.2-unstable, > there are > > some errors about "warn_unused_result". > > Is it the error in code or the error in the compiling > environment? > > Thank you so much. > > > This commit > changeset: 25275:27d63b9f111a > user: Keir Fraser <keir@xen.org> > date: Thu May 10 11:22:18 2012 +0100 > summary: blktap2: Do not build with -O0 > has caused more warnings to be generated by the compiler, > which in turn > leads to build failures. Since this is a compiler specific > thing we are > still tracking them all down. > > Can you confirm which versionof xen-unstable.hg you are using > -- there > have been several fixup patches since the above. If you are > completely > up to date and you still see errors we can dig into whatever > is left. > > Thank you for your reply. I have used "hg pull; hg update" to update > to the latest Xen and it is the version info: > > root@gavin-desktop:~/Xen/xen-4.2-unstable# hg tip > changeset: 25287:54c8c9eaee92 > tag: tip > user: Dietmar Hahn <dietmar.hahn@ts.fujitsu.com> > date: Fri Apr 27 11:09:26 2012 +0200 > summary: docs: add vpmu description to xen-command-line.markdown > > But, it still has the same errors.Thanks, looks like we still need to fix these errors then. I''m not really happy with the following, but it''s no worse than things are today... Bei, I don''t see these build failures, does this work for you? 8<------------------------------------------------------ # HG changeset patch # User Ian Campbell <ian.campbell@citrix.com> # Date 1336751175 -3600 # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 blktap: avoid attribute warn_unused_result build failures. I''m not proud of this, but since none of these callers of read/write have any other error handling and return void themselves (for several links up the call chain AFAICT) and because I don''t really want to get into a massive reworking of blktap2 I suppose it is at least pragmatic Signed-off-by: Ian Campbell <ian.campbell@citrix.com> diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-log.c --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 16:19:16 2012 +0100 +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 16:46:15 2012 +0100 @@ -247,7 +247,7 @@ tlog_flush(void) wsize = ((size + 511) & (~511)); memset(tapdisk_log.buf + size, ''\n'', wsize - size); - write(fd, tapdisk_log.buf, wsize); + (void)write(fd, tapdisk_log.buf, wsize); tapdisk_log.p = tapdisk_log.buf; diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-queue.c --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 16:19:16 2012 +0100 +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 16:46:15 2012 +0100 @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que uint64_t val; if (lio->flags & LIO_FLAG_EVENTFD) - read(lio->event_fd, &val, sizeof(val)); + (void)read(lio->event_fd, &val, sizeof(val)); } static void diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-stream.c --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 16:19:16 2012 +0100 +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 16:46:15 2012 +0100 @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk { int dummy; - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); p->set = 0; } @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s int dummy = 0; if (!p->set) { - write(p->pipe[POLL_WRITE], &dummy, sizeof(dummy)); + (void)write(p->pipe[POLL_WRITE], &dummy, sizeof(dummy)); p->set = 1; } } @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd { unsigned long idx = (unsigned long)tapdisk_stream_request_idx(s, sreq); char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, idx, 0); - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); + (void)write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); } static void diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk2.c --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 2012 +0100 +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 2012 +0100 @@ -79,7 +79,12 @@ main(int argc, char *argv[]) if (optind != argc) usage(argv[0], EINVAL); - chdir("/"); + if (chdir("/")) { + DPRINTF("failed to chdir(/): %d\n", errno); + err = 1; + goto out; + } + tapdisk_start_logging("tapdisk2"); err = tapdisk_server_init();
Bei Guan
2012-May-11 16:10 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
2012/5/11 Ian Campbell <Ian.Campbell@citrix.com>> On Fri, 2012-05-11 at 16:12 +0100, Bei Guan wrote: > > > > 2012/5/11 Ian Campbell <Ian.Campbell@citrix.com> > > On Fri, 2012-05-11 at 15:08 +0100, Bei Guan wrote: > > > When I do the "make install-tools" with xen-4.2-unstable, > > there are > > > some errors about "warn_unused_result". > > > Is it the error in code or the error in the compiling > > environment? > > > Thank you so much. > > > > > > This commit > > changeset: 25275:27d63b9f111a > > user: Keir Fraser <keir@xen.org> > > date: Thu May 10 11:22:18 2012 +0100 > > summary: blktap2: Do not build with -O0 > > has caused more warnings to be generated by the compiler, > > which in turn > > leads to build failures. Since this is a compiler specific > > thing we are > > still tracking them all down. > > > > Can you confirm which versionof xen-unstable.hg you are using > > -- there > > have been several fixup patches since the above. If you are > > completely > > up to date and you still see errors we can dig into whatever > > is left. > > > > Thank you for your reply. I have used "hg pull; hg update" to update > > to the latest Xen and it is the version info: > > > > root@gavin-desktop:~/Xen/xen-4.2-unstable# hg tip > > changeset: 25287:54c8c9eaee92 > > tag: tip > > user: Dietmar Hahn <dietmar.hahn@ts.fujitsu.com> > > date: Fri Apr 27 11:09:26 2012 +0200 > > summary: docs: add vpmu description to xen-command-line.markdown > > > > But, it still has the same errors. > > Thanks, looks like we still need to fix these errors then. > > I''m not really happy with the following, but it''s no worse than things > are today... > > Bei, I don''t see these build failures, does this work for you? >No, it doesn''t work for me. There is the same error. My environment and gcc version are like this: root@gavin-desktop:~# uname -a Linux gavin-desktop 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC 2011 x86_64 GNU/Linux root@gavin-desktop:~# gcc -v Using built-in specs. Target: x86_64-linux-gnu Configured with: ../src/configure -v --with-pkgversion=''Ubuntu 4.4.3-4ubuntu5'' --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 --program-suffix=-4.4 --enable-nls --enable-clocale=gnu --enable-libstdcxx-debug --enable-plugin --enable-objc-gc --disable-werror --with-arch-32=i486 --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu Thread model: posix gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) Thanks, Bei Guan> > 8<------------------------------------------------------ > > # HG changeset patch > # User Ian Campbell <ian.campbell@citrix.com> > # Date 1336751175 -3600 > # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 > # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 > blktap: avoid attribute warn_unused_result build failures. > > I''m not proud of this, but since none of these callers of read/write have > any > other error handling and return void themselves (for several links up the > call > chain AFAICT) and because I don''t really want to get into a massive > reworking > of blktap2 I suppose it is at least pragmatic > > Signed-off-by: Ian Campbell <ian.campbell@citrix.com> > > diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-log.c > --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 16:19:16 2012 > +0100 > +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 16:46:15 2012 > +0100 > @@ -247,7 +247,7 @@ tlog_flush(void) > wsize = ((size + 511) & (~511)); > > memset(tapdisk_log.buf + size, ''\n'', wsize - size); > - write(fd, tapdisk_log.buf, wsize); > + (void)write(fd, tapdisk_log.buf, wsize); > > tapdisk_log.p = tapdisk_log.buf; > > diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-queue.c > --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 16:19:16 2012 > +0100 > +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 16:46:15 2012 > +0100 > @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que > uint64_t val; > > if (lio->flags & LIO_FLAG_EVENTFD) > - read(lio->event_fd, &val, sizeof(val)); > + (void)read(lio->event_fd, &val, sizeof(val)); > } > > static void > diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-stream.c > --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 16:19:16 2012 > +0100 > +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 16:46:15 2012 > +0100 > @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk > { > int dummy; > > - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > p->set = 0; > } > > @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s > int dummy = 0; > > if (!p->set) { > - write(p->pipe[POLL_WRITE], &dummy, sizeof(dummy)); > + (void)write(p->pipe[POLL_WRITE], &dummy, sizeof(dummy)); > p->set = 1; > } > } > @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd > { > unsigned long idx = (unsigned long)tapdisk_stream_request_idx(s, > sreq); > char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, idx, 0); > - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); > + (void)write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); > } > > static void > diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk2.c > --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 2012 +0100 > +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 2012 +0100 > @@ -79,7 +79,12 @@ main(int argc, char *argv[]) > if (optind != argc) > usage(argv[0], EINVAL); > > - chdir("/"); > + if (chdir("/")) { > + DPRINTF("failed to chdir(/): %d\n", errno); > + err = 1; > + goto out; > + } > + > tapdisk_start_logging("tapdisk2"); > > err = tapdisk_server_init(); > > >-- Best Regards, Bei Guan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Keir Fraser
2012-May-11 16:16 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On 11/05/2012 15:16, "Ian Campbell" <Ian.Campbell@citrix.com> wrote:> On Fri, 2012-05-11 at 15:08 +0100, Bei Guan wrote: > When I do the "make > install-tools" with xen-4.2-unstable, there are > some errors about > "warn_unused_result". > Is it the error in code or the error in the compiling > environment? > Thank you so much.This commit changeset:> 25275:27d63b9f111auser: Keir Fraser <keir@xen.org>> date: Thu May 10 11:22:18 2012 +0100summary: blktap2: Do> not build with -O0has caused more warnings to be generated by the compiler,> which in turnleads to build failures. Since this is a compiler specific thing> we arestill tracking them all down. Can you confirm which versionof> xen-unstable.hg you are using -- therehave been several fixup patches since> the above. If you are completelyup to date and you still see errors we can> dig into whatever is left.Ian. (starting to think that this change wasn''t> suitable during afeature freeze....) The original reason for making this fix now was build failure on Fedora 17, due to -O0 implying -D_FORTIFY_SOURCE=2, and some kind of fall-out from that. A simpler feature-freeze fix might be to simply disable -Werror for subtrees that are using -O0. Since this needs backport for 4.0 and 4.1 as well. Or perhaps we are nearly at the end of the bug tail. ;-) -- Keir> > > gcc -O1 > -fno-omit-frame-pointer -m64 -g -fno-strict-aliasing > -std=gnu99 -Wall > -Wstrict-prototypes -Wdeclaration-after-statement > -D__XEN_TOOLS__ -MMD -MF > .tapdisk-queue.o.d -D_LARGEFILE_SOURCE > -D_LARGEFILE64_SOURCE > -fno-optimize-sibling-calls -Werror -g > -Wno-unused -fno-strict-aliasing > -I../include -I../drivers > > -I/root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libxc > -I/root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/include > -D_GNU_SOURCE -DUSE_NFS_LOCKS -fPIC -I > /root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c > -o tapdisk-queue.o tapdisk-queue.c > cc1: warnings being treated as errors > > tapdisk-queue.c: In function tapdisk_lio_ack_event¹: > tapdisk-queue.c:438: > error: ignoring return value of read¹, declared > with attribute > warn_unused_result > make[5]: *** [tapdisk-queue.o] Error 1 > make[5]: Leaving > directory > `/root/Xen/xen-4.2-unstable/tools/blktap2/drivers'' > make[4]: *** > [subdir-install-drivers] Error 2 > make[4]: Leaving directory > `/root/Xen/xen-4.2-unstable/tools/blktap2'' > make[3]: *** [subdirs-install] > Error 2 > make[3]: Leaving directory > `/root/Xen/xen-4.2-unstable/tools/blktap2'' > make[2]: *** > [subdir-install-blktap2] Error 2 > make[2]: Leaving directory > `/root/Xen/xen-4.2-unstable/tools'' > make[1]: *** [subdirs-install] Error 2 > > make[1]: Leaving directory `/root/Xen/xen-4.2-unstable/tools'' > make: *** > [install-tools] Error 2 > ------ > > root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c > -o tapdisk-log.o tapdisk-log.c > cc1: warnings being treated as errors > > tapdisk-log.c: In function tlog_flush¹: > tapdisk-log.c:250: error: ignoring > return value of write¹, declared > with attribute warn_unused_result > > ------ > > /root/Xen/xen-4.2-unstable/tools/blktap2/drivers/../../../tools/libaio/src -c > -o tapdisk2.o tapdisk2.c > cc1: warnings being treated as errors > > tapdisk2.c: In function main¹: > tapdisk2.c:82: error: ignoring return value > of chdir¹, declared with > attribute warn_unused_result > ------ > cc1: > warnings being treated as errors > tapdisk-stream.c: In function > tapdisk_stream_poll_clear¹: > tapdisk-stream.c:148: error: ignoring return > value of read¹, declared > with attribute warn_unused_result > > tapdisk-stream.c: In function tapdisk_stream_poll_set¹: > > tapdisk-stream.c:158: error: ignoring return value of write¹, > declared with > attribute warn_unused_result > tapdisk-stream.c: In function > tapdisk_stream_print_request¹: > tapdisk-stream.c:206: error: ignoring return > value of write¹, > declared with attribute warn_unused_result > > > > Best > Regards, > Bei Guan > >_______________________________________________ Xen-devel mailing> listXen-devel@lists.xen.org http://lists.xen.org/xen-devel
Keir Fraser
2012-May-11 16:18 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On 11/05/2012 17:10, "Bei Guan" <gbtju85@gmail.com> wrote:> 2012/5/11 Ian Campbell <Ian.Campbell@citrix.com> >> On Fri, 2012-05-11 at 16:12 +0100, Bei Guan wrote: >>> >>> 2012/5/11 Ian Campbell <Ian.Campbell@citrix.com>>>> >>> Thank you for your reply. I have used "hg pull; hg update" to update >>> to the latest Xen and it is the version info: >>> >>> root@gavin-desktop:~/Xen/xen-4.2-unstable# hg tip >>> changeset: 25287:54c8c9eaee92 >>> tag: tip >>> user: Dietmar Hahn <dietmar.hahn@ts.fujitsu.com> >>> date: Fri Apr 27 11:09:26 2012 +0200 >>> summary: docs: add vpmu description to xen-command-line.markdown >>> >>> But, it still has the same errors. >> >> Thanks, looks like we still need to fix these errors then. >> >> I''m not really happy with the following, but it''s no worse than things >> are today... >> >> Bei, I don''t see these build failures, does this work for you? > > No, it doesn''t work for me. There is the same error. > My environment and gcc version are like this:I think the problem is that casting the function result to void doesn''t work around this warning. You actually have to do something with the return code. :-( -- Keir> root@gavin-desktop:~# uname -a > Linux gavin-desktop 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC 2011 > x86_64 GNU/Linux > > root@gavin-desktop:~# gcc -v > Using built-in specs. > Target: x86_64-linux-gnu > Configured with: ../src/configure -v --with-pkgversion=''Ubuntu 4.4.3-4ubuntu5'' > --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs > --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared > --enable-multiarch --enable-linker-build-id --with-system-zlib > --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix > --with-gxx-include-dir=/usr/include/c++/4.4 --program-suffix=-4.4 --enable-nls > --enable-clocale=gnu --enable-libstdcxx-debug --enable-plugin --enable-objc-gc > --disable-werror --with-arch-32=i486 --with-tune=generic > --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu > --target=x86_64-linux-gnu > Thread model: posix > gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) > > > > > Thanks, > Bei Guan > > > > >> >> 8<------------------------------------------------------ >> >> # HG changeset patch >> # User Ian Campbell <ian.campbell@citrix.com> >> # Date 1336751175 -3600 >> # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 >> # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 >> blktap: avoid attribute warn_unused_result build failures. >> >> I''m not proud of this, but since none of these callers of read/write have any >> other error handling and return void themselves (for several links up the >> call >> chain AFAICT) and because I don''t really want to get into a massive reworking >> of blktap2 I suppose it is at least pragmatic >> >> Signed-off-by: Ian Campbell <ian.campbell@citrix.com> >> >> diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-log.c >> --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 16:19:16 2012 >> +0100 >> +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 16:46:15 2012 >> +0100 >> @@ -247,7 +247,7 @@ tlog_flush(void) >> wsize = ((size + 511) & (~511)); >> >> memset(tapdisk_log.buf + size, ''\n'', wsize - size); >> - write(fd, tapdisk_log.buf, wsize); >> + (void)write(fd, tapdisk_log.buf, wsize); >> >> tapdisk_log.p = tapdisk_log.buf; >> >> diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-queue.c >> --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 16:19:16 2012 >> +0100 >> +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 16:46:15 2012 >> +0100 >> @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que >> uint64_t val; >> >> if (lio->flags & LIO_FLAG_EVENTFD) >> - read(lio->event_fd, &val, sizeof(val)); >> + (void)read(lio->event_fd, &val, sizeof(val)); >> } >> >> static void >> diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk-stream.c >> --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 16:19:16 2012 >> +0100 >> +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 16:46:15 2012 >> +0100 >> @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk >> { >> int dummy; >> >> - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); >> + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); >> p->set = 0; >> } >> >> @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s >> int dummy = 0; >> >> if (!p->set) { >> - write(p->pipe[POLL_WRITE], &dummy, sizeof(dummy)); >> + (void)write(p->pipe[POLL_WRITE], &dummy, sizeof(dummy)); >> p->set = 1; >> } >> } >> @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd >> { >> unsigned long idx = (unsigned long)tapdisk_stream_request_idx(s, >> sreq); >> char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, idx, 0); >> - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); >> + (void)write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); >> } >> >> static void >> diff -r adc74e492edf -r 15ed8f45c4e5 tools/blktap2/drivers/tapdisk2.c >> --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 2012 +0100 >> +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 2012 +0100 >> @@ -79,7 +79,12 @@ main(int argc, char *argv[]) >> if (optind != argc) >> usage(argv[0], EINVAL); >> >> - chdir("/"); >> + if (chdir("/")) { >> + DPRINTF("failed to chdir(/): %d\n", errno); >> + err = 1; >> + goto out; >> + } >> + >> tapdisk_start_logging("tapdisk2"); >> >> err = tapdisk_server_init(); >> >> > >
Ian Campbell
2012-May-11 16:24 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On Fri, 2012-05-11 at 17:10 +0100, Bei Guan wrote:> No, it doesn''t work for me. There is the same error.Hrm. Since you can reproduce would you mind experimenting a bit to figure out the correct fix for this? Perhaps adding -Wno-unused-result to C flags helps (google suggests it may not)? Or maybe just assigning to a dummy variable. Ian.> My environment and gcc version are like this: > > root@gavin-desktop:~# uname -a > Linux gavin-desktop 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC > 2011 x86_64 GNU/Linux > > root@gavin-desktop:~# gcc -v > Using built-in specs. > Target: x86_64-linux-gnu > Configured with: ../src/configure -v --with-pkgversion=''Ubuntu > 4.4.3-4ubuntu5'' > --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs > --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr > --enable-shared --enable-multiarch --enable-linker-build-id > --with-system-zlib --libexecdir=/usr/lib --without-included-gettext > --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 > --program-suffix=-4.4 --enable-nls --enable-clocale=gnu > --enable-libstdcxx-debug --enable-plugin --enable-objc-gc > --disable-werror --with-arch-32=i486 --with-tune=generic > --enable-checking=release --build=x86_64-linux-gnu > --host=x86_64-linux-gnu --target=x86_64-linux-gnu > Thread model: posix > gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) > > > > > Thanks, > Bei Guan > > > > > > 8<------------------------------------------------------ > > # HG changeset patch > # User Ian Campbell <ian.campbell@citrix.com> > # Date 1336751175 -3600 > # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 > # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 > blktap: avoid attribute warn_unused_result build failures. > > I''m not proud of this, but since none of these callers of > read/write have any > other error handling and return void themselves (for several > links up the call > chain AFAICT) and because I don''t really want to get into a > massive reworking > of blktap2 I suppose it is at least pragmatic > > Signed-off-by: Ian Campbell <ian.campbell@citrix.com> > > diff -r adc74e492edf -r 15ed8f45c4e5 > tools/blktap2/drivers/tapdisk-log.c > --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 > 16:19:16 2012 +0100 > +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 > 16:46:15 2012 +0100 > @@ -247,7 +247,7 @@ tlog_flush(void) > wsize = ((size + 511) & (~511)); > > memset(tapdisk_log.buf + size, ''\n'', wsize - size); > - write(fd, tapdisk_log.buf, wsize); > + (void)write(fd, tapdisk_log.buf, wsize); > > tapdisk_log.p = tapdisk_log.buf; > > diff -r adc74e492edf -r 15ed8f45c4e5 > tools/blktap2/drivers/tapdisk-queue.c > --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 > 16:19:16 2012 +0100 > +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 > 16:46:15 2012 +0100 > @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que > uint64_t val; > > if (lio->flags & LIO_FLAG_EVENTFD) > - read(lio->event_fd, &val, sizeof(val)); > + (void)read(lio->event_fd, &val, sizeof(val)); > } > > static void > diff -r adc74e492edf -r 15ed8f45c4e5 > tools/blktap2/drivers/tapdisk-stream.c > --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 > 16:19:16 2012 +0100 > +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 > 16:46:15 2012 +0100 > @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk > { > int dummy; > > - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > p->set = 0; > } > > @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s > int dummy = 0; > > if (!p->set) { > - write(p->pipe[POLL_WRITE], &dummy, > sizeof(dummy)); > + (void)write(p->pipe[POLL_WRITE], &dummy, > sizeof(dummy)); > p->set = 1; > } > } > @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd > { > unsigned long idx = (unsigned > long)tapdisk_stream_request_idx(s, sreq); > char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, > idx, 0); > - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); > + (void)write(s->out_fd, buf, sreq->secs << > SECTOR_SHIFT); > } > > static void > diff -r adc74e492edf -r 15ed8f45c4e5 > tools/blktap2/drivers/tapdisk2.c > --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 > 2012 +0100 > +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 > 2012 +0100 > @@ -79,7 +79,12 @@ main(int argc, char *argv[]) > if (optind != argc) > usage(argv[0], EINVAL); > > - chdir("/"); > + if (chdir("/")) { > + DPRINTF("failed to chdir(/): %d\n", errno); > + err = 1; > + goto out; > + } > + > tapdisk_start_logging("tapdisk2"); > > err = tapdisk_server_init(); > > > > > > -- > Best Regards, > Bei Guan >
Bei Guan
2012-May-11 16:33 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
2012/5/12 Ian Campbell <Ian.Campbell@citrix.com>> On Fri, 2012-05-11 at 17:10 +0100, Bei Guan wrote: > > > > No, it doesn''t work for me. There is the same error. > > Hrm. Since you can reproduce would you mind experimenting a bit to > figure out the correct fix for this? > > Perhaps adding -Wno-unused-result to C flags helps (google suggests it > may not)? >Yes, I add this option to Makefile. And now it works well. The patch is as the following. Thank you very much. diff -r 54c8c9eaee92 tools/blktap2/drivers/Makefile --- a/tools/blktap2/drivers/Makefile Fri Apr 27 11:09:26 2012 +0200 +++ b/tools/blktap2/drivers/Makefile Sat May 12 00:31:12 2012 +0800 @@ -11,6 +11,7 @@ CFLAGS += -Werror -g CFLAGS += -Wno-unused +CFLAGS += -Wno-unused-result CFLAGS += -fno-strict-aliasing CFLAGS += -I$(BLKTAP_ROOT)/include -I$(BLKTAP_ROOT)/drivers CFLAGS += $(CFLAGS_libxenctrl) Thanks, Bei Guan> > Or maybe just assigning to a dummy variable. > > Ian. > > > My environment and gcc version are like this: > > > > root@gavin-desktop:~# uname -a > > Linux gavin-desktop 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC > > 2011 x86_64 GNU/Linux > > > > root@gavin-desktop:~# gcc -v > > Using built-in specs. > > Target: x86_64-linux-gnu > > Configured with: ../src/configure -v --with-pkgversion=''Ubuntu > > 4.4.3-4ubuntu5'' > > --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs > > --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr > > --enable-shared --enable-multiarch --enable-linker-build-id > > --with-system-zlib --libexecdir=/usr/lib --without-included-gettext > > --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 > > --program-suffix=-4.4 --enable-nls --enable-clocale=gnu > > --enable-libstdcxx-debug --enable-plugin --enable-objc-gc > > --disable-werror --with-arch-32=i486 --with-tune=generic > > --enable-checking=release --build=x86_64-linux-gnu > > --host=x86_64-linux-gnu --target=x86_64-linux-gnu > > Thread model: posix > > gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) > > > > > > > > > > Thanks, > > Bei Guan > > > > > > > > > > > > 8<------------------------------------------------------ > > > > # HG changeset patch > > # User Ian Campbell <ian.campbell@citrix.com> > > # Date 1336751175 -3600 > > # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 > > # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 > > blktap: avoid attribute warn_unused_result build failures. > > > > I''m not proud of this, but since none of these callers of > > read/write have any > > other error handling and return void themselves (for several > > links up the call > > chain AFAICT) and because I don''t really want to get into a > > massive reworking > > of blktap2 I suppose it is at least pragmatic > > > > Signed-off-by: Ian Campbell <ian.campbell@citrix.com> > > > > diff -r adc74e492edf -r 15ed8f45c4e5 > > tools/blktap2/drivers/tapdisk-log.c > > --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 > > 16:19:16 2012 +0100 > > +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 > > 16:46:15 2012 +0100 > > @@ -247,7 +247,7 @@ tlog_flush(void) > > wsize = ((size + 511) & (~511)); > > > > memset(tapdisk_log.buf + size, ''\n'', wsize - size); > > - write(fd, tapdisk_log.buf, wsize); > > + (void)write(fd, tapdisk_log.buf, wsize); > > > > tapdisk_log.p = tapdisk_log.buf; > > > > diff -r adc74e492edf -r 15ed8f45c4e5 > > tools/blktap2/drivers/tapdisk-queue.c > > --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 > > 16:19:16 2012 +0100 > > +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 > > 16:46:15 2012 +0100 > > @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que > > uint64_t val; > > > > if (lio->flags & LIO_FLAG_EVENTFD) > > - read(lio->event_fd, &val, sizeof(val)); > > + (void)read(lio->event_fd, &val, sizeof(val)); > > } > > > > static void > > diff -r adc74e492edf -r 15ed8f45c4e5 > > tools/blktap2/drivers/tapdisk-stream.c > > --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 > > 16:19:16 2012 +0100 > > +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 > > 16:46:15 2012 +0100 > > @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk > > { > > int dummy; > > > > - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > > + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > > p->set = 0; > > } > > > > @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s > > int dummy = 0; > > > > if (!p->set) { > > - write(p->pipe[POLL_WRITE], &dummy, > > sizeof(dummy)); > > + (void)write(p->pipe[POLL_WRITE], &dummy, > > sizeof(dummy)); > > p->set = 1; > > } > > } > > @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd > > { > > unsigned long idx = (unsigned > > long)tapdisk_stream_request_idx(s, sreq); > > char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, > > idx, 0); > > - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); > > + (void)write(s->out_fd, buf, sreq->secs << > > SECTOR_SHIFT); > > } > > > > static void > > diff -r adc74e492edf -r 15ed8f45c4e5 > > tools/blktap2/drivers/tapdisk2.c > > --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 > > 2012 +0100 > > +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 > > 2012 +0100 > > @@ -79,7 +79,12 @@ main(int argc, char *argv[]) > > if (optind != argc) > > usage(argv[0], EINVAL); > > > > - chdir("/"); > > + if (chdir("/")) { > > + DPRINTF("failed to chdir(/): %d\n", errno); > > + err = 1; > > + goto out; > > + } > > + > > tapdisk_start_logging("tapdisk2"); > > > > err = tapdisk_server_init(); > > > > > > > > > > > > -- > > Best Regards, > > Bei Guan > > > > >-- Best Regards, Bei Guan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Keir Fraser
2012-May-11 16:44 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On 11/05/2012 17:24, "Ian Campbell" <Ian.Campbell@citrix.com> wrote:> On Fri, 2012-05-11 at 17:10 +0100, Bei Guan wrote: > > >> No, it doesn''t work for me. There is the same error. > > Hrm. Since you can reproduce would you mind experimenting a bit to > figure out the correct fix for this? > > Perhaps adding -Wno-unused-result to C flags helps (google suggests it > may not)? > > Or maybe just assigning to a dummy variable.Bei, Please try the attached patch. -- Keir> Ian. > >> My environment and gcc version are like this: >> >> root@gavin-desktop:~# uname -a >> Linux gavin-desktop 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC >> 2011 x86_64 GNU/Linux >> >> root@gavin-desktop:~# gcc -v >> Using built-in specs. >> Target: x86_64-linux-gnu >> Configured with: ../src/configure -v --with-pkgversion=''Ubuntu >> 4.4.3-4ubuntu5'' >> --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs >> --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr >> --enable-shared --enable-multiarch --enable-linker-build-id >> --with-system-zlib --libexecdir=/usr/lib --without-included-gettext >> --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 >> --program-suffix=-4.4 --enable-nls --enable-clocale=gnu >> --enable-libstdcxx-debug --enable-plugin --enable-objc-gc >> --disable-werror --with-arch-32=i486 --with-tune=generic >> --enable-checking=release --build=x86_64-linux-gnu >> --host=x86_64-linux-gnu --target=x86_64-linux-gnu >> Thread model: posix >> gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) >> >> >> >> >> Thanks, >> Bei Guan >> >> >> >> >> >> 8<------------------------------------------------------ >> >> # HG changeset patch >> # User Ian Campbell <ian.campbell@citrix.com> >> # Date 1336751175 -3600 >> # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 >> # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 >> blktap: avoid attribute warn_unused_result build failures. >> >> I''m not proud of this, but since none of these callers of >> read/write have any >> other error handling and return void themselves (for several >> links up the call >> chain AFAICT) and because I don''t really want to get into a >> massive reworking >> of blktap2 I suppose it is at least pragmatic >> >> Signed-off-by: Ian Campbell <ian.campbell@citrix.com> >> >> diff -r adc74e492edf -r 15ed8f45c4e5 >> tools/blktap2/drivers/tapdisk-log.c >> --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 >> 16:19:16 2012 +0100 >> +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 >> 16:46:15 2012 +0100 >> @@ -247,7 +247,7 @@ tlog_flush(void) >> wsize = ((size + 511) & (~511)); >> >> memset(tapdisk_log.buf + size, ''\n'', wsize - size); >> - write(fd, tapdisk_log.buf, wsize); >> + (void)write(fd, tapdisk_log.buf, wsize); >> >> tapdisk_log.p = tapdisk_log.buf; >> >> diff -r adc74e492edf -r 15ed8f45c4e5 >> tools/blktap2/drivers/tapdisk-queue.c >> --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 >> 16:19:16 2012 +0100 >> +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 >> 16:46:15 2012 +0100 >> @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que >> uint64_t val; >> >> if (lio->flags & LIO_FLAG_EVENTFD) >> - read(lio->event_fd, &val, sizeof(val)); >> + (void)read(lio->event_fd, &val, sizeof(val)); >> } >> >> static void >> diff -r adc74e492edf -r 15ed8f45c4e5 >> tools/blktap2/drivers/tapdisk-stream.c >> --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 >> 16:19:16 2012 +0100 >> +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 >> 16:46:15 2012 +0100 >> @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk >> { >> int dummy; >> >> - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); >> + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); >> p->set = 0; >> } >> >> @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s >> int dummy = 0; >> >> if (!p->set) { >> - write(p->pipe[POLL_WRITE], &dummy, >> sizeof(dummy)); >> + (void)write(p->pipe[POLL_WRITE], &dummy, >> sizeof(dummy)); >> p->set = 1; >> } >> } >> @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd >> { >> unsigned long idx = (unsigned >> long)tapdisk_stream_request_idx(s, sreq); >> char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, >> idx, 0); >> - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); >> + (void)write(s->out_fd, buf, sreq->secs << >> SECTOR_SHIFT); >> } >> >> static void >> diff -r adc74e492edf -r 15ed8f45c4e5 >> tools/blktap2/drivers/tapdisk2.c >> --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 >> 2012 +0100 >> +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 >> 2012 +0100 >> @@ -79,7 +79,12 @@ main(int argc, char *argv[]) >> if (optind != argc) >> usage(argv[0], EINVAL); >> >> - chdir("/"); >> + if (chdir("/")) { >> + DPRINTF("failed to chdir(/): %d\n", errno); >> + err = 1; >> + goto out; >> + } >> + >> tapdisk_start_logging("tapdisk2"); >> >> err = tapdisk_server_init(); >> >> >> >> >> >> -- >> Best Regards, >> Bei Guan >> > >_______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Keir Fraser
2012-May-11 16:47 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On 11/05/2012 17:33, "Bei Guan" <gbtju85@gmail.com> wrote:> 2012/5/12 Ian Campbell <Ian.Campbell@citrix.com> >> On Fri, 2012-05-11 at 17:10 +0100, Bei Guan wrote: >> >> >>> No, it doesn''t work for me. There is the same error. >> >> Hrm. Since you can reproduce would you mind experimenting a bit to >> figure out the correct fix for this? >> >> Perhaps adding -Wno-unused-result to C flags helps (google suggests it >> may not)? > Yes, I add this option to Makefile. And now it works well. The patch is as the > following. Thank you very much.I''d be worried about more command-line meddling causing even more fallout. Particularly because we haven''t used that option anywhere else so far, so it''s untested by us. E.g., do all gcc support -Wno-unused-result? -- Keir> > diff -r 54c8c9eaee92 tools/blktap2/drivers/Makefile > --- a/tools/blktap2/drivers/Makefile Fri Apr 27 11:09:26 2012 +0200 > +++ b/tools/blktap2/drivers/Makefile Sat May 12 00:31:12 2012 +0800 > @@ -11,6 +11,7 @@ > > CFLAGS += -Werror -g > CFLAGS += -Wno-unused > +CFLAGS += -Wno-unused-result > CFLAGS += -fno-strict-aliasing > CFLAGS += -I$(BLKTAP_ROOT)/include -I$(BLKTAP_ROOT)/drivers > CFLAGS += $(CFLAGS_libxenctrl) > > > > > Thanks, > Bei Guan > > > > >> >> Or maybe just assigning to a dummy variable. >> >> Ian. >> >>> My environment and gcc version are like this: >>> >>> root@gavin-desktop:~# uname -a >>> Linux gavin-desktop 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC >>> 2011 x86_64 GNU/Linux >>> >>> root@gavin-desktop:~# gcc -v >>> Using built-in specs. >>> Target: x86_64-linux-gnu >>> Configured with: ../src/configure -v --with-pkgversion=''Ubuntu >>> 4.4.3-4ubuntu5'' >>> --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs >>> --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr >>> --enable-shared --enable-multiarch --enable-linker-build-id >>> --with-system-zlib --libexecdir=/usr/lib --without-included-gettext >>> --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 >>> --program-suffix=-4.4 --enable-nls --enable-clocale=gnu >>> --enable-libstdcxx-debug --enable-plugin --enable-objc-gc >>> --disable-werror --with-arch-32=i486 --with-tune=generic >>> --enable-checking=release --build=x86_64-linux-gnu >>> --host=x86_64-linux-gnu --target=x86_64-linux-gnu >>> Thread model: posix >>> gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) >>> >>> >>> >>> >>> Thanks, >>> Bei Guan >>> >>> >>> >>> >>> >>> 8<------------------------------------------------------ >>> >>> # HG changeset patch >>> # User Ian Campbell <ian.campbell@citrix.com> >>> # Date 1336751175 -3600 >>> # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 >>> # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 >>> blktap: avoid attribute warn_unused_result build failures. >>> >>> I''m not proud of this, but since none of these callers of >>> read/write have any >>> other error handling and return void themselves (for several >>> links up the call >>> chain AFAICT) and because I don''t really want to get into a >>> massive reworking >>> of blktap2 I suppose it is at least pragmatic >>> >>> Signed-off-by: Ian Campbell <ian.campbell@citrix.com> >>> >>> diff -r adc74e492edf -r 15ed8f45c4e5 >>> tools/blktap2/drivers/tapdisk-log.c >>> --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 >>> 16:19:16 2012 +0100 >>> +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 >>> 16:46:15 2012 +0100 >>> @@ -247,7 +247,7 @@ tlog_flush(void) >>> wsize = ((size + 511) & (~511)); >>> >>> memset(tapdisk_log.buf + size, ''\n'', wsize - size); >>> - write(fd, tapdisk_log.buf, wsize); >>> + (void)write(fd, tapdisk_log.buf, wsize); >>> >>> tapdisk_log.p = tapdisk_log.buf; >>> >>> diff -r adc74e492edf -r 15ed8f45c4e5 >>> tools/blktap2/drivers/tapdisk-queue.c >>> --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 >>> 16:19:16 2012 +0100 >>> +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 >>> 16:46:15 2012 +0100 >>> @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que >>> uint64_t val; >>> >>> if (lio->flags & LIO_FLAG_EVENTFD) >>> - read(lio->event_fd, &val, sizeof(val)); >>> + (void)read(lio->event_fd, &val, sizeof(val)); >>> } >>> >>> static void >>> diff -r adc74e492edf -r 15ed8f45c4e5 >>> tools/blktap2/drivers/tapdisk-stream.c >>> --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 >>> 16:19:16 2012 +0100 >>> +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 >>> 16:46:15 2012 +0100 >>> @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk >>> { >>> int dummy; >>> >>> - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); >>> + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); >>> p->set = 0; >>> } >>> >>> @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s >>> int dummy = 0; >>> >>> if (!p->set) { >>> - write(p->pipe[POLL_WRITE], &dummy, >>> sizeof(dummy)); >>> + (void)write(p->pipe[POLL_WRITE], &dummy, >>> sizeof(dummy)); >>> p->set = 1; >>> } >>> } >>> @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd >>> { >>> unsigned long idx = (unsigned >>> long)tapdisk_stream_request_idx(s, sreq); >>> char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, >>> idx, 0); >>> - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); >>> + (void)write(s->out_fd, buf, sreq->secs << >>> SECTOR_SHIFT); >>> } >>> >>> static void >>> diff -r adc74e492edf -r 15ed8f45c4e5 >>> tools/blktap2/drivers/tapdisk2.c >>> --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 >>> 2012 +0100 >>> +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 >>> 2012 +0100 >>> @@ -79,7 +79,12 @@ main(int argc, char *argv[]) >>> if (optind != argc) >>> usage(argv[0], EINVAL); >>> >>> - chdir("/"); >>> + if (chdir("/")) { >>> + DPRINTF("failed to chdir(/): %d\n", errno); >>> + err = 1; >>> + goto out; >>> + } >>> + >>> tapdisk_start_logging("tapdisk2"); >>> >>> err = tapdisk_server_init(); >>> >>> >>> >>> >>> >>> -- >>> Best Regards, >>> Bei Guan >>> >> >> > >
Bei Guan
2012-May-11 16:57 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
2012/5/12 Keir Fraser <keir@xen.org>> On 11/05/2012 17:24, "Ian Campbell" <Ian.Campbell@citrix.com> wrote: > > > On Fri, 2012-05-11 at 17:10 +0100, Bei Guan wrote: > > > > > >> No, it doesn''t work for me. There is the same error. > > > > Hrm. Since you can reproduce would you mind experimenting a bit to > > figure out the correct fix for this? > > > > Perhaps adding -Wno-unused-result to C flags helps (google suggests it > > may not)? > > > > Or maybe just assigning to a dummy variable. > > Bei, Please try the attached patch. >It works now. Thank you. Thanks, Bei Guan> > -- Keir > > > Ian. > > > >> My environment and gcc version are like this: > >> > >> root@gavin-desktop:~# uname -a > >> Linux gavin-desktop 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC > >> 2011 x86_64 GNU/Linux > >> > >> root@gavin-desktop:~# gcc -v > >> Using built-in specs. > >> Target: x86_64-linux-gnu > >> Configured with: ../src/configure -v --with-pkgversion=''Ubuntu > >> 4.4.3-4ubuntu5'' > >> --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs > >> --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr > >> --enable-shared --enable-multiarch --enable-linker-build-id > >> --with-system-zlib --libexecdir=/usr/lib --without-included-gettext > >> --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 > >> --program-suffix=-4.4 --enable-nls --enable-clocale=gnu > >> --enable-libstdcxx-debug --enable-plugin --enable-objc-gc > >> --disable-werror --with-arch-32=i486 --with-tune=generic > >> --enable-checking=release --build=x86_64-linux-gnu > >> --host=x86_64-linux-gnu --target=x86_64-linux-gnu > >> Thread model: posix > >> gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) > >> > >> > >> > >> > >> Thanks, > >> Bei Guan > >> > >> > >> > >> > >> > >> 8<------------------------------------------------------ > >> > >> # HG changeset patch > >> # User Ian Campbell <ian.campbell@citrix.com> > >> # Date 1336751175 -3600 > >> # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 > >> # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 > >> blktap: avoid attribute warn_unused_result build failures. > >> > >> I''m not proud of this, but since none of these callers of > >> read/write have any > >> other error handling and return void themselves (for several > >> links up the call > >> chain AFAICT) and because I don''t really want to get into a > >> massive reworking > >> of blktap2 I suppose it is at least pragmatic > >> > >> Signed-off-by: Ian Campbell <ian.campbell@citrix.com> > >> > >> diff -r adc74e492edf -r 15ed8f45c4e5 > >> tools/blktap2/drivers/tapdisk-log.c > >> --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 > >> 16:19:16 2012 +0100 > >> +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 > >> 16:46:15 2012 +0100 > >> @@ -247,7 +247,7 @@ tlog_flush(void) > >> wsize = ((size + 511) & (~511)); > >> > >> memset(tapdisk_log.buf + size, ''\n'', wsize - size); > >> - write(fd, tapdisk_log.buf, wsize); > >> + (void)write(fd, tapdisk_log.buf, wsize); > >> > >> tapdisk_log.p = tapdisk_log.buf; > >> > >> diff -r adc74e492edf -r 15ed8f45c4e5 > >> tools/blktap2/drivers/tapdisk-queue.c > >> --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 > >> 16:19:16 2012 +0100 > >> +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 > >> 16:46:15 2012 +0100 > >> @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que > >> uint64_t val; > >> > >> if (lio->flags & LIO_FLAG_EVENTFD) > >> - read(lio->event_fd, &val, sizeof(val)); > >> + (void)read(lio->event_fd, &val, sizeof(val)); > >> } > >> > >> static void > >> diff -r adc74e492edf -r 15ed8f45c4e5 > >> tools/blktap2/drivers/tapdisk-stream.c > >> --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 > >> 16:19:16 2012 +0100 > >> +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 > >> 16:46:15 2012 +0100 > >> @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk > >> { > >> int dummy; > >> > >> - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > >> + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > >> p->set = 0; > >> } > >> > >> @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s > >> int dummy = 0; > >> > >> if (!p->set) { > >> - write(p->pipe[POLL_WRITE], &dummy, > >> sizeof(dummy)); > >> + (void)write(p->pipe[POLL_WRITE], &dummy, > >> sizeof(dummy)); > >> p->set = 1; > >> } > >> } > >> @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd > >> { > >> unsigned long idx = (unsigned > >> long)tapdisk_stream_request_idx(s, sreq); > >> char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, > >> idx, 0); > >> - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); > >> + (void)write(s->out_fd, buf, sreq->secs << > >> SECTOR_SHIFT); > >> } > >> > >> static void > >> diff -r adc74e492edf -r 15ed8f45c4e5 > >> tools/blktap2/drivers/tapdisk2.c > >> --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 > >> 2012 +0100 > >> +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 > >> 2012 +0100 > >> @@ -79,7 +79,12 @@ main(int argc, char *argv[]) > >> if (optind != argc) > >> usage(argv[0], EINVAL); > >> > >> - chdir("/"); > >> + if (chdir("/")) { > >> + DPRINTF("failed to chdir(/): %d\n", errno); > >> + err = 1; > >> + goto out; > >> + } > >> + > >> tapdisk_start_logging("tapdisk2"); > >> > >> err = tapdisk_server_init(); > >> > >> > >> > >> > >> > >> -- > >> Best Regards, > >> Bei Guan > >> > > > > > >-- Best Regards, Bei Guan _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel
Ian Jackson
2012-May-11 17:01 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
~Keir Fraser writes ("Re: [Xen-devel] Errors of doing "make
install-tools" with xen-4.2-unstable?"):> On 11/05/2012 17:33, "Bei Guan" <gbtju85@gmail.com> wrote:
> Yes, I add this option to Makefile. And now it works well. The patch is as
the> > following. Thank you very much.
>
> I''d be worried about more command-line meddling causing even more
fallout.
> Particularly because we haven''t used that option anywhere else so
far, so
> it''s untested by us. E.g., do all gcc support -Wno-unused-result?
Checking the manuals (which are all online), gcc 4.0 doesn''t support
it. So no, relevant gccs don''t.
However the tools build now uses autoconf. I would accept a patch
which used autoconf to determine whether gcc accepts
-Wno-unused-result and to supply it in the blktap2 Makefile if it
does.
Ian.
Ian Campbell
2012-May-12 05:50 UTC
Re: Errors of doing "make install-tools" with xen-4.2-unstable?
On Fri, 2012-05-11 at 17:47 +0100, Keir Fraser wrote:> On 11/05/2012 17:33, "Bei Guan" <gbtju85@gmail.com> wrote: > > > 2012/5/12 Ian Campbell <Ian.Campbell@citrix.com> > >> On Fri, 2012-05-11 at 17:10 +0100, Bei Guan wrote: > >> > >> > >>> No, it doesn''t work for me. There is the same error. > >> > >> Hrm. Since you can reproduce would you mind experimenting a bit to > >> figure out the correct fix for this? > >> > >> Perhaps adding -Wno-unused-result to C flags helps (google suggests it > >> may not)? > > Yes, I add this option to Makefile. And now it works well. The patch is as the > > following. Thank you very much. > > I''d be worried about more command-line meddling causing even more fallout. > Particularly because we haven''t used that option anywhere else so far, so > it''s untested by us. E.g., do all gcc support -Wno-unused-result?We could have used cc-option, but your {read,write}_exact patch looks like a better fix anyway. Ian.> > -- Keir > > > > > diff -r 54c8c9eaee92 tools/blktap2/drivers/Makefile > > --- a/tools/blktap2/drivers/Makefile Fri Apr 27 11:09:26 2012 +0200 > > +++ b/tools/blktap2/drivers/Makefile Sat May 12 00:31:12 2012 +0800 > > @@ -11,6 +11,7 @@ > > > > CFLAGS += -Werror -g > > CFLAGS += -Wno-unused > > +CFLAGS += -Wno-unused-result > > CFLAGS += -fno-strict-aliasing > > CFLAGS += -I$(BLKTAP_ROOT)/include -I$(BLKTAP_ROOT)/drivers > > CFLAGS += $(CFLAGS_libxenctrl) > > > > > > > > > > Thanks, > > Bei Guan > > > > > > > > > >> > >> Or maybe just assigning to a dummy variable. > >> > >> Ian. > >> > >>> My environment and gcc version are like this: > >>> > >>> root@gavin-desktop:~# uname -a > >>> Linux gavin-desktop 2.6.32-5-xen-amd64 #1 SMP Thu May 19 01:16:47 UTC > >>> 2011 x86_64 GNU/Linux > >>> > >>> root@gavin-desktop:~# gcc -v > >>> Using built-in specs. > >>> Target: x86_64-linux-gnu > >>> Configured with: ../src/configure -v --with-pkgversion=''Ubuntu > >>> 4.4.3-4ubuntu5'' > >>> --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs > >>> --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr > >>> --enable-shared --enable-multiarch --enable-linker-build-id > >>> --with-system-zlib --libexecdir=/usr/lib --without-included-gettext > >>> --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 > >>> --program-suffix=-4.4 --enable-nls --enable-clocale=gnu > >>> --enable-libstdcxx-debug --enable-plugin --enable-objc-gc > >>> --disable-werror --with-arch-32=i486 --with-tune=generic > >>> --enable-checking=release --build=x86_64-linux-gnu > >>> --host=x86_64-linux-gnu --target=x86_64-linux-gnu > >>> Thread model: posix > >>> gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) > >>> > >>> > >>> > >>> > >>> Thanks, > >>> Bei Guan > >>> > >>> > >>> > >>> > >>> > >>> 8<------------------------------------------------------ > >>> > >>> # HG changeset patch > >>> # User Ian Campbell <ian.campbell@citrix.com> > >>> # Date 1336751175 -3600 > >>> # Node ID 15ed8f45c4e57a1e206af020e0ff17b792108e99 > >>> # Parent adc74e492edfee6cf44e433e3e93743a3cf71999 > >>> blktap: avoid attribute warn_unused_result build failures. > >>> > >>> I''m not proud of this, but since none of these callers of > >>> read/write have any > >>> other error handling and return void themselves (for several > >>> links up the call > >>> chain AFAICT) and because I don''t really want to get into a > >>> massive reworking > >>> of blktap2 I suppose it is at least pragmatic > >>> > >>> Signed-off-by: Ian Campbell <ian.campbell@citrix.com> > >>> > >>> diff -r adc74e492edf -r 15ed8f45c4e5 > >>> tools/blktap2/drivers/tapdisk-log.c > >>> --- a/tools/blktap2/drivers/tapdisk-log.c Fri May 11 > >>> 16:19:16 2012 +0100 > >>> +++ b/tools/blktap2/drivers/tapdisk-log.c Fri May 11 > >>> 16:46:15 2012 +0100 > >>> @@ -247,7 +247,7 @@ tlog_flush(void) > >>> wsize = ((size + 511) & (~511)); > >>> > >>> memset(tapdisk_log.buf + size, ''\n'', wsize - size); > >>> - write(fd, tapdisk_log.buf, wsize); > >>> + (void)write(fd, tapdisk_log.buf, wsize); > >>> > >>> tapdisk_log.p = tapdisk_log.buf; > >>> > >>> diff -r adc74e492edf -r 15ed8f45c4e5 > >>> tools/blktap2/drivers/tapdisk-queue.c > >>> --- a/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 > >>> 16:19:16 2012 +0100 > >>> +++ b/tools/blktap2/drivers/tapdisk-queue.c Fri May 11 > >>> 16:46:15 2012 +0100 > >>> @@ -435,7 +435,7 @@ tapdisk_lio_ack_event(struct tqueue *que > >>> uint64_t val; > >>> > >>> if (lio->flags & LIO_FLAG_EVENTFD) > >>> - read(lio->event_fd, &val, sizeof(val)); > >>> + (void)read(lio->event_fd, &val, sizeof(val)); > >>> } > >>> > >>> static void > >>> diff -r adc74e492edf -r 15ed8f45c4e5 > >>> tools/blktap2/drivers/tapdisk-stream.c > >>> --- a/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 > >>> 16:19:16 2012 +0100 > >>> +++ b/tools/blktap2/drivers/tapdisk-stream.c Fri May 11 > >>> 16:46:15 2012 +0100 > >>> @@ -145,7 +145,7 @@ tapdisk_stream_poll_clear(struct tapdisk > >>> { > >>> int dummy; > >>> > >>> - read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > >>> + (void)read(p->pipe[POLL_READ], &dummy, sizeof(dummy)); > >>> p->set = 0; > >>> } > >>> > >>> @@ -155,7 +155,7 @@ tapdisk_stream_poll_set(struct tapdisk_s > >>> int dummy = 0; > >>> > >>> if (!p->set) { > >>> - write(p->pipe[POLL_WRITE], &dummy, > >>> sizeof(dummy)); > >>> + (void)write(p->pipe[POLL_WRITE], &dummy, > >>> sizeof(dummy)); > >>> p->set = 1; > >>> } > >>> } > >>> @@ -203,7 +203,7 @@ tapdisk_stream_print_request(struct tapd > >>> { > >>> unsigned long idx = (unsigned > >>> long)tapdisk_stream_request_idx(s, sreq); > >>> char *buf = (char *)MMAP_VADDR(s->vbd->ring.vstart, > >>> idx, 0); > >>> - write(s->out_fd, buf, sreq->secs << SECTOR_SHIFT); > >>> + (void)write(s->out_fd, buf, sreq->secs << > >>> SECTOR_SHIFT); > >>> } > >>> > >>> static void > >>> diff -r adc74e492edf -r 15ed8f45c4e5 > >>> tools/blktap2/drivers/tapdisk2.c > >>> --- a/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:19:16 > >>> 2012 +0100 > >>> +++ b/tools/blktap2/drivers/tapdisk2.c Fri May 11 16:46:15 > >>> 2012 +0100 > >>> @@ -79,7 +79,12 @@ main(int argc, char *argv[]) > >>> if (optind != argc) > >>> usage(argv[0], EINVAL); > >>> > >>> - chdir("/"); > >>> + if (chdir("/")) { > >>> + DPRINTF("failed to chdir(/): %d\n", errno); > >>> + err = 1; > >>> + goto out; > >>> + } > >>> + > >>> tapdisk_start_logging("tapdisk2"); > >>> > >>> err = tapdisk_server_init(); > >>> > >>> > >>> > >>> > >>> > >>> -- > >>> Best Regards, > >>> Bei Guan > >>> > >> > >> > > > > > >