summary refs log tree commit diff stats
path: root/results/classifier/118/all/1801073
diff options
context:
space:
mode:
Diffstat (limited to 'results/classifier/118/all/1801073')
-rw-r--r--results/classifier/118/all/1801073108
1 files changed, 108 insertions, 0 deletions
diff --git a/results/classifier/118/all/1801073 b/results/classifier/118/all/1801073
new file mode 100644
index 000000000..a9b128f7f
--- /dev/null
+++ b/results/classifier/118/all/1801073
@@ -0,0 +1,108 @@
+risc-v: 0.949
+graphic: 0.945
+hypervisor: 0.940
+PID: 0.937
+register: 0.934
+user-level: 0.933
+virtual: 0.932
+permissions: 0.932
+network: 0.930
+KVM: 0.930
+architecture: 0.929
+semantic: 0.927
+VMM: 0.927
+arm: 0.925
+mistranslation: 0.925
+socket: 0.923
+TCG: 0.922
+performance: 0.920
+files: 0.918
+vnc: 0.917
+peripherals: 0.915
+assembly: 0.914
+debug: 0.909
+boot: 0.905
+device: 0.904
+kernel: 0.903
+x86: 0.892
+ppc: 0.884
+i386: 0.710
+
+util/zbin.c:390:37: error: absolute value function 'abs' given an argument of type 'long int' but has parameter of type 'int' which may cause truncation of value [-Werror=absolute-value]
+
+For qemu-3.0.0 I see with GCC 9 compiler a new warning:
+
+$ make -j1 -C roms pxerom VERBOSE=1 V=1
+make: Entering directory '/home/abuild/rpmbuild/BUILD/qemu-3.0.0/roms'
+make -C ipxe/src CONFIG=qemu \
+	PACKAGING_TIMESTAMP=1534273834 \
+	CROSS_COMPILE= \
+	bin/8086100e.rom bin/808610d3.rom bin/80861209.rom bin/10500940.rom bin/10222000.rom bin/10ec8139.rom bin/1af41000.rom bin/15ad07b0.rom
+make[1]: Entering directory '/home/abuild/rpmbuild/BUILD/qemu-3.0.0/roms/ipxe/src'
+gcc  -Wall -W -Wformat-nonliteral -O2 -g -Werror util/zbin.c -llzma -o util/zbin
+util/zbin.c: In function 'process_zinfo_add':
+util/zbin.c:390:37: error: absolute value function 'abs' given an argument of type 'long int' but has parameter of type 'int' which may cause truncation of value [-Werror=absolute-value]
+  390 |      ( ( addend < 0 ) ? "-" : "" ), abs ( addend ), size,
+      |                                     ^~~
+util/zbin.c:398:37: error: absolute value function 'abs' given an argument of type 'long int' but has parameter of type 'int' which may cause truncation of value [-Werror=absolute-value]
+  398 |      ( ( addend < 0 ) ? "-" : "" ), abs ( addend ), size,
+      |                                     ^~~
+util/zbin.c:419:37: error: absolute value function 'abs' given an argument of type 'long int' but has parameter of type 'int' which may cause truncation of value [-Werror=absolute-value]
+  419 |      ( ( addend < 0 ) ? "-" : "" ), abs ( addend ),
+      |                                     ^~~
+cc1: all warnings being treated as errors
+make[1]: *** [Makefile.housekeeping:1368: util/zbin] Error 1
+
+This bug would be better reported to the upstream for the ipxe rom -- we just ship the source code because we ship the binary.
+
+
+Thank you, I contacted IPXE folks.
+
+Should be fixed here:
+http://git.ipxe.org/ipxe.git/commitdiff/956f6a722
+
+Btw. how do you pull changes from ipxe into qemu repo?
+
+On Mon, 21 Jan 2019 at 12:47, Martin Liska <email address hidden> wrote:
+>
+> Should be fixed here:
+> http://git.ipxe.org/ipxe.git/commitdiff/956f6a722
+>
+> Btw. how do you pull changes from ipxe into qemu repo?
+
+The next step is a QEMU patch which updates our git submodule
+to point at a version of IPXE with the fix and also has the
+pre-built binaries corresponding to that new version.
+
+Gerd, it looks like you did the last IPXE update -- could
+we do another one with this fix in it?
+
+thanks
+-- PMM
+
+
+On Mon, Jan 21, 2019 at 12:57:16PM +0000, Peter Maydell wrote:
+> On Mon, 21 Jan 2019 at 12:47, Martin Liska <email address hidden> wrote:
+> >
+> > Should be fixed here:
+> > http://git.ipxe.org/ipxe.git/commitdiff/956f6a722
+> >
+> > Btw. how do you pull changes from ipxe into qemu repo?
+> 
+> The next step is a QEMU patch which updates our git submodule
+> to point at a version of IPXE with the fix and also has the
+> pre-built binaries corresponding to that new version.
+> 
+> Gerd, it looks like you did the last IPXE update -- could
+> we do another one with this fix in it?
+
+Yes, makes sense to do that, last ipxe rebase was more than a year ago.
+
+cheers,
+  Gerd
+
+
+
+In QEMU commit fd813c72392a712022d5d75a9 we upgraded our ipxe, so this bug should now be fixed in git master and in the upcoming 4.0 release.
+
+