Author: emaste
Date: Thu Jul 24 08:28:16 2014
New Revision: 213861
URL: http://llvm.org/viewvc/llvm-project?rev=213861view=rev
Log:
Fix endian test for big-endian hosts
The uint16_t cast truncated the magic value to 0x0304, making the
first byte 0 (eByteOrderInvalid) on big endian hosts.
Author: jingham
Date: Thu Jul 24 11:56:19 2014
New Revision: 213875
URL: http://llvm.org/viewvc/llvm-project?rev=213875view=rev
Log:
We were turning off all these tests on OSX and FreeBSD because of a known (and
fairly unimportant) bug.
Keep a test for that bug, but let the useful parts of the
Remove i486 and i486sx, only support i386 and i686
http://reviews.llvm.org/D4658
Files:
include/lldb/Core/ArchSpec.h
source/Core/ArchSpec.cpp
source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.cpp
source/Plugins/ObjectFile/PECOFF/ObjectFilePECOFF.h
Always happy to help get better code coverage in place. These tests are
particularly nice since they run in everyone's test environment regardless
of host. Those are the nicest tests :-)
On Wed, Jul 23, 2014 at 10:02 PM, Matthew Gardiner m...@csr.com wrote:
Todd Fiala wrote:
And meant to
My pleasure :-)
On Wed, Jul 23, 2014 at 10:03 PM, Matthew Gardiner m...@csr.com wrote:
Todd Fiala wrote:
Done:
svn commit
Sendingdocs/lldb-gdb-remote.txt
Transmitting file data .
Committed revision 213756.
Thanks again!
Matt
Member of the CSR plc group of companies. CSR
Fixed a few bugs
http://reviews.llvm.org/D4641
Files:
include/lldb/Core/ConnectionFileDescriptor.h
include/lldb/Host/Editline.h
include/lldb/Host/File.h
include/lldb/Host/IoObject.h
include/lldb/Host/Socket.h
source/Core/ConnectionFileDescriptor.cpp
Rebased against tip and fixed some comments.
http://reviews.llvm.org/D4641
Files:
include/lldb/Core/ConnectionFileDescriptor.h
include/lldb/Host/Editline.h
include/lldb/Host/File.h
include/lldb/Host/IoObject.h
include/lldb/Host/Socket.h
source/Core/ConnectionFileDescriptor.cpp
IMO we should change the implementation and the documentation for qHostInfo
triple to not specify/use hex encoded strings. debugserver doesn't send
triple. I don't know about llgs or how much work it would be to change the
kalimba gdbserver stub.
On Jul 24, 2014, at 1:39 PM, Jason Molenda
Author: tfiala
Date: Thu Jul 24 18:22:58 2014
New Revision: 213904
URL: http://llvm.org/viewvc/llvm-project?rev=213904view=rev
Log:
Fix configure/make builds on MacOSX.
Change by Keno Fischer.
Modified:
lldb/trunk/source/Host/common/Host.cpp
Modified: lldb/trunk/source/Host/common/Host.cpp
Yep.
On Jul 24, 2014, at 5:28 PM, Todd Fiala tfi...@google.com wrote:
Hey Jason - if Tong verifies this is all still good, is that essentially a
LGTM from you? (i.e. is that something I can get checked in for him?)
On Thu, Jul 24, 2014 at 5:16 PM, Jason Molenda jmole...@apple.com
Hey Jason - if Tong verifies this is all still good, is that essentially a
LGTM from you? (i.e. is that something I can get checked in for him?)
On Thu, Jul 24, 2014 at 5:16 PM, Jason Molenda jmole...@apple.com wrote:
Good catch Tong. I don't think clang generates this instruction pattern
Hey Tong - I'll add that and do all the test runs. I'll get it checked in
unless something breaks.
-Todd
On Thu, Jul 24, 2014 at 5:32 PM, Tong Shen endlessr...@google.com wrote:
Hi Jason,
I don't have any strong preference here, so let's do it the way you see
fit :-)
Your patch works
Thanks Todd!
On Thu, Jul 24, 2014 at 5:33 PM, Todd Fiala tfi...@google.com wrote:
Hey Tong - I'll add that and do all the test runs. I'll get it checked in
unless something breaks.
-Todd
On Thu, Jul 24, 2014 at 5:32 PM, Tong Shen endlessr...@google.com wrote:
Hi Jason,
I don't have
:-)
On Thu, Jul 24, 2014 at 5:34 PM, Tong Shen endlessr...@google.com wrote:
Thanks Todd!
On Thu, Jul 24, 2014 at 5:33 PM, Todd Fiala tfi...@google.com wrote:
Hey Tong - I'll add that and do all the test runs. I'll get it checked
in unless something breaks.
-Todd
On Thu, Jul 24,
Author: tfiala
Date: Thu Jul 24 20:15:34 2014
New Revision: 213914
URL: http://llvm.org/viewvc/llvm-project?rev=213914view=rev
Log:
Fix an x86 assembler stack unwind calculation for non-volatile registers.
This change has the practical effect of fixing some backtrace
scenarios that would fail
15 matches
Mail list logo