search for: macroized

Well, differences of opinion is what makes horse races. Reed On 06/04/2012 04:57 PM, Daniel Berlin wrote: > On Mon, Jun 4, 2012 at 7:53 PM, reed kotler<rkotler at mips.com> wrote: >> On 06/04/2012 03:25 PM, Daniel Berlin wrote: >>> I'm pretty sure neither llvm nor clang have any technical debt at all. >>> >>> On Mon, Jun 4, 2012 at 5:18 PM, reed

On Mon, Jun 4, 2012 at 7:53 PM, reed kotler <rkotler at mips.com> wrote: > On 06/04/2012 03:25 PM, Daniel Berlin wrote: >> >> I'm pretty sure neither llvm nor clang have any technical debt at all. >> >> On Mon, Jun 4, 2012 at 5:18 PM, reed kotler<rkotler at mips.com>  wrote: >>> >>> something to think about as llvm and clang grows.

On 06/04/2012 03:25 PM, Daniel Berlin wrote: > I'm pretty sure neither llvm nor clang have any technical debt at all. > > On Mon, Jun 4, 2012 at 5:18 PM, reed kotler<rkotler at mips.com> wrote: >> something to think about as llvm and clang grows. >> >> http://en.wikipedia.org/wiki/Technical_debt >> _______________________________________________ >>

Can we get back to the substantive discussion about your ideas for lessening the technical debt? On Mon, Jun 4, 2012 at 8:05 PM, reed kotler <rkotler at mips.com> wrote: > Well, differences of opinion is what makes horse races. > > Reed > > > On 06/04/2012 04:57 PM, Daniel Berlin wrote: >> >> On Mon, Jun 4, 2012 at 7:53 PM, reed kotler<rkotler at

...nterrupt mask) is required to be compatible with this requirement. This makes maintenance of the code in entry.S much more easy, and eliminates some very tricky code changes that would otherwise be required. CLI/STI/IRET may all assume a 32 bit flat stack for writing to data areas. STI_SYSEXIT is macroized together because it implicitly relies on the interrupt holdoff property of STI. The IRET from a 16-bit stack must be marked explicitly as IRET16, because at this point there are no flat data segments loaded. It is expected that the IRET translation for a hypervisor will need to access a virtual i...

...nterrupt mask) is required to be compatible with this requirement. This makes maintenance of the code in entry.S much more easy, and eliminates some very tricky code changes that would otherwise be required. CLI/STI/IRET may all assume a 32 bit flat stack for writing to data areas. STI_SYSEXIT is macroized together because it implicitly relies on the interrupt holdoff property of STI. The IRET from a 16-bit stack must be marked explicitly as IRET16, because at this point there are no flat data segments loaded. It is expected that the IRET translation for a hypervisor will need to access a virtual i...

On 06/04/2012 05:17 PM, Daniel Berlin wrote: > Can we get back to the substantive discussion about your ideas for > lessening the technical debt? The lessening requires enlisting people that are willing to do this as opposed to doing fun science like cool optimization. I,for example, find the documentaiton, cleanup and refactoring to be interesting so I don't feel cheated to work on

FWIW, I'm putting together (hopefully to be done by the end of this weekend) a substantial refactoring of the TableGen backend API along with shiny new documentation (reStructuredText with sphinx) of all of TableGen, including documentation about how to write backends and---depending on how adventurous I get---a more detailed coverage of the syntax. Also, Reed, in your TableGen talk, IIRC,

...ert an arbitrary mapped virtual address to a physical frame number. Adapted from code by Chris Wright, added support for large pages. Because of the placement, this needs to be a macro; pgd_offset_k requires init_mm to be defined, and getting the include files right seemed more complicated than a macroized implementation. A more proper location would be in asm/page.h, but that gets confounded by the use of pgd_offset_k() again. Signed-off-by: Zachary Amsden <zach@vmware.com> Index: linux-2.6.13/include/asm-i386/pgtable.h =================================================================== ---...

...ert an arbitrary mapped virtual address to a physical frame number. Adapted from code by Chris Wright, added support for large pages. Because of the placement, this needs to be a macro; pgd_offset_k requires init_mm to be defined, and getting the include files right seemed more complicated than a macroized implementation. A more proper location would be in asm/page.h, but that gets confounded by the use of pgd_offset_k() again. Signed-off-by: Zachary Amsden <zach@vmware.com> Index: linux-2.6.13/include/asm-i386/pgtable.h =================================================================== ---...

...unately since a git source checkout compiles with -Werror I've been having problems because of all of the unused parameter warnings. I've developed some patches that eliminate the problem by marking all of the unused parameters with __attribute__ ((unused)). The attribute markings are all macroized so if you aren't using GCC to compile the macro is just a no-op. There's also a warning of a deprecated function generated by the ffmpeg, I've included a patch for that too. Note that these patches compile, but haven't been tested. swfdec_marshal.c generates unused parameter warni...

Hi Sean, Glad to hear there is clean up of tablegen going on. Just for the record, I don't know what you are referring to regarding some comment of mine at my talk about 10K LOC. I don't know how big tablegen is itself nor how much code has been written in it so I would not have ventured such a guess. The idea of totally replacing the tablegen language came up at the talk during the

I definitely trust what you say now with time to think at your keyboard over what you said on the spot in a live presentation. The comment that I was referring to was: 36:44 of http://llvm.org/devmtg/2012-04-12/videos/Reed_Kotler-mobile.mov "there's not really more than a couple thousand lines of .td ... I mean there's not tons of this code so if we had to use a different one I

A new pixman release 0.11.2 is now available. This is a development version leading up to a stable 0.12.0 release. Notable new features in this release include: - SSE2 implementations of many compositing ops (Andr? Tupinamb?) - VMX implementations for many compositing ops (Luca Barbato) - Opacity based optimization for operators (Antoine

A big pile of updates in this one. Highlights include: - glamor is updated to use OpenGL core profiles if available, which should improve memory usage and performance on modern hardware, and got some other performance improvements for rpi and other GLES platforms - DRI2, DRI3, and Present all received correctness fixes for hangs, crashes, and other weirdness - Xwayland server has been updated

I think we're ready for RC1 at this point, but wanted to give people a chance to scream about "just one more API change" until tomorrow. Let me know if there's something I'm missing; if I don't hear anything, I'll be tagging RC1 in the morning. Aaron Plattner (1): xace: Fix XaceCensorImage to actually censor the right part of the image Adam Jackson (92):