similar to: [LLVMdev] function inlining threshold ?

On Mon, Jul 04, 2005 at 03:32:39PM -0500, Long Fei wrote: > I am using llvm for source-to-source inlining. So I did: > > % llvm-gcc file_a.c file_b.c ... file_n.c -o file > % opt -inline -inline-threshold=1000 < file.bc | llc -march=c > outfile.c > > Can anyone tell me how llvm determines if a function should be > inlined, and what roll does

Hello, I trying TSAN on Darwin on LLVM itself (sanitizing multi-threaded ThinLTO link). However I see two main issues on my debug build: 1) Statistics: the pre/post increment is not safe, it seems to be acknowledge in the code itself: // FIXME: This function and all those that follow carefully use an // atomic operation to update the value safely in the presence of // concurrent

This didn't work as I tried with 197.parser. it works without "-Wl,-disable-opt" switch though. [197.parser]$ llvm-gcc analyze-linkage.c and.c build-disjuncts.c extract-links.c fast-match.c idiom.c main.c massage.c parse.c post-process.c print.c prune.c read-dict.c utilities.c xalloc.c word-file.c strncasecmp.c -Wa,-disable-opt -Wl,-disable-opt -lm -o llvm_parser [197.parser]$

On Mon, 11 Jul 2005, Long Fei wrote: > > This didn't work as I tried with 197.parser. it works without > "-Wl,-disable-opt" switch though. > > [197.parser]$ llvm-gcc analyze-linkage.c and.c build-disjuncts.c > extract-links.c fast-match.c idiom.c main.c massage.c parse.c post-process.c > print.c prune.c read-dict.c utilities.c xalloc.c word-file.c

Hi *, it seems rsync with --fileflags isn't able to work on (already) hardlinked and immutable ("schg") files on FreeBSD. The following scripts will create a simple example for this behaviour: -------------------------------------------------------------- #! /bin/sh # # set -x DIR="/var/tmp/rsync_$(date +%s)/" mkdir "${DIR}/" # Preparing dir_A mkdir

I am investigating some inlining issue, so I did llvm-gcc aaa.c bbb.c ... nnn.c -o output opt -inline -inline-threshold=xxx < output.bc | llc -march=c > output_inline.c 1) I noticed that even if I set xxx to 0 or even a very small negative number, many functions are eliminated. I am wondering if these functions are inlined by the frontend, or identified as deadcode. For instance,

Long Fei wrote: > > I am investigating some inlining issue, so I did > > llvm-gcc aaa.c bbb.c ... nnn.c -o output > opt -inline -inline-threshold=xxx < output.bc | llc -march=c > > output_inline.c I am unsure of whether the LLVM GCC frontend does any inlining. However, I do know that your methods above run the LLVM inlining pass, albeit indirectly. If you use

Dear all is it possible to rsync in a master-slave scenario saving to slave gzip content? i'm not talking about compression during transfer, i'm talking about -saving- the destination in a compressed format. Example: FROM: -folder_A --file_A --file_B TO: -folder_A --file_A.gz --file_B.gz I know that this won't be a "real" rsync between two folder, but it will be an

Hello All - I'm looking for a switch(es) to ensure rsync indexes the pull of files from remotedir, LIFO (last in first out) and then by .ext type where file basenames are similar. example) file_a.xml ctime of 09:01:07 file_a.jar ctime of 09:01:02 get transfered before file_b.xml ctime of 09:01:45 file_b.jar ctime of 09:01:45 and the .xml of basename $file would always be transferred

Hi, I need help :o( I want that my function saves result files in a for()-loop while it runs automatically. the filenames must be saved like: file_1 -> for 1. result file_2 -> for 2. result file_3 -> for 3. result and . . . file_n -> for n. result the file names are the same identified by _1, _2

Andy and I briefly discussed this the other day, we have not yet got chance to list a detailed pass order for the pre- and post- IPO scalar optimizations. This is wish-list in our mind: pre-IPO: based on the ordering he propose, get rid of the inlining (or just inline tiny func), get rid of all loop xforms... post-IPO: get rid of inlining, or maybe we still need it, only

I have a volume at 192.168.0.2 on my local network. I'd like to rsync the entire volume to a backup volume, skip all files already present on the backup volume (same name and the same or earlier timestamp), and include all flags, dates, etc. as applicable to Mac OS X 10.6.x. Can someone tell me the proper command to do this.

An HTML attachment was scrubbed... URL: <http://lists.samba.org/pipermail/rsync/attachments/20091030/c43a29f5/attachment.html>

https://bugzilla.samba.org/show_bug.cgi?id=2554 ------- Additional Comments From wayned@samba.org 2005-03-29 11:50 ------- (In reply to comment #0) >> If a file will be deleted the line starts with "Delete", but when a > file will be copied or overwritten, nothing is displayed. I assume you mean, "no prefix is displayed" since there is something that is

Thank you Easwaran for landing a great diff that enables the use of block frequency information in inliner. As of now, it seems that inliner can exploit the information only when profile data is available. However, instructions such as __builtin_expect may set branch weight metadata as well, which is useful for inliner. I think this problem can be addressed by letting llvm::getInlineCost function

LLVM (2.7 release version) provides 2 implementations for inlining function callsites: - InlineSimple.cpp (-inline): inline simple callsites according to its cost analysis - InlineAlways.cpp (-always-inline): inline all callsites that are marked with "always_inline" attribute. They are both subclasses of Inline.cpp that assumes the function's definition (body) is

(I've already posted this message month ago, but nobody answered, may be it was not delivered to newsgroup?) Hello, All. I wrote an encoder-decoder based on example from OGG-Vorbis SDK. This encoder can encode a large amount of small WAV-files with equal parameters into one sound archive. Then I decode this sound archive back into large amount of small WAV-files. Theese files are

I am interested in analyzing the bytecode code produced for C files. By default, inlining of user and library functions (libc) is done. If I turn off inlining (-disable-inlining in gccas and gccld) then no inlining is done. I want to be able to inline user code but disallow library code to be inlined. In trying to understand the InlineSimple.cpp code, I see that library functions are

On Mar 15, 2006, at 11:15 AM, Chris Lattner wrote: > On Wed, 15 Mar 2006, Vikram S. Adve wrote: >>> Why can't the compiler pass just call InlineFunction(CallSite) on >>> the callsite it wants inlined? The only way that can fail is if >>> LLVM cannot ever inline the call (e.g. it uses varargs). > >> In some cases, that would be fine. But in other

On Thu, 21 Feb 2008, Dale Johannesen wrote: > The defined gcc interface for this is -fno-builtin. It seems not be > to be working in llvm-gcc, however. Please file a reduced testcase in bugzilla, -Chris > >> I am interested in analyzing the bytecode code produced for C files. >> By default, inlining of user and library functions (libc) is done. If >> I turn off