similar to: issue with sample in R 3.6.0.

Below is a patch against the current OpenBSD OpenSSH CVS to workaround a behavior I have observed when converting from SSH 1.2.27 to OpenSSH while using the same old RSA1 host key for protocol 1. In several cases I saw that old SSH sshd reported a host key size of 1024 bits when OpenSSH saw it as 1023 bits. Without the patch, when OpenSSH's ssh client connects to an old SSH sshd it warns

I was experiencing a strange pattern of slowdowns when using sample.int(), where sampling from a one population would sometimes take 1000x longer than taking the same number of samples from a slightly larger population. For my application, this resulted in a runtime of several hours rather than a few seconds. Looking into it, I saw that sample.int() is hardcoded to switch algorithms when the

I'm trying to replace some sshv1 clients and servers in a modular way, and the "Server Lies" warning (when the server says the key has one more bit than it really has) is causing heartache. Per the FAQ, this is relatively benign. Here's a patch that allows an admin or user to disable the warning. - Morty diff -Nur openssh-3.7.1p2/readconf.c

Import "exploiting the new interface in vnc.c" from qemu mainstream. git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6337 c046a42c-6fe2-441c-8c8c-71466251a162 Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> --- diff -r 07a126ac0425 console.c --- a/console.c Mon Feb 16 11:39:06 2009 +0000 +++ b/console.c Mon Feb 16 12:07:19 2009 +0000 @@ -1310,6 +1310,9

On Jan 14, 2010, at 10:13 PM, David Conrad wrote: > Hi, > > On ARMv6T2 this turns cttz into rbit, clz instead of the 4 > instruction sequence it is now. > > I'm not sure if adding RBIT to ARMISD and doing this optimization in > the legalize pass is the best option, but the only better way I > could think of doing it was to add a bitreverse intrinsic to llvm

On Jan 15, 2010, at 10:03 AM, Chris Lattner wrote: > > When/if another target wants this, we could add a ISD::RBIT operation, > it doesn't need to be added at the llvm ir level, Blackfin can add with backwards carry, essentially doing (rbit (add (rbit a), (rbit b))) This is used for FFTs. I wasn't hoping to be able to pattern-match something so complicated.

Could everyone please test the latest snapshots as we will be making a new release soon. If you have any patches you would like us to consider, please resend them to the list ASAP. -d -- | Damien Miller <djm at mindrot.org> \ ``E-mail attachments are the poor man's | http://www.mindrot.org / distributed filesystem'' - Dan Geer

Hi, On ARMv6T2 this turns cttz into rbit, clz instead of the 4 instruction sequence it is now. I'm not sure if adding RBIT to ARMISD and doing this optimization in the legalize pass is the best option, but the only better way I could think of doing it was to add a bitreverse intrinsic to llvm ir, which itself might not be the best option since bitreverse probably isn't too common. Other

On Fri, Jan 15, 2010 at 6:03 PM, Chris Lattner <clattner at apple.com> wrote: > > On Jan 14, 2010, at 10:13 PM, David Conrad wrote: > >> Hi, >> >> On ARMv6T2 this turns cttz into rbit, clz instead of the 4 >> instruction sequence it is now. >> >> I'm not sure if adding RBIT to ARMISD and doing this optimization in >> the legalize pass is

make -r -C /mlsyslinux -f /mlsyslinux/Makefile SRC="/mlsyslinux" \ OBJ=/mlsyslinux objdir=/mlsyslinux bios make[1]: Entering directory `/mlsyslinux' make -r -C /mlsyslinux/bios -f /mlsyslinux/Makefile SRC="/mlsyslinux" \ objdir=/mlsyslinux/bios OBJ=/mlsyslinux/bios HAVE_FIRMWARE=1 \ ARCH=i386 LDLINUX=ldlinux.c32 all make[2]:

On Jan 15, 2010, at 11:37 AM, Richard Osborne wrote: > > On 15 Jan 2010, at 18:03, Chris Lattner wrote: > >> On Jan 14, 2010, at 10:13 PM, David Conrad wrote: >> >>> Other targets that I know of that could potentially benefit from >>> this optimization being global (that have a clz and bitreverse >>> instruction but not ctz) are AVR32 and C64x,

On Jan 15, 2010, at 2:52 PM, Jim Grosbach wrote: > > On Jan 15, 2010, at 11:37 AM, Richard Osborne wrote: > >> >> On 15 Jan 2010, at 18:03, Chris Lattner wrote: >> >>> On Jan 14, 2010, at 10:13 PM, David Conrad wrote: >>> >>>> Other targets that I know of that could potentially benefit from >>>> this optimization being

On 15 Jan 2010, at 18:03, Chris Lattner wrote: > On Jan 14, 2010, at 10:13 PM, David Conrad wrote: > >> Other targets that I know of that could potentially benefit from >> this optimization being global (that have a clz and bitreverse >> instruction but not ctz) are AVR32 and C64x, neither of which llvm >> has backends for yet. > > When/if another

Greetings all, First of all, thanks to all of you for creating such a useful, powerful program. I regularly work with very large datasets (several GB) in R in 64-bit Fedora 8 (details below). I'm lucky to have 16GB RAM available. However if I am not careful and load too much into R's memory, I can crash the whole system. There does not seem to be a check in place that will stop R from

Yesterday I posted the following question (my apologies for not putting a subject line): =================question====================== Hello -- I would like to know of a more efficient way of writing the following piece of code. Thanks. options(stringsAsFactors=FALSE) orig <- c(rep('11111111',100000),rep('22222222',200000),rep('33333333'

On May 7, 2014 11:54 PM, "??" <muliu92 at 163.com> wrote: > make[4]: Entering directory `/mlsyslinux/bios/com32/libupload' > gcc -Wp,-MT,cpio.o,-MD,./.cpio.o.d -std=gnu99 -m32 -march=i386 -mpreferred-stack-boundary=2 -fno-stack-protector -fwrapv -freg-struct-return -Os -fPIC -fno-exceptions -fno-asynchronous-unwind-tables -fno-strict-aliasing -falign-functions=0

On Feb 10, 2005, at 7:38 PM, George W. Gilchrist wrote: > Today I was running a graduate level stats lab using R and we > encountered a > major problem while using the current build of the Cocoa GUI: > >> From the GUI: >> system.time(pbinom(80, 1e5, 806/1e6)) > [1] 14.37 4.94 30.29 0.00 0.00 >> > >> From the command line on the same machine: >>

I thought I was clever. I produce both ps and pdf files from LaTeX (MiKTeX), so sometimes I need to get .eps graphs from R, sometmes .pdf. To rationalize this I wrote the following function to let me easily choos what I want wihtou changing the code in my R-programs too much (here embedded in a small test program): plt <- function( file, type = "pdf", dir =

On Fri, Oct 14, 2016 at 6:10 PM, Hal Finkel <hfinkel at anl.gov> wrote: >> polybench/linear-algebra/kernels/symm, FP_ABSTOLERANCE=1e1 >> polybench/linear-algebra/solvers/gramschmidt, FP_ABSTOLERANCE=1e0 >> What should be a good relative tolerance to set for these two tests? > > What's the minimum relative tolerance that you need for them to pass? Setting

> OK, attempting one more convergence on all this, here's > what I did: > > 1. Reverted back to rev 1.5 of src/libFLAC/ia32/Makefile.am > 2. Applied Matt's last cleanup patch of 8 files. I did not > apply the patch to src/test_unit/main.c since it looks wrong. > main.c is supposed to include the local bitbuffer.h, not > src/libFLAC/private/bitbuffer.h; I think the