similar to: Xen and TLS, -mno-tls-direct-seg-refs needed?

This is a cross post to <xen-devel@lists.sourceforge.net> and <user- mode-linux-user@lists.sourceforge.net> Dear lists, Anybody successfully running cyrus-imapd in FC2/3+Xen/UML? When i tried to run cyrus-imapd in a virtual FC3 (Xen or UML) i got lot''s of DBERROR db4: Berkeley DB library configured to support only DB_PRIVATE environments DBERROR:

Hi, until now I have run Xen with /lib/tls out of the way for both dom0 and domU''s, but now I''m trying to install OpenLDAP in a domU, which requires TLS as default. So now I have at least three options ahead of me: 1. mv /lib/tls.disabled back to /lib/tls and accept the performance penalty (is there any documentation on how heavy this penalty is?) 2. Compile OpenLDAP

Hello, We [the Debian Xen package team] are currently working on Xen packages, and are planning to include them into Debian. A current issue in Xen, is the libc problem. From the Xen wiki [1]: "Xen uses segmentation to provide protection of the memory used for the hypervisor. This results in some performance issues since wrap-around segments as used by glibc need expensive extra handling.

I would like to rebuild glibc specifically with the -mno-tls-direct-seg-refs to allow better performance with Xen. I'm using Centos 4.2 x86_64 on an Athlon 3200+. I'm using the 2.6.9-11.106.unsupported kernel for reiser support. My thought was to download and install the SRPM and then rpmbuild with no changes to make sure it works before adding the above flag. Unfortunately, the

[root@1950 SRPMS]# rpmbuild -bp glibc-2.3.4-2.9xen2.src.rpm ... + mkdir -p /var/tmp/glibc-2.3.4-root/usr/lib/nptl + cp -a libc.a nptl/libpthread.a nptl/libpthread_nonshared.a rt/librt.a /var/tmp/glibc-2.3.4-root/usr/lib/nptl/ + sed ''s| /lib/| /lib/tls/|'' /var/tmp/glibc-2.3.4-root/usr/lib/libc.so sed: can''t read /var/tmp/glibc-2.3.4-root/usr/lib/libc.so: No such file or

Hi all, I''ve performed a benchmark on my server to compare performance with and without Xen, and with and without domUs active. I used lmbench to measure basic hardware operations on math and memory, and system calls like mmap(), fork(), exec(). The results are at http://drainbamage.nl/2009/04/05/xen-benchmark/ . It seems that Xen adds significant overhead to process creation

Guys, I'm trying to compile the Xen 4.0.0 via debian/rules makefile but I got this: administrativo at sid01:~/xen/xen-4/xen-4.0.0$ make -f debian/rules build-tools ..... make -C lib all make[7]: Entering directory `/home/administrativo/xen/xen-4/xen-4.0.0/debian/build/build-tools/tools/blktap2/vhd/lib' make[7]: Nothing to be done for `all'. make[7]: Leaving directory

Hi, I''m still seeing a very strange issue here. First, let''s clarify that the issue has never occurred with the good old xen 3.x and the good old 2.6.18 kernel. So the issue is, that with xen 4.x (including 4.1.1) pretty much any kernel (including kernel from [1] and vanilla 3.0.0, didn''t test the 2.6.18), the machine freezes during a reboot. The machine won''t

Hi, until now, i have been using Xen 3.0.2. Now i upgraded to Xen 3.0.4. Can anybody confirm, that the ordner in the config files for the network interfaces has not changed? My config file used to read: vif = [ ''bridge=xenbr0'', ''bridge=xenbr1'' ] So inside the DomU, eth0 was attached to xenbr0 on the host and eth1 was attached to xenbr1. Now i upgraded to

Dear all, With or without /lib/tls librairies activated, A search in openldap is very slow on XEN 3.0.x . Cpus do nothing . 15000 entries are dumped in 2 minutes whereas same search without xen are dumped in 4 seconds (with far more activitie on CPUs) I have already see one or mail about this problem .  But I don''t see a solution , perhaps a threading problem between Berkeley DB and

I am trying to debug the Xen hypervisor from a second computer over the serial port, but nothing seems to work. Using mercurial, I got xen-3.2-testing.hg. I followed the steps in crashdb.txt in the docs/misc/ folder: set debug=y in Config.mk, crash_debug=y in xen/Rules.mk I also added -fno-omit-frame-pointer to these file as well. I compiled with no errors and booted with minicom connected to

Hi, below is the output of the kernel just before it goes down and crashes. I have to use "xm destroy" to kill it. I have tried the enable/disable the "Optimize for Size" setting in the kernel. I hopes that it''s some kind of compiler bug because of the "invalid opcode" thing. But actually I don''t a clue what this might be caused by. Do you

This time make world on Ubuntu 10.04 gives following error gcc -O2 -fomit-frame-pointer -m32 -march=i686 -fno-strict-aliasing -std=gnu99 -Wall -Wstrict-prototypes -Wno-unused-value -Wdeclaration-after-statement -D__XEN_TOOLS__ -MMD -MF .fsimage.so.d -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -mno-tls-direct-seg-refs -I../../../tools/libfsimage/common/

Everything on Centos 7 seems fine - except - my program. I am trying to compile and run on centos 7. Everything compiles/runs on C6, the executable from C6 runs on C7. when I compile and run on C7 - I get a SEG fault. when I debug: gdb program break main run I get a seg fault at or before main What might I be looking for in this case? Thanks, Jerry

Hey everyone, I''ve searched through all of the previous posts and tried everything mentioned but I am still getting these messages. Is there anything else I can do? My host machine is running CentOS 5 w/ Xen 3.3.1. My Xen VM is also running CentOS 5. I''ve tried doing: # echo ''hwcap 0 nosegneg'' > /etc/ld.so.conf.d/libc6-xen.conf && ldconfig

Hi , All I try to compile xen-unstable.hg(25249:a4e7fce6ee2b) from source repo on ubuntu11.10-amd64. It complains "/usr/include/features.h:323:26 fatal error: bits/predefs.h No such file or directory". I wonder it''s a thing related to 32bits and 64bits. How to solve this problem? Thank you. Here are my steps: [ Install prerequisites. ] sudo apt-get install git sudo apt-get

On Jul 1, 2011, at 2:43 PM, Alistair Lynn wrote: > Hi Andrew- > >> fatal error: error in backend: SSE2 register return with SSE2 disabled > > Is this for 32-bit or 64-bit x86? 64-bit x86. > If it's the latter, the ABI demands > that the return value in this case is in xmm0 - SSE is required. > Well -no-sse -mno-mmx works for EFI as it is pre-boot firmware and

gcc -O2 -fomit-frame-pointer -m32 -march=i686 -fno-strict-aliasing -std=gnu99 -Wall -Wstrict-prototypes -Wno-unused-value -Wdeclaration-after-statement -D__XEN_TOOLS__ -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -mno-tls-direct-seg-refs -Werror -fno-stack-protector -fno-builtin -msoft-float -I../../../tools/include -I. -c -o hvmloader.o hvmloader.c cc1: warnings being treated as errors

Hi Andrew- > Well -no-sse -mno-mmx works for EFI as it is pre-boot firmware and does not have any floating point C code. We use -no-sse and -mno-mmx code to prevent optimized code gen using these registers for optimizations. Whether it's optimised or not doesn't particularly matter, the x86_64 ABI says that floating-point return values go into SSE registers, so that is where LLVM is

Hi Andrew- > fatal error: error in backend: SSE2 register return with SSE2 disabled Is this for 32-bit or 64-bit x86? If it's the latter, the ABI demands that the return value in this case is in xmm0 - SSE is required. Alistair