similar to: Resurrecting the old "klibc-cvs" mailing list

This fixes a long standing issue where it was not possible to do "make usr/klibc/arch/x86_64/longjmp.o" in the kernel. The principle is that all .o files to be part of klibc are listed with klib-y. For each directory a klib.list file is made that specify all .o file and the final AR then adds all .o files to create libc.a. This patch introduce the infrastructure and converts x86_64 to

Hi hpa & others. Following is a list of issues that I hope to be addressed soon so we are in even better shape for -mm inclusion. 1) rebuild initramfs when content changes. > It is a simple matter of copying in usr/Makefile from the latest > -linus kernel and replace the 10 first lines with the content from > klibc Kbuild file. 2) havesyscalls.h is not deleted after make

sparc still needs gcc-3.3 for a working run-init. belows makes use of the exported CC var. Signed-off-by: maximilian attems <maks@sternwelten.at> diff --git a/scripts/Kbuild.klibc b/scripts/Kbuild.klibc index 66e16e7..70e4071 100644 --- a/scripts/Kbuild.klibc +++ b/scripts/Kbuild.klibc @@ -72,7 +72,7 @@ KLIBCROSS := $(CROSS_COMPILE) # binutils KLIBCLD := $(KLIBCROSS)ld

Hi all. Proposal for directory structure in the kernel: usr/include <= _all_ include files usr/include/{net,sys,..} usr/include/arch/$(ARCH) <= arch specific include files *1 usr/klibc <= klibc files usr/klibc/arch/$(ARCH) <= arch specific config and files except include files usr/kinit <= program integrated in the kernel usr/kinit/nfsmount <= used by kinit,

Hi hpa and others. With my ongoing work to prepare for potential integration of klibc in the kernel I wonder if there wold be any interest in using part of the kbuild infrastructure to build the vanilla klibc? The rationale is obvious: - common codebase for Kbuild files, so one only has to fix things in one place - I force klibc developers to create Kbuild files usable in the kernel - Much

Are there any version compatibilities between linux and klibc? I tried to compile the latest klibc (1.5.18) with the latest linux (2.6.32.15) and there is miserable compile failures. It looks like klibc wants some headers to be in include/asm (linked in my case to include/asm-x86) but they are instead in include/asm-generic (it seems all headers but one are in there). I put together this script

After sending an email to klibc at zytor.com, I have to wait for 1~2 days to see the send message. Since other mailing lists respond within seconds, it seems klibc mailing list has a problem. -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://www.zytor.com/pipermail/klibc/attachments/20101025/00310c6b/attachment.html>

gen_initramfs_list.sh failed to generate a dependency list when it was supplied the -d option. Restructured gen_initramfs_list.sh to support dependency info with -d. This also fixes a bug where we did not properly process when more than one dir was specified. Finally update dependency info for the -d case (list_default_initramfs) In usr/Kbuild remove "" around CONFIG_RAMFS_SOURCE so the

Hi Peter & others. Here are three patches that does the following: #1 - Update kbuild part of klibc so make clean works Adds gzip including a sample kbuild file #2 - Factor out definition of usr/ to two variables #3 - Move kbuild files to reflect location in the kernel As requested in earlier mail I need a bit of guidiance of what you expect from the kernel integrated parts of klibc.

From : Erwan Velu <erwan.velu at free.fr> This patch - remove the pci detection code from ethersel - add a find_pci_device() function in the PCI com32 module - make ethersel using the pci com32 module instead of builtin pci detection code Testing and feedback are welcome. Signed-off-by:Erwan Velu <erwan.velu at free.fr> -------------- next part -------------- A non-text attachment

I have a small file of the type that S-PLUS produced and stored in .Data/ directories back in the days before we used R. I'm fairly sure it would become a dataframe if read into Ver 3.4 of that august predecessor to what we use now. (It was Solaris in case that makes any difference.) AFAIK, the format that S-PLUS uses changed after that, so a current version would find it unreadable. If

On Fri, Sep 04, 2015 at 09:15:07AM -0700, syslinux-bot for Erwan Velu wrote: > Commit-ID: 66105926f3d748de827eb863913171efb9624165 > Gitweb: http://www.syslinux.org/commit/66105926f3d748de827eb863913171efb9624165 > Author: Erwan Velu <erwanaliasr1 at gmail.com> > AuthorDate: Wed, 2 Sep 2015 18:26:37 +0200 > Committer: Erwan Velu <erwanaliasr1 at gmail.com> >

From : Erwan Velu <erwan.velu at free.fr> As discussed on IRC, I've been reworking my patch, I resend it. This patch - add a new pci_dev_info structure : It contains additional informations about the pci devices like the product/vendor name and the associated linux kernel module - add a get_name_from_pci_ids() function in pci/scan.c This function reads a pci.ids file from the

Hi all. I have pushed my earlier announced kbuild improvements patch to: bk://linux-sam.bkbits.net/2.5-klibc First cset is the earlier posted cset, whit a few small modifications. One important change is that the usr/Makefile now uses *.a files in initramfs-y assignment. I realised that the purpose was to express what kinit were using / referring. Second cset is a trivial cset that just move

It is supposed to match but sometimes it doesn't; I had the case where I contacted my local server on my public adress and responded from the loopback. So it was the same server from two different addresses which confused then the rest of the code. By adding this print I was able to get the situation immediately and fixed it. That's why I added that information on the debug print.

I'm still pondering with kbuild and klibc. Next in line was to get ipconfig support in the kernel (build wise). A little challenge that is bigger than anticipated was to create a shared executable. This required a far bigger rewrite of Kbuild.klibc than originally planned. The good part is that I now managed to treat linking of objects with single and multiple .o files almost the same.

From : Erwan Velu <erwan.velu at free.fr> This patch add the ability to read the long-mode flag from the processor. This helps you to know if the processor is x86_64 capable or not. So "dmi->processor.cpu_flags.lm " is a boolean that represent the "Long Mode" cpu flag. It also add two bugfix : - some cpu flags were shifted due to some "null" messages in

klibc integrated with the kernel can now be found at: http://linux-sam.bkbits.net Se the repository: klibc. For non-BK users the patches can be retreived using the following links: klibc on top of 2.6.8-rc2: http://linux-sam.bkbits.net:8080/klibc/gnupatch@41098c67CQHEC6lH2E3rfIXcnRW1RA kbuild adaptions to klibc: http://linux-sam.bkbits.net:8080/klibc/gnupatch@4109923a7O1XqfXw_V7JzAmet_AzbA

Hi there, while my memdisk setup works perfectly well an an Intel system it fails like this on a Via-C3 based Mini-ITX system (copied manualy, I hope I did not make a mistake): --cut-- Failed to Free base memory, sorry... MEMDISK 2.08 (Debian, 2003-12-22) Copyright 2001-2003 H. Peter Anvin e820: 0000000000000000 000000000009fc00 1 e820: 000000000009fc00 0000000000000400 1 e820: 00000000000f0000

Got to minimize (barely) my filesystem and kernel and fit it into my usb key (256mb) My kernel was compiles with scsi and usb support built-in. My problem now is i can't mount my root filesystem when i tried to boot. kernel somehow takes time before being able to recognize my usb as a valid scsi device thus displaying VFS: Cannot open root device 08:01 this is my lilo.conf boot=/dev/sda