Older blog entries for acme (starting at number 147)

libdwarves (mostly) explained

In response to a comment I’ve added comments to the ctracer tool source code in an attempt to help people interested in helping improving these tools, check it out!

Syndicated 2007-01-20 15:04:30 from Arnaldo's Ramblings

How many functions receive as a parameter a pointer to a struct?

Implemented a new command line option in pahole: –nr_members, look here for the result using a kernel built for qemu.

Syndicated 2007-01-13 16:04:08 from Arnaldo's Ramblings

Support for more C++ tags

Added support for DW_TAG_reference and DW_AT_specification, still have to work on DW_TAG_namespace and rework the list of tags to make classes within classes to properly be shown in the tools, nowadays its just a flat namespace, which has not been a problem for the tools written so far, but for pdwtags, the dwarf I’m working on now, that will do a complete dump of all tags it is a must.

Syndicated 2007-01-13 13:51:47 from Arnaldo's Ramblings

connection_sock work going on nicely, today sent patches for lots of network families, tomorrow, finally, will try to test the ax25/netrom/rose big patch with bpqether and the ax25 utilities, if all goes well, send the patches to Dave and turn the chainsaw to net/bluetooth/, that will be interesting with all the internal proto infrastructure it has, lets see if we can take advantage of its approach in other protos or if getting what the net/ipv4/ has for a similar purpose is better.

Added my TODO list to this page.

WOW, it has been a long time, and since I'm lazy as hell to use my ghostprotocols.net to create some webpage, diary, etc I'll continue using this nice infrastructure provided by the fine folks at advogato.org :-)

Today I submitted a proposal for OLS 2005 "DCCP on Linux", for those, like a friend who said for some seconds he thought I was going to talk about KDE, no, its not DCOP, its a new internet transport protocol, interested people can take a look at the RFCs drafts at http://www.icir.org/kohler/dcp/.

The proposal is just an attempt to get me from procastrinating and get back to work on this thing, that fits very nicely in my work on TCPfying the poor cousins, i.e. factoring out code from the rich cousins (TCP, yeah) and into the linux kernel core networking code.

OK, lets see how long it takes for me to write another diary entry 8-)

net-exp quiet for a while, but getting good reviews from the friends that have looked at it, if it passes Al Viro's eyes then I'll just chunk it into sensible changesets and start feeding davem, won't be fun tho, over 10K lines of patches for net/ in 2.6.early...
Just got very happy that a serious stab at a 2.6-tiny kernel tree is being maintained by Matt Mackall, have already come up with several ideas from my long nights in net/ land, booting up the kernel with mem=4m on the command line again is sooooooo nice... :) lilliput net here we go :P

Have been working on networking primitives to generalise the wait for data open coded functions in most protocols, initially with sk_wait_event, that allows protocols to just pass the sock, a pointer to a timeout and the condition that has to be checked before releasing the socket lock, going to sleep and then grabbing the lock again, that way we make sure that we process the backlog before testing the condition, i.e. when doing the release_sock if there are packets in the receive_queue they will be flushed, which possibly will make the condition to be true. This is just the start, as there is room for a generic sk_wait_packet and similar functions for waiting for connections, disconnection, etc.

Finished splitting llc into llc_core and llc2, now it is easier to select IPX, Token Ring, Appletalk, more or less like it was in 2.4, but I stil think that this LLC stack needs even more work to be a first class citizen in linux networks stack land, so I'll continue working on it to use dst.h, netfilter (yes, if decnet has why not llc? :-)), etc.

Have been working on a tool that uses sparse, to be part of the kernel build process moving strings in __init/__exit functions to __initdata/__exitdata, saving more bytes when the kernel discards the .init.data ELF section (both when building stuff as modules (in 2.6) or statically) and the .exit.data in static builds.

Also trying to get the Conectiva Linux Alpha port to what is in our current development distro, now kinda stuck in what seems a bug in gcc.

138 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!