Welcome to Catalin(ux) M. BOIE's page! Here you can find some work of mine on Linux kernel & other stuff. E-mail: catab # embedromix dot ro Phone: +40-745-048374 My Fedora repository: rpm -Uhv http://kernel.embedromix.ro/dinorepo-0.0.12-1.noarch.rpm My Oracle/RedHat/CentOS repository: rpm -Uhv http://kernel.embedromix.ro/dinorepo-el-0.0.13-1.noarch.rpm Supported architectures: x86_64, i386, armv7hl, aarch64. For commercial support and customizations, contact us at: Embedromix [Home]  [Networking]  [Crypto]  [Linux kernel patches]  [Userspace]  [Docs]  [CV/Resume]  [Links]  Name Date Status Description Links 0001-IPv6-NAT-Do-not-drop-DNATed-6to4-6rd-packets.patch Sent on 2013-09-14 to netdev list. Accepted in 3.12-rc1 on 2013-09-28 When a router is doing DNAT for 6to4/6rd packets, the latest anti-spoofing commit 218774dc ("ipv6: add anti-spoofing checks for 6to4 and 6rd") will drop them because the IPv6 address embedded does not match the IPv4 destination. This patch will allow them to pass by testing if we have an address that matches on 6to4/6rd interface. I have been hit by this problem using Fedora and IPV6TO4_IPV4ADDR. Also, log the dropped packets (with rate limit). Many thanks to Joe Perches and Hannes Frederic Sowa. Patch (local) git Gmane IPVS-Allow-boot-time-change-of-hash-size.patch Sent on 2008-11-25 to netdev and lvs-devel lists. Accepted in 2.6.34 on 2010-01-05 Allow changing at run-time the number of hash size for IPVS. I want to thank Wikimedia for the tests they made! Patch (local) git Added-support-for-a-no-name-4-ports-multiserial-card.patch 2007-12-16 Resent 2008-06-06 Accepted in 2.6.27 on 2008-07-25 Added support for a no-name 4 ports PCI serial card (10b5:9050, 10b5:1584). Patch was Acked by Alan Cox and Russell King. Patch (local) git e100_sleep_in_invalid_context.patch 2005-07-08 Not needed Tiberiu reported, on rlug, a BUG in e100. The error was: sleeping function called from invalid context at mm/slab.c:2088. The problem is this call path: e100_up - request_irq - ... - proc_create - kmalloc(..., GFP_KERNEL). This patch deffers up/down to keventd. A similar patch was already applied before I sent it upstream. Patch (local) e100_fix_io_ordering.patch 2005-07-08 Accepted on 2006-03-04 Checking e100.c with Documentation/io-ordering.txt, I saw a problem; code calling spin_unlock_irq[restore] _after_ e100_write_flush. This patch corrects this situation moving e100_write_flush before spin_unlock. Patch (local) git bridge_features.patch 2005-05-26 Accepted Select bridge features based on the slaves ones. Also, activate LLTX (Lockless TX), HIGHDMA, SG and TSO. Patch (local) NETDEV_FEAT_CHANGE.patch 2005-05-26 Accepted This patch adds notification for netdevice's features changes. Patch (local) sysfs_net_features_rw.patch 2005-05-21 Not accepted This patch permit features changing using sysfs: echo 0x1004 > /sys/class/net/_device_/features. Not accepted because it's better to use ethtool. Patch (local) dummy_features.patch 2005-05-21 Not accepted Improve dummy device performance. Not accepted because David thinks that "dummy device performance cannot be claimed to be important at all". Patch (local) sch_api_rtnl_unlock_lock-try3.patch 2005-04-01 Accepted This is a rework of Dave's patch to apply cleanly and corrected some device refcount problem. Patch (local) uml-extend-cmd-line-limits.patch 2004-10-27 Not accepted UML has some nasty limits on command line size. This patch allow for a bigger command line. Not accepted because it was rewroten by Jeff or Blaisorblade. Patch (local) cdc-acm-usb-use-uninit-mem-bug 2004-09-?? Accepted Out of bounds access. [Coverity] Patch was accepted by Greg. Patch (local) out-of-bounds-tuner.patch 2004-09-?? Accepted Out of bounds access. [Coverity] Patch was accepted by Gerd with some modifications. Patch (local) usb-serial-use-uninit-mem-bug.patch 2004-09-?? Accepted Out of bounds access. [Coverity] Patch was accepted by Greg. Patch (local)