aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* [TCP]: Fix RFC2465 typo.S P2006-03-281-1/+1
| | | | | Signed-off-by: S P <speattle@yahoo.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* [INET]: Introduce tunnel4/tunnel6Herbert Xu2006-03-2812-247/+344
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Basically this patch moves the generic tunnel protocol stuff out of xfrm4_tunnel/xfrm6_tunnel and moves it into the new files of tunnel4.c and tunnel6 respectively. The reason for this is that the problem that Hugo uncovered is only the tip of the iceberg. The real problem is that when we removed the dependency of ipip on xfrm4_tunnel we didn't really consider the module case at all. For instance, as it is it's possible to build both ipip and xfrm4_tunnel as modules and if the latter is loaded then ipip simply won't load. After considering the alternatives I've decided that the best way out of this is to restore the dependency of ipip on the non-xfrm-specific part of xfrm4_tunnel. This is acceptable IMHO because the intention of the removal was really to be able to use ipip without the xfrm subsystem. This is still preserved by this patch. So now both ipip/xfrm4_tunnel depend on the new tunnel4.c which handles the arbitration between the two. The order of processing is determined by a simple integer which ensures that ipip gets processed before xfrm4_tunnel. The situation for ICMP handling is a little bit more complicated since we may not have enough information to determine who it's for. It's not a big deal at the moment since the xfrm ICMP handlers are basically no-ops. In future we can deal with this when we look at ICMP caching in general. The user-visible change to this is the removal of the TUNNEL Kconfig prompts. This makes sense because it can only be used through IPCOMP as it stands. The addition of the new modules shouldn't introduce any problems since module dependency will cause them to be loaded. Oh and I also turned some unnecessary pskb's in IPv6 related to this patch to skb's. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
* [NET]: deinline 200+ byte inlines in sock.hDenis Vlasenko2006-03-282-87/+97
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Sizes in bytes (allyesconfig, i386) and files where those inlines are used: 238 sock_queue_rcv_skb 2.6.16/net/x25/x25_in.o 238 sock_queue_rcv_skb 2.6.16/net/rose/rose_in.o 238 sock_queue_rcv_skb 2.6.16/net/packet/af_packet.o 238 sock_queue_rcv_skb 2.6.16/net/netrom/nr_in.o 238 sock_queue_rcv_skb 2.6.16/net/llc/llc_sap.o 238 sock_queue_rcv_skb 2.6.16/net/llc/llc_conn.o 238 sock_queue_rcv_skb 2.6.16/net/irda/af_irda.o 238 sock_queue_rcv_skb 2.6.16/net/ipx/af_ipx.o 238 sock_queue_rcv_skb 2.6.16/net/ipv6/udp.o 238 sock_queue_rcv_skb 2.6.16/net/ipv6/raw.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/udp.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/raw.o 238 sock_queue_rcv_skb 2.6.16/net/ipv4/ipmr.o 238 sock_queue_rcv_skb 2.6.16/net/econet/econet.o 238 sock_queue_rcv_skb 2.6.16/net/econet/af_econet.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/sco.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/l2cap.o 238 sock_queue_rcv_skb 2.6.16/net/bluetooth/hci_sock.o 238 sock_queue_rcv_skb 2.6.16/net/ax25/ax25_in.o 238 sock_queue_rcv_skb 2.6.16/net/ax25/af_ax25.o 238 sock_queue_rcv_skb 2.6.16/net/appletalk/ddp.o 238 sock_queue_rcv_skb 2.6.16/drivers/net/pppoe.o 276 sk_receive_skb 2.6.16/net/decnet/dn_nsp_in.o 276 sk_receive_skb 2.6.16/net/dccp/ipv6.o 276 sk_receive_skb 2.6.16/net/dccp/ipv4.o 276 sk_receive_skb 2.6.16/net/dccp/dccp_ipv6.o 276 sk_receive_skb 2.6.16/drivers/net/pppoe.o 209 sk_dst_check 2.6.16/net/ipv6/ip6_output.o 209 sk_dst_check 2.6.16/net/ipv4/udp.o 209 sk_dst_check 2.6.16/net/decnet/dn_nsp_out.o Large inlines with multiple callers: Size Uses Wasted Name and definition ===== ==== ====== ================================================ 238 21 4360 sock_queue_rcv_skb include/net/sock.h 109 10 801 sock_recv_timestamp include/net/sock.h 276 4 768 sk_receive_skb include/net/sock.h 94 8 518 __sk_dst_check include/net/sock.h 209 3 378 sk_dst_check include/net/sock.h 131 4 333 sk_setup_caps include/net/sock.h 152 2 132 sk_stream_alloc_pskb include/net/sock.h 125 2 105 sk_stream_writequeue_purge include/net/sock.h Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
* [ECONET]: Convert away from SOCKOPS_WRAPPEDDavid S. Miller2006-03-281-55/+85
| | | | | | Just use a local econet_mutex instead. Signed-off-by: David S. Miller <davem@davemloft.net>
* [NET]: Fix ipx/econet/appletalk/irda ioctl crashesPetr Vandrovec2006-03-284-0/+86
| | | | | | | | | | | | Fix kernel oopses whenever somebody issues compatible ioctl on AppleTalk, Econet, IPX or IRDA socket. For AppleTalk/Econet/IRDA it restores state in which these sockets were before compat_ioctl was introduced to the socket ops, for IPX it implements support for 4 ioctls which were not implemented before - as these ioctls use structures which match between 32bit and 64bit userspace, no special code is needed, just call 64bit ioctl handler. Signed-off-by: Petr Vandrovec <petr@vandrovec.name> Signed-off-by: David S. Miller <davem@davemloft.net>
* [NET]: Kill Documentation/networking/TODODavid S. Miller2006-03-282-18/+1
| | | | | | | | | | Sorely out of date. Add the linux-net wiki web site to the NETWORKING maintainers entry, on which we maintain the current networking TODO list. Noticed by Randy Dunlap. Signed-off-by: David S. Miller <davem@davemloft.net>
* [TG3]: Update version and reldateMichael Chan2006-03-281-2/+2
| | | | | | | Update version to 3.55. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* [TG3]: Skip timer code during full lockMichael Chan2006-03-281-0/+4
| | | | | | | | Skip the main timer code if interrupts are disabled in the full lock state. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* [TG3]: Speed up SRAM accessMichael Chan2006-03-281-23/+30
| | | | | | | | | | | | | | | | | Speed up SRAM read and write functions if possible by using MMIO instead of config. cycles. With this change, the post reset signature done at the end of D3 power change must now be moved before the D3 power change. IBM reported a problem on powerpc blades during ethtool self test that was caused by the memory test taking excessively long. Config. cycles are very slow on powerpc and the memory test can take more than 10 seconds to complete using config. cycles. As a result, NETDEV WATCHDOG can be triggered during self test and the chip can end up in a funny state. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* [TG3]: Fix PHY loopback on 5700Michael Chan2006-03-281-1/+4
| | | | | | | Fix PHY loopback failure on some 5700 devices. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* [TG3]: Fix bug in 40-bit DMA workaround codeMichael Chan2006-03-281-1/+1
| | | | | | | | Need to check the TG3_FLAG_40BIT_DMA_BUG flag in the workaround code path instead of device flags. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* [TG3]: Fix probe failure due to invalid MAC addressMichael Chan2006-03-281-22/+28
| | | | | | | | | | | | | Some older bootcode in some devices may report 0 MAC address in SRAM when booting up from low power state. This patch fixes the problem by checking for a valid MAC address in SRAM and falling back to NVRAM if necessary. Thanks to walt <wa1ter@myrealbox.com> for reporting the problem and helping to debug it. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
* Merge master.kernel.org:/home/rmk/linux-2.6-armLinus Torvalds2006-03-2874-323/+4937
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | * master.kernel.org:/home/rmk/linux-2.6-arm: [ARM] 3388/1: ixp23xx: add core ixp23xx support [ARM] 3417/1: add support for logicpd pxa270 card engine [ARM] 3387/1: ixp23xx: add defconfig [ARM] 3377/2: add support for intel xsc3 core [ARM] Move ice-dcc code into misc.c [ARM] Fix decompressor serial IO to give CRLF not LFCR [ARM] proc-v6: mark page table walks outer-cacheable, shared. Enable NX. [ARM] nommu: trivial patch for arch/arm/lib/Makefile [ARM] 3416/1: Update LART site URL [ARM] 3415/1: Akita: Add missing EXPORT_SYMBOL [ARM] 3414/1: ep93xx: reset ethernet controller before uncompressing
| * [ARM] 3388/1: ixp23xx: add core ixp23xx supportLennert Buytenhek2006-03-2825-1/+2254
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Patch from Lennert Buytenhek This patch adds support for the Intel ixp23xx series of CPUs. The ixp23xx is an XSC3 based CPU with 512K of L2 cache, a 64bit 66MHz PCI interface, two DDR RAM interfaces, QDR RAM interfaces, two gigabit MACs, two 10/100 MACs, expansion bus, four microengines, a Media and Switch Fabric unit almost identical to the one on the ixp2400, two xscale (8250ish) UARTs and a bunch of other stuff. This patch adds the core ixp23xx support code, and support for the ADI Engineering Roadrunner, Intel IXDP2351, and IP Fabrics Double Espresso platforms. Signed-off-by: Deepak Saxena <dsaxena@plexity.net> Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3417/1: add support for logicpd pxa270 card engineLennert Buytenhek2006-03-285-0/+443
| | | | | | | | | | | | | | | | | | | | Patch from Lennert Buytenhek Add support for the LogicPD PXA270 Card Engine. Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3387/1: ixp23xx: add defconfigLennert Buytenhek2006-03-281-0/+1302
| | | | | | | | | | | | | | | | | | Patch from Lennert Buytenhek Add ixp23xx defconfig. Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3377/2: add support for intel xsc3 coreLennert Buytenhek2006-03-2811-2/+680
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Patch from Lennert Buytenhek This patch adds support for the new XScale v3 core. This is an ARMv5 ISA core with the following additions: - L2 cache - I/O coherency support (on select chipsets) - Low-Locality Reference cache attributes (replaces mini-cache) - Supersections (v6 compatible) - 36-bit addressing (v6 compatible) - Single instruction cache line clean/invalidate - LRU cache replacement (vs round-robin) I attempted to merge the XSC3 support into proc-xscale.S, but XSC3 cores have separate errata and have to handle things like L2, so it is simpler to keep it separate. L2 cache support is currently a build option because the L2 enable bit must be set before we enable the MMU and there is no easy way to capture command line parameters at this point. There are still optimizations that can be done such as using LLR for copypage (in theory using the exisiting mini-cache code) but those can be addressed down the road. Signed-off-by: Deepak Saxena <dsaxena@plexity.net> Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] Move ice-dcc code into misc.cRussell King2006-03-283-22/+14
| | | | | | | | Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] Fix decompressor serial IO to give CRLF not LFCRRussell King2006-03-2824-289/+187
| | | | | | | | | | | | | | | | | | As per the corresponding change to the serial drivers, arrange for ARM decompressors to give CRLF. Move the common putstr code into misc.c such that machines only need to supply "putc" and "flush" functions. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] proc-v6: mark page table walks outer-cacheable, shared. Enable NX.Russell King2006-03-271-4/+12
| | | | | | | | | | | | | | Mark page table walks with outer-cacheable attribute, and enable no-execute in page tables. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] nommu: trivial patch for arch/arm/lib/MakefileHyok S. Choi2006-03-271-1/+1
| | | | | | | | | | | | | | ifeq ($CONFIG_PREEMPT,y) -> ifeq ($(CONFIG_PREEMPT),y) Signed-off-by: Hyok S. Choi <hyok.choi@samsung.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3416/1: Update LART site URLErik Mouw2006-03-274-4/+4
| | | | | | | | | | | | | | | | | | | | | | Patch from Erik Mouw The LART website moved to http://www.lartmaker.nl/. This patch updates the URL in ARM specific files. Signed-off-by: Erik Mouw <erik@bitwizard.nl> Acked-by: Jan-Derk Bakker <jdb@lartmaker.nl> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3415/1: Akita: Add missing EXPORT_SYMBOLRichard Purdie2006-03-261-0/+2
| | | | | | | | | | | | | | | | | | Patch from Richard Purdie Add an EXPORT_SYMBOL for the Akita IO Expander Device. Signed-off-by: Richard Purdie <rpurdie@rpsys.net> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3414/1: ep93xx: reset ethernet controller before uncompressingLennert Buytenhek2006-03-261-1/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Patch from Lennert Buytenhek The Redboot version that cirrus supplies for the cirrus ep93xx doesn't turn off DMA from the ethernet MAC before jumping to linux, which means that we might end up with bits of RX status and packet data scribbled over the uncompressed kernel image. Work around this by resetting the ethernet MAC before we uncompress. We don't usually work around bootloader bugs, but considering that the large majority of ep93xx boards out there have this problem, I figured this it was justified in this case. Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
* | Merge master.kernel.org:/home/rmk/linux-2.6-serialLinus Torvalds2006-03-2810-1483/+190
|\ \ | | | | | | | | | | | | | | | | | | | | | * master.kernel.org:/home/rmk/linux-2.6-serial: [SERIAL] Provide Cirrus EP93xx AMBA PL010 serial support. [SERIAL] amba-pl010: allow platforms to specify modem control method [SERIAL] Remove obsoleted au1x00_uart driver [SERIAL] Small time UART configuration fix for AU1100 processor
| * | [SERIAL] Provide Cirrus EP93xx AMBA PL010 serial support.Lennert Buytenhek2006-03-261-0/+68
| | | | | | | | | | | | | | | Signed-off-by: Lennert Buytenhek <buytenh@wantstofly.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * | [SERIAL] amba-pl010: allow platforms to specify modem control methodRussell King2006-03-263-93/+119
| | | | | | | | | | | | | | | | | | | | | | | | The amba-pl010 hardware does not provide RTS and DTR control lines; it is expected that these will be implemented using GPIO. Allow platforms to supply a function to implement manipulation of modem control lines. Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * | [SERIAL] Remove obsoleted au1x00_uart driverRalf Baechle2006-03-265-1389/+2
| | | | | | | | | | | | | | | | | | | | | As announced in feature-removal-schedule.txt. Signed-off-by: Ralf Baechle <ralf@linux-mips.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * | [SERIAL] Small time UART configuration fix for AU1100 processorFreddy Spierenburg2006-03-261-1/+1
| |/ | | | | | | | | | | | | | | | | The AU1100 processor does not have an internal UART2. Only UART0, UART1 and UART3 exist. This patch removes the non existing UART2 and replaces it with a descriptive comment. Signed-off-by: Freddy Spierenburg <freddy@dusktilldawn.nl> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
* | Merge master.kernel.org:/pub/scm/linux/kernel/git/davej/cpufreqLinus Torvalds2006-03-284-98/+79
|\ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * master.kernel.org:/pub/scm/linux/kernel/git/davej/cpufreq: [CPUFREQ] cpufreq_conservative: keep ignore_nice_load and freq_step values when reselected [CPUFREQ] powernow: remove private for_each_cpu_mask() [CPUFREQ] hotplug cpu fix for powernow-k8 [PATCH] cpufreq_ondemand: add range check [PATCH] cpufreq_ondemand: keep ignore_nice_load value when it is reselected [PATCH] cpufreq_ondemand: Warn if it cannot run due to too long transition latency [PATCH] cpufreq_conservative: alternative initialise approach [PATCH] cpufreq_conservative: make for_each_cpu() safe [PATCH] cpufreq_conservative: alter default responsiveness [PATCH] cpufreq_conservative: aligning of codebase with ondemand
| * | [CPUFREQ] cpufreq_conservative: keep ignore_nice_load and freq_step values ↵Mattia Dongili2006-03-281-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | when reselected Keep the value of ignore_nice_load and freq_step of the conservative governor after the governor is deselected and reselected. Signed-off-by: Mattia Dongili <malattia@linux.it> Signed-off-by: Dave Jones <davej@redhat.com>
| * | [CPUFREQ] powernow: remove private for_each_cpu_mask()Andrew Morton2006-03-271-4/+0
| | | | | | | | | | | | | | | | | | | | | It is unneeded and wrong. Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Dave Jones <davej@redhat.com>
| * | [CPUFREQ] hotplug cpu fix for powernow-k8shin, jacob2006-03-271-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | Andi's previous fix to initialise powernow_data on all siblings will not work properly with CPU Hotplug. Signed-off-by: Jacob Shin <jacob.shin@amd.com> Signed-off-by: Dave Jones <davej@redhat.com>
| * | Merge git://git.kernel.org/pub/scm/linux/kernel/git/brodo/cpufreq-2.6Dave Jones2006-03-272-91/+71
| |\ \
| | * | [PATCH] cpufreq_ondemand: add range checkDominik Brodowski2006-03-261-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Assert that cpufreq_target is, at least, called with the minimum frequency allowed by this policy, not something lower. It triggered problems on ARM. Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
| | * | [PATCH] cpufreq_ondemand: keep ignore_nice_load value when it is reselectedEric Piel2006-03-261-2/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Keep the value of ignore_nice_load of the ondemand governor even after the governor has been deselected and selected back. This is the behavior of the other exported values of the ondemand governor and it's much more user-friendly. Signed-off-by: Eric Piel <eric.piel@tremplin-utc.net> Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
| | * | [PATCH] cpufreq_ondemand: Warn if it cannot run due to too long transition ↵Eric Piel2006-03-261-1/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | latency Display a warning if the ondemand governor can not be selected due to a transition latency of the cpufreq driver which is too long. Signed-off-by: Eric Piel <eric.piel@tremplin-utc.net> Acked-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
| | * | [PATCH] cpufreq_conservative: alternative initialise approachAlexander Clouter2006-03-261-37/+18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Venki, author of cpufreq_ondemand, came up with a neater way to remove the initialiser code from the main loop of my code and out to the point when the governor is actually initialised. Not only does it look but it also feels cleaner, plus its simpler to understand. It also saves a bunch of pointless conditional statements in the main loop. Signed-off-by: Alexander Clouter <alex-kernel@digriz.org.uk> Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
| | * | [PATCH] cpufreq_conservative: make for_each_cpu() safeAlexander Clouter2006-03-261-49/+42
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | All these changes should make cpufreq_conservative safe in regards to the x86 for_each_cpu cpumask.h changes and whatnot. Whilst making it safe a number of pointless for loops related to the cpu mask's were removed. I was never comfortable with all those for loops, especially as the iteration is over the same data again and again for each CPU you had in a single poll, an O(n^2) outcome to frequency scaling. The approach I use is to assume by default no CPU's exist and it sets the requested_freq to zero as a kind of flag, the reasoning is in the source ;) If the CPU is queried and requested_freq is zero then it initialises the variable to current_freq and then continues as if nothing happened which should be the same net effect as before? Signed-off-by: Alexander Clouter <alex-kernel@digriz.org.uk> Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
| | * | [PATCH] cpufreq_conservative: alter default responsivenessAlexander Clouter2006-03-261-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The sensible approach to making conservative less responsive than ondemand :) As mentioned in patch [1/4]. We do not want conservative to shoot through all the frequencies, its point (by default) is to slowly move through them. By default its ten times less responsive. Signed-off-by: Alexander Clouter <alex-kernel@digriz.org.uk> Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
| | * | [PATCH] cpufreq_conservative: aligning of codebase with ondemandAlexander Clouter2006-03-261-26/+27
| | |/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since the conservative govenor was released its codebase has drifted from the the direction and updates that have been applied to the ondemand govornor. This patch addresses the lack of updates in that period and brings conservative back up to date. The resulting diff file between cpufreq_ondemand.c and cpufreq_conservative.c is now much smaller and shows more clearly the differences between the two. Another reason to do this is ages ago, knowingly, I did a piss poor attempt at making conservative less responsive by knocking up DEF_SAMPLING_RATE_LATENCY_MULTIPLIER by two orders of magnitude. I did fix this ages ago but in my dis-organisation I must have toasted the diff and left it the way it was. About two weeks ago a user contacted me saying he was having problems with the conservative governor with his AMD Athlon XP-M 2800+ as /sys/devices/system/cpu/cpu0/cpufreq/conservative showed sampling_rate_min 9950000 sampling_rate_max 1360065408 Nine seconds to decide about changing the frequency....not too responsive :) Signed-off-by: Alexander Clouter <alex-kernel@digriz.org.uk> Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
* | | Merge branch 'cfq-merge' of git://brick.kernel.dk/data/git/linux-2.6-blockLinus Torvalds2006-03-283-180/+220
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | * 'cfq-merge' of git://brick.kernel.dk/data/git/linux-2.6-block: [BLOCK] cfq-iosched: seek and async performance fixes [PATCH] ll_rw_blk: fix 80-col offender in put_io_context() [PATCH] cfq-iosched: small cfq_choose_req() optimization [PATCH] [BLOCK] cfq-iosched: change cfq io context linking from list to tree
| * | | [BLOCK] cfq-iosched: seek and async performance fixesJens Axboe2006-03-282-38/+72
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Detect whether a given process is seeky and if so disable (mostly) the idle window if it is. We still allow just a little idle time, just enough to allow that process to submit a new request. That is needed to maintain fairness across priority groups. In some cases, we could setup several async queues. This is not optimal from a performance POV, since we want all async io in one queue to perform good sorting on it. It also impacted sync queues, as async io got too much slice time. Signed-off-by: Jens Axboe <axboe@suse.de>
| * | | [PATCH] ll_rw_blk: fix 80-col offender in put_io_context()Jens Axboe2006-03-281-1/+3
| | | | | | | | | | | | | | | | | | | | | | | | This makes akpm more happy. Signed-off-by: Jens Axboe <axboe@suse.de>
| * | | [PATCH] cfq-iosched: small cfq_choose_req() optimizationAndreas Mohr2006-03-281-18/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | this is a small optimization to cfq_choose_req() in the CFQ I/O scheduler (this function is a semi-often invoked candidate in an oprofile log): by using a bit mask variable, we can use a simple switch() to check the various cases instead of having to query two variables for each check. Benefit: 251 vs. 285 bytes footprint of cfq_choose_req(). Also, common case 0 (no request wrapping) is now checked first in code. Signed-off-by: Andreas Mohr <andi@lisas.de> Signed-off-by: Jens Axboe <axboe@suse.de>
| * | | [PATCH] [BLOCK] cfq-iosched: change cfq io context linking from list to treeJens Axboe2006-03-283-124/+114
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | On setups with many disks, we spend a considerable amount of time looking up the process-disk mapping on each queue of io. Testing with a NULL based block driver, this costs 40-50% reduction in throughput for 1000 disks. Signed-off-by: Jens Axboe <axboe@suse.de>
* | | Merge master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6Linus Torvalds2006-03-281-2/+20
|\ \ \ | | | | | | | | | | | | | | | | * master.kernel.org:/pub/scm/linux/kernel/git/davem/sparc-2.6: [SPARC64]: Implement futex_atomic_cmpxchg_inatomic().
| * | | [SPARC64]: Implement futex_atomic_cmpxchg_inatomic().David S. Miller2006-03-281-2/+20
| |/ / | | | | | | | | | Signed-off-by: David S. Miller <davem@davemloft.net>
* | | [PATCH] compat_sys_futex() warning fixAndrew Morton2006-03-281-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | kernel/futex_compat.c: In function `compat_sys_futex': kernel/futex_compat.c:140: warning: passing arg 1 of `do_futex' makes integer from pointer without a cast kernel/futex_compat.c:140: warning: passing arg 5 of `do_futex' makes integer from pointer without a cast Not sure what Ingo was thinking of here. Put the casts back in. Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
* | | [PATCH] sound: Remove unneeded kmalloc() return value castsJesper Juhl2006-03-284-7/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Get rid of unnessesary casts of kmalloc() return value in sound/ Signed-off-by: Jesper Juhl <jesper.juhl@gmail.com> Cc: Jaroslav Kysela <perex@perex.cz> Cc: Takashi Iwai <tiwai@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
generated by cgit v1.1 at 2024-09-13 00:15:30 +0000