aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* clean up duplicated alloc/free_thread_infoFUJITA Tomonori2008-07-25
| | | | | | | | | | | | | | | We duplicate alloc/free_thread_info defines on many platforms (the majority uses __get_free_pages/free_pages). This patch defines common defines and removes these duplicated defines. __HAVE_ARCH_THREAD_INFO_ALLOCATOR is introduced for platforms that do something different. Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> Acked-by: Russell King <rmk+kernel@arm.linux.org.uk> Cc: Pekka Enberg <penberg@cs.helsinki.fi> Cc: <linux-arch@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* misc: add HP WMI laptop extras driverMatthew Garrett2008-07-25
| | | | | | | | | | | | | | This driver adds support for reading and configuring certain information on modern HP laptops with WMI BIOS interfaces. It supports enabling and disabling the ambient light sensor, querying attached displays and hard drive temperature, sending events on docking and querying the state of the dock and toggling the state of the wifi, bluetooth and wwan hardware via rfkill. It also makes the little "(i)" button work on machines that send that via WMI rather than via the keyboard controller. Signed-off-by: Matthew Garrett <mjg@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* call_usermodehelper(): increase reliabilityKOSAKI Motohiro2008-07-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Presently call_usermodehelper_setup() uses GFP_ATOMIC. but it can return NULL _very_ easily. GFP_ATOMIC is needed only when we can't sleep. and, GFP_KERNEL is robust and better. thus, I add gfp_mask argument to call_usermodehelper_setup(). So, its callers pass the gfp_t as below: call_usermodehelper() and call_usermodehelper_keys(): depend on 'wait' argument. call_usermodehelper_pipe(): always GFP_KERNEL because always run under process context. orderly_poweroff(): pass to GFP_ATOMIC because may run under interrupt context. Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: "Paul Menage" <menage@google.com> Reviewed-by: Li Zefan <lizf@cn.fujitsu.com> Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Andi Kleen <andi@firstfloor.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* remove some more tipar bitsAdrian Bunk2008-07-25
| | | | | | | | Some bits were missed when the tipar driver was removed. Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* asm-generic/int-ll64.h: always provide __{s,u}64Adrian Bunk2008-07-25
| | | | | | | | | | | | | | | | | Several compilers offer "long long" without claiming to support C99. Considering how frequent __s64/__u64 are used our userspace headers are anyway unusable without __s64/__u64 available. Always offer __s64/__u64 to non-gcc non-C99 compilers - if they provide "long long" that makes the headers compiling and if they don't they are anyway screwed. Signed-off-by: Adrian Bunk <bunk@kernel.org> Acked-by: H. Peter Anvin <hpa@zytor.com> Cc: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* build-kernel-profileo-only-when-requested-cleanupsAndrew Morton2008-07-25
| | | | | | | Cc: Adrian Bunk <bunk@kernel.org> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* build kernel/profile.o only when requestedAdrian Bunk2008-07-25
| | | | | | | | | | | | | | | | | | | | | | Build kernel/profile.o only if CONFIG_PROFILING is enabled. This makes CONFIG_PROFILING=n kernels smaller. As a bonus, some profile_tick() calls and one branch from schedule() are now eliminated with CONFIG_PROFILING=n (but I doubt these are measurable effects). This patch changes the effects of CONFIG_PROFILING=n, but I don't think having more than two choices would be the better choice. This patch also adds the name of the first parameter to the prototypes of profile_{hits,tick}() since I anyway had to add them for the dummy functions. Signed-off-by: Adrian Bunk <bunk@kernel.org> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* list_for_each_rcu must die: networkingPaul E. McKenney2008-07-25
| | | | | | | | | | | | All uses of list_for_each_rcu() can be profitably replaced by the easier-to-use list_for_each_entry_rcu(). This patch makes this change for networking, in preparation for removing the list_for_each_rcu() API entirely. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* kallsyms: fix potential overflow in binary searchVegard Nossum2008-07-25
| | | | | | | | | | | This will probably never trigger... but it won't hurt to be careful. http://googleresearch.blogspot.com/2006/06/extra-extra-read-all-about-it-nearly.html Signed-off-by: Vegard Nossum <vegard.nossum@gmail.com> Cc: Joshua Bloch <jjb@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* introduce HAVE_EFFICIENT_UNALIGNED_ACCESS Kconfig symbolJohannes Berg2008-07-25
| | | | | | | | | | | | | | | | | | | In many cases, especially in networking, it can be beneficial to know at compile time whether the architecture can do unaligned accesses efficiently. This patch introduces a new Kconfig symbol HAVE_EFFICIENT_UNALIGNED_ACCESS for that purpose and adds it to the powerpc and x86 architectures. Also add some documentation about alignment and networking, and especially one intended use of this symbol. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Sam Ravnborg <sam@ravnborg.org> Acked-by: Ingo Molnar <mingo@elte.hu> [x86 architecture part] Cc: <linux-arch@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* lists: remove a redundant conditional definition of list_add()Robert P. J. Day2008-07-25
| | | | | | | | | | | | Remove the conditional surrounding the definition of list_add() from list.h since, if you define CONFIG_DEBUG_LIST, the definition you will subsequently pick up from lib/list_debug.c will be absolutely identical, at which point you can remove that redundant definition from list_debug.c as well. Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca> Cc: Dave Jones <davej@codemonkey.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* lib: allow memparse() to accept a NULL and ignorable second parmRobert P. J. Day2008-07-25
| | | | | | | | | | | | | | | | | | | Extend memparse() to allow the caller to use a NULL second parameter, which would represent no interest in returning the address of the end of the parsed string. In numerous cases, callers invoke memparse() to parse a possibly-suffixed string (such as "64K" or "2G" or whatever) and define a character pointer to accept the end pointer being returned by memparse() even though they have no interest in it and promptly throw it away. This (backward-compatible) enhancement allows callers to use NULL in the cases where they just don't care about getting back that end pointer. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* init/: delete hard-coded setting and testing of BUILD_CRAMDISKRobert P. J. Day2008-07-25
| | | | | | | | | There seems to be little point in explicitly setting, then testing the macro BUILD_CRAMDISK within the context of a single source file. Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Remove apparently unused fd1772.h header file.Robert P. J. Day2008-07-25
| | | | | | | | | | | | This header file has been unused for quite some time, and the corresponding source files appear to have been removed back in commit 99eb8a550dbccc0e1f6c7e866fe421810e0585f6 ("Remove the arm26 port") Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca> Cc: Adrian Bunk <bunk@stusta.de> Cc: Ian Molton <spyro@f2s.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* init/do_mounts.c should #include <linux/initrd.h>Adrian Bunk2008-07-25
| | | | | | | | | Every file should include the headers containing the externs for its global code (in this case for rd_doload). Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* fix checkstack.pl arch detectionEric Sandeen2008-07-25
| | | | | | | | | | uname -m was leaving a newline in $arch, and not passing the tests. Also, printing the unknown arch on failure is probably helpful. Signed-off-by: Eric Sandeen <sandeen@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* find dynamic stack allocations in checkstack.plEric Sandeen2008-07-25
| | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, checkstack.pl only looks for fixed subtractions from the stack pointer. However, things like this: void function(int size) { char stackbuster[size << 2]; ... are certainly worth pointing out, I think. This could perhaps be done more cleanly, and the following patch only adds "dynamic" REs for x86 and x86_64, but it works: 0x00b0 crypto_cbc_decrypt_inplace [cbc]: Dynamic (%rax) 0x00ad crypto_pcbc_decrypt_inplace [pcbc]: Dynamic (%rax) 0x02f6 crypto_pcbc_encrypt_inplace [pcbc]: Dynamic (%rax) 0x036c _crypto_xcbc_digest_setkey [xcbc]: Dynamic (%rax) ... (Inspired by Keith Owens' old stack-check script) Signed-off-by: Eric Sandeen <sandeen@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* lzo: use get/put_unaligned_* helpersHarvey Harrison2008-07-25
| | | | | | | Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Acked-by: Richard Purdie <rpurdie@rpsys.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* include: use get/put_unaligned_* helpersHarvey Harrison2008-07-25
| | | | | | | Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Cc: "John W. Linville" <linville@tuxdriver.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* fs: ldm.[ch] use get_unaligned_* helpersHarvey Harrison2008-07-25
| | | | | | | | | Replace the private BE16/BE32/BE64 macros with direct calls to get_unaligned_be16/32/64. Signed-off-by: Harvey Harrison <harvey.harrison@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* locking: add typecheck on irqsave and friends for correct flagsSteven Rostedt2008-07-25
| | | | | | | | | | | | | | | | There haave been several areas in the kernel where an int has been used for flags in local_irq_save() and friends instead of a long. This can cause some hard to debug problems on some architectures. This patch adds a typecheck inside the irqsave and restore functions to flag these cases. [akpm@linux-foundation.org: coding-style fixes] [akpm@linux-foundation.org: build fix] Signed-off-by: Steven Rostedt <srostedt@redhat.com> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* split the typecheck macros out of include/linux/kernel.hAndrew Morton2008-07-25
| | | | | | | | | | Needed to fix up a recursive include snafu in locking-add-typecheck-on-irqsave-and-friends-for-correct-flags.patch Cc: Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* flag parameters: fix compile error of sys_epoll_create1Wang Chen2008-07-25
| | | | | | | | | | | | | | | | | | GEN .version CHK include/linux/compile.h UPD include/linux/compile.h CC init/version.o LD init/built-in.o LD vmlinux arch/x86/kernel/built-in.o: In function `sys_call_table': (.rodata+0x8a4): undefined reference to `sys_epoll_create1' make: *** [vmlinux] Error 1 Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com> Cc: Ulrich Drepper <drepper@redhat.com> Cc: Davide Libenzi <davidel@xmailserver.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* mfd: don't use memzeroAndrew Morton2008-07-25
| | | | | | | | | | For it doesn't exist on i386. Cc: Ian Molton <spyro@f2s.com> Cc: Dmitry Baryshkov <dbaryshkov@gmail.com> Cc: Russell King <rmk+kernel@arm.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* endian: Always evaluate arguments.David Miller2008-07-25
| | | | | | | | | | | | | | | | | | | | Changeset 7fa897b91a3ea0f16c2873b869d7a0eef05acff4 ("ide: trivial sparse annotations") created an IDE bootup regression on big-endian systems. In drivers/ide/ide-iops.c, function ide_fixstring() we now have the loop: for (p = end ; p != s;) be16_to_cpus((u16 *)(p -= 2)); which will never terminate on big-endian because in such a configuration be16_to_cpus() evaluates to "do { } while (0)" Therefore, always evaluate the arguments to nop endian transformation operations. Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* char/xilinx_hwicap/xilinx_hwicap.c: Removed duplicated includeHuang Weiyi2008-07-25
| | | | | | | | Removed duplicated include file <linux/version.h> in char/xilinx_hwicap/xilinx_hwicap.c. Signed-off-by: Huang Weiyi <weiyi.huang@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Undo duplicate "m68k: drivers/input/serio/hp_sdc.c needs <linux/semaphore.h>"Linus Torvalds2008-07-25
| | | | | | | | | | | | Both commits 0f17e4c796e89d1f69f13b653aba60e6ccfb8ae0 ("Add missing semaphore.h includes") and 4933d07531711e399d8d578036aa9fc1be2f9b20 ("m68k: drivers/input/serio/hp_sdc.c needs <linux/semaphore.h>") added a We only really need one ;) Reported-by: Huang Weiyi <weiyi.huang@gmail.com> Requested-by: Dmitry Torokhov <dmitry.torokhov@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linusLinus Torvalds2008-07-24
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus: virtio: Add transport feature handling stub for virtio_ring. virtio: Rename set_features to finalize_features virtio: Formally reserve bits 28-31 to be 'transport' features. s390: use virtio_console for KVM on s390 virtio: console as a config option virtio_console: use virtqueue notification for hvc_console hvc_console: rework setup to replace irq functions with callbacks virtio_blk: check for hardsector size from host virtio: Use bus_type probe and remove methods virtio: don't always force a notification when ring is full virtio: clarify that ABI is usable by any implementations virtio: Recycle unused recv buffer pages for large skbs in net driver virtio net: Allow receiving SG packets virtio net: Add ethtool ops for SG/GSO virtio: fix virtio_net xmit of freed skb bug
| * virtio: Add transport feature handling stub for virtio_ring.Rusty Russell2008-07-24
| | | | | | | | | | | | | | | | To prepare for virtio_ring transport feature bits, hook in a call in all the users to manipulate them. This currently just clears all the bits, since it doesn't understand any features. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * virtio: Rename set_features to finalize_featuresRusty Russell2008-07-24
| | | | | | | | | | | | | | | | | | Rather than explicitly handing the features to the lower-level, we just hand the virtio_device and have it set the features. This make it clear that it has the chance to manipulate the features of the device at this point (and that all feature negotiation is already done). Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * virtio: Formally reserve bits 28-31 to be 'transport' features.Rusty Russell2008-07-24
| | | | | | | | | | | | | | We assign feature bits as required, but it makes sense to reserve some for the particular transport, rather than the particular device. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * s390: use virtio_console for KVM on s390Christian Borntraeger2008-07-24
| | | | | | | | | | | | | | | | | | This patch enables virtio_console as the default console on kvm for s390. We currently use the same notify hack as lguest for early console output. I will try to address this for lguest and s390 later. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * virtio: console as a config optionChristian Borntraeger2008-07-24
| | | | | | | | | | | | | | | | | | | | | | I also added a small Kconfig change that allows the user to specify the virtio console in menuconfig. (Fixes to export symbols from Stephen Rothwell <sfr@canb.auug.org.au>) (Fixes for CONFIG_VIRTIO_CONSOLE=y vs CONFIG_VIRTIO=m from Christian himself) Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Cc: Stephen Rothwell <sfr@canb.auug.org.au>
| * virtio_console: use virtqueue notification for hvc_consoleChristian Borntraeger2008-07-24
| | | | | | | | | | | | | | | | This patch exploits the new notifier callbacks of the hvc_console. We can use the virtio callbacks instead of the polling code. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * hvc_console: rework setup to replace irq functions with callbacksChristian Borntraeger2008-07-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch tries to change hvc_console to not use request_irq/free_irq if the backend does not use irqs. This allows virtio_console to use hvc_console without having a linker reference to request_irq/free_irq. In addition, together with patch 2/3 it improves the performance for virtio console input. (an earlier version of this patch was tested by Yajin on lguest) The irq specific code is moved to hvc_irq.c and selected by the drivers that use irqs (System p, System i, XEN). I replaced "int irq" with the opaque "int data". The request_irq and free_irq calls are replaced with notifier_add and notifier_del. I have also changed the code a bit to call the notifier_add and notifier_del inside the spinlock area as the callbacks are found via hp->ops. Changes since last version: o remove ifdef o reintroduce "irq_requested" as "notified" o cleanups, sparse.. I did not move the timer based polling into a separate polling scheme. I played with several variants, but it seems we need to sleep/schedule in a thread even for irq based consoles, as there are throttleing and buffer size constraints. I also kept hvc_struct defined in hvc_console.h so that hvc_irq.c can access the irq_requested element. Feedback is appreciated. virtio_console is currently the only available console for kvm on s390. I plan to push this change as soon as all affected parties agree on it. I would love to get test results from System p, Xen etc. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * virtio_blk: check for hardsector size from hostChristian Borntraeger2008-07-24
| | | | | | | | | | | | | | | | | | | | | | | | Currently virtio_blk assumes a 512 byte hard sector size. This can cause trouble / performance issues if the backing has a different block size (like a file on an ext3 file system formatted with 4k block size or a dasd). Lets add a feature flag that tells the guest to use a different hard sector size than 512 byte. Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * virtio: Use bus_type probe and remove methodsMark McLoughlin2008-07-24
| | | | | | | | | | | | | | | | | | Hook up to the probe() and remove() methods in bus_type rather than device_driver. The latter has been preferred since 2.6.16. Signed-off-by: Mark McLoughlin <markmc@redhat.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * virtio: don't always force a notification when ring is fullRusty Russell2008-07-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We force notification when the ring is full, even if the host has indicated it doesn't want to know. This seemed like a good idea at the time: if we fill the transmit ring, we should tell the host immediately. Unfortunately this logic also applies to the receiving ring, which is refilled constantly. We should introduce real notification thesholds to replace this logic. Meanwhile, removing the logic altogether breaks the heuristics which KVM uses, so we use a hack: only notify if there are outgoing parts of the new buffer. Here are the number of exits with lguest's crappy network implementation: Before: network xmit 7859051 recv 236420 After: network xmit 7858610 recv 118136 Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * virtio: clarify that ABI is usable by any implementationsRusty Russell2008-07-24
| | | | | | | | | | | | | | | | | | | | | | | | | | We want others to implement and use virtio, so it makes sense to BSD license the non-__KERNEL__ parts of the headers to make this crystal clear. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Acked-by: Christian Borntraeger <borntraeger@de.ibm.com> Acked-by: Mark McLoughlin <markmc@redhat.com> Acked-by: Ryan Harper <ryanh@us.ibm.com> Acked-by: Eric Van Hensbergen <ericvh@gmail.com> Acked-by: Anthony Liguori <aliguori@us.ibm.com>
| * virtio: Recycle unused recv buffer pages for large skbs in net driverRusty Russell2008-07-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If we hack the virtio_net driver to always allocate full-sized (64k+) skbuffs, the driver slows down (lguest numbers): Time to receive 1GB (small buffers): 10.85 seconds Time to receive 1GB (64k+ buffers): 24.75 seconds Of course, large buffers use up more space in the ring, so we increase that from 128 to 2048: Time to receive 1GB (64k+ buffers, 2k ring): 16.61 seconds If we recycle pages rather than using alloc_page/free_page: Time to receive 1GB (64k+ buffers, 2k ring, recycle pages): 10.81 seconds This demonstrates that with efficient allocation, we don't need to have a separate "small buffer" queue. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
| * virtio net: Allow receiving SG packetsHerbert Xu2008-07-24
| | | | | | | | | | | | | | | | | | | | Finally this patch lets virtio_net receive GSO packets in addition to sending them. This can definitely be optimised for the non-GSO case. For comparison the Xen approach stores one page in each skb and uses subsequent skb's pages to construct an SG skb instead of preallocating the maximum amount of pages per skb. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (added feature bits)
| * virtio net: Add ethtool ops for SG/GSOHerbert Xu2008-07-24
| | | | | | | | | | | | | | | | This patch adds some basic ethtool operations to virtio_net so I could test SG without GSO (which was really useful because TSO turned out to be buggy :) Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (remove MTU setting)
| * virtio: fix virtio_net xmit of freed skb bugMark McLoughlin2008-07-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | On Mon, 2008-05-26 at 17:42 +1000, Rusty Russell wrote: > If we fail to transmit a packet, we assume the queue is full and put > the skb into last_xmit_skb. However, if more space frees up before we > xmit it, we loop, and the result can be transmitting the same skb twice. > > Fix is simple: set skb to NULL if we've used it in some way, and check > before sending. ... > diff -r 564237b31993 drivers/net/virtio_net.c > --- a/drivers/net/virtio_net.c Mon May 19 12:22:00 2008 +1000 > +++ b/drivers/net/virtio_net.c Mon May 19 12:24:58 2008 +1000 > @@ -287,21 +287,25 @@ again: > free_old_xmit_skbs(vi); > > /* If we has a buffer left over from last time, send it now. */ > - if (vi->last_xmit_skb) { > + if (unlikely(vi->last_xmit_skb)) { > if (xmit_skb(vi, vi->last_xmit_skb) != 0) { > /* Drop this skb: we only queue one. */ > vi->dev->stats.tx_dropped++; > kfree_skb(skb); > + skb = NULL; > goto stop_queue; > } > vi->last_xmit_skb = NULL; With this, may drop an skb and then later in the function discover that we could have sent it after all. Poor wee skb :) How about the incremental patch below? Cheers, Mark. Subject: [PATCH] virtio_net: Delay dropping tx skbs Currently we drop the skb in start_xmit() if we have a queued buffer and fail to transmit it. However, if we delay dropping it until we've stopped the queue and enabled the tx notification callback, then there is a chance space might become available for it. Signed-off-by: Mark McLoughlin <markmc@redhat.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
* | Label kthread_create() with printf attribute tag.Rusty Russell2008-07-24
| | | | | | | | | | | | | | Obvious misc patch been in my queue (& linux-next) for over a cycle. Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | fix fs/nfs/nfsroot.c compilationAdrian Bunk2008-07-24
| | | | | | | | | | | | | | | | | | | | | | | | | | This fixes the following compile error caused by commit f9247273cb69ba101877e946d2d83044409cc8c5 ("UFS: add const to parser token table"): CC fs/nfs/nfsroot.o /home/bunk/linux/kernel-2.6/git/linux-2.6/fs/nfs/nfsroot.c:130: error: tokens causes a section type conflict make[3]: *** [fs/nfs/nfsroot.o] Error 1 Signed-off-by: Adrian Bunk <bunk@kernel.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | x86/oprofile/nmi_int: add Nehalem to list of ppro coresLinus Torvalds2008-07-24
| | | | | | | | | | | | | | | | | | ..otherwise oprofile will fall back on that poor timer interrupt. Also replace the unreadable chain of if-statements with a "switch()" statement instead. It generates better code, and is a lot clearer. Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* | x86-64: Clean up 'save/restore_i387()' usageLinus Torvalds2008-07-24
|/ | | | | | | | | | | | | | | | | | | | Suresh Siddha wants to fix a possible FPU leakage in error conditions, but the fact that save/restore_i387() are inlines in a header file makes that harder to do than necessary. So start off with an obvious cleanup. This just moves the x86-64 version of save/restore_i387() out of the header file, and moves it to the only file that it is actually used in: arch/x86/kernel/signal_64.c. So exposing it in a header file was wrong to begin with. [ Side note: I'd like to fix up some of the games we play with the 32-bit version of these functions too, but that's a separate matter. The 32-bit versions are shared - under different names at that! - by both the native x86-32 code and the x86-64 32-bit compatibility code ] Acked-by: Suresh Siddha <suresh.b.siddha@intel.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* Merge git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6Linus Torvalds2008-07-24
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * git://git.kernel.org/pub/scm/linux/kernel/git/bart/ide-2.6: (76 commits) ide: use proper printk() KERN_* levels in ide-probe.c ide: fix for EATA SCSI HBA in ATA emulating mode ide: remove stale comments from drivers/ide/Makefile ide: enable local IRQs in all handlers for TASKFILE_NO_DATA data phase ide-scsi: remove kmalloced struct request ht6560b: remove old history ht6560b: update email address ide-cd: fix oops when using growisofs gayle: release resources on ide_host_add() failure palm_bk3710: add UltraDMA/100 support ide: trivial sparse annotations ide: ide-tape.c sparse annotations and unaligned access removal ide: drop 'name' parameter from ->init_chipset method ide: prefix messages from IDE PCI host drivers by driver name it821x: remove DECLARE_ITE_DEV() macro it8213: remove DECLARE_ITE_DEV() macro ide: include PCI device name in messages from IDE PCI host drivers ide: remove <asm/ide.h> for some archs ide-generic: remove ide_default_{io_base,irq}() inlines (take 3) ide-generic: is no longer needed on ppc32 ...
| * ide: use proper printk() KERN_* levels in ide-probe.cBartlomiej Zolnierkiewicz2008-07-24
| | | | | | | | | | | | | | | | While at it: - fixup printk() messages in save_match() and hwif_init(). Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
| * ide: fix for EATA SCSI HBA in ATA emulating modeBartlomiej Zolnierkiewicz2008-07-24
| | | | | | | | | | | | | | | | | | | | | | | | IDE probing code used to skip devices attached to EATA SCSI HBA in ATA emulating mode but because of warm-plug support port I/O resources are no longer freed if no devices are detected on a port and the decision about the driver to use is left up to the user. Remove no longer valid EATA SCSI HBA quirk from do_identify(). Noticed-by: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Bartlomiej Zolnierkiewicz <bzolnier@gmail.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-12-17 01:59:29 -0500