- 29 Jun, 2005 26 commits
-
-
Mingming Cao authored
Currently in ext3 block reservation code, the global filesystem reservation tree lock (rsv_block) is hold during the process of searching for a space to make a new reservation window, including while scaning the block bitmap to verify if the avalible window has a free block. Holding the lock during bitmap scan is unnecessary and could possibly cause scalability issue and latency issues. This patch tries to address this by dropping the lock before scan the bitmap. Before that we need to reserve the open window in case someone else is targetting at the same window. Question was should we reserve the whole free reservable space or just the window size we need. Reserve the whole free reservable space will possibly force other threads which intended to do block allocation nearby move to another block group(cause bad layout). In this patch, we just reserve the desired size before drop the lock and scan the block bitmap. This patch fixed a ext3 reservation latency issue seen on a cvs check out test. Patch is tested with many fsx, tiobench, dbench and untar a kernel test. Signed-Off-By:
Mingming Cao <cmm@us.ibm.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Nick Piggin authored
get_io_context needlessly turned off interrupts and checked for racing io context creations. Both of which aren't needed, because the io context can only be created while in process context of the current process. Also, split the function in 2. A light version, current_io_context does not elevate the reference count specifically, but can be used when in process context, because the process holds a reference itself. Signed-off-by:
Nick Piggin <nickpiggin@yahoo.com.au> Cc: Jens Axboe <axboe@suse.de> Signed-off-by:
-
Nick Piggin authored
Change around locking a bit for a result of 1-2 less spin lock unlock pairs in request submission paths. Signed-off-by:
-
Nick Piggin authored
In the case where the request is not able to be merged by the elevator, don't retake the lock and retry the merge mechanism after allocating a new request. Instead assume that the chance of a merge remains slim, and now that we've done most of the work allocating a request we may as well just go with it. Also be rid of the GFP_ATOMIC allocation: we've got working mempools for the block layer now, so let's save atomic memory for things like networking. Lastly, in get_request_wait, do an initial get_request call before going into the waitqueue. This is reported to help efficiency. Signed-off-by:
-
KAMBAROV, ZAUR authored
We add a check of the return value of tty_ldisc_ref(), which is checked 7 out of 8 times, e.g.: 149 ld = tty_ldisc_ref(tty); 150 if (ld != NULL) { 151 if (ld->set_termios) 152 (ld->set_termios)(tty, &old_termios); 153 tty_ldisc_deref(ld); 154 } This defect was found automatically by Coverity Prevent, a static analysis tool. (akpm: presumably `ld' is never NULL. Oh well) Signed-off-by:Zaur Kambarov <zkambarov@coverity.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by:
-
KAMBAROV, ZAUR authored
The return value of "match_int" is checked 27 out of 28 times In lib/parser.c 142 /** 143 * match_int: - scan a decimal representation of an integer from a substring_t 144 * @s: substring_t to be scanned 145 * @result: resulting integer on success 146 * 147 * Description: Attempts to parse the &substring_t @s as a decimal integer. On 148 * success, sets @result to the integer represented by the string and returns 0. 149 * Returns either -ENOMEM or -EINVAL on failure. 150 */ 151 int match_int(substring_t *s, int *result) 152 { 153 return match_number(s, result, 0); 154 } Signed-off-by: -
KAMBAROV, ZAUR authored
"dir" was dereferenced before null check Signed-off-by:
-
KAMBAROV, ZAUR authored
We fix the check in 1084, which was 1084 if (addr->channel > IPMI_NUM_CHANNELS) { 1085 spin_lock_irqsave(&intf->counter_lock, flags); 1086 intf->sent_invalid_commands++; 1087 spin_unlock_irqrestore(&intf->counter_lock, flags); 1088 rv = -EINVAL; 1089 goto out_err; 1090 } addr->channel is used in 1092 if (intf->channels[addr->channel].medium Definitions involved: 221 struct ipmi_channel channels[IPMI_MAX_CHANNELS]; 134 #define IPMI_MAX_CHANNELS 8 In /linux-2.6.12-rc6/include/linux/ipmi.h 148 #define IPMI_NUM_CHANNELS 0x10 Signed-off-by: -
KAMBAROV, ZAUR authored
The check in 627 BUG_ON(index > SG_MEMPOOL_NR); with SG_MEMPOOL_NR defined in 32 #define SG_MEMPOOL_NR (sizeof(scsi_sg_pools)/sizeof(struct scsi_host_sg_pool)) was not sufficient. sgp, set in 629 sgp = scsi_sg_pools + index; is dereferenced in 630 mempool_free(sgl, sgp->pool); Signed-off-by:
-
KAMBAROV, ZAUR authored
Variable "c" was declared as an unsigned int, but used in: 125 for (i=0 ; (c=read(fd, buf, sizeof(buf)))>0 ; i+=c ) 126 if (write(1, buf, c) != c) 127 die("Write call failed"); (akpm: read() can return -1. If it does, we fill the disk up with garbage). Signed-off-by: -
GOTO Masanori authored
This patch for usb_ch9.h includes linux/types.h instead of asm/types.h so that __le16 and so on is explicitly defined. It also cleans up non standard // comment. Signed-off-by:
GOTO Masanori <gotom@debian.or.jp> Cc: Greg KH <greg@kroah.com> Signed-off-by:
-
GOTO Masanori authored
This patch lets i2c-dev.h include linux/compiler.h so that __user is defined. Signed-off-by:
-
GOTO Masanori authored
This patch cleans up asm-ppc64/byteorder.h to enable ___arch__swab16 and ___arch__swab32 which are marked TODO currently. It removes ___arch__swab64 because ppc64 does not have short instruction combinations for swab64, the recent gcc generates enough smart code that is equivalent to hand assembled code under my tests. Signed-off-by:
-
Sébastien Dugu authored
In the case of buffered AIO, in the aio retry path (aio_run_iocb), when the retry method returns EIOCBRETRY the kicked iocb is added to the context run list but is never queued onto the work queue. The request therefore is never completed. This patch fixes that by adding the appropriate call to aio_queue_work in aio_run_aiocb so that subsequent retries will be handled by the aio worker thread. Signed-off-by:
Sébastien Dugué <sebastien.dugue@bull.net> Acked-by:
Benjamin LaHaise <benjamin.c.lahaise@intel.com> Signed-off-by:
-
Christoph Hellwig authored
Looks like it sneaked back with the NFS ACL merge.. Signed-off-by:
Christoph Hellwig <hch@lst.de> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Neil Brown <neilb@cse.unsw.edu.au> Signed-off-by:
-
Adrian Bunk authored
This patch makes some needlessly global code static. Signed-off-by:
Adrian Bunk <bunk@stusta.de> Signed-off-by:
Armin Schindler <armin@melware.de> Signed-off-by:
Karsten Keil <kkeil@suse.de> Signed-off-by:
-
Pekka J Enberg authored
Signed-off-by:
Pekka Enberg <penberg@cs.helsinki.fi> Signed-off-by:
-
Andrew Morton authored
In file included from drivers/media/dvb/ttpci/av7110_hw.c:38: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type In file included from drivers/media/dvb/ttpci/av7110_v4l.c:36: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type In file included from drivers/media/dvb/ttpci/av7110_av.c:37: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type drivers/isdn/icn/icn.c:719:4: warning: #warning TODO test headroom or use skb->nb to flag ACK In file included from drivers/media/dvb/ttpci/av7110_ca.c:39: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type In file included from drivers/media/dvb/ttpci/av7110.c:41: include/linux/byteorder/swabb.h:96: warning: type qualifiers ignored on function return type include/linux/byteorder/swabb.h:110: warning: type qualifiers ignored on function return type Does declaring a function to return a const value actually mean something to gcc? Dunno. Kill it and replace sone `__inline__'s with `inline' too. Signed-off-by:
-
Andrew Morton authored
drivers/isdn/hisax/hfc4s8s_l1.c:317: warning: type qualifiers ignored on function return type drivers/isdn/hisax/hfc4s8s_l1.c:329: warning: type qualifiers ignored on function return type Signed-off-by:
-
Stanislaw W. Gruszka authored
Avoid race occurs when some process have open file descriptor for class device attributes and already firmware allocated memory are freed. Don't allow negative loading timeout. Signed-off-by:
Stanislaw W. Gruszka <stf_xl@wp.pl> Signed-off-by:
-
Emmanuel Colbus authored
This code uses the x86 (non-AMD-ELAN) value of CLOCK_TICK_RATE instead of CLOCK_TICK_RATE itself, which is wrong for other archs. Signed-off-by:
Emmanuel Colbus <emmanuel.colbus@ensimag.imag.fr> Signed-off-by:
-
Oleg Nesterov authored
As Steven Rostedt pointed out, there are 2 problems with ITIMER_REAL timers. 1. do_setitimer() does not call del_timer_sync() in case when the timer is not pending (it_real_value() returns 0). This is wrong, the timer may still be running, and it can rearm itself. 2. It calls del_timer_sync() with tsk->sighand->siglock held. This is deadlockable, because timer's handler needs this lock too. Signed-off-by:
Oleg Nesterov <oleg@tv-sign.ru> Acked-by:
Steven Rostedt <rostedt@goodmis.org> Cc: Ingo Molnar <mingo@elte.hu> Signed-off-by:
-
Lennert Buytenhek authored
Fix the same typo in the ixp4xx and ixp2000 watchdog drivers. Signed-off-by:
Lennert Buytenhek <buytenh+lkml@wantstofly.org> Signed-off-by:
-
Luc Van Oostenryck authored
In the setup function, the delay variable is initialized with ints[2], but ints is declared as: int ints[2]; Since the module parameter should correspond to: tipar=timeout,delay I suppose that the following patch fix the problem. Signed-off-by:
Luc Van Oostenryck <luc.vanoostenryck@looxix.net> Signed-off-by:
-
Luca Falavigna authored
Use msleep() in a few places. Signed-off-by:
Luca Falavigna <dktrkranz@gmail.com> Acked-by:
Ingo Molnar <mingo@elte.hu> Acked-by:
Jeff Garzik <jgarzik@pobox.com> Signed-off-by:
-
Wen-chien Jesse Sung authored
This up() should be down() instead. Signed-off-by:
Wen-chien Jesse Sung <jesse@cola.voip.idv.tw> Signed-off-by:
-
- 28 Jun, 2005 14 commits
-
-
Linus Torvalds authored
-
Linus Torvalds authored
-
Ingo Molnar authored
This patch tweaks idle thread setup semantics a bit: instead of setting NEED_RESCHED in init_idle(), we do an explicit schedule() before calling into cpu_idle(). This patch, while having no negative side-effects, enables wider use of cond_resched()s. (which might happen in the stock kernel too, but it's particulary important for voluntary-preempt) Signed-off-by:
-
Jens Axboe authored
Currently we cap request allocations at q->nr_requests, but we allow a batching io context to allocate up to 32 more (default setting). This can flood the queue with request allocations, with only a few batching processes. The real fix would be to limit the number of batchers, but as that isn't currently tracked, I suggest we just cap the maximum number of allocated requests to eg 50% over the limit. This was observed in real life, users typically see this as vmstat bo numbers going off the wall with seconds of no queueing afterwards. Behaviour this bursty is not beneficial. Signed-off-by:
Jens Axboe <axboe@suse.de> Signed-off-by:
-
Neil Brown authored
insert a missing bio_put when writting the md superblock. Without this we have a steady growth in the "bio" slab. Signed-off-by:
Neil Brown <neilb@cse.unsw.edu.au> Signed-off-by:
-
Alexey Dobriyan authored
Signed-off-by:
Alexey Dobriyan <adobriyan@gmail.com> Cc: Eric Biederman <ebiederm@xmission.com> Signed-off-by:
-
Andrew Morton authored
Signed-off-by:
-
Ben Dooks authored
Patch from Ben Dooks Fix the IRQ_LCD so that it is marked as valid since we no longer de-mux this in the main IRQ handler. Signed-off-by:
Ben Dooks <ben-linux@fluff.org> Signed-off-by:
Russell King <rmk+kernel@arm.linux.org.uk>
-
Tony Lindgren authored
Patch from Tony Lindgren This patch was suggested by RMK, and adds a warning on the accuracy of timekeeping when using dynamic tick on some platforms. Depending on the timer implementation, dynamic tick may affect the accuracy of timekeeping. Currently at least OMAP is known to have accurate timekeeping with dynamic tick. Signed-off-by:
Tony Lindgren <tony@atomide.com> Signed-off-by:
-
Ben Dooks authored
Patch from Ben Dooks This provides the s3c24xx audio platform data which can be supplied from any of the board specific drivers. Signed-off-by:
-
Richard Purdie authored
Patch from Richard Purdie Add functions to generate backtraces of both kernel and user processes which allows oprofile's call graphing functionality to be used on arm. This requires unstripped binaries/libs which use a frame pointer. Signed-off-by: Richard Purdie Signed-off-by:
Zwane Mwaikambo <zwane@arm.linux.org.uk> Signed-off-by:
-
Linus Torvalds authored
Merge head 'upstream-20050628-1' of rsync://rsync.kernel.org/pub/scm/linux/kernel/git/jgarzik/libata-dev
-
-
Greg Kroah-Hartman authored
Now that we have access to the whole MCFG table, let's properly use it for all pci device accesses (as that's what it is there for, some boxes don't put all the busses into one entry.) If, for some reason, the table is incorrect, we fallback to the "old style" of mmconfig accesses, namely, we just assume the first entry in the table is the one for us, and blindly use it. Signed-off-by:Greg Kroah-Hartman <gregkh@suse.de>
-
