- 19 Jul, 2007 40 commits
-
-
Mingming Cao authored
Looking at the current linus-git tree jbd_debug() define in include/linux/jbd2.h extern u8 journal_enable_debug; #define jbd_debug(n, f, a...) \ do { \ if ((n) <= journal_enable_debug) { \ printk (KERN_DEBUG "(%s, %d): %s: ", \ __FILE__, __LINE__, __FUNCTION__); \ printk (f, ## a); \ } \ } while (0) > fs/ext4/inode.c: In function âext4_write_inodeâ: > fs/ext4/inode.c:2906: warning: comparison is always true due to limited > range of data type > > fs/jbd2/recovery.c: In function âjbd2_journal_recoverâ: > fs/jbd2/recovery.c:254: warning: comparison is always true due to > limited range of data type > fs/jbd2/recovery.c:257: warning: comparison is always true due to > limited range of data type > > fs/jbd2/recovery.c: In function âjbd2_journal_skip_recoveryâ: > fs/jbd2/recovery.c:301: warning: comparison is always true due to > limited range of data type > Noticed all warnings are occurs when the debug level is 0. Then found the "jbd2: Move jbd2-debug file to debugfs" patch http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=0f49d5d019afa4e94253bfc92f0daca3badb990b changed the jbd2_journal_enable_debug from int type to u8, makes the jbd_debug comparision is always true when the debugging level is 0. Thus the compile warning occurs. Thought about changing the jbd2_journal_enable_debug data type back to int, but can't, because the jbd2-debug is moved to debug fs, where calling debugfs_create_u8() to create the debugfs entry needs the value to be u8 type. Even if we changed the data type back to int, the code is still buggy, kernel should not print jbd2 debug message if the jbd2_journal_enable_debug is set to 0. But this is not the case. The fix is change the level of debugging to 1. The same should fixed in ext3/JBD, but currently ext3 jbd-debug via /proc fs is broken, so we probably should fix it all together. Signed-off-by:Mingming Cao <cmm@us.ibm.com> Cc: Jeff Garzik <jeff@garzik.org> Cc: Theodore Tso <tytso@mit.edu> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Andy Whitcroft authored
This version brings a number of new checks, and a number of bug fixes. Of note: - warnings for multiple assignments per line - warnings for multiple declarations per line - checks for single statement blocks with braces This patch includes an update for feature-removal-schedule.txt to better target checks. Andy Whitcroft (12): Version: 0.08 only apply printk checks where there is a string literal allow suppression of errors for when no patch is found warn about multiple assignments warn on declaration of multiple variables check for kfree() with needless null check check for single statement braced blocks check for aggregate initialisation on the next line handle the => operator check for spaces between function name and open parenthesis move to explicit Check: entries in feature-removal-schedule.txt handle pointer attributes Signed-off-by:Andy Whitcroft <apw@shadowen.org> Signed-off-by:
-
Rolf Eike Beer authored
Signed-off-by:
Rolf Eike Beer <eike-kernel@sf-tec.de> Signed-off-by:
-
Ingo Molnar authored
Update the MAINTAINERS file: Thomas Gleixner has been the de-facto maintainer of POSIX timers and clocks for quite some time. Signed-off-by:
Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Adrian Bunk authored
Signed-off-by:
Adrian Bunk <bunk@stusta.de> Cc: Tom Zanussi <zanussi@us.ibm.com> Signed-off-by:
-
Kawai, Hidehiro authored
This patch adds the documentation for /proc/<pid>/coredump_filter. Signed-off-by:
Hidehiro Kawai <hidehiro.kawai.ez@hitachi.com> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Cc: David Howells <dhowells@redhat.com> Cc: Hugh Dickins <hugh@veritas.com> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: "Randy.Dunlap" <rdunlap@xenotime.net> Signed-off-by:
-
Kawai, Hidehiro authored
This patch enables core dump filtering for ELF-FDPIC-formatted core file. Signed-off-by:
-
Kawai, Hidehiro authored
This patch removes an unused argument from elf_fdpic_dump_segments(). Signed-off-by:
-
Kawai, Hidehiro authored
This patch enables core dump filtering for ELF-formatted core file. Signed-off-by:
-
Kawai, Hidehiro authored
This patch adds an interface to set/reset flags which determines each memory segment should be dumped or not when a core file is generated. /proc/<pid>/coredump_filter file is provided to access the flags. You can change the flag status for a particular process by writing to or reading from the file. The flag status is inherited to the child process when it is created. Signed-off-by:
-
Kawai, Hidehiro authored
This patch changes mm_struct.dumpable to a pair of bit flags. set_dumpable() converts three-value dumpable to two flags and stores it into lower two bits of mm_struct.flags instead of mm_struct.dumpable. get_dumpable() behaves in the opposite way. [akpm@linux-foundation.org: export set_dumpable] Signed-off-by:
-
Kawai, Hidehiro authored
This patch series is version 5 of the core dump masking feature, which controls which VMAs should be dumped based on their memory types and per-process flags. I adopted most of Andrew's suggestion at the previous version. He also suggested using system call instead of /proc/<pid>/ interface, I decided to use the latter continuously because adding new system call with pid argument will give a big impact on the kernel. You can access the per-process flags via /proc/<pid>/coredump_filter interface. coredump_filter represents a bitmask of memory types, and if a bit is set, VMAs of corresponding memory type are written into a core file when the process is dumped. The bitmask is inherited from the parent process when a process is created. The original purpose is to avoid longtime system slowdown when a number of processes which share a huge shared memory are dumped at the same time. To achieve this purpose, this patch series adds an ability to suppress dumping anonymous shared memory for specified processes. In this version, three other memory types are also supported. Here are the coredump_filter bits: bit 0: anonymous private memory bit 1: anonymous shared memory bit 2: file-backed private memory bit 3: file-backed shared memory The default value of coredump_filter is 0x3. This means the new core dump routine has the same behavior as conventional behavior by default. In this version, coredump_filter bits and mm.dumpable are merged into mm.flags, and it is accessed by atomic bitops. The supported core file formats are ELF and ELF-FDPIC. ELF has been tested, but ELF-FDPIC has not been built and tested because I don't have the test environment. This patch limits a value of suid_dumpable sysctl to the range of 0 to 2. Signed-off-by:
-
Randy Dunlap authored
Remove include/linux/rmap.h from kernel-api.tmpl since it no longer contains kernel-doc. Fixes this warning: Warning(linux-2.6.22//include/linux/rmap.h): no structured comments found Signed-off-by:
Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by:
-
Randy Dunlap authored
If a parameter description begins with a '.', this indicates a "request" for "man" mode output (*roff), so it needs special handling. Problem case is in include/asm-i386/atomic.h for function atomic_add_unless(): * @u: ...unless v is equal to u. This parameter description is currently not printed in man mode output. [akpm@linux-foundation.org: cleanup] Signed-off-by:
-
Randy Dunlap authored
Strip C99-style comments from the input stream. /*...*/ comments are already stripped. C99 comments confuse the kernel-doc script. Also update some comments. Signed-off-by:
-
Randy Dunlap authored
Fix kernel-doc warning: Warning(linux-2.6.22-rc2-git2/include/linux/skbuff.h:316): No description found for parameter '}' which is caused by nested anonymous structs/unions ending with: }; }; Signed-off-by:
-
Randy Dunlap authored
Add kernel-doc tools info in Makefile. Signed-off-by:
-
Josef 'Jeff' Sipek authored
Signed-off-by:
Josef 'Jeff' Sipek <jsipek@cs.sunysb.edu> Cc: Al Viro <viro@zeniv.linux.org.uk> Acked-by:
Christoph Hellwig <hch@lst.de> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Cc: Neil Brown <neilb@suse.de> Cc: Michael Halcrow <mhalcrow@us.ibm.com> Signed-off-by:
-
Josef 'Jeff' Sipek authored
Signed-off-by:
-
Josef 'Jeff' Sipek authored
use vfs_path_lookup instead of open-coding the necessary functionality. Signed-off-by:
NeilBrown <neilb@suse.de> Cc: Al Viro <viro@zeniv.linux.org.uk> Acked-by:
-
Josef 'Jeff' Sipek authored
use vfs_path_lookup instead of open-coding the necessary functionality. Signed-off-by:
Trond Myklebust <Trond.Myklebust@netapp.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Acked-by:
-
Josef 'Jeff' Sipek authored
Stackable file systems, among others, frequently need to lookup paths or path components starting from an arbitrary point in the namespace (identified by a dentry and a vfsmount). Currently, such file systems use lookup_one_len, which is frowned upon [1] as it does not pass the lookup intent along; not passing a lookup intent, for example, can trigger BUG_ON's when stacking on top of NFSv4. The first patch introduces a new lookup function to allow lookup starting from an arbitrary point in the namespace. This approach has been suggested by Christoph Hellwig [2]. The second patch changes sunrpc to use vfs_path_lookup. The third patch changes nfsctl.c to use vfs_path_lookup. The fourth patch marks link_path_walk static. The fifth, and last patch, unexports path_walk because it is no longer unnecessary to call it directly, and using the new vfs_path_lookup is cleaner. For example, the following snippet of code, looks up "some/path/component" in a directory pointed to by parent_{dentry,vfsmnt}: err = vfs_path_lookup(parent_dentry, parent_vfsmnt, "some/path/component", 0, &nd); if (!err) { /* exits */ ... /* once done, release the references */ path_release(&nd); } else if (err == -ENOENT) { /* doesn't exist */ } else { /* other error */ } VFS functions such as lookup_create can be used on the nameidata structure to pass the create intent to the file system. Signed-off-by: -
Ollie Wild authored
Remove the arg+env limit of MAX_ARG_PAGES by copying the strings directly from the old mm into the new mm. We create the new mm before the binfmt code runs, and place the new stack at the very top of the address space. Once the binfmt code runs and figures out where the stack should be, we move it downwards. It is a bit peculiar in that we have one task with two mm's, one of which is inactive. [a.p.zijlstra@chello.nl: limit stack size] Signed-off-by:
Ollie Wild <aaw@google.com> Signed-off-by:
Peter Zijlstra <a.p.zijlstra@chello.nl> Cc: <linux-arch@vger.kernel.org> Cc: Hugh Dickins <hugh@veritas.com> [bunk@stusta.de: unexport bprm_mm_init] Signed-off-by:
-
Peter Zijlstra authored
The purpose of audit_bprm() is to log the argv array to a userspace daemon at the end of the execve system call. Since user-space hasn't had time to run, this array is still in pristine state on the process' stack; so no need to copy it, we can just grab it from there. In order to minimize the damage to audit_log_*() copy each string into a temporary kernel buffer first. Currently the audit code requires that the full argument vector fits in a single packet. So currently it does clip the argv size to a (sysctl) limit, but only when execve auditing is enabled. If the audit protocol gets extended to allow for multiple packets this check can be removed. Signed-off-by:
-
Peter Zijlstra authored
New arch macro STACK_TOP_MAX it gives the larges valid stack address for the architecture in question. It differs from STACK_TOP in that it will not distinguish between personalities but will always return the largest possible address. This is used to create the initial stack on execve, which we will move down to the proper location once the binfmt code has figured out where that is. Signed-off-by:
-
Fenghua Yu authored
Currently most of the per cpu data, which is accessed by different cpus, has a ____cacheline_aligned_in_smp attribute. Move all this data to the new per cpu shared data section: .data.percpu.shared_aligned. This will seperate the percpu data which is referenced frequently by other cpus from the local only percpu data. Signed-off-by:
Fenghua Yu <fenghua.yu@intel.com> Acked-by:
Suresh Siddha <suresh.b.siddha@intel.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Christoph Lameter <clameter@sgi.com> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by:
-
Fenghua Yu authored
per cpu data section contains two types of data. One set which is exclusively accessed by the local cpu and the other set which is per cpu, but also shared by remote cpus. In the current kernel, these two sets are not clearely separated out. This can potentially cause the same data cacheline shared between the two sets of data, which will result in unnecessary bouncing of the cacheline between cpus. One way to fix the problem is to cacheline align the remotely accessed per cpu data, both at the beginning and at the end. Because of the padding at both ends, this will likely cause some memory wastage and also the interface to achieve this is not clean. This patch: Moves the remotely accessed per cpu data (which is currently marked as ____cacheline_aligned_in_smp) into a different section, where all the data elements are cacheline aligned. And as such, this differentiates the local only data and remotely accessed data cleanly. Signed-off-by:
-
Michael Ellerman authored
I realise jprobes are a razor-blades-included type of interface, but that doesn't mean we can't try and make them safer to use. This guy I know once wrote code like this: struct jprobe jp = { .kp.symbol_name = "foo", .entry = "jprobe_foo" }; And then his kernel exploded. Oops. This patch adds an arch hook, arch_deref_entry_point() (I don't like it either) which takes the void * in a struct jprobe, and gives back the text address that it represents. We can then use that in register_jprobe() to check that the entry point we're passed is actually in the kernel text, rather than just some random value. Signed-off-by:Michael Ellerman <michael@ellerman.id.au> Cc: Prasanna S Panchamukhi <prasanna@in.ibm.com> Acked-by:
Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> Cc: David S. Miller <davem@davemloft.net> Signed-off-by:
-
Michael Ellerman authored
AFAICT now that jprobe.entry is a void *, JPROBE_ENTRY doesn't do anything useful - so remove it .. I've left a do-nothing version so that out-of-tree jprobes code will still compile without modifications. Signed-off-by:
-
Michael Ellerman authored
Currently jprobe.entry is a kprobe_opcode_t *, but that's a lie. On some platforms it doesn't point to an opcode at all, it points to a function descriptor. It's really a pointer to something that the arch code can turn into a function entry point. And that's what actually happens, none of the generic code ever looks at jprobe.entry, it's only ever dereferenced by arch code. So just make it a void *. Signed-off-by:
-
Fengguang Wu authored
Rename some file_ra_state variables and remove some accessors. It results in much simpler code. Kudos to Rusty! Signed-off-by:
Fengguang Wu <wfg@mail.ustc.edu.cn> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by:
-
Rusty Russell authored
Split ondemand readahead interface into two functions. I think this makes it a little clearer for non-readahead experts (like Rusty). Internally they both call ondemand_readahead(), but the page argument is changed to an obvious boolean flag. Signed-off-by:
Rusty Russell <rusty@rustcorp.com.au> Signed-off-by:
-
Fengguang Wu authored
Share the same page flag bit for PG_readahead and PG_reclaim. One is used only on file reads, another is only for emergency writes. One is used mostly for fresh/young pages, another is for old pages. Combinations of possible interactions are: a) clear PG_reclaim => implicit clear of PG_readahead it will delay an asynchronous readahead into a synchronous one it actually does _good_ for readahead: the pages will be reclaimed soon, it's readahead thrashing! in this case, synchronous readahead makes more sense. b) clear PG_readahead => implicit clear of PG_reclaim one(and only one) page will not be reclaimed in time it can be avoided by checking PageWriteback(page) in readahead first c) set PG_reclaim => implicit set of PG_readahead will confuse readahead and make it restart the size rampup process it's a trivial problem, and can mostly be avoided by checking PageWriteback(page) first in readahead d) set PG_readahead => implicit set of PG_reclaim PG_readahead will never be set on already cached pages. PG_reclaim will always be cleared on dirtying a page. so not a problem. In summary, a) we get better behavior b,d) possible interactions can be avoided c) racy condition exists that might affect readahead, but the chance is _really_ low, and the hurt on readahead is trivial. Compound pages also use PG_reclaim, but for now they do not interact with reclaim/readahead code. Signed-off-by:
-
Fengguang Wu authored
Pass real splice size to page_cache_readahead_ondemand(). The splice code works in chunks of 16 pages internally. The readahead code should be told of the overall splice size, instead of the internal chunk size. Otherwize bad things may happen. Imagine some 17-page random splice reads. The code before this patch will result in two readahead calls: readahead(16); readahead(1); That leads to one 16-page I/O and one 32-page I/O: one extra I/O and 31 readahead miss pages. Signed-off-by:
-
Fengguang Wu authored
Move synchronous page_cache_readahead_ondemand() call out of splice loop. This avoids one pointless page allocation/insertion in case of non-zero ra_pages, or many pointless readahead calls in case of zero ra_pages. Note that if a user sets ra_pages to less than PIPE_BUFFERS=16 pages, he will not get expected readahead behavior anyway. The splice code works in batches of 16 pages, which can be taken as another form of synchronous readahead. Signed-off-by:
-
Fengguang Wu authored
Remove the old readahead algorithm. Signed-off-by:
-
Fengguang Wu authored
Convert ext3/ext4 dir reads to use on-demand readahead. Readahead for dirs operates _not_ on file level, but on blockdev level. This makes a difference when the data blocks are not continuous. And the read routine is somehow opaque: there's no handy info about the status of current page. So a simplified call scheme is employed: to call into readahead whenever the current page falls out of readahead windows. Signed-off-by:
-
Fengguang Wu authored
Convert splice reads to use on-demand readahead. Signed-off-by:
-
Fengguang Wu authored
Convert filemap reads to use on-demand readahead. The new call scheme is to - call readahead on non-cached page - call readahead on look-ahead page - update prev_index when finished with the read request Signed-off-by:
-
Fengguang Wu authored
This is a minimal readahead algorithm that aims to replace the current one. It is more flexible and reliable, while maintaining almost the same behavior and performance. Also it is full integrated with adaptive readahead. It is designed to be called on demand: - on a missing page, to do synchronous readahead - on a lookahead page, to do asynchronous readahead In this way it eliminated the awkward workarounds for cache hit/miss, readahead thrashing, retried read, and unaligned read. It also adopts the data structure introduced by adaptive readahead, parameterizes readahead pipelining with `lookahead_index', and reduces the current/ahead windows to one single window. HEURISTICS The logic deals with four cases: - sequential-next found a consistent readahead window, so push it forward - random standalone small read, so read as is - sequential-first create a new readahead window for a sequential/oversize request - lookahead-clueless hit a lookahead page not associated with the readahead window, so create a new readahead window and ramp it up In each case, three parameters are determined: - readahead index: where the next readahead begins - readahead size: how much to readahead - lookahead size: when to do the next readahead (for pipelining) BEHAVIORS The old behaviors are maximally preserved for trivial sequential/random reads. Notable changes are: - It no longer imposes strict sequential checks. It might help some interleaved cases, and clustered random reads. It does introduce risks of a random lookahead hit triggering an unexpected readahead. But in general it is more likely to do good than to do evil. - Interleaved reads are supported in a minimal way. Their chances of being detected and proper handled are still low. - Readahead thrashings are better handled. The current readahead leads to tiny average I/O sizes, because it never turn back for the thrashed pages. They have to be fault in by do_generic_mapping_read() one by one. Whereas the on-demand readahead will redo readahead for them. OVERHEADS The new code reduced the overheads of - excessively calling the readahead routine on small sized reads (the current readahead code insists on seeing all requests) - doing a lot of pointless page-cache lookups for small cached files (the current readahead only turns itself off after 256 cache hits, unfortunately most files are < 1MB, so never see that chance) That accounts for speedup of - 0.3% on 1-page sequential reads on sparse file - 1.2% on 1-page cache hot sequential reads - 3.2% on 256-page cache hot sequential reads - 1.3% on cache hot `tar /lib` However, it does introduce one extra page-cache lookup per cache miss, which impacts random reads slightly. That's 1% overheads for 1-page random reads on sparse file. PERFORMANCE The basic benchmark setup is - 2.6.20 kernel with on-demand readahead - 1MB max readahead size - 2.9GHz Intel Core 2 CPU - 2GB memory - 160G/8M Hitachi SATA II 7200 RPM disk The benchmarks show that - it maintains the same performance for trivial sequential/random reads - sysbench/OLTP performance on MySQL gains up to 8% - performance on readahead thrashing gains up to 3 times iozone throughput (KB/s): roughly the same ========================================== iozone -c -t1 -s 4096m -r 64k 2.6.20 on-demand gain first run " Initial write " 61437.27 64521.53 +5.0% " Rewrite " 47893.02 48335.20 +0.9% " Read " 62111.84 62141.49 +0.0% " Re-read " 62242.66 62193.17 -0.1% " Reverse Read " 50031.46 49989.79 -0.1% " Stride read " 8657.61 8652.81 -0.1% " Random read " 13914.28 13898.23 -0.1% " Mixed workload " 19069.27 19033.32 -0.2% " Random write " 14849.80 14104.38 -5.0% " Pwrite " 62955.30 65701.57 +4.4% " Pread " 62209.99 62256.26 +0.1% second run " Initial write " 60810.31 66258.69 +9.0% " Rewrite " 49373.89 57833.66 +17.1% " Read " 62059.39 62251.28 +0.3% " Re-read " 62264.32 62256.82 -0.0% " Reverse Read " 49970.96 50565.72 +1.2% " Stride read " 8654.81 8638.45 -0.2% " Random read " 13901.44 13949.91 +0.3% " Mixed workload " 19041.32 19092.04 +0.3% " Random write " 14019.99 14161.72 +1.0% " Pwrite " 64121.67 68224.17 +6.4% " Pread " 62225.08 62274.28 +0.1% In summary, writes are unstable, reads are pretty close on average: access pattern 2.6.20 on-demand gain Read 62085.61 62196.38 +0.2% Re-read 62253.49 62224.99 -0.0% Reverse Read 50001.21 50277.75 +0.6% Stride read 8656.21 8645.63 -0.1% Random read 13907.86 13924.07 +0.1% Mixed workload 19055.29 19062.68 +0.0% Pread 62217.53 62265.27 +0.1% aio-stress: roughly the same ============================ aio-stress -l -s4096 -r128 -t1 -o1 knoppix511-dvd-cn.iso aio-stress -l -s4096 -r128 -t1 -o3 knoppix511-dvd-cn.iso 2.6.20 on-demand delta sequential 92.57s 92.54s -0.0% random 311.87s 312.15s +0.1% sysbench fileio: roughly the same ================================= sysbench --test=fileio --file-io-mode=async --file-test-mode=rndrw \ --file-total-size=4G --file-block-size=64K \ --num-threads=001 --max-requests=10000 --max-time=900 run threads 2.6.20 on-demand delta first run 1 59.1974s 59.2262s +0.0% 2 58.0575s 58.2269s +0.3% 4 48.0545s 47.1164s -2.0% 8 41.0684s 41.2229s +0.4% 16 35.8817s 36.4448s +1.6% 32 32.6614s 32.8240s +0.5% 64 23.7601s 24.1481s +1.6% 128 24.3719s 23.8225s -2.3% 256 23.2366s 22.0488s -5.1% second run 1 59.6720s 59.5671s -0.2% 8 41.5158s 41.9541s +1.1% 64 25.0200s 23.9634s -4.2% 256 22.5491s 20.9486s -7.1% Note that the numbers are not very stable because of the writes. The overall performance is close when we sum all seconds up: sum all up 495.046s 491.514s -0.7% sysbench oltp (trans/sec): up to 8% gain ======================================== sysbench --test=oltp --oltp-table-size=10000000 --oltp-read-only \ --mysql-socket=/var/run/mysqld/mysqld.sock \ --mysql-user=root --mysql-password=readahead \ --num-threads=064 --max-requests=10000 --max-time=900 run 10000-transactions run threads 2.6.20 on-demand gain 1 62.81 64.56 +2.8% 2 67.97 70.93 +4.4% 4 81.81 85.87 +5.0% 8 94.60 97.89 +3.5% 16 99.07 104.68 +5.7% 32 95.93 104.28 +8.7% 64 96.48 103.68 +7.5% 5000-transactions run 1 48.21 48.65 +0.9% 8 68.60 70.19 +2.3% 64 70.57 74.72 +5.9% 2000-transactions run 1 37.57 38.04 +1.3% 2 38.43 38.99 +1.5% 4 45.39 46.45 +2.3% 8 51.64 52.36 +1.4% 16 54.39 55.18 +1.5% 32 52.13 54.49 +4.5% 64 54.13 54.61 +0.9% That's interesting results. Some investigations show that - MySQL is accessing the db file non-uniformly: some parts are more hot than others - It is mostly doing 4-page random reads, and sometimes doing two reads in a row, the latter one triggers a 16-page readahead. - The on-demand readahead leaves many lookahead pages (flagged PG_readahead) there. Many of them will be hit, and trigger more readahead pages. Which might save more seeks. - Naturally, the readahead windows tend to lie in hot areas, and the lookahead pages in hot areas is more likely to be hit. - The more overall read density, the more possible gain. That also explains the adaptive readahead tricks for clustered random reads. readahead thrashing: 3 times better =================================== We boot kernel with "mem=128m single", and start a 100KB/s stream on every second, until reaching 200 streams. max throughput min avg I/O size 2.6.20: 5MB/s 16KB on-demand: 15MB/s 140KB Signed-off-by:
-
