search for: 520mb

...e Samba client, both on the same exact ZFS dataset 2) The ZFS dataset only writes async, converting sync to async writes at all times 3) That same dataset being shared through Samba, also only performs async writes (strict sync = no) 4) With these settings, on the host the benchmarks tops out at 520MB/s. On the Samba client (again, writing to the same end destination) I get 40 MB/s tops. I feel like those benchmark scores should be much closer given the above test. I've ruled out the network being an issue since a) they're hosted on the same host via a Linux bridge (which means it neve...

...the host on this dataset, it will use RAM to do the write, drastically speeding up random writes. The below benchmark command on the ZFS host: fio --direct=1 --sync=1 --rw=randwrite --bs=4k --numjobs=1 --iodepth=1 --runtime=60 --group_reporting --name=sambatest --time_based Has an average speed of 520MB/s (which is the maximum speed of my SATA SSD). Despite requesting a sync write, ZFS turns it in an async write, dramatically speeding it up. Clearly the results are great when I directly benchmark from the host into the sync=disabled ZFS dataset. But this doesn't translate to real-world Samba s...

...check remotely. dmesg's first lines: Linux version 2.6.18-164.el5xen (mockbuild at builder16.centos.org) (gcc version 4.1.2 20080704 (Red Hat 4.1.2-46)) #1 SMP Thu Sep 3 04:47:32 EDT 2009 BIOS-provided physical RAM map: Xen: 0000000000000000 - 0000000020800000 (usable) 0MB HIGHMEM available. 520MB LOWMEM available. Using x86 segment limits to approximate NX protection On node 0 totalpages: 133120 DMA zone: 133120 pages, LIFO batch:31 DMI 2.3 present. ACPI: RSDP (v000 IBM ) @ 0x000fdfc0 ACPI: RSDT (v001 IBM SERONYXP 0x00001000 IBM 0x45444f43) @ 0x7ffdff...

...c` side-vectors for 7M line >> table entries. The cost of these is ~180B each, for another >> ~600MB. >> >> If we integrate a side-table of `MDLineTable`s into its uniquing, >> the overhead is only ~12B / line table entry, or ~80MB. This saves >> 520MB. >> >> This is somewhat perpendicular to redesigning the metadata format, >> but IMO it's worth doing as soon as it's possible. >> >> 4. Create `GenericDebugMDNode`, a transitional subclass of `MDUser` >> through an intermediate class `DebugMDNod...

...ws that we have 3.5M entries in the `DebugLoc` side-vectors for 7M line table entries. The cost of these is ~180B each, for another ~600MB. If we integrate a side-table of `MDLineTable`s into its uniquing, the overhead is only ~12B / line table entry, or ~80MB. This saves 520MB. This is somewhat perpendicular to redesigning the metadata format, but IMO it's worth doing as soon as it's possible. 4. Create `GenericDebugMDNode`, a transitional subclass of `MDUser` through an intermediate class `DebugMDNode` with an allocation-time-optional `Callbac...

...onfig file "domain1.xm". Started domain domain1 Linux version 2.6.16-xen (root@localhost.localdomain) (gcc version 3.4.6 20060404 (Red Hat 3.4.6-3)) #1 SMP Fri Sep 8 18:40:45 MST 2006 BIOS-provided physical RAM map: Xen: 0000000000000000 - 0000000020000000 (usable) 0MB HIGHMEM available. 520MB LOWMEM available. ACPI in unprivileged domain disabled IRQ lockup detection disabled Built 1 zonelists Kernel command line: ip=:1.2.3.4::::eth0:dhcp root=/dev/sda1 ro 3 Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done. Initializing CPU#0 PID hash tab...

...ch fork+execs it > when it needs to symbolize a stack trace. It successfully symbolizes ASan report for model Chromium crash. If Chromium is build with -gline-tables-only (500Mb), > it runs in less than a second (most time is spent on loading the binary in memory), (unsurprisingly) uses about 520Mb of memory and provides stack trace > with file/line/column and full function names. > > Do you think it is the way to go for us now? > Since no one else responded that I could see... I think this sounds great :) -eric

...me exact ZFS dataset > > 2) The ZFS dataset only writes async, converting sync to async writes at all times > > 3) That same dataset being shared through Samba, also only performs async writes (strict sync = no) > > 4) With these settings, on the host the benchmarks tops out at 520MB/s. On the Samba client (again, writing to the same end destination) I get 40 MB/s tops. I feel like those benchmark scores should be much closer given the above test. > > > I've ruled out the network being an issue since a) they're hosted on the same host via a Linux bridge (whic...

...> > 2. The ZFS dataset only writes async, converting sync to async writes at all times > > > > 3. That same dataset being shared through Samba, also only performs async writes (strict sync = no) > > > > 4. With these settings, on the host the benchmarks tops out at 520MB/s. On the Samba client (again, writing to the same end destination) I get 40 MB/s tops. I feel like those benchmark scores should be much closer given the above test. > > > > > > I've ruled out the network being an issue since a) they're hosted on the same host via a Linux...

...is dataset, it will use RAM to do the write, drastically speeding up random writes. > The below benchmark command on the ZFS host: > fio --direct=1 --sync=1 --rw=randwrite --bs=4k --numjobs=1 --iodepth=1 --runtime=60 --group_reporting --name=sambatest --time_based > Has an average speed of 520MB/s (which is the maximum speed of my SATA SSD). Despite requesting a sync write, ZFS turns it in an async write, dramatically speeding it up. Clearly the results are great when I directly benchmark from the host into the sync=disabled ZFS dataset. But this doesn't translate to real-world Samba s...

...t;> table entries. The cost of these is ~180B each, for another > >> ~600MB. > >> > >> If we integrate a side-table of `MDLineTable`s into its uniquing, > >> the overhead is only ~12B / line table entry, or ~80MB. This saves > >> 520MB. > >> > >> This is somewhat perpendicular to redesigning the metadata format, > >> but IMO it's worth doing as soon as it's possible. > >> > >> 4. Create `GenericDebugMDNode`, a transitional subclass of `MDUser` > >> through...

...ies in the `DebugLoc` side-vectors for 7M line > table entries. The cost of these is ~180B each, for another > ~600MB. > > If we integrate a side-table of `MDLineTable`s into its uniquing, > the overhead is only ~12B / line table entry, or ~80MB. This saves > 520MB. > > This is somewhat perpendicular to redesigning the metadata format, > but IMO it's worth doing as soon as it's possible. > > 4. Create `GenericDebugMDNode`, a transitional subclass of `MDUser` > through an intermediate class `DebugMDNode` with an >...

On Thu, Jun 21, 2012 at 12:52 PM, Chandler Carruth <chandlerc at google.com>wrote: > On Thu, Jun 21, 2012 at 1:42 AM, Kostya Serebryany <kcc at google.com> wrote: > >> Can we alter the build system so that when building a run-time library it >> modifies all .cpp files like this: >> namespace FOO { >> <file body> >> } >> This

...d domain ptt Linux version 2.6.18-6-xen-686 (Debian 2.6.18.dfsg.1-18etch1) (waldi at debian.org) (gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)) #1 SMP Sun Feb 10 22:43:13 UTC 2008 BIOS-provided physical RAM map: Xen: 0000000000000000 - 0000000020800000 (usable) 0MB HIGHMEM available. 520MB LOWMEM available. NX (Execute Disable) protection: active ACPI in unprivileged domain disabled Built 1 zonelists. Total pages: 133120 Kernel command line: root=/dev/sda1 ro Enabling fast FPU save and restore... done. Enabling unmasked SIMD FPU exception support... done. Initializing CPU#0 PID has...

(resurrecting the thread, as much is discussed here already) Formulating Kostya's suggestion: What do you think of compiling LLVM sources into ASan/TSan runtime by just taking the library sources, providing custom compiler (target) flags *and* a flag "-Dllvm=__sanitizer_llvm"? Yeah, it's hacky and applicable to LLVM libs, but OTOH we don't plan to use smth else for now (in