IRC log search results for memcpy

• 2016-12-13 (Tuesday):

  shorne	stekern: olofk: nor next series of changes I am thinking the memcpy memset funcs, some change from sebastien macke and getting our gpio driver into staging (and updated to latest gpio api)	05:37

• 2016-09-14 (Wednesday):

  ZipCPU	Let's see ... I annotated the disassembly of the memcpy function.	15:04
  ZipCPU	olofk: Here's an annotated disassembly of the memcpy function: https://justpaste.it/yci0	15:11
  _franck_	I started to port this one to or1k assembly some times ago: http://git.musl-libc.org/cgit/musl/tree/src/string/memcpy.c	15:13
  ZipCPU	I'm now using -O3 and I've verified that the strcmp, memcpy, and strcpy calls are all the optimized newlib versions	16:26
  shorne	pick eb6b230 openrisc: Add optimized memcpy routine	18:32
  shorne	_franck_: you can look at my memcpy routine, I also send it to the kernel list and got some response on it	18:34

• 2016-08-28 (Sunday):

  ZipCPU	OpenRISC has no memcpy() or strcmp() instructions like the VAX did, so ... you'd expect a bit of a difference.	08:37
  ZipCPU	kc5tja: Let's discuss loop unrolling for a moment.  When measuring the ZipCPU's performance, I unrolled the loops of the strcmp, strcpy, and memcpy manually.	14:38

• 2016-05-20 (Friday):

  zump	The HW guys use some sort of system called AXI. I have to somehow remove that in each verilog block and replace it with, you know, memcpy.	10:18

• 2016-03-24 (Thursday):

  shorne	so... from the response from Jonas I spent the past day and a half looking into adding a memcpy builtin to the openrisc gcc backent	08:06

• 2016-03-23 (Wednesday):

  olofk	I've been thinking about that too. It's a bit annoying that we have to add memcpy and friends to all OSes and libc implementations. A single one in gcc (+llvm I guess) would be nice	08:59

• 2016-03-21 (Monday):

  shorne	olofk: I did some more testing on my memcpy routines. Intestingly doing the microblaze type analigned word copies is not that much better than byte copies it seems	02:59
  shorne	anyway some numbers:  memcpy word copies (supporting non-alignment) with loop unrolls, during linux boot avg cycles 1880.3	03:05
  shorne	memcpy  byte copies, during linux boot avg cycles 7603.0	03:06
  shorne	The memcpy stuff is here:	03:46
  olofk	Or did you mean memcpy?	05:31
  shorne	olofk: thats what it loks like. sorry memcpy...	05:31
  olofk	Would be great to get that upstreamed as well. memcpy will probably improve things during runtime as well	05:32
  wallento	I get memcpy undefined :)	06:12
  wallento	ERROR: "memcpy" [crypto/sha256_generic.ko] undefined!	06:17
  wallento	ERROR: "memcpy" [crypto/jitterentropy_rng.ko] undefined!	06:17
  wallento	ERROR: "memcpy" [crypto/hmac.ko] undefined!	06:17
  wallento	ERROR: "memcpy" [crypto/echainiv.ko] undefined!	06:17
  wallento	ERROR: "memcpy" [crypto/drbg.ko] undefined!	06:17

• 2016-03-20 (Sunday):

  shorne	I have been playing with memcpy optimization for linux... something is bit strange	04:57
  shorne	nice, finally fixed some stupid bugs in my memcpy	06:51

• 2016-03-18 (Friday):

  shorne	olofk: you did memset, did you have a look at memcpy?	17:22
  olofk	shorne: I never got around to do memcpy. It was slightly more complicated than memset, so I let it be	17:26
  shorne	microblaze. this one http://lxr.free-electrons.com/source/arch/microblaze/lib/memcpy.c#L36?	17:30
  olofk	Four times faster memcpy should probably help a bit in many cases	18:12
  shorne	So I am playing with memcpy, for now I think the microblaze one is a bit too much and I just dont want to copy. Also it doesnt unroll loops	19:12

• 2015-03-29 (Sunday):

  dalias	(for example if we wanted to compare different memcpys)	16:44
  olofk	But for something like memcpy, you would probably have to factor in memory latency as well.	16:48

• 2015-03-13 (Friday):

  sheridp	I'm hitting an undefined reference to memcpy as a result of the following line: int colList [4] = {0,1,2,3};	23:27
  sheridp	does anyone know where to grab memcpy from?	23:27

• 2015-02-06 (Friday):

  _franck__	olofk: while you are working on this you should try the C version of memcpy from  musl	12:48
  _franck__	it's the same king of algorithm you can find in arch/microblaze/lib/memcpy.c	12:49

• 2015-02-03 (Tuesday):

  _franck__	dalias: olofk : what I started to do some times ago is to optimize the assembly version of memcpy from musl	09:02
  olofk	_franck__: You do memcpy and I do memset then :)	09:03

• 2014-11-14 (Friday):

  olofk	IIRC memcpy and memset where the biggest contributors. I did some experiments to optimize that, and by stealing the memset code from Microblaze, it dropped down from the top 10	08:28
  olofk	Can you use DMA for memcpy easily under Linux btw?	08:32
  olofk	True. I haven't got a clue how large the common memcpy is	08:33
  poke53282	http://stackoverflow.com/questions/25521422/dma-memcpy-operation-in-linux	08:39
  olofk	And DMA for memcpy seems not worth the trouble, even though I think that our break-even block size would be much smaller than 1MB since we have smaller caches than an x86	08:45

• 2014-10-16 (Thursday):

  maxpaln	we are trying to assess performance on a memcpy. The way it is currently implemented (possibly by the C code or more likely by the assembler as interpretted by the compiler) is at a word-by-word basis. So copying a block of, say, 1KB, requires a 1000x loop that reads a byte and writes it somewhere else.	13:05
  _franck__	maxpaln: are talking about memcpy from Linux or from the libc ?	13:09

• 2014-08-07 (Thursday):

  poke53282	I will figure out the correct code line tomorrow. It must be one before the memcpy (case REL_COPY:) if I remember correctly the disasm code.	06:13

• 2014-06-12 (Thursday):

  _franck__	I started something but I switched to something else. I compiled musl memcpy, disassembled it and started to optimize it in order to put the result in the kernel	12:59

• 2014-05-15 (Thursday):

  dalias	_franck_, fyi, i have a pretty generic high-performance "C" implementation of memcpy that does the word shuffling for alignment, as part of musl libc	00:46
  dalias	it's modeled after the way you'd do it in asm for most risc isa's, and is competitive with the asm implementations for most (iirc it's something like 80-90% of the speed of android's memcpy.s on arm)	00:47
  stekern	_franck_: Linux doesn't use libgcc for memcpy in the generic case: http://lxr.free-electrons.com/source/lib/string.c#L589	03:29
  stekern	and if you have written an efficient copy_to/from_user (we haven't), then you basically have an efficient memcpy too	03:33
  _franck_	(memcpy) should we have an arch specific memcpy then ? looks like most (all ?) of Linux arch that have a memcopy do it in assembler	16:31
  _franck_	dalias: (memcpy) good to know. If I had to make an asm memcpy, I would compile it down and tweaked by hand like pgavin said	16:35
  _franck_	well, memcpy complied with -O3 gives me 387 lines of assembly code... I hope there is some room for optimization	16:44
  _franck_	I don't understand something in the calling of "void *memcpy(void *restrict dest, const void *restrict src, size_t n)"	19:42

• 2014-05-14 (Wednesday):

  _franck_	stekern: blueCmd : is that a reason why we don't have __HAVE_ARCH_MEMCPY (with an asm version of memcpy doing word transfert and handling aligment) in the kernel ?	22:13
  olofk	http://www.juliusbaxter.net/openrisc-irc/search?q=memcpy	22:14
  _franck_	I would have said that the Kernel doesn't use the GCC memcpy but now I'm not sure. It is linked with libgcc right ?	22:19

• 2014-01-24 (Friday):

  stekern	jonibo: you mentioned a assembly version memcpy in openrisc linux, but I can't seem to find that, only the __copy_to/from_user, which AFAICT does about what memcpy do, but it's not used for memcpy	12:24
  jonibo	yeah... fair enough... I thought I remembered an existing memcpy but apparently it doesn't exist... just the "user" mem copy functions	12:24
  stekern	interestingly enough, my grepping yesterday also showed that arm can be configured to use their memcpy implementation for the copy_to/from_user	12:27
  jonibo	i think maybe GCC has the memcpy function (byte-by-byte)... I'm pretty sure I've seen it somewhere	12:28
  jonibo	yeah, but that should be fine for memcpy too because you should simply never hit an exception there...	12:33

• 2014-01-23 (Thursday):

  jonibo	anyway, this is one of the BIG pieces missing pieces of openrisc GCC... optimized versions of the builtin functions... strcpy, memcpy, et cetera, et cetera...	12:54
  stekern	or did you just mean that we are using generic versions of memcpy et al?	14:36
  jonibo	stekern: I meant that the functions memcpy, strchr, memset, et al are either not optmized or poorly optimized... memcpy has an assembly version, but it's really quite poor (and I think GCC has the same poor version)	15:39
  jonibo	memcpy currently just copies byte-by-byte	15:45

• 2013-07-14 (Sunday):

  poke53281	not that I know. unaligned memcpy or memset maybe	03:10

• 2013-05-29 (Wednesday):

  juliusb	memcpy()?	08:45

• 2012-06-12 (Tuesday):

  stekern	oh, and I have got it to run the dhrystone test as fast as gcc-4.8, by tweaking the limit where memcpys vs inserted load/stores goes	20:49

• 2012-05-13 (Sunday):

  stekern	clang was way behind at 0.5, I checked what slowed it down so much and it was the strcpys being done as memcpys, could be me doing something suboptimal in our backend	12:09

73 matches in 2795 log files with 136823 lines (1.3 seconds).