search for: strides

Hello, I understand the need for a proper stride setting when formatting a filesystem on a RAID device. However, is there any problem in using a stride setting when formatting a filesystem on a regular non-RAID, non-SSD, just plain-vanilla-single-disk block device? I'm sure there isn't any benefit to it, but I'm curious if there is any harm. The reason I ask is I'm looking at

Sorry for the spam. Copy-paste didn't capture the Subject properly. Resending with the correct Subject so that the thread is captured properly. -----Original Message----- From: Saito, Hideki Sent: Wednesday, June 15, 2016 1:39 PM To: 'llvm-dev at lists.llvm.org' <llvm-dev at lists.llvm.org> Subject: RE: [llvm-dev] [Proposal][RFC] Strided Memory Access Ashutosh, First,

Dear all, In our compiler we use a modified version LLVM Polly, which is very sensitive to proper code generation. Among the number of limitations, the loop region (enclosed by phi node on induction variable and branch) is required to be free of additional memory-dependent branches. In other words, there must be no conditional "br" instructions below phi nodes. The problem we are facing

>Vectorizer's output should be as clean as vector code can be so that analyses and optimizers downstream can >do a great job optimizing. Guess I should clarify this philosophical position of mine. In terms of vector code optimization that complicates the output of vectorizer: If vectorizer is the best place to perform the optimization, it should do so. This includes the cases like

One common concern raised for cases where Loop Vectorizer generate bigger types than target supported: Based on VF currently we check the cost and generate the expected set of instruction[s] for bigger type. It has two challenges for bigger types cost is not always correct and code generation may not generate efficient instruction[s]. Probably can depend on the support provided by below RFC by

...and >> count==1). > > > For the same issue in a different context, I recently implemented a scheme using run-length-encoding but using a variable stride. So for a run of alternate words, you still get a single entry, but with stride 16 instead of 8. In my application, most cases of strides > 8 are a run of only 2 or 3 but there are a few cases of dozens or hundreds with a stride of 16. My case is a solution tailored to exactly one application (a kernel), so there is a closed sample set that's all that matters and the trade-off between simplicity of the analysis and compactnes...

As a strong advocate of logical vector representation, I'm counting on community liking Michael's RFC and that'll proceed sooner than later. I plan to pitch in (e.g., perf experiments). >Probably can depend on the support provided by below RFC by Michael: > "Allow loop vectorizer to choose vector widths that generate illegal types" >In that case Loop Vectorizer will

The driver doesn't consider framebuffer pitch and offset, leading to a wrong display in case offset != 0 or pitch != width * bpp. Fix it. Signed-off-by: Gerd Hoffmann <kraxel at redhat.com> --- drivers/gpu/drm/bochs/bochs.h | 2 +- drivers/gpu/drm/bochs/bochs_hw.c | 14 ++++++++++---- drivers/gpu/drm/bochs/bochs_kms.c | 3 ++- 3 files changed, 13 insertions(+), 6 deletions(-)

I'm seeing the following failures with "make check" (x86-32 linux): FAIL: test/CodeGen/X86/fold-mul-lohi.ll Failed with exit(1) at line 2 while running: llvm-as < test/CodeGen/X86/fold-mul-lohi.ll | llc -march=x86-64 | not grep lea leaq B, %rsi leaq A, %r8 leaq P, %rsi child process exited abnormally FAIL:

Hi, is there any place where the possible values for the strides in the yuv_buffer structure is specified? or just in the source code? ;) I could compress a video with, lets say, yuvbuf.y_stride = yuvbuf.y_width, so the yuvbuf.y[0] really started in the start of the y channel data. I watched the video in several players, and everythings' ok. But, when progr...

I agree that a single vector index is sufficient for many cases. Matt Pharr (from the ISPC compiler), showed me an interesting case where there is a single pointer into an array. In this case we need to have two indices, where the first index is zero. Once the basic patch is in, we can start looking at adding support for arrays and multiple indices. Nadav -----Original Message----- From: David

Fixed. Thanks. Evan On Feb 11, 2008, at 2:35 AM, Duncan Sands wrote: > I'm seeing the following failures with "make check" (x86-32 linux): > > FAIL: test/CodeGen/X86/fold-mul-lohi.ll > Failed with exit(1) at line 2 > while running: llvm-as < test/CodeGen/X86/fold-mul-lohi.ll | llc - > march=x86-64 | not grep lea > leaq B, %rsi > leaq

Hi, I am trying to educate myself on prefetching/readahead algorithm for Lustre''s read. For a starter I only have two simple questions. 1 - Does Lustre detect linear or random I/O pattern or it always triggers readahead? 2 - If readahead is triggered, how many pages are read in addition to what is necessary? Thanks, Arifa.

Fixes FDO#55294. --- src/gallium/drivers/nv30/nv30_miptree.c | 3 +-- src/gallium/drivers/nv50/nv50_miptree.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/src/gallium/drivers/nv30/nv30_miptree.c b/src/gallium/drivers/nv30/nv30_miptree.c index 5a9a63b..9700fa8 100644 --- a/src/gallium/drivers/nv30/nv30_miptree.c +++ b/src/gallium/drivers/nv30/nv30_miptree.c @@ -56,8

All, I have a question about calculating the value for the -E stride option to mke2fs. The mke2fs man page says stride=stripe-size Configure the filesystem for a RAID array with stripe-size filesystem blocks per stripe. So stride = size of stripe/blocksize. The size of a stripe is the RAID chunk size * the number of drives in the RAID. My question: are parity disks

Hi Evan, In -relocation-model=static mode, those tests are now getting code like this leaq A, %rsi movss %xmm0, (%rsi,%rdx,4) instead of this: movss %xmm0, A(,%rdx,4) This is specifically what these tests were written to catch :-). Running them with -relocation-model=pic is hiding the real bug. Dan On Feb 11, 2008, at 11:22 PM, Evan Cheng wrote: > Fixed.

...ctorize it. > Oh, so that's what I was missing. When Nadav said about pointer reduction, I thought that was how we were should be dealing with memory PHIs in the end. I'll see how stride 1 pointer induction traverses the code and where to add stride N (but not sooner than I try to teach strides to non-pointer cases). I'll also add the reduction case, like: for (i .. N/3) { r += a[3*i] ..; r += a[3*i+1] ..; r += a[3*i+2] ..; } return r; And see how it should work. (later, too). Basically for pointer inductions we store the start value. When we come to > actually vectorize...

Fixed. However, I wonder if we are doing the right / smart codegen for static codegen. AMD64 ABI document seems to indicate rip relative addressing should be used even in this case (see page 38). You know about about Linux addressing mode than I do. Please check. Thanks, Evan On Feb 12, 2008, at 10:10 AM, Dan Gohman wrote: > Hi Evan, > > In -relocation-model=static mode, those

I''ve looked long and hard for an answer to this and am stuck. I have setup a Windows XP DomU on Xen 3.2-1 running on Debian and I can connect fine to the instance via VNC but I get no networking within the windows instance as it says the IP address conflicts with another on the network. All my linux instances network fine, and I''m confused as to the issue. Can anybody provide

...rue. As it stands, all my examples can't vectorize because of the extra memory PHI, and your example below can't vectorize because it can't find the array bounds. I'm assuming that, as soon as I teach the validation to accept your loop (given additional checks), and teach about the strides and costs, it will vectorize. But if I go back to my original code, it won't, because of the reduction PHI. So, again, I agree with you, one step at a time, I'll work with your loop, because it's the straightest path from here. But at some time, I'll have to either identify the mem...