similar to: [LLVMdev] loop fusion

Is there a transformation in LLVM that will perform loop fusion? http://en.wikipedia.org/wiki/Loop_fusion I have the following program, in which I would like the 2 loops (iterating the same number of times) to be merged into 1, after which other nice optimizations such as mem2reg will apply: ; ModuleID = 'test' define void @vector([16 x float]* nocapture %arg, [16 x float]* nocapture

I did find this note from 2004 that references a LoopFusion pass, but I can't find it in the source code: http://nondot.org/sabre/LLVMNotes/LoopOptimizerNotes.txt A little background - I'm working on a SIMD runtime, where I have a scalar program but need to execute it on multiple independent data elements. One approach is to generate a loop for each operation, then rely on the

# uname -a Linux dhcppc1 3.0.1-xxxx-std-ipv6-64 #1 SMP Sun Aug 14 17:06:21 CEST 2011 x86_64 x86_64 x86_64 GNU/Linux mkdir test5 cd test5 dd if=/dev/null of=img5 bs=1 seek=2G dd if=/dev/null of=img6 bs=1 seek=2G losetup /dev/loop2 img5 losetup /dev/loop3 img6 mkfs.btrfs -d raid1 -m raid1 /dev/loop2 /dev/loop3 btrfs device scan btrfs filesystem show Label: none uuid:

Hello list, Currently, i use a partition image to run my XenU (/dev/hda2) and "fdisk -l" return nothing and can''t open /dev/hda . I want to use a full hard drive image (not partition image) from my Xen0 as hda in my XenU. I want to be able to run fdisk on hda in my XenU to list and regenerate partitions. I tried with this (/img is my hard drive image): disk =

On Mon, Sep 21, 2015 at 11:13 AM, Wei Mi <wmi at google.com> wrote: > I have the same worry as Philip and Hal that the new LoopExitValues > pass may increase some live range significantly in certain cases > because it reuses value cross outerloop iterations. Like the following > hypothetical case, the value reuse will create a live range living > across loop2, loop3, .... But

The following C test program was compiled using LLVM with -O3 option and MSVC with /O2. The MSVC one is about 600 times faster than the one compiled with the LLVM. We can see that the for loop in MSVC assembler is solved in the optimization pass more efficiently than that in LLVM. Is there an way to get a optimization result in LLVM like that of the MSVC? Manoel Teixeira #include

virtnet_poll is called from soft IRQ and it tries to allocate GFP_ATOMIC memory (through try_fill_recv). This allocation can fail and we are falling back to schedule_delayed_work in that case. Let's add __GFP_NOWARN to the allocation flags to get rid of the allocator complains for failed allocations: [22798.508903] The following is only an harmless informational message. [22798.508909]

virtnet_poll is called from soft IRQ and it tries to allocate GFP_ATOMIC memory (through try_fill_recv). This allocation can fail and we are falling back to schedule_delayed_work in that case. Let's add __GFP_NOWARN to the allocation flags to get rid of the allocator complains for failed allocations: [22798.508903] The following is only an harmless informational message. [22798.508909]

Hi, I am trying for quite some time to create a bootable hd disk image with mkdiskimage (included in the syslinux distribution) and the combination isolinux/memdisk. It just won't work:-} I have done the following: [snip] # ./mkdiskimage -d -o disk 256 32 1 640 # losetup -o 640 /dev/loop3 disk # mount /dev/loop3 /mnt/3 # mount | grep loop3 /dev/loop3 on /mnt/3 type vfat (rw) :: copy

Hi Folks, Let's keep this optimization alive. To summarize: several folks voiced general support, but with questions about why existing optimizations do not already catch this case. Deep dive by Wei Mi showed that the optimization is most likely not a clean-up of LSR sloppiness, but something new. Follow-up by myself confirmed that the redundancy eliminated the LoopExitValues pass exists in

Il 30/01/19 16:49, Simon Matter ha scritto: >> On 01/30/19 03:45, Alessandro Baggi wrote: >>> Il 29/01/19 20:42, mark ha scritto: >>>> Alessandro Baggi wrote: >>>>> Il 29/01/19 18:47, mark ha scritto: >>>>>> Alessandro Baggi wrote: >>>>>>> Il 29/01/19 15:03, mark ha scritto: >>>>>>>

On Apr 23, 2013, at 10:37 AM, Shuxin Yang <shuxin.llvm at gmail.com> wrote: > As far as I can understand of the code, the Reassociate tries to achieve this result by its "ranking" mechanism. > > If it dose not, it is not hard to achieve this result, just restructure the expression in a way such that > the earlier definition of the sub-expression is permute earlier in

Hi! I recently tried to mount a filesystem in RAID1 mode using loopback devices. I followed the instructions at [1]. Here''s exactly what I''ve done: $ dd if=/dev/zero of=raid1_0.img bs=1M count=500 $ dd if=/dev/zero of=raid1_1.img bs=1M count=500 $ mkfs.btrfs -m raid1 -d raid1 raid1_0.img raid1_1.img $ losetup /dev/loop0 raid1_0.img $ losetup /dev/loop1 raid1_1.img $ mount -t

--- Josh Coalson <xflac@yahoo.com> wrote: > > -- Miroslav Lichvar <lichvarm@phoenix.inf.upol.cz> wrote: > > On Tue, Dec 17, 2002 at 01:01:08PM -0800, Josh Coalson wrote: > > > --- Miroslav Lichvar <lichvarm@phoenix.inf.upol.cz> wrote: > > > > Ok, what about enabling the 3dnow function in libFLAC by > default? > > > > I think time

As far as I can understand of the code, the Reassociate tries to achieve this result by its "ranking" mechanism. If it dose not, it is not hard to achieve this result, just restructure the expression in a way such that the earlier definition of the sub-expression is permute earlier in the resulting expr. e.g. outer-loop1 x= outer-loop2 y =

Continuing on with my earlier post . . . after looking through code of JBD, is the following perhaps the difference in why the md5 values differ; When a journalled filesystem that uses jbd is mounted the journal b_count is incremented by one? *EVEN* if the volume was mounted read only, this b_count is still increased by one? curious as ever! lt __________________________________ Do you

It's an interesting problem. The best stuff I've seen published is by Cooper, Eckhart, & Kennedy, in PACT '08. Cooper gives a nice intro in one of his lectures: http://www.cs.rice.edu/~keith/512/2012/Lectures/26ReassocII-1up.pdf I can't tell, quickly, what's going on in Reassociate; as usual, the documentation resolutely avoids giving any credit for the ideas. Why is that?

On Apr 25, 2013, at 10:51 AM, Preston Briggs <preston.briggs at gmail.com> wrote: > It's an interesting problem. > The best stuff I've seen published is by Cooper, Eckhart, & Kennedy, in PACT '08. > Cooper gives a nice intro in one of his lectures: http://www.cs.rice.edu/~keith/512/2012/Lectures/26ReassocII-1up.pdf > I can't tell, quickly, what's going on

> On 01/30/19 03:45, Alessandro Baggi wrote: >> Il 29/01/19 20:42, mark ha scritto: >>> Alessandro Baggi wrote: >>>> Il 29/01/19 18:47, mark ha scritto: >>>>> Alessandro Baggi wrote: >>>>>> Il 29/01/19 15:03, mark ha scritto: >>>>>> >>>>>>> I've no idea what happened, but the box I was working

I need direction. I don't know what to google for anymore. I am unable to get my SmartUPS 2200XL to be recognized. My suspicions are that it's UDEV or Kernel related. I've updated the kernel to the highest version I feel comfortable with due to other things that run on the server, and I've updated UDEV a bit too. When I plug in the USB cable from the UPS into the server, I