search for: shifter

...uot;allowed" to use it, if that's the right word. I checked up on it more and indeed the mainstream targets like ARM use that node type in custom lowering code, so that should solve that. Perhaps in the future I might submit a patch for TargetLowering to set whether the target has a barrel shifter, because I believe that should alter the default emitted DAGs considerably. For example, I tried lowering SHL %a, 5 to a series of SHL %a, 1's at one point and the framework somewhat hilariously (to me) recombined them. If CodeGen knew not to do that, it could save some effort for people trying...

I forgot to mention that I used EXTRACT_ELEMENT in my backend to get the high and low parts of an SDValue. On 10 Nov 2013, at 17:50, Steve Montgomery <stephen.montgomery3 at btinternet.com> wrote: > I had a similar problem with a backend for the 68HC12 family which also has no barrel shifter. Some 68HC12 CPUs support shift for just one of the 16-bit registers and only support rotation on the 2 8-bit subregs of that 16-bit register. That means the only practical solution for 32-bit shifts is to lower to a libcall but my situation for 16-bit shifts sounds similar to yours for 32-bit shif...

I had a similar problem with a backend for the 68HC12 family which also has no barrel shifter. Some 68HC12 CPUs support shift for just one of the 16-bit registers and only support rotation on the 2 8-bit subregs of that 16-bit register. That means the only practical solution for 32-bit shifts is to lower to a libcall but my situation for 16-bit shifts sounds similar to yours for 32-bit shif...

...then it will expand the node into a more nodes as follows in an example: shl i32 %a , 2 => high_sdvalue = (or (shr %b, 14), (shl %c, 2) ) => low_sdvalue = (shl %b, 2) where %b would be the lower 16 bits of %a, and %c is the upper 16 bits of %a. Since this target doesn't have a barrel shifter, it would be a huge performance hit to construct the DAG in this way, due to the large number of right shifts). Thus, I wish to mark ISD::SHL/SHR as "Custom" in setOperationAction(), but I'm unsure of how to manually split the 32-bit shiftee SDValue into two 16-bit values. The standar...

...ng to convince them to let me spend some company time building an llvm back end :-) I have read the documentation and taken a look on the X86 back end. It looks quiet simple to get a very minimal back end working. I am considering for a first version: 1) only support the 32 bits ABI 2) Ignore the shifter (i.e use only 8 bits immediates) 3) Ignore the fancier addressing modes Without this, the generated code with not be very good but the back end will be very simple. The only part that appears to require more work is describing the instructions set. Am I underestimating how hard it is? How much wo...

...an > do it with NUT. > I don't think anyone has tried connecting to the serial port on the RPi expansion header, but NUT should not have a problem (as long as the driver is not expecting flow control pins). However, the serial port is TTL levels, so you would need a TTL-to-RS-232 level shifter chip. The USB hub would be a lot simpler, IMHO. -- Charles Lepple clepple at gmail

...have read the documentation and taken a look on the X86 back end. It > looks quiet simple to get a very minimal back end working. I am > considering for a first version: One of the RISC backends will make the work look a lot simpler :) > 1) only support the 32 bits ABI > 2) Ignore the shifter (i.e use only 8 bits immediates) > 3) Ignore the fancier addressing modes > > Without this, the generated code with not be very good but the back > end will be very simple. The only part that appears to require more > work is describing the instructions set. Actually, I think the m...

Hi there, I have a question about a raspberry pi as NUT server and a "relay". I explain my situation: I have an UPS with USB port, so I connect it to the rasp. The problem is that I have a device manageable only via USB port. How can I handle this configuration? My assumptions are: 1) split the USB cable from UPS to rasp and device, but I don't know if it works 2) connect the UPS

...si %R0<kill>, 18, pred:14, pred:%noreg, opt:%noreg** **    %R1<def> = LEApcrelJT <jt#0>, pred:14, pred:%noreg** **    BR_JTm %R0<kill>, %R1<kill>, 0, <jt#0>; mem:LD4[JumpTable]* So basically I want to catch the pattern with the possible simplification using the shifter, but I'm not even sure that I am looking into this issue at the right optimization level. Maybe this idea should be implemented in a higher level, or as a fixup in ARMConstantIslands, like the Thumb jumptable optimizations mentioned in the Bugzilla issue. I hope someone more familiar with th...

Compile with -O3 -m32 (see <https://godbolt.org/z/yCpBpM>): long lsign(long x) { return (x > 0) - (x < 0); } long long llsign(long long x) { return (x > 0) - (x < 0); } While the code generated for the "long" version of this function is quite OK, the code for the "long long" version misses an obvious optimisation: lsign: # @lsign mov

...x 240dda102845..b562a8cd61bf 100644 --- a/drivers/gpu/drm/omapdrm/displays/Kconfig +++ b/drivers/gpu/drm/omapdrm/displays/Kconfig @@ -8,18 +8,18 @@ config DRM_OMAP_ENCODER_OPA362 through a GPIO. config DRM_OMAP_ENCODER_TPD12S015 - tristate "TPD12S015 HDMI ESD protection and level shifter" + tristate "TPD12S015 HDMI ESD protection and level shifter" help Driver for TPD12S015, which offers HDMI ESD protection and level shifting. config DRM_OMAP_CONNECTOR_HDMI - tristate "HDMI Connector" + tristate "HDMI Connector" help Driv...

...and might not produce anything usable for a year ;-) > I have read the documentation and taken a look on the X86 back end. It > looks quiet simple to get a very minimal back end working. I am > considering for a first version: > > 1) only support the 32 bits ABI > 2) Ignore the shifter (i.e use only 8 bits immediates) > 3) Ignore the fancier addressing modes > > Without this, the generated code with not be very good but the back > end will be very simple. The only part that appears to require more > work is describing the instructions set. > > Am I underest...

    Lately I replaced my 10+-year-old server with a nano-ITX-type one that has USB but not old serial ports.  I connected my FE700VA to DB25-to-9-to-USB cables, but haven't been able to set it up. Apparently my DB9-to-USB cable (Prolific Technology, Inc. PL2303 Serial Port) has issues needing old software (setserial?) and some new software (fconfigure) being changed.  Has anyone managed

...nally felt they did a good job of power filtering. I use a NUT master running on a Raspberry Pi 3 which is easy to maintain and gets current updates via its Debian roots. I have two environments, one with a much older UPS which has only serial, but the addition of a tiny card with a MAX3232 level-shifter https://www.sparkfun.com/products/11189 allows the RPi to communicate nicely, the second environment uses a USB connection from the RPi to the UPS. Both work very well with NUT, and then communicate with NUT clients on the other systems. I like that the RPi allows me to separate the NUT master fu...

I've been looking over the documentation and several installation guides for NUT, and I'm starting to get a little bit of a feel for what it is. But I'm still not clear whether what I'd *like* to accomplish is actually practical using NUT. Perhaps someone here can tell me if they've done something similar, or if I'm way off base. I would like to set up a Linux host as a

...> can do it with NUT. > I don't think anyone has tried connecting to the serial port on the RPi expansion header, but NUT should not have a problem (as long as the driver is not expecting flow control pins). However, the serial port is TTL levels, so you would need a TTL-to-RS-232 level shifter chip. The USB hub would be a lot simpler, IMHO. -- Charles Lepple clepple at gmail

...x 240dda102845..b562a8cd61bf 100644 --- a/drivers/gpu/drm/omapdrm/displays/Kconfig +++ b/drivers/gpu/drm/omapdrm/displays/Kconfig @@ -8,18 +8,18 @@ config DRM_OMAP_ENCODER_OPA362 through a GPIO. config DRM_OMAP_ENCODER_TPD12S015 - tristate "TPD12S015 HDMI ESD protection and level shifter" + tristate "TPD12S015 HDMI ESD protection and level shifter" help Driver for TPD12S015, which offers HDMI ESD protection and level shifting. config DRM_OMAP_CONNECTOR_HDMI - tristate "HDMI Connector" + tristate "HDMI Connector" help Driv...

...ost the APC and Tripp-Lite MIBs, so that should cover case 1. (The three-phase names are similar to the single-phase names.) I have done a fair amount of testing with a Raspberry Pi and USB, and it is no worse than USB on a PC, so that covers at least part of case 2. I think you would need a level-shifter to do RS-232 serial with a RPi, and you may need to fiddle with the cable if modem control lines are involved, but it should be possible. I will point out that we sometimes run into regressions with firmware upgrades on SNMP network cards, but often it is something that can be fixed by downgrading...

...nning on a Raspberry Pi 3 which is easy to maintain and gets > current updates via its Debian roots. Ah . . another good reason to get a RPi going . . > I have two environments, one > with a much older UPS which has only serial, but the addition of a > tiny card with a MAX3232 level-shifter > https://www.sparkfun.com/products/11189 allows the RPi to communicate > nicely, the second environment uses a USB connection from the RPi to > the UPS. Both work very well with NUT, and then communicate with NUT > clients on the other systems. I like that the RPi allows me to > s...

...x 240dda102845..b562a8cd61bf 100644 --- a/drivers/gpu/drm/omapdrm/displays/Kconfig +++ b/drivers/gpu/drm/omapdrm/displays/Kconfig @@ -8,18 +8,18 @@ config DRM_OMAP_ENCODER_OPA362 through a GPIO. config DRM_OMAP_ENCODER_TPD12S015 - tristate "TPD12S015 HDMI ESD protection and level shifter" + tristate "TPD12S015 HDMI ESD protection and level shifter" help Driver for TPD12S015, which offers HDMI ESD protection and level shifting. config DRM_OMAP_CONNECTOR_HDMI - tristate "HDMI Connector" + tristate "HDMI Connector" help Driv...