[gem5-dev] Change in gem5/gem5[develop]: arch,cpu: Separate printing and serialization of VecPredReg.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41999 ) Change subject: arch,cpu: Separate printing and serialization of VecPredReg. .. arch,cpu: Separate printing and

[gem5-dev] Change in gem5/gem5[develop]: sim: Don't needlessly recreate ISA types in InstRecord.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/42002 ) Change subject: sim: Don't needlessly recreate ISA types in InstRecord. .. sim: Don't needlessly recreate ISA

[gem5-dev] Change in gem5/gem5[develop]: cpu: Use the built in << for VecReg and VecPredReg in ExeTrace.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/42001 ) Change subject: cpu: Use the built in << for VecReg and VecPredReg in ExeTrace. .. cpu: Use the built in <<

[gem5-dev] Change in gem5/gem5[develop]: arch: Remove unnecessary "typename"s from VecPredRegT.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41998 ) Change subject: arch: Remove unnecessary "typename"s from VecPredRegT. .. arch: Remove unnecessary "typename"s

[gem5-dev] Change in gem5/gem5[develop]: arch: Simplify and correct style of VecReg types.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41993 ) Change subject: arch: Simplify and correct style of VecReg types. .. arch: Simplify and correct style of VecReg

[gem5-dev] Change in gem5/gem5[develop]: arch: Break the dependence between (non)-predicate vector regs.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41997 ) Change subject: arch: Break the dependence between (non)-predicate vector regs. .. arch: Break the dependence

[gem5-dev] Re: vector register indexing modes and renaming?

Another question/clarification: Does any data actually get shared between the two rename modes? I think you said there is not, but now I can't find that. Would it work just as well to have two register files which operate entirely independently? From what I can tell the "V" registers of Neon in

[gem5-dev] Change in gem5/gem5[develop]: arch: Delete a few unused vector register types/constants.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/42003 ) Change subject: arch: Delete a few unused vector register types/constants. .. arch: Delete a few unused vector

[gem5-dev] Change in gem5/gem5[develop]: arch,sim: Get rid of unused "Packed" vector predicate registers.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41996 ) Change subject: arch,sim: Get rid of unused "Packed" vector predicate registers. .. arch,sim: Get rid of

[gem5-dev] Change in gem5/gem5[develop]: arch,cpu: Stop using << and to_number for VecReg serialization.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41994 ) Change subject: arch,cpu: Stop using << and to_number for VecReg serialization. .. arch,cpu: Stop using << and

[gem5-dev] Change in gem5/gem5[develop]: arch: Collapse unused size parameter from "as" VecPredReg method.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/42000 ) Change subject: arch: Collapse unused size parameter from "as" VecPredReg method. .. arch: Collapse unused

[gem5-dev] Re: Build failed in Jenkins: Weekly #9

I believe I've got the issue sorted for now. The fail_if was simply checking if the driver was set to a not-null value, however, we didn't explicitly set the driver to null anywhere. Now, the thing that's more interesting is that on certain processors, the driver was still null the first time

[gem5-dev] Change in gem5/gem5[develop]: arch,cpu,gpu-compute: Further simplify VecRegContainer.

Gabe Black has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41995 ) Change subject: arch,cpu,gpu-compute: Further simplify VecRegContainer. .. arch,cpu,gpu-compute: Further simplify

[gem5-dev] Change in gem5/gem5[develop]: gpu-compute: Explicitly set driver to NULL in constructor

Kyle Roarty has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41973 ) Change subject: gpu-compute: Explicitly set driver to NULL in constructor .. gpu-compute: Explicitly set driver

[gem5-dev] Build failed in Jenkins: Weekly #9

See Changes: [gabe.black] cpu,mem: Add or remove parenthesis to make the compiler happy. [gabe.black] misc: Fix mismatched struct/class "tags" and reenable that warning. [baz21] arch-arm,sim: make compile on FreeBSD

[gem5-dev] Change in gem5/gem5[develop]: sim,base-stats: Fix leading "." bug when obtaining requestors

Bobby R. Bruce has submitted this change. ( https://gem5-review.googlesource.com/c/public/gem5/+/41513 ) Change subject: sim,base-stats: Fix leading "." bug when obtaining requestors .. sim,base-stats: Fix leading "." bug

[gem5-dev] Change in gem5/gem5[develop]: base-stats,python: Add Units to the Python Stats

Bobby R. Bruce has submitted this change. ( https://gem5-review.googlesource.com/c/public/gem5/+/41754 ) Change subject: base-stats,python: Add Units to the Python Stats .. base-stats,python: Add Units to the Python Stats

[gem5-dev] Change in gem5/gem5[develop]: base-stats: Fixed System "work_item" stat name

Bobby R. Bruce has submitted this change. ( https://gem5-review.googlesource.com/c/public/gem5/+/41833 ) Change subject: base-stats: Fixed System "work_item" stat name .. base-stats: Fixed System "work_item" stat name The

[gem5-dev] Change in gem5/gem5[develop]: base-stats,python: Add Python Stats

Bobby R. Bruce has submitted this change. ( https://gem5-review.googlesource.com/c/public/gem5/+/38615 ) Change subject: base-stats,python: Add Python Stats .. base-stats,python: Add Python Stats This model is used to store

[gem5-dev] Change in gem5/gem5[develop]: base-stats,python: Expose a stat's unit via PyBind11

Bobby R. Bruce has submitted this change. ( https://gem5-review.googlesource.com/c/public/gem5/+/41753 ) Change subject: base-stats,python: Expose a stat's unit via PyBind11 .. base-stats,python: Expose a stat's unit via

[gem5-dev] Change in gem5/gem5[develop]: arch-x86: Add ACPI support for MADT

Maximilian Stein has uploaded this change for review. ( https://gem5-review.googlesource.com/c/public/gem5/+/41953 ) Change subject: arch-x86: Add ACPI support for MADT .. arch-x86: Add ACPI support for MADT This extends

[gem5-dev] Change in gem5/gem5[develop]: arch-riscv,misc: Add missing overrides for clang compilation

Bobby R. Bruce has submitted this change. ( https://gem5-review.googlesource.com/c/public/gem5/+/41913 ) Change subject: arch-riscv,misc: Add missing overrides for clang compilation .. arch-riscv,misc: Add missing overrides

[gem5-dev] Re: Build failed in Jenkins: Weekly #9

Hi Bobby, I've been trying to replicate it myself, but I've had no luck doing so. I built using the same commit (e100156a51deda7f50873faba069c2c3a6cde2e7) and command that were shown in the build log. I also used the pre-built square binary from dist.gem5, and used the same run command as

[gem5-dev] Re: Build failed in Jenkins: Weekly #9

The script running on Jenkins is as follows: ``` docker run -u $UID:$GUID --volume $(pwd):$(pwd) -w $(pwd) gcr.io/gem5-test/gcn-gpu:latest \ bash -c "scons build/GCN3_X86/gem5.opt -j50 || (rm -rf build && scons build/GCN3_X86/gem5.opt -j50)" wget -qN

[gem5-dev] Re: Build failed in Jenkins: Weekly #9

Matt, Kyle The weekly tests are failing with a "fatal: fatal condition driver occurred: Should not overwrite driver." error." when running the GCN3 square tests. I've managed to recreate this issue on my end, so it appears to be genuine. Could you look into this? Kind regards, Bobby -- Dr.