Currently working at the Samsung Austin R&D Center on the Samsung next generation ARM cores.
- User Since
- Jan 5 2016, 9:21 AM (181 w, 1 d)
Thu, Jun 20
Wed, Jun 19
Tue, Jun 18
May 2 2019
Apr 30 2019
Apr 25 2019
Yes, default, whether explicit or implicit, counts too, for the resources used inside the target only care about different addresses. In you example, a typical branch predictor will see up to 3 target addresses. Of course, branch predictors are trained by branches that are actually taken, so only target addresses that are executed count. Which opens the window for future work involving FDO or just the static heuristics.
Apr 19 2019
Apr 5 2019
Apr 4 2019
I'm still collecting data on other architectures, but I've observed improvements between 3 and 15% in some SPEC benchmarks, such as 253.perlbmk and 400.perlbench, and proprietary ones on AArch64.
If anyone could give me a tip on how to avoid the review to be collapsed to just the part of whichever repo was hit first by a commit, I'd appreciate it.
Apr 3 2019
That's a good idea to follow up on, following the pattern of hasOptNone().
Apr 2 2019
Mar 29 2019
Mar 27 2019
Mar 26 2019
Mar 25 2019
Mar 22 2019
Mar 21 2019
Mar 20 2019
Mar 11 2019
LGTM, after the minor issues below are addressed.
Feb 27 2019
TargetLoweringBase::isFPImmLegal() could have a default value of false for ForCodeSize.
Feb 11 2019
Commited extra float C99 functions in rGf4a369596f7b.
Feb 7 2019
After the nsz check is addressed, it LGTM.
Feb 6 2019
Separated functions between VC19 and float.
Feb 5 2019
Assume VC19 as the default. However, I wonder if this should be the default earlier, in the clang driver.
Restrict the improvements to VS2015.