p4lang-p4c-nightly

p4lang-p4c-nightly:hanw/dpdk-pna

  1. Git
  2. lp:p4lang-p4c-nightly
  3. hanw/dpdk-pna
Last commit made on 2021-01-21
Get this branch:
git clone -b hanw/dpdk-pna https://git.launchpad.net/p4lang-p4c-nightly

Branch merges

Branch information

Name:
hanw/dpdk-pna
Repository:
lp:p4lang-p4c-nightly

Recent commits

37cff47... by Han Wang <email address hidden>

initial support

3b67fd6... by Mihai Budiu <email address hidden>

Add don't cares to the list of type variable bindings (#2633)

8405141... by dbrajovic-syrmia <email address hidden>

Convert SimplifyDefUse to a PassRepeated (#2610) (#2621)

This (second) PR is a continuation of the discussion in https://github.com/p4lang/p4c/pull/2610.
Originally suggested as part of the solution, it was decided to become a separate concern.
As a result, the SimplifyDefUse pass runs until convergence, instead of only once.

ba2aec7... by robert lytton <email address hidden>

Remove compiler warnings (#2626)

90194b6... by Han Wang <email address hidden>

add ceil_log2 to gmputil.h (#2625)

* Revert "remove duplicated library function in favor of big_int version (#2622)"

This reverts commit 426f512755848db4b9d8a04e761fc2204d79b232.

* add ceil_log2 to gmputil.h

7289a6b... by anasyrmia <email address hidden>

Fixes #2613 predication issue (#2624)

2bf2f61... by Chris Dodd <email address hidden>

cleanup PassManger construction (#2618)

Co-authored-by: Chrid Dodd <email address hidden>

426f512... by Han Wang <email address hidden>

remove duplicated library function in favor of big_int version (#2622)

1eb9c3a... by dbrajovic-syrmia <email address hidden>

Remove redundant slice statements during SimplifyDefUse (#2583) (#2610)

* Overwritten slice statements skipped
* Improved logging in simplifyDefUse.* and def_use.* files.

17c63d2... by Jed Liu <email address hidden>

Add INode::checkedTo for checked casts. (#2620)

Added a convenience method `INode::checkedTo` for checked casts. The
effect of `n->checkedTo<N>()` is similar to `&n->as<N>()`, except the
former produces an error message that is easier to debug. The former
also has clearer intent: the latter could be easily mistaken as being
equivalent to `n->to<N>()`.