|
|
Created:
3 years, 11 months ago by Buck Modified:
3 years, 6 months ago Reviewers:
Ryan Hamilton CC:
chromium-reviews, cbentzel+watch_chromium.org, tfarina, alyssar1 Target Ref:
refs/pending/heads/master Project:
chromium Visibility:
Public. |
Descriptionnet/QUIC: inline gypi file into BUILD.gn file and upstream proto-quic.
This is a redo of https://codereview.chromium.org/2475673004/.
It is based of current net.gypi, and adds (suprisingly little?)
proto-quic specific logic to net/BUILD.gn. That is guarded by a new gn
build argument "is_proto_quic".
On the proto-quic side of things, this will deprecate the merge_net_* part of the update.py script entirely. Instead, this version of BUILD.gn should be directly usable there, with the caveat that gn be invoked as follows during the proto-quic merge:
gn --args="is_proto_quic=true" gen out/Default
I will update the instructions there in a followup cl (in proto-quic
github), as well as nuking most if not all of modified_files/.
It is also worth pointing out that this abandons the previous proto-quic
strategy of blacklisting new files under net/ by default. I think that
only added maintenance complexity and did not significantly reduce
proto-quic size or build times.
I have tested this under proto-quic, but plan to do a new merge to
double check, once this has landed.
BUG=
Review-Url: https://codereview.chromium.org/2641413003
Cr-Commit-Position: refs/heads/master@{#445552}
Committed: https://chromium.googlesource.com/chromium/src/+/73f7240b56a264d017e8bf9f2a19c71f8a3e356d
Patch Set 1 #Patch Set 2 : Fix some leaked typos from proto-quic. #
Total comments: 1
Messages
Total messages: 24 (15 generated)
Description was changed from ========== net/QUIC: inline gypi file into BUILD.gn file and upstream proto-quic. This is a redo of https://codereview.chromium.org/2475673004/. It is based of current net.gypi, and adds (suprisingly little?) proto-quic specific logic to net/BUILD.gn. That is guarded by a new gn build argument "is_proto_quic". On the proto-quic side of things, this will deprecate the update.py script entirely. Instead, this version of BUILD.gn should be directly usable there, with the caveat that gn be invoked as follows during the proto-quic merge: gn --args="is_proto_quic=true" gen out/Default I will update the instructions there in a followup cl (in proto-quic github), as well as nuking most if not all of modified_files/. It is also worth pointing out that this abandons the previous proto-quic strategy of blacklisting new files under net/ by default. I think that only added maintenance complexity and did not significantly reduce proto-quic size or build times. I have tested this under proto-quic, but plan to do a new merge to double check, once this has landed. BUG= ========== to ========== net/QUIC: inline gypi file into BUILD.gn file and upstream proto-quic. This is a redo of https://codereview.chromium.org/2475673004/. It is based of current net.gypi, and adds (suprisingly little?) proto-quic specific logic to net/BUILD.gn. That is guarded by a new gn build argument "is_proto_quic". On the proto-quic side of things, this will deprecate the update.py script entirely. Instead, this version of BUILD.gn should be directly usable there, with the caveat that gn be invoked as follows during the proto-quic merge: gn --args="is_proto_quic=true" gen out/Default I will update the instructions there in a followup cl (in proto-quic github), as well as nuking most if not all of modified_files/. It is also worth pointing out that this abandons the previous proto-quic strategy of blacklisting new files under net/ by default. I think that only added maintenance complexity and did not significantly reduce proto-quic size or build times. I have tested this under proto-quic, but plan to do a new merge to double check, once this has landed. BUG= ==========
ckrasic@chromium.org changed reviewers: + dpranke@chromium.org, jri@chromium.org, rch@chromium.org
The CQ bit was checked by ckrasic@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: Try jobs failed on following builders: ios-device on master.tryserver.chromium.mac (JOB_FAILED, http://build.chromium.org/p/tryserver.chromium.mac/builders/ios-device/builds...)
Description was changed from ========== net/QUIC: inline gypi file into BUILD.gn file and upstream proto-quic. This is a redo of https://codereview.chromium.org/2475673004/. It is based of current net.gypi, and adds (suprisingly little?) proto-quic specific logic to net/BUILD.gn. That is guarded by a new gn build argument "is_proto_quic". On the proto-quic side of things, this will deprecate the update.py script entirely. Instead, this version of BUILD.gn should be directly usable there, with the caveat that gn be invoked as follows during the proto-quic merge: gn --args="is_proto_quic=true" gen out/Default I will update the instructions there in a followup cl (in proto-quic github), as well as nuking most if not all of modified_files/. It is also worth pointing out that this abandons the previous proto-quic strategy of blacklisting new files under net/ by default. I think that only added maintenance complexity and did not significantly reduce proto-quic size or build times. I have tested this under proto-quic, but plan to do a new merge to double check, once this has landed. BUG= ========== to ========== net/QUIC: inline gypi file into BUILD.gn file and upstream proto-quic. This is a redo of https://codereview.chromium.org/2475673004/. It is based of current net.gypi, and adds (suprisingly little?) proto-quic specific logic to net/BUILD.gn. That is guarded by a new gn build argument "is_proto_quic". On the proto-quic side of things, this will deprecate the merge_net_* part of the update.py script entirely. Instead, this version of BUILD.gn should be directly usable there, with the caveat that gn be invoked as follows during the proto-quic merge: gn --args="is_proto_quic=true" gen out/Default I will update the instructions there in a followup cl (in proto-quic github), as well as nuking most if not all of modified_files/. It is also worth pointing out that this abandons the previous proto-quic strategy of blacklisting new files under net/ by default. I think that only added maintenance complexity and did not significantly reduce proto-quic size or build times. I have tested this under proto-quic, but plan to do a new merge to double check, once this has landed. BUG= ==========
The CQ bit was checked by ckrasic@chromium.org to run a CQ dry run
Dry run: CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
The CQ bit was unchecked by commit-bot@chromium.org
Dry run: This issue passed the CQ dry run.
Exciting! https://codereview.chromium.org/2641413003/diff/20001/net/net.gypi File net/net.gypi (left): https://codereview.chromium.org/2641413003/diff/20001/net/net.gypi#oldcode2391 net/net.gypi:2391: # running net/data/update_net_gypi.py. Since update_net_gypi.py is being deleted, is it still possible to regenerate this list of files?
On 2017/01/22 16:20:02, Ryan Hamilton wrote: > Exciting! > > https://codereview.chromium.org/2641413003/diff/20001/net/net.gypi > File net/net.gypi (left): > > https://codereview.chromium.org/2641413003/diff/20001/net/net.gypi#oldcode2391 > net/net.gypi:2391: # running net/data/update_net_gypi.py. > Since update_net_gypi.py is being deleted, is it still possible to regenerate > this list of files? I believe that net/BUILD.gn will become the new source of truth. Also, I was looking at gn's reference, and it appears it has capabilities to generate file lists for given targets, etc.. It might be usable to look into further pruning the proto-quic repo to just those things absolutely needed to build the proto-quic targets, that is if there is room for improvement.
On 2017/01/22 17:26:50, Buck wrote: > On 2017/01/22 16:20:02, Ryan Hamilton wrote: > > Exciting! > > > > https://codereview.chromium.org/2641413003/diff/20001/net/net.gypi > > File net/net.gypi (left): > > > > https://codereview.chromium.org/2641413003/diff/20001/net/net.gypi#oldcode2391 > > net/net.gypi:2391: # running net/data/update_net_gypi.py. > > Since update_net_gypi.py is being deleted, is it still possible to regenerate > > this list of files? > > I believe that net/BUILD.gn will become the new source of truth. Perhaps I'm not understanding how this script is used, but my sense is that it runs through the filesystem looking for files to include, and allowing the resulting list of files to be used in net.gypi (which is included by both BUILD.gn and the now nonexistent net.gyp). I *assume* that this script exists because it's annoying to maintain this list of files by hand. (https://codereview.chromium.org/1923203002) Is that no longer required? > Also, I was looking at gn's reference, and it appears it has capabilities to > generate file lists for given targets, etc.. It might be usable to look into > further pruning the proto-quic repo to just those things absolutely needed to > build the proto-quic targets, that is if there is room for improvement. Cool!
On 2017/01/23 03:30:14, Ryan Hamilton wrote: > On 2017/01/22 17:26:50, Buck wrote: > > On 2017/01/22 16:20:02, Ryan Hamilton wrote: > > > Exciting! > > > > > > https://codereview.chromium.org/2641413003/diff/20001/net/net.gypi > > > File net/net.gypi (left): > > > > > > > https://codereview.chromium.org/2641413003/diff/20001/net/net.gypi#oldcode2391 > > > net/net.gypi:2391: # running net/data/update_net_gypi.py. > > > Since update_net_gypi.py is being deleted, is it still possible to > regenerate > > > this list of files? > > > > I believe that net/BUILD.gn will become the new source of truth. > > Perhaps I'm not understanding how this script is used, but my sense is that it > runs through the filesystem looking for files to include, and allowing the > resulting list of files to be used in net.gypi (which is included by both > BUILD.gn and the now nonexistent net.gyp). I *assume* that this script exists > because it's annoying to maintain this list of files by hand. > (https://codereview.chromium.org/1923203002) Is that no longer required? > > > Also, I was looking at gn's reference, and it appears it has capabilities to > > generate file lists for given targets, etc.. It might be usable to look into > > further pruning the proto-quic repo to just those things absolutely needed to > > build the proto-quic targets, that is if there is room for improvement. > > Cool! My understanding was that net/net.gypi was maintained by hand, and the script was used to include the files into net/BUILD.gn. After this CL,the net/net.gypi file is gone, and the list of files lives directly in net/BUILD.gn. I used the script to generate the current file lists, and then hand inlined from there to put this cl together.
lgtm
ckrasic@chromium.org changed reviewers: - dpranke@chromium.org, jri@chromium.org
The CQ bit was checked by ckrasic@chromium.org
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...
CQ is committing da patch. Bot data: {"patchset_id": 20001, "attempt_start_ts": 1485211125269890, "parent_rev": "24b30bfd5ac15c60a29477638c91ab0fb8159055", "commit_rev": "73f7240b56a264d017e8bf9f2a19c71f8a3e356d"}
Message was sent while issue was closed.
Description was changed from ========== net/QUIC: inline gypi file into BUILD.gn file and upstream proto-quic. This is a redo of https://codereview.chromium.org/2475673004/. It is based of current net.gypi, and adds (suprisingly little?) proto-quic specific logic to net/BUILD.gn. That is guarded by a new gn build argument "is_proto_quic". On the proto-quic side of things, this will deprecate the merge_net_* part of the update.py script entirely. Instead, this version of BUILD.gn should be directly usable there, with the caveat that gn be invoked as follows during the proto-quic merge: gn --args="is_proto_quic=true" gen out/Default I will update the instructions there in a followup cl (in proto-quic github), as well as nuking most if not all of modified_files/. It is also worth pointing out that this abandons the previous proto-quic strategy of blacklisting new files under net/ by default. I think that only added maintenance complexity and did not significantly reduce proto-quic size or build times. I have tested this under proto-quic, but plan to do a new merge to double check, once this has landed. BUG= ========== to ========== net/QUIC: inline gypi file into BUILD.gn file and upstream proto-quic. This is a redo of https://codereview.chromium.org/2475673004/. It is based of current net.gypi, and adds (suprisingly little?) proto-quic specific logic to net/BUILD.gn. That is guarded by a new gn build argument "is_proto_quic". On the proto-quic side of things, this will deprecate the merge_net_* part of the update.py script entirely. Instead, this version of BUILD.gn should be directly usable there, with the caveat that gn be invoked as follows during the proto-quic merge: gn --args="is_proto_quic=true" gen out/Default I will update the instructions there in a followup cl (in proto-quic github), as well as nuking most if not all of modified_files/. It is also worth pointing out that this abandons the previous proto-quic strategy of blacklisting new files under net/ by default. I think that only added maintenance complexity and did not significantly reduce proto-quic size or build times. I have tested this under proto-quic, but plan to do a new merge to double check, once this has landed. BUG= Review-Url: https://codereview.chromium.org/2641413003 Cr-Commit-Position: refs/heads/master@{#445552} Committed: https://chromium.googlesource.com/chromium/src/+/73f7240b56a264d017e8bf9f2a19... ==========
Message was sent while issue was closed.
Committed patchset #2 (id:20001) as https://chromium.googlesource.com/chromium/src/+/73f7240b56a264d017e8bf9f2a19...
Message was sent while issue was closed.
Really late drive-by, but can we get some docs on what "is_proto_quic" is? I haven't the foggiest. I assume it's never used for chromium builds, but not seeing any way to actually deduce that from source. |