Bug 170825

Summary:

WebAssembly: limit slow memories

Product:

WebKit

Reporter:

JF Bastien <jfbastien>

Component:

JavaScriptCore

Assignee:

JF Bastien <jfbastien>

Status:

RESOLVED FIXED

Severity:

Normal

CC:

commit-queue, fpizlo, ggaren, jfbastien, keith_miller, mark.lam, msaboff, saam

Priority:

Version:

WebKit Nightly Build

Hardware:

Unspecified

OS:

Unspecified

Bug Depends on:

170628

Bug Blocks:

159775

Attachments:

Description	Flags
patch	saam: review+, saam: commit-queue-
patch	commit-queue: commit-queue-
patch	none

JF Bastien

Reported 2017-04-13 13:42:16 PDT

Bug #170628 limits the number of fast memories, partly because ASLR. The code then falls back to slow memories. It first tries to virtually allocated any declared maximum (and in there, physically the initial), and if that fails it tries to physically allocate the initial without any extra. This can still be used to cause a bunch of virtual allocation. We should probably impose a soft limit on slow memories as well. I think that limit should be against whole-process sum of virtually allocated slow memory (don't forget to include calls to grow!). I'm not sure what the number should be, but I'll guess something around what the maximum for fast memory virtual allocation is. Anything exceeding that limit would cause allocation/grow to fail.

Attachments
patch (3.91 KB, patch) 2017-04-19 01:35 PDT, JF Bastien	saam: review+ saam: commit-queue-	Details Formatted Diff Diff
patch (6.18 KB, patch) 2017-04-19 10:31 PDT, JF Bastien	commit-queue: commit-queue-	Details Formatted Diff Diff
patch (6.06 KB, patch) 2017-04-19 10:39 PDT, JF Bastien	no flags	Details Formatted Diff Diff
Show Obsolete (2) View All Add attachment proposed patch, testcase, etc.

JF Bastien

Comment 1 2017-04-19 01:35:15 PDT

Created attachment 307473 [details] patch

Saam Barati

Comment 2 2017-04-19 09:49:49 PDT

Comment on attachment 307473 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=307473&action=review r=me. I'd like to at least see a test we run on Mac, and if it's flaky, we can consider what to do next. But let's land it before assuming it'll be flaky. > Source/JavaScriptCore/ChangeLog:21 > + I haven't added a test because the bots will likely be unhappy / What about at least making it run on Mac? > Source/JavaScriptCore/wasm/WasmMemory.cpp:203 > + size_t maximum = fastMemoryAllocationSoftLimit * Memory::fastMappedBytes(); I would really add a new variable for this, even if initially it's the same value. That way we can change them independent of one another.

JF Bastien

Comment 3 2017-04-19 10:31:31 PDT

Created attachment 307488 [details] patch Add test and factor out function as suggested.

WebKit Commit Bot

Comment 4 2017-04-19 10:33:30 PDT

Comment on attachment 307488 [details] patch Rejecting attachment 307488 [details] from commit-queue. Failed to run "['/Volumes/Data/EWS/WebKit/Tools/Scripts/webkit-patch', '--status-host=webkit-queues.webkit.org', '--bot-id=webkit-cq-03', 'apply-attachment', '--no-update', '--non-interactive', 307488, '--port=mac']" exit_code: 2 cwd: /Volumes/Data/EWS/WebKit Last 500 characters of output: succeeded at 1 with fuzz 3. patching file JSTests/wasm.yaml Hunk #1 FAILED at 1. Hunk #2 succeeded at 31 (offset 2 lines). 1 out of 2 hunks FAILED -- saving rejects to file JSTests/wasm.yaml.rej patching file JSTests/wasm/stress/oom.js patching file Source/JavaScriptCore/ChangeLog Hunk #1 succeeded at 1 with fuzz 3. patching file Source/JavaScriptCore/wasm/WasmMemory.cpp Failed to run "[u'/Volumes/Data/EWS/WebKit/Tools/Scripts/svn-apply', '--force']" exit_code: 1 cwd: /Volumes/Data/EWS/WebKit Full output: http://webkit-queues.webkit.org/results/3564516

JF Bastien

Comment 5 2017-04-19 10:39:01 PDT

Created attachment 307489 [details] patch Rebase. I self-merge-conflicted.

WebKit Commit Bot

Comment 6 2017-04-19 11:45:41 PDT

The commit-queue encountered the following flaky tests while processing attachment 307489 [details]: The commit-queue is continuing to process your patch.

WebKit Commit Bot

Comment 7 2017-04-19 11:45:49 PDT

The commit-queue encountered the following flaky tests while processing attachment 307489 [details]: media/modern-media-controls/macos-fullscreen-media-controls/macos-fullscreen-media-controls-buttons-styles.html bug 168317 (author: graouts@apple.com) The commit-queue is continuing to process your patch.

WebKit Commit Bot

Comment 8 2017-04-19 12:38:55 PDT

Comment on attachment 307489 [details] patch Clearing flags on attachment: 307489 Committed r215525: <http://trac.webkit.org/changeset/215525>

WebKit Commit Bot

Comment 9 2017-04-19 12:38:57 PDT

All reviewed patches have been landed. Closing bug.

Note You need to log in before you can comment on or make changes to this bug.