NotificationsYou must be signed in to change notification settings
Fork6
Star31

Commitd3bbc4b

committed

Add valgrind suppressions for wcsrtombs optimizations

wcsrtombs (called through wchar2char from common functions like lower,upper, etc.) uses various optimizations that may look like access touninitialized data, triggering valgrind reports.For example AVX2 instructions load data in 256-bit chunks, and gconvdoes something similar with 32-bit chunks. This is faster than accessingthe bytes one by one, and the uninitialized part of the buffer is notactually used. So suppress the bogus reports.The exact stack depends on possible optimizations - it might be AVX, SSE(as in the report by Aleksander Alekseev) or something else. Hence thelast frame is wildcarded, to deal with this.Backpatch all the way back to 9.4.Author: Tomas VondraDiscussion:https://www.postgresql.org/message-id/flat/90ac0452-e907-e7a4-b3c8-15bd33780e62%402ndquadrant.comDiscussion:https://www.postgresql.org/message-id/20180220150838.GD18315@e733.localdomain

1 parent37afc07 commitd3bbc4bCopy full SHA for d3bbc4b

File tree

1 file changed

+36

-0

lines changed

src/tools
- valgrind.supp

1 file changed

+36

-0

lines changed

`‎src/tools/valgrind.supp`

Lines changed: 36 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -212,3 +212,39 @@`
`212`	`212`	`Memcheck:Cond`
`213`	`213`	`fun:PyObject_Realloc`
`214`	`214`	`}`
	`215`	`+`
	`216`	`+# wcsrtombs uses some clever optimizations internally, which to valgrind`
	`217`	`+# may look like access to uninitialized data. For example AVX2 instructions`
	`218`	`+# load data in 256-bit chunks, irrespectedly of wchar length. gconv does`
	`219`	`+# somethink similar by loading data in 32bit chunks and then shifting the`
	`220`	`+# data internally. Neither of those actually uses the uninitialized part`
	`221`	`+# of the buffer, as far as we know.`
	`222`	`+#`
	`223`	`+# https://www.postgresql.org/message-id/90ac0452-e907-e7a4-b3c8-15bd33780e62@2ndquadrant.com`
	`224`	`+`
	`225`	`+{`
	`226`	`+ wcsnlen_optimized`
	`227`	`+ Memcheck:Cond`
	`228`	`+ ...`
	`229`	`+ fun:wcsrtombs`
	`230`	`+ fun:wcstombs`
	`231`	`+ fun:wchar2char`
	`232`	`+}`
	`233`	`+`
	`234`	`+{`
	`235`	`+ wcsnlen_optimized_addr32`
	`236`	`+ Memcheck:Addr32`
	`237`	`+ ...`
	`238`	`+ fun:wcsrtombs`
	`239`	`+ fun:wcstombs`
	`240`	`+ fun:wchar2char`
	`241`	`+}`
	`242`	`+`
	`243`	`+{`
	`244`	`+ gconv_transform_internal`
	`245`	`+ Memcheck:Cond`
	`246`	`+ fun:__gconv_transform_internal_utf8`
	`247`	`+ fun:wcsrtombs`
	`248`	`+ fun:wcstombs`
	`249`	`+ fun:wchar2char`
	`250`	`+}`

0 commit comments

Comments

(0)

Movatterモバイル変換

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commitd3bbc4b

File tree

1 file changed

1 file changed

`‎src/tools/valgrind.supp`

0 commit comments