view src/testdir/test77a.in @ 34105:c4ad4778946c v9.1.0017

patch 9.1.0017: [security]: use-after-free in eval1_emsg() Commit: https://github.com/vim/vim/commit/28d71b566a29ceea3a2d05bcee9264ed5d630d42 Author: Yegappan Lakshmanan <yegappan@yahoo.com> Date: Fri Jan 12 17:21:55 2024 +0100 patch 9.1.0017: [security]: use-after-free in eval1_emsg() Problem: use-after-free in eval1_emsg() when an empty line follows a lambda (by @yu3s) Solution: only set evalarg->eval_using_cmdline = FALSE when the *arg pointer is not null fixes: #13833 closes: #13841 Signed-off-by: Yegappan Lakshmanan <yegappan@yahoo.com> Signed-off-by: Christian Brabandt <cb@256bit.org>
author Christian Brabandt <cb@256bit.org>
date Fri, 12 Jan 2024 17:30:06 +0100
parents e705ea6e855b
children
line wrap: on
line source

Inserts 2 million lines with consecutive integers starting from 1
(essentially, the output of GNU's seq 1 2000000), writes them to Xtest
and writes its cksum to test.out.

We need 2 million lines to trigger a call to mf_hash_grow().  If it would mess
up the lines the checksum would differ.

cksum is part of POSIX and so should be available on most Unixes.
If it isn't available then the test will be skipped.

VMS does not have CKSUM but has a built in CHECKSUM - it should be used
STARTTEST
:silent! while 0
:  e! test.ok
:  w! test.out
:  qa!
:silent! endwhile
:if !has("vms")
: e! test.ok
: w! test.out
: qa!
:endif
:set fileformat=unix undolevels=-1
ggdG
:let i = 1
:while i <= 2000000 | call append(i, range(i, i + 99)) | let i += 100 | endwhile
ggdd
:w! Xtest.
:r !@test77a.com Xtest.
:s/\s/ /g
:set fileformat&
:.w! test.out
:qa!
ENDTEST