summaryrefslogtreecommitdiff
path: root/open_issues/_san.mdwn
diff options
context:
space:
mode:
authorSamuel Thibault <samuel.thibault@ens-lyon.org>2024-08-01 00:57:37 +0200
committerSamuel Thibault <samuel.thibault@ens-lyon.org>2024-08-01 00:57:37 +0200
commit4ff7a6c4a504a7801bd33221fa56cdfa7f87ccdc (patch)
treefaef27573728f3389f2cebb8af0ce871979abc83 /open_issues/_san.mdwn
parentabdefb07cfe378c57fcc2995aa88d11f4c6976e0 (diff)
completeHEADmaster
Diffstat (limited to 'open_issues/_san.mdwn')
-rw-r--r--open_issues/_san.mdwn7
1 files changed, 4 insertions, 3 deletions
diff --git a/open_issues/_san.mdwn b/open_issues/_san.mdwn
index d514cfb2..374e89fe 100644
--- a/open_issues/_san.mdwn
+++ b/open_issues/_san.mdwn
@@ -35,11 +35,12 @@ as:
* Undefined Behavior Sanitizer (UBsan; `-fsanitize=undefined`)
Porting these to the Hurd is not a trivial task, for they have intimate
-knowledge about the operating system kernel they're running on, and from a
-first look they reimplement a lot of [[/glibc]] by directly using
+knowledge about the operating system kernel they're running on, and
+reimplement the needed parts of [[/glibc]] by directly using
[[system_call]]s -- which is basically a no-go on GNU Hurd.
-Samuel took some look at it and got some basic support for libubsan and libasan:
+Samuel took some look at it and got some basic support for libubsan and libasan
+by making it call glibc still:
<https://people.debian.org/~sthibault/hurd-i386/libubsan-hurd.diff>
which allowed to fix some issues in the hurd code.