This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Tunables for 2.23?
- From: Andi Kleen <andi at firstfloor dot org>
- To: "Paul E. Murphy" <murphyp at linux dot vnet dot ibm dot com>
- Cc: sid at reserved-bit dot com, "libc-alpha at sourceware dot org" <libc-alpha at sourceware dot org>, Carlos O'Donell <carlos at redhat dot com>, "munroesj at linux dot vnet dot ibm dot com" <munroesj at linux dot vnet dot ibm dot com>, Tulio Magno Quites Machado Filho <tuliom at linux dot vnet dot ibm dot com>, Andi Kleen <andi at firstfloor dot org>
- Date: Fri, 8 Jan 2016 03:20:37 +0100
- Subject: Re: Tunables for 2.23?
- Authentication-results: sourceware.org; auth=none
- References: <568C2AE6 dot 90002 at linux dot vnet dot ibm dot com>
I finally tracked down now why the tunables patchkit disabled elision
on x86. The register function always calls the set function with
an empty string, and atoi returns 0 for that.
That results in all the elision configuration variables being
zeroed.
With this patch -- avoiding calling set_func for empty strings --
the tunables work as expected for me on x86.
-Andi
diff --git a/tunables/tunables.c b/tunables/tunables.c
index 6e5aaa9..e8a5690 100644
--- a/tunables/tunables.c
+++ b/tunables/tunables.c
@@ -182,7 +182,7 @@ tunable_register (tunable_id_t id, tunable_setter_t set_func, char **envp)
tunable_t *cur = &tunable_list[id];
cur->set = set_func;
- if (cur->val != NULL)
+ if (cur->val != NULL && *cur->val)
{
set_func (cur->val);
cur->initialized = true;
--
ak@linux.intel.com -- Speaking for myself only.