summaryrefslogtreecommitdiff
path: root/hurd/history/hurd-flash5
diff options
context:
space:
mode:
authorThomas Schwinge <tschwinge@gnu.org>2008-11-06 12:29:44 +0100
committerThomas Schwinge <tschwinge@gnu.org>2008-11-06 12:29:44 +0100
commit4b35c398ab01c2e5ecf49d9fccc44c215f850a0e (patch)
tree2af42b0c7d4d03582ec23d9f92706b70948e4139 /hurd/history/hurd-flash5
parent2213166ab5b58cf89f3b07eded20ab2a03d9c681 (diff)
Integrate history.html and hurd-announcements.html, plus all Hurd flashes and announcements.
Diffstat (limited to 'hurd/history/hurd-flash5')
-rw-r--r--hurd/history/hurd-flash523
1 files changed, 23 insertions, 0 deletions
diff --git a/hurd/history/hurd-flash5 b/hurd/history/hurd-flash5
new file mode 100644
index 00000000..041a0ef7
--- /dev/null
+++ b/hurd/history/hurd-flash5
@@ -0,0 +1,23 @@
+From: mib@gnu.ai.mit.edu (Michael I Bushnell)
+Message-Id: <9409210619.AA17570@churchy.gnu.ai.mit.edu>
+To: "Lots of potentially interested people and" <nobody@gnu.ai.mit.edu>
+Subject: New milestone acheived by the GNU Hurd
+X-Tom-Swiftie: "I can't get this fire started," Tom said woodenly.
+
+
+I have just successfully compiled and run a null C program on the
+Hurd. This is using GCC native as one would normally use GCC.
+
+Sadly, it took quite a while (too long, in fact) to read the large
+archives that make up the GNU C library, but I think I know where the
+substantial inefficiency is.
+
+Once that is done, I would be happy to label this a "self-hosting
+system". But not just yet.
+
+The last bug preventing this was an error in dealing with files over
+about 8 M; this came about because in order to link a program one
+needed the GNU C library, which is over 9M when symbols are included.
+
+ -mib
+