[Bug runtime/31117] transport failures

["wcohen at redhat dot com" <sourceware-bugzilla@sourceware.org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=31117

--- Comment #5 from William Cohen <wcohen at redhat dot com> ---
Created attachment 15285
  --> https://sourceware.org/bugzilla/attachment.cgi?id=15285&action=edit
Patch reduce transport drops

The patch retries reserving space if it gets to the end of a subbuffer and
there isn't enough room for the header.  There probably needs to be some tweaks
to the code in staprun/relay.c to not count the padding as lost bytes when it
is looking for the the STAP_TRACE_MAGIC.

The attached patch improves the test results for the systemtap.map tests that
were transporting more than PAGE_SIZE amount of data.  It also improves the
test results for the syscall tests.

For:

sudo make installcheck RUNTESTFLAGS="systemtap.maps/*.exp"

Before the patch:

                === systemtap Summary ===

# of expected passes            83
# of unexpected failures        6
# of unsupported tests          1

                === systemtap Summary ===

# of expected passes            86
# of unexpected failures        3
# of unsupported tests          1

@@ -106 +95 @@
-FAIL: systemtap.maps/map_hash_SSI.stp
+PASS: systemtap.maps/map_hash_SSI.stp
@@ -108 +97 @@
-FAIL: systemtap.maps/map_hash_stat_II.stp
+PASS: systemtap.maps/map_hash_stat_II.stp
@@ -110 +99 @@
-FAIL: systemtap.maps/map_hash_stat_SI.stp
+PASS: systemtap.maps/map_hash_stat_SI.stp


The syscall test results look much better also.  However, on RHEL9/f39 there
are a number of unrelated failures on x86_64 for the 32-bit tests because the
check for the syscall names/numbers added by git commit
7fe1b7bb8e665c9edd38eaf34a499618ee763293

-- 
You are receiving this mail because:
You are the assignee for the bug.