diff options
| author | Ali Saidi <Ali.Saidi@ARM.com> | 2012-06-07 10:59:03 -0400 |
|---|---|---|
| committer | Ali Saidi <Ali.Saidi@ARM.com> | 2012-06-07 10:59:03 -0400 |
| commit | c80cd4136e3fec00c8448bc0dea20a65b182a259 (patch) | |
| tree | 26e4649233c5236eb0b19a2f54ae020ae323c2cf /src/mem/tport.hh | |
| parent | 7183c3fd56c4f0e8e4137795bb91dcaff312f591 (diff) | |
| download | gem5-c80cd4136e3fec00c8448bc0dea20a65b182a259.tar.xz | |
mem: Delay deleting of incoming packets by one call.
This patch is a temporary fix until Andreas' four-phase patches
get reviewed and committed. Removing FastAlloc seems to have exposed
an issue which previously was reasonable rare in which packets are freed
before the sending cache is done with them. This change puts incoming packets
no a pendingDelete queue which are deleted at the start of the next call and
thus breaks the dependency between when the caller returns true and when the
packet is actually used by the sending cache.
Running valgrind on a multi-core linux boot and the memtester results in no
valgrind warnings.
Diffstat (limited to 'src/mem/tport.hh')
| -rw-r--r-- | src/mem/tport.hh | 8 |
1 files changed, 8 insertions, 0 deletions
diff --git a/src/mem/tport.hh b/src/mem/tport.hh index db5a074fb..1f08d1a91 100644 --- a/src/mem/tport.hh +++ b/src/mem/tport.hh @@ -73,6 +73,14 @@ class SimpleTimingPort : public QueuedSlavePort virtual Tick recvAtomic(PacketPtr pkt) = 0; + /** + * @todo this is a temporary workaround until the 4-phase code is committed. + * upstream caches need this packet until true is returned, so hold it for + * deletion until a subsequent call + */ + std::vector<PacketPtr> pendingDelete; + + public: /** |