To elide several races between reading and writing the metadata array for different types of transactions, all L2XactTrackers can now sink Voluntary Releases (writebacks from the L1 in the current implementation). These writebacks are merged with the ongoing transaction, and the merging tracker supplies an acknowledgment of the writeback in addition to its ongoing activities. This change involved another refactoring of the control logic for allocating new trackers and routing incoming Acquires and Releases. BroadcastHub uses the new routing logic, but still processes all voluntary releases through the VoluntaryReleaseTracker (not a problem because there are no metadata update races).
Closes#18Closes#20
This uses an reorder queue but doesn't check to ensure that the data it fetches
from the queue is actually in the queue before using it. It seems that during
correct operation this never breaks, but I'm trying to get the backup memory
port working again and this assertion fails with it enabled (without the
assertion the core just gets a bogus data beat dies).
Closes#16
Exposes new parameter field SplitMetadata to determine whether the metadata array in a particular cache is stored in a single SeqMem or with one SeqMem per way.
Closes#14
This was recently changed to write out physical addresses for SCR file entries,
but to bring up the chip we need SCR offsets so we can write the uncore SCR
file over HTIF. This changes the map generator to generate both.
Without this change things happened to work anyway because the high bits were
getting dropped by the SCR file.
Right now there's no way to ensure that SCR addresses don't conflict within
RocketChip. Since upstream only has one of them this isn't a big deal, but we
want to add a whole bunch more to control all the IP on Hurricane.
This patch adds some Scala code to allocate registers inside the SCR file,
ensure they don't conflict, to provide names for SCRs, attach registers to the
SCR file, and generate a C header file that contains the addresses of every SCR
on a chip.
With this patch we'll be able to get rid of that constant in the testbench.
This also allows us to kill one of the Raven diffs, which is does pretty much
the same thing (just in a second SCR file, and hacked in).
Prior to this commit, the L2 cache banks used the lower bits of the
block address as the set index. However, the lower bits are also used to
route addresses to different banks. As a result, in multi-bank
configurations, only a fraction of the sets in each bank could be
accessed. This commit fixes that problem by using the bits ahead of the
bank index as the set index, so that all sets in the cache can be
accessed.
