From: George Cherian george.cherian@marvell.com

[ Upstream commit 32837d8a8f63eb95dcb9cd005524a27f06478832 ]

Some Cavium endpoints are implemented as multi-function devices without ACS capability, but they actually don't support peer-to-peer transactions.

Add ACS quirks to declare DMA isolation for the following devices:

- BGX device found on Octeon-TX (8xxx) - CGX device found on Octeon-TX2 (9xxx) - RPM device found on Octeon-TX3 (10xxx)

Link: https://lore.kernel.org/r/20210810122425.1115156-1-george.cherian@marvell.co... Signed-off-by: George Cherian george.cherian@marvell.com Signed-off-by: Bjorn Helgaas bhelgaas@google.com Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/pci/quirks.c | 4 ++++ 1 file changed, 4 insertions(+)

diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index 7e873b6b7d55..10348d754313 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -4777,6 +4777,10 @@ static const struct pci_dev_acs_enabled { { 0x10df, 0x720, pci_quirk_mf_endpoint_acs }, /* Emulex Skyhawk-R */ /* Cavium ThunderX */ { PCI_VENDOR_ID_CAVIUM, PCI_ANY_ID, pci_quirk_cavium_acs }, + /* Cavium multi-function devices */ + { PCI_VENDOR_ID_CAVIUM, 0xA026, pci_quirk_mf_endpoint_acs }, + { PCI_VENDOR_ID_CAVIUM, 0xA059, pci_quirk_mf_endpoint_acs }, + { PCI_VENDOR_ID_CAVIUM, 0xA060, pci_quirk_mf_endpoint_acs }, /* APM X-Gene */ { PCI_VENDOR_ID_AMCC, 0xE004, pci_quirk_xgene_acs }, /* Ampere Computing */

From: Paolo Valente paolo.valente@linaro.org

[ Upstream commit 2d52c58b9c9bdae0ca3df6a1eab5745ab3f7d80b ]

The function bfq_setup_merge prepares the merging between two bfq_queues, say bfqq and new_bfqq. To this goal, it assigns bfqq->new_bfqq = new_bfqq. Then, each time some I/O for bfqq arrives, the process that generated that I/O is disassociated from bfqq and associated with new_bfqq (merging is actually a redirection). In this respect, bfq_setup_merge increases new_bfqq->ref in advance, adding the number of processes that are expected to be associated with new_bfqq.

Unfortunately, the stable-merging mechanism interferes with this setup. After bfqq->new_bfqq has been set by bfq_setup_merge, and before all the expected processes have been associated with bfqq->new_bfqq, bfqq may happen to be stably merged with a different queue than the current bfqq->new_bfqq. In this case, bfqq->new_bfqq gets changed. So, some of the processes that have been already accounted for in the ref counter of the previous new_bfqq will not be associated with that queue. This creates an unbalance, because those references will never be decremented.

This commit fixes this issue by reestablishing the previous, natural behaviour: once bfqq->new_bfqq has been set, it will not be changed until all expected redirections have occurred.

Signed-off-by: Davide Zini davidezini2@gmail.com Signed-off-by: Paolo Valente paolo.valente@linaro.org Link: https://lore.kernel.org/r/20210802141352.74353-2-paolo.valente@linaro.org Signed-off-by: Jens Axboe axboe@kernel.dk Signed-off-by: Sasha Levin sashal@kernel.org --- block/bfq-iosched.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index 5b3e5483c657..047eb0cce7b5 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -2137,6 +2137,15 @@ bfq_setup_merge(struct bfq_queue *bfqq, struct bfq_queue *new_bfqq) * are likely to increase the throughput. */ bfqq->new_bfqq = new_bfqq; + /* + * The above assignment schedules the following redirections: + * each time some I/O for bfqq arrives, the process that + * generated that I/O is disassociated from bfqq and + * associated with new_bfqq. Here we increases new_bfqq->ref + * in advance, adding the number of processes that are + * expected to be associated with new_bfqq as they happen to + * issue I/O. + */ new_bfqq->ref += process_refs; return new_bfqq; } @@ -2196,6 +2205,10 @@ bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq, { struct bfq_queue *in_service_bfqq, *new_bfqq;

+ /* if a merge has already been setup, then proceed with that first */ + if (bfqq->new_bfqq) + return bfqq->new_bfqq; + /* * Prevent bfqq from being merged if it has been created too * long ago. The idea is that true cooperating processes, and @@ -2210,9 +2223,6 @@ bfq_setup_cooperator(struct bfq_data *bfqd, struct bfq_queue *bfqq, if (bfq_too_late_for_merging(bfqq)) return NULL;

- if (bfqq->new_bfqq) - return bfqq->new_bfqq; - if (!io_struct || unlikely(bfqq == &bfqd->oom_bfqq)) return NULL;

From: Yang Li yang.lee@linux.alibaba.com

[ Upstream commit 0097ae5f7af5684f961a5f803ff7ad3e6f933668 ]

When the function IS_ALIGNED() returns false, the value of ret is 0. So, we set ret to -EINVAL to indicate this error.

Clean up smatch warning: drivers/ntb/test/ntb_perf.c:602 perf_setup_inbuf() warn: missing error code 'ret'.

Reported-by: Abaci Robot abaci@linux.alibaba.com Signed-off-by: Yang Li yang.lee@linux.alibaba.com Reviewed-by: Serge Semin fancer.lancer@gmail.com Signed-off-by: Jon Mason jdmason@kudzu.us Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/ntb/test/ntb_perf.c | 1 + 1 file changed, 1 insertion(+)

diff --git a/drivers/ntb/test/ntb_perf.c b/drivers/ntb/test/ntb_perf.c index ad5d3919435c..87a41d0ededc 100644 --- a/drivers/ntb/test/ntb_perf.c +++ b/drivers/ntb/test/ntb_perf.c @@ -600,6 +600,7 @@ static int perf_setup_inbuf(struct perf_peer *peer) return -ENOMEM; } if (!IS_ALIGNED(peer->inbuf_xlat, xlat_align)) { + ret = -EINVAL; dev_err(&perf->ntb->dev, "Unaligned inbuf allocated\n"); goto err_free_inbuf; }