From 437a310b22244d4e0b78665c3042e5d1c0f45306 Mon Sep 17 00:00:00 2001 From: Cristian Marussi Date: Wed, 20 Dec 2023 17:21:12 +0000 Subject: [PATCH 1/6] firmware: arm_scmi: Check mailbox/SMT channel for consistency On reception of a completion interrupt the shared memory area is accessed to retrieve the message header at first and then, if the message sequence number identifies a transaction which is still pending, the related payload is fetched too. When an SCMI command times out the channel ownership remains with the platform until eventually a late reply is received and, as a consequence, any further transmission attempt remains pending, waiting for the channel to be relinquished by the platform. Once that late reply is received the channel ownership is given back to the agent and any pending request is then allowed to proceed and overwrite the SMT area of the just delivered late reply; then the wait for the reply to the new request starts. It has been observed that the spurious IRQ related to the late reply can be wrongly associated with the freshly enqueued request: when that happens the SCMI stack in-flight lookup procedure is fooled by the fact that the message header now present in the SMT area is related to the new pending transaction, even though the real reply has still to arrive. This race-condition on the A2P channel can be detected by looking at the channel status bits: a genuine reply from the platform will have set the channel free bit before triggering the completion IRQ. Add a consistency check to validate such condition in the A2P ISR. Reported-by: Xinglong Yang Closes: https://lore.kernel.org/all/PUZPR06MB54981E6FA00D82BFDBB864FBF08DA@PUZPR06MB5498.apcprd06.prod.outlook.com/ Fixes: 5c8a47a5a91d ("firmware: arm_scmi: Make scmi core independent of the transport type") Cc: stable@vger.kernel.org # 5.15+ Signed-off-by: Cristian Marussi Tested-by: Xinglong Yang Link: https://lore.kernel.org/r/20231220172112.763539-1-cristian.marussi@arm.com Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/common.h | 1 + drivers/firmware/arm_scmi/mailbox.c | 14 ++++++++++++++ drivers/firmware/arm_scmi/shmem.c | 6 ++++++ 3 files changed, 21 insertions(+) diff --git a/drivers/firmware/arm_scmi/common.h b/drivers/firmware/arm_scmi/common.h index c46dc5215af7..00b165d1f502 100644 --- a/drivers/firmware/arm_scmi/common.h +++ b/drivers/firmware/arm_scmi/common.h @@ -314,6 +314,7 @@ void shmem_fetch_notification(struct scmi_shared_mem __iomem *shmem, void shmem_clear_channel(struct scmi_shared_mem __iomem *shmem); bool shmem_poll_done(struct scmi_shared_mem __iomem *shmem, struct scmi_xfer *xfer); +bool shmem_channel_free(struct scmi_shared_mem __iomem *shmem); /* declarations for message passing transports */ struct scmi_msg_payld; diff --git a/drivers/firmware/arm_scmi/mailbox.c b/drivers/firmware/arm_scmi/mailbox.c index 19246ed1f01f..b8d470417e8f 100644 --- a/drivers/firmware/arm_scmi/mailbox.c +++ b/drivers/firmware/arm_scmi/mailbox.c @@ -45,6 +45,20 @@ static void rx_callback(struct mbox_client *cl, void *m) { struct scmi_mailbox *smbox = client_to_scmi_mailbox(cl); + /* + * An A2P IRQ is NOT valid when received while the platform still has + * the ownership of the channel, because the platform at first releases + * the SMT channel and then sends the completion interrupt. + * + * This addresses a possible race condition in which a spurious IRQ from + * a previous timed-out reply which arrived late could be wrongly + * associated with the next pending transaction. + */ + if (cl->knows_txdone && !shmem_channel_free(smbox->shmem)) { + dev_warn(smbox->cinfo->dev, "Ignoring spurious A2P IRQ !\n"); + return; + } + scmi_rx_callback(smbox->cinfo, shmem_read_header(smbox->shmem), NULL); } diff --git a/drivers/firmware/arm_scmi/shmem.c b/drivers/firmware/arm_scmi/shmem.c index 87b4f4d35f06..517d52fb3bcb 100644 --- a/drivers/firmware/arm_scmi/shmem.c +++ b/drivers/firmware/arm_scmi/shmem.c @@ -122,3 +122,9 @@ bool shmem_poll_done(struct scmi_shared_mem __iomem *shmem, (SCMI_SHMEM_CHAN_STAT_CHANNEL_ERROR | SCMI_SHMEM_CHAN_STAT_CHANNEL_FREE); } + +bool shmem_channel_free(struct scmi_shared_mem __iomem *shmem) +{ + return (ioread32(&shmem->channel_status) & + SCMI_SHMEM_CHAN_STAT_CHANNEL_FREE); +} From 0726fcc8d4af75441b38aaa082f820e63b3a8748 Mon Sep 17 00:00:00 2001 From: Tanzir Hasan Date: Tue, 26 Dec 2023 22:52:03 +0000 Subject: [PATCH 2/6] firmware: arm_scmi: Replace asm-generic/bug.h with linux/bug.h linux/bug.h includes asm-generic/bug.h already and hence replacing asm-generic/bug.h with linux/bug.h will not regress any build. Also, it is always better to avoid header file inclusion from asm-generic if possible. Suggested-by: Al Viro Signed-off-by: Tanzir Hasan Link: https://lore.kernel.org/r/20231226-shmem-v1-1-ea15ce81d8ba@google.com Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/shmem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/firmware/arm_scmi/shmem.c b/drivers/firmware/arm_scmi/shmem.c index 517d52fb3bcb..8bf495bcad09 100644 --- a/drivers/firmware/arm_scmi/shmem.c +++ b/drivers/firmware/arm_scmi/shmem.c @@ -10,7 +10,7 @@ #include #include -#include +#include #include "common.h" From e8ef4bbe39b9576a73f104f6af743fb9c7b624ba Mon Sep 17 00:00:00 2001 From: Cristian Marussi Date: Mon, 8 Jan 2024 18:50:49 +0000 Subject: [PATCH 3/6] firmware: arm_scmi: Use xa_insert() to store opps When storing opps by level or index use xa_insert() instead of xa_store() and add error-checking to spot bad duplicates indexes possibly wrongly provided by the platform firmware. Fixes: 31c7c1397a33 ("firmware: arm_scmi: Add v3.2 perf level indexing mode support") Signed-off-by: Cristian Marussi Link: https://lore.kernel.org/r/20240108185050.1628687-1-cristian.marussi@arm.com Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/perf.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/drivers/firmware/arm_scmi/perf.c b/drivers/firmware/arm_scmi/perf.c index 8ea2a7b3d35d..211e8e0aef2c 100644 --- a/drivers/firmware/arm_scmi/perf.c +++ b/drivers/firmware/arm_scmi/perf.c @@ -350,8 +350,8 @@ process_response_opp(struct scmi_opp *opp, unsigned int loop_idx, } static inline void -process_response_opp_v4(struct perf_dom_info *dom, struct scmi_opp *opp, - unsigned int loop_idx, +process_response_opp_v4(struct device *dev, struct perf_dom_info *dom, + struct scmi_opp *opp, unsigned int loop_idx, const struct scmi_msg_resp_perf_describe_levels_v4 *r) { opp->perf = le32_to_cpu(r->opp[loop_idx].perf_val); @@ -362,10 +362,23 @@ process_response_opp_v4(struct perf_dom_info *dom, struct scmi_opp *opp, /* Note that PERF v4 reports always five 32-bit words */ opp->indicative_freq = le32_to_cpu(r->opp[loop_idx].indicative_freq); if (dom->level_indexing_mode) { + int ret; + opp->level_index = le32_to_cpu(r->opp[loop_idx].level_index); - xa_store(&dom->opps_by_idx, opp->level_index, opp, GFP_KERNEL); - xa_store(&dom->opps_by_lvl, opp->perf, opp, GFP_KERNEL); + ret = xa_insert(&dom->opps_by_idx, opp->level_index, opp, + GFP_KERNEL); + if (ret) + dev_warn(dev, + "Failed to add opps_by_idx at %d - ret:%d\n", + opp->level_index, ret); + + ret = xa_insert(&dom->opps_by_lvl, opp->perf, opp, GFP_KERNEL); + if (ret) + dev_warn(dev, + "Failed to add opps_by_lvl at %d - ret:%d\n", + opp->perf, ret); + hash_add(dom->opps_by_freq, &opp->hash, opp->indicative_freq); } } @@ -382,7 +395,7 @@ iter_perf_levels_process_response(const struct scmi_protocol_handle *ph, if (PROTOCOL_REV_MAJOR(p->version) <= 0x3) process_response_opp(opp, st->loop_idx, response); else - process_response_opp_v4(p->perf_dom, opp, st->loop_idx, + process_response_opp_v4(ph->dev, p->perf_dom, opp, st->loop_idx, response); p->perf_dom->opp_count++; From b5dc0ffd36560dbadaed9a3d9fd7838055d62d74 Mon Sep 17 00:00:00 2001 From: Cristian Marussi Date: Mon, 8 Jan 2024 18:50:50 +0000 Subject: [PATCH 4/6] firmware: arm_scmi: Use xa_insert() when saving raw queues Use xa_insert() when saving per-channel raw queues to better check for duplicates. Fixes: 7860701d1e6e ("firmware: arm_scmi: Add per-channel raw injection support") Signed-off-by: Cristian Marussi Link: https://lore.kernel.org/r/20240108185050.1628687-2-cristian.marussi@arm.com Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/raw_mode.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/firmware/arm_scmi/raw_mode.c b/drivers/firmware/arm_scmi/raw_mode.c index 0493aa3c12bf..350573518503 100644 --- a/drivers/firmware/arm_scmi/raw_mode.c +++ b/drivers/firmware/arm_scmi/raw_mode.c @@ -1111,7 +1111,6 @@ static int scmi_raw_mode_setup(struct scmi_raw_mode_info *raw, int i; for (i = 0; i < num_chans; i++) { - void *xret; struct scmi_raw_queue *q; q = scmi_raw_queue_init(raw); @@ -1120,13 +1119,12 @@ static int scmi_raw_mode_setup(struct scmi_raw_mode_info *raw, goto err_xa; } - xret = xa_store(&raw->chans_q, channels[i], q, + ret = xa_insert(&raw->chans_q, channels[i], q, GFP_KERNEL); - if (xa_err(xret)) { + if (ret) { dev_err(dev, "Fail to allocate Raw queue 0x%02X\n", channels[i]); - ret = xa_err(xret); goto err_xa; } } @@ -1322,6 +1320,12 @@ void scmi_raw_message_report(void *r, struct scmi_xfer *xfer, dev = raw->handle->dev; q = scmi_raw_queue_select(raw, idx, SCMI_XFER_IS_CHAN_SET(xfer) ? chan_id : 0); + if (!q) { + dev_warn(dev, + "RAW[%d] - NO queue for chan 0x%X. Dropping report.\n", + idx, chan_id); + return; + } /* * Grab the msg_q_lock upfront to avoid a possible race between From 27600c96e2ffa6c1b2cb378ddc75c6620c628d04 Mon Sep 17 00:00:00 2001 From: Cristian Marussi Date: Tue, 9 Jan 2024 15:01:06 +0000 Subject: [PATCH 5/6] firmware: arm_scmi: Fix the clock protocol version for v3.2 The clock protocol version as per the SCMI v3.2 specification is 0x30000. Enable the v3.0 clock protocol features only when clock protocol version equals 0x30000. The previous beta version of the spec had this value set to 0x20001 and th same value trickled down from the initial development. The version update were missed in the driver. Fixes: e49e314a2cf7 ("firmware: arm_scmi: Add clock v3.2 CONFIG_SET support") Signed-off-by: Cristian Marussi Link: https://lore.kernel.org/r/20240109150106.2066739-1-cristian.marussi@arm.com Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/clock.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/firmware/arm_scmi/clock.c b/drivers/firmware/arm_scmi/clock.c index c0644558042a..d6357513163c 100644 --- a/drivers/firmware/arm_scmi/clock.c +++ b/drivers/firmware/arm_scmi/clock.c @@ -954,8 +954,7 @@ static int scmi_clock_protocol_init(const struct scmi_protocol_handle *ph) scmi_clock_describe_rates_get(ph, clkid, clk); } - if (PROTOCOL_REV_MAJOR(version) >= 0x2 && - PROTOCOL_REV_MINOR(version) >= 0x1) { + if (PROTOCOL_REV_MAJOR(version) >= 0x3) { cinfo->clock_config_set = scmi_clock_config_set_v2; cinfo->clock_config_get = scmi_clock_config_get_v2; } else { From 6bd1b3fede83d8ba5314886062a9bfdada5102a9 Mon Sep 17 00:00:00 2001 From: Cristian Marussi Date: Tue, 9 Jan 2024 18:17:16 +0000 Subject: [PATCH 6/6] firmware: arm_scmi: Fix the clock protocol supported version Rollback currently supported SCMI clock protocol version to v2.0 since some of the mandatory v3.0 features are indeed still not supported yet. Fixes: b5efc28a754d ("firmware: arm_scmi: Add protocol versioning checks") Signed-off-by: Cristian Marussi Link: https://lore.kernel.org/r/20240109181716.2338636-1-cristian.marussi@arm.com Signed-off-by: Sudeep Holla --- drivers/firmware/arm_scmi/clock.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/firmware/arm_scmi/clock.c b/drivers/firmware/arm_scmi/clock.c index d6357513163c..e2050adbf85c 100644 --- a/drivers/firmware/arm_scmi/clock.c +++ b/drivers/firmware/arm_scmi/clock.c @@ -13,7 +13,7 @@ #include "notify.h" /* Updated only after ALL the mandatory features for that version are merged */ -#define SCMI_PROTOCOL_SUPPORTED_VERSION 0x20001 +#define SCMI_PROTOCOL_SUPPORTED_VERSION 0x20000 enum scmi_clock_protocol_cmd { CLOCK_ATTRIBUTES = 0x3,