From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.kernel.org (aws-us-west-2-korg-mail-1.web.codeaurora.org [10.30.226.201]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 8D2CB3EDE68; Tue, 12 May 2026 17:51:23 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=10.30.226.201 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1778608283; cv=none; b=SL3PMbsWBs89fPBKw2YDsXNqxxNrj4yyTE40A1FM3QeHqhIVrsuMStO85s6r4xXNDYenmcqeHytvQ4pK6/F0v1UMqgkDQi0UeJMK9x399LRA1/4jJXbHzPa6AiZ8NaqxEjAF6OPFiTOFdDxrtKHqq1fCuCT+EcfWqtFUYqM/n8k= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1778608283; c=relaxed/simple; bh=pZo2/NnhvoX9VYHRNERE1g+Q21iHflEIfUcTroEJBOQ=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=HCHGapiS0X/kBtTx+cQaDOw/O6LigZBcHxbzJFLd/ImbAsiNsBo8VN1OQzEHNDGnvvuT9foaz2qbqiOO/zDyoRO2RWBS9rPFbVlw9tRQ3QZaq7xe3a6+bt/hOxojtogD/aVP+DG/hbfO0MpCBM2K4epb3Zh0kFbTN/3jMsX3Sbc= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b=dG2oTm9e; arc=none smtp.client-ip=10.30.226.201 Authentication-Results: smtp.subspace.kernel.org; dkim=pass (1024-bit key) header.d=linuxfoundation.org header.i=@linuxfoundation.org header.b="dG2oTm9e" Received: by smtp.kernel.org (Postfix) with ESMTPSA id 232D6C2BCB0; Tue, 12 May 2026 17:51:22 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=linuxfoundation.org; s=korg; t=1778608283; bh=pZo2/NnhvoX9VYHRNERE1g+Q21iHflEIfUcTroEJBOQ=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=dG2oTm9ey6g3/bMDjM3wdNpQJmNXmHLIu6Bdm9MyCji+4DpjAwYY7dC24IRPrkB2M 2Nvy5/QIDOtLL7ihwxd+C4E31MdbSCNFlC77qdCSC4MrBZOd3yHE7TFk5bPnrGLwOm OLYaBv4TeSaH8VaVieWPTuwgiUOsLQTuwyGIPjOM= From: Greg Kroah-Hartman To: stable@vger.kernel.org Cc: Greg Kroah-Hartman , patches@lists.linux.dev, Matt Fleming , Corey Minyard Subject: [PATCH 6.18 002/270] ipmi: Add limits to event and receive message requests Date: Tue, 12 May 2026 19:36:43 +0200 Message-ID: <20260512173938.506500264@linuxfoundation.org> X-Mailer: git-send-email 2.54.0 In-Reply-To: <20260512173938.452574370@linuxfoundation.org> References: <20260512173938.452574370@linuxfoundation.org> User-Agent: quilt/0.69 X-stable: review X-Patchwork-Hint: ignore Precedence: bulk X-Mailing-List: stable@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit 6.18-stable review patch. If anyone has any objections, please let me know. ------------------ From: Corey Minyard commit c4cca236968683eb0d59abfb12d5c7e4d8514227 upstream. The driver would just fetch events and receive messages until the BMC said it was done. To avoid issues with BMCs that never say they are done, add a limit of 10 fetches at a time. In addition, an si interface has an attn state it can return from the hardware which is supposed to cause a flag fetch to see if the driver needs to fetch events or message or a few other things. If the attn bit gets stuck, it's a similar problem. So allow messages in between flag fetches so the driver itself doesn't get stuck. This is a more general fix than the previous fix for the specific bad BMC, but should fix the more general issue of a BMC that won't stop saying it has data. This has been there from the beginning of the driver. It's not a bug per-se, but it is accounting for bugs in BMCs. Reported-by: Matt Fleming Closes: https://lore.kernel.org/lkml/20260415115930.3428942-1-matt@readmodwrite.com/ Fixes: <1da177e4c3f4> ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Signed-off-by: Corey Minyard Signed-off-by: Greg Kroah-Hartman --- drivers/char/ipmi/ipmi_si_intf.c | 54 +++++++++++++++++++++++++++++++-------- drivers/char/ipmi/ipmi_ssif.c | 23 +++++++++++++++- 2 files changed, 64 insertions(+), 13 deletions(-) --- a/drivers/char/ipmi/ipmi_si_intf.c +++ b/drivers/char/ipmi/ipmi_si_intf.c @@ -168,6 +168,10 @@ struct smi_info { OEM2_DATA_AVAIL) unsigned char msg_flags; + /* When requesting events and messages, don't do it forever. */ + unsigned int num_requests_in_a_row; + bool last_was_flag_fetch; + /* Does the BMC have an event buffer? */ bool has_event_buffer; @@ -411,7 +415,10 @@ static void start_getting_msg_queue(stru start_new_msg(smi_info, smi_info->curr_msg->data, smi_info->curr_msg->data_size); - smi_info->si_state = SI_GETTING_MESSAGES; + if (smi_info->si_state != SI_GETTING_MESSAGES) { + smi_info->num_requests_in_a_row = 0; + smi_info->si_state = SI_GETTING_MESSAGES; + } } static void start_getting_events(struct smi_info *smi_info) @@ -422,7 +429,10 @@ static void start_getting_events(struct start_new_msg(smi_info, smi_info->curr_msg->data, smi_info->curr_msg->data_size); - smi_info->si_state = SI_GETTING_EVENTS; + if (smi_info->si_state != SI_GETTING_EVENTS) { + smi_info->num_requests_in_a_row = 0; + smi_info->si_state = SI_GETTING_EVENTS; + } } /* @@ -596,6 +606,7 @@ static void handle_transaction_done(stru smi_info->si_state = SI_NORMAL; } else { smi_info->msg_flags = msg[3]; + smi_info->last_was_flag_fetch = true; handle_flags(smi_info); } break; @@ -641,6 +652,11 @@ static void handle_transaction_done(stru } else { smi_inc_stat(smi_info, events); + smi_info->num_requests_in_a_row++; + if (smi_info->num_requests_in_a_row > 10) + /* Stop if we do this too many times. */ + smi_info->msg_flags &= ~EVENT_MSG_BUFFER_FULL; + /* * Do this before we deliver the message * because delivering the message releases the @@ -679,6 +695,11 @@ static void handle_transaction_done(stru } else { smi_inc_stat(smi_info, incoming_messages); + smi_info->num_requests_in_a_row++; + if (smi_info->num_requests_in_a_row > 10) + /* Stop if we do this too many times. */ + smi_info->msg_flags &= ~RECEIVE_MSG_AVAIL; + /* * Do this before we deliver the message * because delivering the message releases the @@ -821,6 +842,26 @@ restart: } /* + * If we are currently idle, or if the last thing that was + * done was a flag fetch and there is a message pending, try + * to start the next message. + * + * We do the waiting message check to avoid a stuck flag + * completely wedging the driver. Let a message through + * in between flag operations if that happens. + */ + if (si_sm_result == SI_SM_IDLE || + (si_sm_result == SI_SM_ATTN && smi_info->waiting_msg && + smi_info->last_was_flag_fetch)) { + smi_info->last_was_flag_fetch = false; + smi_inc_stat(smi_info, idles); + + si_sm_result = start_next_msg(smi_info); + if (si_sm_result != SI_SM_IDLE) + goto restart; + } + + /* * We prefer handling attn over new messages. But don't do * this if there is not yet an upper layer to handle anything. */ @@ -847,15 +888,6 @@ restart: } } - /* If we are currently idle, try to start the next message. */ - if (si_sm_result == SI_SM_IDLE) { - smi_inc_stat(smi_info, idles); - - si_sm_result = start_next_msg(smi_info); - if (si_sm_result != SI_SM_IDLE) - goto restart; - } - if ((si_sm_result == SI_SM_IDLE) && (atomic_read(&smi_info->req_events))) { /* --- a/drivers/char/ipmi/ipmi_ssif.c +++ b/drivers/char/ipmi/ipmi_ssif.c @@ -225,6 +225,9 @@ struct ssif_info { bool has_event_buffer; bool supports_alert; + /* When requesting events and messages, don't do it forever. */ + unsigned int num_requests_in_a_row; + /* * Used to tell what we should do with alerts. If we are * waiting on a response, read the data immediately. @@ -413,7 +416,10 @@ static void start_event_fetch(struct ssi } ssif_info->curr_msg = msg; - ssif_info->ssif_state = SSIF_GETTING_EVENTS; + if (ssif_info->ssif_state != SSIF_GETTING_EVENTS) { + ssif_info->num_requests_in_a_row = 0; + ssif_info->ssif_state = SSIF_GETTING_EVENTS; + } ipmi_ssif_unlock_cond(ssif_info, flags); msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2); @@ -436,7 +442,10 @@ static void start_recv_msg_fetch(struct } ssif_info->curr_msg = msg; - ssif_info->ssif_state = SSIF_GETTING_MESSAGES; + if (ssif_info->ssif_state != SSIF_GETTING_MESSAGES) { + ssif_info->num_requests_in_a_row = 0; + ssif_info->ssif_state = SSIF_GETTING_MESSAGES; + } ipmi_ssif_unlock_cond(ssif_info, flags); msg->data[0] = (IPMI_NETFN_APP_REQUEST << 2); @@ -843,6 +852,11 @@ static void msg_done_handler(struct ssif ssif_info->msg_flags &= ~EVENT_MSG_BUFFER_FULL; handle_flags(ssif_info, flags); } else { + ssif_info->num_requests_in_a_row++; + if (ssif_info->num_requests_in_a_row > 10) + /* Stop if we do this too many times. */ + ssif_info->msg_flags &= ~EVENT_MSG_BUFFER_FULL; + handle_flags(ssif_info, flags); ssif_inc_stat(ssif_info, events); deliver_recv_msg(ssif_info, msg); @@ -876,6 +890,11 @@ static void msg_done_handler(struct ssif ssif_info->msg_flags &= ~RECEIVE_MSG_AVAIL; handle_flags(ssif_info, flags); } else { + ssif_info->num_requests_in_a_row++; + if (ssif_info->num_requests_in_a_row > 10) + /* Stop if we do this too many times. */ + ssif_info->msg_flags &= ~RECEIVE_MSG_AVAIL; + ssif_inc_stat(ssif_info, incoming_messages); handle_flags(ssif_info, flags); deliver_recv_msg(ssif_info, msg);