Linux SCSI subsystem development
 help / color / mirror / Atom feed
From: sashiko-bot@kernel.org
To: "Nilesh Javali" <njavali@marvell.com>
Cc: linux-scsi@vger.kernel.org
Subject: Re: [PATCH v3 63/88] scsi: qla2xxx: Zero mailbox struct in qla2x00_get_firmware_state()
Date: Tue, 07 Jul 2026 07:18:15 +0000	[thread overview]
Message-ID: <20260707071816.062DD1F000E9@smtp.kernel.org> (raw)
In-Reply-To: <20260707055435.2680300-64-njavali@marvell.com>

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:

Pre-existing issues:
- [High] Uninitialized kernel stack memory disclosure vulnerability remains unfixed in the sysfs handler `qla2x00_fw_state_show()` on the success path.
--

--- Patch [63]: [PATCH v3 63/88] scsi: qla2xxx: Zero mailbox struct in qla2x00_get_firmware_state() ---
commit 9d1cc31f70b4b1282c00f58b1e61ed39b581f84d
Author: Nilesh Javali <njavali@marvell.com>

scsi: qla2xxx: Zero mailbox struct in qla2x00_get_firmware_state()

This commit zero-initializes the mailbox command structure in
qla2x00_get_firmware_state() to prevent leaking uninitialized kernel stack
memory on failure paths. It ensures that when early returns occur, userspace
receives deterministic zeroed state rather than stack contents.

> diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c
> index ba4a4764de1fb..ab5648eb5f206 100644
> --- a/drivers/scsi/qla2xxx/qla_mbx.c
> +++ b/drivers/scsi/qla2xxx/qla_mbx.c
> @@ -2276,6 +2276,8 @@ qla2x00_get_firmware_state(scsi_qla_host_t *vha, uint16_t *states)
>  	if (!ha->flags.fw_started)
>  		return QLA_FUNCTION_FAILED;
>  
> +	memset(&mc, 0, sizeof(mc));
> +
>  	mcp->mb[0] = MBC_GET_FIRMWARE_STATE;
>  	mcp->out_mb = MBX_0;
>  	if (IS_FWI2_CAPABLE(vha->hw))

[Severity: High]
This is a pre-existing issue, but does this still leak uninitialized stack
memory on the success path for adapters that are not FWI2 capable?

Looking at the caller qla2x00_fw_state_show() in
drivers/scsi/qla2xxx/qla_attr.c, an uninitialized array is declared on the
stack and passed to qla2x00_get_firmware_state():

drivers/scsi/qla2xxx/qla_attr.c:qla2x00_fw_state_show() {
	uint16_t state[16];
        ...
	rval = qla2x00_get_firmware_state(vha, state);
        ...
}

When qla2x00_get_firmware_state() completes successfully, it only populates
states[1] through states[5] if the adapter is FWI2 capable:

drivers/scsi/qla2xxx/qla_mbx.c:qla2x00_get_firmware_state() {
	states[0] = mcp->mb[1];
	if (IS_FWI2_CAPABLE(vha->hw)) {
		states[1] = mcp->mb[2];
		states[2] = mcp->mb[3];  /* SFP info */
		states[3] = mcp->mb[4];
		states[4] = mcp->mb[5];
		states[5] = mcp->mb[6];  /* DPORT status */
        ...
}

However, qla2x00_fw_state_show() unconditionally formats all six elements into
the sysfs output buffer, which could expose uninitialized stack memory to
userspace for non-FWI2 adapters:

drivers/scsi/qla2xxx/qla_attr.c:qla2x00_fw_state_show() {
        ...
	return scnprintf(buf, PAGE_SIZE, "0x%x 0x%x 0x%x 0x%x 0x%x 0x%x\n",
	    state[0], state[1], state[2], state[3], state[4], state[5]);
}

Should the state array be zero-initialized in qla2x00_fw_state_show(), or
should qla2x00_get_firmware_state() guarantee all returned values are
initialized regardless of the adapter capability?

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260707055435.2680300-1-njavali@marvell.com?part=63

  reply	other threads:[~2026-07-07  7:18 UTC|newest]

Thread overview: 159+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-07-07  5:53 [PATCH v3 00/88] scsi: qla2xxx: Add QLA29xx series adapter support Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 01/88] scsi: qla2xxx: Add 29xx series PCI device ID support Nilesh Javali
2026-07-07  6:21   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 02/88] scsi: qla2xxx: Add flash read/write interface for 29xx Nilesh Javali
2026-07-07  8:24   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 03/88] scsi: qla2xxx: Add NVRAM config support for 29xx adapters Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 04/88] scsi: qla2xxx: Add 29xx support in queue initialisation path Nilesh Javali
2026-07-07  6:31   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 05/88] scsi: qla2xxx: Add FC operational firmware load for 29xx Nilesh Javali
2026-07-07  6:17   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 06/88] scsi: qla2xxx: Remove redundant VPD flash read in sysfs read path Nilesh Javali
2026-07-07  6:13   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 07/88] scsi: qla2xxx: Add flash block read/write BSG support for 29xx Nilesh Javali
2026-07-07  6:21   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 08/88] scsi: qla2xxx: Add BSG MPI firmware load/dump " Nilesh Javali
2026-07-07  6:13   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 09/88] scsi: qla2xxx: Add 128-byte IOCB definitions " Nilesh Javali
2026-07-07  6:14   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 10/88] scsi: qla2xxx: Add extended status continuation and marker IOCBs Nilesh Javali
2026-07-07  6:10   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 11/88] scsi: qla2xxx: Update IO path to use 128-byte IOCBs for 29xx Nilesh Javali
2026-07-07 10:27   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 12/88] scsi: qla2xxx: Skip image-set-valid attribute " Nilesh Javali
2026-07-07  6:12   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 13/88] scsi: qla2xxx: Skip unsupported sysfs attributes " Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 14/88] scsi: qla2xxx: Enable get_fw_version mailbox " Nilesh Javali
2026-07-07  6:12   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 15/88] scsi: qla2xxx: Extend execute_fw mailbox to include 29xx Nilesh Javali
2026-07-07  6:23   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 16/88] scsi: qla2xxx: Enable get_adapter_id mailbox for 29xx Nilesh Javali
2026-07-07  6:15   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 17/88] scsi: qla2xxx: Enable init_firmware " Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 18/88] scsi: qla2xxx: Enable get_firmware_state " Nilesh Javali
2026-07-07  6:15   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 19/88] scsi: qla2xxx: Enable serdes, resource count and FCE trace " Nilesh Javali
2026-07-07  6:36   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 20/88] scsi: qla2xxx: Enable set_els_cmds and echo_test " Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 21/88] scsi: qla2xxx: Add support for QLA29XX in data rate functions Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 22/88] scsi: qla2xxx: Enable qla2x00_shutdown for 29xx Nilesh Javali
2026-07-07  6:27   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 23/88] scsi: qla2xxx: Use ring-slot helpers in __qla2x00_alloc_iocbs Nilesh Javali
2026-07-07  6:26   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 24/88] scsi: qla2xxx: Add support for QLA29XX in memory allocation Nilesh Javali
2026-07-07  6:22   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 25/88] scsi: qla2xxx: Handle sts_cont_entry_ext_t for 29xx adapters Nilesh Javali
2026-07-07  6:31   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 26/88] scsi: qla2xxx: Update handling of status entries for 29xx series Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 27/88] scsi: qla2xxx: Enhance ct_entry_24xx_ext iocb handling " Nilesh Javali
2026-07-07  6:40   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 28/88] scsi: qla2xxx: Enhance purex_entry " Nilesh Javali
2026-07-07  6:35   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 29/88] scsi: qla2xxx: Update handling of ELS IOCBs " Nilesh Javali
2026-07-07  6:35   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 30/88] scsi: qla2xxx: Add size check for ELS status entry layout on 29xx Nilesh Javali
2026-07-07  6:31   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 31/88] scsi: qla2xxx: Add 29xx extended logio IOCB support Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 32/88] scsi: qla2xxx: Enhance task management IOCB handling for 29xx series Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 33/88] scsi: qla2xxx: Add abort command " Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 34/88] scsi: qla2xxx: Enhance ABTS processing " Nilesh Javali
2026-07-07  6:46   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 35/88] scsi: qla2xxx: Update VP control IOCB handling " Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 36/88] scsi: qla2xxx: Add build-time size check for VP config IOCB layout Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 37/88] scsi: qla2xxx: Add size check for extended VP report ID entry Nilesh Javali
2026-07-07  6:46   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 38/88] scsi: qla2xxx: Add LS4 pass-through IOCB handling for 29xx series Nilesh Javali
2026-07-07  6:45   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 39/88] scsi: qla2xxx: Adjust feature gating in BSG paths for 29xx support Nilesh Javali
2026-07-07  6:48   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 40/88] scsi: qla2xxx: Fix queue teardown NULL dma_free and bitmap locking Nilesh Javali
2026-07-07  6:50   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 41/88] scsi: qla2xxx: Replace __le16 bitfields with scalar and accessors Nilesh Javali
2026-07-07  6:49   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 42/88] scsi: qla2xxx: Fix endianness annotations in vp_rpt_id_entry structures Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 43/88] scsi: qla2xxx: Use 64-bit FPM word counters for 29xx host stats Nilesh Javali
2026-07-07  6:49   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 44/88] scsi: qla2xxx: Add 64G/128G port speed setting support Nilesh Javali
2026-07-07  6:52   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 45/88] scsi: qla2xxx: Fix 64G link speed reporting in get_data_rate Nilesh Javali
2026-07-07  5:53 ` [PATCH v3 46/88] scsi: qla2xxx: edif: Fix NULL pointer deref in RX SA delete check Nilesh Javali
2026-07-07  7:09   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 47/88] scsi: qla2xxx: Fix Name Server logout detection on FWI2 adapters Nilesh Javali
2026-07-07  6:53   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 48/88] scsi: qla2xxx: Bound VP index against VP_CTRL IOCB bitmap size Nilesh Javali
2026-07-07  6:55   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 49/88] scsi: qla2xxx: Check entry_status in qla24xx_modify_vp_config() Nilesh Javali
2026-07-07  6:54   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 50/88] scsi: qla2xxx: Hold vport reference in qla24xx_report_id_acquisition() Nilesh Javali
2026-07-07  7:03   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 51/88] scsi: qla2xxx: Initialize NVMe abort_work once at submission Nilesh Javali
2026-07-07  7:00   ` sashiko-bot
2026-07-07  5:53 ` [PATCH v3 52/88] scsi: qla2xxx: Hold qpair lock when sending NVMe LS reject Nilesh Javali
2026-07-07  7:07   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 53/88] scsi: qla2xxx: Zero dport diagnostics buffer to avoid info leak Nilesh Javali
2026-07-07  7:06   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 54/88] scsi: qla2xxx: Fix BSG job leak on validate flash image error path Nilesh Javali
2026-07-07  7:03   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 55/88] scsi: qla2xxx: Bound image count in qla2x00_update_fru_versions() Nilesh Javali
2026-07-07  7:05   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 56/88] scsi: qla2xxx: Clamp MSI-X derived queue counts to avoid truncation Nilesh Javali
2026-07-07  7:06   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 57/88] scsi: qla2xxx: Improve firmware dump data capture Nilesh Javali
2026-07-07  7:18   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 58/88] scsi: qla2xxx: Serialize flash version read in reset handler Nilesh Javali
2026-07-07  5:54 ` [PATCH v3 59/88] scsi: qla2xxx: Fix use-after-free of qpair work on queue teardown Nilesh Javali
2026-07-07  7:08   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 60/88] scsi: qla2xxx: Clarify MPI optrom address/length units Nilesh Javali
2026-07-07  5:54 ` [PATCH v3 61/88] scsi: qla2xxx: Fix cs84xx use-after-free on host teardown Nilesh Javali
2026-07-07  7:11   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 62/88] scsi: qla2xxx: Don't query firmware state while chip is down Nilesh Javali
2026-07-07  7:03   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 63/88] scsi: qla2xxx: Zero mailbox struct in qla2x00_get_firmware_state() Nilesh Javali
2026-07-07  7:18   ` sashiko-bot [this message]
2026-07-07  5:54 ` [PATCH v3 64/88] scsi: qla2xxx: Fix FCE trace enable parsing in debugfs Nilesh Javali
2026-07-07  7:11   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 65/88] scsi: qla2xxx: Fix FCE trace use-after-free during firmware dump Nilesh Javali
2026-07-07  7:25   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 66/88] scsi: qla2xxx: Use memset_io() to clear QLAFX00 request ring slot Nilesh Javali
2026-07-07  7:26   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 67/88] scsi: qla2xxx: Null out freed pointers in qla2x00_mem_alloc() error path Nilesh Javali
2026-07-07  7:25   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 68/88] scsi: qla2xxx: Fix response queue over-consumption in __qla_consume_iocb() Nilesh Javali
2026-07-07  5:54 ` [PATCH v3 69/88] scsi: qla2xxx: Fix soft lockup polling continuation IOCB signature Nilesh Javali
2026-07-07  7:19   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 70/88] scsi: qla2xxx: Bound rsp_info_len to avoid OOB sense-data read Nilesh Javali
2026-07-07  7:24   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 71/88] scsi: qla2xxx: Avoid req_q_map double-read in qla2x00_error_entry() Nilesh Javali
2026-07-07  7:21   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 72/88] scsi: qla2xxx: Quiesce response IRQ before freeing request queue Nilesh Javali
2026-07-07  7:22   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 73/88] scsi: qla2xxx: Reject non-SCSI SRB on status IOCB fast path Nilesh Javali
2026-07-07  5:54 ` [PATCH v3 74/88] scsi: qla2xxx: Clamp max_npiv_vports to VP_CTRL bitmap capacity Nilesh Javali
2026-07-07  7:34   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 75/88] scsi: qla2xxx: Avoid double completion in async IOCB timeout Nilesh Javali
2026-07-07  7:37   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 76/88] scsi: qla2xxx: Skip vport under deletion in report ID acquisition Nilesh Javali
2026-07-07  7:27   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 77/88] scsi: qla2xxx: Drop vport reference under lock " Nilesh Javali
2026-07-07  7:33   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 78/88] scsi: qla2xxx: Hold vport_slock for host map update " Nilesh Javali
2026-07-07  7:28   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 79/88] scsi: qla2xxx: Fix NVMe abort reference leak on repeated abort Nilesh Javali
2026-07-07  7:27   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 80/88] scsi: qla2xxx: Skip NVMe LS reject IOCB when FW not started Nilesh Javali
2026-07-07  7:38   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 81/88] scsi: qla2xxx: Unlink NVMe unsol ctx before freeing on LS reject error Nilesh Javali
2026-07-07  7:34   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 82/88] scsi: qla2xxx: Serialize NVMe unsol ctx list with a per-fcport lock Nilesh Javali
2026-07-07  7:38   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 83/88] scsi: qla2xxx: Use coherent DMA buffer for D_Port diagnostics Nilesh Javali
2026-07-07  7:34   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 84/88] scsi: qla2xxx: Zero-init bsg stack buffers to avoid info leak Nilesh Javali
2026-07-07  7:37   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 85/88] scsi: qla2xxx: Validate BSG request_len before reading vendor_cmd[] Nilesh Javali
2026-07-07  7:46   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 86/88] scsi: qla2xxx: Zero SFP DMA buffer in FRU/I2C bsg handlers Nilesh Javali
2026-07-07  7:38   ` sashiko-bot
2026-07-07  5:54 ` [PATCH v3 87/88] scsi: qla2xxx: Bound i2c->length in I2C " Nilesh Javali
2026-07-07  5:54 ` [PATCH v3 88/88] scsi: qla2xxx: Update version to 12.00.00.2607b1 Nilesh Javali
2026-07-07  7:35   ` sashiko-bot

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20260707071816.062DD1F000E9@smtp.kernel.org \
    --to=sashiko-bot@kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=njavali@marvell.com \
    --cc=sashiko-reviews@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox