Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Dan Carpenter <dan.carpenter@oracle.com>
To: kbuild@lists.01.org, Sean Z Huang <sean.z.huang@intel.com>,
	Intel-gfx@lists.freedesktop.org
Cc: "Huang, Sean Z" <sean.z.huang@intel.com>,
	kbuild-all@lists.01.org, Dan Carpenter <error27@gmail.com>
Subject: Re: [Intel-gfx] [PATCH 05/27] drm/i915/pxp: Enable ioctl action to set the ring3 context
Date: Mon, 16 Nov 2020 12:46:06 +0300	[thread overview]
Message-ID: <20201116094606.GX29398@kadam> (raw)
In-Reply-To: <20201114014537.25495-5-sean.z.huang@intel.com>

[-- Attachment #1: Type: text/plain, Size: 5905 bytes --]

Hi Sean,

url:    https://github.com/0day-ci/linux/commits/Sean-Z-Huang/drm-i915-pxp-Introduce-Intel-PXP-component/20201114-094926
base:    92edc4aef86780a8ad01b092c6d6630bb3cb423d
config: i386-randconfig-m021-20201115 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-15) 9.3.0

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>

smatch warnings:
drivers/gpu/drm/i915/pxp/intel_pxp.c:62 i915_pxp_ops_ioctl() error: we previously assumed 'i915' could be null (see line 20)

vim +/i915 +62 drivers/gpu/drm/i915/pxp/intel_pxp.c

1d109ada10e82c3 Huang, Sean Z 2020-11-13  11  int i915_pxp_ops_ioctl(struct drm_device *dev, void *data, struct drm_file *drmfile)
1d109ada10e82c3 Huang, Sean Z 2020-11-13  12  {
1d109ada10e82c3 Huang, Sean Z 2020-11-13  13  	int ret;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  14  	struct pxp_info pxp_info = {0};
1d109ada10e82c3 Huang, Sean Z 2020-11-13  15  	struct drm_i915_pxp_ops *pxp_ops = data;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  16  	struct drm_i915_private *i915 = to_i915(dev);
1d109ada10e82c3 Huang, Sean Z 2020-11-13  17  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  18  	drm_dbg(&i915->drm, ">>> %s\n", __func__);
                                                         ^^^^^^^^^
If "i915" is NULL then this will crash.

1d109ada10e82c3 Huang, Sean Z 2020-11-13  19  
1d109ada10e82c3 Huang, Sean Z 2020-11-13 @20  	if (!i915 || !drmfile || !pxp_ops || pxp_ops->pxp_info_size != sizeof(pxp_info)) {
                                                     ^^^^
Check too late.

1d109ada10e82c3 Huang, Sean Z 2020-11-13  21  		drm_dbg(&i915->drm, "Failed to %s, invalid params\n", __func__);
1d109ada10e82c3 Huang, Sean Z 2020-11-13  22  		ret = -EINVAL;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  23  		goto end;

This will unlock a lock that we are not holding.  This should just
"return -EINVAL;".  All the stuff at "goto end;" is pointless or buggy.

1d109ada10e82c3 Huang, Sean Z 2020-11-13  24  	}
1d109ada10e82c3 Huang, Sean Z 2020-11-13  25  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  26  	if (copy_from_user(&pxp_info, (void __user *)pxp_ops->pxp_info_ptr, sizeof(pxp_info)) != 0) {
1d109ada10e82c3 Huang, Sean Z 2020-11-13  27  		ret = -EFAULT;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  28  		goto end;
                                                        ^^^^^^^^
This will unlock.  Same.  Just return -EFAULT;

1d109ada10e82c3 Huang, Sean Z 2020-11-13  29  	}
1d109ada10e82c3 Huang, Sean Z 2020-11-13  30  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  31  	drm_dbg(&i915->drm, "i915 pxp ioctl call with action=[%d]\n", pxp_info.action);
1d109ada10e82c3 Huang, Sean Z 2020-11-13  32  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  33  	mutex_lock(&i915->pxp.r0ctx->ctx_mutex);
1d109ada10e82c3 Huang, Sean Z 2020-11-13  34  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  35  	if (i915->pxp.r0ctx->global_state_in_suspend) {
1d109ada10e82c3 Huang, Sean Z 2020-11-13  36  		drm_dbg(&i915->drm, "Return failure due to state in suspend\n");
1d109ada10e82c3 Huang, Sean Z 2020-11-13  37  		pxp_info.sm_status = PXP_SM_STATUS_SESSION_NOT_AVAILABLE;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  38  		ret = 0;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  39  		goto end;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  40  	}
1d109ada10e82c3 Huang, Sean Z 2020-11-13  41  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  42  	if (i915->pxp.r0ctx->global_state_attacked) {
1d109ada10e82c3 Huang, Sean Z 2020-11-13  43  		drm_dbg(&i915->drm, "Retry required due to state attacked\n");
1d109ada10e82c3 Huang, Sean Z 2020-11-13  44  		pxp_info.sm_status = PXP_SM_STATUS_RETRY_REQUIRED;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  45  		ret = 0;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  46  		goto end;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  47  	}
1d109ada10e82c3 Huang, Sean Z 2020-11-13  48  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  49  	switch (pxp_info.action) {
1d109ada10e82c3 Huang, Sean Z 2020-11-13  50  	case PXP_ACTION_SET_R3_CONTEXT:
1d109ada10e82c3 Huang, Sean Z 2020-11-13  51  	{
1d109ada10e82c3 Huang, Sean Z 2020-11-13  52  		ret = intel_pxp_set_r3ctx(i915, pxp_info.set_r3ctx);
1d109ada10e82c3 Huang, Sean Z 2020-11-13  53  		break;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  54  	}
1d109ada10e82c3 Huang, Sean Z 2020-11-13  55  	default:
1d109ada10e82c3 Huang, Sean Z 2020-11-13  56  		drm_dbg(&i915->drm, "Failed to %s due to bad params\n", __func__);
1d109ada10e82c3 Huang, Sean Z 2020-11-13  57  		ret = -EINVAL;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  58  		goto end;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  59  	}
1d109ada10e82c3 Huang, Sean Z 2020-11-13  60  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  61  end:
1d109ada10e82c3 Huang, Sean Z 2020-11-13 @62  	mutex_unlock(&i915->pxp.r0ctx->ctx_mutex);
                                                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

1d109ada10e82c3 Huang, Sean Z 2020-11-13  63  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  64  	if (ret == 0)
1d109ada10e82c3 Huang, Sean Z 2020-11-13  65  		if (copy_to_user((void __user *)pxp_ops->pxp_info_ptr, &pxp_info, sizeof(pxp_info)) != 0)
1d109ada10e82c3 Huang, Sean Z 2020-11-13  66  			ret = -EFAULT;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  67  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  68  	if (ret)
1d109ada10e82c3 Huang, Sean Z 2020-11-13  69  		dev_err(&dev->pdev->dev, "pid=%d, ret = %d\n", task_pid_nr(current), ret);
1d109ada10e82c3 Huang, Sean Z 2020-11-13  70  
1d109ada10e82c3 Huang, Sean Z 2020-11-13  71  	drm_dbg(&i915->drm, "<<< %s\n", __func__);

Delete this printk() and use ftrace for this information.

1d109ada10e82c3 Huang, Sean Z 2020-11-13  72  	return ret;
1d109ada10e82c3 Huang, Sean Z 2020-11-13  73  }

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 46446 bytes --]

[-- Attachment #3: Type: text/plain, Size: 160 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2020-11-16  9:46 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-11-14  1:45 [Intel-gfx] [PATCH 01/27] drm/i915/pxp: Introduce Intel PXP component Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 02/27] drm/i915/pxp: Enable PXP irq worker and callback stub Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 03/27] drm/i915/pxp: Add PXP context for logical hardware states Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 04/27] drm/i915/pxp: set KCR reg init during the boot time Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 05/27] drm/i915/pxp: Enable ioctl action to set the ring3 context Sean Z Huang
2020-11-14  4:23   ` kernel test robot
2020-11-16  9:46   ` Dan Carpenter [this message]
2020-11-14  1:45 ` [Intel-gfx] [PATCH 06/27] drm/i915: Rename the whitelist to allowlist Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 07/27] drm/i915/pxp: Add PXP-related registers into allowlist Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 08/27] drm/i915/pxp: Read register to check hardware session state Sean Z Huang
2020-11-14  5:04   ` kernel test robot
2020-11-14  5:04   ` [Intel-gfx] [RFC PATCH] drm/i915/pxp: pxp_sm_reg_read() can be static kernel test robot
2020-11-14  5:14   ` [Intel-gfx] [PATCH 08/27] drm/i915/pxp: Read register to check hardware session state kernel test robot
2020-11-14  1:45 ` [Intel-gfx] [PATCH 09/27] drm/i915/pxp: Implement funcs to get/set PXP tag Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 10/27] drm/i915/pxp: Enable ioctl action to reserve session slot Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 11/27] drm/i915/pxp: Enable ioctl action to set session in play Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 12/27] drm/i915/pxp: Func to send hardware session termination Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 13/27] drm/i915/pxp: Enable ioctl action to terminate the session Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 14/27] drm/i915/pxp: Enable ioctl action to query PXP tag Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 15/27] drm/i915/pxp: Destroy all type0 sessions upon teardown Sean Z Huang
2020-11-14  5:51   ` kernel test robot
2020-11-14  5:51   ` [Intel-gfx] [RFC PATCH] drm/i915/pxp: intel_pxp_sm_destroy_all_sw_sessions() can be static kernel test robot
2020-11-14  6:28   ` [Intel-gfx] [PATCH 15/27] drm/i915/pxp: Destroy all type0 sessions upon teardown kernel test robot
2020-11-14  1:45 ` [Intel-gfx] [PATCH 16/27] drm/i915/pxp: Termiante the session upon app crash Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 17/27] drm/i915/pxp: Enable PXP power management Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 18/27] drm/i915/pxp: Implement funcs to create the TEE channel Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 19/27] drm/i915/pxp: Enable ioctl action to send TEE commands Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 20/27] drm/i915/pxp: Create the arbitrary session after boot Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 21/27] drm/i915/pxp: Add i915 trace logs for PXP operations Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 22/27] drm/i915/pxp: Expose session state for display protection flip Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 23/27] mei: bus: enable pavp device Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 24/27] mei: pxp: export pavp client to me client bus Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 25/27] drm/i915/uapi: introduce drm_i915_gem_create_ext for TGL Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 26/27] drm/i915/pavp: User interface for Protected buffer Sean Z Huang
2020-11-14  1:45 ` [Intel-gfx] [PATCH 27/27] drm/i915/pxp: Add plane decryption support Sean Z Huang
2020-11-14  1:59 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for series starting with [01/27] drm/i915/pxp: Introduce Intel PXP component Patchwork
  -- strict thread matches above, loose matches on Subject: below --
2020-11-15 20:23 [Intel-gfx] [PATCH 01/27] " Huang, Sean Z
2020-11-15 20:23 ` [Intel-gfx] [PATCH 05/27] drm/i915/pxp: Enable ioctl action to set the ring3 context Huang, Sean Z
2020-11-15 21:07 [Intel-gfx] [PATCH 01/27] drm/i915/pxp: Introduce Intel PXP component Huang, Sean Z
2020-11-15 21:07 ` [Intel-gfx] [PATCH 05/27] drm/i915/pxp: Enable ioctl action to set the ring3 context Huang, Sean Z
2020-11-16 10:22   ` Joonas Lahtinen
2020-11-16 10:53     ` Chris Wilson

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=20201116094606.GX29398@kadam \
    --to=dan.carpenter@oracle.com \
    --cc=Intel-gfx@lists.freedesktop.org \
    --cc=error27@gmail.com \
    --cc=kbuild-all@lists.01.org \
    --cc=kbuild@lists.01.org \
    --cc=sean.z.huang@intel.com \
    /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