linux-kernel - drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c:258:2-8: preceding lock on line 251 (fwd)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Message-ID: <alpine.DEB.2.22.394.2312272155490.3107@hadrien>
Date: Wed, 27 Dec 2023 21:56:46 +0100 (CET)
From: Julia Lawall <julia.lawall@...ia.fr>
To: Ben Skeggs <bskeggs@...hat.com>, Lyude Paul <lyude@...hat.com>
cc: linux-kernel@...r.kernel.org, oe-kbuild-all@...ts.linux.dev
Subject: drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c:258:2-8: preceding
 lock on line 251 (fwd)

Hello,

There is no bug here, but it could be nice to refer to the lock in the
same way at all call sites (see line 258.

julia

---------- Forwarded message ----------
Date: Wed, 27 Dec 2023 22:34:45 +0800
From: kernel test robot <lkp@...el.com>
To: oe-kbuild@...ts.linux.dev
Cc: lkp@...el.com, Julia Lawall <julia.lawall@...ia.fr>
Subject: drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c:258:2-8: preceding lock
     on line 251

BCC: lkp@...el.com
CC: oe-kbuild-all@...ts.linux.dev
CC: linux-kernel@...r.kernel.org
TO: Ben Skeggs <bskeggs@...hat.com>
CC: Lyude Paul <lyude@...hat.com>

tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
head:   fbafc3e621c3f4ded43720fdb1d6ce1728ec664e
commit: f48dd2936138882d7755cbbc5d9984015c75980c drm/nouveau/fifo: add new engine context tracking
date:   1 year, 2 months ago
:::::: branch date: 2 days ago
:::::: commit date: 1 year, 2 months ago
config: loongarch-randconfig-r062-20231222 (https://download.01.org/0day-ci/archive/20231227/202312272217.k0uoS8R5-lkp@intel.com/config)
compiler: loongarch64-linux-gcc (GCC) 13.2.0

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@...el.com>
| Reported-by: Julia Lawall <julia.lawall@...ia.fr>
| Closes: https://lore.kernel.org/r/202312272217.k0uoS8R5-lkp@intel.com/

cocci warnings: (new ones prefixed by >>)
>> drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c:258:2-8: preceding lock on line 251

vim +258 drivers/gpu/drm/nouveau/nvkm/engine/fifo/chan.c

f48dd293613888 Ben Skeggs 2022-06-01  240
f48dd293613888 Ben Skeggs 2022-06-01  241  int
f48dd293613888 Ben Skeggs 2022-06-01  242  nvkm_chan_cctx_get(struct nvkm_chan *chan, struct nvkm_engn *engn, struct nvkm_cctx **pcctx,
f48dd293613888 Ben Skeggs 2022-06-01  243  		   struct nvkm_client *client)
f48dd293613888 Ben Skeggs 2022-06-01  244  {
f48dd293613888 Ben Skeggs 2022-06-01  245  	struct nvkm_cgrp *cgrp = chan->cgrp;
f48dd293613888 Ben Skeggs 2022-06-01  246  	struct nvkm_vctx *vctx;
f48dd293613888 Ben Skeggs 2022-06-01  247  	struct nvkm_cctx *cctx;
f48dd293613888 Ben Skeggs 2022-06-01  248  	int ret;
f48dd293613888 Ben Skeggs 2022-06-01  249
f48dd293613888 Ben Skeggs 2022-06-01  250  	/* Look for an existing channel context for this engine+VEID. */
f48dd293613888 Ben Skeggs 2022-06-01 @251  	mutex_lock(&cgrp->mutex);
f48dd293613888 Ben Skeggs 2022-06-01  252  	cctx = nvkm_list_find(cctx, &chan->cctxs, head,
f48dd293613888 Ben Skeggs 2022-06-01  253  			      cctx->vctx->ectx->engn == engn && cctx->vctx->vmm == chan->vmm);
f48dd293613888 Ben Skeggs 2022-06-01  254  	if (cctx) {
f48dd293613888 Ben Skeggs 2022-06-01  255  		refcount_inc(&cctx->refs);
f48dd293613888 Ben Skeggs 2022-06-01  256  		*pcctx = cctx;
f48dd293613888 Ben Skeggs 2022-06-01  257  		mutex_unlock(&chan->cgrp->mutex);
f48dd293613888 Ben Skeggs 2022-06-01 @258  		return 0;
f48dd293613888 Ben Skeggs 2022-06-01  259  	}
f48dd293613888 Ben Skeggs 2022-06-01  260
f48dd293613888 Ben Skeggs 2022-06-01  261  	/* Nope - create a fresh one.  But, sub-context first. */
f48dd293613888 Ben Skeggs 2022-06-01  262  	ret = nvkm_cgrp_vctx_get(cgrp, engn, chan, &vctx, client);
f48dd293613888 Ben Skeggs 2022-06-01  263  	if (ret) {
f48dd293613888 Ben Skeggs 2022-06-01  264  		CHAN_ERROR(chan, "vctx %d[%s]: %d", engn->id, engn->engine->subdev.name, ret);
f48dd293613888 Ben Skeggs 2022-06-01  265  		goto done;
f48dd293613888 Ben Skeggs 2022-06-01  266  	}
f48dd293613888 Ben Skeggs 2022-06-01  267
f48dd293613888 Ben Skeggs 2022-06-01  268  	/* Now, create the channel context - to track engine binding. */
f48dd293613888 Ben Skeggs 2022-06-01  269  	CHAN_TRACE(chan, "ctor cctx %d[%s]", engn->id, engn->engine->subdev.name);
f48dd293613888 Ben Skeggs 2022-06-01  270  	if (!(cctx = *pcctx = kzalloc(sizeof(*cctx), GFP_KERNEL))) {
f48dd293613888 Ben Skeggs 2022-06-01  271  		nvkm_cgrp_vctx_put(cgrp, &vctx);
f48dd293613888 Ben Skeggs 2022-06-01  272  		ret = -ENOMEM;
f48dd293613888 Ben Skeggs 2022-06-01  273  		goto done;
f48dd293613888 Ben Skeggs 2022-06-01  274  	}
f48dd293613888 Ben Skeggs 2022-06-01  275
f48dd293613888 Ben Skeggs 2022-06-01  276  	cctx->vctx = vctx;
f48dd293613888 Ben Skeggs 2022-06-01  277  	refcount_set(&cctx->refs, 1);
f48dd293613888 Ben Skeggs 2022-06-01  278  	refcount_set(&cctx->uses, 0);
f48dd293613888 Ben Skeggs 2022-06-01  279  	list_add_tail(&cctx->head, &chan->cctxs);
f48dd293613888 Ben Skeggs 2022-06-01  280  done:
f48dd293613888 Ben Skeggs 2022-06-01  281  	mutex_unlock(&cgrp->mutex);
f48dd293613888 Ben Skeggs 2022-06-01  282  	return ret;
f48dd293613888 Ben Skeggs 2022-06-01  283  }
f48dd293613888 Ben Skeggs 2022-06-01  284

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki