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>] [<thread-prev] [day] [month] [year] [list]
Date:   Fri, 15 Nov 2019 14:51:55 -0500
From:   Olga Kornievskaia <aglo@...ch.edu>
To:     Colin Ian King <colin.king@...onical.com>
Cc:     Olga Kornievskaia <kolga@...app.com>,
        Trond Myklebust <trond.myklebust@...merspace.com>,
        Anna Schumaker <anna.schumaker@...app.com>,
        linux-nfs <linux-nfs@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: NFS: inter ssc open (memory leak detected)

On Fri, Nov 15, 2019 at 4:24 AM Colin Ian King <colin.king@...onical.com> wrote:
>
> Hi,
>
> Static analysis with Coverity has detected a memory leak in the
> following commit:
>
> commit ec4b0925089826af45e99cdf78a8ac84c1d005f1
> Author: Olga Kornievskaia <kolga@...app.com>
> Date:   Tue Oct 8 16:33:53 2019 -0400
>
>     NFS: inter ssc open
>
>
> In function nfs42_ssc_open(), fs/nfs/nfs4file.c, analysis is as follows:
>
>    3. alloc_fn: Storage is returned from allocation function kzalloc.
>    4. var_assign: Assigning: read_name = storage returned from
> kzalloc(len, 3136U).
> 336        read_name = kzalloc(len, GFP_NOFS);
>
>    5. Condition read_name == NULL, taking false branch.
> 337        if (read_name == NULL)
> 338                goto out;
>
>    6. noescape: Resource read_name is not freed or pointed-to in snprintf.
> 339        snprintf(read_name, len, SSC_READ_NAME_BODY, read_name_gen++);
> 340
> 341        r_ino = nfs_fhget(ss_mnt->mnt_root->d_inode->i_sb, src_fh,
> &fattr,
> 342                        NULL);
>
>    7. Condition IS_ERR(r_ino), taking true branch.
> 343        if (IS_ERR(r_ino)) {
> 344                res = ERR_CAST(r_ino);
>
>    8. Jumping to label out.
> 345                goto out;
> 346        }
> 347
> 348        filep = alloc_file_pseudo(r_ino, ss_mnt, read_name, FMODE_READ,
> 349                                     r_ino->i_fop);
> 350        if (IS_ERR(filep)) {
> 351                res = ERR_CAST(filep);
> 352                goto out;
> 353        }
> 354        filep->f_mode |= FMODE_READ;
> 355
> 356        ctx = alloc_nfs_open_context(filep->f_path.dentry, filep->f_mode,
> 357                                        filep);
> 358        if (IS_ERR(ctx)) {
> 359                res = ERR_CAST(ctx);
> 360                goto out_filep;
> 361        }
> 362
> 363        res = ERR_PTR(-EINVAL);
> 364        sp = nfs4_get_state_owner(server, ctx->cred, GFP_KERNEL);
> 365        if (sp == NULL)
> 366                goto out_ctx;
> 367
> 368        ctx->state = nfs4_get_open_state(r_ino, sp);
> 369        if (ctx->state == NULL)
> 370                goto out_stateowner;
> 371
> 372        set_bit(NFS_SRV_SSC_COPY_STATE, &ctx->state->flags);
> 373        set_bit(NFS_OPEN_STATE, &ctx->state->flags);
> 374        memcpy(&ctx->state->open_stateid.other, &stateid->other,
> 375               NFS4_STATEID_OTHER_SIZE);
> 376        update_open_stateid(ctx->state, stateid, NULL, filep->f_mode);
> 377
> 378        nfs_file_set_open_context(filep, ctx);
> 379        put_nfs_open_context(ctx);
> 380
> 381        file_ra_state_init(&filep->f_ra,
> filep->f_mapping->host->i_mapping);
> 382        res = filep;
> 383out:
>
>    CID 91575 (#1-2 of 2): Resource leak (RESOURCE_LEAK)
>
> 9. leaked_storage: Variable read_name going out of scope leaks the
> storage it points to.
>
> 384        return res;
>
>
> Looks like there are several return paths to out: that leak the
> allocation of read_name.

Thanks will fix.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ