From: Clemens Ladisch <clemens@ladisch.de>
To: sudarshan.bisht@nokia.com
Cc: alsa-devel@alsa-project.org
Subject: Re: [PATCH 1/1] alsa-lib: fixed coverity reported issues under "RESOURCE_LEAK" checker.
Date: Thu, 17 Mar 2011 12:53:55 +0100 [thread overview]
Message-ID: <4D81F653.5070906@ladisch.de> (raw)
In-Reply-To: <1300354477-1088-2-git-send-email-sudarshan.bisht@nokia.com>
sudarshan.bisht@nokia.com wrote:
> --- a/modules/mixer/simple/sbase.c
> +++ b/modules/mixer/simple/sbase.c
> @@ -377,6 +377,7 @@ static int simple_event_add1(snd_mixer_class_t *class,
> if (ctype != SND_CTL_ELEM_TYPE_BOOLEAN) {
> __invalid_type:
> snd_mixer_selem_id_free(id);
> + free(hsimple);
> return -EINVAL;
> }
> break;
OK
> --- a/modules/mixer/simple/sbasedl.c
> +++ b/modules/mixer/simple/sbasedl.c
> @@ -99,7 +99,7 @@ int mixer_simple_basic_dlopen(snd_mixer_class_t *class,
> __error:
> if (initflag)
> free(priv);
> - if (h == NULL)
> + if (h)
> snd_dlclose(h);
> free(xlib);
> return -ENXIO;
OK
> --- a/src/conf.c
> +++ b/src/conf.c
> @@ -638,6 +638,7 @@ static int get_char_skip_comments(input_t *input)
> fd = malloc(sizeof(*fd));
> if (!fd) {
> free(str);
> + snd_input_close(in);
> return -ENOMEM;
> }
OK
> @@ -4589,6 +4590,7 @@ static int parse_args(snd_config_t *subs, const char *str, snd_config_t *defs)
> if (err < 0) {
> _err:
> free(val);
> + snd_config_delete(sub);
> return err;
> }
When this code is reached through the label _err, sub is not
necessarily a valid pointer.
> --- a/src/control/control_hw.c
> +++ b/src/control/control_hw.c
> @@ -230,8 +230,10 @@ static int snd_ctl_hw_elem_tlv(snd_ctl_t *handle, int op_flag,
> return -errno;
> }
> if (op_flag == 0) {
> - if (xtlv->tlv[1] + 2 * sizeof(unsigned int) > tlv_size)
> + if (xtlv->tlv[1] + 2 * sizeof(unsigned int) > tlv_size) {
> + free(xtlv);
> return -EFAULT;
> + }
OK, but the indent of the closing brace is wrong.
> --- a/src/pcm/pcm_file.c
> +++ b/src/pcm/pcm_file.c
> @@ -218,7 +218,7 @@ static int snd_pcm_file_open_output_file(snd_pcm_file_t *file)
>
> if (file->final_fname[0] == '|') {
> /* pipe mode */
> - FILE *pipe;
> + FILE *pipe = NULL;
> /* clearing */
> pipe = popen(file->final_fname + 1, "w");
> if (!pipe) {
This isn't necessary because the variable is set in the next line.
> @@ -226,7 +226,8 @@ static int snd_pcm_file_open_output_file(snd_pcm_file_t *file)
> file->final_fname);
> return -errno;
> }
> - fd = fileno(pipe);
> + fd = dup(fileno(pipe));
> + fclose(pipe);
> } else {
> if (file->trunc)
> fd = open(file->final_fname, O_WRONLY|O_CREAT|O_TRUNC,
OK, but dup() might fail.
> --- a/src/pcm/pcm_ladspa.c
> +++ b/src/pcm/pcm_ladspa.c
> @@ -750,8 +750,10 @@ static int snd_pcm_ladspa_allocate_memory(snd_pcm_t *pcm, snd_pcm_ladspa_t *lads
> if (instance->input.data == NULL ||
> instance->input.m_data == NULL ||
> instance->output.data == NULL ||
> - instance->output.m_data == NULL)
> + instance->output.m_data == NULL) {
> + free(pchannels);
> return -ENOMEM;
> + }
OK
> @@ -761,8 +763,10 @@ static int snd_pcm_ladspa_allocate_memory(snd_pcm_t *pcm, snd_pcm_ladspa_t *lads
> instance->input.data[idx] = pchannels[chn];
> if (instance->input.data[idx] == NULL) {
> instance->input.data[idx] = snd_pcm_ladspa_allocate_zero(ladspa, 0);
> - if (instance->input.data[idx] == NULL)
> - return -ENOMEM;
> + if (instance->input.data[idx] == NULL) {
> + free(pchannels);
> + return -ENOMEM;
> + }
> }
> }
OK
> @@ -770,8 +774,10 @@ static int snd_pcm_ladspa_allocate_memory(snd_pcm_t *pcm, snd_pcm_ladspa_t *lads
> /* FIXME/OPTIMIZE: check if we can remove double alloc */
> /* if LADSPA plugin has no broken inplace */
> instance->output.data[idx] = malloc(sizeof(LADSPA_Data) * ladspa->allocated);
> - if (instance->output.data[idx] == NULL)
> - return -ENOMEM;
> + if (instance->output.data[idx] == NULL) {
> + free(pchannels);
> + return -ENOMEM;
> + }
> pchannels[chn] = instance->output.m_data[idx] = instance->output.data[idx];
> }
> }
OK
> @@ -793,8 +799,10 @@ static int snd_pcm_ladspa_allocate_memory(snd_pcm_t *pcm, snd_pcm_ladspa_t *lads
> instance->output.data[idx] = NULL;
> } else {
> instance->output.data[idx] = snd_pcm_ladspa_allocate_zero(ladspa, 1);
> - if (instance->output.data[idx] == NULL)
> + if (instance->output.data[idx] == NULL) {
> + free(pchannels);
> return -ENOMEM;
> + }
> }
> }
> }
OK
> --- a/src/pcm/pcm_plug.c
> +++ b/src/pcm/pcm_plug.c
> @@ -1298,6 +1298,7 @@ int _snd_pcm_plug_open(snd_pcm_t **pcmp, const char *name,
> err = snd_pcm_route_load_ttable(tt, ttable, csize, ssize, &cused, &sused, -1);
> if (err < 0) {
> snd_config_delete(sconf);
> + free(ttable);
> return err;
> }
OK
> @@ -1310,8 +1311,10 @@ int _snd_pcm_plug_open(snd_pcm_t **pcmp, const char *name,
>
> err = snd_pcm_open_slave(&spcm, root, sconf, stream, mode, conf);
> snd_config_delete(sconf);
> - if (err < 0)
> + if (err < 0) {
> + free(ttable);
> return err;
> + }
OK, but the following snd_pcm_plug_open() does not guarantee that ttable
is freed if an error happens.
> diff --git a/src/pcm/pcm_rate.c b/src/pcm/pcm_rate.c
> index 70e30e5..065c9da 100644
> --- a/src/pcm/pcm_rate.c
> +++ b/src/pcm/pcm_rate.c
> @@ -1392,6 +1392,7 @@ int snd_pcm_rate_open(snd_pcm_t **pcmp, const char *name,
> } else {
> SNDERR("Invalid type for rate converter");
> snd_pcm_close(pcm);
> + free(rate);
> return -EINVAL;
> }
OK, but the following three error cases don't free rate either.
> --- a/src/rawmidi/rawmidi.c
> +++ b/src/rawmidi/rawmidi.c
> @@ -255,8 +255,10 @@ static int snd_rawmidi_open_conf(snd_rawmidi_t **inputp, snd_rawmidi_t **outputp
> snd_config_delete(type_conf);
> if (err >= 0)
> err = open_func(inputp, outputp, name, rawmidi_root, rawmidi_conf, mode);
> - if (err < 0)
> + if (err < 0) {
> + snd_dlclose(h);
> return err;
> + }
> if (inputp) {
> (*inputp)->dl_handle = h; h = NULL;
> snd_rawmidi_params_default(*inputp, ¶ms);
OK
Regards,
Clemens
next prev parent reply other threads:[~2011-03-17 11:52 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-03-17 9:34 [PATCH 0/1] Fixed issues/defects reported by Coverity tool sudarshan.bisht
2011-03-17 9:34 ` [PATCH 1/1] alsa-lib: fixed coverity reported issues under "RESOURCE_LEAK" checker sudarshan.bisht
2011-03-17 11:53 ` Clemens Ladisch [this message]
2011-03-18 8:23 ` Sudarshan Bisht
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=4D81F653.5070906@ladisch.de \
--to=clemens@ladisch.de \
--cc=alsa-devel@alsa-project.org \
--cc=sudarshan.bisht@nokia.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;
as well as URLs for NNTP newsgroup(s).