staging: comedi: drivers: Fix memory leak in gsc_hpdi_auto_attach
Subsystems (2)
PATCH 1/1 staging: comedi: drivers: Fix memory leak in gsc_hpdi_auto_attach
In the implementation of gsc_hpdi_auto_attach(), the allocated dma description is leaks in case of alignment error, or failure of gsc_hpdi_setup_dma_descriptors() or comedi_alloc_subdevices(). Release devpriv->dma_desc via dma_free_coherent().
--- drivers/staging/comedi/drivers/gsc_hpdi.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/staging/comedi/drivers/gsc_hpdi.c +13 -3
@@ -633,16 +633,17 @@ static int gsc_hpdi_auto_attach(struct comedi_device *dev,if (devpriv->dma_desc_phys_addr &0xf ) {dev_warn (dev->class_dev," dma descriptors not quad-word aligned (bug)\n" ); -return -EIO ; + retval = -EIO ; + goto release_dma_desc; } retval =gsc_hpdi_setup_dma_descriptors (dev,0x1000 );if (retval <0 ) -return retval; + goto release_dma_desc; retval =comedi_alloc_subdevices (dev,1 );if (retval) -return retval; + goto release_dma_desc;/* Digital I/O subdevice */ s = &dev->subdevices[0 ];
@@ -660,6 +661,15 @@ static int gsc_hpdi_auto_attach(struct comedi_device *dev, s->cancel = gsc_hpdi_cancel;return gsc_hpdi_init (dev); + +release_dma_desc: +if (devpriv->dma_desc) +dma_free_coherent (&pcidev->dev, +sizeof (struct plx_dma_desc ) * +NUM_DMA_DESCRIPTORS , + devpriv->dma_desc, + devpriv->dma_desc_phys_addr); +return retval; }static void gsc_hpdi_detach (struct comedi_device *dev)
Signature · 1 line
2.17.1
Quoted 2 lines
In the implementation of
gsc_hpdi_auto_attach(), the allocated dma description is leaks in case of alignment error, …
Please avoid a typo in this change message.
…
+++ b/drivers/staging/comedi/drivers/gsc_hpdi.c
…
Quoted 6 lines
drivers/staging/comedi/drivers/gsc_hpdi.c +2
-0@@ -660,6 +661,15 @@ static int gsc_hpdi_auto_attach(struct comedi_device *dev, s->cancel = gsc_hpdi_cancel;return gsc_hpdi_init (dev); + +release_dma_desc:
drivers/staging/comedi/drivers/gsc_hpdi.c static int gsc_hpdi_auto_attach(struct comedi_device *dev, · line 675
Quoted 10 lines
drivers/staging/comedi/drivers/gsc_hpdi.c +7
-0+if (devpriv->dma_desc) +dma_free_coherent (&pcidev->dev, +sizeof (struct plx_dma_desc ) * +NUM_DMA_DESCRIPTORS , + devpriv->dma_desc, + devpriv->dma_desc_phys_addr); +return retval; }static void gsc_hpdi_detach (struct comedi_device *dev)
I got the impression that return values from calls of the function “dma_alloc_coherent” should be checked before.
* Would you like to add null pointer checks at other source code places?
* Should the jump targets be accordingly adjusted then for the completion of the desired exception handling?
Regards,
Markus
Quoted 6 lines · Navid Emamdoost
In the implementation of
gsc_hpdi_auto_attach(), the allocated dma description is leaks in case of alignment error, or failure ofgsc_hpdi_setup_dma_descriptors()orcomedi_alloc_subdevices(). Releasedevpriv->dma_descviadma_free_coherent().Signed-off-by Navid Emamdoost <[email protected]>
Actually, there is no memory leak (although there is another problem that I'll mention below). If the "auto_attach" handler gsc_hpdi_auto_attach() returns an error, then the "detach" handler gsc_hpdi_detach() will be called automatically to clean up. (This is true for all comedi drivers). gsc_hpdi_detach() calls gsc_hpdi_free_dma() to free the DMA buffers and DMA descriptors.
Quoted 36 lines
--- drivers/staging/comedi/drivers/gsc_hpdi.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)drivers/staging/comedi/drivers/gsc_hpdi.c +6
-3@@ -633,16 +633,17 @@ static int gsc_hpdi_auto_attach(struct comedi_device *dev,if (devpriv->dma_desc_phys_addr &0xf ) {dev_warn (dev->class_dev," dma descriptors not quad-word aligned (bug)\n" ); -return -EIO ; + retval = -EIO ; + goto release_dma_desc; } retval =gsc_hpdi_setup_dma_descriptors (dev,0x1000 );if (retval <0 ) -return retval; + goto release_dma_desc; retval =comedi_alloc_subdevices (dev,1 );if (retval) -return retval; + goto release_dma_desc;/* Digital I/O subdevice */ s = &dev->subdevices[0 ];@@ -660,6 +661,15 @@ static int gsc_hpdi_auto_attach(struct comedi_device *dev, s->cancel = gsc_hpdi_cancel;return gsc_hpdi_init (dev); + +release_dma_desc:
drivers/staging/comedi/drivers/gsc_hpdi.c static int gsc_hpdi_auto_attach(struct comedi_device *dev, · line 675
Quoted 10 lines
drivers/staging/comedi/drivers/gsc_hpdi.c +7
-0+if (devpriv->dma_desc) +dma_free_coherent (&pcidev->dev, +sizeof (struct plx_dma_desc ) * +NUM_DMA_DESCRIPTORS , + devpriv->dma_desc, + devpriv->dma_desc_phys_addr); +return retval; }static void gsc_hpdi_detach (struct comedi_device *dev)
This patch could actually result in devpriv->dma_desc being freed twice - once in the 'release_dma_desc:' code and again when gsc_hpdi_detach() is called externally as part of the clean-up.
The real bug in the original code is that it does not check whether any of the calls to dma_alloc_coherent() returned NULL. If any of the calls to dma_alloc_coherent() returns NULL, gsc_hpdi_auto_attach() needs to return an error (-ENOMEM). The subsequent call to gsc_hpdi_detach() will then free whatever DMA coherent buffers where allocated.
Signature · 4 lines
-=( Ian Abbott <[email protected]> || Web: www.mev.co.uk )=- -=( MEV Ltd. is a company registered in England & Wales. )=- -=( Registered number: 02862268. Registered address: )=- -=( 15 West Park Road, Bramhall, STOCKPORT, SK7 3JZ, UK. )=-
Ian, thanks for your feedback.
Quoted 7 lines · Ian Abbott
Quoted 5 lines · Navid Emamdoost
In the implementation of
gsc_hpdi_auto_attach(), the allocated dma description is leaks in case of alignment error, or failure ofgsc_hpdi_setup_dma_descriptors()orcomedi_alloc_subdevices(). Releasedevpriv->dma_descviadma_free_coherent().
Quoted 9 lines
Quoted 1 line
Signed-off-by Navid Emamdoost <[email protected]>Actually, there is no memory leak (although there is another problem that I'll mention below). If the "
auto_attach" handlergsc_hpdi_auto_attach()returns an error, then the "detach" handlergsc_hpdi_detach()will be called automatically to clean up. (This is true for all comedi drivers).gsc_hpdi_detach()callsgsc_hpdi_free_dma()to free the DMA buffers and DMA descriptors.
I was aware that comedi_alloc_devpriv() is a resource managed allocation, but was not sure how subsequent dma_desc allocation will be handled when device detach.
Quoted 51 lines
Quoted 36 lines
--- drivers/staging/comedi/drivers/gsc_hpdi.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)drivers/staging/comedi/drivers/gsc_hpdi.c +6
-3@@ -633,16 +633,17 @@ static int gsc_hpdi_auto_attach(struct comedi_device *dev,if (devpriv->dma_desc_phys_addr &0xf ) {dev_warn (dev->class_dev," dma descriptors not quad-word aligned (bug)\n" ); -return -EIO ; + retval = -EIO ; + goto release_dma_desc; } retval =gsc_hpdi_setup_dma_descriptors (dev,0x1000 );if (retval <0 ) -return retval; + goto release_dma_desc; retval =comedi_alloc_subdevices (dev,1 );if (retval) -return retval; + goto release_dma_desc;/* Digital I/O subdevice */ s = &dev->subdevices[0 ];@@ -660,6 +661,15 @@ static int gsc_hpdi_auto_attach(struct comedi_device *dev, s->cancel = gsc_hpdi_cancel;return gsc_hpdi_init (dev); + +release_dma_desc:drivers/staging/comedi/drivers/gsc_hpdi.c static int gsc_hpdi_auto_attach(struct comedi_device *dev, · line 675
Quoted 10 lines
drivers/staging/comedi/drivers/gsc_hpdi.c +7
-0+if (devpriv->dma_desc) +dma_free_coherent (&pcidev->dev, +sizeof (struct plx_dma_desc ) * +NUM_DMA_DESCRIPTORS , + devpriv->dma_desc, + devpriv->dma_desc_phys_addr); +return retval; }static void gsc_hpdi_detach (struct comedi_device *dev)This patch could actually result in
devpriv->dma_descbeing freed twice
- once in the 'release_dma_desc:' code and again whengsc_hpdi_detach()is called externally as part of the clean-up.
Quoted 6 lines
The real bug in the original code is that it does not check whether any of the calls to
dma_alloc_coherent()returned NULL. If any of the calls todma_alloc_coherent()returns NULL,gsc_hpdi_auto_attach()needs to return an error (-ENOMEM). The subsequent call togsc_hpdi_detach()will then free whatever DMA coherent buffers where allocated.
Yes, this potential null deref is another type of bug, which I will send a patch for separately.
Quoted 5 lines
--
-=( Ian Abbott <[email protected]> || Web: www.mev.co.uk )=- -=( MEV Ltd. is a company registered in England & Wales. )=-
-=( Registered number: 02862268. Registered address: )=-
-=( 15 West Park Road, Bramhall, STOCKPORT, SK7 3JZ, UK. )=-
Signature · 1 line
Navid.