Re: [PATCH 1/3] usb: chipidea: Fix DMA overwrite for Tegra

From: "Michał Mirosław" <mirq-linux@rere.qmqm.pl>
To: Peter Chen <peter.chen@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Thierry Reding <treding@nvidia.com>,
	Dmitry Osipenko <digetx@gmail.com>,
	Peter Geis <pgwipeout@gmail.com>,
	Jonathan Hunter <jonathanh@nvidia.com>,
	linux-usb@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org,
	linux-tegra@vger.kernel.org
Subject: Re: [PATCH 1/3] usb: chipidea: Fix DMA overwrite for Tegra
Date: Tue, 26 Sep 2023 01:53:39 +0200	[thread overview]
Message-ID: <ZRIdg29ADFmKnlAQ@qmqm.qmqm.pl> (raw)
In-Reply-To: <20230925114522.GA2070044@nchen-desktop>

On Mon, Sep 25, 2023 at 07:45:22PM +0800, Peter Chen wrote:
> On 23-09-23 21:41:55, Michał Mirosław wrote:
> > Tegra USB controllers seem to issue DMA in doubleword-sized chunks and thus
> > may write past the buffer provided. This is detected by SLUB:
> > 
> > =============================================================================
> > BUG kmalloc-64 (Tainted: G    B             ): kmalloc Redzone overwritten
> > -----------------------------------------------------------------------------
> > 
> > 0x8555cd02-0x8555cd03 @offset=3330. First byte 0x0 instead of 0xcc
> > Allocated in usb_get_status+0x2b/0xac age=1 cpu=3 pid=41
> >  __kmem_cache_alloc_node+0x12f/0x1e4
> >  __kmalloc+0x33/0x8c
> >  usb_get_status+0x2b/0xac
> >  hub_probe+0x5e9/0xcec
> >  usb_probe_interface+0xbf/0x21c
> >  really_probe+0xa5/0x2c4
> >  __driver_probe_device+0x75/0x174
> >  driver_probe_device+0x31/0x94
> >  __device_attach_driver+0x65/0xc0
> >  bus_for_each_drv+0x4b/0x74
> >  __device_attach+0x69/0x120
> >  bus_probe_device+0x65/0x6c
> >  device_add+0x48b/0x5f8
> >  usb_set_configuration+0x37b/0x6b4
> >  usb_generic_driver_probe+0x37/0x68
> >  usb_probe_device+0x35/0xb4
> > Slab 0xbf622b80 objects=21 used=18 fp=0x8555cdc0 flags=0x800(slab|zone=0)
> > Object 0x8555cd00 @offset=3328 fp=0x00000000
> > 
> > Redzone  8555ccc0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
> > Redzone  8555ccd0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
> > Redzone  8555cce0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
> > Redzone  8555ccf0: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
> > Object   8555cd00: 01 00 00 00 cc cc cc cc cc cc cc cc cc cc cc cc  ................
> > Object   8555cd10: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
> > Object   8555cd20: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
> > Object   8555cd30: cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc cc  ................
> > Redzone  8555cd40: cc cc cc cc                                      ....
> > Padding  8555cd74: 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a 5a              ZZZZZZZZZZZZ
> > CPU: 3 PID: 41 Comm: kworker/3:1 Tainted: G    B              6.6.0-rc1mq-00118-g59786f827ea1 #1115
> > Hardware name: NVIDIA Tegra SoC (Flattened Device Tree)
> > Workqueue: usb_hub_wq hub_event
> > [<8010ca28>] (unwind_backtrace) from [<801090a5>] (show_stack+0x11/0x14)
> > [<801090a5>] (show_stack) from [<805da2fb>] (dump_stack_lvl+0x4d/0x7c)
> > [<805da2fb>] (dump_stack_lvl) from [<8026464f>] (check_bytes_and_report+0xb3/0xe4)
> > [<8026464f>] (check_bytes_and_report) from [<802648e1>] (check_object+0x261/0x290)
> > [<802648e1>] (check_object) from [<802671b1>] (free_to_partial_list+0x105/0x3f8)
> > [<802671b1>] (free_to_partial_list) from [<80268613>] (__kmem_cache_free+0x103/0x128)
> > [<80268613>] (__kmem_cache_free) from [<80425a67>] (usb_get_status+0x73/0xac)
> > [<80425a67>] (usb_get_status) from [<80421b31>] (hub_probe+0x5e9/0xcec)
> > [<80421b31>] (hub_probe) from [<80428bbb>] (usb_probe_interface+0xbf/0x21c)
> > [<80428bbb>] (usb_probe_interface) from [<803ee13d>] (really_probe+0xa5/0x2c4)
> > [<803ee13d>] (really_probe) from [<803ee3d1>] (__driver_probe_device+0x75/0x174)
> > [<803ee3d1>] (__driver_probe_device) from [<803ee501>] (driver_probe_device+0x31/0x94)
> > usb 1-1: device descriptor read/8, error -71
> > 
> > Fixes: fc53d5279094 ("usb: chipidea: tegra: Support host mode")
> > Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
> > ---
> >  drivers/usb/chipidea/host.c | 7 ++++---
> >  1 file changed, 4 insertions(+), 3 deletions(-)
> > 
> > diff --git a/drivers/usb/chipidea/host.c b/drivers/usb/chipidea/host.c
> > index 08af26b762a2..abddd39d1ff1 100644
> > --- a/drivers/usb/chipidea/host.c
> > +++ b/drivers/usb/chipidea/host.c
> > @@ -411,12 +411,13 @@ static int ci_hdrc_alloc_dma_aligned_buffer(struct urb *urb, gfp_t mem_flags)
> >  	const unsigned int ci_hdrc_usb_dma_align = 32;
> >  	size_t kmalloc_size;
> >  
> > -	if (urb->num_sgs || urb->sg || urb->transfer_buffer_length == 0 ||
> > -	    !((uintptr_t)urb->transfer_buffer & (ci_hdrc_usb_dma_align - 1)))
> > +	if (urb->num_sgs || urb->sg || urb->transfer_buffer_length == 0)
> > +		return 0;
> > +	if (!((uintptr_t)urb->transfer_buffer & (ci_hdrc_usb_dma_align - 1)) && !(urb->transfer_buffer_length & 3))
> >  		return 0;
> >  
> >  	/* Allocate a buffer with enough padding for alignment */
> > -	kmalloc_size = urb->transfer_buffer_length +
> > +	kmalloc_size = ALIGN(urb->transfer_buffer_length, 4) +
> >  		       sizeof(struct ci_hdrc_dma_aligned_buffer) +
> >  		       ci_hdrc_usb_dma_align - 1;
> >  
> 
> Would you please explain why you make these changes?

Can you point out what's unclear in the commit message? This is to fix
a buffer overflow by DMA from the USB controller as it seems to write
data 32-bit word at a time. What the patch does is extend the workaround
code to account for the extra room needed at the tail of the buffer.

Best Regards
Michał Mirosław