mbox series

[SRU,F,v2,0/3] CVE-2024-26787

Message ID 20240815081951.10261-1-massimiliano.pellizzer@canonical.com
Headers show
Series CVE-2024-26787 | expand

Message

Massimiliano Pellizzer Aug. 15, 2024, 8:19 a.m. UTC 
[Impact]

mmc: mmci: stm32: fix DMA API overlapping mappings warning

Turning on CONFIG_DMA_API_DEBUG_SG results in the following warning:

DMA-API: mmci-pl18x 48220000.mmc: cacheline tracking EEXIST,
overlapping mappings aren't supported
WARNING: CPU: 1 PID: 51 at kernel/dma/debug.c:568
add_dma_entry+0x234/0x2f4
Modules linked in:
CPU: 1 PID: 51 Comm: kworker/1:2 Not tainted 6.1.28 #1
Hardware name: STMicroelectronics STM32MP257F-EV1 Evaluation Board (DT)
Workqueue: events_freezable mmc_rescan
Call trace:
add_dma_entry+0x234/0x2f4
debug_dma_map_sg+0x198/0x350
__dma_map_sg_attrs+0xa0/0x110
dma_map_sg_attrs+0x10/0x2c
sdmmc_idma_prep_data+0x80/0xc0
mmci_prep_data+0x38/0x84
mmci_start_data+0x108/0x2dc
mmci_request+0xe4/0x190
__mmc_start_request+0x68/0x140
mmc_start_request+0x94/0xc0
mmc_wait_for_req+0x70/0x100
mmc_send_tuning+0x108/0x1ac
sdmmc_execute_tuning+0x14c/0x210
mmc_execute_tuning+0x48/0xec
mmc_sd_init_uhs_card.part.0+0x208/0x464
mmc_sd_init_card+0x318/0x89c
mmc_attach_sd+0xe4/0x180
mmc_rescan+0x244/0x320

DMA API debug brings to light leaking dma-mappings as dma_map_sg and
dma_unmap_sg are not correctly balanced.

If an error occurs in mmci_cmd_irq function, only mmci_dma_error
function is called and as this API is not managed on stm32 variant,
dma_unmap_sg is never called in this error path.

[Fix]

Noble:  Not affected
Jammy:  Fixed
Focal:  Clean cherry pick of two prereqs, and clean cherry pick of the
fix commit
Bionic: Not affected
Xenial: Not affected

[Test Case]

Compile and boot tested.

[Where problems could occur]

The fix affects the driver responsible for managing the MMC/SD host
controller found on STM32 microcontrollers. An issue with this fix
therefore may lead to data corruption during SD/MMC card operations and
possibly to system crashes.

Christophe Kerello (1):
  mmc: mmci: stm32: fix DMA API overlapping mappings warning

Ludovic Barre (1):
  mmc: mmci_sdmmc: Rename sdmmc_priv struct to sdmmc_idma

Yann Gautier (1):
  mmc: mmci: stm32: use a buffer for unaligned DMA requests

 drivers/mmc/host/mmci_stm32_sdmmc.c | 118 +++++++++++++++++++++++-----
 1 file changed, 98 insertions(+), 20 deletions(-)

Comments

Manuel Diewald Aug. 15, 2024, 3:29 p.m. UTC | #1 
On Thu, Aug 15, 2024 at 10:19:16AM +0200, Massimiliano Pellizzer wrote:
> [Impact]
> 
> mmc: mmci: stm32: fix DMA API overlapping mappings warning
> 
> Turning on CONFIG_DMA_API_DEBUG_SG results in the following warning:
> 
> DMA-API: mmci-pl18x 48220000.mmc: cacheline tracking EEXIST,
> overlapping mappings aren't supported
> WARNING: CPU: 1 PID: 51 at kernel/dma/debug.c:568
> add_dma_entry+0x234/0x2f4
> Modules linked in:
> CPU: 1 PID: 51 Comm: kworker/1:2 Not tainted 6.1.28 #1
> Hardware name: STMicroelectronics STM32MP257F-EV1 Evaluation Board (DT)
> Workqueue: events_freezable mmc_rescan
> Call trace:
> add_dma_entry+0x234/0x2f4
> debug_dma_map_sg+0x198/0x350
> __dma_map_sg_attrs+0xa0/0x110
> dma_map_sg_attrs+0x10/0x2c
> sdmmc_idma_prep_data+0x80/0xc0
> mmci_prep_data+0x38/0x84
> mmci_start_data+0x108/0x2dc
> mmci_request+0xe4/0x190
> __mmc_start_request+0x68/0x140
> mmc_start_request+0x94/0xc0
> mmc_wait_for_req+0x70/0x100
> mmc_send_tuning+0x108/0x1ac
> sdmmc_execute_tuning+0x14c/0x210
> mmc_execute_tuning+0x48/0xec
> mmc_sd_init_uhs_card.part.0+0x208/0x464
> mmc_sd_init_card+0x318/0x89c
> mmc_attach_sd+0xe4/0x180
> mmc_rescan+0x244/0x320
> 
> DMA API debug brings to light leaking dma-mappings as dma_map_sg and
> dma_unmap_sg are not correctly balanced.
> 
> If an error occurs in mmci_cmd_irq function, only mmci_dma_error
> function is called and as this API is not managed on stm32 variant,
> dma_unmap_sg is never called in this error path.
> 
> [Fix]
> 
> Noble:  Not affected
> Jammy:  Fixed
> Focal:  Clean cherry pick of two prereqs, and clean cherry pick of the
> fix commit
> Bionic: Not affected
> Xenial: Not affected
> 
> [Test Case]
> 
> Compile and boot tested.
> 
> [Where problems could occur]
> 
> The fix affects the driver responsible for managing the MMC/SD host
> controller found on STM32 microcontrollers. An issue with this fix
> therefore may lead to data corruption during SD/MMC card operations and
> possibly to system crashes.
> 
> Christophe Kerello (1):
>   mmc: mmci: stm32: fix DMA API overlapping mappings warning
> 
> Ludovic Barre (1):
>   mmc: mmci_sdmmc: Rename sdmmc_priv struct to sdmmc_idma
> 
> Yann Gautier (1):
>   mmc: mmci: stm32: use a buffer for unaligned DMA requests
> 
>  drivers/mmc/host/mmci_stm32_sdmmc.c | 118 +++++++++++++++++++++++-----
>  1 file changed, 98 insertions(+), 20 deletions(-)
> 
> -- 
> 2.43.0
> 
> 
> -- 
> kernel-team mailing list
> kernel-team@lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team

Acked-by: Manuel Diewald <manuel.diewald@canonical.com>
Kuan-Ying Lee Aug. 16, 2024, 2:15 a.m. UTC | #2 
On Thu, Aug 15, 2024 at 10:19:16AM +0200, Massimiliano Pellizzer wrote:
> [Impact]
> 
> mmc: mmci: stm32: fix DMA API overlapping mappings warning
> 
> Turning on CONFIG_DMA_API_DEBUG_SG results in the following warning:
> 
> DMA-API: mmci-pl18x 48220000.mmc: cacheline tracking EEXIST,
> overlapping mappings aren't supported
> WARNING: CPU: 1 PID: 51 at kernel/dma/debug.c:568
> add_dma_entry+0x234/0x2f4
> Modules linked in:
> CPU: 1 PID: 51 Comm: kworker/1:2 Not tainted 6.1.28 #1
> Hardware name: STMicroelectronics STM32MP257F-EV1 Evaluation Board (DT)
> Workqueue: events_freezable mmc_rescan
> Call trace:
> add_dma_entry+0x234/0x2f4
> debug_dma_map_sg+0x198/0x350
> __dma_map_sg_attrs+0xa0/0x110
> dma_map_sg_attrs+0x10/0x2c
> sdmmc_idma_prep_data+0x80/0xc0
> mmci_prep_data+0x38/0x84
> mmci_start_data+0x108/0x2dc
> mmci_request+0xe4/0x190
> __mmc_start_request+0x68/0x140
> mmc_start_request+0x94/0xc0
> mmc_wait_for_req+0x70/0x100
> mmc_send_tuning+0x108/0x1ac
> sdmmc_execute_tuning+0x14c/0x210
> mmc_execute_tuning+0x48/0xec
> mmc_sd_init_uhs_card.part.0+0x208/0x464
> mmc_sd_init_card+0x318/0x89c
> mmc_attach_sd+0xe4/0x180
> mmc_rescan+0x244/0x320
> 
> DMA API debug brings to light leaking dma-mappings as dma_map_sg and
> dma_unmap_sg are not correctly balanced.
> 
> If an error occurs in mmci_cmd_irq function, only mmci_dma_error
> function is called and as this API is not managed on stm32 variant,
> dma_unmap_sg is never called in this error path.
> 
> [Fix]
> 
> Noble:  Not affected
> Jammy:  Fixed
> Focal:  Clean cherry pick of two prereqs, and clean cherry pick of the
> fix commit
> Bionic: Not affected
> Xenial: Not affected
> 
> [Test Case]
> 
> Compile and boot tested.
> 
> [Where problems could occur]
> 
> The fix affects the driver responsible for managing the MMC/SD host
> controller found on STM32 microcontrollers. An issue with this fix
> therefore may lead to data corruption during SD/MMC card operations and
> possibly to system crashes.
> 
> Christophe Kerello (1):
>   mmc: mmci: stm32: fix DMA API overlapping mappings warning
> 
> Ludovic Barre (1):
>   mmc: mmci_sdmmc: Rename sdmmc_priv struct to sdmmc_idma
> 
> Yann Gautier (1):
>   mmc: mmci: stm32: use a buffer for unaligned DMA requests
> 
>  drivers/mmc/host/mmci_stm32_sdmmc.c | 118 +++++++++++++++++++++++-----
>  1 file changed, 98 insertions(+), 20 deletions(-)

Acked-by: Kuan-Ying Lee <kuan-ying.lee@canonical.com>
Stefan Bader Aug. 16, 2024, 1:44 p.m. UTC | #3 
On 15.08.24 10:19, Massimiliano Pellizzer wrote:
> [Impact]
> 
> mmc: mmci: stm32: fix DMA API overlapping mappings warning
> 
> Turning on CONFIG_DMA_API_DEBUG_SG results in the following warning:
> 
> DMA-API: mmci-pl18x 48220000.mmc: cacheline tracking EEXIST,
> overlapping mappings aren't supported
> WARNING: CPU: 1 PID: 51 at kernel/dma/debug.c:568
> add_dma_entry+0x234/0x2f4
> Modules linked in:
> CPU: 1 PID: 51 Comm: kworker/1:2 Not tainted 6.1.28 #1
> Hardware name: STMicroelectronics STM32MP257F-EV1 Evaluation Board (DT)
> Workqueue: events_freezable mmc_rescan
> Call trace:
> add_dma_entry+0x234/0x2f4
> debug_dma_map_sg+0x198/0x350
> __dma_map_sg_attrs+0xa0/0x110
> dma_map_sg_attrs+0x10/0x2c
> sdmmc_idma_prep_data+0x80/0xc0
> mmci_prep_data+0x38/0x84
> mmci_start_data+0x108/0x2dc
> mmci_request+0xe4/0x190
> __mmc_start_request+0x68/0x140
> mmc_start_request+0x94/0xc0
> mmc_wait_for_req+0x70/0x100
> mmc_send_tuning+0x108/0x1ac
> sdmmc_execute_tuning+0x14c/0x210
> mmc_execute_tuning+0x48/0xec
> mmc_sd_init_uhs_card.part.0+0x208/0x464
> mmc_sd_init_card+0x318/0x89c
> mmc_attach_sd+0xe4/0x180
> mmc_rescan+0x244/0x320
> 
> DMA API debug brings to light leaking dma-mappings as dma_map_sg and
> dma_unmap_sg are not correctly balanced.
> 
> If an error occurs in mmci_cmd_irq function, only mmci_dma_error
> function is called and as this API is not managed on stm32 variant,
> dma_unmap_sg is never called in this error path.
> 
> [Fix]
> 
> Noble:  Not affected
> Jammy:  Fixed
> Focal:  Clean cherry pick of two prereqs, and clean cherry pick of the
> fix commit
> Bionic: Not affected
> Xenial: Not affected
> 
> [Test Case]
> 
> Compile and boot tested.
> 
> [Where problems could occur]
> 
> The fix affects the driver responsible for managing the MMC/SD host
> controller found on STM32 microcontrollers. An issue with this fix
> therefore may lead to data corruption during SD/MMC card operations and
> possibly to system crashes.
> 
> Christophe Kerello (1):
>    mmc: mmci: stm32: fix DMA API overlapping mappings warning
> 
> Ludovic Barre (1):
>    mmc: mmci_sdmmc: Rename sdmmc_priv struct to sdmmc_idma
> 
> Yann Gautier (1):
>    mmc: mmci: stm32: use a buffer for unaligned DMA requests
> 
>   drivers/mmc/host/mmci_stm32_sdmmc.c | 118 +++++++++++++++++++++++-----
>   1 file changed, 98 insertions(+), 20 deletions(-)
> 

Applied to focal:linux/master-next. Thanks.

-Stefan