Connecting master to an output port when GTH driver module is not loaded triggers a NULL dereference:
RIP: 0010:intel_th_set_output+0x35/0x70 [intel_th] Call Trace: ? sth_stm_link+0x12/0x20 [intel_th_sth] stm_source_link_store+0x164/0x270 [stm_core] dev_attr_store+0x17/0x30 sysfs_kf_write+0x3e/0x50 kernfs_fop_write+0xda/0x1b0 __vfs_write+0x1b/0x40 vfs_write+0xb9/0x1a0 ksys_write+0x67/0xe0 __x64_sys_write+0x1a/0x20 do_syscall_64+0x57/0x1d0 entry_SYSCALL_64_after_hwframe+0x44/0xa9
Make sure the module in question is loaded and return an error if not.
Signed-off-by: Alexander Shishkin alexander.shishkin@linux.intel.com Fixes: 39f4034693b7c ("intel_th: Add driver infrastructure for Intel(R) Trace Hub devices") Reviewed-by: Andy Shevchenko andriy.shevchenko@linux.intel.com Reported-by: Ammy Yi ammy.yi@intel.com Tested-by: Ammy Yi ammy.yi@intel.com Cc: stable@vger.kernel.org # v4.4 --- drivers/hwtracing/intel_th/core.c | 21 ++++++++++++++++++--- drivers/hwtracing/intel_th/sth.c | 4 +--- 2 files changed, 19 insertions(+), 6 deletions(-)
diff --git a/drivers/hwtracing/intel_th/core.c b/drivers/hwtracing/intel_th/core.c index ca232ec565e8..c9ac3dc65113 100644 --- a/drivers/hwtracing/intel_th/core.c +++ b/drivers/hwtracing/intel_th/core.c @@ -1021,15 +1021,30 @@ int intel_th_set_output(struct intel_th_device *thdev, { struct intel_th_device *hub = to_intel_th_hub(thdev); struct intel_th_driver *hubdrv = to_intel_th_driver(hub->dev.driver); + int ret;
/* In host mode, this is up to the external debugger, do nothing. */ if (hub->host_mode) return 0;
- if (!hubdrv->set_output) - return -ENOTSUPP; + /* + * hub is instantiated together with the source device that + * calls here, so guaranteed to be present. + */ + hubdrv = to_intel_th_driver(hub->dev.driver); + if (!hubdrv || !try_module_get(hubdrv->driver.owner)) + return -EINVAL; + + if (!hubdrv->set_output) { + ret = -ENOTSUPP; + goto out; + } + + ret = hubdrv->set_output(hub, master);
- return hubdrv->set_output(hub, master); +out: + module_put(hubdrv->driver.owner); + return ret; } EXPORT_SYMBOL_GPL(intel_th_set_output);
diff --git a/drivers/hwtracing/intel_th/sth.c b/drivers/hwtracing/intel_th/sth.c index 3a1f4e650378..a1529f571491 100644 --- a/drivers/hwtracing/intel_th/sth.c +++ b/drivers/hwtracing/intel_th/sth.c @@ -161,9 +161,7 @@ static int sth_stm_link(struct stm_data *stm_data, unsigned int master, { struct sth_device *sth = container_of(stm_data, struct sth_device, stm);
- intel_th_set_output(to_intel_th_device(sth->dev), master); - - return 0; + return intel_th_set_output(to_intel_th_device(sth->dev), master); }
static int intel_th_sw_init(struct sth_device *sth)
Hi
[This is an automated email]
This commit has been processed because it contains a "Fixes:" tag fixing commit: 39f4034693b7 ("intel_th: Add driver infrastructure for Intel(R) Trace Hub devices").
The bot has tested the following trees: v5.7.7, v5.4.50, v4.19.131, v4.14.187, v4.9.229, v4.4.229.
v5.7.7: Build OK! v5.4.50: Build OK! v4.19.131: Build OK! v4.14.187: Failed to apply! Possible dependencies: c2d2c7de972d7 ("intel_th: Don't touch switch routing in host mode")
v4.9.229: Failed to apply! Possible dependencies: c2d2c7de972d7 ("intel_th: Don't touch switch routing in host mode")
v4.4.229: Failed to apply! Possible dependencies: c2d2c7de972d7 ("intel_th: Don't touch switch routing in host mode")
NOTE: The patch will not be queued to stable trees until it is upstream.
How should we proceed with this patch?
linux-stable-mirror@lists.linaro.org