In `crb_acpi_add()`, we call `acpi_get_table()` to retrieve the ACPI table entry. `acpi_put_table()` is called on the error path to avoid a memory leak, but the current implementation does not call `acpi_put_table()` when the `length` field of `struct acpi_table_header` is not valid, which leads to a memory leak. Although this memory leak only occurrs when the firmware misconfigured the ACPI table, it would still be nice to have this fix.
Signed-off-by: Joe Hattori <[email protected]> --- drivers/char/tpm/tpm_crb.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/char/tpm/tpm_crb.c b/drivers/char/tpm/tpm_crb.c index ea085b14ab7c..68fe28208331 100644 --- a/drivers/char/tpm/tpm_crb.c +++ b/drivers/char/tpm/tpm_crb.c @@ -738,10 +738,14 @@ static int crb_acpi_add(struct acpi_device *device) status = acpi_get_table(ACPI_SIG_TPM2, 1, (struct acpi_table_header **) &buf); - if (ACPI_FAILURE(status) || buf->header.length < sizeof(*buf)) { + if (ACPI_FAILURE(status)) { dev_err(dev, FW_BUG "failed to get TPM2 ACPI table\n"); return -EINVAL; } + if (buf->header.length < sizeof(*buf)) { + rc = -EINVAL; + goto out; + } /* Should the FIFO driver handle this? */ sm = buf->start_method; -- 2.34.1
