| Message ID | 20231216010729.2904751-8-tusharsu@linux.microsoft.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | ima: kexec: measure events between kexec load and execute | expand |
On Fri, 2023-12-15 at 17:07 -0800, Tushar Sugandhi wrote: > There could be a potential mismatch between IMA measurements and TPM PCR > quotes caused by the indeterminate interval between kexec 'load' and > 'execute'. Memory allocated at kexec 'load' for IMA log buffer may run > out. It can lead to missing events in the IMA log after a soft reboot to > the new Kernel, resulting in TPM PCR quotes mismatch and remote > attestation failures. > > Define two new IMA events, 'kexec_load' and 'kexec_execute', to be > measured as critical data at kexec 'load' and 'execute' respectively. > > These events serve as markers in the IMA log to verify the IMA events are > captured between kexec 'load' and 'execute' window. The presence of a > 'kexec_load' event in between the last two 'boot_aggregate' events in the > IMA log implies this is a kexec soft reboot, and not a cold-boot. And the > absence of 'kexec_execute' event after kexec soft reboot implies missing > events in that window which results in inconsistency with TPM PCR quotes, > necessitating a cold boot for further successful remote attestation. > > Signed-off-by: Tushar Sugandhi <tusharsu@linux.microsoft.com> > --- > security/integrity/ima/ima_kexec.c | 23 ++++++++++++++++++++++- > 1 file changed, 22 insertions(+), 1 deletion(-) > > diff --git a/security/integrity/ima/ima_kexec.c b/security/integrity/ima/ima_kexec.c > index 063da9c834a0..47da41a66dcc 100644 > --- a/security/integrity/ima/ima_kexec.c > +++ b/security/integrity/ima/ima_kexec.c > @@ -17,6 +17,8 @@ > #include "ima.h" > > #ifdef CONFIG_IMA_KEXEC > +#define IMA_KEXEC_EVENT_LEN 128 > + > struct seq_file ima_kexec_file; > static void *ima_kexec_buffer; > static size_t kexec_segment_size; > @@ -33,6 +35,8 @@ void ima_free_kexec_file_buf(struct seq_file *sf) > > static int ima_alloc_kexec_file_buf(size_t segment_size) > { > + char ima_kexec_event[IMA_KEXEC_EVENT_LEN]; > + > /* > * kexec 'load' may be called multiple times. > * Free and realloc the buffer only if the segment_size is > @@ -42,7 +46,7 @@ static int ima_alloc_kexec_file_buf(size_t segment_size) > ima_kexec_file.size == segment_size && > ima_kexec_file.read_pos == 0 && > ima_kexec_file.count == sizeof(struct ima_kexec_hdr)) > - return 0; > + goto out; > > ima_free_kexec_file_buf(&ima_kexec_file); > > @@ -55,6 +59,13 @@ static int ima_alloc_kexec_file_buf(size_t segment_size) > ima_kexec_file.read_pos = 0; > ima_kexec_file.count = sizeof(struct ima_kexec_hdr); /* reserved space */ > > +out: > + scnprintf(ima_kexec_event, IMA_KEXEC_EVENT_LEN, > + "kexec_segment_size=%lu;", segment_size); > + > + ima_measure_critical_data("ima_kexec", "kexec_load", ima_kexec_event, > + strlen(ima_kexec_event), false, NULL, 0); > + > return 0; > } > > @@ -179,6 +190,7 @@ void ima_add_kexec_buffer(struct kimage *image) > static int ima_update_kexec_buffer(struct notifier_block *self, > unsigned long action, void *data) > { > + char ima_kexec_event[IMA_KEXEC_EVENT_LEN]; > void *buf = NULL; > size_t buf_size; > bool resume = false; > @@ -194,6 +206,15 @@ static int ima_update_kexec_buffer(struct notifier_block *self, > return ret; > } > > + buf_size = ima_get_binary_runtime_size(); > + scnprintf(ima_kexec_event, IMA_KEXEC_EVENT_LEN, > + "kexec_segment_size=%lu;ima_binary_runtime_size=%lu;", > + kexec_segment_size, buf_size); > + > + ima_measure_critical_data("ima_kexec", "kexec_execute", > + ima_kexec_event, strlen(ima_kexec_event), > + false, NULL, 0); > + Please consider including the number of measurement records as well. > ima_measurements_suspend(); > > ret = ima_dump_measurement_list(&buf_size, &buf,
On 12/20/23 12:41, Mimi Zohar wrote: > On Fri, 2023-12-15 at 17:07 -0800, Tushar Sugandhi wrote: >> There could be a potential mismatch between IMA measurements and TPM PCR >> quotes caused by the indeterminate interval between kexec 'load' and >> 'execute'. Memory allocated at kexec 'load' for IMA log buffer may run >> out. It can lead to missing events in the IMA log after a soft reboot to >> the new Kernel, resulting in TPM PCR quotes mismatch and remote >> attestation failures. >> >> Define two new IMA events, 'kexec_load' and 'kexec_execute', to be >> measured as critical data at kexec 'load' and 'execute' respectively. >> >> These events serve as markers in the IMA log to verify the IMA events are >> captured between kexec 'load' and 'execute' window. The presence of a >> 'kexec_load' event in between the last two 'boot_aggregate' events in the >> IMA log implies this is a kexec soft reboot, and not a cold-boot. And the >> absence of 'kexec_execute' event after kexec soft reboot implies missing >> events in that window which results in inconsistency with TPM PCR quotes, >> necessitating a cold boot for further successful remote attestation. >> >> Signed-off-by: Tushar Sugandhi <tusharsu@linux.microsoft.com> >> --- >> security/integrity/ima/ima_kexec.c | 23 ++++++++++++++++++++++- >> 1 file changed, 22 insertions(+), 1 deletion(-) >> >> diff --git a/security/integrity/ima/ima_kexec.c b/security/integrity/ima/ima_kexec.c >> index 063da9c834a0..47da41a66dcc 100644 >> --- a/security/integrity/ima/ima_kexec.c >> +++ b/security/integrity/ima/ima_kexec.c >> @@ -17,6 +17,8 @@ >> #include "ima.h" >> >> #ifdef CONFIG_IMA_KEXEC >> +#define IMA_KEXEC_EVENT_LEN 128 >> + >> struct seq_file ima_kexec_file; >> static void *ima_kexec_buffer; >> static size_t kexec_segment_size; >> @@ -33,6 +35,8 @@ void ima_free_kexec_file_buf(struct seq_file *sf) >> >> static int ima_alloc_kexec_file_buf(size_t segment_size) >> { >> + char ima_kexec_event[IMA_KEXEC_EVENT_LEN]; >> + >> /* >> * kexec 'load' may be called multiple times. >> * Free and realloc the buffer only if the segment_size is >> @@ -42,7 +46,7 @@ static int ima_alloc_kexec_file_buf(size_t segment_size) >> ima_kexec_file.size == segment_size && >> ima_kexec_file.read_pos == 0 && >> ima_kexec_file.count == sizeof(struct ima_kexec_hdr)) >> - return 0; >> + goto out; >> >> ima_free_kexec_file_buf(&ima_kexec_file); >> >> @@ -55,6 +59,13 @@ static int ima_alloc_kexec_file_buf(size_t segment_size) >> ima_kexec_file.read_pos = 0; >> ima_kexec_file.count = sizeof(struct ima_kexec_hdr); /* reserved space */ >> >> +out: >> + scnprintf(ima_kexec_event, IMA_KEXEC_EVENT_LEN, >> + "kexec_segment_size=%lu;", segment_size); >> + >> + ima_measure_critical_data("ima_kexec", "kexec_load", ima_kexec_event, >> + strlen(ima_kexec_event), false, NULL, 0); >> + >> return 0; >> } >> >> @@ -179,6 +190,7 @@ void ima_add_kexec_buffer(struct kimage *image) >> static int ima_update_kexec_buffer(struct notifier_block *self, >> unsigned long action, void *data) >> { >> + char ima_kexec_event[IMA_KEXEC_EVENT_LEN]; >> void *buf = NULL; >> size_t buf_size; >> bool resume = false; >> @@ -194,6 +206,15 @@ static int ima_update_kexec_buffer(struct notifier_block *self, >> return ret; >> } >> >> + buf_size = ima_get_binary_runtime_size(); >> + scnprintf(ima_kexec_event, IMA_KEXEC_EVENT_LEN, >> + "kexec_segment_size=%lu;ima_binary_runtime_size=%lu;", >> + kexec_segment_size, buf_size); >> + >> + ima_measure_critical_data("ima_kexec", "kexec_execute", >> + ima_kexec_event, strlen(ima_kexec_event), >> + false, NULL, 0); >> + > > Please consider including the number of measurement records as well. Will do. I think that would be valuable information. Per my understanding, I will have to use the function ima_show_measurements_count() or ima_show_htable_value() to get the number of measurement records value[1]. So I will have to expose that function from "ima_fs.c" to "ima_kexec.c". Hope that's ok. [1] https://elixir.bootlin.com/linux/latest/sourcesecurity/integrity/ima/ima_fs.c static ssize_t ima_show_measurements_count(struct file *filp, char __user *buf, size_t count, loff_t *ppos) { return ima_show_htable_value(buf, count, ppos, &ima_htable.len); } ~Tushar > >> ima_measurements_suspend(); >> >> ret = ima_dump_measurement_list(&buf_size, &buf, >
On Fri, 2024-01-05 at 12:22 -0800, Tushar Sugandhi wrote: > >> @@ -194,6 +206,15 @@ static int ima_update_kexec_buffer(struct notifier_block *self, > >> return ret; > >> } > >> > >> + buf_size = ima_get_binary_runtime_size(); > >> + scnprintf(ima_kexec_event, IMA_KEXEC_EVENT_LEN, > >> + "kexec_segment_size=%lu;ima_binary_runtime_size=%lu;", > >> + kexec_segment_size, buf_size); > >> + > >> + ima_measure_critical_data("ima_kexec", "kexec_execute", > >> + ima_kexec_event, strlen(ima_kexec_event), > >> + false, NULL, 0); > >> + > > > > Please consider including the number of measurement records as well. > Will do. I think that would be valuable information. > > Per my understanding, I will have to use the function > ima_show_measurements_count() or ima_show_htable_value() to get the > number of measurement records value[1]. So I will have to expose that > function from "ima_fs.c" to "ima_kexec.c". Hope that's ok. > > [1] > https://elixir.bootlin.com/linux/latest/sourcesecurity/integrity/ima/ima_fs.c > > > static ssize_t ima_show_measurements_count(struct file *filp, > char __user *buf, > size_t count, loff_t *ppos) > { > return ima_show_htable_value(buf, count, ppos, &ima_htable.len); I don't see a need to expose this function. ima_htable is defined in ima.h. You can read the ima_htable.len directly, as ima_show_htable_value does.
On 1/7/24 06:24, Mimi Zohar wrote: > On Fri, 2024-01-05 at 12:22 -0800, Tushar Sugandhi wrote: >>>> @@ -194,6 +206,15 @@ static int ima_update_kexec_buffer(struct notifier_block *self, >>>> return ret; >>>> } >>>> >>>> + buf_size = ima_get_binary_runtime_size(); >>>> + scnprintf(ima_kexec_event, IMA_KEXEC_EVENT_LEN, >>>> + "kexec_segment_size=%lu;ima_binary_runtime_size=%lu;", >>>> + kexec_segment_size, buf_size); >>>> + >>>> + ima_measure_critical_data("ima_kexec", "kexec_execute", >>>> + ima_kexec_event, strlen(ima_kexec_event), >>>> + false, NULL, 0); >>>> + >>> >>> Please consider including the number of measurement records as well. >> Will do. I think that would be valuable information. >> >> Per my understanding, I will have to use the function >> ima_show_measurements_count() or ima_show_htable_value() to get the >> number of measurement records value[1]. So I will have to expose that >> function from "ima_fs.c" to "ima_kexec.c". Hope that's ok. >> >> [1] >> https://elixir.bootlin.com/linux/latest/sourcesecurity/integrity/ima/ima_fs.c >> >> >> static ssize_t ima_show_measurements_count(struct file *filp, >> char __user *buf, >> size_t count, loff_t *ppos) >> { >> return ima_show_htable_value(buf, count, ppos, &ima_htable.len); > > I don't see a need to expose this function. ima_htable is defined in ima.h. > You can read the ima_htable.len directly, as ima_show_htable_value does. > Agreed. Thanks for the pointer. That's what I concluded too when I was implementing this change. I will use ima_htable.len directly. ~Tushar
diff --git a/security/integrity/ima/ima_kexec.c b/security/integrity/ima/ima_kexec.c index 063da9c834a0..47da41a66dcc 100644 --- a/security/integrity/ima/ima_kexec.c +++ b/security/integrity/ima/ima_kexec.c @@ -17,6 +17,8 @@ #include "ima.h" #ifdef CONFIG_IMA_KEXEC +#define IMA_KEXEC_EVENT_LEN 128 + struct seq_file ima_kexec_file; static void *ima_kexec_buffer; static size_t kexec_segment_size; @@ -33,6 +35,8 @@ void ima_free_kexec_file_buf(struct seq_file *sf) static int ima_alloc_kexec_file_buf(size_t segment_size) { + char ima_kexec_event[IMA_KEXEC_EVENT_LEN]; + /* * kexec 'load' may be called multiple times. * Free and realloc the buffer only if the segment_size is @@ -42,7 +46,7 @@ static int ima_alloc_kexec_file_buf(size_t segment_size) ima_kexec_file.size == segment_size && ima_kexec_file.read_pos == 0 && ima_kexec_file.count == sizeof(struct ima_kexec_hdr)) - return 0; + goto out; ima_free_kexec_file_buf(&ima_kexec_file); @@ -55,6 +59,13 @@ static int ima_alloc_kexec_file_buf(size_t segment_size) ima_kexec_file.read_pos = 0; ima_kexec_file.count = sizeof(struct ima_kexec_hdr); /* reserved space */ +out: + scnprintf(ima_kexec_event, IMA_KEXEC_EVENT_LEN, + "kexec_segment_size=%lu;", segment_size); + + ima_measure_critical_data("ima_kexec", "kexec_load", ima_kexec_event, + strlen(ima_kexec_event), false, NULL, 0); + return 0; } @@ -179,6 +190,7 @@ void ima_add_kexec_buffer(struct kimage *image) static int ima_update_kexec_buffer(struct notifier_block *self, unsigned long action, void *data) { + char ima_kexec_event[IMA_KEXEC_EVENT_LEN]; void *buf = NULL; size_t buf_size; bool resume = false; @@ -194,6 +206,15 @@ static int ima_update_kexec_buffer(struct notifier_block *self, return ret; } + buf_size = ima_get_binary_runtime_size(); + scnprintf(ima_kexec_event, IMA_KEXEC_EVENT_LEN, + "kexec_segment_size=%lu;ima_binary_runtime_size=%lu;", + kexec_segment_size, buf_size); + + ima_measure_critical_data("ima_kexec", "kexec_execute", + ima_kexec_event, strlen(ima_kexec_event), + false, NULL, 0); + ima_measurements_suspend(); ret = ima_dump_measurement_list(&buf_size, &buf,
There could be a potential mismatch between IMA measurements and TPM PCR quotes caused by the indeterminate interval between kexec 'load' and 'execute'. Memory allocated at kexec 'load' for IMA log buffer may run out. It can lead to missing events in the IMA log after a soft reboot to the new Kernel, resulting in TPM PCR quotes mismatch and remote attestation failures. Define two new IMA events, 'kexec_load' and 'kexec_execute', to be measured as critical data at kexec 'load' and 'execute' respectively. These events serve as markers in the IMA log to verify the IMA events are captured between kexec 'load' and 'execute' window. The presence of a 'kexec_load' event in between the last two 'boot_aggregate' events in the IMA log implies this is a kexec soft reboot, and not a cold-boot. And the absence of 'kexec_execute' event after kexec soft reboot implies missing events in that window which results in inconsistency with TPM PCR quotes, necessitating a cold boot for further successful remote attestation. Signed-off-by: Tushar Sugandhi <tusharsu@linux.microsoft.com> --- security/integrity/ima/ima_kexec.c | 23 ++++++++++++++++++++++- 1 file changed, 22 insertions(+), 1 deletion(-)