| Message ID | 20211022131057.1308851-2-scgl@linux.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Test spec exception interception | expand |
On 22/10/2021 15.10, Janis Schoetterl-Glausch wrote: > Check that specification exceptions cause intercepts when > specification exception interpretation is off. > Check that specification exceptions caused by program new PSWs > cause interceptions. > We cannot assert that non program new PSW specification exceptions > are interpreted because whether interpretation occurs or not is > configuration dependent. > > Signed-off-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com> > Reviewed-by: Janosch Frank <frankja@de.ibm.com> > Reviewed-by: Thomas Huth <thuth@redhat.com> > --- > s390x/Makefile | 2 + > lib/s390x/sie.h | 1 + > s390x/snippets/c/spec_ex.c | 21 ++++++++++ > s390x/spec_ex-sie.c | 82 ++++++++++++++++++++++++++++++++++++++ > s390x/unittests.cfg | 3 ++ > 5 files changed, 109 insertions(+) > create mode 100644 s390x/snippets/c/spec_ex.c > create mode 100644 s390x/spec_ex-sie.c Thanks, I've pushed it to the repository now. Thomas
On 22/10/2021 15.10, Janis Schoetterl-Glausch wrote: > Check that specification exceptions cause intercepts when > specification exception interpretation is off. > Check that specification exceptions caused by program new PSWs > cause interceptions. > We cannot assert that non program new PSW specification exceptions > are interpreted because whether interpretation occurs or not is > configuration dependent. > > Signed-off-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com> > Reviewed-by: Janosch Frank <frankja@de.ibm.com> > Reviewed-by: Thomas Huth <thuth@redhat.com> > --- ... > + report_prefix_push("on"); > + vm.sblk->ecb |= ECB_SPECI; > + reset_guest(); > + sie(&vm); > + /* interpretation on -> configuration dependent if initial exception causes > + * interception, but invalid new program PSW must > + */ > + report(vm.sblk->icptcode == ICPT_PROGI > + && vm.sblk->iprcc == PGM_INT_CODE_SPECIFICATION, > + "Received specification exception intercept"); > + if (vm.sblk->gpsw.addr == 0xdeadbeee) > + report_info("Interpreted initial exception, intercepted invalid program new PSW exception"); > + else > + report_info("Did not interpret initial exception"); Hi Janis! While using this test in our downstream verification of the backport of the related kernel patch, it occurred that the way of only reporting the interpreted exception via report_info() is rather unfortunate for using this test in automatic regression runs. For such regression runs, it would be good if the test would be marked with FAIL if the exception was not interpreted. I know, the interpretation facility is not always there, but still would it be somehow possible to add such a mode? E.g. by checking the machine generation (is this always available with z15 and newer?) and maybe adding a CLI option to force the hard check (so that e.g. "-f" triggers the failure if the exception has not been interpreted, while running the test without "-f" would still do the old behavior instead)? Thomas
On 11/11/21 08:47, Thomas Huth wrote: > On 22/10/2021 15.10, Janis Schoetterl-Glausch wrote: >> Check that specification exceptions cause intercepts when >> specification exception interpretation is off. >> Check that specification exceptions caused by program new PSWs >> cause interceptions. >> We cannot assert that non program new PSW specification exceptions >> are interpreted because whether interpretation occurs or not is >> configuration dependent. >> >> Signed-off-by: Janis Schoetterl-Glausch <scgl@linux.ibm.com> >> Reviewed-by: Janosch Frank <frankja@de.ibm.com> >> Reviewed-by: Thomas Huth <thuth@redhat.com> >> --- > ... >> + report_prefix_push("on"); >> + vm.sblk->ecb |= ECB_SPECI; >> + reset_guest(); >> + sie(&vm); >> + /* interpretation on -> configuration dependent if initial exception causes >> + * interception, but invalid new program PSW must >> + */ >> + report(vm.sblk->icptcode == ICPT_PROGI >> + && vm.sblk->iprcc == PGM_INT_CODE_SPECIFICATION, >> + "Received specification exception intercept"); >> + if (vm.sblk->gpsw.addr == 0xdeadbeee) >> + report_info("Interpreted initial exception, intercepted invalid program new PSW exception"); >> + else >> + report_info("Did not interpret initial exception"); > > Hi Janis! > > While using this test in our downstream verification of the backport of the related kernel patch, it occurred that the way of only reporting the interpreted exception via report_info() is rather unfortunate for using this test in automatic regression runs. For such regression runs, it would be good if the test would be marked with FAIL if the exception was not interpreted. I know, the interpretation facility is not always there, but still would it be somehow possible to add such a mode? E.g. by checking the machine generation (is this always available with z15 and newer?) and maybe adding a CLI option to force the hard check (so that e.g. "-f" triggers the failure if the exception has not been interpreted, while running the test without "-f" would still do the old behavior instead)? > > Thomas > Sounds good, I'll look into it.
diff --git a/s390x/Makefile b/s390x/Makefile index d18b08b..f95f2e6 100644 --- a/s390x/Makefile +++ b/s390x/Makefile @@ -24,6 +24,7 @@ tests += $(TEST_DIR)/mvpg.elf tests += $(TEST_DIR)/uv-host.elf tests += $(TEST_DIR)/edat.elf tests += $(TEST_DIR)/mvpg-sie.elf +tests += $(TEST_DIR)/spec_ex-sie.elf tests_binary = $(patsubst %.elf,%.bin,$(tests)) ifneq ($(HOST_KEY_DOCUMENT),) @@ -86,6 +87,7 @@ snippet_asmlib = $(SNIPPET_DIR)/c/cstart.o lib/auxinfo.o # perquisites (=guests) for the snippet hosts. # $(TEST_DIR)/<snippet-host>.elf: snippets = $(SNIPPET_DIR)/<c/asm>/<snippet>.gbin $(TEST_DIR)/mvpg-sie.elf: snippets = $(SNIPPET_DIR)/c/mvpg-snippet.gbin +$(TEST_DIR)/spec_ex-sie.elf: snippets = $(SNIPPET_DIR)/c/spec_ex.gbin $(SNIPPET_DIR)/asm/%.gbin: $(SNIPPET_DIR)/asm/%.o $(FLATLIBS) $(OBJCOPY) -O binary $(patsubst %.gbin,%.o,$@) $@ diff --git a/lib/s390x/sie.h b/lib/s390x/sie.h index ca514ef..7ef7251 100644 --- a/lib/s390x/sie.h +++ b/lib/s390x/sie.h @@ -98,6 +98,7 @@ struct kvm_s390_sie_block { uint8_t fpf; /* 0x0060 */ #define ECB_GS 0x40 #define ECB_TE 0x10 +#define ECB_SPECI 0x08 #define ECB_SRSI 0x04 #define ECB_HOSTPROTINT 0x02 uint8_t ecb; /* 0x0061 */ diff --git a/s390x/snippets/c/spec_ex.c b/s390x/snippets/c/spec_ex.c new file mode 100644 index 0000000..71655dd --- /dev/null +++ b/s390x/snippets/c/spec_ex.c @@ -0,0 +1,21 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright IBM Corp. 2021 + * + * Snippet used by specification exception interception test. + */ +#include <libcflat.h> +#include <bitops.h> +#include <asm/arch_def.h> + +__attribute__((section(".text"))) int main(void) +{ + struct lowcore *lowcore = (struct lowcore *) 0; + uint64_t bad_psw = 0; + + /* PSW bit 12 has no name or meaning and must be 0 */ + lowcore->pgm_new_psw.mask = BIT(63 - 12); + lowcore->pgm_new_psw.addr = 0xdeadbeee; + asm volatile ("lpsw %0" :: "Q"(bad_psw)); + return 0; +} diff --git a/s390x/spec_ex-sie.c b/s390x/spec_ex-sie.c new file mode 100644 index 0000000..5dea411 --- /dev/null +++ b/s390x/spec_ex-sie.c @@ -0,0 +1,82 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright IBM Corp. 2021 + * + * Specification exception interception test. + * Checks that specification exception interceptions occur as expected when + * specification exception interpretation is off/on. + */ +#include <libcflat.h> +#include <sclp.h> +#include <asm/page.h> +#include <asm/arch_def.h> +#include <alloc_page.h> +#include <vm.h> +#include <sie.h> +#include <snippet.h> + +static struct vm vm; +extern const char SNIPPET_NAME_START(c, spec_ex)[]; +extern const char SNIPPET_NAME_END(c, spec_ex)[]; + +static void setup_guest(void) +{ + char *guest; + int binary_size = SNIPPET_LEN(c, spec_ex); + + setup_vm(); + guest = alloc_pages(8); + memcpy(guest, SNIPPET_NAME_START(c, spec_ex), binary_size); + sie_guest_create(&vm, (uint64_t) guest, HPAGE_SIZE); +} + +static void reset_guest(void) +{ + vm.sblk->gpsw = snippet_psw; + vm.sblk->icptcode = 0; +} + +static void test_spec_ex_sie(void) +{ + setup_guest(); + + report_prefix_push("SIE spec ex interpretation"); + report_prefix_push("off"); + reset_guest(); + sie(&vm); + /* interpretation off -> initial exception must cause interception */ + report(vm.sblk->icptcode == ICPT_PROGI + && vm.sblk->iprcc == PGM_INT_CODE_SPECIFICATION + && vm.sblk->gpsw.addr != 0xdeadbeee, + "Received specification exception intercept for initial exception"); + report_prefix_pop(); + + report_prefix_push("on"); + vm.sblk->ecb |= ECB_SPECI; + reset_guest(); + sie(&vm); + /* interpretation on -> configuration dependent if initial exception causes + * interception, but invalid new program PSW must + */ + report(vm.sblk->icptcode == ICPT_PROGI + && vm.sblk->iprcc == PGM_INT_CODE_SPECIFICATION, + "Received specification exception intercept"); + if (vm.sblk->gpsw.addr == 0xdeadbeee) + report_info("Interpreted initial exception, intercepted invalid program new PSW exception"); + else + report_info("Did not interpret initial exception"); + report_prefix_pop(); + report_prefix_pop(); +} + +int main(int argc, char **argv) +{ + if (!sclp_facilities.has_sief2) { + report_skip("SIEF2 facility unavailable"); + goto out; + } + + test_spec_ex_sie(); +out: + return report_summary(); +} diff --git a/s390x/unittests.cfg b/s390x/unittests.cfg index 9e1802f..3b454b7 100644 --- a/s390x/unittests.cfg +++ b/s390x/unittests.cfg @@ -109,3 +109,6 @@ file = edat.elf [mvpg-sie] file = mvpg-sie.elf + +[spec_ex-sie] +file = spec_ex-sie.elf