Message ID | 1589818051-20549-12-git-send-email-pmorel@linux.ibm.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | s390x: Testing the Channel Subsystem I/O | expand |
On Mon, 18 May 2020 18:07:30 +0200 Pierre Morel <pmorel@linux.ibm.com> wrote: > We add a new css_lib file to contain the I/O functions we may > share with different tests. > First function is the subchannel_enable() function. > > When a channel is enabled we can start a SENSE_ID command using > the SSCH instruction to recognize the control unit and device. > > This tests the success of SSCH, the I/O interruption and the TSCH > instructions. > > The test expects a device with a control unit type of 0xC0CA as the > first subchannel of the CSS. It might make sense to extend this to be able to check for any expected type (e.g. 0x3832, should my suggestion to split css tests and css-pong tests make sense.) > > Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> > --- > lib/s390x/css.h | 20 ++++++ > lib/s390x/css_lib.c | 55 +++++++++++++++++ > s390x/Makefile | 1 + > s390x/css.c | 145 ++++++++++++++++++++++++++++++++++++++++++++ > 4 files changed, 221 insertions(+) > create mode 100644 lib/s390x/css_lib.c (...) > +int enable_subchannel(unsigned int sid) > +{ > + struct schib schib; > + struct pmcw *pmcw = &schib.pmcw; > + int try_count = 5; > + int cc; > + > + if (!(sid & SID_ONE)) > + return -1; Hm... this error is indistinguishable for the caller from a cc 1 for the msch. Use something else (as this is a coding error)? > + > + cc = stsch(sid, &schib); > + if (cc) > + return -cc; > + > + do { > + pmcw->flags |= PMCW_ENABLE; > + > + cc = msch(sid, &schib); > + if (cc) > + return -cc; > + > + cc = stsch(sid, &schib); > + if (cc) > + return -cc; > + > + } while (!(pmcw->flags & PMCW_ENABLE) && --try_count); > + > + return try_count; How useful is that information for the caller? I don't see the code below making use of it. > +} > + > +int start_ccw1_chain(unsigned int sid, struct ccw1 *ccw) > +{ > + struct orb orb; > + > + orb.intparm = sid; Just an idea: If you use something else here (maybe the cpa), and set the intparm to the sid in msch, you can test something else: Does msch properly set the intparm, and is that intparm overwritten by a successful ssch, until the next ssch or msch comes around? > + orb.ctrl = ORB_F_INIT_IRQ|ORB_F_FORMAT|ORB_F_LPM_DFLT; > + orb.cpa = (unsigned int) (unsigned long)ccw; Use a struct initializer, so that unset fields are 0? > + > + return ssch(sid, &orb); > +} (...) > +/* > + * test_sense > + * Pre-requisits: > + * - We need the QEMU PONG device as the first recognized > + * device by the enumeration. > + * - ./s390x-run s390x/css.elf -device ccw-pong,cu_type=0xc0ca > + */ > +static void test_sense(void) > +{ > + int ret; > + > + if (!test_device_sid) { > + report_skip("No device"); > + return; > + } > + > + ret = enable_subchannel(test_device_sid); > + if (ret < 0) { > + report(0, > + "Could not enable the subchannel: %08x", > + test_device_sid); > + return; > + } > + > + ret = register_io_int_func(irq_io); > + if (ret) { > + report(0, "Could not register IRQ handler"); > + goto unreg_cb; > + } > + > + lowcore->io_int_param = 0; > + > + ret = start_subchannel(CCW_CMD_SENSE_ID, &senseid, sizeof(senseid), > + CCW_F_SLI); Clear senseid, before actually sending the program? > + if (!ret) { > + report(0, "ssch failed for SENSE ID on sch %08x", > + test_device_sid); > + goto unreg_cb; > + } > + > + wait_for_interrupt(PSW_MASK_IO); > + > + if (lowcore->io_int_param != test_device_sid) > + goto unreg_cb; > + > + report_info("reserved %02x cu_type %04x cu_model %02x dev_type %04x dev_model %02x", > + senseid.reserved, senseid.cu_type, senseid.cu_model, > + senseid.dev_type, senseid.dev_model); > + I'd also recommend checking that senseid.reserved is indeed 0xff -- in combination with senseid clearing before the ssch, that ensures that the senseid structure has actually been written to and is not pure garbage. (It's also a cu type agnostic test :) It also might make sense to check how much data you actually got, as you set SLI. > + report((senseid.cu_type == PONG_CU), > + "cu_type: expect 0x%04x got 0x%04x", > + PONG_CU_TYPE, senseid.cu_type); > + > +unreg_cb: > + unregister_io_int_func(irq_io); > +} > + > static struct { > const char *name; > void (*func)(void); > } tests[] = { > { "enumerate (stsch)", test_enumerate }, > { "enable (msch)", test_enable }, > + { "sense (ssch/tsch)", test_sense }, > { NULL, NULL } > }; > > @@ -145,6 +289,7 @@ int main(int argc, char *argv[]) > int i; > > report_prefix_push("Channel Subsystem"); > + enable_io_isc(); > for (i = 0; tests[i].name; i++) { > report_prefix_push(tests[i].name); > tests[i].func();
On 2020-05-27 12:09, Cornelia Huck wrote: > On Mon, 18 May 2020 18:07:30 +0200 > Pierre Morel <pmorel@linux.ibm.com> wrote: > >> We add a new css_lib file to contain the I/O functions we may >> share with different tests. >> First function is the subchannel_enable() function. >> >> When a channel is enabled we can start a SENSE_ID command using >> the SSCH instruction to recognize the control unit and device. >> >> This tests the success of SSCH, the I/O interruption and the TSCH >> instructions. >> >> The test expects a device with a control unit type of 0xC0CA as the >> first subchannel of the CSS. > > It might make sense to extend this to be able to check for any expected > type (e.g. 0x3832, should my suggestion to split css tests and css-pong > tests make sense.) right. > >> >> Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> >> --- >> lib/s390x/css.h | 20 ++++++ >> lib/s390x/css_lib.c | 55 +++++++++++++++++ >> s390x/Makefile | 1 + >> s390x/css.c | 145 ++++++++++++++++++++++++++++++++++++++++++++ >> 4 files changed, 221 insertions(+) >> create mode 100644 lib/s390x/css_lib.c > > (...) > >> +int enable_subchannel(unsigned int sid) >> +{ >> + struct schib schib; >> + struct pmcw *pmcw = &schib.pmcw; >> + int try_count = 5; >> + int cc; >> + >> + if (!(sid & SID_ONE)) >> + return -1; > > Hm... this error is indistinguishable for the caller from a cc 1 for > the msch. Use something else (as this is a coding error)? right it is a coding error -> assert ? > >> + >> + cc = stsch(sid, &schib); >> + if (cc) >> + return -cc; >> + >> + do { >> + pmcw->flags |= PMCW_ENABLE; >> + >> + cc = msch(sid, &schib); >> + if (cc) >> + return -cc; >> + >> + cc = stsch(sid, &schib); >> + if (cc) >> + return -cc; >> + >> + } while (!(pmcw->flags & PMCW_ENABLE) && --try_count); >> + >> + return try_count; > > How useful is that information for the caller? I don't see the code > below making use of it. right, I will change the fail cases to a report_info and return 0 in case of success. > >> +} >> + >> +int start_ccw1_chain(unsigned int sid, struct ccw1 *ccw) >> +{ >> + struct orb orb; >> + >> + orb.intparm = sid; > > Just an idea: If you use something else here (maybe the cpa), and set > the intparm to the sid in msch, you can test something else: Does msch > properly set the intparm, and is that intparm overwritten by a > successful ssch, until the next ssch or msch comes around? good idea. Using cpa is all what we need at the current development. > >> + orb.ctrl = ORB_F_INIT_IRQ|ORB_F_FORMAT|ORB_F_LPM_DFLT; >> + orb.cpa = (unsigned int) (unsigned long)ccw; > > Use a struct initializer, so that unset fields are 0? not only more beautifull but effective. thanks! > >> + >> + return ssch(sid, &orb); >> +} > > (...) > >> +/* >> + * test_sense >> + * Pre-requisits: >> + * - We need the QEMU PONG device as the first recognized >> + * device by the enumeration. >> + * - ./s390x-run s390x/css.elf -device ccw-pong,cu_type=0xc0ca >> + */ >> +static void test_sense(void) >> +{ >> + int ret; >> + >> + if (!test_device_sid) { >> + report_skip("No device"); >> + return; >> + } >> + >> + ret = enable_subchannel(test_device_sid); >> + if (ret < 0) { >> + report(0, >> + "Could not enable the subchannel: %08x", >> + test_device_sid); >> + return; >> + } >> + >> + ret = register_io_int_func(irq_io); >> + if (ret) { >> + report(0, "Could not register IRQ handler"); >> + goto unreg_cb; >> + } >> + >> + lowcore->io_int_param = 0; >> + >> + ret = start_subchannel(CCW_CMD_SENSE_ID, &senseid, sizeof(senseid), >> + CCW_F_SLI); > > Clear senseid, before actually sending the program? yes. > >> + if (!ret) { >> + report(0, "ssch failed for SENSE ID on sch %08x", >> + test_device_sid); >> + goto unreg_cb; >> + } >> + >> + wait_for_interrupt(PSW_MASK_IO); >> + >> + if (lowcore->io_int_param != test_device_sid) >> + goto unreg_cb; >> + >> + report_info("reserved %02x cu_type %04x cu_model %02x dev_type %04x dev_model %02x", >> + senseid.reserved, senseid.cu_type, senseid.cu_model, >> + senseid.dev_type, senseid.dev_model); >> + > > I'd also recommend checking that senseid.reserved is indeed 0xff -- in > combination with senseid clearing before the ssch, that ensures that > the senseid structure has actually been written to and is not pure > garbage. (It's also a cu type agnostic test :) good idea, thanks. > > It also might make sense to check how much data you actually got, as > you set SLI. > > Yes, will do. Thanks for the comments, I make the changes for the next revision. Regards, Pierre
On Fri, 5 Jun 2020 09:37:39 +0200 Pierre Morel <pmorel@linux.ibm.com> wrote: > On 2020-05-27 12:09, Cornelia Huck wrote: > > On Mon, 18 May 2020 18:07:30 +0200 > > Pierre Morel <pmorel@linux.ibm.com> wrote: > > > >> We add a new css_lib file to contain the I/O functions we may > >> share with different tests. > >> First function is the subchannel_enable() function. > >> > >> When a channel is enabled we can start a SENSE_ID command using > >> the SSCH instruction to recognize the control unit and device. > >> > >> This tests the success of SSCH, the I/O interruption and the TSCH > >> instructions. > >> > >> The test expects a device with a control unit type of 0xC0CA as the > >> first subchannel of the CSS. > > > > It might make sense to extend this to be able to check for any expected > > type (e.g. 0x3832, should my suggestion to split css tests and css-pong > > tests make sense.) > > right. > > > > >> > >> Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> > >> --- > >> lib/s390x/css.h | 20 ++++++ > >> lib/s390x/css_lib.c | 55 +++++++++++++++++ > >> s390x/Makefile | 1 + > >> s390x/css.c | 145 ++++++++++++++++++++++++++++++++++++++++++++ > >> 4 files changed, 221 insertions(+) > >> create mode 100644 lib/s390x/css_lib.c > > > > (...) > > > >> +int enable_subchannel(unsigned int sid) > >> +{ > >> + struct schib schib; > >> + struct pmcw *pmcw = &schib.pmcw; > >> + int try_count = 5; > >> + int cc; > >> + > >> + if (!(sid & SID_ONE)) > >> + return -1; > > > > Hm... this error is indistinguishable for the caller from a cc 1 for > > the msch. Use something else (as this is a coding error)? > > right it is a coding error -> assert ? Sounds good to me.
diff --git a/lib/s390x/css.h b/lib/s390x/css.h index e0d3a98..f4df7bc 100644 --- a/lib/s390x/css.h +++ b/lib/s390x/css.h @@ -99,6 +99,19 @@ struct irb { uint32_t emw[8]; } __attribute__ ((aligned(4))); +#define CCW_CMD_SENSE_ID 0xe4 +#define PONG_CU 0xc0ca +struct senseid { + /* common part */ + uint8_t reserved; /* always 0x'FF' */ + uint16_t cu_type; /* control unit type */ + uint8_t cu_model; /* control unit model */ + uint16_t dev_type; /* device type */ + uint8_t dev_model; /* device model */ + uint8_t unused; /* padding byte */ + uint8_t padding[256 - 10]; /* Extra padding for CCW */ +} __attribute__ ((aligned(4))) __attribute__ ((packed)); + /* CSS low level access functions */ static inline int ssch(unsigned long schid, struct orb *addr) @@ -256,4 +269,11 @@ static inline struct ccw *dump_ccw(struct ccw *cp) return NULL; } #endif /* DEBUG_CSS */ + +#define SID_ONE 0x00010000 + +/* Library functions */ +int enable_subchannel(unsigned int sid); +int start_ccw1_chain(unsigned int sid, struct ccw1 *ccw); + #endif diff --git a/lib/s390x/css_lib.c b/lib/s390x/css_lib.c new file mode 100644 index 0000000..80b9359 --- /dev/null +++ b/lib/s390x/css_lib.c @@ -0,0 +1,55 @@ +/* + * Channel subsystem library functions + * + * Copyright (c) 2020 IBM Corp. + * + * Authors: + * Pierre Morel <pmorel@linux.ibm.com> + * + * This code is free software; you can redistribute it and/or modify it + * under the terms of the GNU General Public License version 2. + */ +#include <stdint.h> +#include <stddef.h> +#include <css.h> + +int enable_subchannel(unsigned int sid) +{ + struct schib schib; + struct pmcw *pmcw = &schib.pmcw; + int try_count = 5; + int cc; + + if (!(sid & SID_ONE)) + return -1; + + cc = stsch(sid, &schib); + if (cc) + return -cc; + + do { + pmcw->flags |= PMCW_ENABLE; + + cc = msch(sid, &schib); + if (cc) + return -cc; + + cc = stsch(sid, &schib); + if (cc) + return -cc; + + } while (!(pmcw->flags & PMCW_ENABLE) && --try_count); + + return try_count; +} + +int start_ccw1_chain(unsigned int sid, struct ccw1 *ccw) +{ + struct orb orb; + + orb.intparm = sid; + orb.ctrl = ORB_F_INIT_IRQ|ORB_F_FORMAT|ORB_F_LPM_DFLT; + orb.cpa = (unsigned int) (unsigned long)ccw; + + return ssch(sid, &orb); +} diff --git a/s390x/Makefile b/s390x/Makefile index baebf18..166cb5c 100644 --- a/s390x/Makefile +++ b/s390x/Makefile @@ -53,6 +53,7 @@ cflatobjs += lib/s390x/interrupt.o cflatobjs += lib/s390x/mmu.o cflatobjs += lib/s390x/smp.o cflatobjs += lib/s390x/css_dump.o +cflatobjs += lib/s390x/css_lib.o OBJDIRS += lib/s390x diff --git a/s390x/css.c b/s390x/css.c index 1b60a47..c94a916 100644 --- a/s390x/css.c +++ b/s390x/css.c @@ -20,8 +20,24 @@ #include <css.h> +#define PONG_CU_TYPE 0xc0ca + +struct lowcore *lowcore = (void *)0x0; + static struct schib schib; static int test_device_sid; +#define NUM_CCW 100 +static struct ccw1 ccw[NUM_CCW]; +static struct irb irb; +static struct senseid senseid; + +static void set_io_irq_subclass_mask(uint64_t const new_mask) +{ + asm volatile ( + "lctlg %%c6, %%c6, %[source]\n" + : /* No outputs */ + : [source] "R" (new_mask)); +} static void test_enumerate(void) { @@ -131,12 +147,140 @@ retry: test_device_sid, retry_count, pmcw->flags); } +static void enable_io_isc(void) +{ + /* Let's enable all ISCs for I/O interrupt */ + set_io_irq_subclass_mask(0x00000000ff000000); +} + +static void irq_io(void) +{ + int ret = 0; + char *flags; + int sid; + + report_prefix_push("Interrupt"); + /* Lowlevel set the SID as interrupt parameter. */ + if (lowcore->io_int_param != test_device_sid) { + report(0, + "Bad io_int_param: %x expected %x", + lowcore->io_int_param, test_device_sid); + goto pop; + } + report_prefix_pop(); + + report_prefix_push("tsch"); + sid = lowcore->subsys_id_word; + ret = tsch(sid, &irb); + switch (ret) { + case 1: + dump_irb(&irb); + flags = dump_scsw_flags(irb.scsw.ctrl); + report(0, + "I/O interrupt, CC 1 but tsch reporting sch %08x as not status pending: %s", + sid, flags); + break; + case 2: + report(0, "tsch returns unexpected CC 2"); + break; + case 3: + report(0, "tsch reporting sch %08x as not operational", sid); + break; + case 0: + /* Stay humble on success */ + break; + } +pop: + report_prefix_pop(); + lowcore->io_old_psw.mask &= ~PSW_MASK_WAIT; +} + +static int start_subchannel(int code, void *data, int count, + unsigned char flags) +{ + int ret; + + report_prefix_push("start_senseid"); + /* Build the CCW chain with a single CCW */ + ccw[0].code = code; + ccw[0].flags = flags; /* No flags need to be set */ + ccw[0].count = count; + ccw[0].data_address = (int)(unsigned long)data; + + ret = start_ccw1_chain(test_device_sid, ccw); + if (ret) { + report(0, "start_ccw_chain failed ret=%d", ret); + report_prefix_pop(); + return 0; + } + report_prefix_pop(); + return 1; +} + +/* + * test_sense + * Pre-requisits: + * - We need the QEMU PONG device as the first recognized + * device by the enumeration. + * - ./s390x-run s390x/css.elf -device ccw-pong,cu_type=0xc0ca + */ +static void test_sense(void) +{ + int ret; + + if (!test_device_sid) { + report_skip("No device"); + return; + } + + ret = enable_subchannel(test_device_sid); + if (ret < 0) { + report(0, + "Could not enable the subchannel: %08x", + test_device_sid); + return; + } + + ret = register_io_int_func(irq_io); + if (ret) { + report(0, "Could not register IRQ handler"); + goto unreg_cb; + } + + lowcore->io_int_param = 0; + + ret = start_subchannel(CCW_CMD_SENSE_ID, &senseid, sizeof(senseid), + CCW_F_SLI); + if (!ret) { + report(0, "ssch failed for SENSE ID on sch %08x", + test_device_sid); + goto unreg_cb; + } + + wait_for_interrupt(PSW_MASK_IO); + + if (lowcore->io_int_param != test_device_sid) + goto unreg_cb; + + report_info("reserved %02x cu_type %04x cu_model %02x dev_type %04x dev_model %02x", + senseid.reserved, senseid.cu_type, senseid.cu_model, + senseid.dev_type, senseid.dev_model); + + report((senseid.cu_type == PONG_CU), + "cu_type: expect 0x%04x got 0x%04x", + PONG_CU_TYPE, senseid.cu_type); + +unreg_cb: + unregister_io_int_func(irq_io); +} + static struct { const char *name; void (*func)(void); } tests[] = { { "enumerate (stsch)", test_enumerate }, { "enable (msch)", test_enable }, + { "sense (ssch/tsch)", test_sense }, { NULL, NULL } }; @@ -145,6 +289,7 @@ int main(int argc, char *argv[]) int i; report_prefix_push("Channel Subsystem"); + enable_io_isc(); for (i = 0; tests[i].name; i++) { report_prefix_push(tests[i].name); tests[i].func();
We add a new css_lib file to contain the I/O functions we may share with different tests. First function is the subchannel_enable() function. When a channel is enabled we can start a SENSE_ID command using the SSCH instruction to recognize the control unit and device. This tests the success of SSCH, the I/O interruption and the TSCH instructions. The test expects a device with a control unit type of 0xC0CA as the first subchannel of the CSS. Signed-off-by: Pierre Morel <pmorel@linux.ibm.com> --- lib/s390x/css.h | 20 ++++++ lib/s390x/css_lib.c | 55 +++++++++++++++++ s390x/Makefile | 1 + s390x/css.c | 145 ++++++++++++++++++++++++++++++++++++++++++++ 4 files changed, 221 insertions(+) create mode 100644 lib/s390x/css_lib.c