| Message ID | 20200915174639.1366226-1-philmd@redhat.com |
|---|---|
| State | New |
| Headers | show |
| Series | hw/scsi/lsi53c895a: Sanitize some trace events format | expand |
ping? On 9/15/20 7:46 PM, Philippe Mathieu-Daudé wrote: > Make some lsi53c895a trace events more understandable. > > Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> > --- > hw/scsi/trace-events | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/hw/scsi/trace-events b/hw/scsi/trace-events > index 9a4a60ca635..ac4209f361e 100644 > --- a/hw/scsi/trace-events > +++ b/hw/scsi/trace-events > @@ -234,9 +234,9 @@ spapr_vscsi_do_crq(unsigned c0, unsigned c1) "crq: %02x %02x ..." > > # lsi53c895a.c > lsi_reset(void) "Reset" > -lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 0x%02x0x%02x" > +lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 1:0x%02x 0:0x%02x" > lsi_update_irq_disconnected(void) "Handled IRQs & disconnected, looking for pending processes" > -lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt 0x%02x0x%02x prev 0x%02x0x%02x" > +lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt stat 1:0x%02x 0:0x%02x sist 1:0x%02x 0:0x%02x" > lsi_script_dma_interrupt(uint8_t stat, uint8_t dstat) "DMA Interrupt 0x%x prev 0x%x" > lsi_bad_phase_jump(uint32_t dsp) "Data phase mismatch jump to 0x%"PRIX32 > lsi_bad_phase_interrupt(void) "Phase mismatch interrupt" >
On Tue, Sep 15, 2020 at 07:46:39PM +0200, Philippe Mathieu-Daudé wrote: > Make some lsi53c895a trace events more understandable. > > Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> > --- > hw/scsi/trace-events | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) Reviewed-by: Daniel P. Berrangé <berrange@redhat.com> Regards, Daniel
Le 15/09/2020 à 19:46, Philippe Mathieu-Daudé a écrit : > Make some lsi53c895a trace events more understandable. > > Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> > --- > hw/scsi/trace-events | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/hw/scsi/trace-events b/hw/scsi/trace-events > index 9a4a60ca635..ac4209f361e 100644 > --- a/hw/scsi/trace-events > +++ b/hw/scsi/trace-events > @@ -234,9 +234,9 @@ spapr_vscsi_do_crq(unsigned c0, unsigned c1) "crq: %02x %02x ..." > > # lsi53c895a.c > lsi_reset(void) "Reset" > -lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 0x%02x0x%02x" > +lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 1:0x%02x 0:0x%02x" > lsi_update_irq_disconnected(void) "Handled IRQs & disconnected, looking for pending processes" > -lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt 0x%02x0x%02x prev 0x%02x0x%02x" > +lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt stat 1:0x%02x 0:0x%02x sist 1:0x%02x 0:0x%02x" > lsi_script_dma_interrupt(uint8_t stat, uint8_t dstat) "DMA Interrupt 0x%x prev 0x%x" > lsi_bad_phase_jump(uint32_t dsp) "Data phase mismatch jump to 0x%"PRIX32 > lsi_bad_phase_interrupt(void) "Phase mismatch interrupt" > Applied to my trivial-patches branch. Thanks, Laurent
On 15/09/20 19:46, Philippe Mathieu-Daudé wrote: > Make some lsi53c895a trace events more understandable. > > Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> > --- > hw/scsi/trace-events | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/hw/scsi/trace-events b/hw/scsi/trace-events > index 9a4a60ca635..ac4209f361e 100644 > --- a/hw/scsi/trace-events > +++ b/hw/scsi/trace-events > @@ -234,9 +234,9 @@ spapr_vscsi_do_crq(unsigned c0, unsigned c1) "crq: %02x %02x ..." > > # lsi53c895a.c > lsi_reset(void) "Reset" > -lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 0x%02x0x%02x" > +lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 1:0x%02x 0:0x%02x" > lsi_update_irq_disconnected(void) "Handled IRQs & disconnected, looking for pending processes" > -lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt 0x%02x0x%02x prev 0x%02x0x%02x" > +lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt stat 1:0x%02x 0:0x%02x sist 1:0x%02x 0:0x%02x" > lsi_script_dma_interrupt(uint8_t stat, uint8_t dstat) "DMA Interrupt 0x%x prev 0x%x" > lsi_bad_phase_jump(uint32_t dsp) "Data phase mismatch jump to 0x%"PRIX32 > lsi_bad_phase_interrupt(void) "Phase mismatch interrupt" > The idea of the patch is that SIST0 and SIST1 are usually read in pairs. So it's a bit weird in the tracepoint format but it's actually more understandable this way when you see it printed. Laurent, can you please unqueue this? Paolo
Le 25/09/2020 à 22:12, Paolo Bonzini a écrit : > On 15/09/20 19:46, Philippe Mathieu-Daudé wrote: >> Make some lsi53c895a trace events more understandable. >> >> Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> >> --- >> hw/scsi/trace-events | 4 ++-- >> 1 file changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/hw/scsi/trace-events b/hw/scsi/trace-events >> index 9a4a60ca635..ac4209f361e 100644 >> --- a/hw/scsi/trace-events >> +++ b/hw/scsi/trace-events >> @@ -234,9 +234,9 @@ spapr_vscsi_do_crq(unsigned c0, unsigned c1) "crq: %02x %02x ..." >> >> # lsi53c895a.c >> lsi_reset(void) "Reset" >> -lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 0x%02x0x%02x" >> +lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 1:0x%02x 0:0x%02x" >> lsi_update_irq_disconnected(void) "Handled IRQs & disconnected, looking for pending processes" >> -lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt 0x%02x0x%02x prev 0x%02x0x%02x" >> +lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt stat 1:0x%02x 0:0x%02x sist 1:0x%02x 0:0x%02x" >> lsi_script_dma_interrupt(uint8_t stat, uint8_t dstat) "DMA Interrupt 0x%x prev 0x%x" >> lsi_bad_phase_jump(uint32_t dsp) "Data phase mismatch jump to 0x%"PRIX32 >> lsi_bad_phase_interrupt(void) "Phase mismatch interrupt" >> > > The idea of the patch is that SIST0 and SIST1 are usually read in pairs. > So it's a bit weird in the tracepoint format but it's actually more > understandable this way when you see it printed. > > Laurent, can you please unqueue this? No problem. Done. Thanks, Laurent
diff --git a/hw/scsi/trace-events b/hw/scsi/trace-events index 9a4a60ca635..ac4209f361e 100644 --- a/hw/scsi/trace-events +++ b/hw/scsi/trace-events @@ -234,9 +234,9 @@ spapr_vscsi_do_crq(unsigned c0, unsigned c1) "crq: %02x %02x ..." # lsi53c895a.c lsi_reset(void) "Reset" -lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 0x%02x0x%02x" +lsi_update_irq(int level, uint8_t dstat, uint8_t sist1, uint8_t sist0) "Update IRQ level %d dstat 0x%02x sist 1:0x%02x 0:0x%02x" lsi_update_irq_disconnected(void) "Handled IRQs & disconnected, looking for pending processes" -lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt 0x%02x0x%02x prev 0x%02x0x%02x" +lsi_script_scsi_interrupt(uint8_t stat1, uint8_t stat0, uint8_t sist1, uint8_t sist0) "SCSI Interrupt stat 1:0x%02x 0:0x%02x sist 1:0x%02x 0:0x%02x" lsi_script_dma_interrupt(uint8_t stat, uint8_t dstat) "DMA Interrupt 0x%x prev 0x%x" lsi_bad_phase_jump(uint32_t dsp) "Data phase mismatch jump to 0x%"PRIX32 lsi_bad_phase_interrupt(void) "Phase mismatch interrupt"
Make some lsi53c895a trace events more understandable. Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com> --- hw/scsi/trace-events | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)