2
0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-21 11:44:01 +08:00

IB/qib: Defer HCA error events to tasklet

With ib_qib options:

    options ib_qib krcvqs=1 pcie_caps=0x51 rcvhdrcnt=4096 singleport=1 ibmtu=4

a run of ib_write_bw -a yields the following:

    ------------------------------------------------------------------
     #bytes     #iterations    BW peak[MB/sec]    BW average[MB/sec]
     1048576   5000           2910.64            229.80
    ------------------------------------------------------------------

The top cpu use in a profile is:

    CPU: Intel Architectural Perfmon, speed 2400.15 MHz (estimated)
    Counted CPU_CLK_UNHALTED events (Clock cycles when not halted) with a unit mask
    of 0x00 (No unit mask) count 1002300
    Counted LLC_MISSES events (Last level cache demand requests from this core that
    missed the LLC) with a unit mask of 0x41 (No unit mask) count 10000
    samples  %        samples  %        app name                 symbol name
    15237    29.2642  964      17.1195  ib_qib.ko                qib_7322intr
    12320    23.6618  1040     18.4692  ib_qib.ko                handle_7322_errors
    4106      7.8860  0              0  vmlinux                  vsnprintf


Analysis of the stats, profile, the code, and the annotated profile indicate:
 - All of the overflow interrupts (one per packet overflow) are
   serviced on CPU0 with no mitigation on the frequency.
 - All of the receive interrupts are being serviced by CPU0.  (That is
   the way truescale.cmds statically allocates the kctx IRQs to CPU)
 - The code is spending all of its time servicing QIB_I_C_ERROR
   RcvEgrFullErr interrupts on CPU0, starving the packet receive
   processing.
 - The decode_err routine is very inefficient, using a printf variant
   to format a "%s" and continues to loop when the errs mask has been
   cleared.
 - Both qib_7322intr and handle_7322_errors read pci registers, which
   is very inefficient.

The fix does the following:
 - Adds a tasklet to service QIB_I_C_ERROR
 - Replaces the very inefficient scnprintf() with a memcpy().  A field
   is added to qib_hwerror_msgs to save the sizeof("string") at
   compile time so that a strlen is not needed during err_decode().
 - The most frequent errors (Overflows) are serviced first to exit the
   loop as early as possible.
 - The loop now exits as soon as the errs mask is clear rather than
   fruitlessly looping through the msp array.

With this fix the performance changes to:

    ------------------------------------------------------------------
     #bytes     #iterations    BW peak[MB/sec]    BW average[MB/sec]
     1048576   5000           2990.64            2941.35
    ------------------------------------------------------------------

During testing of the error handling overflow patch, it was determined
that some CPU's were slower when servicing both overflow and receive
interrupts on CPU0 with different MSI interrupt vectors.

This patch adds an option (krcvq01_no_msi) to not use a dedicated MSI
interrupt for kctx's < 2 and to service them on the default interrupt.
For some CPUs, the cost of the interrupt enter/exit is more costly
than then the additional PCI read in the default handler.

Signed-off-by: Mike Marciniszyn <mike.marciniszyn@qlogic.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This commit is contained in:
Mike Marciniszyn 2011-07-21 13:21:16 +00:00 committed by Roland Dreier
parent e800bd032c
commit e67306a380
2 changed files with 53 additions and 21 deletions

View File

@ -1012,6 +1012,8 @@ struct qib_devdata {
u8 psxmitwait_supported; u8 psxmitwait_supported;
/* cycle length of PS* counters in HW (in picoseconds) */ /* cycle length of PS* counters in HW (in picoseconds) */
u16 psxmitwait_check_rate; u16 psxmitwait_check_rate;
/* high volume overflow errors defered to tasklet */
struct tasklet_struct error_tasklet;
}; };
/* hol_state values */ /* hol_state values */
@ -1433,6 +1435,7 @@ extern struct mutex qib_mutex;
struct qib_hwerror_msgs { struct qib_hwerror_msgs {
u64 mask; u64 mask;
const char *msg; const char *msg;
size_t sz;
}; };
#define QLOGIC_IB_HWE_MSG(a, b) { .mask = a, .msg = b } #define QLOGIC_IB_HWE_MSG(a, b) { .mask = a, .msg = b }

View File

@ -114,6 +114,10 @@ static ushort qib_singleport;
module_param_named(singleport, qib_singleport, ushort, S_IRUGO); module_param_named(singleport, qib_singleport, ushort, S_IRUGO);
MODULE_PARM_DESC(singleport, "Use only IB port 1; more per-port buffer space"); MODULE_PARM_DESC(singleport, "Use only IB port 1; more per-port buffer space");
static ushort qib_krcvq01_no_msi;
module_param_named(krcvq01_no_msi, qib_krcvq01_no_msi, ushort, S_IRUGO);
MODULE_PARM_DESC(krcvq01_no_msi, "No MSI for kctx < 2");
/* /*
* Receive header queue sizes * Receive header queue sizes
*/ */
@ -1106,9 +1110,9 @@ static inline u32 read_7322_creg32_port(const struct qib_pportdata *ppd,
#define AUTONEG_TRIES 3 /* sequential retries to negotiate DDR */ #define AUTONEG_TRIES 3 /* sequential retries to negotiate DDR */
#define HWE_AUTO(fldname) { .mask = SYM_MASK(HwErrMask, fldname##Mask), \ #define HWE_AUTO(fldname) { .mask = SYM_MASK(HwErrMask, fldname##Mask), \
.msg = #fldname } .msg = #fldname , .sz = sizeof(#fldname) }
#define HWE_AUTO_P(fldname, port) { .mask = SYM_MASK(HwErrMask, \ #define HWE_AUTO_P(fldname, port) { .mask = SYM_MASK(HwErrMask, \
fldname##Mask##_##port), .msg = #fldname } fldname##Mask##_##port), .msg = #fldname , .sz = sizeof(#fldname) }
static const struct qib_hwerror_msgs qib_7322_hwerror_msgs[] = { static const struct qib_hwerror_msgs qib_7322_hwerror_msgs[] = {
HWE_AUTO_P(IBSerdesPClkNotDetect, 1), HWE_AUTO_P(IBSerdesPClkNotDetect, 1),
HWE_AUTO_P(IBSerdesPClkNotDetect, 0), HWE_AUTO_P(IBSerdesPClkNotDetect, 0),
@ -1126,14 +1130,16 @@ static const struct qib_hwerror_msgs qib_7322_hwerror_msgs[] = {
HWE_AUTO_P(IBCBusFromSPCParityErr, 0), HWE_AUTO_P(IBCBusFromSPCParityErr, 0),
HWE_AUTO(statusValidNoEop), HWE_AUTO(statusValidNoEop),
HWE_AUTO(LATriggered), HWE_AUTO(LATriggered),
{ .mask = 0 } { .mask = 0, .sz = 0 }
}; };
#define E_AUTO(fldname) { .mask = SYM_MASK(ErrMask, fldname##Mask), \ #define E_AUTO(fldname) { .mask = SYM_MASK(ErrMask, fldname##Mask), \
.msg = #fldname } .msg = #fldname, .sz = sizeof(#fldname) }
#define E_P_AUTO(fldname) { .mask = SYM_MASK(ErrMask_0, fldname##Mask), \ #define E_P_AUTO(fldname) { .mask = SYM_MASK(ErrMask_0, fldname##Mask), \
.msg = #fldname } .msg = #fldname, .sz = sizeof(#fldname) }
static const struct qib_hwerror_msgs qib_7322error_msgs[] = { static const struct qib_hwerror_msgs qib_7322error_msgs[] = {
E_AUTO(RcvEgrFullErr),
E_AUTO(RcvHdrFullErr),
E_AUTO(ResetNegated), E_AUTO(ResetNegated),
E_AUTO(HardwareErr), E_AUTO(HardwareErr),
E_AUTO(InvalidAddrErr), E_AUTO(InvalidAddrErr),
@ -1146,9 +1152,7 @@ static const struct qib_hwerror_msgs qib_7322error_msgs[] = {
E_AUTO(SendSpecialTriggerErr), E_AUTO(SendSpecialTriggerErr),
E_AUTO(SDmaWrongPortErr), E_AUTO(SDmaWrongPortErr),
E_AUTO(SDmaBufMaskDuplicateErr), E_AUTO(SDmaBufMaskDuplicateErr),
E_AUTO(RcvHdrFullErr), { .mask = 0, .sz = 0 }
E_AUTO(RcvEgrFullErr),
{ .mask = 0 }
}; };
static const struct qib_hwerror_msgs qib_7322p_error_msgs[] = { static const struct qib_hwerror_msgs qib_7322p_error_msgs[] = {
@ -1158,7 +1162,8 @@ static const struct qib_hwerror_msgs qib_7322p_error_msgs[] = {
/* /*
* SDmaHaltErr is not really an error, make it clearer; * SDmaHaltErr is not really an error, make it clearer;
*/ */
{.mask = SYM_MASK(ErrMask_0, SDmaHaltErrMask), .msg = "SDmaHalted"}, {.mask = SYM_MASK(ErrMask_0, SDmaHaltErrMask), .msg = "SDmaHalted",
.sz = 11},
E_P_AUTO(SDmaDescAddrMisalignErr), E_P_AUTO(SDmaDescAddrMisalignErr),
E_P_AUTO(SDmaUnexpDataErr), E_P_AUTO(SDmaUnexpDataErr),
E_P_AUTO(SDmaMissingDwErr), E_P_AUTO(SDmaMissingDwErr),
@ -1194,7 +1199,7 @@ static const struct qib_hwerror_msgs qib_7322p_error_msgs[] = {
E_P_AUTO(RcvICRCErr), E_P_AUTO(RcvICRCErr),
E_P_AUTO(RcvVCRCErr), E_P_AUTO(RcvVCRCErr),
E_P_AUTO(RcvFormatErr), E_P_AUTO(RcvFormatErr),
{ .mask = 0 } { .mask = 0, .sz = 0 }
}; };
/* /*
@ -1202,17 +1207,17 @@ static const struct qib_hwerror_msgs qib_7322p_error_msgs[] = {
* context * context
*/ */
#define INTR_AUTO(fldname) { .mask = SYM_MASK(IntMask, fldname##Mask), \ #define INTR_AUTO(fldname) { .mask = SYM_MASK(IntMask, fldname##Mask), \
.msg = #fldname } .msg = #fldname, .sz = sizeof(#fldname) }
/* Below generates "auto-message" for interrupts specific to a port */ /* Below generates "auto-message" for interrupts specific to a port */
#define INTR_AUTO_P(fldname) { .mask = MASK_ACROSS(\ #define INTR_AUTO_P(fldname) { .mask = MASK_ACROSS(\
SYM_LSB(IntMask, fldname##Mask##_0), \ SYM_LSB(IntMask, fldname##Mask##_0), \
SYM_LSB(IntMask, fldname##Mask##_1)), \ SYM_LSB(IntMask, fldname##Mask##_1)), \
.msg = #fldname "_P" } .msg = #fldname "_P", .sz = sizeof(#fldname "_P") }
/* For some reason, the SerDesTrimDone bits are reversed */ /* For some reason, the SerDesTrimDone bits are reversed */
#define INTR_AUTO_PI(fldname) { .mask = MASK_ACROSS(\ #define INTR_AUTO_PI(fldname) { .mask = MASK_ACROSS(\
SYM_LSB(IntMask, fldname##Mask##_1), \ SYM_LSB(IntMask, fldname##Mask##_1), \
SYM_LSB(IntMask, fldname##Mask##_0)), \ SYM_LSB(IntMask, fldname##Mask##_0)), \
.msg = #fldname "_P" } .msg = #fldname "_P", .sz = sizeof(#fldname "_P") }
/* /*
* Below generates "auto-message" for interrupts specific to a context, * Below generates "auto-message" for interrupts specific to a context,
* with ctxt-number appended * with ctxt-number appended
@ -1220,7 +1225,7 @@ static const struct qib_hwerror_msgs qib_7322p_error_msgs[] = {
#define INTR_AUTO_C(fldname) { .mask = MASK_ACROSS(\ #define INTR_AUTO_C(fldname) { .mask = MASK_ACROSS(\
SYM_LSB(IntMask, fldname##0IntMask), \ SYM_LSB(IntMask, fldname##0IntMask), \
SYM_LSB(IntMask, fldname##17IntMask)), \ SYM_LSB(IntMask, fldname##17IntMask)), \
.msg = #fldname "_C"} .msg = #fldname "_C", .sz = sizeof(#fldname "_C") }
static const struct qib_hwerror_msgs qib_7322_intr_msgs[] = { static const struct qib_hwerror_msgs qib_7322_intr_msgs[] = {
INTR_AUTO_P(SDmaInt), INTR_AUTO_P(SDmaInt),
@ -1234,11 +1239,12 @@ static const struct qib_hwerror_msgs qib_7322_intr_msgs[] = {
INTR_AUTO_P(SendDoneInt), INTR_AUTO_P(SendDoneInt),
INTR_AUTO(SendBufAvailInt), INTR_AUTO(SendBufAvailInt),
INTR_AUTO_C(RcvAvail), INTR_AUTO_C(RcvAvail),
{ .mask = 0 } { .mask = 0, .sz = 0 }
}; };
#define TXSYMPTOM_AUTO_P(fldname) \ #define TXSYMPTOM_AUTO_P(fldname) \
{ .mask = SYM_MASK(SendHdrErrSymptom_0, fldname), .msg = #fldname } { .mask = SYM_MASK(SendHdrErrSymptom_0, fldname), \
.msg = #fldname, .sz = sizeof(#fldname) }
static const struct qib_hwerror_msgs hdrchk_msgs[] = { static const struct qib_hwerror_msgs hdrchk_msgs[] = {
TXSYMPTOM_AUTO_P(NonKeyPacket), TXSYMPTOM_AUTO_P(NonKeyPacket),
TXSYMPTOM_AUTO_P(GRHFail), TXSYMPTOM_AUTO_P(GRHFail),
@ -1247,7 +1253,7 @@ static const struct qib_hwerror_msgs hdrchk_msgs[] = {
TXSYMPTOM_AUTO_P(SLIDFail), TXSYMPTOM_AUTO_P(SLIDFail),
TXSYMPTOM_AUTO_P(RawIPV6), TXSYMPTOM_AUTO_P(RawIPV6),
TXSYMPTOM_AUTO_P(PacketTooSmall), TXSYMPTOM_AUTO_P(PacketTooSmall),
{ .mask = 0 } { .mask = 0, .sz = 0 }
}; };
#define IBA7322_HDRHEAD_PKTINT_SHIFT 32 /* interrupt cnt in upper 32 bits */ #define IBA7322_HDRHEAD_PKTINT_SHIFT 32 /* interrupt cnt in upper 32 bits */
@ -1292,7 +1298,7 @@ static void err_decode(char *msg, size_t len, u64 errs,
u64 these, lmask; u64 these, lmask;
int took, multi, n = 0; int took, multi, n = 0;
while (msp && msp->mask) { while (errs && msp && msp->mask) {
multi = (msp->mask & (msp->mask - 1)); multi = (msp->mask & (msp->mask - 1));
while (errs & msp->mask) { while (errs & msp->mask) {
these = (errs & msp->mask); these = (errs & msp->mask);
@ -1303,9 +1309,14 @@ static void err_decode(char *msg, size_t len, u64 errs,
*msg++ = ','; *msg++ = ',';
len--; len--;
} }
took = scnprintf(msg, len, "%s", msp->msg); BUG_ON(!msp->sz);
/* msp->sz counts the nul */
took = min_t(size_t, msp->sz - (size_t)1, len);
memcpy(msg, msp->msg, took);
len -= took; len -= took;
msg += took; msg += took;
if (len)
*msg = '\0';
} }
errs &= ~lmask; errs &= ~lmask;
if (len && multi) { if (len && multi) {
@ -1643,6 +1654,14 @@ done:
return; return;
} }
static void qib_error_tasklet(unsigned long data)
{
struct qib_devdata *dd = (struct qib_devdata *)data;
handle_7322_errors(dd);
qib_write_kreg(dd, kr_errmask, dd->cspec->errormask);
}
static void reenable_chase(unsigned long opaque) static void reenable_chase(unsigned long opaque)
{ {
struct qib_pportdata *ppd = (struct qib_pportdata *)opaque; struct qib_pportdata *ppd = (struct qib_pportdata *)opaque;
@ -2724,8 +2743,10 @@ static noinline void unlikely_7322_intr(struct qib_devdata *dd, u64 istat)
unknown_7322_ibits(dd, istat); unknown_7322_ibits(dd, istat);
if (istat & QIB_I_GPIO) if (istat & QIB_I_GPIO)
unknown_7322_gpio_intr(dd); unknown_7322_gpio_intr(dd);
if (istat & QIB_I_C_ERROR) if (istat & QIB_I_C_ERROR) {
handle_7322_errors(dd); qib_write_kreg(dd, kr_errmask, 0ULL);
tasklet_schedule(&dd->error_tasklet);
}
if (istat & INT_MASK_P(Err, 0) && dd->rcd[0]) if (istat & INT_MASK_P(Err, 0) && dd->rcd[0])
handle_7322_p_errors(dd->rcd[0]->ppd); handle_7322_p_errors(dd->rcd[0]->ppd);
if (istat & INT_MASK_P(Err, 1) && dd->rcd[1]) if (istat & INT_MASK_P(Err, 1) && dd->rcd[1])
@ -3124,6 +3145,8 @@ try_intx:
arg = dd->rcd[ctxt]; arg = dd->rcd[ctxt];
if (!arg) if (!arg)
continue; continue;
if (qib_krcvq01_no_msi && ctxt < 2)
continue;
lsb = QIB_I_RCVAVAIL_LSB + ctxt; lsb = QIB_I_RCVAVAIL_LSB + ctxt;
handler = qib_7322pintr; handler = qib_7322pintr;
name = QIB_DRV_NAME " (kctx)"; name = QIB_DRV_NAME " (kctx)";
@ -3158,6 +3181,8 @@ try_intx:
for (i = 0; i < ARRAY_SIZE(redirect); i++) for (i = 0; i < ARRAY_SIZE(redirect); i++)
qib_write_kreg(dd, kr_intredirect + i, redirect[i]); qib_write_kreg(dd, kr_intredirect + i, redirect[i]);
dd->cspec->main_int_mask = mask; dd->cspec->main_int_mask = mask;
tasklet_init(&dd->error_tasklet, qib_error_tasklet,
(unsigned long)dd);
bail:; bail:;
} }
@ -6787,6 +6812,10 @@ struct qib_devdata *qib_init_iba7322_funcs(struct pci_dev *pdev,
(i >= ARRAY_SIZE(irq_table) && (i >= ARRAY_SIZE(irq_table) &&
dd->rcd[i - ARRAY_SIZE(irq_table)])) dd->rcd[i - ARRAY_SIZE(irq_table)]))
actual_cnt++; actual_cnt++;
/* reduce by ctxt's < 2 */
if (qib_krcvq01_no_msi)
actual_cnt -= dd->num_pports;
tabsize = actual_cnt; tabsize = actual_cnt;
dd->cspec->msix_entries = kmalloc(tabsize * dd->cspec->msix_entries = kmalloc(tabsize *
sizeof(struct msix_entry), GFP_KERNEL); sizeof(struct msix_entry), GFP_KERNEL);