arm_mpam: Apply T241-MPAM-6 to 63-bit counters
PATCH 1/1 arm_mpam: Apply T241-MPAM-6 to 63-bit counters
T241-MPAM-6 causes all MBWU counter formats to count 64-byte requests instead of bytes. Commit dc48eb1ff27c excluded the 63-bit MSMON_MBWU_LWD format while scaling the shorter counters. Systems selecting the preferred 63-bit counter consequently report bandwidth values that are 64 times too small.
Apply the scale to both the sampled value and overflow correction for the 63-bit format. Unsigned arithmetic retains modulo-u64 behavior when the scaled counter range exceeds u64.
--- drivers/resctrl/mpam_devices.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)
drivers/resctrl/mpam_devices.c +2 -4
@@ -1196,8 +1196,7 @@ static u64 mpam_msmon_overflow_val(enum mpam_device_features type, {u64 overflow_val =__mpam_msmon_overflow_val (type); -if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc) && - type != mpam_feat_msmon_mbwu_63counter) +if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc)) overflow_val *=64 ;return overflow_val;
@@ -1293,8 +1292,7 @@ static void __ris_msmon_read(void *arg) now =FIELD_GET (MSMON___VALUE , now); } -if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc) && - m->type != mpam_feat_msmon_mbwu_63counter) +if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc)) now *=64 ;if (nrdy)
Signature · 1 line
2.43.0
Quoted 6 lines · Shanker Donthineni
T241-MPAM-6 causes all MBWU counter formats to count 64-byte requests instead of bytes. Commit
dc48eb1ff27cexcluded the 63-bitMSMON_MBWU_LWDformat while scaling the shorter counters. Systems selecting the preferred 63-bit counter consequently report bandwidth values that are 64 times too small.
Quoted 7 lines
Apply the scale to both the sampled value and overflow correction for the 63-bit format. Unsigned arithmetic retains modulo-u64 behavior when the scaled counter range exceeds u64.
Fixes dc48eb1ff27c ("arm_mpam: Add workaround for T241-MPAM-6")Signed-off-by Shanker Donthineni <[email protected]>
This is a hardware quirk fix and should be in stable.
Thanks.
-Fenghua
Quoted 28 lines
--- drivers/resctrl/mpam_devices.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)drivers/resctrl/mpam_devices.c +2
-4@@ -1196,8 +1196,7 @@ static u64 mpam_msmon_overflow_val(enum mpam_device_features type, {u64 overflow_val =__mpam_msmon_overflow_val (type); -if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc) && - type != mpam_feat_msmon_mbwu_63counter) +if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc)) overflow_val *=64 ;return overflow_val;@@ -1293,8 +1292,7 @@ static void __ris_msmon_read(void *arg) now =FIELD_GET (MSMON___VALUE , now); } -if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc) && - m->type != mpam_feat_msmon_mbwu_63counter) +if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc)) now *=64 ;if (nrdy)
Ben Horgan
maintainerHi Shanker,
Quoted 2 lines · Shanker Donthineni
T241-MPAM-6 causes all MBWU counter formats to count 64-byte requests instead of bytes. Commit
dc48eb1ff27cexcluded the 63-bit
As described in that commit message, T241-MPAM-6 does not affect LWD counters. Is there another canonical description of T241-MPAM-6. Otherwise, this just looks like you've discovered a new h/w bug :)
Quoted 3 lines
MSMON_MBWU_LWDformat while scaling the shorter counters. Systems selecting the preferred 63-bit counter consequently report bandwidth values that are 64 times too small.
So, the MSC with IIDR MPAM_IIDR_NVIDIA_T241 has different sizes of long bandwidth counters between RIS. This seems a bit unusual. Please could you share which RIS have 44 bit long counters and which have 63 bits. (MPAMF_MBWUMON_IDR.LWD gives this)
Thanks,
Ben
Quoted 36 lines
Apply the scale to both the sampled value and overflow correction for the 63-bit format. Unsigned arithmetic retains modulo-u64 behavior when the scaled counter range exceeds u64.
Fixes dc48eb1ff27c ("arm_mpam: Add workaround for T241-MPAM-6")Signed-off-by Shanker Donthineni <[email protected]>--- drivers/resctrl/mpam_devices.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)drivers/resctrl/mpam_devices.c +2
-4@@ -1196,8 +1196,7 @@ static u64 mpam_msmon_overflow_val(enum mpam_device_features type, {u64 overflow_val =__mpam_msmon_overflow_val (type); -if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc) && - type != mpam_feat_msmon_mbwu_63counter) +if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc)) overflow_val *=64 ;return overflow_val;@@ -1293,8 +1292,7 @@ static void __ris_msmon_read(void *arg) now =FIELD_GET (MSMON___VALUE , now); } -if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc) && - m->type != mpam_feat_msmon_mbwu_63counter) +if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc)) now *=64 ;if (nrdy)
Hi Ben,
Quoted 11 lines · Ben Horgan
External email: Use caution opening links or attachments
Hi Shanker,
Quoted 2 lines · Shanker Donthineni
T241-MPAM-6 causes all MBWU counter formats to count 64-byte requests instead of bytes. Commit
dc48eb1ff27cexcluded the 63-bitAs described in that commit message, T241-MPAM-6 does not affect LWD counters. Is there another canonical description of T241-MPAM-6. Otherwise, this just looks like you've discovered a new h/w bug :)
The original commit message for dc48eb1ff27c explicitly states that T241-MPAM-6 affects both MSMON_MBWU and MSMON_MBWU_L:
"The registers MSMON_MBWU_L and MSMON_MBWU return the number of requests rather than the number of bytes transferred."
LWD (when HAS_LONG=1) selects the width/format of MSMON_MBWU_L; it does not identify a separate register outside the scope described above.
Quoted 6 lines
Quoted 3 lines
MSMON_MBWU_LWDformat while scaling the shorter counters. Systems selecting the preferred 63-bit counter consequently report bandwidth values that are 64 times too small.So, the MSC with IIDR
MPAM_IIDR_NVIDIA_T241has different sizes of long bandwidth counters between RIS. This seems a bit unusual. Please could you share which RIS have 44 bit long counters and which have 63 bits. (MPAMF_MBWUMON_IDR.LWD gives this)
T241 does not implement RIS (MPAMF_IDR.HAS_RIS=0). Its L3 MSC implements a 63-bit long bandwidth counter, as indicated by HAS_LONG=1 and LWD=1; therefore, there are no per-RIS differences in counter width.
The mpam_feat_msmon_mbwu_63counter exclusion in dc48eb1ff27c means the workaround is not applied to the long-counter format implemented by the T241 L3 MSC. This patch removes that exclusion and makes the implementation consistent with the original erratum description and commit message.
Quoted 39 lines
Thanks,
Ben
Quoted 35 lines
Apply the scale to both the sampled value and overflow correction for the 63-bit format. Unsigned arithmetic retains modulo-u64 behavior when the scaled counter range exceeds u64.
Fixes dc48eb1ff27c ("arm_mpam: Add workaround for T241-MPAM-6")Signed-off-by Shanker Donthineni <[email protected]>--- drivers/resctrl/mpam_devices.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-)drivers/resctrl/mpam_devices.c +2
-4@@ -1196,8 +1196,7 @@ static u64 mpam_msmon_overflow_val(enum mpam_device_features type, {u64 overflow_val =__mpam_msmon_overflow_val (type); -if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc) && - type != mpam_feat_msmon_mbwu_63counter) +if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc)) overflow_val *=64 ;return overflow_val;@@ -1293,8 +1292,7 @@ static void __ris_msmon_read(void *arg) now =FIELD_GET (MSMON___VALUE , now); } -if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc) && - m->type != mpam_feat_msmon_mbwu_63counter) +if (mpam_has_quirk (T241_MBW_COUNTER_SCALE_64 , msc)) now *=64 ;if (nrdy)
Hi Shanker,
Quoted 14 lines · Shanker Donthineni
Hi Ben,
Quoted 11 lines · Ben Horgan
External email: Use caution opening links or attachments
Hi Shanker,
Quoted 2 lines · Shanker Donthineni
T241-MPAM-6 causes all MBWU counter formats to count 64-byte requests instead of bytes. Commit
dc48eb1ff27cexcluded the 63-bitAs described in that commit message, T241-MPAM-6 does not affect LWD counters. Is there another canonical description of T241-MPAM-6. Otherwise, this just looks like you've discovered a new h/w bug :)
Quoted 8 lines
The original commit message for
dc48eb1ff27cexplicitly states that T241-MPAM-6 affects bothMSMON_MBWUandMSMON_MBWU_L:"The registers
MSMON_MBWU_LandMSMON_MBWUreturn the number of requests rather than the number of bytes transferred."LWD (when HAS_LONG=1) selects the width/format of
MSMON_MBWU_L; it does not identify a separate register outside the scope described above.
Oh yes, you are right. The L_VALUE/LWD_VALUE naming is just internal to the driver.
Quoted 10 lines
Quoted 7 lines
Quoted 4 lines
MSMON_MBWU_LWDformat while scaling the shorter counters. Systems selecting the preferred 63-bit counter consequently report bandwidth values that are 64 times too small.So, the MSC with IIDR
MPAM_IIDR_NVIDIA_T241has different sizes of long bandwidth counters between RIS. This seems a bit unusual. Please could you share which RIS have 44 bit long counters and which have 63 bits. (MPAMF_MBWUMON_IDR.LWD gives this)T241 does not implement RIS (MPAMF_IDR.HAS_RIS=0). Its L3 MSC implements a 63-bit long bandwidth counter, as indicated by HAS_LONG=1 and LWD=1; therefore, there are no per-RIS differences in counter width.
Ok, so there are no 44 bit counters?
Thanks,
Ben
Quoted 45 lines
The
mpam_feat_msmon_mbwu_63counterexclusion indc48eb1ff27cmeans the workaround is not applied to the long-counter format implemented by the T241 L3 MSC. This patch removes that exclusion and makes the implementation consistent with the original erratum description and commit message.Quoted 39 lines
Thanks,
Ben
Quoted 35 lines
Apply the scale to both the sampled value and overflow correction for the 63-bit format. Unsigned arithmetic retains modulo-u64 behavior when the scaled counter range exceeds u64.
Fixes dc48eb1ff27c ("arm_mpam: Add workaround for T241-MPAM-6")Signed-off-by Shanker Donthineni <[email protected]>---
drivers/resctrl/mpam_devices.c| 6 ++----1 file changed, 2 insertions(+), 4 deletions(-)drivers/resctrl/mpam_devices.c +2
-4@@ -1196,8 +1196,7 @@ static u64 mpam_msmon_overflow_val(enum mpam_device_features type, { u64 overflow_val = __mpam_msmon_overflow_val(type); - if (mpam_has_quirk(T241_MBW_COUNTER_SCALE_64, msc) && - type != mpam_feat_msmon_mbwu_63counter) + if (mpam_has_quirk(T241_MBW_COUNTER_SCALE_64, msc)) overflow_val *= 64; return overflow_val;@@ -1293,8 +1292,7 @@ static void __ris_msmon_read(void *arg) now = FIELD_GET(MSMON___VALUE, now); } - if (mpam_has_quirk(T241_MBW_COUNTER_SCALE_64, msc) && - m->type != mpam_feat_msmon_mbwu_63counter) + if (mpam_has_quirk(T241_MBW_COUNTER_SCALE_64, msc)) now *= 64; if (nrdy)