Timeline for Intel CPU bug in the 90s
Current License: CC BY-SA 4.0
34 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Nov 2 at 14:28 | comment | added | cup | It is very rare that a bank would use sine or cosine - that is more for engineering companies. Banks would normally use square roots, and the 4 math functions. and maybe a bit of bit-shifting for the cryptology. The arithmetic that monetary applications use don't get more complicated than that but they do require high precision. | |
| Nov 2 at 9:57 | answer | added | Ofek Shilon | timeline score: 2 | |
| S Feb 18, 2023 at 21:46 | history | suggested | Sep Roland | CC BY-SA 4.0 |
corrected grammar and spelling ("did not find" vs "haven't found")
|
| Feb 18, 2023 at 19:18 | review | Suggested edits | |||
| S Feb 18, 2023 at 21:46 | |||||
| Oct 15, 2020 at 17:26 | comment | added | supercat | ...but rounding would bump the value up to 1000.0, which actually has four digits to the left of the decimal). | |
| Oct 15, 2020 at 17:25 | comment | added | supercat |
@DougWarren: What's the difference between Windows 3.1 and Windows 3.11? The calculator knows. Go to the Windows Calculator and type 3.11-3.1=. On Windows 3.10, the calculator will display "0.00", On Windows 3.11, the calculator will display "0.01". This issue is somewhat similar to a bug that was in in the printf for Turbo C 2.0 but fixed in 2.1, where printf("%1.1f", 999.96); would output 000.0 [it would determine that the value was at least 100 but less than 1,000 and thus needed three digits to the left of the decimal...
|
|
| S Oct 15, 2020 at 16:17 | history | suggested | Andreas Rejbrand | CC BY-SA 4.0 |
improved grammar and typography
|
| Oct 15, 2020 at 15:32 | review | Suggested edits | |||
| S Oct 15, 2020 at 16:17 | |||||
| Oct 14, 2020 at 17:36 | comment | added | Mooing Duck | More recently, AMD Ryzen 3000 had an issue where random number generation would always return -1. arstechnica.com/gadgets/2019/10/… | |
| Oct 14, 2020 at 13:22 | comment | added | Doug Warren | A joke I heard at the time was: "At Intel, quality is job 0.999999999." | |
| Oct 14, 2020 at 3:26 | comment | added | Peter Cordes | So basically Skylake now (with those microcode mitigations) has some performance glass-jaws that can limit throughput for small loops, which didn't exist in previous CPUs like Haswell. Ice Lake fixes some of that, (e.g. re-enable the loop buffer), but unfortunately many of the skylake-derived uarches like Coffee Lake don't. | |
| Oct 14, 2020 at 3:13 | comment | added | Peter Cordes | @forest: and also disabled the hardware lock elision part of TSX, and sometimes RTM. That might partly be due to a new class of vulnerabilities being discovered (Spectre and then MDS) that weren't even anticipated when Skylake was being validated. But regardless, TSX has been a real Charlie Brown football situation: present in Haswell, disabled due to bugs. Present in Broadwell, disabled due to a different bug I think? Present in Skylake, then disabled due to security(?) bugs. | |
| Oct 14, 2020 at 3:08 | comment | added | Peter Cordes |
@forest: More recently, former Intel principal engineer, François Piednoël has said Skylake was more buggy than most previous designs. That matches up with the 2014 timeline that those anonymous sources report for Intel making changes to reduce QA / validation time: Skylake launched in mid 2015. It has bugs that required microcode updates to disable the loop buffer LSD, promote mfence to serializing OoO exec, disable the uop cache for JCC at a 32-byte boundary
|
|
| S Oct 14, 2020 at 2:41 | history | edited | Peter Cordes |
fixed grammar of last sentence
|
|
| S Oct 14, 2020 at 2:41 | history | suggested | CommunityBot | CC BY-SA 4.0 |
fixed grammar of last sentence
|
| Oct 13, 2020 at 20:49 | review | Suggested edits | |||
| S Oct 14, 2020 at 2:41 | |||||
| Oct 13, 2020 at 16:19 | answer | added | supercat | timeline score: 26 | |
| Oct 13, 2020 at 15:44 | vote | accept | gomd | ||
| Oct 13, 2020 at 15:33 | comment | added | reirab | @user3528438 So true about the errata. These can be really annoying when you run into them. A digital signal processor that we used on a project back over a decade ago - a product which we still manufacture to this day - had a couple of rather unfortunate silicon anomalies. One of them sometimes resulted in corruption of the DMA control registers when using the USB controller's DMA mode 0. The only listed workaround was to use mode 1. The other resulted in occasional corruption of the DMA control registers when using mode 1... with a listed workaround of using mode 0... | |
| Oct 13, 2020 at 11:32 | history | edited | Raffzahn | CC BY-SA 4.0 |
deleted 8 characters in body
|
| Oct 13, 2020 at 10:56 | answer | added | Bagnus | timeline score: 32 | |
| Oct 13, 2020 at 2:53 | comment | added | forest | Unfortunately, Intel has recently gotten rid of its validation team: danluu.com/cpu-bugs/#update | |
| Oct 12, 2020 at 22:30 | history | became hot network question | |||
| Oct 12, 2020 at 21:43 | comment | added | user3528438 | Also every CPU has a tremendous bugs historically and now, google for "xxxx processor errata", e.g. ARM A77 errata is 59 pages: developer.arm.com/documentation/101992/0009 | |
| Oct 12, 2020 at 21:38 | comment | added | user3528438 | FYI fsin is a different bug that came much later, caused hard coded pi not large enough to handle very large inputs: cpushack.com/2014/10/15/… | |
| Oct 12, 2020 at 20:08 | comment | added | Tommy |
@mannaggia off the top of my head, since markets react to numbers they're in part big feedback systems; wherever there is feedback there is a chance for harmonic motion. So a classic case of if f''(x) = -f(x) then f(x) = cos(x)? I'm not a quant, I'm speculating wildly.
|
|
| Oct 12, 2020 at 18:04 | comment | added | mannaggia | There were also bugs in the 80386 with 32-bit integer multiplication in the A1 and A2 stepping, but that was in the mid 80's. Not sure how sin/cos would be used in banking... | |
| Oct 12, 2020 at 16:55 | comment | added | Euro Micelli | Such bugs do happen (see infamous FDIV bug mentioned by others), but this particular story appears to be a bit distorted. It's hard to imagine what use could a bank have for trigonometric functions, and the values of trigonometric functions have been tabulated for decades if not centuries to high precision - any discrepancy could be easily verified without hiring someone to examine algorithms. The FDIV bug is not a close match because Intel was indeed wrong on that one (and the error was not algorithmic but a missing column in a lookup table). Looking forward to a closer match. | |
| Oct 12, 2020 at 15:30 | comment | added | peterh | There was a problem with the floating point multiplication. I did not hear about a sin/cos bug. | |
| Oct 12, 2020 at 14:35 | answer | added | Stephen Kitt | timeline score: 91 | |
| Oct 12, 2020 at 14:34 | answer | added | dave | timeline score: 4 | |
| Oct 12, 2020 at 14:30 | review | First posts | |||
| Oct 12, 2020 at 15:30 | |||||
| Oct 12, 2020 at 14:30 | comment | added | dave | It's true. I got a CPU change out of it. I'll see if I can find a reference. | |
| Oct 12, 2020 at 14:26 | history | asked | gomd | CC BY-SA 4.0 |