text stringlengths 14 502 |
|---|
[910.10 --> 910.64] and Amazon, |
[910.98 --> 912.08] and all these huge companies |
[912.08 --> 914.02] all came together to try to work on this. |
[915.00 --> 917.24] And they've been working on it actually |
[917.24 --> 918.96] for quite a while. |
[920.72 --> 921.34] There's, |
[921.70 --> 923.44] this is where we get into, |
[924.82 --> 926.72] there's a lot of different information |
[926.72 --> 928.92] and it's a little bit confusing |
[928.92 --> 930.48] what happened next, |
[930.48 --> 932.30] but the public figured out. |
[932.72 --> 933.74] And the public figured out |
[933.74 --> 936.22] not that long ago. |
[936.78 --> 940.40] The main article that really blew up |
[940.40 --> 941.32] in a lot of people's minds |
[941.32 --> 942.78] because it was much more public facing, |
[942.90 --> 943.36] I believe, |
[943.48 --> 944.66] was the Register article, |
[945.40 --> 946.28] register.co.uk, |
[946.44 --> 948.54] which is posted in the WAN doc, |
[948.60 --> 949.32] if you want to check it out, |
[949.38 --> 949.88] on the forum. |
[949.88 --> 951.36] That was, |
[951.36 --> 952.32] as far as I know, |
[952.44 --> 953.72] one of the biggest explosions |
[953.72 --> 955.78] in terms of more people knowing. |
[956.26 --> 957.32] There was talk about it |
[957.32 --> 960.38] on more like programmer-focused back channels. |
[960.52 --> 962.74] This was a big news outlet bringing it on. |
[962.96 --> 964.10] I could be wrong about that. |
[964.18 --> 965.18] It could have been someone else first. |
[965.34 --> 966.46] This is the first one I saw. |
[966.50 --> 966.70] Same. |
[966.70 --> 967.66] And I think it's the first one |
[967.66 --> 968.28] most people saw. |
[969.68 --> 971.20] It got to there, |
[971.34 --> 972.98] as far as I know, |
[973.38 --> 976.42] because a programmer at AMD |
[976.42 --> 979.46] pushed code to the Linux kernel. |
[980.10 --> 980.94] Let's see here. |
[981.82 --> 982.16] Ah. |
[983.74 --> 984.26] Yeah. |
[984.34 --> 985.12] So there was an embargo |
[985.12 --> 986.18] on all this kind of stuff |
[986.18 --> 988.18] through the Google Project Zero platform. |
[988.78 --> 989.94] Not everyone agrees |
[989.94 --> 991.20] with having embargoes |
[991.20 --> 992.08] on these types of things. |
[992.20 --> 993.82] Some people position themselves |
[993.82 --> 995.92] with all information should be open. |
[996.28 --> 996.82] Everyone should know |
[996.82 --> 997.84] everything that's going on. |
[999.02 --> 1000.10] And your opinion on that |
[1000.10 --> 1001.26] is whatever, Zs. |
[1001.34 --> 1002.44] But there was an embargo. |
[1003.06 --> 1006.58] And it wasn't technically broken. |
[1006.88 --> 1007.18] I don't know. |
[1007.22 --> 1009.04] You can frame that however you want. |
[1010.62 --> 1010.94] But... |
[1010.94 --> 1012.26] Yeah. |
[1012.34 --> 1012.66] So it says, |
[1012.92 --> 1013.64] it's true that AMD |
[1013.64 --> 1014.78] didn't actually reveal |
[1014.78 --> 1016.12] the details of the flaw |
[1016.12 --> 1017.68] before the embargo was up. |
[1017.98 --> 1019.64] But one of the company's developers |
[1019.64 --> 1021.00] came very close |
[1021.00 --> 1021.64] and close enough |
[1021.64 --> 1023.48] to infer what was actually going on. |
[1024.58 --> 1025.44] An AMD developer |
[1025.44 --> 1026.82] contributed a Linux patch |
[1026.82 --> 1028.38] that excluded AMD chips |
[1028.38 --> 1030.00] from the meltdown mitigation. |
[1030.62 --> 1031.90] In the note with the patch, |
[1031.98 --> 1032.66] the developer wrote, |
[1032.98 --> 1034.54] the AMD microarchitecture |
[1034.54 --> 1036.74] does not allow memory references, |
[1037.00 --> 1038.40] including speculative references |
[1038.40 --> 1040.68] that access higher privileged data |
[1040.68 --> 1041.44] when running |
[1041.44 --> 1042.92] in a lesser privileged mode |
[1042.92 --> 1044.04] when that access |
[1044.04 --> 1045.64] would result in a page fault. |
[1046.02 --> 1047.00] So with that information, |
[1047.44 --> 1048.40] knowing that people |
[1048.40 --> 1050.12] are writing code |
[1050.12 --> 1051.28] and it's extremely important, |
[1051.28 --> 1053.30] having that |
[1053.30 --> 1054.42] basically is like, |
[1054.50 --> 1055.92] we are not vulnerable to this |
[1055.92 --> 1057.00] and knowing that people |
[1057.00 --> 1058.00] are working on something |
[1058.00 --> 1058.62] very major |
[1058.62 --> 1059.34] and he's like |
[1059.34 --> 1060.68] excluding themselves from it, |
[1060.68 --> 1061.52] you can kind of figure out |
[1061.52 --> 1062.20] what's going on. |
[1063.00 --> 1064.22] But then the guy |
[1064.22 --> 1065.28] kind of got |
[1065.28 --> 1066.56] murked by a, |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.