Skip to content
  • Categories
  • Recent
  • Tags
  • Popular
  • World
  • Users
  • Groups
Skins
  • Light
  • Brite
  • Cerulean
  • Cosmo
  • Flatly
  • Journal
  • Litera
  • Lumen
  • Lux
  • Materia
  • Minty
  • Morph
  • Pulse
  • Sandstone
  • Simplex
  • Sketchy
  • Spacelab
  • United
  • Yeti
  • Zephyr
  • Dark
  • Cyborg
  • Darkly
  • Quartz
  • Slate
  • Solar
  • Superhero
  • Vapor

  • Default (Cyborg)
  • No Skin
Collapse
Brand Logo

CIRCLE WITH A DOT

  1. Home
  2. Uncategorized
  3. "Fun bug of the month, mesa edition, episode may"

"Fun bug of the month, mesa edition, episode may"

Scheduled Pinned Locked Moved Uncategorized
25 Posts 11 Posters 33 Views
  • Oldest to Newest
  • Newest to Oldest
  • Most Votes
Reply
  • Reply as topic
Log in to reply
This topic has been deleted. Only users with topic management privileges can see it.
  • karolherbst@chaos.socialK karolherbst@chaos.social

    "Fun bug of the month, mesa edition, episode may"

    so if you do "uint64_t some_var = 1 << 31;" in C you get "0xffffffff80000000" as the value, because that's super obvious and not confusing at all.

    It's pretty funny getting reminded how non-intuitive and broken C is from time to time.

    J This user is from outside of this forum
    J This user is from outside of this forum
    jann@infosec.exchange
    wrote last edited by
    #5

    @karolherbst I think that's UB? see C99 6.5.7 "Bitwise shift operators" - the LHS is signed and the result of the computation is not representable in the result type

    J 1 Reply Last reply
    0
    • J jann@infosec.exchange

      @karolherbst I think that's UB? see C99 6.5.7 "Bitwise shift operators" - the LHS is signed and the result of the computation is not representable in the result type

      J This user is from outside of this forum
      J This user is from outside of this forum
      jann@infosec.exchange
      wrote last edited by
      #6

      @karolherbst but apparently gcc has decided to not treat it as UB, except when using UBSAN: https://gcc.gnu.org/onlinedocs/gcc/Integers-implementation.html

      karolherbst@chaos.socialK P 2 Replies Last reply
      0
      • J jann@infosec.exchange

        @karolherbst but apparently gcc has decided to not treat it as UB, except when using UBSAN: https://gcc.gnu.org/onlinedocs/gcc/Integers-implementation.html

        karolherbst@chaos.socialK This user is from outside of this forum
        karolherbst@chaos.socialK This user is from outside of this forum
        karolherbst@chaos.social
        wrote last edited by
        #7

        @jann yeah technically it's UB, but there is only so much you can optimize with a 1-2 instruction pattern that it doesn't really matter in practice, because most impls will do the same (more or less).

        Like there is UB and then there is UB.

        J david_chisnall@infosec.exchangeD 2 Replies Last reply
        0
        • karolherbst@chaos.socialK karolherbst@chaos.social

          @jann yeah technically it's UB, but there is only so much you can optimize with a 1-2 instruction pattern that it doesn't really matter in practice, because most impls will do the same (more or less).

          Like there is UB and then there is UB.

          J This user is from outside of this forum
          J This user is from outside of this forum
          jann@infosec.exchange
          wrote last edited by
          #8

          @karolherbst yeah, I guess my point is that, for the code you showed, a C compiler would be well within its rights to refuse to build that code or complain about it, so this is not entirely the language's fault

          karolherbst@chaos.socialK 1 Reply Last reply
          0
          • J jann@infosec.exchange

            @karolherbst but apparently gcc has decided to not treat it as UB, except when using UBSAN: https://gcc.gnu.org/onlinedocs/gcc/Integers-implementation.html

            P This user is from outside of this forum
            P This user is from outside of this forum
            pinskia@hachyderm.io
            wrote last edited by
            #9

            @jann @karolherbst

            It was not UB in C90. That is why it was UB without ubsan ...

            1 Reply Last reply
            0
            • J jann@infosec.exchange

              @karolherbst yeah, I guess my point is that, for the code you showed, a C compiler would be well within its rights to refuse to build that code or complain about it, so this is not entirely the language's fault

              karolherbst@chaos.socialK This user is from outside of this forum
              karolherbst@chaos.socialK This user is from outside of this forum
              karolherbst@chaos.social
              wrote last edited by
              #10

              @jann ohh it's totally the languages fault even if it wouldn't be UB, because that's just the worst way to specify this.

              Like it's just a design bug really. And no matter how much this is UB or not won't change that.

              1 Reply Last reply
              0
              • trilader@chaos.socialT This user is from outside of this forum
                trilader@chaos.socialT This user is from outside of this forum
                trilader@chaos.social
                wrote last edited by
                #11

                @puppethead @karolherbst When not using the U (or L) suffix 1<<31 triggers clang's -Wshift-sign-overflow warning. However that warning is not enabled by default and gcc doesn't support it at all.

                1 Reply Last reply
                0
                • karolherbst@chaos.socialK karolherbst@chaos.social

                  @jann yeah technically it's UB, but there is only so much you can optimize with a 1-2 instruction pattern that it doesn't really matter in practice, because most impls will do the same (more or less).

                  Like there is UB and then there is UB.

                  david_chisnall@infosec.exchangeD This user is from outside of this forum
                  david_chisnall@infosec.exchangeD This user is from outside of this forum
                  david_chisnall@infosec.exchange
                  wrote last edited by
                  #12

                  @karolherbst @jann

                  It’s UB in the general case because, if the operand is not a constant, you want to lower it to a shift instruction but C works with targets that have different number representations. Ones or twos complements, or explicit sign bits are all permitted, but all of these will give different behaviours if you flip the top bit.

                  For wider shifts, different ISAs had different semantics for shifts wider than the register, so C made that fully undefined.

                  This combination lets you lower source-level shifts to a shift instruction.

                  C also doesn’t mandate that this be constant evaluated unless the result is used as a constant, so there’s no way to force implementations to diagnose the UB at compile time for this case. But, as a QoI issue, it is permitted and compilers should.

                  karolherbst@chaos.socialK 1 Reply Last reply
                  0
                  • trilader@chaos.socialT trilader@chaos.social

                    @karolherbst For my understanding: That's default int promotion + sign extend on 64 bit extension? Would 1L << 31L fix this or is there other pitfalls with that?

                    ewhac@mastodon.socialE This user is from outside of this forum
                    ewhac@mastodon.socialE This user is from outside of this forum
                    ewhac@mastodon.social
                    wrote last edited by
                    #13

                    @trilader @karolherbst "Um, actually..."

                    I believe it would work as expected with:

                    1U << 31;

                    The unexpected part is that sign extension from 32 to 64 bits takes place before reinterpretation to unsigned. The C99 standard is admittedly opaque on this point. If you make the rvalue unsigned as well, then you get the (presumably) expected result.

                    (Just tested it with GCC 13.3 -- it works.)

                    trilader@chaos.socialT 1 Reply Last reply
                    0
                    • ewhac@mastodon.socialE ewhac@mastodon.social

                      @trilader @karolherbst "Um, actually..."

                      I believe it would work as expected with:

                      1U << 31;

                      The unexpected part is that sign extension from 32 to 64 bits takes place before reinterpretation to unsigned. The C99 standard is admittedly opaque on this point. If you make the rvalue unsigned as well, then you get the (presumably) expected result.

                      (Just tested it with GCC 13.3 -- it works.)

                      trilader@chaos.socialT This user is from outside of this forum
                      trilader@chaos.socialT This user is from outside of this forum
                      trilader@chaos.social
                      wrote last edited by
                      #14

                      @ewhac @karolherbst Yes. In the other thread leg of the original post I also posted about that clang even warns you about the behavior of 1 << 31 without U or L suffix, provided you enable the right, off by default, warning that GCC doesn't have. Another leg notes that GCC catches this at runtime with ubsan enabled.

                      1 Reply Last reply
                      0
                      • karolherbst@chaos.socialK karolherbst@chaos.social

                        "Fun bug of the month, mesa edition, episode may"

                        so if you do "uint64_t some_var = 1 << 31;" in C you get "0xffffffff80000000" as the value, because that's super obvious and not confusing at all.

                        It's pretty funny getting reminded how non-intuitive and broken C is from time to time.

                        pavel@social.kernel.orgP This user is from outside of this forum
                        pavel@social.kernel.orgP This user is from outside of this forum
                        pavel@social.kernel.org
                        wrote last edited by
                        #15
                        @karolherbst C is 1973 or so. If you believe it is confusing, try assembly :-).
                        1 Reply Last reply
                        0
                        • trilader@chaos.socialT trilader@chaos.social

                          @karolherbst For my understanding: That's default int promotion + sign extend on 64 bit extension? Would 1L << 31L fix this or is there other pitfalls with that?

                          pavel@social.kernel.orgP This user is from outside of this forum
                          pavel@social.kernel.orgP This user is from outside of this forum
                          pavel@social.kernel.org
                          wrote last edited by
                          #16
                          @trilader @karolherbst Yes, int promotion. 1L<<31L would still be broken on 32-bit architectures (as long is 32 bit there). You'd need 1LL or something, AFAICT.
                          1 Reply Last reply
                          0
                          • trilader@chaos.socialT trilader@chaos.social

                            @karolherbst For my understanding: That's default int promotion + sign extend on 64 bit extension? Would 1L << 31L fix this or is there other pitfalls with that?

                            pavel@social.kernel.orgP This user is from outside of this forum
                            pavel@social.kernel.orgP This user is from outside of this forum
                            pavel@social.kernel.org
                            wrote last edited by
                            #17
                            @trilader @karolherbst Actually right solution would be uint64_t some_var = ((uint 64_t)1) << 31; AFAICT.
                            karolherbst@chaos.socialK 1 Reply Last reply
                            0
                            • david_chisnall@infosec.exchangeD david_chisnall@infosec.exchange

                              @karolherbst @jann

                              It’s UB in the general case because, if the operand is not a constant, you want to lower it to a shift instruction but C works with targets that have different number representations. Ones or twos complements, or explicit sign bits are all permitted, but all of these will give different behaviours if you flip the top bit.

                              For wider shifts, different ISAs had different semantics for shifts wider than the register, so C made that fully undefined.

                              This combination lets you lower source-level shifts to a shift instruction.

                              C also doesn’t mandate that this be constant evaluated unless the result is used as a constant, so there’s no way to force implementations to diagnose the UB at compile time for this case. But, as a QoI issue, it is permitted and compilers should.

                              karolherbst@chaos.socialK This user is from outside of this forum
                              karolherbst@chaos.socialK This user is from outside of this forum
                              karolherbst@chaos.social
                              wrote last edited by
                              #18

                              @david_chisnall @jann at least C23 fixes one part of this by requiring two's complement for integers.

                              But also, I just wished C would mandate that constants are just assumed to be of the "expected" type, because in 99.999999% of all cases a programmer really meant the obvious thing with "uint64_t x = 1 << 31".

                              But I guess we'll just keep those horrible semantics C has in a couple of areas, because nobody want to fix those things, because "it could break things".

                              blp@framapiaf.orgB 1 Reply Last reply
                              0
                              • pavel@social.kernel.orgP pavel@social.kernel.org
                                @trilader @karolherbst Actually right solution would be uint64_t some_var = ((uint 64_t)1) << 31; AFAICT.
                                karolherbst@chaos.socialK This user is from outside of this forum
                                karolherbst@chaos.socialK This user is from outside of this forum
                                karolherbst@chaos.social
                                wrote last edited by
                                #19

                                @pavel @trilader the actual right solution would be to fix the language 😛

                                pavel@social.kernel.orgP 1 Reply Last reply
                                0
                                • karolherbst@chaos.socialK karolherbst@chaos.social

                                  @david_chisnall @jann at least C23 fixes one part of this by requiring two's complement for integers.

                                  But also, I just wished C would mandate that constants are just assumed to be of the "expected" type, because in 99.999999% of all cases a programmer really meant the obvious thing with "uint64_t x = 1 << 31".

                                  But I guess we'll just keep those horrible semantics C has in a couple of areas, because nobody want to fix those things, because "it could break things".

                                  blp@framapiaf.orgB This user is from outside of this forum
                                  blp@framapiaf.orgB This user is from outside of this forum
                                  blp@framapiaf.org
                                  wrote last edited by
                                  #20

                                  @karolherbst @david_chisnall @jann That particular change would break a lot!

                                  1 Reply Last reply
                                  0
                                  • karolherbst@chaos.socialK karolherbst@chaos.social

                                    "Fun bug of the month, mesa edition, episode may"

                                    so if you do "uint64_t some_var = 1 << 31;" in C you get "0xffffffff80000000" as the value, because that's super obvious and not confusing at all.

                                    It's pretty funny getting reminded how non-intuitive and broken C is from time to time.

                                    L This user is from outside of this forum
                                    L This user is from outside of this forum
                                    lalufu@mastodon.social
                                    wrote last edited by
                                    #21

                                    @karolherbst I recently read https://www.os2museum.com/wp/bitfield-pitfalls/ which is a similar pitfall?

                                    1 Reply Last reply
                                    0
                                    • karolherbst@chaos.socialK karolherbst@chaos.social

                                      @pavel @trilader the actual right solution would be to fix the language 😛

                                      pavel@social.kernel.orgP This user is from outside of this forum
                                      pavel@social.kernel.orgP This user is from outside of this forum
                                      pavel@social.kernel.org
                                      wrote last edited by
                                      #22
                                      @karolherbst @trilader Yeah, travelling to 1972 and changing the language would be best :-).
                                      1 Reply Last reply
                                      0
                                      • karolherbst@chaos.socialK karolherbst@chaos.social

                                        "Fun bug of the month, mesa edition, episode may"

                                        so if you do "uint64_t some_var = 1 << 31;" in C you get "0xffffffff80000000" as the value, because that's super obvious and not confusing at all.

                                        It's pretty funny getting reminded how non-intuitive and broken C is from time to time.

                                        lkundrak@metalhead.clubL This user is from outside of this forum
                                        lkundrak@metalhead.clubL This user is from outside of this forum
                                        lkundrak@metalhead.club
                                        wrote last edited by
                                        #23

                                        @karolherbst taking off my "understands sign extension" badge

                                        wore it with pride, but the pride was misplaced

                                        1 Reply Last reply
                                        0
                                        • trilader@chaos.socialT trilader@chaos.social

                                          @karolherbst For my understanding: That's default int promotion + sign extend on 64 bit extension? Would 1L << 31L fix this or is there other pitfalls with that?

                                          lkundrak@metalhead.clubL This user is from outside of this forum
                                          lkundrak@metalhead.clubL This user is from outside of this forum
                                          lkundrak@metalhead.club
                                          wrote last edited by
                                          #24

                                          @trilader @karolherbst 1L << alone would do the trick on real world 64-bit machines, but i think compiler is still fully allowed to do the wrong thing. msvc perhaps still has 32-bit longs on 64-bit platforms?

                                          i think you need to make sure it's unsigned so that sign extension has no chance of occurring, so my money is on 1U << 31?

                                          1 Reply Last reply
                                          0
                                          Reply
                                          • Reply as topic
                                          Log in to reply
                                          • Oldest to Newest
                                          • Newest to Oldest
                                          • Most Votes


                                          • Login

                                          • Login or register to search.
                                          • First post
                                            Last post
                                          0
                                          • Categories
                                          • Recent
                                          • Tags
                                          • Popular
                                          • World
                                          • Users
                                          • Groups