Just to be clear: the current code is OK and correct for non-compacted buffers. Since we currently disable the compacted buffers, this patch has no effect on current kernels.
This patch fixes the (currently unused) calculation for sizing the compacted-format buffer. I can either send it now, or try to make sure it gets picked up by whoever goes back and re-implents XSAVES/compact-format support. On 08/06/2015 12:15 AM, Ingo Molnar wrote: > * Dave Hansen <[email protected]> wrote: >>> I realize that the calculation and what CPUID gives us should match, but >>> it's >>> not really good for the kernel to not know the precise layout of a critical >>> task context data structure ... >> >> There is no architectural guarantee that the sum of xstate sizes will be the >> same as what comes out of that CPUID leaf. It would be nice, but it's not >> architectural and I've run in to platforms where that assumption does not >> hold. > > WHY? >From a real dmesg: [ 0.000000] x86/fpu: xstate_offset[2]: 0240, xstate_sizes[2]: 0100 [ 0.000000] x86/fpu: xstate_offset[3]: 03c0, xstate_sizes[3]: 0040 [ 0.000000] x86/fpu: xstate_offset[4]: 0400, xstate_sizes[4]: 0040 ... Note: 0x240 + 0x100 != 0x3c0. > What sense does it make to have a blob we don't know the exact layout of? How > will > debuggers or user-space in general be able to print (and change) the register > values if they don't know the layout? Ingo, we know the layout. We know where every component is. We know how big each component is. This patch does not change the fact that we calculate and store that. The *ONLY* thing it does it not derive the total (compacted) buffer size from that layout since we have another simpler way of getting it. In fact, it makes the compacted format size calculation work in an analogous way to the non-compacted one that works today! > If 'compacted' format means "binary blob only the CPU can decode, not the > kernel" > then our answer is "uhm, no, thank you, we'll use standard format instead" ... Nobody is saying that. We need to read its contents (like with MPX), and the CPU tells us everything we need in order to decode it. > And no, "it's not Intel architectural" is a stupid and somewhat circular > argument > IMHO: the kernel always knew how to decompose CPU context dumps and you'll > have to > come up with a damn better reason to break that than pointing at some text in > an > Intel document. setup_xstate_features() still populates xstate_offsets[] which tells us where to find each field. This patch does not change that. This only changes how we size the (compacted) buffer, not how we decompose it. >>> So can we turn this into 'double check the CPUID size and print a warning >>> on >>> mismatch' kind of boot time sanity check? Preferably for all XSAVE* data >>> formats we can run into. I'd be fine with applying such a patch ahead of >>> enabling compaction again. >> >> I don't think that is sufficient. >> >> There are 4 reasons to apply this patch that I can think of: >> 1. There is no architectural guarantee that the calculation (sum of >> xstate sizes) will match what CPUID gives us as the size of the >> buffer. I've seen this in practice. > > So the context layout and structure on such CPUs has to be mapped and > properly > taken into account in the size calculation. How can GDB or any other (kernel) > debugger display (and change) individual fields reliably if the layout is not > known? Anything wanting to decode the buffer needs to read the format from CPUID and decode it that way. GDB from what I hear actually does it wrong and there are folks looking to fix it. >> 2. The alignment bit indicates that there is space used in the buffer >> which is not part of a state component. The current code does not >> take that in to account. > > Then it has to be taken it into account - just like user-space has to take it > into > account if it wants to display (and change) individual fields... This isn't the end of the world (it's just a few more lines and cpuid calls), but it is unnecessary work. >> 3. The code is currently asking for the size of an XSAVE-produced >> buffer. The code will be wrong the moment we switch to XSAVES >> because XSAVES saves more things than XSAVE and uses more space. > > This will have to be fixed before we move to compacted format. This patch is intended to help us move in the direction of re-enabling the compacted format. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
