| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | [iobuf] Ensure I/O buffer data sits within unshared cachelines | Michael Brown | 2025-07-07 | 1 | -1/+7 |
| | | | | | | | | | | | | On platforms where DMA devices are not in the same coherency domain as the CPU cache, we must ensure that DMA I/O buffers do not share cachelines with other data. Align the start and end of I/O buffers to IOB_ZLEN, which is larger than any cacheline size we expect to encounter. Signed-off-by: Michael Brown <mcb30@ipxe.org> | ||||
| * | [iobuf] Improve robustness of I/O buffer allocation | Michael Brown | 2016-02-11 | 1 | -0/+136 |
| Guard against various corner cases (such as zero-length buffers, zero alignments, and integer overflow when rounding up allocation lengths and alignments) and ensure that the struct io_buffer is correctly aligned even when the caller requests a non-zero alignment for the I/O buffer itself. Add self-tests to verify that the resulting alignments and lengths are correct for a range of allocations. Signed-off-by: Michael Brown <mcb30@ipxe.org> | |||||
